Re: 중간엽 줄기세포의 생존율을 높이는 NO(산화질소)

[문형철]

Intracellular delivery of nitric oxide enhances the therapeutic efficacy of mesenchymal stem cells for myocardial infarction

Tian Hao https://orcid.org/0009-0007-8657-0424, Guangbo Ji https://orcid.org/0009-0000-7079-8896, Meng Qian https://orcid.org/0009-0006-9135-4593, Qiu Xuan Li, Haoyan Huang, Shiyu Deng, Pei Liu, Weiliang Deng https://orcid.org/0009-0001-8402-7280, Yongzhen Wei, [...], and Qiang Zhao https://orcid.org/0000-0003-4656-6002 +9 authorsAuthors Info & Affiliations

Science Advances

29 Nov 2023

Vol 9, Issue 48

DOI: 10.1126/sciadv.adi9967

4,3612

Metrics

Total Downloads4,361

• Last 6 Months1,270
• Last 12 Months3,995

Total Citations2

• Last 6 Months1
• Last 12 Months2

NotificationsBookmark

• • Abstract
  • INTRODUCTION
  • RESULTS
  • DISCUSSION
  • MATERIALS AND METHODS
  • Acknowledgments
  • Supplementary Materials
  • REFERENCES AND NOTES
  • eLetters (0)
• • Information & Authors
  • Metrics & Citations
  • View Options
  • References
  • Media
  • Tables
  • Share

Abstract

Cell therapy by autologous mesenchymal stem cells (MSCs) is a clinically acceptable strategy for treating various diseases. Unfortunately, the therapeutic efficacy is largely affected by the low quality of MSCs collected from patients. Here, we showed that the gene expression‎ of MSCs from patients with diabetes was differentially regulated compared to that of MSCs from healthy controls. Then, MSCs were genetically engineered to catalyze an NO prodrug to release NO intracellularly. Compared to extracellular NO conversion, intracellular NO delivery effectively prolonged survival and enhanced the paracrine function of MSCs, as demonstrated by in vitro and in vivo assays. The enhanced therapeutic efficacy of engineered MSCs combined with intracellular NO delivery was further confirmed in mouse and rat models of myocardial infarction, and a clinically relevant cell administration paradigm through secondary thoracotomy has been attempted.

요약

자가 중간엽 줄기세포(MSC)를 이용한 세포 치료는 

다양한 질병을 치료하기 위해 임상적으로 허용되는 전략입니다. 

안타깝게도 치료 효과는 

환자로부터 채취한 MSC의 낮은 품질에 크게 영향을 받습니다. 

본 연구에서는 

당뇨병 환자에서 채취한 MSC의 유전자 발현이 

건강한 대조군에서 채취한 MSC와 비교하여 다르게 조절된다는 것을 보여주었습니다. 

그런 다음 MSC가 NO 전구체를 촉매하여 세포 내로 NO를 방출하도록 유전적으로 조작했습니다. 

세포 외 NO 전환과 비교하여 세포 내 NO 전달은 

시험관 및 생체 내 분석에서 입증된 바와 같이 생존을 효과적으로 연장하고 

MSC의 파라크린 기능을 향상시켰습니다. 

세포 내 NO 전달과 결합된 엔지니어링 MSC의 향상된 치료 효능은 

심근경색 마우스 및 쥐 모델에서 추가로 확인되었으며, 

이차 개흉술을 통한 임상 관련 세포 투여 패러다임이 시도되고 있습니다.

SIGN UP FOR THE AWARD-WINNING SCIENCEADVISER NEWSLETTER

The latest news, commentary, and research, free to your inbox daily

Sign up

INTRODUCTION

Mesenchymal stem cells (MSCs) are pluripotent stem cells with high self-renewal abilities and multidirectional differentiation potential (1, 2). They are extensively distributed throughout the body and serve a variety of purposes, including tissue regeneration (3), immunoregulation (4), and angiogenesis (5). However, an important challenge for stem cell therapy is the low survival rate of stem cells after transplantation, which is associated with nonspecific homing of cells and ischemia/hypoxia at the injury site (6). In addition, transplanted stem cells cannot fully exert their paracrine effects in pathological environments, which seriously limits the clinical application of stem cells.

The proangiogenic function of MSCs is a key factor contributing to the treatment of ischemic diseases. Generally, MSCs can stimulate local angiogenesis in ischemic tissue by secreting cytokines, including vascular endothelial growth factor (VEGF), transforming growth factor β (TGFβ), and basic fibroblast growth factor (bFGF), to induce endothelial cells to form tubular structures (7). In addition, MSCs can recruit pericytes and smooth muscle cells to promote neovascularization (8, 9). Recently, MSCs have been reported to participate in the construction of blood vessels by directly differentiating into endothelial or smooth muscle cells (10).

As an important signaling molecule, nitric oxide (NO) plays a crucial role in the maintenance of vascular homeostasis by inhibiting thrombus formation and promoting angiogenesis (11, 12). Recently, increasing attention has been given to the regulation of stem cells by NO due to its multiple biological functions (13, 14). NO can regulate the proliferation of stem cells by regulating the activities of cyclin and mitotic receptors, as well as their downstream pathways (15, 16). On the other hand, NO can regulate the expression‎ of angiogenic cytokines and immunomodulatory factors to improve the paracrine performance of stem cells (17). Additionally, recent studies have demonstrated that NO can regulate the differentiation behavior of stem cells through the phosphatidylinositol 3-kinase (PI3K)/AKT, guanosine 3′,5′-monophosphate (cGMP), and other signaling pathways (18).

As a result, NO-releasing biomaterials have been used as delivery carriers for stem cells to enhance their survival and regulate paracrine functions (19). However, NO, which is a gaseous molecule, easily diffuses and has a high level of instability. Furthermore, the physiologic function of NO is dose dependent, and an overdose of NO often leads to notable cytotoxicity (20). Thus, optimizing the beneficial effects of NO to strengthen the therapeutic efficacy of stem cells by tuning their release profile should be taken into account.

In our previous study, based on the chemical biology principle of “bump-and-hole,” we designed and prepared an enzyme-prodrug delivery system and achieved targeted delivery of NO at the lesion site in two different ischemic disease models (21). Here, MSCs were further modified by gene transfection to express a catalytic enzyme (A4-β-GalH363A). The engineered MSCs (eMSCs) were transplanted using an injectable hyaluronic acid (HA) hydrogel as the carrier, while the NO prodrug was injected through the tail vein to achieve controlled release of NO catalyzed by the enzymes in eMSCs. The therapeutic efficacy of MSCs combined with exogenous NO delivery was eval‎uated in mouse and rat models of myocardial infarction (MI) with an emphasis on comparing the therapeutic efficacy of two different NO administration methods (intracellular or extracellular), and the underlying mechanism of their myocardial protective effect was further explored.

소개

중간엽 줄기세포(MSC)는

높은 자기 재생 능력과 다방향 분화 잠재력을 가진 만능 줄기세포입니다(1, 2).

체내에 광범위하게 분포하며

조직 재생(3), 면역 조절(4), 혈관 신생(5) 등 다양한 용도로 활용되고 있습니다.

그러나

줄기세포 치료의 중요한 과제는

이식 후 줄기세포의 낮은 생존율이며,

이는 세포의 비특이적 이동 및 손상 부위의 허혈/저산소증과 관련이 있습니다(6).

또한 이식된 줄기세포는

병리학적인 환경에서 파라크린 효과를 충분히 발휘할 수 없어

줄기세포의 임상 적용에 심각한 제한이 있습니다.

MSC의 혈관 신생 기능은

허혈성 질환 치료에 기여하는 핵심 요소입니다.

일반적으로

MSC는

혈관 내피 성장 인자(VEGF), 형질 전환 성장 인자 β(TGFβ), 기저 섬유아세포 성장 인자(bFGF) 등의

사이토카인을 분비하여

내피 세포가 관 구조를 형성하도록 유도함으로써

허혈성 조직에서 국소 혈관 생성을 자극할 수 있습니다(7).

또한,

MSC는

혈관 주변 세포와 평활근 세포를 모집하여 신생 혈관 형성을 촉진할 수 있습니다(8, 9).

최근에는

MSC가 내피세포나 평활근세포로 직접 분화하여

혈관 생성에 관여하는 것으로 보고되고 있습니다(10).

산화질소(NO)는

중요한 신호 분자로서

혈전 형성을 억제하고

혈관 신생을 촉진하여 혈관 항상성 유지에 중요한 역할을 합니다(11, 12).

최근에는

다양한 생물학적 기능으로 인해

NO에 의한 줄기세포의 조절에 대한 관심이 높아지고 있습니다(13, 14).

NO는

사이클린과 유사 분열 수용체의 활성과 그 하류 경로를 조절하여

줄기 세포의 증식을 조절할 수 있습니다 (15, 16).

한편,

NO는

혈관 신생성 사이토카인과 면역 조절 인자의 발현을 조절하여

줄기세포의 파라크린 성능을 향상시킬 수 있습니다(17).

또한

최근 연구에 따르면

NO는

포스파티딜이노시톨 3-키나아제(PI3K)/AKT, 구아노신 3′,5′- 모노포스페이트(cGMP) 및

기타 신호 경로를 통해

줄기세포의 분화 행동을 조절할 수 있음이 입증되었습니다(18).

그 결과,

NO 방출 생체 물질은

줄기세포의 생존을 향상시키고

파라크린 기능을 조절하기 위한 전달체로 사용되어 왔습니다(19).

그러나

기체 분자인 NO는 쉽게 확산되고 불안정성이 높습니다.

또한

NO의 생리적 기능은 용량에 따라 달라지며,

NO를 과다 복용하면 종종 현저한 세포 독성을 유발합니다(20).

따라서

줄기세포의 방출 프로파일을 조정하여

줄기세포의 치료 효능을 강화하기 위해

NO의 유익한 효과를 최적화하는 것이 고려되어야 합니다.

이전 연구에서는

“범프 앤 홀”의 화학 생물학적 원리를 기반으로

효소-약물 전달 시스템을 설계하고 준비하여

두 가지 허혈성 질환 모델에서 병변 부위에 NO의 표적 전달을 달성했습니다(21).

여기서

MSC는 촉매 효소(A4-β-GalH363A)를 발현하도록 유전자 형질전환을 통해 추가로 변형되었습니다.

이렇게 조작된 MSC(eMSC)를 주사 가능한 히알루론산(HA) 하이드로젤을 운반체로 사용하여 이식하고,

꼬리정맥을 통해 NO 전구체를 주입하여 eMSC의 효소에 의해 촉매된 NO의 방출을 제어했습니다.

두 가지 다른 NO 투여 방법(세포 내 또는 세포 외)의 치료 효능을 비교하는 데 중점을 두고

심근경색(MI) 마우스 및 쥐 모델에서 외인성 NO 전달과 결합된 MSC의 치료 효능을 평가하고

심근 보호 효과의 근본 메커니즘을 추가로 탐색했습니다.

RESULTS

The gene expression‎ of MSCs collected from patients is differentially regulated

In the clinic, the therapeutic efficacy of autologous stem cell transplantation is largely affected by the low quality (including cell survival and paracrine function) of stem cells collected from patients due to chronic diseases. Diabetes mellitus is a chronic medical condition that can lead to a variety of complications, and these complications can affect various parts of the body, such as the kidney, lower limb, and heart. Diabetes predisposes affected individuals to a spectrum of cardiovascular complications, and one of the most debilitating in terms of prognosis is heart failure (22).

Accordingly, although autologous MSCs have been widely accepted as a promising strategy for treating various complications associated with diabetes, the therapeutic efficacy is largely affected by the quality of stem cells collected from the patients themselves. The gene levels in adipose-derived MSCs (ADMSCs) from patients with diabetes and healthy individuals were first compared by transcriptome sequencing. The heatmap shows that multiple genes in MSCs from patients with diabetes were up- or down-regulated compared to healthy controls (Fig. 1A). Gene ontology (GO) enrichment analysis revealed that the most differentially up-regulated genes were related to tumor necrosis factor (TNF) signaling pathways (Fig. 1B). Subsequently, we performed gene set enrichment analysis (GSEA) based on the RNA-sequencing results. GSEA revealed that inflammatory target genes were highly enriched in MSCs from patients with diabetes, with a normalized enrichment score (NES) of 1.65 (P < 0.01) (Fig. 1C). Apoptosis-related genes were also highly enriched, with an NES of 1.77 (P < 0.01) (Fig. 1D). In addition, through enrichment analysis, we found that proangiogenic genes in MSCs were greatly down-regulated in patients with diabetes (Fig. 1E).

Fig. 1. Transcriptome sequencing of adipose mesenchymal stem cells (MSCs) collected from patients with diabetes and healthy controls.

(A) Heatmap showing the differentially expressed genes of MSCs from patients with diabetes and healthy controls (n = 3). (B) Gene ontology (GO) analysis of the up-regulated transcriptome of MSCs from patients with diabetes. (C and D) Gene set enrichment analysis (GSEA) was performed to determine the enrichment of inflammation (C) and apoptosis (D) target genes in the diabetic group. (E) Heatmap showing angiogenesis-related genes in the two groups (n = 3). (F) Schematic illustration demonstrating the difference in MSCs between patients with diabetes and healthy controls at the gene level.

Open in viewer

eMSCs are constructed to express mutant β-galactosidase

Here, we first designed and constructed plasmids carrying the mutant β-galactosidase (A4-β-GalH363A) target gene and luciferase–red fluorescent protein (RFP) dual reporter genes, which could be further used for in vivo imaging. eMSCs expressing A4-β-GalH363A were constructed by infecting MSCs with lentiviruses obtained from human embryonic kidney 293T cells (Fig. 2A). Immunofluorescence staining for RFP confirmed that A4-β-GalH363A was successfully expressed by eMSCs (Fig. 2B). The subcellular fraction and intracellular distribution of enzymes expressed by the eMSCs was determined by Western blotting (Fig. 2C). In contrast to natural β-galactosidases, which are widely distributed within cells, A4-β-GalH363A was mainly confined to the nucleus of eMSCs.

Fig. 2. Intracellular expression‎ and localization of A4-β-GalH363A.

(A) Schematic diagram of lentivirus packaging and mesenchymal stem cell (MSC) infection. (B) Immunofluorescence staining of β-Gal and A4-β-GalH363A in eMSCs. Scale bar, 50 μm. (C) The distribution of two different enzymes in engineered MSCs (eMSCs) was analyzed by Western blotting.

Open in viewer

Intracellular release of NO is realized via decomposition of the 6-OMeGal-Ph-NO prodrug

Since the mutant β-galactosidase was confined to the nucleus in eMSCs, we designed a prodrug by introducing a lipid-soluble self-decomposition chain into 6-OMeGal-NO to improve its oil and water distribution coefficient; therefore, the resultant NO donor 6-OMeGal-Ph-NO could penetrate the cell membrane and decompose and release NO under the catalysis of A4-β-GalH363A (Fig. 3A and Supplementary Materials). An in vitro release assay showed that the 6-OMeGal-Ph-NO prodrug was efficiently recognized and converted by A4-β-GalH363A with a cumulative release ratio of approximately 97.3%, while nearly no release was observed in the presence of wild-type β-galactosidase (Fig. 3B). To confirm‎ intracellular NO release, eMSCs were preincubated with an NO fluorescent probe (DAF-AM DA) and then treated with different NO prodrugs. The changes in fluorescence signals were examined by time-delay cell imaging (Fig. 3C). The results indicated that the fluorescence intensity continuously increased in the group that was treated with 6-OMeGal-Ph-NO, indicating conversion into NO (Fig. 3D). In contrast, no detectable changes were identified in the β-Gal-NO group because the high water solubility restricted its entry into eMSCs. Next, intracellular and extracellular release of NO in eMSCs was assessed (Fig. 3E). The quantity of intracellular NO was measured by electron paramagnetic resonance (EPR) using ferrous N-diethyl dithiocarbamate (DETC2-Fe) as the spin-trapping reagent. The resultant NO adduct (DETC2-Fe-NO) exhibited a characteristic triplet EPR signal (aN = 13.06 G, giso = 2.041) at room temperature. Quantitative analysis showed that the NO level was significantly (P < 0.001 or 0.0001) higher in eMSCs treated with 6-OMeGal-Ph-NO than in the β-Gal-NO and control groups (Fig. 3F). Furthermore, in the group treated with β-Gal-NO, the release of NO was mainly catalyzed by β-galactosidase that translocated from the cytoplasm in eMSCs, and the extracellular release profile of β-Gal-NO was confirmed by detecting the NO level in the cell culture medium with the NO-sensitive near-infrared fluorescence probe (23); it was significantly (P < 0.01) higher than that in the 6-OMeGal-Ph-NO and control groups (Fig. 3G). To determine the uptake of the NO prodrug by eMSCs, we incubated 6-OMeGal-Ph-NO with eMSCs, and the concentration in the culture medium was determined at different time points. The results reflected that approximately 45% of 6-OMeGal-Ph-NO was incorporated into the eMSCs within 12 hours (fig. S1).

Fig. 3. Intracellular generation of nitric oxide (NO) from the NO prodrug under the catalysis of A4-β-GalH363A expressed by engineered mesenchymal stem cells (eMSCs).

(A) Synthesis of two NO prodrugs with different enzyme response abilities and cellular permeabilities. (B) In vitro release profile of NO from the NO prodrug (6-OMeGal-Ph-NO) in the presence of β-Gal or A4-β-GalH363A. (C) Schematic illustration of intracellular NO imaging by using an NO fluorescence probe (DAF-AM DA). (D) Representative time-lapse images of NO generation from two different prodrugs in eMSCs and quantification of the fluorescence intensity (n = 6). ***P < 0.001, ****P < 0.0001 versus 6-OMeGal-Ph-NO group. (E) Schematic illustration showing the detection of intracellular and extracellular NO generation differentially. (F) Representative electron paramagnetic resonance (EPR) spectra and quantification of intracellular NO generation by measuring the DETC2-Fe-NO complex using 2,2,5,5-tetramethyl piperidine 1-oxyl (TEMPO) as a standard (n = 3). (G) Relative quantification of NO production in the medium determined using the near-infrared fluorescence probe (n = 4). Data are expressed as the mean ± SEM. Significant differences were detected by one-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

Expand for more

Open in viewer

Intracellular delivery of NO inhibits apoptosis and enhances the proangiogenic activity of eMSCs

Overproduction of reactive oxygen species due to cellular oxidative stress has been accepted as an important factor contributing to apoptosis in transplanted cells in ischemic tissue. Therefore, we assessed the protective effect of exogenously administered NO on the survival of eMSCs with H2O2-induced oxidative stress and focused on comparing the protection provided by intracellular and extracellular NO administration. The results showed that H2O2 stimulated apoptosis, and delivery of NO via extracellular and intracellular strategies significantly (P < 0.01 or 0.001) reduced apoptosis in eMSCs stimulated by oxidative stress, and the highest fluorescence signal was observed in response to intracellular NO delivery (Fig. 4A).

Fig. 4. Intracellular delivery of nitric oxide (NO) inhibits apoptosis of engineered mesenchymal stem cells (eMSCs).

(A) Bioluminescence imaging (BLI) was used to detect the effect of NO delivery on cell apoptosis stimulated by different concentrations of H2O2, and the fluorescence signals were further quantified (n = 3). **P < 0.01, ***P < 0.001, ****P < 0.0001 versus control group; #P < 0.05 versus β-Gal-NO group. (B) Flow cytometry assay of cell viability and apoptosis of eMSCs after H2O2 stimulation, and quantification of mean percent values of apoptotic cells (n = 3). (C) The expression‎ of apoptosis-related protein (BCL2, Bax, Bad, caspase3, and cleaved caspase3) by eMSCs was detected after H2O2 stimulation by Western blots (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001. (D) Representative BLI photographs reflecting the retention of eMSCs with and without intracellular NO delivery after in vivo transplantation as well as the quantitative analysis of signals (n = 3). Data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01.

Open in viewer

The survival of eMSCs after treatment with various NO delivery strategies was compared by using flow cytometry. Intracellular NO delivery had the most remarkable antiapoptotic effect on cells stimulated by oxidative stress (P < 0.001) (Fig. 4B). The expression‎ of apoptosis-related genes by eMSCs followed a similar trend; NO delivery (extracellular and intracellular) effectively promoted the expression‎ of the antiapoptotic gene BCL2 in eMSCs after oxidative stress stimulation, while the expression‎ levels of proapoptotic genes were reduced accordingly. Intracellular NO delivery via administration of the 6-OMeGal-Ph-NO prodrug exerted a more pronounced antiapoptotic effect at both the gene and protein levels (Fig. 4C and fig. S2A), which may be because the intracellular generation of NO directly activated the antioxidant system in cells to resist oxidative stress damage and inhibit further apoptosis.

The expression‎ of proangiogenic genes, including ANGPT1, ANGPT2, FGF2, VEGFA, and KDR, in eMSCs was further detected by reverse transcription polymerase chain reaction (RT-PCR). The results showed that the expression‎ level of proangiogenic genes was significantly (P < 0.05, 0.01, or 0.001) higher in eMSCs treated with the 6-OMeGal-Ph-NO prodrug than in the other groups, indicating the enhanced proangiogenic functions of eMSCs after intracellular NO delivery (fig. S2B).

We further eval‎uated the effect of NO delivery on the in vivo retention of eMSCs after orthotopic transplantation in the myocardial tissue of mice. As shown in Fig. 4D, intracellular delivery of NO effectively prolonged the retention of eMSCs within the myocardium, and an evident bioluminescence imaging (BLI) signal corresponding to the retention of eMSCs was observed 7 days after transplantation compared to the counterpart without administration of the NO prodrug. To further eval‎uate the translational potential of eMSCs in clinical settings, we used MSCs derived from diabetic patients and conducted a series of assays related to cell survival and paracrine function. The findings indicated that intracellular delivery of NO also confers advantages in the attenuation of cell apoptosis under stress conditions, thereby prolonging the in vivo retention of eMSCs (fig. S3).

Intracellular delivery of NO ameliorates myocardial injury in MI mice after treatment with eMSCs

The therapeutic efficacy of MSCs combined with exogenous NO was further eval‎uated in a mouse MI model (Fig. 5A and fig. S4). The inflammatory response in the early stage (3 days) was first detected by hematoxylin-eosin (H&E) staining and CD68 immunofluorescence staining. The results demonstrated that severe inflammatory cell infiltration occurred in the injured myocardium of MI mice, and it was effectively alleviated after eMSC treatment. More prominent restoration in the injured myocardium was observed after further administration of NO (Fig. 5, B and C), confirm‎ing the inhibitory effect on inflammation after MI provided by the combination of MSCs and NO. Moreover, this inhibitory effect was more significant (P < 0.01 or 0.0001) in the group with intracellular NO delivery than in the group with extracellular NO delivery.

Fig. 5. Intracellular delivery of nitric oxide (NO) ameliorates myocardial injury in myocardial infarction (MI) mice after treatment with engineered mesenchymal stem cells (eMSCs) via regulation of the inflammation and stimulation of angiogenesis.

(A) Experimental schedule for the treatment of MI in a mouse model. (B) Hematoxylin-eosin (H&E) staining was performed to detect inflammatory cell infiltration in the early stage of MI (n = 6). Scale bar, 100 μm. (C) Representative images of CD68 immunofluorescence staining (green) and quantification of CD68+ macrophages in injured myocardium (n = 6). Scale bar, 25 μm. (D and E) Flow cytometry was performed to detect peritoneal macrophage polarization 7 days after surgery followed by different treatments. TNFα- and CD206-positive ascites macrophages (markers of M1 and M2 macrophage phenotypes, respectively) were quantified accordingly (n = 3). (F) Representative images of α–Smooth muscle actin (α-SMA) immunofluorescence staining and the quantification of α-SMA+ arterioles (n = 6). Scale bar, 100 μm. (G) Representative images of von Willebrand factor (vWF) immunofluorescence staining and the quantification of vWF+ capillaries (n = 6). Scale bar, 100 μm. Data are expressed as the mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

Expand for more

Open in viewer

It has been widely accepted that a higher proportion of M2-type macrophages is more conducive to the repair of tissue damage. Thus, peritoneal macrophages were extracted at 7 days after surgery and examined by flow cytometry to assess the polarization of macrophages in MI mice that received different treatments. MI modeling leads to a marked increase in the polarization of macrophages toward the M1 phenotype (TNF-α positive); however, the ratio of M1-type macrophages was moderately reduced after treatment with eMSCs. Further administration of NO via the intracellular delivery method significantly (P < 0.001 or 0.01) inhibited the polarization of macrophages toward the M1 phenotype while increasing the proportion of M2-type macrophages compared to the control and eMSC groups (Fig. 5, D and E). Additionally, we conducted immunofluorescence staining in heart section (fig. S5). The results revealed that intracellular delivery of NO also induces the polarization of macrophages into the M2 phenotype within the heart.

In vitro studies demonstrated that exogenous NO could improve the proangiogenic capacity of eMSCs. Here, we further explored the influence of the combined delivery of exogenous NO and eMSCs on the reconstruction of the vascular network at the site of infarction. α–Smooth muscle actin (α-SMA)–positive arterioles and von Willebrand factor (vWF)–positive small vessels in MI mice after the different treatments were detected by immunofluorescence staining (Fig. 5, F and G). Treatment with eMSCs efficiently promoted angiogenesis in the injured myocardium, and more prominent enhancement was observed in response to further treatment with intracellular NO. This finding was further supported by the expression‎ of angiogenesis-related genes in the border zone of the infarcted heart (fig. S6).

Intracellular delivery of NO improves heart function and inhibits adverse myocardial remodeling in MI mice after treatment with eMSCs

Ultrasound and histological analyses were performed to eval‎uate the long-term recovery of cardiac function after MI. Cardiac injury was first eval‎uated by 2,3,5-triphenyltetrazolium chloride (TTC) staining (Fig. 6A). Treatment with eMSCs moderately repressed MI compared to the acute myocardial infarction (AMI) group, but a more pronounced inhibitory effect was observed in the group with further intracellular NO delivery. Left ventricular function was assessed by echocardiography at different time points. As shown in Fig. 6B, after 1 day of MI, the left ventricle in each group was markedly enlarged, cardiac function decreased rapidly, and deterioration of heart function continued for 28 days without detectable restoration in the AMI group. However, eMSC treatment could restore left ventricular systolic function and reduce ventricular dilation, as shown by the increase in left ventricular ejection fraction (LV-EF) and fraction shortening (LV-FS), as well as the decrease in left ventricular end-diastolic diameter (LVIDd) and left ventricular end-diastolic volume (LV-EDV) to a certain extent. In the group treated with eMSCs and intracellular NO delivery (NO-eMSCs), LV-EF and LV-FS were effectively recovered, while LVIDd and LV-EDV were significantly (P < 0.001 or 0.0001) enhanced compared to the AMI group.

Fig. 6. Intracellular delivery of nitric oxide (NO) improves heart function and reduces adverse cardiac remodeling in myocardial infarction (MI) mice after treatment with engineered mesenchymal stem cells (eMSCs).

(A) Representative images of 2,3,5-triphenyltetrazolium chloride (TTC) staining and quantification of the infarct area (n = 3). Scale bar, 2 mm. (B) Cardiac echo measurement was performed at different time points after surgery, and cardiac function indicators of left ventricular ejection fraction (LV-EF), left ventricular fractional shortening (LV-FS), left ventricular internal diameter at end diastole (LVIDd), and left ventricular end-diastolic volume (LV-EDV) were eval‎uated accordingly (n = 6). ***P < 0.001, ****P < 0.0001 versus acute myocardial infarction (AMI) group. (C) Masson’s trichrome staining was performed, and the infarct size was quantified accordingly (n = 6). (D) Collagen deposition in the hearts was detected by Sirius Red staining (n = 6). Scale bar, 100 μm. (E) Representative images of wheat germ agglutinin (WGA) immunofluorescence staining and quantification of the cross-sectional area of cardiomyocytes (n = 6). Scale bar, 50 μm. (F) Representative images of immunofluorescence staining (red) for the gap junction protein (Cx43) and the quantification of the intensity of red fluorescence to the whole area of images (n = 6). Scale bar, 25 μm. Data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

Expand for more

Open in viewer

Masson staining was used to detect the degree of myocardial fibrosis after MI. Severe myocardial fibrosis was observed in MI mice compared to the sham group (Fig. 6C). In addition, the ventricular wall became thinner, which was closely related to impaired left ventricular systolic function after MI, as demonstrated by echocardiography. However, these adverse cardiac remodeling events were ameliorated after eMSC treatment and accompanied by a lowered degree of myocardial fibrosis and a thickened ventricular wall in contrast to the AMI group. Notably, treatment with eMSCs and intracellular NO delivery exerted the most prominent inhibitory effect on myocardial fibrosis after MI (Fig. 6C).

Collagen deposition in MI mice was assessed by Sirius staining (Fig. 6D), and the results showed that MI resulted in severe collagen deposition in the injured myocardium compared to the sham group. It was effectively reduced after eMSC treatment, and the inhibitory effect of eMSC plus intracellular NO delivery was significantly higher than that in the other two groups (P < 0.001 or 0.0001).

Next, wheat germ agglutinin (WGA) staining was carried out to eval‎uate myocardial cell hypertrophy 28 days after MI (Fig. 6E). The cross-sectional area of cardiomyocytes was increased in MI mice in contrast to the sham operation group due to compensatory hypertrophy in the heart to maintain the normal rate of cardiac ejection. Hypertrophy was significantly (P < 0.05) mitigated after treatment with eMSCs, especially in the presence of exogenous NO (P < 0.001), indicating an ideal therapeutic effect on inhibiting myocardial cell hypertrophy and adverse ventricular remodeling by the combination of eMSCs and NO.

Gap junctions (GJs) are the main connections between cardiomyocytes in the heart, and Cx43 is the main GJ protein in ventricular muscle in the heart (24). Studies have shown that the absence of Cx43 leads to the occurrence of cardiac ventricular arrhythmia, which can develop into heart failure (25). After 28 days of MI, immunofluorescence staining for Cx43 revealed abundant and uniform distribution of GJ proteins in the sham group, whereas MI injury resulted in a marked decrease in the expression‎ of Cx43 (Fig. 6F). Despite the moderate inhibitory effect provided by the administration of eMSCs, further delivery of NO via the intracellular method significantly enhanced (P < 0.001 or 0.01) the expression‎ of Cx43 compared to that in the AMI or eMSC groups.

Intracellular delivery of NO enhances the therapeutic efficacy of eMSCs in a rat MI model

Although the outcome in a mouse model supported the beneficial effect of NO via intracellular delivery on enhancing the therapeutic efficacy of MSCs for MI, immediate administration of stem cells after MI is different from the clinical treatment of MI due to the limitation of the administration paradigm in mouse models. In addition, 3 to 7 days after MI is the outbreak period of the inflammatory response. For this reason, we established a rat model of MI and conducted secondary thoracotomy 3 days after surgery (Fig. 7A), and eMSCs were delivered via an injectable HA hydrogel as the carrier (Fig. 7B). Lactate dehydrogenase (LDH), a crucial marker for assessing the extent of myocardial damage, exhibited an initial elevation within 2 to 48 hours following the onset of MI, reaching its zenith between 2 and 5 days after MI. We collected blood samples from the orbital venous plexus of rats 5 days after MI to measure serum LDH levels (fig. S7). The findings revealed a sharp increase in serum LDH levels due to MI. However, treatment with NO-eMSCs significantly reduced serum LDH levels, indicating an attenuation of cardiac injury.

Fig. 7. Intracellular delivery of nitric oxide (NO) enhances the therapeutic efficacy of engineered mesenchymal stem cells (eMSCs) in a rat myocardial infarction (MI) model.

(A) Experimental schedule for the treatment of rat MI. (B) Representative images showing the second thoracotomy in rats after MI. (C) Cardiac echo measurement was performed at different time points after surgery, and cardiac function indicators of left ventricular ejection fraction (LV-EF), left ventricular fractional shortening (LV-FS), left ventricular internal diameter at end diastole (LVIDd), and left ventricular end-diastolic volume (LV-EDV) were eval‎uated accordingly (n = 6). **P < 0.01, ***P < 0.001 versus acute myocardial infarction (AMI) group. (D) Representative images of Masson’s trichrome staining and quantification of the infarct size and infarct thickness (n = 6). **P < 0.01, ***P < 0.001, ****P < 0.0001 versus AMI group. (E) Representative images of wheat germ agglutinin (WGA) immunofluorescence staining and quantification of the cross-sectional area of cardiomyocytes (n = 6). Scale bar, 50 μm. (F) Representative images of α–Smooth muscle actin (α-SMA) immunofluorescence staining and quantification of α-SMA+ arterioles (n = 6). Scale bar, 100 μm. Data are expressed as the mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

Open in viewer

Echocardiography was performed to detect heart function at different time points after MI (Fig. 7C). At 4 weeks after surgery, the anterior wall of the left ventricle was completely infarcted, and the ventricular cavity was increased dramatically. Quantitative analysis further indicated that both EF and FS were decreased markedly after surgery. Treatment with MSCs could restore the systolic function of the heart, and the highest recovery rate (60 to 80%) was observed in the group treated with eMSCs and intracellular NO delivery. Ventricular dilatation was also effectively restored after combined treatment with eMSCs and NO, which was consistent with the results obtained in the mouse model.

Histological analyses showed that treatment with eMSCs combined with NO efficiently reduced the degree of fibrosis (Fig. 7D) and collagen deposition (fig. S8A). Furthermore, it alleviated infarct size after MI and restored ventricular cavity morphology of the heart with a significant enhancement in the thickness of the infarcted left ventricular wall and the interventricular septum (IVS), as eval‎uated by Masson (Fig. 7D) and H&E staining (fig. S9). Twenty-eight days after MI, WGA staining also showed that the combination of eMSCs and NO significantly (P < 0.01) inhibited cardiomyocyte hypertrophy in contrast to the AMI group, thus inhibiting further myocardial systolic dysfunction (Fig. 7E). The expression‎ of the GJ protein Cx43 (fig. S8B) followed a similar trend to that in mouse models; that is, acute MI led to a marked decrease in the distribution of Cx43 in the myocardium, and treatment with eMSCs up-regulated the expression‎ of Cx43 to a certain extent. Further delivery of NO via the intracellular method produced a more significant (P < 0.05 or 0.01) effect on the up-regulation of Cx43 in the injured myocardium compared to the eMSC and AMI groups, thereby facilitating the connection between cardiomyocytes and further inhibiting the occurrence of arrhythmias and heart failure.

Tissue repair after MI is often closely related to angiogenesis, which begins at the infarct border and extends to the center of the infarction (26). Therefore, we further compared the proangiogenic effect of eMSCs with and without NO delivery on the damaged margin of the infarcted myocardium 28 days after MI in rats (Fig. 7F and fig. S8C). The combination of eMSCs and NO remarkably promoted angiogenesis and reconstruction of the vascular network compared to the group treated with eMSCs only, which is beneficial to the repair of myocardial injury after MI.

DISCUSSION

Cell therapy based on MSCs has proven to be a powerful solution for treating degenerative diseases and tissue damage (27–29). Despite the advantages of autologous stem cells over allogeneic stem cells, including the absence of immune rejection, the decreased survival and impaired paracrine functions of stem cells collected from patients with chronic diseases greatly limit their clinical use (7, 30, 31). Here, we first showed that MSCs collected from patients with diabetes exhibited marked up-regulation of apoptosis- and inflammation-related genes and down-regulation of proangiogenic genes, affecting the efficacy of cell therapy. As a result, genetic engineering strategies are often required to enhance the therapeutic efficacy of autologous stem cells. A recent study revealed that eMSCs, engineered to express PD-L1 on their surface and secrete CTLA4-Ig (immunoglobulin) as an extracellular factor, exhibited immunoprotective properties, which improved the outcome of both syngeneic and allogeneic islet transplantation in diabetic mice (32).

NO is involved in a variety of physiological processes. Studies have shown that as an important signaling molecule, NO plays a pivotal role in regulating stem cell behavior (33–35), including cell survival, migration, differentiation, and paracrine behavior. These factors affect the interaction of stem cells with other cells and the tissue microenvironment. Previously, different types of NO-releasing biomaterials, such as injectable hydrogels, have been prepared by us and other groups (36–40), and further studies have shown that the combination of NO and MSCs is more effective in treating various diseases than MSC therapy alone. In addition, it has been reported that pretreatment of MSCs with NO-releasing biomaterials could enhance the therapeutic efficacy of MSCs and their secreted exosomes because of their enhanced proangiogenic functions (41).

Due to the spatiotemporal characteristics of NO (42), precise delivery of NO in a site-specific and controllable manner holds great importance in the regulatory effect of exogenously administered NO. In addition to the controlled release rate, the site at which NO is generated is also a key factor due to the relative half-life and limited diffusion distance (43, 44). It is reasonable to speculate that intracellular and extracellular NO delivery may lead to different outcomes when regulating the survival and function of MSCs. In our previous work, an enzyme-prodrug delivery system was designed based on a bump-and-hole strategy (21). The mutant galactosidase (A4-β-GalH363A) enables the targeted delivery of NO, thus reducing the side effects due to the unspecific decomposition of the NO prodrug and enhancing the therapeutic efficacy. Here, we transfected a plasmid expressing mutant galactosidase into MSCs and successfully constructed eMSCs. The enzyme expressed by MSCs could catalyze the decomposition of the 6-OMe-galactose–protected NO prodrug and release NO intracellularly.

Western blotting and fluorescence imaging demonstrated that the expression‎ of the engineered enzyme was confined to the nucleus of MSCs, while wild-type β-galactosidase was widely distributed in the cytoplasm, including the lysosome and perinuclear region (45, 46). Since the corresponding prodrug for wild-type β-galactosidase is highly hydrophilic, it fails to enter MSCs and releases NO extracellularly by enzymes that translocate from the cell. In contrast, the prodrug for mutant galactosidase is cell penetrating because of the modified molecular structure; therefore, it can enter MSCs and release NO intracellularly under the catalysis of the corresponding enzyme expressed by the cells. Accordingly, two different NO delivery paradigms were successfully developed in this study and further confirmed by a series of eval‎uations, including cell imaging and electronic paramagnetic resonance. Further in vitro and in vivo assays indicated that in contrast to extracellular NO delivery, intracellular administration of NO enhanced cell survival and the paracrine effects of MSCs, including inhibiting apoptosis and supporting angiogenesis.

Next, we established a mouse MI model to systematically eval‎uate the therapeutic efficacy of MSCs combined with exogenous NO. The results showed that intracellular delivery of NO prolonged the retention of eMSCs after myocardial orthotopic transplantation. In addition, the combination of eMSCs and intracellular NO delivery improved cardiac function after MI and reduced adverse ventricular remodeling compared to the group treated with MSCs only. Additionally, it could effectively restore the reconstruction of the blood vessel network and further promote the repair of the infarcted myocardium.

To gain further insight into the translational potential of the combinatory therapeutic strategy developed in this study, a rat model of MI was established, and MSCs were administered by a second thoracotomy after 3 days to mimic the clinical use of MSCs for the treatment of MI (47, 48). Clinically, acute MI is typically due to the rupture of coronary atherosclerotic plaque and the formation of thrombus, which causes coronary artery obstruction. After the acute phase of MI, adverse ventricular remodeling further affects the prognosis of patients, which is specifically characterized as a decrease in ventricular wall thickness and myocardial tension in the MI area, myocardial hypertrophy in the noninfarction area, and a change in the morphology of the ventricular cavity, thus leading to arrhythmia and further development into heart failure. The efficacy of MSCs in managing arrhythmias remains a topic of ongoing debate. Some researchers argue that MSCs do not appear to reduce or prevent arrhythmias, with the antiarrhythmic or proarrhythmic potential of MSCs primarily relying on paracrine factors (49). Conversely, other studies suggest that MSCs themselves may play a role in the post-MI recovery process (50). In our study, we observed an evident up-regulation of Cx43 expression‎ after NO-eMSC treatment, which is a potential target associated with antiarrhythmic effects. Further investigation is still required to comprehensively explore the antiarrhythmic potential of NO-eMSCs. In line with the enhanced therapeutic efficacy in the mouse model, intracellular delivery of NO showed enormous advantages in the rat MI model by inhibiting apoptosis and enhancing the paracrine function of MSCs.

In summary, we first showed that survival and paracrine function were reduced in MSCs collected from patients with diabetes, which could greatly affect therapeutic efficacy. Accordingly, eMSCs were successfully constructed, and the mutant β-galactosidase expressed by the cells enabled the intracellular generation of NO via the conversion of an exogenous NO prodrug. In vitro and in vivo assays indicated that intracellular delivery of NO effectively enhanced the survival of transplanted MSCs and promoted the paracrine function of MSCs, which was further confirmed by the enhanced therapeutic efficacy in mouse and rat models of MI compared to the group treated with MSCs only. This synergistic strategy provides an option for the treatment of MI by autologous MSCs in the clinic.

MATERIALS AND METHODSRNA sequencing analysis

RNA sequencing was performed by the BGI (Shenzhen, China). Briefly, RNA from the ADMSCs of healthy people and patients with diabetes was extracted using TRIzol reagent (Yeasen, China). RNA samples were sequenced on the BGISEQ platform. The raw data containing low-quality reads, adaptor sequences, and high levels of N bases were filtered before analysis. Then, the clean reads were mapped to the reference genome using HISAT, and Bowtie2 was used to align the clean reads to the reference genes. The reference genome source is National Center for Biotechnology Information (NCBI), and the reference genome version is GCF_000001405.39_GRCh38.p13. The expression‎ levels of genes were quantified to identify differentially expressed genes by RNA-Seq by expectation maximization (RSEM). The analyses of hierarchical clustering and heatmap were performed using the online Dr. Tom system (biosys.bgi.com) to compare differential gene expression‎ of ADMSCs in healthy people and patients with diabetes. According to the KEGG_pathway annotation classification, the phyper function in R software was used for enrichment analysis, the P value was calculated, and then false discovery rate (FDR) was performed on the P value to obtain a Q value. Generally, a Q value of ≤0.05 was regarded as significant enrichment. GSEA was used to analyze significant differences in gene expression‎ between inflammatory and apoptosis-related pathways. Expression‎ cluster heatmap was used to analyze the expression‎ of genes associated with angiogenesis.

Measurement of NO release

The NO-releasing profile was determined by the Griess kit assay. In brief, 50 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were dissolved in phosphate-buffered saline (PBS) buffer (pH 7.4), and enzymes were added to the solutions at a concentration of 0.005 mg/ml. At each predetermined time interval, 50 ml of solution was transferred into a 96-well plate, and 50 ml of Griess I and 50 ml of Griess II were added thereafter. The azo compound of purple color was formed, and the absorbance was measured at a wavelength of 540 nm using an iMark microplate reader (Bio-Rad, USA).

Cell cultureMesenchymal stem cells

MSCs derived from human umbilical cord were obtained from Health-Biotech, maintained in Dulbecco's modified Eagle’s medium (DMEM)/F12 (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, USA) and 1% penicillin- streptomycin solution (Gibco, USA), and placed in a cell culture chamber containing 5% CO2 at 37°C.

Human embryonic kidney 293T cells

Human embryonic kidney (HEK) 293T cells were obtained from the American Type Culture Collection (ATCC), maintained in high-glucose DMEM (Gibco, USA) with 10% FBS (Gibco, USA) and 1% penicillin-streptomycin solution (Gibco, USA), and placed in a cell culture chamber containing 5% CO2 at 37°C.

Construction of eMSCs

The coding sequence of mutant β-galactosidase (A4-β-GalH363A) can be obtained from the previous publication (21). The lentivirus packaging system containing A4-β-GalH363A sequence and Rluc-RFP sequence was constructed by Wuhan Miaolingbio Co. Ltd. The constructed lentivirus plasmid containing the target gene and the package gene (psPAX2 and pMD2.G) was transfected into HEK 293T cells through Lipo2000, and the supernatant was collected to obtain the virus solution. After removing impurities, the virus solution was mixed 1:1 with fresh MSC medium, and polybrene (10 μg/ml) was added. MSCs were infected with virus through incubation in the mixture medium. The infection efficiency was observed under an inverted fluorescence microscope, and the expression‎ of target protein was determined by Western blotting.

Cell immunofluorescence staining

eMSCs were inoculated in 24-well plates. Cells were fixed with 4% paraformaldehyde and blocked in 4% bovine serum albumin in PBS for 30 min at room temperature. Then, the cells were incubated with primary antibodies overnight at 4°C. The bound primary antibodies were displayed by incubation with the secondary antibodies for 2 hours at room temperature. Cells were counterstained with DAPI (4′,6-diamidino-2-phenylindole)–containing Fluoromount-G and visualized with a fluorescence microscope (Axio Imager Z1). Antibodies used include anti–β-galactosidase (1:100, A1863, Abclonal) and anti-RFP (1:100, PA1-986, Invitrogen).

Western blot

eMSCs were collected, and total protein was extracted using radioimmunoprecipitation assay (RIPA) lysate containing protease inhibitor (Solarbio, China). Cytoplasmic protein and nucleoprotein were extracted using a nucleoprotein extraction kit containing protease inhibitors (Solarbio, China). The protein concentration was quantified using a BCA protein assay kit (Solarbio, China). The samples were diluted with 4× SDS–polyacrylamide gel electrophoresis (PAGE) loading buffer and boiled in boiling water for 8 min. Then, 30 μg of protein was isolated by 10% SDS-PAGE electrophoresis. The isolated proteins were transferred to an Immobilon-P Transfer membrane (Millipore, USA) and incubated with the primary antibody overnight at 4°C and then with the secondary antibody at room temperature for 2 hours. The bands were detected with chemiluminescent horseradish peroxidase substrate (Millipore, USA). Signals were generated by using an enhanced chemiluminescence (ECL) reagent (Millipore, USA) and were captured by using the Tanon-5200 Chemiluminescence Imaging System (Tanon, China). The antibodies used included anti–β-galactosidase (1:1000, A1863, Abclonal), anti–His-tag (1:1000, 12698S, Cell Signaling Technology), anti–glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (1:1000, AC001, Abclonal), anti–histone H3 (1:1000, 4499S, Cell Signaling Technology), anti–β-actin (1:1000, UM4001, Utibody), anti-BCL2 (1:1000, WL01556, Wanleibio), anti-Bax (1:1000, 50599-2-lg, Proteintech), anti-Bad (1:1000, WL02140, Wanleibio), and anti-caspase/cleaved caspase3 (1:1000, WL02117, Wanleibio).

6-OMeGal-Ph-NO uptake by eMSCs

eMSCs were inoculated in 12-well plates, and 50 μM substrate (6-OMeGal-Ph-NO) was added per well. At predetermined time points (0, 1, 3, 6, and 12 hours), an appropriate amount of culture medium was collected, and an excess of A4-β-GalH363A was added immediately to fully catalyze the decomposition of the remaining substrate at room temperature. The amount of 6-OMeGal-Ph-NO substrate in the culture medium was determined by Griess kit assay.

Real-time imaging of intracellular NO

The fluorescence emission associated with NO in the cytosol was detected by an electron-multiplying charge-coupled device (DU-897D-CS0-BV; Andor, Belfast, UK) connected to an inverted fluorescence microscope (Axio Observer D1; Carl Zeiss, Oberkochen, Germany). Intracellular NO imaging was performed using an NO fluorescence probe, DAF-AM DA (Beyotime, China), according to the manufacturer’s instruction. eMSCs were inoculated in small confocal dishes in advance, and the experiment was conducted when the cell density reached 80%. First, the medium was collected for later use. After two gentle washes with PBS, 5 μM DAF-AM DA solution was incubated at 37°C for 30 min in the dark. Then, cells were gently washed with PBS twice. The previously collected medium was added anew, and cell imaging was performed immediately. At the 488-nm excitation wavelength, pictures were taken every 5 s. After stable shooting for 2 min, 100 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were added. The filming for cell fluorescence was continued for 4 min. The final fluorescence intensity was determined without the background fluorescence value. The proportion of change in fluorescence intensity of each cell in the visual field was calculated.

Intracellular NO detection

Intracellular NO radicals (NO•) were detected using EPR as described (51, 52). In brief, sodium DETC (4.5 mg) and FeSO4•7H2O were dissolved in two separate volumes (10 μl) of deoxygenated Krebs/Hepes solution. Equal volumes of these parent solutions were rapidly mixed and aspirated into Eppendorf combi tips. The 0.5 mM Fe•(DETC)2 colloid solution had a yellow-brownish color with a slight opalescence in light. No aggregate formation was observed, at least during the first 30 min. eMSCs were rinsed with modified Krebs/Hepes buffer and incubated with freshly prepared NO•-specific spin trap Fe•(DETC)2 colloid (0.5 mM) for 30 min. Meanwhile, 100 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were added to the buffer. Gently collected cell suspensions were snap-frozen in liquid nitrogen. Ethyl acetate (200 μl) was added, and the cells were ultrasonically broken to extract DETC2-Fe-NO. The ethyl acetate extract was concentrated with nitrogen and transferred to a 50-μl capillary, and then the X-band EPR was measured at room temperature. The following acquisition parameters were used: modulation frequency, 100 kHz; microwave power, 10 mW; modulation amplitude, 2 G; number of scans, 60. The double-integrated area of the EPR spectra was calibrated into concentrations of DETC2-Fe-NO using TEMPO (2,2,5,5-tetramethyl piperidine 1-oxyl) as a standard. EPR spectral simulation was conducted by the WINSIM program.

Extracellular NO detection

eMSCs were treated with β-Gal-NO or 6-OMeGal-Ph-NO (30 μM). The production of NO in the medium of each group was detected 6 hours after incubation with NO-sensitive near-infrared fluorescence probe (5 μM). The NO production of medium in different groups was compared by the relative fluorescence intensity under the excitation at 750 nm (emission at 800 nm).

Cell apoptosis detection

To test the protective effect of NO delivery on cellular oxidative stress stimulation, 30 μM NO substrate (6-OMeGal-Ph-NO) was added to the medium in advance. Then, H2O2 with different concentrations (100, 200, 400, and 600 μM) was added to stimulate the lentivirus-infected eMSCs. BLI was performed immediately after addition of the luciferase substrate coelenterin to eval‎uate cell apoptosis. Additionally, eMSCs treated with 200 mM H2O2 were stimulated for 24 hours to induce cell apoptosis. An Annexin V/PI assay kit (Solarbio) was used to detect eMSC apoptosis.

BLI detection of cell retention

BLI and luciferase substrates were used in mice to eval‎uate the retention of NO-eMSCsGluc/RFP in cardiac orthotopic transplantation. The mice after eMSC injections were anesthetized with 1.5% isoflurane and injected with coelenterin through the caudal vein at 150 mg/kg. After injection, the mice were immediately placed in a BLI system to detect cell retention in the myocardium.

Animals

C57BL/6 mice (male, 8 weeks old) and Sprague-Dawley rats (male, 8 weeks old) were purchased from SPF (Beijing) Biotechnology Co. Ltd., Beijing, China. Animals were randomly grouped for treated and untreated controls. All experiments and animal procedures were approved by the Animal Experiments Ethical Committee of Nankai University and carried out in conformity with the Guide for Care and Use of Laboratory Animals.

MI in mice and rats

Surgical induction of MI was performed on C57BL/6 mice (male, 8 weeks old) as previously described with some modifications. Briefly, mice were anesthetized with 2% isoflurane, followed by fixation to a heating pad (37°C) at supine position, and then ventilated with a mechanical ventilation system (Hallowell EMC Microvent I, USA) set at 110 breaths per minute with a tidal volume of 0.2 to 0.3 ml. The third intercostal space over the left chest and the heart was exposed. After left thoracotomy, the left anterior descending coronary artery was ligated with a 6-0 silk ligature. The left main descending coronary artery was sutured and tied with a slip knot at a site approximately 3 mm from its origin. Cardiac ischemia was confirmed by the presence of myocardial blanching. Immediately, eMSCs encapsulated with HA hydrogel were injected into the myocardium of mice through three-point injection around the infarct zone. The chest cavity was closed to restore negative pressure and prevent pneumothorax.

Similar MI surgery was performed on Sprague-Dawley rats (male, 8 weeks old) first. Briefly, rats were anesthetized via intraperitoneal injection of 10% chloral hydrate (350 mg/kg), followed by fixation to a heating pad (37°C) at supine position. Then, they were ventilated with a mechanical ventilation system (Hallowell EMC Microvent I, USA) set at 110 breaths per minute with a tidal volume of 6 ml. The third intercostal space over the left chest and the heart was exposed. After left thoracotomy, the left anterior descending coronary artery was ligated with a 6-0 silk ligature. The left main descending coronary artery was sutured and tied with a slip knot at a site approximately 3 mm from its origin. Cardiac ischemia was confirmed by the presence of myocardial blanching. The chest cavity was closed to restore negative pressure and prevent pneumothorax. Three days after surgery, secondary thoracotomy was performed, and eMSCs were injected into the myocardium around the infarct zone. The chest cavity was closed to restore negative pressure and prevent pneumothorax.

Mice and rats in the AMI group only received MI without eMSC injection, while sham-operated mice only experienced thoracotomy without MI surgery or eMSC injection.

At 1, 3, 5, and 7 days after myocardial injection of eMSCs, the prodrug was injected through the tail vein. Mice were injected with 100 μl of prodrug (1 mg/ml) each time, and rats were injected with 200 μl of prodrug (1 mg/ml) each time.

TTC staining

Two days after surgery, a thoracotomy was performed. The heart was quickly excised after quick freezing for 15 min and sliced at 1 mm thickness. Afterward, the sections were incubated with 1.5% TTC (Sigma-Aldrich) solution at 37°C in an incubator for 15 to 30 min and then with a 4% formaldehyde solution for 2 hours. The normal myocardial tissue was red, while the ischemic myocardium was white. The size of the infarcted myocardium, which was white or pale, was measured by ImageJ software.

Cardiac function assessment

Transthoracic echocardiography was performed with the Vevo 2100 Imaging System (Fuji Film Visual Sonics Inc., Canada) equipped with an MS-250/400 imaging transducer. The baseline cardiac function of mice and rats was measured at 3 days before surgery. Cardiac function was analyzed at days 1 and 28 after MI surgery with different treatments, as reported previously. Mice or rats were slightly anesthetized in a box with isoflurane. Their limbs were fixed in a supine position on the echo mat, and the chest hair was removed by depilating cream. Then, mice or rats were anesthetized by inhalation of isoflurane (0.5 to 1%) mixed with oxygen to maintain the heart rate at approximately 500 to 600, and M-mode echocardiography was performed. The left ventricular internal diameter at end-diastole (LVIDd) and systole (LVIDs) were obtained by measuring the long axis and the short axis. Accordingly, the cardiac parameters LV-EF, LV-FS, LV-EDV, and LV end-systole volume (LV-ESV) were determined. The echocardiography measurement was carried out in a double-blind manner.

Histological analysis

At the indicated time points, mice and rats were anesthetized via intraperitoneal injection of chloral hydrate, and a thoracotomy was performed. The hearts were fixed with trans-cardiac perfusion of saline and immersed in 4% paraformaldehyde over 24 hours. The heart tissue samples were dehydrated with gradient alcohol and xylene, embedded in paraffin blocks, and cut into sections in 5 μm thickness.

The paraffin-embedded sections were stained with Masson trichrome, H&E, and Sirius Red following a standard protocol. Immunofluorescence staining was performed on paraffin-embedded sections of the heart tissue samples. After deparaffinization and heat-mediated antigen retrieval‎ in citrate solution, the samples were washed with PBS three times and incubated with blocking serum, which was used to avoid nonspecific binding, at room temperature for 30 min. The sections were incubated with specific antibodies diluted in goat serum at 4°C overnight. On the second day, the sections were rewarmed at room temperature for 1 hour and washed with PBS three times. Afterward, the sections were incubated with Alexa Fluor–coupled secondary antibodies for 2 hours at room temperature. After washing with PBS, the sections were counterstained with DAPI-containing Fluoromount-G (SouthernBiotech, USA) and coverslipped. The antibodies used included anti–α-SMA (1:100, ab5694, Abcam), anti-vWF (1:100, ab6694, Abcam), anti–α-actinin (1:100, ab9475, Abcam), anti-Connexin43 (1:100, ab11370, Abcam), anti-CD68 (1:100, ab125212, Abcam), WGA (1:500, FL-1021, Novus Biologicals), anti-iNOS (1:100, ab178945, Abcam), and anti-CD206 (1:100, ab64693, Abcam).

Macrophage isolation and detection

Three days before euthanasia, mice were intraperitoneally injected with 2 ml of 4% thioglycolate. Three days later, the mice were sacrificed by cervical dislocation and immersed in 75% alcohol and then transferred to an ultraclean workbench. The mouse limb was fixed in the supine position, and the mouse abdominal wall was carefully cut open with the peritoneal. PBS [1% penicillin-streptomycin (PS)] was injected intraperitoneally to collect the cell suspension, which was centrifuged at 2000 rpm for 10 min. After discarding the supernatant, the cells were incubated with anti-F4/80/TNF-α and anti-F4/80/CD206 antibodies. FlowJo software was used to analyze the results of flow cytometry.

Quantitative real-time PCR

Total RNA samples from the cells were prepared using TRIeasy Total RNA Extraction Reagent (Yeasen, China) according to the manufacturer’s instructions. Heart tissue samples were collected at the indicated time points after MI surgery.

The tissue samples were dissected at the border zone of the left ventricle and frozen in liquid nitrogen immediately. Afterward, the total RNA was extracted with TRIzol reagent, as mentioned before. The concentration of the RNA was measured with a NanoDrop spectrophotometer (NanoDrop Technologies, USA). The complementary cDNA was synthesized using a first-strand cDNA synthesis kit (Yeasen, China). Quantitative real-time PCR was performed on a CFX96 real-time PCR system (Bio-Rad, Hercules, USA) with an SYBR Green–based real-time detection system (Yeasen, China). The relative gene expression‎ of mRNA was expressed as 2−(△△CT) and normalized to GAPDH as an internal control. Each reaction was performed in triplicate to obtain an average value, and the changes in relative gene expression‎ normalized to the internal control levels were determined. The highly purified primers used in this experiment were commercially synthesized (Sango, China). The sequences of the primers used in this experiment are summarized in the Supplementary Materials.

Statistics

All data are presented as the mean ± SEM from at least three independent experiments. Comparisons between two groups were performed by Student’s t test, and comparisons among more than two groups were performed by one-way or two-way analysis of variance (ANOVA). Statistical analyses were performed with GraphPad Prism software 7.0, and a statistical significance level of less than 0.05 was accepted.

Acknowledgments

Funding: This study is supported by the National Key R&D Program of China (2018YFE0200503), the National Natural Science Foundation of China (nos. 81925021, 82330066, 81921004, and U2004126), and the Tianjin Natural Science Foundation (21JCZDJC00240).

Author contributions: Q.Z. and Z.L. conceived the original concept and initiated this project. Q.Z., Z.L., and F.G. designed the experiment and supervised the entire project. S.W. collected human adipose mesenchymal stem cells. M.Q. synthesized all NO prodrugs and probes. P.L. prepared engineered enzymes. T.H., G.J., and Q.X.L. established mouse and rat MI models. T.H. and G.J. performed histological analysis. G.J. and W.D. carried out in vitro cell experiments. S.D. carried out NO cell imaging under the supervision of L.P. T.H., G.J., and M.Q. analyzed data under the supervision of Q.Z. H.H. helped with lentivirus packaging and cell infection. W.G. and T.L. helped in establishing animal MI models. Y.W., J.H., J.C., and J.T. helped with data collection. T.H. and Q.Z. wrote the paper with input from other authors.

Competing interests: The authors declare that they have no competing interests.

Data and materials availability: All data needed to eval‎uate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Transcriptome sequencing dataset is available at https://doi.org/10.5061/dryad.tqjq2bw5b.

Supplementary MaterialsThis PDF file includes:

Supplementary Text

Figs. S1 to S9

Table S1

• Download
• 1.94 MB

REFERENCES AND NOTES

1

M. F. Pittenger, A. M. Mackay, S. C. Beck, R. K. Jaiswal, R. Douglas, J. D. Mosca, M. A. Moorman, D. W. Simonetti, S. Craig, D. R. Marshak, Multilineage potential of adult human mesenchymal stem cells. Science 284, 143–147 (1999).

GO TO REFERENCE

Crossref

PubMed

Web of Science

Google Scholar

2

A. Uccelli, L. Moretta, V. Pistoia, Mesenchymal stem cells in health and disease. Nat. Rev. Immunol. 8, 726–736 (2008).

GO TO REFERENCE

Crossref

PubMed

Web of Science

Google Scholar

3

Y. Wang, J. Fang, B. Liu, C. Shao, Y. Shi, Reciprocal regulation of mesenchymal stem cells and immune responses. Cell Stem Cell 29, 1515–1530 (2022).

GO TO REFERENCE

Crossref

PubMed

Web of Science

Google Scholar

4

Y. Wang, X. Chen, W. Cao, Y. Shi, Plasticity of mesenchymal stem cells in immunomodulation: Pathological and therapeutic implications. Nat. Immunol. 15, 1009–1016 (2014).

GO TO REFERENCE

Crossref

PubMed

Web of Science

Google Scholar

Show all references

(0)eLetters

eLetters is a forum for ongoing peer review. eLetters are not edited, proofread, or indexed, but they are screened. eLetters should provide substantive and scholarly commentary on the article. Neither embedded figures nor equations with special characters can be submitted, and we discourage the use of figures and equations within eLetters in general. If a figure or equation is essential, please include within the text of the eLetter a link to the figure, equation, or full text with special characters at a public repository with versioning, such as Zenodo. Please read our Terms of Service before submitting an eLetter.

Log In to Submit a Response

No eLetters have been published for this article yet.

Recommended articles from TrendMD

• In vivo priming of human mesenchymal stem cells with hepatocyte growth factor–engineered mesenchymal stem cells promotes therapeutic potential for cardiac repai...

  Bong-Woo Park, Sci Adv, 2020

• Shattering barriers toward clinically meaningful MSC therapies

  Oren Levy, Sci Adv, 2020

• Nanovesicles derived from iron oxide nanoparticles–incorporated mesenchymal stem cells for cardiac repair

  Ju-Ro Lee, Sci Adv, 2020

• Exendin-4 gene modification and microscaffold encapsulation promote self-persistence and antidiabetic activity of MSCs

  Yuanyuan Zhang, Sci Adv, 2021

• Human gingival mesenchymal stem cells retain their growth and immunomodulatory characteristics independent of donor age

  Jay R. Dave, Sci Adv, 2022

• The secretion profile of mesenchymal stem cells and potential applications in treating human diseases

  Yuyi Han, Signal Transduction and Targeted Therapy, 2022

• Mesenchymal Stem Cell Therapy for Ischemic Heart Disease: Systematic Review and Meta-analysis

  Hyunsuk Jeong, International Journal of Stem Cells

• Mesenchymal Stem Cell and MicroRNA Therapy of Musculoskeletal Diseases

  International Journal of Stem Cells

• Excessive branched-chain amino acid accumulation restricts mesenchymal stem cell-based therapy efficacy in myocardial infarction

  Fuyang Zhang, Signal Transduction and Targeted Therapy, 2022

  Intracellular delivery of nitric oxide enhances the therapeutic efficacy of mesenchymal stem cells for myocardial infarction

  Tian Hao https://orcid.org/0009-0007-8657-0424, Guangbo Ji https://orcid.org/0009-0000-7079-8896, Meng Qian https://orcid.org/0009-0006-9135-4593, Qiu Xuan Li, Haoyan Huang, Shiyu Deng, Pei Liu, Weiliang Deng https://orcid.org/0009-0001-8402-7280, Yongzhen Wei, [...], and Qiang Zhao https://orcid.org/0000-0003-4656-6002 +9 authorsAuthors Info & Affiliations

  Science Advances

  29 Nov 2023

  Vol 9, Issue 48

  DOI: 10.1126/sciadv.adi9967

  4,3612

  Metrics

  Total Downloads4,361

  • Last 6 Months1,270
  • Last 12 Months3,995

  Total Citations2

  • Last 6 Months1
  • Last 12 Months2

  NotificationsBookmark

  • • Abstract
    • INTRODUCTION
    • RESULTS
    • DISCUSSION
    • MATERIALS AND METHODS
    • Acknowledgments
    • Supplementary Materials
    • REFERENCES AND NOTES
    • eLetters (0)
  • • Information & Authors
    • Metrics & Citations
    • View Options
    • References
    • Media
    • Tables
    • Share

  Abstract

  Cell therapy by autologous mesenchymal stem cells (MSCs) is a clinically acceptable strategy for treating various diseases. Unfortunately, the therapeutic efficacy is largely affected by the low quality of MSCs collected from patients. Here, we showed that the gene expression‎ of MSCs from patients with diabetes was differentially regulated compared to that of MSCs from healthy controls. Then, MSCs were genetically engineered to catalyze an NO prodrug to release NO intracellularly. Compared to extracellular NO conversion, intracellular NO delivery effectively prolonged survival and enhanced the paracrine function of MSCs, as demonstrated by in vitro and in vivo assays. The enhanced therapeutic efficacy of engineered MSCs combined with intracellular NO delivery was further confirmed in mouse and rat models of myocardial infarction, and a clinically relevant cell administration paradigm through secondary thoracotomy has been attempted.

  SIGN UP FOR THE AWARD-WINNING SCIENCEADVISER NEWSLETTER

  The latest news, commentary, and research, free to your inbox daily

  Sign up

  INTRODUCTION

  Mesenchymal stem cells (MSCs) are pluripotent stem cells with high self-renewal abilities and multidirectional differentiation potential (1, 2). They are extensively distributed throughout the body and serve a variety of purposes, including tissue regeneration (3), immunoregulation (4), and angiogenesis (5). However, an important challenge for stem cell therapy is the low survival rate of stem cells after transplantation, which is associated with nonspecific homing of cells and ischemia/hypoxia at the injury site (6). In addition, transplanted stem cells cannot fully exert their paracrine effects in pathological environments, which seriously limits the clinical application of stem cells.

  The proangiogenic function of MSCs is a key factor contributing to the treatment of ischemic diseases. Generally, MSCs can stimulate local angiogenesis in ischemic tissue by secreting cytokines, including vascular endothelial growth factor (VEGF), transforming growth factor β (TGFβ), and basic fibroblast growth factor (bFGF), to induce endothelial cells to form tubular structures (7). In addition, MSCs can recruit pericytes and smooth muscle cells to promote neovascularization (8, 9). Recently, MSCs have been reported to participate in the construction of blood vessels by directly differentiating into endothelial or smooth muscle cells (10).

  As an important signaling molecule, nitric oxide (NO) plays a crucial role in the maintenance of vascular homeostasis by inhibiting thrombus formation and promoting angiogenesis (11, 12). Recently, increasing attention has been given to the regulation of stem cells by NO due to its multiple biological functions (13, 14). NO can regulate the proliferation of stem cells by regulating the activities of cyclin and mitotic receptors, as well as their downstream pathways (15, 16). On the other hand, NO can regulate the expression‎ of angiogenic cytokines and immunomodulatory factors to improve the paracrine performance of stem cells (17). Additionally, recent studies have demonstrated that NO can regulate the differentiation behavior of stem cells through the phosphatidylinositol 3-kinase (PI3K)/AKT, guanosine 3′,5′-monophosphate (cGMP), and other signaling pathways (18).

  As a result, NO-releasing biomaterials have been used as delivery carriers for stem cells to enhance their survival and regulate paracrine functions (19). However, NO, which is a gaseous molecule, easily diffuses and has a high level of instability. Furthermore, the physiologic function of NO is dose dependent, and an overdose of NO often leads to notable cytotoxicity (20). Thus, optimizing the beneficial effects of NO to strengthen the therapeutic efficacy of stem cells by tuning their release profile should be taken into account.

  In our previous study, based on the chemical biology principle of “bump-and-hole,” we designed and prepared an enzyme-prodrug delivery system and achieved targeted delivery of NO at the lesion site in two different ischemic disease models (21). Here, MSCs were further modified by gene transfection to express a catalytic enzyme (A4-β-GalH363A). The engineered MSCs (eMSCs) were transplanted using an injectable hyaluronic acid (HA) hydrogel as the carrier, while the NO prodrug was injected through the tail vein to achieve controlled release of NO catalyzed by the enzymes in eMSCs. The therapeutic efficacy of MSCs combined with exogenous NO delivery was eval‎uated in mouse and rat models of myocardial infarction (MI) with an emphasis on comparing the therapeutic efficacy of two different NO administration methods (intracellular or extracellular), and the underlying mechanism of their myocardial protective effect was further explored.

  RESULTSThe gene expression‎ of MSCs collected from patients is differentially regulated

  In the clinic, the therapeutic efficacy of autologous stem cell transplantation is largely affected by the low quality (including cell survival and paracrine function) of stem cells collected from patients due to chronic diseases. Diabetes mellitus is a chronic medical condition that can lead to a variety of complications, and these complications can affect various parts of the body, such as the kidney, lower limb, and heart. Diabetes predisposes affected individuals to a spectrum of cardiovascular complications, and one of the most debilitating in terms of prognosis is heart failure (22).

  Accordingly, although autologous MSCs have been widely accepted as a promising strategy for treating various complications associated with diabetes, the therapeutic efficacy is largely affected by the quality of stem cells collected from the patients themselves. The gene levels in adipose-derived MSCs (ADMSCs) from patients with diabetes and healthy individuals were first compared by transcriptome sequencing. The heatmap shows that multiple genes in MSCs from patients with diabetes were up- or down-regulated compared to healthy controls (Fig. 1A). Gene ontology (GO) enrichment analysis revealed that the most differentially up-regulated genes were related to tumor necrosis factor (TNF) signaling pathways (Fig. 1B). Subsequently, we performed gene set enrichment analysis (GSEA) based on the RNA-sequencing results. GSEA revealed that inflammatory target genes were highly enriched in MSCs from patients with diabetes, with a normalized enrichment score (NES) of 1.65 (P < 0.01) (Fig. 1C). Apoptosis-related genes were also highly enriched, with an NES of 1.77 (P < 0.01) (Fig. 1D). In addition, through enrichment analysis, we found that proangiogenic genes in MSCs were greatly down-regulated in patients with diabetes (Fig. 1E).

  

  Fig. 1. Transcriptome sequencing of adipose mesenchymal stem cells (MSCs) collected from patients with diabetes and healthy controls.

  (A) Heatmap showing the differentially expressed genes of MSCs from patients with diabetes and healthy controls (n = 3). (B) Gene ontology (GO) analysis of the up-regulated transcriptome of MSCs from patients with diabetes. (C and D) Gene set enrichment analysis (GSEA) was performed to determine the enrichment of inflammation (C) and apoptosis (D) target genes in the diabetic group. (E) Heatmap showing angiogenesis-related genes in the two groups (n = 3). (F) Schematic illustration demonstrating the difference in MSCs between patients with diabetes and healthy controls at the gene level.

  Open in viewer

  eMSCs are constructed to express mutant β-galactosidase

  Here, we first designed and constructed plasmids carrying the mutant β-galactosidase (A4-β-GalH363A) target gene and luciferase–red fluorescent protein (RFP) dual reporter genes, which could be further used for in vivo imaging. eMSCs expressing A4-β-GalH363A were constructed by infecting MSCs with lentiviruses obtained from human embryonic kidney 293T cells (Fig. 2A). Immunofluorescence staining for RFP confirmed that A4-β-GalH363A was successfully expressed by eMSCs (Fig. 2B). The subcellular fraction and intracellular distribution of enzymes expressed by the eMSCs was determined by Western blotting (Fig. 2C). In contrast to natural β-galactosidases, which are widely distributed within cells, A4-β-GalH363A was mainly confined to the nucleus of eMSCs.

  

  Fig. 2. Intracellular expression‎ and localization of A4-β-GalH363A.

  (A) Schematic diagram of lentivirus packaging and mesenchymal stem cell (MSC) infection. (B) Immunofluorescence staining of β-Gal and A4-β-GalH363A in eMSCs. Scale bar, 50 μm. (C) The distribution of two different enzymes in engineered MSCs (eMSCs) was analyzed by Western blotting.

  Open in viewer

  Intracellular release of NO is realized via decomposition of the 6-OMeGal-Ph-NO prodrug

  Since the mutant β-galactosidase was confined to the nucleus in eMSCs, we designed a prodrug by introducing a lipid-soluble self-decomposition chain into 6-OMeGal-NO to improve its oil and water distribution coefficient; therefore, the resultant NO donor 6-OMeGal-Ph-NO could penetrate the cell membrane and decompose and release NO under the catalysis of A4-β-GalH363A (Fig. 3A and Supplementary Materials). An in vitro release assay showed that the 6-OMeGal-Ph-NO prodrug was efficiently recognized and converted by A4-β-GalH363A with a cumulative release ratio of approximately 97.3%, while nearly no release was observed in the presence of wild-type β-galactosidase (Fig. 3B). To confirm‎ intracellular NO release, eMSCs were preincubated with an NO fluorescent probe (DAF-AM DA) and then treated with different NO prodrugs. The changes in fluorescence signals were examined by time-delay cell imaging (Fig. 3C). The results indicated that the fluorescence intensity continuously increased in the group that was treated with 6-OMeGal-Ph-NO, indicating conversion into NO (Fig. 3D). In contrast, no detectable changes were identified in the β-Gal-NO group because the high water solubility restricted its entry into eMSCs. Next, intracellular and extracellular release of NO in eMSCs was assessed (Fig. 3E). The quantity of intracellular NO was measured by electron paramagnetic resonance (EPR) using ferrous N-diethyl dithiocarbamate (DETC2-Fe) as the spin-trapping reagent. The resultant NO adduct (DETC2-Fe-NO) exhibited a characteristic triplet EPR signal (aN = 13.06 G, giso = 2.041) at room temperature. Quantitative analysis showed that the NO level was significantly (P < 0.001 or 0.0001) higher in eMSCs treated with 6-OMeGal-Ph-NO than in the β-Gal-NO and control groups (Fig. 3F). Furthermore, in the group treated with β-Gal-NO, the release of NO was mainly catalyzed by β-galactosidase that translocated from the cytoplasm in eMSCs, and the extracellular release profile of β-Gal-NO was confirmed by detecting the NO level in the cell culture medium with the NO-sensitive near-infrared fluorescence probe (23); it was significantly (P < 0.01) higher than that in the 6-OMeGal-Ph-NO and control groups (Fig. 3G). To determine the uptake of the NO prodrug by eMSCs, we incubated 6-OMeGal-Ph-NO with eMSCs, and the concentration in the culture medium was determined at different time points. The results reflected that approximately 45% of 6-OMeGal-Ph-NO was incorporated into the eMSCs within 12 hours (fig. S1).

  

  Fig. 3. Intracellular generation of nitric oxide (NO) from the NO prodrug under the catalysis of A4-β-GalH363A expressed by engineered mesenchymal stem cells (eMSCs).

  (A) Synthesis of two NO prodrugs with different enzyme response abilities and cellular permeabilities. (B) In vitro release profile of NO from the NO prodrug (6-OMeGal-Ph-NO) in the presence of β-Gal or A4-β-GalH363A. (C) Schematic illustration of intracellular NO imaging by using an NO fluorescence probe (DAF-AM DA). (D) Representative time-lapse images of NO generation from two different prodrugs in eMSCs and quantification of the fluorescence intensity (n = 6). ***P < 0.001, ****P < 0.0001 versus 6-OMeGal-Ph-NO group. (E) Schematic illustration showing the detection of intracellular and extracellular NO generation differentially. (F) Representative electron paramagnetic resonance (EPR) spectra and quantification of intracellular NO generation by measuring the DETC2-Fe-NO complex using 2,2,5,5-tetramethyl piperidine 1-oxyl (TEMPO) as a standard (n = 3). (G) Relative quantification of NO production in the medium determined using the near-infrared fluorescence probe (n = 4). Data are expressed as the mean ± SEM. Significant differences were detected by one-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

  Expand for more

  Open in viewer

  Intracellular delivery of NO inhibits apoptosis and enhances the proangiogenic activity of eMSCs

  Overproduction of reactive oxygen species due to cellular oxidative stress has been accepted as an important factor contributing to apoptosis in transplanted cells in ischemic tissue. Therefore, we assessed the protective effect of exogenously administered NO on the survival of eMSCs with H2O2-induced oxidative stress and focused on comparing the protection provided by intracellular and extracellular NO administration. The results showed that H2O2 stimulated apoptosis, and delivery of NO via extracellular and intracellular strategies significantly (P < 0.01 or 0.001) reduced apoptosis in eMSCs stimulated by oxidative stress, and the highest fluorescence signal was observed in response to intracellular NO delivery (Fig. 4A).

  

  Fig. 4. Intracellular delivery of nitric oxide (NO) inhibits apoptosis of engineered mesenchymal stem cells (eMSCs).

  (A) Bioluminescence imaging (BLI) was used to detect the effect of NO delivery on cell apoptosis stimulated by different concentrations of H2O2, and the fluorescence signals were further quantified (n = 3). **P < 0.01, ***P < 0.001, ****P < 0.0001 versus control group; #P < 0.05 versus β-Gal-NO group. (B) Flow cytometry assay of cell viability and apoptosis of eMSCs after H2O2 stimulation, and quantification of mean percent values of apoptotic cells (n = 3). (C) The expression‎ of apoptosis-related protein (BCL2, Bax, Bad, caspase3, and cleaved caspase3) by eMSCs was detected after H2O2 stimulation by Western blots (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001. (D) Representative BLI photographs reflecting the retention of eMSCs with and without intracellular NO delivery after in vivo transplantation as well as the quantitative analysis of signals (n = 3). Data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01.

  Open in viewer

  The survival of eMSCs after treatment with various NO delivery strategies was compared by using flow cytometry. Intracellular NO delivery had the most remarkable antiapoptotic effect on cells stimulated by oxidative stress (P < 0.001) (Fig. 4B). The expression‎ of apoptosis-related genes by eMSCs followed a similar trend; NO delivery (extracellular and intracellular) effectively promoted the expression‎ of the antiapoptotic gene BCL2 in eMSCs after oxidative stress stimulation, while the expression‎ levels of proapoptotic genes were reduced accordingly. Intracellular NO delivery via administration of the 6-OMeGal-Ph-NO prodrug exerted a more pronounced antiapoptotic effect at both the gene and protein levels (Fig. 4C and fig. S2A), which may be because the intracellular generation of NO directly activated the antioxidant system in cells to resist oxidative stress damage and inhibit further apoptosis.

  The expression‎ of proangiogenic genes, including ANGPT1, ANGPT2, FGF2, VEGFA, and KDR, in eMSCs was further detected by reverse transcription polymerase chain reaction (RT-PCR). The results showed that the expression‎ level of proangiogenic genes was significantly (P < 0.05, 0.01, or 0.001) higher in eMSCs treated with the 6-OMeGal-Ph-NO prodrug than in the other groups, indicating the enhanced proangiogenic functions of eMSCs after intracellular NO delivery (fig. S2B).

  We further eval‎uated the effect of NO delivery on the in vivo retention of eMSCs after orthotopic transplantation in the myocardial tissue of mice. As shown in Fig. 4D, intracellular delivery of NO effectively prolonged the retention of eMSCs within the myocardium, and an evident bioluminescence imaging (BLI) signal corresponding to the retention of eMSCs was observed 7 days after transplantation compared to the counterpart without administration of the NO prodrug. To further eval‎uate the translational potential of eMSCs in clinical settings, we used MSCs derived from diabetic patients and conducted a series of assays related to cell survival and paracrine function. The findings indicated that intracellular delivery of NO also confers advantages in the attenuation of cell apoptosis under stress conditions, thereby prolonging the in vivo retention of eMSCs (fig. S3).

  Intracellular delivery of NO ameliorates myocardial injury in MI mice after treatment with eMSCs

  The therapeutic efficacy of MSCs combined with exogenous NO was further eval‎uated in a mouse MI model (Fig. 5A and fig. S4). The inflammatory response in the early stage (3 days) was first detected by hematoxylin-eosin (H&E) staining and CD68 immunofluorescence staining. The results demonstrated that severe inflammatory cell infiltration occurred in the injured myocardium of MI mice, and it was effectively alleviated after eMSC treatment. More prominent restoration in the injured myocardium was observed after further administration of NO (Fig. 5, B and C), confirm‎ing the inhibitory effect on inflammation after MI provided by the combination of MSCs and NO. Moreover, this inhibitory effect was more significant (P < 0.01 or 0.0001) in the group with intracellular NO delivery than in the group with extracellular NO delivery.

  

  Fig. 5. Intracellular delivery of nitric oxide (NO) ameliorates myocardial injury in myocardial infarction (MI) mice after treatment with engineered mesenchymal stem cells (eMSCs) via regulation of the inflammation and stimulation of angiogenesis.

  (A) Experimental schedule for the treatment of MI in a mouse model. (B) Hematoxylin-eosin (H&E) staining was performed to detect inflammatory cell infiltration in the early stage of MI (n = 6). Scale bar, 100 μm. (C) Representative images of CD68 immunofluorescence staining (green) and quantification of CD68+ macrophages in injured myocardium (n = 6). Scale bar, 25 μm. (D and E) Flow cytometry was performed to detect peritoneal macrophage polarization 7 days after surgery followed by different treatments. TNFα- and CD206-positive ascites macrophages (markers of M1 and M2 macrophage phenotypes, respectively) were quantified accordingly (n = 3). (F) Representative images of α–Smooth muscle actin (α-SMA) immunofluorescence staining and the quantification of α-SMA+ arterioles (n = 6). Scale bar, 100 μm. (G) Representative images of von Willebrand factor (vWF) immunofluorescence staining and the quantification of vWF+ capillaries (n = 6). Scale bar, 100 μm. Data are expressed as the mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

  Expand for more

  Open in viewer

  It has been widely accepted that a higher proportion of M2-type macrophages is more conducive to the repair of tissue damage. Thus, peritoneal macrophages were extracted at 7 days after surgery and examined by flow cytometry to assess the polarization of macrophages in MI mice that received different treatments. MI modeling leads to a marked increase in the polarization of macrophages toward the M1 phenotype (TNF-α positive); however, the ratio of M1-type macrophages was moderately reduced after treatment with eMSCs. Further administration of NO via the intracellular delivery method significantly (P < 0.001 or 0.01) inhibited the polarization of macrophages toward the M1 phenotype while increasing the proportion of M2-type macrophages compared to the control and eMSC groups (Fig. 5, D and E). Additionally, we conducted immunofluorescence staining in heart section (fig. S5). The results revealed that intracellular delivery of NO also induces the polarization of macrophages into the M2 phenotype within the heart.

  In vitro studies demonstrated that exogenous NO could improve the proangiogenic capacity of eMSCs. Here, we further explored the influence of the combined delivery of exogenous NO and eMSCs on the reconstruction of the vascular network at the site of infarction. α–Smooth muscle actin (α-SMA)–positive arterioles and von Willebrand factor (vWF)–positive small vessels in MI mice after the different treatments were detected by immunofluorescence staining (Fig. 5, F and G). Treatment with eMSCs efficiently promoted angiogenesis in the injured myocardium, and more prominent enhancement was observed in response to further treatment with intracellular NO. This finding was further supported by the expression‎ of angiogenesis-related genes in the border zone of the infarcted heart (fig. S6).

  Intracellular delivery of NO improves heart function and inhibits adverse myocardial remodeling in MI mice after treatment with eMSCs

  Ultrasound and histological analyses were performed to eval‎uate the long-term recovery of cardiac function after MI. Cardiac injury was first eval‎uated by 2,3,5-triphenyltetrazolium chloride (TTC) staining (Fig. 6A). Treatment with eMSCs moderately repressed MI compared to the acute myocardial infarction (AMI) group, but a more pronounced inhibitory effect was observed in the group with further intracellular NO delivery. Left ventricular function was assessed by echocardiography at different time points. As shown in Fig. 6B, after 1 day of MI, the left ventricle in each group was markedly enlarged, cardiac function decreased rapidly, and deterioration of heart function continued for 28 days without detectable restoration in the AMI group. However, eMSC treatment could restore left ventricular systolic function and reduce ventricular dilation, as shown by the increase in left ventricular ejection fraction (LV-EF) and fraction shortening (LV-FS), as well as the decrease in left ventricular end-diastolic diameter (LVIDd) and left ventricular end-diastolic volume (LV-EDV) to a certain extent. In the group treated with eMSCs and intracellular NO delivery (NO-eMSCs), LV-EF and LV-FS were effectively recovered, while LVIDd and LV-EDV were significantly (P < 0.001 or 0.0001) enhanced compared to the AMI group.

  

  Fig. 6. Intracellular delivery of nitric oxide (NO) improves heart function and reduces adverse cardiac remodeling in myocardial infarction (MI) mice after treatment with engineered mesenchymal stem cells (eMSCs).

  (A) Representative images of 2,3,5-triphenyltetrazolium chloride (TTC) staining and quantification of the infarct area (n = 3). Scale bar, 2 mm. (B) Cardiac echo measurement was performed at different time points after surgery, and cardiac function indicators of left ventricular ejection fraction (LV-EF), left ventricular fractional shortening (LV-FS), left ventricular internal diameter at end diastole (LVIDd), and left ventricular end-diastolic volume (LV-EDV) were eval‎uated accordingly (n = 6). ***P < 0.001, ****P < 0.0001 versus acute myocardial infarction (AMI) group. (C) Masson’s trichrome staining was performed, and the infarct size was quantified accordingly (n = 6). (D) Collagen deposition in the hearts was detected by Sirius Red staining (n = 6). Scale bar, 100 μm. (E) Representative images of wheat germ agglutinin (WGA) immunofluorescence staining and quantification of the cross-sectional area of cardiomyocytes (n = 6). Scale bar, 50 μm. (F) Representative images of immunofluorescence staining (red) for the gap junction protein (Cx43) and the quantification of the intensity of red fluorescence to the whole area of images (n = 6). Scale bar, 25 μm. Data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

  Expand for more

  Open in viewer

  Masson staining was used to detect the degree of myocardial fibrosis after MI. Severe myocardial fibrosis was observed in MI mice compared to the sham group (Fig. 6C). In addition, the ventricular wall became thinner, which was closely related to impaired left ventricular systolic function after MI, as demonstrated by echocardiography. However, these adverse cardiac remodeling events were ameliorated after eMSC treatment and accompanied by a lowered degree of myocardial fibrosis and a thickened ventricular wall in contrast to the AMI group. Notably, treatment with eMSCs and intracellular NO delivery exerted the most prominent inhibitory effect on myocardial fibrosis after MI (Fig. 6C).

  Collagen deposition in MI mice was assessed by Sirius staining (Fig. 6D), and the results showed that MI resulted in severe collagen deposition in the injured myocardium compared to the sham group. It was effectively reduced after eMSC treatment, and the inhibitory effect of eMSC plus intracellular NO delivery was significantly higher than that in the other two groups (P < 0.001 or 0.0001).

  Next, wheat germ agglutinin (WGA) staining was carried out to eval‎uate myocardial cell hypertrophy 28 days after MI (Fig. 6E). The cross-sectional area of cardiomyocytes was increased in MI mice in contrast to the sham operation group due to compensatory hypertrophy in the heart to maintain the normal rate of cardiac ejection. Hypertrophy was significantly (P < 0.05) mitigated after treatment with eMSCs, especially in the presence of exogenous NO (P < 0.001), indicating an ideal therapeutic effect on inhibiting myocardial cell hypertrophy and adverse ventricular remodeling by the combination of eMSCs and NO.

  Gap junctions (GJs) are the main connections between cardiomyocytes in the heart, and Cx43 is the main GJ protein in ventricular muscle in the heart (24). Studies have shown that the absence of Cx43 leads to the occurrence of cardiac ventricular arrhythmia, which can develop into heart failure (25). After 28 days of MI, immunofluorescence staining for Cx43 revealed abundant and uniform distribution of GJ proteins in the sham group, whereas MI injury resulted in a marked decrease in the expression‎ of Cx43 (Fig. 6F). Despite the moderate inhibitory effect provided by the administration of eMSCs, further delivery of NO via the intracellular method significantly enhanced (P < 0.001 or 0.01) the expression‎ of Cx43 compared to that in the AMI or eMSC groups.

  Intracellular delivery of NO enhances the therapeutic efficacy of eMSCs in a rat MI model

  Although the outcome in a mouse model supported the beneficial effect of NO via intracellular delivery on enhancing the therapeutic efficacy of MSCs for MI, immediate administration of stem cells after MI is different from the clinical treatment of MI due to the limitation of the administration paradigm in mouse models. In addition, 3 to 7 days after MI is the outbreak period of the inflammatory response. For this reason, we established a rat model of MI and conducted secondary thoracotomy 3 days after surgery (Fig. 7A), and eMSCs were delivered via an injectable HA hydrogel as the carrier (Fig. 7B). Lactate dehydrogenase (LDH), a crucial marker for assessing the extent of myocardial damage, exhibited an initial elevation within 2 to 48 hours following the onset of MI, reaching its zenith between 2 and 5 days after MI. We collected blood samples from the orbital venous plexus of rats 5 days after MI to measure serum LDH levels (fig. S7). The findings revealed a sharp increase in serum LDH levels due to MI. However, treatment with NO-eMSCs significantly reduced serum LDH levels, indicating an attenuation of cardiac injury.

  

  Fig. 7. Intracellular delivery of nitric oxide (NO) enhances the therapeutic efficacy of engineered mesenchymal stem cells (eMSCs) in a rat myocardial infarction (MI) model.

  (A) Experimental schedule for the treatment of rat MI. (B) Representative images showing the second thoracotomy in rats after MI. (C) Cardiac echo measurement was performed at different time points after surgery, and cardiac function indicators of left ventricular ejection fraction (LV-EF), left ventricular fractional shortening (LV-FS), left ventricular internal diameter at end diastole (LVIDd), and left ventricular end-diastolic volume (LV-EDV) were eval‎uated accordingly (n = 6). **P < 0.01, ***P < 0.001 versus acute myocardial infarction (AMI) group. (D) Representative images of Masson’s trichrome staining and quantification of the infarct size and infarct thickness (n = 6). **P < 0.01, ***P < 0.001, ****P < 0.0001 versus AMI group. (E) Representative images of wheat germ agglutinin (WGA) immunofluorescence staining and quantification of the cross-sectional area of cardiomyocytes (n = 6). Scale bar, 50 μm. (F) Representative images of α–Smooth muscle actin (α-SMA) immunofluorescence staining and quantification of α-SMA+ arterioles (n = 6). Scale bar, 100 μm. Data are expressed as the mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

  Open in viewer

  Echocardiography was performed to detect heart function at different time points after MI (Fig. 7C). At 4 weeks after surgery, the anterior wall of the left ventricle was completely infarcted, and the ventricular cavity was increased dramatically. Quantitative analysis further indicated that both EF and FS were decreased markedly after surgery. Treatment with MSCs could restore the systolic function of the heart, and the highest recovery rate (60 to 80%) was observed in the group treated with eMSCs and intracellular NO delivery. Ventricular dilatation was also effectively restored after combined treatment with eMSCs and NO, which was consistent with the results obtained in the mouse model.

  Histological analyses showed that treatment with eMSCs combined with NO efficiently reduced the degree of fibrosis (Fig. 7D) and collagen deposition (fig. S8A). Furthermore, it alleviated infarct size after MI and restored ventricular cavity morphology of the heart with a significant enhancement in the thickness of the infarcted left ventricular wall and the interventricular septum (IVS), as eval‎uated by Masson (Fig. 7D) and H&E staining (fig. S9). Twenty-eight days after MI, WGA staining also showed that the combination of eMSCs and NO significantly (P < 0.01) inhibited cardiomyocyte hypertrophy in contrast to the AMI group, thus inhibiting further myocardial systolic dysfunction (Fig. 7E). The expression‎ of the GJ protein Cx43 (fig. S8B) followed a similar trend to that in mouse models; that is, acute MI led to a marked decrease in the distribution of Cx43 in the myocardium, and treatment with eMSCs up-regulated the expression‎ of Cx43 to a certain extent. Further delivery of NO via the intracellular method produced a more significant (P < 0.05 or 0.01) effect on the up-regulation of Cx43 in the injured myocardium compared to the eMSC and AMI groups, thereby facilitating the connection between cardiomyocytes and further inhibiting the occurrence of arrhythmias and heart failure.

  Tissue repair after MI is often closely related to angiogenesis, which begins at the infarct border and extends to the center of the infarction (26). Therefore, we further compared the proangiogenic effect of eMSCs with and without NO delivery on the damaged margin of the infarcted myocardium 28 days after MI in rats (Fig. 7F and fig. S8C). The combination of eMSCs and NO remarkably promoted angiogenesis and reconstruction of the vascular network compared to the group treated with eMSCs only, which is beneficial to the repair of myocardial injury after MI.

  DISCUSSION

  Cell therapy based on MSCs has proven to be a powerful solution for treating degenerative diseases and tissue damage (27–29). Despite the advantages of autologous stem cells over allogeneic stem cells, including the absence of immune rejection, the decreased survival and impaired paracrine functions of stem cells collected from patients with chronic diseases greatly limit their clinical use (7, 30, 31). Here, we first showed that MSCs collected from patients with diabetes exhibited marked up-regulation of apoptosis- and inflammation-related genes and down-regulation of proangiogenic genes, affecting the efficacy of cell therapy. As a result, genetic engineering strategies are often required to enhance the therapeutic efficacy of autologous stem cells. A recent study revealed that eMSCs, engineered to express PD-L1 on their surface and secrete CTLA4-Ig (immunoglobulin) as an extracellular factor, exhibited immunoprotective properties, which improved the outcome of both syngeneic and allogeneic islet transplantation in diabetic mice (32).

  NO is involved in a variety of physiological processes. Studies have shown that as an important signaling molecule, NO plays a pivotal role in regulating stem cell behavior (33–35), including cell survival, migration, differentiation, and paracrine behavior. These factors affect the interaction of stem cells with other cells and the tissue microenvironment. Previously, different types of NO-releasing biomaterials, such as injectable hydrogels, have been prepared by us and other groups (36–40), and further studies have shown that the combination of NO and MSCs is more effective in treating various diseases than MSC therapy alone. In addition, it has been reported that pretreatment of MSCs with NO-releasing biomaterials could enhance the therapeutic efficacy of MSCs and their secreted exosomes because of their enhanced proangiogenic functions (41).

  Due to the spatiotemporal characteristics of NO (42), precise delivery of NO in a site-specific and controllable manner holds great importance in the regulatory effect of exogenously administered NO. In addition to the controlled release rate, the site at which NO is generated is also a key factor due to the relative half-life and limited diffusion distance (43, 44). It is reasonable to speculate that intracellular and extracellular NO delivery may lead to different outcomes when regulating the survival and function of MSCs. In our previous work, an enzyme-prodrug delivery system was designed based on a bump-and-hole strategy (21). The mutant galactosidase (A4-β-GalH363A) enables the targeted delivery of NO, thus reducing the side effects due to the unspecific decomposition of the NO prodrug and enhancing the therapeutic efficacy. Here, we transfected a plasmid expressing mutant galactosidase into MSCs and successfully constructed eMSCs. The enzyme expressed by MSCs could catalyze the decomposition of the 6-OMe-galactose–protected NO prodrug and release NO intracellularly.

  Western blotting and fluorescence imaging demonstrated that the expression‎ of the engineered enzyme was confined to the nucleus of MSCs, while wild-type β-galactosidase was widely distributed in the cytoplasm, including the lysosome and perinuclear region (45, 46). Since the corresponding prodrug for wild-type β-galactosidase is highly hydrophilic, it fails to enter MSCs and releases NO extracellularly by enzymes that translocate from the cell. In contrast, the prodrug for mutant galactosidase is cell penetrating because of the modified molecular structure; therefore, it can enter MSCs and release NO intracellularly under the catalysis of the corresponding enzyme expressed by the cells. Accordingly, two different NO delivery paradigms were successfully developed in this study and further confirmed by a series of eval‎uations, including cell imaging and electronic paramagnetic resonance. Further in vitro and in vivo assays indicated that in contrast to extracellular NO delivery, intracellular administration of NO enhanced cell survival and the paracrine effects of MSCs, including inhibiting apoptosis and supporting angiogenesis.

  Next, we established a mouse MI model to systematically eval‎uate the therapeutic efficacy of MSCs combined with exogenous NO. The results showed that intracellular delivery of NO prolonged the retention of eMSCs after myocardial orthotopic transplantation. In addition, the combination of eMSCs and intracellular NO delivery improved cardiac function after MI and reduced adverse ventricular remodeling compared to the group treated with MSCs only. Additionally, it could effectively restore the reconstruction of the blood vessel network and further promote the repair of the infarcted myocardium.

  To gain further insight into the translational potential of the combinatory therapeutic strategy developed in this study, a rat model of MI was established, and MSCs were administered by a second thoracotomy after 3 days to mimic the clinical use of MSCs for the treatment of MI (47, 48). Clinically, acute MI is typically due to the rupture of coronary atherosclerotic plaque and the formation of thrombus, which causes coronary artery obstruction. After the acute phase of MI, adverse ventricular remodeling further affects the prognosis of patients, which is specifically characterized as a decrease in ventricular wall thickness and myocardial tension in the MI area, myocardial hypertrophy in the noninfarction area, and a change in the morphology of the ventricular cavity, thus leading to arrhythmia and further development into heart failure. The efficacy of MSCs in managing arrhythmias remains a topic of ongoing debate. Some researchers argue that MSCs do not appear to reduce or prevent arrhythmias, with the antiarrhythmic or proarrhythmic potential of MSCs primarily relying on paracrine factors (49). Conversely, other studies suggest that MSCs themselves may play a role in the post-MI recovery process (50). In our study, we observed an evident up-regulation of Cx43 expression‎ after NO-eMSC treatment, which is a potential target associated with antiarrhythmic effects. Further investigation is still required to comprehensively explore the antiarrhythmic potential of NO-eMSCs. In line with the enhanced therapeutic efficacy in the mouse model, intracellular delivery of NO showed enormous advantages in the rat MI model by inhibiting apoptosis and enhancing the paracrine function of MSCs.

  In summary, we first showed that survival and paracrine function were reduced in MSCs collected from patients with diabetes, which could greatly affect therapeutic efficacy. Accordingly, eMSCs were successfully constructed, and the mutant β-galactosidase expressed by the cells enabled the intracellular generation of NO via the conversion of an exogenous NO prodrug. In vitro and in vivo assays indicated that intracellular delivery of NO effectively enhanced the survival of transplanted MSCs and promoted the paracrine function of MSCs, which was further confirmed by the enhanced therapeutic efficacy in mouse and rat models of MI compared to the group treated with MSCs only. This synergistic strategy provides an option for the treatment of MI by autologous MSCs in the clinic.

  MATERIALS AND METHODSRNA sequencing analysis

  RNA sequencing was performed by the BGI (Shenzhen, China). Briefly, RNA from the ADMSCs of healthy people and patients with diabetes was extracted using TRIzol reagent (Yeasen, China). RNA samples were sequenced on the BGISEQ platform. The raw data containing low-quality reads, adaptor sequences, and high levels of N bases were filtered before analysis. Then, the clean reads were mapped to the reference genome using HISAT, and Bowtie2 was used to align the clean reads to the reference genes. The reference genome source is National Center for Biotechnology Information (NCBI), and the reference genome version is GCF_000001405.39_GRCh38.p13. The expression‎ levels of genes were quantified to identify differentially expressed genes by RNA-Seq by expectation maximization (RSEM). The analyses of hierarchical clustering and heatmap were performed using the online Dr. Tom system (biosys.bgi.com) to compare differential gene expression‎ of ADMSCs in healthy people and patients with diabetes. According to the KEGG_pathway annotation classification, the phyper function in R software was used for enrichment analysis, the P value was calculated, and then false discovery rate (FDR) was performed on the P value to obtain a Q value. Generally, a Q value of ≤0.05 was regarded as significant enrichment. GSEA was used to analyze significant differences in gene expression‎ between inflammatory and apoptosis-related pathways. Expression‎ cluster heatmap was used to analyze the expression‎ of genes associated with angiogenesis.

  Measurement of NO release

  The NO-releasing profile was determined by the Griess kit assay. In brief, 50 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were dissolved in phosphate-buffered saline (PBS) buffer (pH 7.4), and enzymes were added to the solutions at a concentration of 0.005 mg/ml. At each predetermined time interval, 50 ml of solution was transferred into a 96-well plate, and 50 ml of Griess I and 50 ml of Griess II were added thereafter. The azo compound of purple color was formed, and the absorbance was measured at a wavelength of 540 nm using an iMark microplate reader (Bio-Rad, USA).

  Cell cultureMesenchymal stem cells

  MSCs derived from human umbilical cord were obtained from Health-Biotech, maintained in Dulbecco's modified Eagle’s medium (DMEM)/F12 (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, USA) and 1% penicillin- streptomycin solution (Gibco, USA), and placed in a cell culture chamber containing 5% CO2 at 37°C.

  Human embryonic kidney 293T cells

  Human embryonic kidney (HEK) 293T cells were obtained from the American Type Culture Collection (ATCC), maintained in high-glucose DMEM (Gibco, USA) with 10% FBS (Gibco, USA) and 1% penicillin-streptomycin solution (Gibco, USA), and placed in a cell culture chamber containing 5% CO2 at 37°C.

  Construction of eMSCs

  The coding sequence of mutant β-galactosidase (A4-β-GalH363A) can be obtained from the previous publication (21). The lentivirus packaging system containing A4-β-GalH363A sequence and Rluc-RFP sequence was constructed by Wuhan Miaolingbio Co. Ltd. The constructed lentivirus plasmid containing the target gene and the package gene (psPAX2 and pMD2.G) was transfected into HEK 293T cells through Lipo2000, and the supernatant was collected to obtain the virus solution. After removing impurities, the virus solution was mixed 1:1 with fresh MSC medium, and polybrene (10 μg/ml) was added. MSCs were infected with virus through incubation in the mixture medium. The infection efficiency was observed under an inverted fluorescence microscope, and the expression‎ of target protein was determined by Western blotting.

  Cell immunofluorescence staining

  eMSCs were inoculated in 24-well plates. Cells were fixed with 4% paraformaldehyde and blocked in 4% bovine serum albumin in PBS for 30 min at room temperature. Then, the cells were incubated with primary antibodies overnight at 4°C. The bound primary antibodies were displayed by incubation with the secondary antibodies for 2 hours at room temperature. Cells were counterstained with DAPI (4′,6-diamidino-2-phenylindole)–containing Fluoromount-G and visualized with a fluorescence microscope (Axio Imager Z1). Antibodies used include anti–β-galactosidase (1:100, A1863, Abclonal) and anti-RFP (1:100, PA1-986, Invitrogen).

  Western blot

  eMSCs were collected, and total protein was extracted using radioimmunoprecipitation assay (RIPA) lysate containing protease inhibitor (Solarbio, China). Cytoplasmic protein and nucleoprotein were extracted using a nucleoprotein extraction kit containing protease inhibitors (Solarbio, China). The protein concentration was quantified using a BCA protein assay kit (Solarbio, China). The samples were diluted with 4× SDS–polyacrylamide gel electrophoresis (PAGE) loading buffer and boiled in boiling water for 8 min. Then, 30 μg of protein was isolated by 10% SDS-PAGE electrophoresis. The isolated proteins were transferred to an Immobilon-P Transfer membrane (Millipore, USA) and incubated with the primary antibody overnight at 4°C and then with the secondary antibody at room temperature for 2 hours. The bands were detected with chemiluminescent horseradish peroxidase substrate (Millipore, USA). Signals were generated by using an enhanced chemiluminescence (ECL) reagent (Millipore, USA) and were captured by using the Tanon-5200 Chemiluminescence Imaging System (Tanon, China). The antibodies used included anti–β-galactosidase (1:1000, A1863, Abclonal), anti–His-tag (1:1000, 12698S, Cell Signaling Technology), anti–glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (1:1000, AC001, Abclonal), anti–histone H3 (1:1000, 4499S, Cell Signaling Technology), anti–β-actin (1:1000, UM4001, Utibody), anti-BCL2 (1:1000, WL01556, Wanleibio), anti-Bax (1:1000, 50599-2-lg, Proteintech), anti-Bad (1:1000, WL02140, Wanleibio), and anti-caspase/cleaved caspase3 (1:1000, WL02117, Wanleibio).

  6-OMeGal-Ph-NO uptake by eMSCs

  eMSCs were inoculated in 12-well plates, and 50 μM substrate (6-OMeGal-Ph-NO) was added per well. At predetermined time points (0, 1, 3, 6, and 12 hours), an appropriate amount of culture medium was collected, and an excess of A4-β-GalH363A was added immediately to fully catalyze the decomposition of the remaining substrate at room temperature. The amount of 6-OMeGal-Ph-NO substrate in the culture medium was determined by Griess kit assay.

  Real-time imaging of intracellular NO

  The fluorescence emission associated with NO in the cytosol was detected by an electron-multiplying charge-coupled device (DU-897D-CS0-BV; Andor, Belfast, UK) connected to an inverted fluorescence microscope (Axio Observer D1; Carl Zeiss, Oberkochen, Germany). Intracellular NO imaging was performed using an NO fluorescence probe, DAF-AM DA (Beyotime, China), according to the manufacturer’s instruction. eMSCs were inoculated in small confocal dishes in advance, and the experiment was conducted when the cell density reached 80%. First, the medium was collected for later use. After two gentle washes with PBS, 5 μM DAF-AM DA solution was incubated at 37°C for 30 min in the dark. Then, cells were gently washed with PBS twice. The previously collected medium was added anew, and cell imaging was performed immediately. At the 488-nm excitation wavelength, pictures were taken every 5 s. After stable shooting for 2 min, 100 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were added. The filming for cell fluorescence was continued for 4 min. The final fluorescence intensity was determined without the background fluorescence value. The proportion of change in fluorescence intensity of each cell in the visual field was calculated.

  Intracellular NO detection

  Intracellular NO radicals (NO•) were detected using EPR as described (51, 52). In brief, sodium DETC (4.5 mg) and FeSO4•7H2O were dissolved in two separate volumes (10 μl) of deoxygenated Krebs/Hepes solution. Equal volumes of these parent solutions were rapidly mixed and aspirated into Eppendorf combi tips. The 0.5 mM Fe•(DETC)2 colloid solution had a yellow-brownish color with a slight opalescence in light. No aggregate formation was observed, at least during the first 30 min. eMSCs were rinsed with modified Krebs/Hepes buffer and incubated with freshly prepared NO•-specific spin trap Fe•(DETC)2 colloid (0.5 mM) for 30 min. Meanwhile, 100 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were added to the buffer. Gently collected cell suspensions were snap-frozen in liquid nitrogen. Ethyl acetate (200 μl) was added, and the cells were ultrasonically broken to extract DETC2-Fe-NO. The ethyl acetate extract was concentrated with nitrogen and transferred to a 50-μl capillary, and then the X-band EPR was measured at room temperature. The following acquisition parameters were used: modulation frequency, 100 kHz; microwave power, 10 mW; modulation amplitude, 2 G; number of scans, 60. The double-integrated area of the EPR spectra was calibrated into concentrations of DETC2-Fe-NO using TEMPO (2,2,5,5-tetramethyl piperidine 1-oxyl) as a standard. EPR spectral simulation was conducted by the WINSIM program.

  Extracellular NO detection

  eMSCs were treated with β-Gal-NO or 6-OMeGal-Ph-NO (30 μM). The production of NO in the medium of each group was detected 6 hours after incubation with NO-sensitive near-infrared fluorescence probe (5 μM). The NO production of medium in different groups was compared by the relative fluorescence intensity under the excitation at 750 nm (emission at 800 nm).

  Cell apoptosis detection

  To test the protective effect of NO delivery on cellular oxidative stress stimulation, 30 μM NO substrate (6-OMeGal-Ph-NO) was added to the medium in advance. Then, H2O2 with different concentrations (100, 200, 400, and 600 μM) was added to stimulate the lentivirus-infected eMSCs. BLI was performed immediately after addition of the luciferase substrate coelenterin to eval‎uate cell apoptosis. Additionally, eMSCs treated with 200 mM H2O2 were stimulated for 24 hours to induce cell apoptosis. An Annexin V/PI assay kit (Solarbio) was used to detect eMSC apoptosis.

  BLI detection of cell retention

  BLI and luciferase substrates were used in mice to eval‎uate the retention of NO-eMSCsGluc/RFP in cardiac orthotopic transplantation. The mice after eMSC injections were anesthetized with 1.5% isoflurane and injected with coelenterin through the caudal vein at 150 mg/kg. After injection, the mice were immediately placed in a BLI system to detect cell retention in the myocardium.

  Animals

  C57BL/6 mice (male, 8 weeks old) and Sprague-Dawley rats (male, 8 weeks old) were purchased from SPF (Beijing) Biotechnology Co. Ltd., Beijing, China. Animals were randomly grouped for treated and untreated controls. All experiments and animal procedures were approved by the Animal Experiments Ethical Committee of Nankai University and carried out in conformity with the Guide for Care and Use of Laboratory Animals.

  MI in mice and rats

  Surgical induction of MI was performed on C57BL/6 mice (male, 8 weeks old) as previously described with some modifications. Briefly, mice were anesthetized with 2% isoflurane, followed by fixation to a heating pad (37°C) at supine position, and then ventilated with a mechanical ventilation system (Hallowell EMC Microvent I, USA) set at 110 breaths per minute with a tidal volume of 0.2 to 0.3 ml. The third intercostal space over the left chest and the heart was exposed. After left thoracotomy, the left anterior descending coronary artery was ligated with a 6-0 silk ligature. The left main descending coronary artery was sutured and tied with a slip knot at a site approximately 3 mm from its origin. Cardiac ischemia was confirmed by the presence of myocardial blanching. Immediately, eMSCs encapsulated with HA hydrogel were injected into the myocardium of mice through three-point injection around the infarct zone. The chest cavity was closed to restore negative pressure and prevent pneumothorax.

  Similar MI surgery was performed on Sprague-Dawley rats (male, 8 weeks old) first. Briefly, rats were anesthetized via intraperitoneal injection of 10% chloral hydrate (350 mg/kg), followed by fixation to a heating pad (37°C) at supine position. Then, they were ventilated with a mechanical ventilation system (Hallowell EMC Microvent I, USA) set at 110 breaths per minute with a tidal volume of 6 ml. The third intercostal space over the left chest and the heart was exposed. After left thoracotomy, the left anterior descending coronary artery was ligated with a 6-0 silk ligature. The left main descending coronary artery was sutured and tied with a slip knot at a site approximately 3 mm from its origin. Cardiac ischemia was confirmed by the presence of myocardial blanching. The chest cavity was closed to restore negative pressure and prevent pneumothorax. Three days after surgery, secondary thoracotomy was performed, and eMSCs were injected into the myocardium around the infarct zone. The chest cavity was closed to restore negative pressure and prevent pneumothorax.

  Mice and rats in the AMI group only received MI without eMSC injection, while sham-operated mice only experienced thoracotomy without MI surgery or eMSC injection.

  At 1, 3, 5, and 7 days after myocardial injection of eMSCs, the prodrug was injected through the tail vein. Mice were injected with 100 μl of prodrug (1 mg/ml) each time, and rats were injected with 200 μl of prodrug (1 mg/ml) each time.

  TTC staining

  Two days after surgery, a thoracotomy was performed. The heart was quickly excised after quick freezing for 15 min and sliced at 1 mm thickness. Afterward, the sections were incubated with 1.5% TTC (Sigma-Aldrich) solution at 37°C in an incubator for 15 to 30 min and then with a 4% formaldehyde solution for 2 hours. The normal myocardial tissue was red, while the ischemic myocardium was white. The size of the infarcted myocardium, which was white or pale, was measured by ImageJ software.

  Cardiac function assessment

  Transthoracic echocardiography was performed with the Vevo 2100 Imaging System (Fuji Film Visual Sonics Inc., Canada) equipped with an MS-250/400 imaging transducer. The baseline cardiac function of mice and rats was measured at 3 days before surgery. Cardiac function was analyzed at days 1 and 28 after MI surgery with different treatments, as reported previously. Mice or rats were slightly anesthetized in a box with isoflurane. Their limbs were fixed in a supine position on the echo mat, and the chest hair was removed by depilating cream. Then, mice or rats were anesthetized by inhalation of isoflurane (0.5 to 1%) mixed with oxygen to maintain the heart rate at approximately 500 to 600, and M-mode echocardiography was performed. The left ventricular internal diameter at end-diastole (LVIDd) and systole (LVIDs) were obtained by measuring the long axis and the short axis. Accordingly, the cardiac parameters LV-EF, LV-FS, LV-EDV, and LV end-systole volume (LV-ESV) were determined. The echocardiography measurement was carried out in a double-blind manner.

  Histological analysis

  At the indicated time points, mice and rats were anesthetized via intraperitoneal injection of chloral hydrate, and a thoracotomy was performed. The hearts were fixed with trans-cardiac perfusion of saline and immersed in 4% paraformaldehyde over 24 hours. The heart tissue samples were dehydrated with gradient alcohol and xylene, embedded in paraffin blocks, and cut into sections in 5 μm thickness.

  The paraffin-embedded sections were stained with Masson trichrome, H&E, and Sirius Red following a standard protocol. Immunofluorescence staining was performed on paraffin-embedded sections of the heart tissue samples. After deparaffinization and heat-mediated antigen retrieval‎ in citrate solution, the samples were washed with PBS three times and incubated with blocking serum, which was used to avoid nonspecific binding, at room temperature for 30 min. The sections were incubated with specific antibodies diluted in goat serum at 4°C overnight. On the second day, the sections were rewarmed at room temperature for 1 hour and washed with PBS three times. Afterward, the sections were incubated with Alexa Fluor–coupled secondary antibodies for 2 hours at room temperature. After washing with PBS, the sections were counterstained with DAPI-containing Fluoromount-G (SouthernBiotech, USA) and coverslipped. The antibodies used included anti–α-SMA (1:100, ab5694, Abcam), anti-vWF (1:100, ab6694, Abcam), anti–α-actinin (1:100, ab9475, Abcam), anti-Connexin43 (1:100, ab11370, Abcam), anti-CD68 (1:100, ab125212, Abcam), WGA (1:500, FL-1021, Novus Biologicals), anti-iNOS (1:100, ab178945, Abcam), and anti-CD206 (1:100, ab64693, Abcam).

  Macrophage isolation and detection

  Three days before euthanasia, mice were intraperitoneally injected with 2 ml of 4% thioglycolate. Three days later, the mice were sacrificed by cervical dislocation and immersed in 75% alcohol and then transferred to an ultraclean workbench. The mouse limb was fixed in the supine position, and the mouse abdominal wall was carefully cut open with the peritoneal. PBS [1% penicillin-streptomycin (PS)] was injected intraperitoneally to collect the cell suspension, which was centrifuged at 2000 rpm for 10 min. After discarding the supernatant, the cells were incubated with anti-F4/80/TNF-α and anti-F4/80/CD206 antibodies. FlowJo software was used to analyze the results of flow cytometry.

  Quantitative real-time PCR

  Total RNA samples from the cells were prepared using TRIeasy Total RNA Extraction Reagent (Yeasen, China) according to the manufacturer’s instructions. Heart tissue samples were collected at the indicated time points after MI surgery.

  The tissue samples were dissected at the border zone of the left ventricle and frozen in liquid nitrogen immediately. Afterward, the total RNA was extracted with TRIzol reagent, as mentioned before. The concentration of the RNA was measured with a NanoDrop spectrophotometer (NanoDrop Technologies, USA). The complementary cDNA was synthesized using a first-strand cDNA synthesis kit (Yeasen, China). Quantitative real-time PCR was performed on a CFX96 real-time PCR system (Bio-Rad, Hercules, USA) with an SYBR Green–based real-time detection system (Yeasen, China). The relative gene expression‎ of mRNA was expressed as 2−(△△CT) and normalized to GAPDH as an internal control. Each reaction was performed in triplicate to obtain an average value, and the changes in relative gene expression‎ normalized to the internal control levels were determined. The highly purified primers used in this experiment were commercially synthesized (Sango, China). The sequences of the primers used in this experiment are summarized in the Supplementary Materials.

  Statistics

  All data are presented as the mean ± SEM from at least three independent experiments. Comparisons between two groups were performed by Student’s t test, and comparisons among more than two groups were performed by one-way or two-way analysis of variance (ANOVA). Statistical analyses were performed with GraphPad Prism software 7.0, and a statistical significance level of less than 0.05 was accepted.

  Acknowledgments

  Funding: This study is supported by the National Key R&D Program of China (2018YFE0200503), the National Natural Science Foundation of China (nos. 81925021, 82330066, 81921004, and U2004126), and the Tianjin Natural Science Foundation (21JCZDJC00240).

  Author contributions: Q.Z. and Z.L. conceived the original concept and initiated this project. Q.Z., Z.L., and F.G. designed the experiment and supervised the entire project. S.W. collected human adipose mesenchymal stem cells. M.Q. synthesized all NO prodrugs and probes. P.L. prepared engineered enzymes. T.H., G.J., and Q.X.L. established mouse and rat MI models. T.H. and G.J. performed histological analysis. G.J. and W.D. carried out in vitro cell experiments. S.D. carried out NO cell imaging under the supervision of L.P. T.H., G.J., and M.Q. analyzed data under the supervision of Q.Z. H.H. helped with lentivirus packaging and cell infection. W.G. and T.L. helped in establishing animal MI models. Y.W., J.H., J.C., and J.T. helped with data collection. T.H. and Q.Z. wrote the paper with input from other authors.

  Competing interests: The authors declare that they have no competing interests.

  Data and materials availability: All data needed to eval‎uate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Transcriptome sequencing dataset is available at https://doi.org/10.5061/dryad.tqjq2bw5b.

  Supplementary MaterialsThis PDF file includes:

  Supplementary Text

  Figs. S1 to S9

  Table S1

  • Download
  • 1.94 MB

  REFERENCES AND NOTES

  1

  M. F. Pittenger, A. M. Mackay, S. C. Beck, R. K. Jaiswal, R. Douglas, J. D. Mosca, M. A. Moorman, D. W. Simonetti, S. Craig, D. R. Marshak, Multilineage potential of adult human mesenchymal stem cells. Science 284, 143–147 (1999).

  GO TO REFERENCE

  Crossref

  PubMed

  Web of Science

  Google Scholar

  2

  A. Uccelli, L. Moretta, V. Pistoia, Mesenchymal stem cells in health and disease. Nat. Rev. Immunol. 8, 726–736 (2008).

  GO TO REFERENCE

  Crossref

  PubMed

  Web of Science

  Google Scholar

  3

  Y. Wang, J. Fang, B. Liu, C. Shao, Y. Shi, Reciprocal regulation of mesenchymal stem cells and immune responses. Cell Stem Cell 29, 1515–1530 (2022).

  GO TO REFERENCE

  Crossref

  PubMed

  Web of Science

  Google Scholar

  4

  Y. Wang, X. Chen, W. Cao, Y. Shi, Plasticity of mesenchymal stem cells in immunomodulation: Pathological and therapeutic implications. Nat. Immunol. 15, 1009–1016 (2014).

  GO TO REFERENCE

  Crossref

  PubMed

  Web of Science

  Google Scholar

  Show all references

  (0)eLetters

  eLetters is a forum for ongoing peer review. eLetters are not edited, proofread, or indexed, but they are screened. eLetters should provide substantive and scholarly commentary on the article. Neither embedded figures nor equations with special characters can be submitted, and we discourage the use of figures and equations within eLetters in general. If a figure or equation is essential, please include within the text of the eLetter a link to the figure, equation, or full text with special characters at a public repository with versioning, such as Zenodo. Please read our Terms of Service before submitting an eLetter.

  Log In to Submit a Response

  No eLetters have been published for this article yet.

  Recommended articles from TrendMD

  • In vivo priming of human mesenchymal stem cells with hepatocyte growth factor–engineered mesenchymal stem cells promotes therapeutic potential for cardiac repai...

    Bong-Woo Park, Sci Adv, 2020

  • Shattering barriers toward clinically meaningful MSC therapies

    Oren Levy, Sci Adv, 2020

  • Nanovesicles derived from iron oxide nanoparticles–incorporated mesenchymal stem cells for cardiac repair

    Ju-Ro Lee, Sci Adv, 2020

  • Exendin-4 gene modification and microscaffold encapsulation promote self-persistence and antidiabetic activity of MSCs

    Yuanyuan Zhang, Sci Adv, 2021

  • Human gingival mesenchymal stem cells retain their growth and immunomodulatory characteristics independent of donor age

    Jay R. Dave, Sci Adv, 2022

  • The secretion profile of mesenchymal stem cells and potential applications in treating human diseases

    Yuyi Han, Signal Transduction and Targeted Therapy, 2022

  • Mesenchymal Stem Cell Therapy for Ischemic Heart Disease: Systematic Review and Meta-analysis

    Hyunsuk Jeong, International Journal of Stem Cells

  • Mesenchymal Stem Cell and MicroRNA Therapy of Musculoskeletal Diseases

    International Journal of Stem Cells

  • Excessive branched-chain amino acid accumulation restricts mesenchymal stem cell-based therapy efficacy in myocardial infarction

    Fuyang Zhang, Signal Transduction and Targeted Therapy, 2022

    Intracellular delivery of nitric oxide enhances the therapeutic efficacy of mesenchymal stem cells for myocardial infarction

    Tian Hao https://orcid.org/0009-0007-8657-0424, Guangbo Ji https://orcid.org/0009-0000-7079-8896, Meng Qian https://orcid.org/0009-0006-9135-4593, Qiu Xuan Li, Haoyan Huang, Shiyu Deng, Pei Liu, Weiliang Deng https://orcid.org/0009-0001-8402-7280, Yongzhen Wei, [...], and Qiang Zhao https://orcid.org/0000-0003-4656-6002 +9 authorsAuthors Info & Affiliations

    Science Advances

    29 Nov 2023

    Vol 9, Issue 48

    DOI: 10.1126/sciadv.adi9967

    4,3612

    Metrics

    Total Downloads4,361

    • Last 6 Months1,270
    • Last 12 Months3,995

    Total Citations2

    • Last 6 Months1
    • Last 12 Months2

    NotificationsBookmark

    • • Abstract
      • INTRODUCTION
      • RESULTS
      • DISCUSSION
      • MATERIALS AND METHODS
      • Acknowledgments
      • Supplementary Materials
      • REFERENCES AND NOTES
      • eLetters (0)
    • • Information & Authors
      • Metrics & Citations
      • View Options
      • References
      • Media
      • Tables
      • Share

    Abstract

    Cell therapy by autologous mesenchymal stem cells (MSCs) is a clinically acceptable strategy for treating various diseases. Unfortunately, the therapeutic efficacy is largely affected by the low quality of MSCs collected from patients. Here, we showed that the gene expression‎ of MSCs from patients with diabetes was differentially regulated compared to that of MSCs from healthy controls. Then, MSCs were genetically engineered to catalyze an NO prodrug to release NO intracellularly. Compared to extracellular NO conversion, intracellular NO delivery effectively prolonged survival and enhanced the paracrine function of MSCs, as demonstrated by in vitro and in vivo assays. The enhanced therapeutic efficacy of engineered MSCs combined with intracellular NO delivery was further confirmed in mouse and rat models of myocardial infarction, and a clinically relevant cell administration paradigm through secondary thoracotomy has been attempted.

    SIGN UP FOR THE AWARD-WINNING SCIENCEADVISER NEWSLETTER

    The latest news, commentary, and research, free to your inbox daily

    Sign up

    INTRODUCTION

    Mesenchymal stem cells (MSCs) are pluripotent stem cells with high self-renewal abilities and multidirectional differentiation potential (1, 2). They are extensively distributed throughout the body and serve a variety of purposes, including tissue regeneration (3), immunoregulation (4), and angiogenesis (5). However, an important challenge for stem cell therapy is the low survival rate of stem cells after transplantation, which is associated with nonspecific homing of cells and ischemia/hypoxia at the injury site (6). In addition, transplanted stem cells cannot fully exert their paracrine effects in pathological environments, which seriously limits the clinical application of stem cells.

    The proangiogenic function of MSCs is a key factor contributing to the treatment of ischemic diseases. Generally, MSCs can stimulate local angiogenesis in ischemic tissue by secreting cytokines, including vascular endothelial growth factor (VEGF), transforming growth factor β (TGFβ), and basic fibroblast growth factor (bFGF), to induce endothelial cells to form tubular structures (7). In addition, MSCs can recruit pericytes and smooth muscle cells to promote neovascularization (8, 9). Recently, MSCs have been reported to participate in the construction of blood vessels by directly differentiating into endothelial or smooth muscle cells (10).

    As an important signaling molecule, nitric oxide (NO) plays a crucial role in the maintenance of vascular homeostasis by inhibiting thrombus formation and promoting angiogenesis (11, 12). Recently, increasing attention has been given to the regulation of stem cells by NO due to its multiple biological functions (13, 14). NO can regulate the proliferation of stem cells by regulating the activities of cyclin and mitotic receptors, as well as their downstream pathways (15, 16). On the other hand, NO can regulate the expression‎ of angiogenic cytokines and immunomodulatory factors to improve the paracrine performance of stem cells (17). Additionally, recent studies have demonstrated that NO can regulate the differentiation behavior of stem cells through the phosphatidylinositol 3-kinase (PI3K)/AKT, guanosine 3′,5′-monophosphate (cGMP), and other signaling pathways (18).

    As a result, NO-releasing biomaterials have been used as delivery carriers for stem cells to enhance their survival and regulate paracrine functions (19). However, NO, which is a gaseous molecule, easily diffuses and has a high level of instability. Furthermore, the physiologic function of NO is dose dependent, and an overdose of NO often leads to notable cytotoxicity (20). Thus, optimizing the beneficial effects of NO to strengthen the therapeutic efficacy of stem cells by tuning their release profile should be taken into account.

    In our previous study, based on the chemical biology principle of “bump-and-hole,” we designed and prepared an enzyme-prodrug delivery system and achieved targeted delivery of NO at the lesion site in two different ischemic disease models (21). Here, MSCs were further modified by gene transfection to express a catalytic enzyme (A4-β-GalH363A). The engineered MSCs (eMSCs) were transplanted using an injectable hyaluronic acid (HA) hydrogel as the carrier, while the NO prodrug was injected through the tail vein to achieve controlled release of NO catalyzed by the enzymes in eMSCs. The therapeutic efficacy of MSCs combined with exogenous NO delivery was eval‎uated in mouse and rat models of myocardial infarction (MI) with an emphasis on comparing the therapeutic efficacy of two different NO administration methods (intracellular or extracellular), and the underlying mechanism of their myocardial protective effect was further explored.

    RESULTSThe gene expression‎ of MSCs collected from patients is differentially regulated

    In the clinic, the therapeutic efficacy of autologous stem cell transplantation is largely affected by the low quality (including cell survival and paracrine function) of stem cells collected from patients due to chronic diseases. Diabetes mellitus is a chronic medical condition that can lead to a variety of complications, and these complications can affect various parts of the body, such as the kidney, lower limb, and heart. Diabetes predisposes affected individuals to a spectrum of cardiovascular complications, and one of the most debilitating in terms of prognosis is heart failure (22).

    Accordingly, although autologous MSCs have been widely accepted as a promising strategy for treating various complications associated with diabetes, the therapeutic efficacy is largely affected by the quality of stem cells collected from the patients themselves. The gene levels in adipose-derived MSCs (ADMSCs) from patients with diabetes and healthy individuals were first compared by transcriptome sequencing. The heatmap shows that multiple genes in MSCs from patients with diabetes were up- or down-regulated compared to healthy controls (Fig. 1A). Gene ontology (GO) enrichment analysis revealed that the most differentially up-regulated genes were related to tumor necrosis factor (TNF) signaling pathways (Fig. 1B). Subsequently, we performed gene set enrichment analysis (GSEA) based on the RNA-sequencing results. GSEA revealed that inflammatory target genes were highly enriched in MSCs from patients with diabetes, with a normalized enrichment score (NES) of 1.65 (P < 0.01) (Fig. 1C). Apoptosis-related genes were also highly enriched, with an NES of 1.77 (P < 0.01) (Fig. 1D). In addition, through enrichment analysis, we found that proangiogenic genes in MSCs were greatly down-regulated in patients with diabetes (Fig. 1E).

    

    Fig. 1. Transcriptome sequencing of adipose mesenchymal stem cells (MSCs) collected from patients with diabetes and healthy controls.

    (A) Heatmap showing the differentially expressed genes of MSCs from patients with diabetes and healthy controls (n = 3). (B) Gene ontology (GO) analysis of the up-regulated transcriptome of MSCs from patients with diabetes. (C and D) Gene set enrichment analysis (GSEA) was performed to determine the enrichment of inflammation (C) and apoptosis (D) target genes in the diabetic group. (E) Heatmap showing angiogenesis-related genes in the two groups (n = 3). (F) Schematic illustration demonstrating the difference in MSCs between patients with diabetes and healthy controls at the gene level.

    Open in viewer

    eMSCs are constructed to express mutant β-galactosidase

    Here, we first designed and constructed plasmids carrying the mutant β-galactosidase (A4-β-GalH363A) target gene and luciferase–red fluorescent protein (RFP) dual reporter genes, which could be further used for in vivo imaging. eMSCs expressing A4-β-GalH363A were constructed by infecting MSCs with lentiviruses obtained from human embryonic kidney 293T cells (Fig. 2A). Immunofluorescence staining for RFP confirmed that A4-β-GalH363A was successfully expressed by eMSCs (Fig. 2B). The subcellular fraction and intracellular distribution of enzymes expressed by the eMSCs was determined by Western blotting (Fig. 2C). In contrast to natural β-galactosidases, which are widely distributed within cells, A4-β-GalH363A was mainly confined to the nucleus of eMSCs.

    

    Fig. 2. Intracellular expression‎ and localization of A4-β-GalH363A.

    (A) Schematic diagram of lentivirus packaging and mesenchymal stem cell (MSC) infection. (B) Immunofluorescence staining of β-Gal and A4-β-GalH363A in eMSCs. Scale bar, 50 μm. (C) The distribution of two different enzymes in engineered MSCs (eMSCs) was analyzed by Western blotting.

    Open in viewer

    Intracellular release of NO is realized via decomposition of the 6-OMeGal-Ph-NO prodrug

    Since the mutant β-galactosidase was confined to the nucleus in eMSCs, we designed a prodrug by introducing a lipid-soluble self-decomposition chain into 6-OMeGal-NO to improve its oil and water distribution coefficient; therefore, the resultant NO donor 6-OMeGal-Ph-NO could penetrate the cell membrane and decompose and release NO under the catalysis of A4-β-GalH363A (Fig. 3A and Supplementary Materials). An in vitro release assay showed that the 6-OMeGal-Ph-NO prodrug was efficiently recognized and converted by A4-β-GalH363A with a cumulative release ratio of approximately 97.3%, while nearly no release was observed in the presence of wild-type β-galactosidase (Fig. 3B). To confirm‎ intracellular NO release, eMSCs were preincubated with an NO fluorescent probe (DAF-AM DA) and then treated with different NO prodrugs. The changes in fluorescence signals were examined by time-delay cell imaging (Fig. 3C). The results indicated that the fluorescence intensity continuously increased in the group that was treated with 6-OMeGal-Ph-NO, indicating conversion into NO (Fig. 3D). In contrast, no detectable changes were identified in the β-Gal-NO group because the high water solubility restricted its entry into eMSCs. Next, intracellular and extracellular release of NO in eMSCs was assessed (Fig. 3E). The quantity of intracellular NO was measured by electron paramagnetic resonance (EPR) using ferrous N-diethyl dithiocarbamate (DETC2-Fe) as the spin-trapping reagent. The resultant NO adduct (DETC2-Fe-NO) exhibited a characteristic triplet EPR signal (aN = 13.06 G, giso = 2.041) at room temperature. Quantitative analysis showed that the NO level was significantly (P < 0.001 or 0.0001) higher in eMSCs treated with 6-OMeGal-Ph-NO than in the β-Gal-NO and control groups (Fig. 3F). Furthermore, in the group treated with β-Gal-NO, the release of NO was mainly catalyzed by β-galactosidase that translocated from the cytoplasm in eMSCs, and the extracellular release profile of β-Gal-NO was confirmed by detecting the NO level in the cell culture medium with the NO-sensitive near-infrared fluorescence probe (23); it was significantly (P < 0.01) higher than that in the 6-OMeGal-Ph-NO and control groups (Fig. 3G). To determine the uptake of the NO prodrug by eMSCs, we incubated 6-OMeGal-Ph-NO with eMSCs, and the concentration in the culture medium was determined at different time points. The results reflected that approximately 45% of 6-OMeGal-Ph-NO was incorporated into the eMSCs within 12 hours (fig. S1).

    

    Fig. 3. Intracellular generation of nitric oxide (NO) from the NO prodrug under the catalysis of A4-β-GalH363A expressed by engineered mesenchymal stem cells (eMSCs).

    (A) Synthesis of two NO prodrugs with different enzyme response abilities and cellular permeabilities. (B) In vitro release profile of NO from the NO prodrug (6-OMeGal-Ph-NO) in the presence of β-Gal or A4-β-GalH363A. (C) Schematic illustration of intracellular NO imaging by using an NO fluorescence probe (DAF-AM DA). (D) Representative time-lapse images of NO generation from two different prodrugs in eMSCs and quantification of the fluorescence intensity (n = 6). ***P < 0.001, ****P < 0.0001 versus 6-OMeGal-Ph-NO group. (E) Schematic illustration showing the detection of intracellular and extracellular NO generation differentially. (F) Representative electron paramagnetic resonance (EPR) spectra and quantification of intracellular NO generation by measuring the DETC2-Fe-NO complex using 2,2,5,5-tetramethyl piperidine 1-oxyl (TEMPO) as a standard (n = 3). (G) Relative quantification of NO production in the medium determined using the near-infrared fluorescence probe (n = 4). Data are expressed as the mean ± SEM. Significant differences were detected by one-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

    Expand for more

    Open in viewer

    Intracellular delivery of NO inhibits apoptosis and enhances the proangiogenic activity of eMSCs

    Overproduction of reactive oxygen species due to cellular oxidative stress has been accepted as an important factor contributing to apoptosis in transplanted cells in ischemic tissue. Therefore, we assessed the protective effect of exogenously administered NO on the survival of eMSCs with H2O2-induced oxidative stress and focused on comparing the protection provided by intracellular and extracellular NO administration. The results showed that H2O2 stimulated apoptosis, and delivery of NO via extracellular and intracellular strategies significantly (P < 0.01 or 0.001) reduced apoptosis in eMSCs stimulated by oxidative stress, and the highest fluorescence signal was observed in response to intracellular NO delivery (Fig. 4A).

    

    Fig. 4. Intracellular delivery of nitric oxide (NO) inhibits apoptosis of engineered mesenchymal stem cells (eMSCs).

    (A) Bioluminescence imaging (BLI) was used to detect the effect of NO delivery on cell apoptosis stimulated by different concentrations of H2O2, and the fluorescence signals were further quantified (n = 3). **P < 0.01, ***P < 0.001, ****P < 0.0001 versus control group; #P < 0.05 versus β-Gal-NO group. (B) Flow cytometry assay of cell viability and apoptosis of eMSCs after H2O2 stimulation, and quantification of mean percent values of apoptotic cells (n = 3). (C) The expression‎ of apoptosis-related protein (BCL2, Bax, Bad, caspase3, and cleaved caspase3) by eMSCs was detected after H2O2 stimulation by Western blots (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001. (D) Representative BLI photographs reflecting the retention of eMSCs with and without intracellular NO delivery after in vivo transplantation as well as the quantitative analysis of signals (n = 3). Data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01.

    Open in viewer

    The survival of eMSCs after treatment with various NO delivery strategies was compared by using flow cytometry. Intracellular NO delivery had the most remarkable antiapoptotic effect on cells stimulated by oxidative stress (P < 0.001) (Fig. 4B). The expression‎ of apoptosis-related genes by eMSCs followed a similar trend; NO delivery (extracellular and intracellular) effectively promoted the expression‎ of the antiapoptotic gene BCL2 in eMSCs after oxidative stress stimulation, while the expression‎ levels of proapoptotic genes were reduced accordingly. Intracellular NO delivery via administration of the 6-OMeGal-Ph-NO prodrug exerted a more pronounced antiapoptotic effect at both the gene and protein levels (Fig. 4C and fig. S2A), which may be because the intracellular generation of NO directly activated the antioxidant system in cells to resist oxidative stress damage and inhibit further apoptosis.

    The expression‎ of proangiogenic genes, including ANGPT1, ANGPT2, FGF2, VEGFA, and KDR, in eMSCs was further detected by reverse transcription polymerase chain reaction (RT-PCR). The results showed that the expression‎ level of proangiogenic genes was significantly (P < 0.05, 0.01, or 0.001) higher in eMSCs treated with the 6-OMeGal-Ph-NO prodrug than in the other groups, indicating the enhanced proangiogenic functions of eMSCs after intracellular NO delivery (fig. S2B).

    We further eval‎uated the effect of NO delivery on the in vivo retention of eMSCs after orthotopic transplantation in the myocardial tissue of mice. As shown in Fig. 4D, intracellular delivery of NO effectively prolonged the retention of eMSCs within the myocardium, and an evident bioluminescence imaging (BLI) signal corresponding to the retention of eMSCs was observed 7 days after transplantation compared to the counterpart without administration of the NO prodrug. To further eval‎uate the translational potential of eMSCs in clinical settings, we used MSCs derived from diabetic patients and conducted a series of assays related to cell survival and paracrine function. The findings indicated that intracellular delivery of NO also confers advantages in the attenuation of cell apoptosis under stress conditions, thereby prolonging the in vivo retention of eMSCs (fig. S3).

    Intracellular delivery of NO ameliorates myocardial injury in MI mice after treatment with eMSCs

    The therapeutic efficacy of MSCs combined with exogenous NO was further eval‎uated in a mouse MI model (Fig. 5A and fig. S4). The inflammatory response in the early stage (3 days) was first detected by hematoxylin-eosin (H&E) staining and CD68 immunofluorescence staining. The results demonstrated that severe inflammatory cell infiltration occurred in the injured myocardium of MI mice, and it was effectively alleviated after eMSC treatment. More prominent restoration in the injured myocardium was observed after further administration of NO (Fig. 5, B and C), confirm‎ing the inhibitory effect on inflammation after MI provided by the combination of MSCs and NO. Moreover, this inhibitory effect was more significant (P < 0.01 or 0.0001) in the group with intracellular NO delivery than in the group with extracellular NO delivery.

    

    Fig. 5. Intracellular delivery of nitric oxide (NO) ameliorates myocardial injury in myocardial infarction (MI) mice after treatment with engineered mesenchymal stem cells (eMSCs) via regulation of the inflammation and stimulation of angiogenesis.

    (A) Experimental schedule for the treatment of MI in a mouse model. (B) Hematoxylin-eosin (H&E) staining was performed to detect inflammatory cell infiltration in the early stage of MI (n = 6). Scale bar, 100 μm. (C) Representative images of CD68 immunofluorescence staining (green) and quantification of CD68+ macrophages in injured myocardium (n = 6). Scale bar, 25 μm. (D and E) Flow cytometry was performed to detect peritoneal macrophage polarization 7 days after surgery followed by different treatments. TNFα- and CD206-positive ascites macrophages (markers of M1 and M2 macrophage phenotypes, respectively) were quantified accordingly (n = 3). (F) Representative images of α–Smooth muscle actin (α-SMA) immunofluorescence staining and the quantification of α-SMA+ arterioles (n = 6). Scale bar, 100 μm. (G) Representative images of von Willebrand factor (vWF) immunofluorescence staining and the quantification of vWF+ capillaries (n = 6). Scale bar, 100 μm. Data are expressed as the mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

    Expand for more

    Open in viewer

    It has been widely accepted that a higher proportion of M2-type macrophages is more conducive to the repair of tissue damage. Thus, peritoneal macrophages were extracted at 7 days after surgery and examined by flow cytometry to assess the polarization of macrophages in MI mice that received different treatments. MI modeling leads to a marked increase in the polarization of macrophages toward the M1 phenotype (TNF-α positive); however, the ratio of M1-type macrophages was moderately reduced after treatment with eMSCs. Further administration of NO via the intracellular delivery method significantly (P < 0.001 or 0.01) inhibited the polarization of macrophages toward the M1 phenotype while increasing the proportion of M2-type macrophages compared to the control and eMSC groups (Fig. 5, D and E). Additionally, we conducted immunofluorescence staining in heart section (fig. S5). The results revealed that intracellular delivery of NO also induces the polarization of macrophages into the M2 phenotype within the heart.

    In vitro studies demonstrated that exogenous NO could improve the proangiogenic capacity of eMSCs. Here, we further explored the influence of the combined delivery of exogenous NO and eMSCs on the reconstruction of the vascular network at the site of infarction. α–Smooth muscle actin (α-SMA)–positive arterioles and von Willebrand factor (vWF)–positive small vessels in MI mice after the different treatments were detected by immunofluorescence staining (Fig. 5, F and G). Treatment with eMSCs efficiently promoted angiogenesis in the injured myocardium, and more prominent enhancement was observed in response to further treatment with intracellular NO. This finding was further supported by the expression‎ of angiogenesis-related genes in the border zone of the infarcted heart (fig. S6).

    Intracellular delivery of NO improves heart function and inhibits adverse myocardial remodeling in MI mice after treatment with eMSCs

    Ultrasound and histological analyses were performed to eval‎uate the long-term recovery of cardiac function after MI. Cardiac injury was first eval‎uated by 2,3,5-triphenyltetrazolium chloride (TTC) staining (Fig. 6A). Treatment with eMSCs moderately repressed MI compared to the acute myocardial infarction (AMI) group, but a more pronounced inhibitory effect was observed in the group with further intracellular NO delivery. Left ventricular function was assessed by echocardiography at different time points. As shown in Fig. 6B, after 1 day of MI, the left ventricle in each group was markedly enlarged, cardiac function decreased rapidly, and deterioration of heart function continued for 28 days without detectable restoration in the AMI group. However, eMSC treatment could restore left ventricular systolic function and reduce ventricular dilation, as shown by the increase in left ventricular ejection fraction (LV-EF) and fraction shortening (LV-FS), as well as the decrease in left ventricular end-diastolic diameter (LVIDd) and left ventricular end-diastolic volume (LV-EDV) to a certain extent. In the group treated with eMSCs and intracellular NO delivery (NO-eMSCs), LV-EF and LV-FS were effectively recovered, while LVIDd and LV-EDV were significantly (P < 0.001 or 0.0001) enhanced compared to the AMI group.

    

    Fig. 6. Intracellular delivery of nitric oxide (NO) improves heart function and reduces adverse cardiac remodeling in myocardial infarction (MI) mice after treatment with engineered mesenchymal stem cells (eMSCs).

    (A) Representative images of 2,3,5-triphenyltetrazolium chloride (TTC) staining and quantification of the infarct area (n = 3). Scale bar, 2 mm. (B) Cardiac echo measurement was performed at different time points after surgery, and cardiac function indicators of left ventricular ejection fraction (LV-EF), left ventricular fractional shortening (LV-FS), left ventricular internal diameter at end diastole (LVIDd), and left ventricular end-diastolic volume (LV-EDV) were eval‎uated accordingly (n = 6). ***P < 0.001, ****P < 0.0001 versus acute myocardial infarction (AMI) group. (C) Masson’s trichrome staining was performed, and the infarct size was quantified accordingly (n = 6). (D) Collagen deposition in the hearts was detected by Sirius Red staining (n = 6). Scale bar, 100 μm. (E) Representative images of wheat germ agglutinin (WGA) immunofluorescence staining and quantification of the cross-sectional area of cardiomyocytes (n = 6). Scale bar, 50 μm. (F) Representative images of immunofluorescence staining (red) for the gap junction protein (Cx43) and the quantification of the intensity of red fluorescence to the whole area of images (n = 6). Scale bar, 25 μm. Data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

    Expand for more

    Open in viewer

    Masson staining was used to detect the degree of myocardial fibrosis after MI. Severe myocardial fibrosis was observed in MI mice compared to the sham group (Fig. 6C). In addition, the ventricular wall became thinner, which was closely related to impaired left ventricular systolic function after MI, as demonstrated by echocardiography. However, these adverse cardiac remodeling events were ameliorated after eMSC treatment and accompanied by a lowered degree of myocardial fibrosis and a thickened ventricular wall in contrast to the AMI group. Notably, treatment with eMSCs and intracellular NO delivery exerted the most prominent inhibitory effect on myocardial fibrosis after MI (Fig. 6C).

    Collagen deposition in MI mice was assessed by Sirius staining (Fig. 6D), and the results showed that MI resulted in severe collagen deposition in the injured myocardium compared to the sham group. It was effectively reduced after eMSC treatment, and the inhibitory effect of eMSC plus intracellular NO delivery was significantly higher than that in the other two groups (P < 0.001 or 0.0001).

    Next, wheat germ agglutinin (WGA) staining was carried out to eval‎uate myocardial cell hypertrophy 28 days after MI (Fig. 6E). The cross-sectional area of cardiomyocytes was increased in MI mice in contrast to the sham operation group due to compensatory hypertrophy in the heart to maintain the normal rate of cardiac ejection. Hypertrophy was significantly (P < 0.05) mitigated after treatment with eMSCs, especially in the presence of exogenous NO (P < 0.001), indicating an ideal therapeutic effect on inhibiting myocardial cell hypertrophy and adverse ventricular remodeling by the combination of eMSCs and NO.

    Gap junctions (GJs) are the main connections between cardiomyocytes in the heart, and Cx43 is the main GJ protein in ventricular muscle in the heart (24). Studies have shown that the absence of Cx43 leads to the occurrence of cardiac ventricular arrhythmia, which can develop into heart failure (25). After 28 days of MI, immunofluorescence staining for Cx43 revealed abundant and uniform distribution of GJ proteins in the sham group, whereas MI injury resulted in a marked decrease in the expression‎ of Cx43 (Fig. 6F). Despite the moderate inhibitory effect provided by the administration of eMSCs, further delivery of NO via the intracellular method significantly enhanced (P < 0.001 or 0.01) the expression‎ of Cx43 compared to that in the AMI or eMSC groups.

    Intracellular delivery of NO enhances the therapeutic efficacy of eMSCs in a rat MI model

    Although the outcome in a mouse model supported the beneficial effect of NO via intracellular delivery on enhancing the therapeutic efficacy of MSCs for MI, immediate administration of stem cells after MI is different from the clinical treatment of MI due to the limitation of the administration paradigm in mouse models. In addition, 3 to 7 days after MI is the outbreak period of the inflammatory response. For this reason, we established a rat model of MI and conducted secondary thoracotomy 3 days after surgery (Fig. 7A), and eMSCs were delivered via an injectable HA hydrogel as the carrier (Fig. 7B). Lactate dehydrogenase (LDH), a crucial marker for assessing the extent of myocardial damage, exhibited an initial elevation within 2 to 48 hours following the onset of MI, reaching its zenith between 2 and 5 days after MI. We collected blood samples from the orbital venous plexus of rats 5 days after MI to measure serum LDH levels (fig. S7). The findings revealed a sharp increase in serum LDH levels due to MI. However, treatment with NO-eMSCs significantly reduced serum LDH levels, indicating an attenuation of cardiac injury.

    

    Fig. 7. Intracellular delivery of nitric oxide (NO) enhances the therapeutic efficacy of engineered mesenchymal stem cells (eMSCs) in a rat myocardial infarction (MI) model.

    (A) Experimental schedule for the treatment of rat MI. (B) Representative images showing the second thoracotomy in rats after MI. (C) Cardiac echo measurement was performed at different time points after surgery, and cardiac function indicators of left ventricular ejection fraction (LV-EF), left ventricular fractional shortening (LV-FS), left ventricular internal diameter at end diastole (LVIDd), and left ventricular end-diastolic volume (LV-EDV) were eval‎uated accordingly (n = 6). **P < 0.01, ***P < 0.001 versus acute myocardial infarction (AMI) group. (D) Representative images of Masson’s trichrome staining and quantification of the infarct size and infarct thickness (n = 6). **P < 0.01, ***P < 0.001, ****P < 0.0001 versus AMI group. (E) Representative images of wheat germ agglutinin (WGA) immunofluorescence staining and quantification of the cross-sectional area of cardiomyocytes (n = 6). Scale bar, 50 μm. (F) Representative images of α–Smooth muscle actin (α-SMA) immunofluorescence staining and quantification of α-SMA+ arterioles (n = 6). Scale bar, 100 μm. Data are expressed as the mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

    Open in viewer

    Echocardiography was performed to detect heart function at different time points after MI (Fig. 7C). At 4 weeks after surgery, the anterior wall of the left ventricle was completely infarcted, and the ventricular cavity was increased dramatically. Quantitative analysis further indicated that both EF and FS were decreased markedly after surgery. Treatment with MSCs could restore the systolic function of the heart, and the highest recovery rate (60 to 80%) was observed in the group treated with eMSCs and intracellular NO delivery. Ventricular dilatation was also effectively restored after combined treatment with eMSCs and NO, which was consistent with the results obtained in the mouse model.

    Histological analyses showed that treatment with eMSCs combined with NO efficiently reduced the degree of fibrosis (Fig. 7D) and collagen deposition (fig. S8A). Furthermore, it alleviated infarct size after MI and restored ventricular cavity morphology of the heart with a significant enhancement in the thickness of the infarcted left ventricular wall and the interventricular septum (IVS), as eval‎uated by Masson (Fig. 7D) and H&E staining (fig. S9). Twenty-eight days after MI, WGA staining also showed that the combination of eMSCs and NO significantly (P < 0.01) inhibited cardiomyocyte hypertrophy in contrast to the AMI group, thus inhibiting further myocardial systolic dysfunction (Fig. 7E). The expression‎ of the GJ protein Cx43 (fig. S8B) followed a similar trend to that in mouse models; that is, acute MI led to a marked decrease in the distribution of Cx43 in the myocardium, and treatment with eMSCs up-regulated the expression‎ of Cx43 to a certain extent. Further delivery of NO via the intracellular method produced a more significant (P < 0.05 or 0.01) effect on the up-regulation of Cx43 in the injured myocardium compared to the eMSC and AMI groups, thereby facilitating the connection between cardiomyocytes and further inhibiting the occurrence of arrhythmias and heart failure.

    Tissue repair after MI is often closely related to angiogenesis, which begins at the infarct border and extends to the center of the infarction (26). Therefore, we further compared the proangiogenic effect of eMSCs with and without NO delivery on the damaged margin of the infarcted myocardium 28 days after MI in rats (Fig. 7F and fig. S8C). The combination of eMSCs and NO remarkably promoted angiogenesis and reconstruction of the vascular network compared to the group treated with eMSCs only, which is beneficial to the repair of myocardial injury after MI.

    DISCUSSION

    Cell therapy based on MSCs has proven to be a powerful solution for treating degenerative diseases and tissue damage (27–29). Despite the advantages of autologous stem cells over allogeneic stem cells, including the absence of immune rejection, the decreased survival and impaired paracrine functions of stem cells collected from patients with chronic diseases greatly limit their clinical use (7, 30, 31). Here, we first showed that MSCs collected from patients with diabetes exhibited marked up-regulation of apoptosis- and inflammation-related genes and down-regulation of proangiogenic genes, affecting the efficacy of cell therapy. As a result, genetic engineering strategies are often required to enhance the therapeutic efficacy of autologous stem cells. A recent study revealed that eMSCs, engineered to express PD-L1 on their surface and secrete CTLA4-Ig (immunoglobulin) as an extracellular factor, exhibited immunoprotective properties, which improved the outcome of both syngeneic and allogeneic islet transplantation in diabetic mice (32).

    NO is involved in a variety of physiological processes. Studies have shown that as an important signaling molecule, NO plays a pivotal role in regulating stem cell behavior (33–35), including cell survival, migration, differentiation, and paracrine behavior. These factors affect the interaction of stem cells with other cells and the tissue microenvironment. Previously, different types of NO-releasing biomaterials, such as injectable hydrogels, have been prepared by us and other groups (36–40), and further studies have shown that the combination of NO and MSCs is more effective in treating various diseases than MSC therapy alone. In addition, it has been reported that pretreatment of MSCs with NO-releasing biomaterials could enhance the therapeutic efficacy of MSCs and their secreted exosomes because of their enhanced proangiogenic functions (41).

    Due to the spatiotemporal characteristics of NO (42), precise delivery of NO in a site-specific and controllable manner holds great importance in the regulatory effect of exogenously administered NO. In addition to the controlled release rate, the site at which NO is generated is also a key factor due to the relative half-life and limited diffusion distance (43, 44). It is reasonable to speculate that intracellular and extracellular NO delivery may lead to different outcomes when regulating the survival and function of MSCs. In our previous work, an enzyme-prodrug delivery system was designed based on a bump-and-hole strategy (21). The mutant galactosidase (A4-β-GalH363A) enables the targeted delivery of NO, thus reducing the side effects due to the unspecific decomposition of the NO prodrug and enhancing the therapeutic efficacy. Here, we transfected a plasmid expressing mutant galactosidase into MSCs and successfully constructed eMSCs. The enzyme expressed by MSCs could catalyze the decomposition of the 6-OMe-galactose–protected NO prodrug and release NO intracellularly.

    Western blotting and fluorescence imaging demonstrated that the expression‎ of the engineered enzyme was confined to the nucleus of MSCs, while wild-type β-galactosidase was widely distributed in the cytoplasm, including the lysosome and perinuclear region (45, 46). Since the corresponding prodrug for wild-type β-galactosidase is highly hydrophilic, it fails to enter MSCs and releases NO extracellularly by enzymes that translocate from the cell. In contrast, the prodrug for mutant galactosidase is cell penetrating because of the modified molecular structure; therefore, it can enter MSCs and release NO intracellularly under the catalysis of the corresponding enzyme expressed by the cells. Accordingly, two different NO delivery paradigms were successfully developed in this study and further confirmed by a series of eval‎uations, including cell imaging and electronic paramagnetic resonance. Further in vitro and in vivo assays indicated that in contrast to extracellular NO delivery, intracellular administration of NO enhanced cell survival and the paracrine effects of MSCs, including inhibiting apoptosis and supporting angiogenesis.

    Next, we established a mouse MI model to systematically eval‎uate the therapeutic efficacy of MSCs combined with exogenous NO. The results showed that intracellular delivery of NO prolonged the retention of eMSCs after myocardial orthotopic transplantation. In addition, the combination of eMSCs and intracellular NO delivery improved cardiac function after MI and reduced adverse ventricular remodeling compared to the group treated with MSCs only. Additionally, it could effectively restore the reconstruction of the blood vessel network and further promote the repair of the infarcted myocardium.

    To gain further insight into the translational potential of the combinatory therapeutic strategy developed in this study, a rat model of MI was established, and MSCs were administered by a second thoracotomy after 3 days to mimic the clinical use of MSCs for the treatment of MI (47, 48). Clinically, acute MI is typically due to the rupture of coronary atherosclerotic plaque and the formation of thrombus, which causes coronary artery obstruction. After the acute phase of MI, adverse ventricular remodeling further affects the prognosis of patients, which is specifically characterized as a decrease in ventricular wall thickness and myocardial tension in the MI area, myocardial hypertrophy in the noninfarction area, and a change in the morphology of the ventricular cavity, thus leading to arrhythmia and further development into heart failure. The efficacy of MSCs in managing arrhythmias remains a topic of ongoing debate. Some researchers argue that MSCs do not appear to reduce or prevent arrhythmias, with the antiarrhythmic or proarrhythmic potential of MSCs primarily relying on paracrine factors (49). Conversely, other studies suggest that MSCs themselves may play a role in the post-MI recovery process (50). In our study, we observed an evident up-regulation of Cx43 expression‎ after NO-eMSC treatment, which is a potential target associated with antiarrhythmic effects. Further investigation is still required to comprehensively explore the antiarrhythmic potential of NO-eMSCs. In line with the enhanced therapeutic efficacy in the mouse model, intracellular delivery of NO showed enormous advantages in the rat MI model by inhibiting apoptosis and enhancing the paracrine function of MSCs.

    In summary, we first showed that survival and paracrine function were reduced in MSCs collected from patients with diabetes, which could greatly affect therapeutic efficacy. Accordingly, eMSCs were successfully constructed, and the mutant β-galactosidase expressed by the cells enabled the intracellular generation of NO via the conversion of an exogenous NO prodrug. In vitro and in vivo assays indicated that intracellular delivery of NO effectively enhanced the survival of transplanted MSCs and promoted the paracrine function of MSCs, which was further confirmed by the enhanced therapeutic efficacy in mouse and rat models of MI compared to the group treated with MSCs only. This synergistic strategy provides an option for the treatment of MI by autologous MSCs in the clinic.

    MATERIALS AND METHODSRNA sequencing analysis

    RNA sequencing was performed by the BGI (Shenzhen, China). Briefly, RNA from the ADMSCs of healthy people and patients with diabetes was extracted using TRIzol reagent (Yeasen, China). RNA samples were sequenced on the BGISEQ platform. The raw data containing low-quality reads, adaptor sequences, and high levels of N bases were filtered before analysis. Then, the clean reads were mapped to the reference genome using HISAT, and Bowtie2 was used to align the clean reads to the reference genes. The reference genome source is National Center for Biotechnology Information (NCBI), and the reference genome version is GCF_000001405.39_GRCh38.p13. The expression‎ levels of genes were quantified to identify differentially expressed genes by RNA-Seq by expectation maximization (RSEM). The analyses of hierarchical clustering and heatmap were performed using the online Dr. Tom system (biosys.bgi.com) to compare differential gene expression‎ of ADMSCs in healthy people and patients with diabetes. According to the KEGG_pathway annotation classification, the phyper function in R software was used for enrichment analysis, the P value was calculated, and then false discovery rate (FDR) was performed on the P value to obtain a Q value. Generally, a Q value of ≤0.05 was regarded as significant enrichment. GSEA was used to analyze significant differences in gene expression‎ between inflammatory and apoptosis-related pathways. Expression‎ cluster heatmap was used to analyze the expression‎ of genes associated with angiogenesis.

    Measurement of NO release

    The NO-releasing profile was determined by the Griess kit assay. In brief, 50 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were dissolved in phosphate-buffered saline (PBS) buffer (pH 7.4), and enzymes were added to the solutions at a concentration of 0.005 mg/ml. At each predetermined time interval, 50 ml of solution was transferred into a 96-well plate, and 50 ml of Griess I and 50 ml of Griess II were added thereafter. The azo compound of purple color was formed, and the absorbance was measured at a wavelength of 540 nm using an iMark microplate reader (Bio-Rad, USA).

    Cell cultureMesenchymal stem cells

    MSCs derived from human umbilical cord were obtained from Health-Biotech, maintained in Dulbecco's modified Eagle’s medium (DMEM)/F12 (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, USA) and 1% penicillin- streptomycin solution (Gibco, USA), and placed in a cell culture chamber containing 5% CO2 at 37°C.

    Human embryonic kidney 293T cells

    Human embryonic kidney (HEK) 293T cells were obtained from the American Type Culture Collection (ATCC), maintained in high-glucose DMEM (Gibco, USA) with 10% FBS (Gibco, USA) and 1% penicillin-streptomycin solution (Gibco, USA), and placed in a cell culture chamber containing 5% CO2 at 37°C.

    Construction of eMSCs

    The coding sequence of mutant β-galactosidase (A4-β-GalH363A) can be obtained from the previous publication (21). The lentivirus packaging system containing A4-β-GalH363A sequence and Rluc-RFP sequence was constructed by Wuhan Miaolingbio Co. Ltd. The constructed lentivirus plasmid containing the target gene and the package gene (psPAX2 and pMD2.G) was transfected into HEK 293T cells through Lipo2000, and the supernatant was collected to obtain the virus solution. After removing impurities, the virus solution was mixed 1:1 with fresh MSC medium, and polybrene (10 μg/ml) was added. MSCs were infected with virus through incubation in the mixture medium. The infection efficiency was observed under an inverted fluorescence microscope, and the expression‎ of target protein was determined by Western blotting.

    Cell immunofluorescence staining

    eMSCs were inoculated in 24-well plates. Cells were fixed with 4% paraformaldehyde and blocked in 4% bovine serum albumin in PBS for 30 min at room temperature. Then, the cells were incubated with primary antibodies overnight at 4°C. The bound primary antibodies were displayed by incubation with the secondary antibodies for 2 hours at room temperature. Cells were counterstained with DAPI (4′,6-diamidino-2-phenylindole)–containing Fluoromount-G and visualized with a fluorescence microscope (Axio Imager Z1). Antibodies used include anti–β-galactosidase (1:100, A1863, Abclonal) and anti-RFP (1:100, PA1-986, Invitrogen).

    Western blot

    eMSCs were collected, and total protein was extracted using radioimmunoprecipitation assay (RIPA) lysate containing protease inhibitor (Solarbio, China). Cytoplasmic protein and nucleoprotein were extracted using a nucleoprotein extraction kit containing protease inhibitors (Solarbio, China). The protein concentration was quantified using a BCA protein assay kit (Solarbio, China). The samples were diluted with 4× SDS–polyacrylamide gel electrophoresis (PAGE) loading buffer and boiled in boiling water for 8 min. Then, 30 μg of protein was isolated by 10% SDS-PAGE electrophoresis. The isolated proteins were transferred to an Immobilon-P Transfer membrane (Millipore, USA) and incubated with the primary antibody overnight at 4°C and then with the secondary antibody at room temperature for 2 hours. The bands were detected with chemiluminescent horseradish peroxidase substrate (Millipore, USA). Signals were generated by using an enhanced chemiluminescence (ECL) reagent (Millipore, USA) and were captured by using the Tanon-5200 Chemiluminescence Imaging System (Tanon, China). The antibodies used included anti–β-galactosidase (1:1000, A1863, Abclonal), anti–His-tag (1:1000, 12698S, Cell Signaling Technology), anti–glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (1:1000, AC001, Abclonal), anti–histone H3 (1:1000, 4499S, Cell Signaling Technology), anti–β-actin (1:1000, UM4001, Utibody), anti-BCL2 (1:1000, WL01556, Wanleibio), anti-Bax (1:1000, 50599-2-lg, Proteintech), anti-Bad (1:1000, WL02140, Wanleibio), and anti-caspase/cleaved caspase3 (1:1000, WL02117, Wanleibio).

    6-OMeGal-Ph-NO uptake by eMSCs

    eMSCs were inoculated in 12-well plates, and 50 μM substrate (6-OMeGal-Ph-NO) was added per well. At predetermined time points (0, 1, 3, 6, and 12 hours), an appropriate amount of culture medium was collected, and an excess of A4-β-GalH363A was added immediately to fully catalyze the decomposition of the remaining substrate at room temperature. The amount of 6-OMeGal-Ph-NO substrate in the culture medium was determined by Griess kit assay.

    Real-time imaging of intracellular NO

    The fluorescence emission associated with NO in the cytosol was detected by an electron-multiplying charge-coupled device (DU-897D-CS0-BV; Andor, Belfast, UK) connected to an inverted fluorescence microscope (Axio Observer D1; Carl Zeiss, Oberkochen, Germany). Intracellular NO imaging was performed using an NO fluorescence probe, DAF-AM DA (Beyotime, China), according to the manufacturer’s instruction. eMSCs were inoculated in small confocal dishes in advance, and the experiment was conducted when the cell density reached 80%. First, the medium was collected for later use. After two gentle washes with PBS, 5 μM DAF-AM DA solution was incubated at 37°C for 30 min in the dark. Then, cells were gently washed with PBS twice. The previously collected medium was added anew, and cell imaging was performed immediately. At the 488-nm excitation wavelength, pictures were taken every 5 s. After stable shooting for 2 min, 100 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were added. The filming for cell fluorescence was continued for 4 min. The final fluorescence intensity was determined without the background fluorescence value. The proportion of change in fluorescence intensity of each cell in the visual field was calculated.

    Intracellular NO detection

    Intracellular NO radicals (NO•) were detected using EPR as described (51, 52). In brief, sodium DETC (4.5 mg) and FeSO4•7H2O were dissolved in two separate volumes (10 μl) of deoxygenated Krebs/Hepes solution. Equal volumes of these parent solutions were rapidly mixed and aspirated into Eppendorf combi tips. The 0.5 mM Fe•(DETC)2 colloid solution had a yellow-brownish color with a slight opalescence in light. No aggregate formation was observed, at least during the first 30 min. eMSCs were rinsed with modified Krebs/Hepes buffer and incubated with freshly prepared NO•-specific spin trap Fe•(DETC)2 colloid (0.5 mM) for 30 min. Meanwhile, 100 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were added to the buffer. Gently collected cell suspensions were snap-frozen in liquid nitrogen. Ethyl acetate (200 μl) was added, and the cells were ultrasonically broken to extract DETC2-Fe-NO. The ethyl acetate extract was concentrated with nitrogen and transferred to a 50-μl capillary, and then the X-band EPR was measured at room temperature. The following acquisition parameters were used: modulation frequency, 100 kHz; microwave power, 10 mW; modulation amplitude, 2 G; number of scans, 60. The double-integrated area of the EPR spectra was calibrated into concentrations of DETC2-Fe-NO using TEMPO (2,2,5,5-tetramethyl piperidine 1-oxyl) as a standard. EPR spectral simulation was conducted by the WINSIM program.

    Extracellular NO detection

    eMSCs were treated with β-Gal-NO or 6-OMeGal-Ph-NO (30 μM). The production of NO in the medium of each group was detected 6 hours after incubation with NO-sensitive near-infrared fluorescence probe (5 μM). The NO production of medium in different groups was compared by the relative fluorescence intensity under the excitation at 750 nm (emission at 800 nm).

    Cell apoptosis detection

    To test the protective effect of NO delivery on cellular oxidative stress stimulation, 30 μM NO substrate (6-OMeGal-Ph-NO) was added to the medium in advance. Then, H2O2 with different concentrations (100, 200, 400, and 600 μM) was added to stimulate the lentivirus-infected eMSCs. BLI was performed immediately after addition of the luciferase substrate coelenterin to eval‎uate cell apoptosis. Additionally, eMSCs treated with 200 mM H2O2 were stimulated for 24 hours to induce cell apoptosis. An Annexin V/PI assay kit (Solarbio) was used to detect eMSC apoptosis.

    BLI detection of cell retention

    BLI and luciferase substrates were used in mice to eval‎uate the retention of NO-eMSCsGluc/RFP in cardiac orthotopic transplantation. The mice after eMSC injections were anesthetized with 1.5% isoflurane and injected with coelenterin through the caudal vein at 150 mg/kg. After injection, the mice were immediately placed in a BLI system to detect cell retention in the myocardium.

    Animals

    C57BL/6 mice (male, 8 weeks old) and Sprague-Dawley rats (male, 8 weeks old) were purchased from SPF (Beijing) Biotechnology Co. Ltd., Beijing, China. Animals were randomly grouped for treated and untreated controls. All experiments and animal procedures were approved by the Animal Experiments Ethical Committee of Nankai University and carried out in conformity with the Guide for Care and Use of Laboratory Animals.

    MI in mice and rats

    Surgical induction of MI was performed on C57BL/6 mice (male, 8 weeks old) as previously described with some modifications. Briefly, mice were anesthetized with 2% isoflurane, followed by fixation to a heating pad (37°C) at supine position, and then ventilated with a mechanical ventilation system (Hallowell EMC Microvent I, USA) set at 110 breaths per minute with a tidal volume of 0.2 to 0.3 ml. The third intercostal space over the left chest and the heart was exposed. After left thoracotomy, the left anterior descending coronary artery was ligated with a 6-0 silk ligature. The left main descending coronary artery was sutured and tied with a slip knot at a site approximately 3 mm from its origin. Cardiac ischemia was confirmed by the presence of myocardial blanching. Immediately, eMSCs encapsulated with HA hydrogel were injected into the myocardium of mice through three-point injection around the infarct zone. The chest cavity was closed to restore negative pressure and prevent pneumothorax.

    Similar MI surgery was performed on Sprague-Dawley rats (male, 8 weeks old) first. Briefly, rats were anesthetized via intraperitoneal injection of 10% chloral hydrate (350 mg/kg), followed by fixation to a heating pad (37°C) at supine position. Then, they were ventilated with a mechanical ventilation system (Hallowell EMC Microvent I, USA) set at 110 breaths per minute with a tidal volume of 6 ml. The third intercostal space over the left chest and the heart was exposed. After left thoracotomy, the left anterior descending coronary artery was ligated with a 6-0 silk ligature. The left main descending coronary artery was sutured and tied with a slip knot at a site approximately 3 mm from its origin. Cardiac ischemia was confirmed by the presence of myocardial blanching. The chest cavity was closed to restore negative pressure and prevent pneumothorax. Three days after surgery, secondary thoracotomy was performed, and eMSCs were injected into the myocardium around the infarct zone. The chest cavity was closed to restore negative pressure and prevent pneumothorax.

    Mice and rats in the AMI group only received MI without eMSC injection, while sham-operated mice only experienced thoracotomy without MI surgery or eMSC injection.

    At 1, 3, 5, and 7 days after myocardial injection of eMSCs, the prodrug was injected through the tail vein. Mice were injected with 100 μl of prodrug (1 mg/ml) each time, and rats were injected with 200 μl of prodrug (1 mg/ml) each time.

    TTC staining

    Two days after surgery, a thoracotomy was performed. The heart was quickly excised after quick freezing for 15 min and sliced at 1 mm thickness. Afterward, the sections were incubated with 1.5% TTC (Sigma-Aldrich) solution at 37°C in an incubator for 15 to 30 min and then with a 4% formaldehyde solution for 2 hours. The normal myocardial tissue was red, while the ischemic myocardium was white. The size of the infarcted myocardium, which was white or pale, was measured by ImageJ software.

    Cardiac function assessment

    Transthoracic echocardiography was performed with the Vevo 2100 Imaging System (Fuji Film Visual Sonics Inc., Canada) equipped with an MS-250/400 imaging transducer. The baseline cardiac function of mice and rats was measured at 3 days before surgery. Cardiac function was analyzed at days 1 and 28 after MI surgery with different treatments, as reported previously. Mice or rats were slightly anesthetized in a box with isoflurane. Their limbs were fixed in a supine position on the echo mat, and the chest hair was removed by depilating cream. Then, mice or rats were anesthetized by inhalation of isoflurane (0.5 to 1%) mixed with oxygen to maintain the heart rate at approximately 500 to 600, and M-mode echocardiography was performed. The left ventricular internal diameter at end-diastole (LVIDd) and systole (LVIDs) were obtained by measuring the long axis and the short axis. Accordingly, the cardiac parameters LV-EF, LV-FS, LV-EDV, and LV end-systole volume (LV-ESV) were determined. The echocardiography measurement was carried out in a double-blind manner.

    Histological analysis

    At the indicated time points, mice and rats were anesthetized via intraperitoneal injection of chloral hydrate, and a thoracotomy was performed. The hearts were fixed with trans-cardiac perfusion of saline and immersed in 4% paraformaldehyde over 24 hours. The heart tissue samples were dehydrated with gradient alcohol and xylene, embedded in paraffin blocks, and cut into sections in 5 μm thickness.

    The paraffin-embedded sections were stained with Masson trichrome, H&E, and Sirius Red following a standard protocol. Immunofluorescence staining was performed on paraffin-embedded sections of the heart tissue samples. After deparaffinization and heat-mediated antigen retrieval‎ in citrate solution, the samples were washed with PBS three times and incubated with blocking serum, which was used to avoid nonspecific binding, at room temperature for 30 min. The sections were incubated with specific antibodies diluted in goat serum at 4°C overnight. On the second day, the sections were rewarmed at room temperature for 1 hour and washed with PBS three times. Afterward, the sections were incubated with Alexa Fluor–coupled secondary antibodies for 2 hours at room temperature. After washing with PBS, the sections were counterstained with DAPI-containing Fluoromount-G (SouthernBiotech, USA) and coverslipped. The antibodies used included anti–α-SMA (1:100, ab5694, Abcam), anti-vWF (1:100, ab6694, Abcam), anti–α-actinin (1:100, ab9475, Abcam), anti-Connexin43 (1:100, ab11370, Abcam), anti-CD68 (1:100, ab125212, Abcam), WGA (1:500, FL-1021, Novus Biologicals), anti-iNOS (1:100, ab178945, Abcam), and anti-CD206 (1:100, ab64693, Abcam).

    Macrophage isolation and detection

    Three days before euthanasia, mice were intraperitoneally injected with 2 ml of 4% thioglycolate. Three days later, the mice were sacrificed by cervical dislocation and immersed in 75% alcohol and then transferred to an ultraclean workbench. The mouse limb was fixed in the supine position, and the mouse abdominal wall was carefully cut open with the peritoneal. PBS [1% penicillin-streptomycin (PS)] was injected intraperitoneally to collect the cell suspension, which was centrifuged at 2000 rpm for 10 min. After discarding the supernatant, the cells were incubated with anti-F4/80/TNF-α and anti-F4/80/CD206 antibodies. FlowJo software was used to analyze the results of flow cytometry.

    Quantitative real-time PCR

    Total RNA samples from the cells were prepared using TRIeasy Total RNA Extraction Reagent (Yeasen, China) according to the manufacturer’s instructions. Heart tissue samples were collected at the indicated time points after MI surgery.

    The tissue samples were dissected at the border zone of the left ventricle and frozen in liquid nitrogen immediately. Afterward, the total RNA was extracted with TRIzol reagent, as mentioned before. The concentration of the RNA was measured with a NanoDrop spectrophotometer (NanoDrop Technologies, USA). The complementary cDNA was synthesized using a first-strand cDNA synthesis kit (Yeasen, China). Quantitative real-time PCR was performed on a CFX96 real-time PCR system (Bio-Rad, Hercules, USA) with an SYBR Green–based real-time detection system (Yeasen, China). The relative gene expression‎ of mRNA was expressed as 2−(△△CT) and normalized to GAPDH as an internal control. Each reaction was performed in triplicate to obtain an average value, and the changes in relative gene expression‎ normalized to the internal control levels were determined. The highly purified primers used in this experiment were commercially synthesized (Sango, China). The sequences of the primers used in this experiment are summarized in the Supplementary Materials.

    Statistics

    All data are presented as the mean ± SEM from at least three independent experiments. Comparisons between two groups were performed by Student’s t test, and comparisons among more than two groups were performed by one-way or two-way analysis of variance (ANOVA). Statistical analyses were performed with GraphPad Prism software 7.0, and a statistical significance level of less than 0.05 was accepted.

    Acknowledgments

    Funding: This study is supported by the National Key R&D Program of China (2018YFE0200503), the National Natural Science Foundation of China (nos. 81925021, 82330066, 81921004, and U2004126), and the Tianjin Natural Science Foundation (21JCZDJC00240).

    Author contributions: Q.Z. and Z.L. conceived the original concept and initiated this project. Q.Z., Z.L., and F.G. designed the experiment and supervised the entire project. S.W. collected human adipose mesenchymal stem cells. M.Q. synthesized all NO prodrugs and probes. P.L. prepared engineered enzymes. T.H., G.J., and Q.X.L. established mouse and rat MI models. T.H. and G.J. performed histological analysis. G.J. and W.D. carried out in vitro cell experiments. S.D. carried out NO cell imaging under the supervision of L.P. T.H., G.J., and M.Q. analyzed data under the supervision of Q.Z. H.H. helped with lentivirus packaging and cell infection. W.G. and T.L. helped in establishing animal MI models. Y.W., J.H., J.C., and J.T. helped with data collection. T.H. and Q.Z. wrote the paper with input from other authors.

    Competing interests: The authors declare that they have no competing interests.

    Data and materials availability: All data needed to eval‎uate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Transcriptome sequencing dataset is available at https://doi.org/10.5061/dryad.tqjq2bw5b.

    Supplementary MaterialsThis PDF file includes:

    Supplementary Text

    Figs. S1 to S9

    Table S1

    • Download
    • 1.94 MB

    REFERENCES AND NOTES

    1

    M. F. Pittenger, A. M. Mackay, S. C. Beck, R. K. Jaiswal, R. Douglas, J. D. Mosca, M. A. Moorman, D. W. Simonetti, S. Craig, D. R. Marshak, Multilineage potential of adult human mesenchymal stem cells. Science 284, 143–147 (1999).

    GO TO REFERENCE

    Crossref

    PubMed

    Web of Science

    Google Scholar

    2

    A. Uccelli, L. Moretta, V. Pistoia, Mesenchymal stem cells in health and disease. Nat. Rev. Immunol. 8, 726–736 (2008).

    GO TO REFERENCE

    Crossref

    PubMed

    Web of Science

    Google Scholar

    3

    Y. Wang, J. Fang, B. Liu, C. Shao, Y. Shi, Reciprocal regulation of mesenchymal stem cells and immune responses. Cell Stem Cell 29, 1515–1530 (2022).

    GO TO REFERENCE

    Crossref

    PubMed

    Web of Science

    Google Scholar

    4

    Y. Wang, X. Chen, W. Cao, Y. Shi, Plasticity of mesenchymal stem cells in immunomodulation: Pathological and therapeutic implications. Nat. Immunol. 15, 1009–1016 (2014).

    GO TO REFERENCE

    Crossref

    PubMed

    Web of Science

    Google Scholar

    Show all references

    (0)eLetters

    eLetters is a forum for ongoing peer review. eLetters are not edited, proofread, or indexed, but they are screened. eLetters should provide substantive and scholarly commentary on the article. Neither embedded figures nor equations with special characters can be submitted, and we discourage the use of figures and equations within eLetters in general. If a figure or equation is essential, please include within the text of the eLetter a link to the figure, equation, or full text with special characters at a public repository with versioning, such as Zenodo. Please read our Terms of Service before submitting an eLetter.

    Log In to Submit a Response

    No eLetters have been published for this article yet.

    Recommended articles from TrendMD

    • In vivo priming of human mesenchymal stem cells with hepatocyte growth factor–engineered mesenchymal stem cells promotes therapeutic potential for cardiac repai...

      Bong-Woo Park, Sci Adv, 2020

    • Shattering barriers toward clinically meaningful MSC therapies

      Oren Levy, Sci Adv, 2020

    • Nanovesicles derived from iron oxide nanoparticles–incorporated mesenchymal stem cells for cardiac repair

      Ju-Ro Lee, Sci Adv, 2020

    • Exendin-4 gene modification and microscaffold encapsulation promote self-persistence and antidiabetic activity of MSCs

      Yuanyuan Zhang, Sci Adv, 2021

    • Human gingival mesenchymal stem cells retain their growth and immunomodulatory characteristics independent of donor age

      Jay R. Dave, Sci Adv, 2022

    • The secretion profile of mesenchymal stem cells and potential applications in treating human diseases

      Yuyi Han, Signal Transduction and Targeted Therapy, 2022

    • Mesenchymal Stem Cell Therapy for Ischemic Heart Disease: Systematic Review and Meta-analysis

      Hyunsuk Jeong, International Journal of Stem Cells

    • Mesenchymal Stem Cell and MicroRNA Therapy of Musculoskeletal Diseases

      International Journal of Stem Cells

    • Excessive branched-chain amino acid accumulation restricts mesenchymal stem cell-based therapy efficacy in myocardial infarction

      Fuyang Zhang, Signal Transduction and Targeted Therapy, 2022

      Intracellular delivery of nitric oxide enhances the therapeutic efficacy of mesenchymal stem cells for myocardial infarction

      Tian Hao https://orcid.org/0009-0007-8657-0424, Guangbo Ji https://orcid.org/0009-0000-7079-8896, Meng Qian https://orcid.org/0009-0006-9135-4593, Qiu Xuan Li, Haoyan Huang, Shiyu Deng, Pei Liu, Weiliang Deng https://orcid.org/0009-0001-8402-7280, Yongzhen Wei, [...], and Qiang Zhao https://orcid.org/0000-0003-4656-6002 +9 authorsAuthors Info & Affiliations

      Science Advances

      29 Nov 2023

      Vol 9, Issue 48

      DOI: 10.1126/sciadv.adi9967

      4,3612

      Metrics

      Total Downloads4,361

      • Last 6 Months1,270
      • Last 12 Months3,995

      Total Citations2

      • Last 6 Months1
      • Last 12 Months2

      NotificationsBookmark

      • • Abstract
        • INTRODUCTION
        • RESULTS
        • DISCUSSION
        • MATERIALS AND METHODS
        • Acknowledgments
        • Supplementary Materials
        • REFERENCES AND NOTES
        • eLetters (0)
      • • Information & Authors
        • Metrics & Citations
        • View Options
        • References
        • Media
        • Tables
        • Share

      Abstract

      Cell therapy by autologous mesenchymal stem cells (MSCs) is a clinically acceptable strategy for treating various diseases. Unfortunately, the therapeutic efficacy is largely affected by the low quality of MSCs collected from patients. Here, we showed that the gene expression‎ of MSCs from patients with diabetes was differentially regulated compared to that of MSCs from healthy controls. Then, MSCs were genetically engineered to catalyze an NO prodrug to release NO intracellularly. Compared to extracellular NO conversion, intracellular NO delivery effectively prolonged survival and enhanced the paracrine function of MSCs, as demonstrated by in vitro and in vivo assays. The enhanced therapeutic efficacy of engineered MSCs combined with intracellular NO delivery was further confirmed in mouse and rat models of myocardial infarction, and a clinically relevant cell administration paradigm through secondary thoracotomy has been attempted.

      SIGN UP FOR THE AWARD-WINNING SCIENCEADVISER NEWSLETTER

      The latest news, commentary, and research, free to your inbox daily

      Sign up

      INTRODUCTION

      Mesenchymal stem cells (MSCs) are pluripotent stem cells with high self-renewal abilities and multidirectional differentiation potential (1, 2). They are extensively distributed throughout the body and serve a variety of purposes, including tissue regeneration (3), immunoregulation (4), and angiogenesis (5). However, an important challenge for stem cell therapy is the low survival rate of stem cells after transplantation, which is associated with nonspecific homing of cells and ischemia/hypoxia at the injury site (6). In addition, transplanted stem cells cannot fully exert their paracrine effects in pathological environments, which seriously limits the clinical application of stem cells.

      The proangiogenic function of MSCs is a key factor contributing to the treatment of ischemic diseases. Generally, MSCs can stimulate local angiogenesis in ischemic tissue by secreting cytokines, including vascular endothelial growth factor (VEGF), transforming growth factor β (TGFβ), and basic fibroblast growth factor (bFGF), to induce endothelial cells to form tubular structures (7). In addition, MSCs can recruit pericytes and smooth muscle cells to promote neovascularization (8, 9). Recently, MSCs have been reported to participate in the construction of blood vessels by directly differentiating into endothelial or smooth muscle cells (10).

      As an important signaling molecule, nitric oxide (NO) plays a crucial role in the maintenance of vascular homeostasis by inhibiting thrombus formation and promoting angiogenesis (11, 12). Recently, increasing attention has been given to the regulation of stem cells by NO due to its multiple biological functions (13, 14). NO can regulate the proliferation of stem cells by regulating the activities of cyclin and mitotic receptors, as well as their downstream pathways (15, 16). On the other hand, NO can regulate the expression‎ of angiogenic cytokines and immunomodulatory factors to improve the paracrine performance of stem cells (17). Additionally, recent studies have demonstrated that NO can regulate the differentiation behavior of stem cells through the phosphatidylinositol 3-kinase (PI3K)/AKT, guanosine 3′,5′-monophosphate (cGMP), and other signaling pathways (18).

      As a result, NO-releasing biomaterials have been used as delivery carriers for stem cells to enhance their survival and regulate paracrine functions (19). However, NO, which is a gaseous molecule, easily diffuses and has a high level of instability. Furthermore, the physiologic function of NO is dose dependent, and an overdose of NO often leads to notable cytotoxicity (20). Thus, optimizing the beneficial effects of NO to strengthen the therapeutic efficacy of stem cells by tuning their release profile should be taken into account.

      In our previous study, based on the chemical biology principle of “bump-and-hole,” we designed and prepared an enzyme-prodrug delivery system and achieved targeted delivery of NO at the lesion site in two different ischemic disease models (21). Here, MSCs were further modified by gene transfection to express a catalytic enzyme (A4-β-GalH363A). The engineered MSCs (eMSCs) were transplanted using an injectable hyaluronic acid (HA) hydrogel as the carrier, while the NO prodrug was injected through the tail vein to achieve controlled release of NO catalyzed by the enzymes in eMSCs. The therapeutic efficacy of MSCs combined with exogenous NO delivery was eval‎uated in mouse and rat models of myocardial infarction (MI) with an emphasis on comparing the therapeutic efficacy of two different NO administration methods (intracellular or extracellular), and the underlying mechanism of their myocardial protective effect was further explored.

      RESULTSThe gene expression‎ of MSCs collected from patients is differentially regulated

      In the clinic, the therapeutic efficacy of autologous stem cell transplantation is largely affected by the low quality (including cell survival and paracrine function) of stem cells collected from patients due to chronic diseases. Diabetes mellitus is a chronic medical condition that can lead to a variety of complications, and these complications can affect various parts of the body, such as the kidney, lower limb, and heart. Diabetes predisposes affected individuals to a spectrum of cardiovascular complications, and one of the most debilitating in terms of prognosis is heart failure (22).

      Accordingly, although autologous MSCs have been widely accepted as a promising strategy for treating various complications associated with diabetes, the therapeutic efficacy is largely affected by the quality of stem cells collected from the patients themselves. The gene levels in adipose-derived MSCs (ADMSCs) from patients with diabetes and healthy individuals were first compared by transcriptome sequencing. The heatmap shows that multiple genes in MSCs from patients with diabetes were up- or down-regulated compared to healthy controls (Fig. 1A). Gene ontology (GO) enrichment analysis revealed that the most differentially up-regulated genes were related to tumor necrosis factor (TNF) signaling pathways (Fig. 1B). Subsequently, we performed gene set enrichment analysis (GSEA) based on the RNA-sequencing results. GSEA revealed that inflammatory target genes were highly enriched in MSCs from patients with diabetes, with a normalized enrichment score (NES) of 1.65 (P < 0.01) (Fig. 1C). Apoptosis-related genes were also highly enriched, with an NES of 1.77 (P < 0.01) (Fig. 1D). In addition, through enrichment analysis, we found that proangiogenic genes in MSCs were greatly down-regulated in patients with diabetes (Fig. 1E).

      

      Fig. 1. Transcriptome sequencing of adipose mesenchymal stem cells (MSCs) collected from patients with diabetes and healthy controls.

      (A) Heatmap showing the differentially expressed genes of MSCs from patients with diabetes and healthy controls (n = 3). (B) Gene ontology (GO) analysis of the up-regulated transcriptome of MSCs from patients with diabetes. (C and D) Gene set enrichment analysis (GSEA) was performed to determine the enrichment of inflammation (C) and apoptosis (D) target genes in the diabetic group. (E) Heatmap showing angiogenesis-related genes in the two groups (n = 3). (F) Schematic illustration demonstrating the difference in MSCs between patients with diabetes and healthy controls at the gene level.

      Open in viewer

      eMSCs are constructed to express mutant β-galactosidase

      Here, we first designed and constructed plasmids carrying the mutant β-galactosidase (A4-β-GalH363A) target gene and luciferase–red fluorescent protein (RFP) dual reporter genes, which could be further used for in vivo imaging. eMSCs expressing A4-β-GalH363A were constructed by infecting MSCs with lentiviruses obtained from human embryonic kidney 293T cells (Fig. 2A). Immunofluorescence staining for RFP confirmed that A4-β-GalH363A was successfully expressed by eMSCs (Fig. 2B). The subcellular fraction and intracellular distribution of enzymes expressed by the eMSCs was determined by Western blotting (Fig. 2C). In contrast to natural β-galactosidases, which are widely distributed within cells, A4-β-GalH363A was mainly confined to the nucleus of eMSCs.

      

      Fig. 2. Intracellular expression‎ and localization of A4-β-GalH363A.

      (A) Schematic diagram of lentivirus packaging and mesenchymal stem cell (MSC) infection. (B) Immunofluorescence staining of β-Gal and A4-β-GalH363A in eMSCs. Scale bar, 50 μm. (C) The distribution of two different enzymes in engineered MSCs (eMSCs) was analyzed by Western blotting.

      Open in viewer

      Intracellular release of NO is realized via decomposition of the 6-OMeGal-Ph-NO prodrug

      Since the mutant β-galactosidase was confined to the nucleus in eMSCs, we designed a prodrug by introducing a lipid-soluble self-decomposition chain into 6-OMeGal-NO to improve its oil and water distribution coefficient; therefore, the resultant NO donor 6-OMeGal-Ph-NO could penetrate the cell membrane and decompose and release NO under the catalysis of A4-β-GalH363A (Fig. 3A and Supplementary Materials). An in vitro release assay showed that the 6-OMeGal-Ph-NO prodrug was efficiently recognized and converted by A4-β-GalH363A with a cumulative release ratio of approximately 97.3%, while nearly no release was observed in the presence of wild-type β-galactosidase (Fig. 3B). To confirm‎ intracellular NO release, eMSCs were preincubated with an NO fluorescent probe (DAF-AM DA) and then treated with different NO prodrugs. The changes in fluorescence signals were examined by time-delay cell imaging (Fig. 3C). The results indicated that the fluorescence intensity continuously increased in the group that was treated with 6-OMeGal-Ph-NO, indicating conversion into NO (Fig. 3D). In contrast, no detectable changes were identified in the β-Gal-NO group because the high water solubility restricted its entry into eMSCs. Next, intracellular and extracellular release of NO in eMSCs was assessed (Fig. 3E). The quantity of intracellular NO was measured by electron paramagnetic resonance (EPR) using ferrous N-diethyl dithiocarbamate (DETC2-Fe) as the spin-trapping reagent. The resultant NO adduct (DETC2-Fe-NO) exhibited a characteristic triplet EPR signal (aN = 13.06 G, giso = 2.041) at room temperature. Quantitative analysis showed that the NO level was significantly (P < 0.001 or 0.0001) higher in eMSCs treated with 6-OMeGal-Ph-NO than in the β-Gal-NO and control groups (Fig. 3F). Furthermore, in the group treated with β-Gal-NO, the release of NO was mainly catalyzed by β-galactosidase that translocated from the cytoplasm in eMSCs, and the extracellular release profile of β-Gal-NO was confirmed by detecting the NO level in the cell culture medium with the NO-sensitive near-infrared fluorescence probe (23); it was significantly (P < 0.01) higher than that in the 6-OMeGal-Ph-NO and control groups (Fig. 3G). To determine the uptake of the NO prodrug by eMSCs, we incubated 6-OMeGal-Ph-NO with eMSCs, and the concentration in the culture medium was determined at different time points. The results reflected that approximately 45% of 6-OMeGal-Ph-NO was incorporated into the eMSCs within 12 hours (fig. S1).

      

      Fig. 3. Intracellular generation of nitric oxide (NO) from the NO prodrug under the catalysis of A4-β-GalH363A expressed by engineered mesenchymal stem cells (eMSCs).

      (A) Synthesis of two NO prodrugs with different enzyme response abilities and cellular permeabilities. (B) In vitro release profile of NO from the NO prodrug (6-OMeGal-Ph-NO) in the presence of β-Gal or A4-β-GalH363A. (C) Schematic illustration of intracellular NO imaging by using an NO fluorescence probe (DAF-AM DA). (D) Representative time-lapse images of NO generation from two different prodrugs in eMSCs and quantification of the fluorescence intensity (n = 6). ***P < 0.001, ****P < 0.0001 versus 6-OMeGal-Ph-NO group. (E) Schematic illustration showing the detection of intracellular and extracellular NO generation differentially. (F) Representative electron paramagnetic resonance (EPR) spectra and quantification of intracellular NO generation by measuring the DETC2-Fe-NO complex using 2,2,5,5-tetramethyl piperidine 1-oxyl (TEMPO) as a standard (n = 3). (G) Relative quantification of NO production in the medium determined using the near-infrared fluorescence probe (n = 4). Data are expressed as the mean ± SEM. Significant differences were detected by one-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

      Expand for more

      Open in viewer

      Intracellular delivery of NO inhibits apoptosis and enhances the proangiogenic activity of eMSCs

      Overproduction of reactive oxygen species due to cellular oxidative stress has been accepted as an important factor contributing to apoptosis in transplanted cells in ischemic tissue. Therefore, we assessed the protective effect of exogenously administered NO on the survival of eMSCs with H2O2-induced oxidative stress and focused on comparing the protection provided by intracellular and extracellular NO administration. The results showed that H2O2 stimulated apoptosis, and delivery of NO via extracellular and intracellular strategies significantly (P < 0.01 or 0.001) reduced apoptosis in eMSCs stimulated by oxidative stress, and the highest fluorescence signal was observed in response to intracellular NO delivery (Fig. 4A).

      

      Fig. 4. Intracellular delivery of nitric oxide (NO) inhibits apoptosis of engineered mesenchymal stem cells (eMSCs).

      (A) Bioluminescence imaging (BLI) was used to detect the effect of NO delivery on cell apoptosis stimulated by different concentrations of H2O2, and the fluorescence signals were further quantified (n = 3). **P < 0.01, ***P < 0.001, ****P < 0.0001 versus control group; #P < 0.05 versus β-Gal-NO group. (B) Flow cytometry assay of cell viability and apoptosis of eMSCs after H2O2 stimulation, and quantification of mean percent values of apoptotic cells (n = 3). (C) The expression‎ of apoptosis-related protein (BCL2, Bax, Bad, caspase3, and cleaved caspase3) by eMSCs was detected after H2O2 stimulation by Western blots (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001. (D) Representative BLI photographs reflecting the retention of eMSCs with and without intracellular NO delivery after in vivo transplantation as well as the quantitative analysis of signals (n = 3). Data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01.

      Open in viewer

      The survival of eMSCs after treatment with various NO delivery strategies was compared by using flow cytometry. Intracellular NO delivery had the most remarkable antiapoptotic effect on cells stimulated by oxidative stress (P < 0.001) (Fig. 4B). The expression‎ of apoptosis-related genes by eMSCs followed a similar trend; NO delivery (extracellular and intracellular) effectively promoted the expression‎ of the antiapoptotic gene BCL2 in eMSCs after oxidative stress stimulation, while the expression‎ levels of proapoptotic genes were reduced accordingly. Intracellular NO delivery via administration of the 6-OMeGal-Ph-NO prodrug exerted a more pronounced antiapoptotic effect at both the gene and protein levels (Fig. 4C and fig. S2A), which may be because the intracellular generation of NO directly activated the antioxidant system in cells to resist oxidative stress damage and inhibit further apoptosis.

      The expression‎ of proangiogenic genes, including ANGPT1, ANGPT2, FGF2, VEGFA, and KDR, in eMSCs was further detected by reverse transcription polymerase chain reaction (RT-PCR). The results showed that the expression‎ level of proangiogenic genes was significantly (P < 0.05, 0.01, or 0.001) higher in eMSCs treated with the 6-OMeGal-Ph-NO prodrug than in the other groups, indicating the enhanced proangiogenic functions of eMSCs after intracellular NO delivery (fig. S2B).

      We further eval‎uated the effect of NO delivery on the in vivo retention of eMSCs after orthotopic transplantation in the myocardial tissue of mice. As shown in Fig. 4D, intracellular delivery of NO effectively prolonged the retention of eMSCs within the myocardium, and an evident bioluminescence imaging (BLI) signal corresponding to the retention of eMSCs was observed 7 days after transplantation compared to the counterpart without administration of the NO prodrug. To further eval‎uate the translational potential of eMSCs in clinical settings, we used MSCs derived from diabetic patients and conducted a series of assays related to cell survival and paracrine function. The findings indicated that intracellular delivery of NO also confers advantages in the attenuation of cell apoptosis under stress conditions, thereby prolonging the in vivo retention of eMSCs (fig. S3).

      Intracellular delivery of NO ameliorates myocardial injury in MI mice after treatment with eMSCs

      The therapeutic efficacy of MSCs combined with exogenous NO was further eval‎uated in a mouse MI model (Fig. 5A and fig. S4). The inflammatory response in the early stage (3 days) was first detected by hematoxylin-eosin (H&E) staining and CD68 immunofluorescence staining. The results demonstrated that severe inflammatory cell infiltration occurred in the injured myocardium of MI mice, and it was effectively alleviated after eMSC treatment. More prominent restoration in the injured myocardium was observed after further administration of NO (Fig. 5, B and C), confirm‎ing the inhibitory effect on inflammation after MI provided by the combination of MSCs and NO. Moreover, this inhibitory effect was more significant (P < 0.01 or 0.0001) in the group with intracellular NO delivery than in the group with extracellular NO delivery.

      

      Fig. 5. Intracellular delivery of nitric oxide (NO) ameliorates myocardial injury in myocardial infarction (MI) mice after treatment with engineered mesenchymal stem cells (eMSCs) via regulation of the inflammation and stimulation of angiogenesis.

      (A) Experimental schedule for the treatment of MI in a mouse model. (B) Hematoxylin-eosin (H&E) staining was performed to detect inflammatory cell infiltration in the early stage of MI (n = 6). Scale bar, 100 μm. (C) Representative images of CD68 immunofluorescence staining (green) and quantification of CD68+ macrophages in injured myocardium (n = 6). Scale bar, 25 μm. (D and E) Flow cytometry was performed to detect peritoneal macrophage polarization 7 days after surgery followed by different treatments. TNFα- and CD206-positive ascites macrophages (markers of M1 and M2 macrophage phenotypes, respectively) were quantified accordingly (n = 3). (F) Representative images of α–Smooth muscle actin (α-SMA) immunofluorescence staining and the quantification of α-SMA+ arterioles (n = 6). Scale bar, 100 μm. (G) Representative images of von Willebrand factor (vWF) immunofluorescence staining and the quantification of vWF+ capillaries (n = 6). Scale bar, 100 μm. Data are expressed as the mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

      Expand for more

      Open in viewer

      It has been widely accepted that a higher proportion of M2-type macrophages is more conducive to the repair of tissue damage. Thus, peritoneal macrophages were extracted at 7 days after surgery and examined by flow cytometry to assess the polarization of macrophages in MI mice that received different treatments. MI modeling leads to a marked increase in the polarization of macrophages toward the M1 phenotype (TNF-α positive); however, the ratio of M1-type macrophages was moderately reduced after treatment with eMSCs. Further administration of NO via the intracellular delivery method significantly (P < 0.001 or 0.01) inhibited the polarization of macrophages toward the M1 phenotype while increasing the proportion of M2-type macrophages compared to the control and eMSC groups (Fig. 5, D and E). Additionally, we conducted immunofluorescence staining in heart section (fig. S5). The results revealed that intracellular delivery of NO also induces the polarization of macrophages into the M2 phenotype within the heart.

      In vitro studies demonstrated that exogenous NO could improve the proangiogenic capacity of eMSCs. Here, we further explored the influence of the combined delivery of exogenous NO and eMSCs on the reconstruction of the vascular network at the site of infarction. α–Smooth muscle actin (α-SMA)–positive arterioles and von Willebrand factor (vWF)–positive small vessels in MI mice after the different treatments were detected by immunofluorescence staining (Fig. 5, F and G). Treatment with eMSCs efficiently promoted angiogenesis in the injured myocardium, and more prominent enhancement was observed in response to further treatment with intracellular NO. This finding was further supported by the expression‎ of angiogenesis-related genes in the border zone of the infarcted heart (fig. S6).

      Intracellular delivery of NO improves heart function and inhibits adverse myocardial remodeling in MI mice after treatment with eMSCs

      Ultrasound and histological analyses were performed to eval‎uate the long-term recovery of cardiac function after MI. Cardiac injury was first eval‎uated by 2,3,5-triphenyltetrazolium chloride (TTC) staining (Fig. 6A). Treatment with eMSCs moderately repressed MI compared to the acute myocardial infarction (AMI) group, but a more pronounced inhibitory effect was observed in the group with further intracellular NO delivery. Left ventricular function was assessed by echocardiography at different time points. As shown in Fig. 6B, after 1 day of MI, the left ventricle in each group was markedly enlarged, cardiac function decreased rapidly, and deterioration of heart function continued for 28 days without detectable restoration in the AMI group. However, eMSC treatment could restore left ventricular systolic function and reduce ventricular dilation, as shown by the increase in left ventricular ejection fraction (LV-EF) and fraction shortening (LV-FS), as well as the decrease in left ventricular end-diastolic diameter (LVIDd) and left ventricular end-diastolic volume (LV-EDV) to a certain extent. In the group treated with eMSCs and intracellular NO delivery (NO-eMSCs), LV-EF and LV-FS were effectively recovered, while LVIDd and LV-EDV were significantly (P < 0.001 or 0.0001) enhanced compared to the AMI group.

      

      Fig. 6. Intracellular delivery of nitric oxide (NO) improves heart function and reduces adverse cardiac remodeling in myocardial infarction (MI) mice after treatment with engineered mesenchymal stem cells (eMSCs).

      (A) Representative images of 2,3,5-triphenyltetrazolium chloride (TTC) staining and quantification of the infarct area (n = 3). Scale bar, 2 mm. (B) Cardiac echo measurement was performed at different time points after surgery, and cardiac function indicators of left ventricular ejection fraction (LV-EF), left ventricular fractional shortening (LV-FS), left ventricular internal diameter at end diastole (LVIDd), and left ventricular end-diastolic volume (LV-EDV) were eval‎uated accordingly (n = 6). ***P < 0.001, ****P < 0.0001 versus acute myocardial infarction (AMI) group. (C) Masson’s trichrome staining was performed, and the infarct size was quantified accordingly (n = 6). (D) Collagen deposition in the hearts was detected by Sirius Red staining (n = 6). Scale bar, 100 μm. (E) Representative images of wheat germ agglutinin (WGA) immunofluorescence staining and quantification of the cross-sectional area of cardiomyocytes (n = 6). Scale bar, 50 μm. (F) Representative images of immunofluorescence staining (red) for the gap junction protein (Cx43) and the quantification of the intensity of red fluorescence to the whole area of images (n = 6). Scale bar, 25 μm. Data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

      Expand for more

      Open in viewer

      Masson staining was used to detect the degree of myocardial fibrosis after MI. Severe myocardial fibrosis was observed in MI mice compared to the sham group (Fig. 6C). In addition, the ventricular wall became thinner, which was closely related to impaired left ventricular systolic function after MI, as demonstrated by echocardiography. However, these adverse cardiac remodeling events were ameliorated after eMSC treatment and accompanied by a lowered degree of myocardial fibrosis and a thickened ventricular wall in contrast to the AMI group. Notably, treatment with eMSCs and intracellular NO delivery exerted the most prominent inhibitory effect on myocardial fibrosis after MI (Fig. 6C).

      Collagen deposition in MI mice was assessed by Sirius staining (Fig. 6D), and the results showed that MI resulted in severe collagen deposition in the injured myocardium compared to the sham group. It was effectively reduced after eMSC treatment, and the inhibitory effect of eMSC plus intracellular NO delivery was significantly higher than that in the other two groups (P < 0.001 or 0.0001).

      Next, wheat germ agglutinin (WGA) staining was carried out to eval‎uate myocardial cell hypertrophy 28 days after MI (Fig. 6E). The cross-sectional area of cardiomyocytes was increased in MI mice in contrast to the sham operation group due to compensatory hypertrophy in the heart to maintain the normal rate of cardiac ejection. Hypertrophy was significantly (P < 0.05) mitigated after treatment with eMSCs, especially in the presence of exogenous NO (P < 0.001), indicating an ideal therapeutic effect on inhibiting myocardial cell hypertrophy and adverse ventricular remodeling by the combination of eMSCs and NO.

      Gap junctions (GJs) are the main connections between cardiomyocytes in the heart, and Cx43 is the main GJ protein in ventricular muscle in the heart (24). Studies have shown that the absence of Cx43 leads to the occurrence of cardiac ventricular arrhythmia, which can develop into heart failure (25). After 28 days of MI, immunofluorescence staining for Cx43 revealed abundant and uniform distribution of GJ proteins in the sham group, whereas MI injury resulted in a marked decrease in the expression‎ of Cx43 (Fig. 6F). Despite the moderate inhibitory effect provided by the administration of eMSCs, further delivery of NO via the intracellular method significantly enhanced (P < 0.001 or 0.01) the expression‎ of Cx43 compared to that in the AMI or eMSC groups.

      Intracellular delivery of NO enhances the therapeutic efficacy of eMSCs in a rat MI model

      Although the outcome in a mouse model supported the beneficial effect of NO via intracellular delivery on enhancing the therapeutic efficacy of MSCs for MI, immediate administration of stem cells after MI is different from the clinical treatment of MI due to the limitation of the administration paradigm in mouse models. In addition, 3 to 7 days after MI is the outbreak period of the inflammatory response. For this reason, we established a rat model of MI and conducted secondary thoracotomy 3 days after surgery (Fig. 7A), and eMSCs were delivered via an injectable HA hydrogel as the carrier (Fig. 7B). Lactate dehydrogenase (LDH), a crucial marker for assessing the extent of myocardial damage, exhibited an initial elevation within 2 to 48 hours following the onset of MI, reaching its zenith between 2 and 5 days after MI. We collected blood samples from the orbital venous plexus of rats 5 days after MI to measure serum LDH levels (fig. S7). The findings revealed a sharp increase in serum LDH levels due to MI. However, treatment with NO-eMSCs significantly reduced serum LDH levels, indicating an attenuation of cardiac injury.

      

      Fig. 7. Intracellular delivery of nitric oxide (NO) enhances the therapeutic efficacy of engineered mesenchymal stem cells (eMSCs) in a rat myocardial infarction (MI) model.

      (A) Experimental schedule for the treatment of rat MI. (B) Representative images showing the second thoracotomy in rats after MI. (C) Cardiac echo measurement was performed at different time points after surgery, and cardiac function indicators of left ventricular ejection fraction (LV-EF), left ventricular fractional shortening (LV-FS), left ventricular internal diameter at end diastole (LVIDd), and left ventricular end-diastolic volume (LV-EDV) were eval‎uated accordingly (n = 6). **P < 0.01, ***P < 0.001 versus acute myocardial infarction (AMI) group. (D) Representative images of Masson’s trichrome staining and quantification of the infarct size and infarct thickness (n = 6). **P < 0.01, ***P < 0.001, ****P < 0.0001 versus AMI group. (E) Representative images of wheat germ agglutinin (WGA) immunofluorescence staining and quantification of the cross-sectional area of cardiomyocytes (n = 6). Scale bar, 50 μm. (F) Representative images of α–Smooth muscle actin (α-SMA) immunofluorescence staining and quantification of α-SMA+ arterioles (n = 6). Scale bar, 100 μm. Data are expressed as the mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

      Open in viewer

      Echocardiography was performed to detect heart function at different time points after MI (Fig. 7C). At 4 weeks after surgery, the anterior wall of the left ventricle was completely infarcted, and the ventricular cavity was increased dramatically. Quantitative analysis further indicated that both EF and FS were decreased markedly after surgery. Treatment with MSCs could restore the systolic function of the heart, and the highest recovery rate (60 to 80%) was observed in the group treated with eMSCs and intracellular NO delivery. Ventricular dilatation was also effectively restored after combined treatment with eMSCs and NO, which was consistent with the results obtained in the mouse model.

      Histological analyses showed that treatment with eMSCs combined with NO efficiently reduced the degree of fibrosis (Fig. 7D) and collagen deposition (fig. S8A). Furthermore, it alleviated infarct size after MI and restored ventricular cavity morphology of the heart with a significant enhancement in the thickness of the infarcted left ventricular wall and the interventricular septum (IVS), as eval‎uated by Masson (Fig. 7D) and H&E staining (fig. S9). Twenty-eight days after MI, WGA staining also showed that the combination of eMSCs and NO significantly (P < 0.01) inhibited cardiomyocyte hypertrophy in contrast to the AMI group, thus inhibiting further myocardial systolic dysfunction (Fig. 7E). The expression‎ of the GJ protein Cx43 (fig. S8B) followed a similar trend to that in mouse models; that is, acute MI led to a marked decrease in the distribution of Cx43 in the myocardium, and treatment with eMSCs up-regulated the expression‎ of Cx43 to a certain extent. Further delivery of NO via the intracellular method produced a more significant (P < 0.05 or 0.01) effect on the up-regulation of Cx43 in the injured myocardium compared to the eMSC and AMI groups, thereby facilitating the connection between cardiomyocytes and further inhibiting the occurrence of arrhythmias and heart failure.

      Tissue repair after MI is often closely related to angiogenesis, which begins at the infarct border and extends to the center of the infarction (26). Therefore, we further compared the proangiogenic effect of eMSCs with and without NO delivery on the damaged margin of the infarcted myocardium 28 days after MI in rats (Fig. 7F and fig. S8C). The combination of eMSCs and NO remarkably promoted angiogenesis and reconstruction of the vascular network compared to the group treated with eMSCs only, which is beneficial to the repair of myocardial injury after MI.

      DISCUSSION

      Cell therapy based on MSCs has proven to be a powerful solution for treating degenerative diseases and tissue damage (27–29). Despite the advantages of autologous stem cells over allogeneic stem cells, including the absence of immune rejection, the decreased survival and impaired paracrine functions of stem cells collected from patients with chronic diseases greatly limit their clinical use (7, 30, 31). Here, we first showed that MSCs collected from patients with diabetes exhibited marked up-regulation of apoptosis- and inflammation-related genes and down-regulation of proangiogenic genes, affecting the efficacy of cell therapy. As a result, genetic engineering strategies are often required to enhance the therapeutic efficacy of autologous stem cells. A recent study revealed that eMSCs, engineered to express PD-L1 on their surface and secrete CTLA4-Ig (immunoglobulin) as an extracellular factor, exhibited immunoprotective properties, which improved the outcome of both syngeneic and allogeneic islet transplantation in diabetic mice (32).

      NO is involved in a variety of physiological processes. Studies have shown that as an important signaling molecule, NO plays a pivotal role in regulating stem cell behavior (33–35), including cell survival, migration, differentiation, and paracrine behavior. These factors affect the interaction of stem cells with other cells and the tissue microenvironment. Previously, different types of NO-releasing biomaterials, such as injectable hydrogels, have been prepared by us and other groups (36–40), and further studies have shown that the combination of NO and MSCs is more effective in treating various diseases than MSC therapy alone. In addition, it has been reported that pretreatment of MSCs with NO-releasing biomaterials could enhance the therapeutic efficacy of MSCs and their secreted exosomes because of their enhanced proangiogenic functions (41).

      Due to the spatiotemporal characteristics of NO (42), precise delivery of NO in a site-specific and controllable manner holds great importance in the regulatory effect of exogenously administered NO. In addition to the controlled release rate, the site at which NO is generated is also a key factor due to the relative half-life and limited diffusion distance (43, 44). It is reasonable to speculate that intracellular and extracellular NO delivery may lead to different outcomes when regulating the survival and function of MSCs. In our previous work, an enzyme-prodrug delivery system was designed based on a bump-and-hole strategy (21). The mutant galactosidase (A4-β-GalH363A) enables the targeted delivery of NO, thus reducing the side effects due to the unspecific decomposition of the NO prodrug and enhancing the therapeutic efficacy. Here, we transfected a plasmid expressing mutant galactosidase into MSCs and successfully constructed eMSCs. The enzyme expressed by MSCs could catalyze the decomposition of the 6-OMe-galactose–protected NO prodrug and release NO intracellularly.

      Western blotting and fluorescence imaging demonstrated that the expression‎ of the engineered enzyme was confined to the nucleus of MSCs, while wild-type β-galactosidase was widely distributed in the cytoplasm, including the lysosome and perinuclear region (45, 46). Since the corresponding prodrug for wild-type β-galactosidase is highly hydrophilic, it fails to enter MSCs and releases NO extracellularly by enzymes that translocate from the cell. In contrast, the prodrug for mutant galactosidase is cell penetrating because of the modified molecular structure; therefore, it can enter MSCs and release NO intracellularly under the catalysis of the corresponding enzyme expressed by the cells. Accordingly, two different NO delivery paradigms were successfully developed in this study and further confirmed by a series of eval‎uations, including cell imaging and electronic paramagnetic resonance. Further in vitro and in vivo assays indicated that in contrast to extracellular NO delivery, intracellular administration of NO enhanced cell survival and the paracrine effects of MSCs, including inhibiting apoptosis and supporting angiogenesis.

      Next, we established a mouse MI model to systematically eval‎uate the therapeutic efficacy of MSCs combined with exogenous NO. The results showed that intracellular delivery of NO prolonged the retention of eMSCs after myocardial orthotopic transplantation. In addition, the combination of eMSCs and intracellular NO delivery improved cardiac function after MI and reduced adverse ventricular remodeling compared to the group treated with MSCs only. Additionally, it could effectively restore the reconstruction of the blood vessel network and further promote the repair of the infarcted myocardium.

      To gain further insight into the translational potential of the combinatory therapeutic strategy developed in this study, a rat model of MI was established, and MSCs were administered by a second thoracotomy after 3 days to mimic the clinical use of MSCs for the treatment of MI (47, 48). Clinically, acute MI is typically due to the rupture of coronary atherosclerotic plaque and the formation of thrombus, which causes coronary artery obstruction. After the acute phase of MI, adverse ventricular remodeling further affects the prognosis of patients, which is specifically characterized as a decrease in ventricular wall thickness and myocardial tension in the MI area, myocardial hypertrophy in the noninfarction area, and a change in the morphology of the ventricular cavity, thus leading to arrhythmia and further development into heart failure. The efficacy of MSCs in managing arrhythmias remains a topic of ongoing debate. Some researchers argue that MSCs do not appear to reduce or prevent arrhythmias, with the antiarrhythmic or proarrhythmic potential of MSCs primarily relying on paracrine factors (49). Conversely, other studies suggest that MSCs themselves may play a role in the post-MI recovery process (50). In our study, we observed an evident up-regulation of Cx43 expression‎ after NO-eMSC treatment, which is a potential target associated with antiarrhythmic effects. Further investigation is still required to comprehensively explore the antiarrhythmic potential of NO-eMSCs. In line with the enhanced therapeutic efficacy in the mouse model, intracellular delivery of NO showed enormous advantages in the rat MI model by inhibiting apoptosis and enhancing the paracrine function of MSCs.

      In summary, we first showed that survival and paracrine function were reduced in MSCs collected from patients with diabetes, which could greatly affect therapeutic efficacy. Accordingly, eMSCs were successfully constructed, and the mutant β-galactosidase expressed by the cells enabled the intracellular generation of NO via the conversion of an exogenous NO prodrug. In vitro and in vivo assays indicated that intracellular delivery of NO effectively enhanced the survival of transplanted MSCs and promoted the paracrine function of MSCs, which was further confirmed by the enhanced therapeutic efficacy in mouse and rat models of MI compared to the group treated with MSCs only. This synergistic strategy provides an option for the treatment of MI by autologous MSCs in the clinic.

      MATERIALS AND METHODSRNA sequencing analysis

      RNA sequencing was performed by the BGI (Shenzhen, China). Briefly, RNA from the ADMSCs of healthy people and patients with diabetes was extracted using TRIzol reagent (Yeasen, China). RNA samples were sequenced on the BGISEQ platform. The raw data containing low-quality reads, adaptor sequences, and high levels of N bases were filtered before analysis. Then, the clean reads were mapped to the reference genome using HISAT, and Bowtie2 was used to align the clean reads to the reference genes. The reference genome source is National Center for Biotechnology Information (NCBI), and the reference genome version is GCF_000001405.39_GRCh38.p13. The expression‎ levels of genes were quantified to identify differentially expressed genes by RNA-Seq by expectation maximization (RSEM). The analyses of hierarchical clustering and heatmap were performed using the online Dr. Tom system (biosys.bgi.com) to compare differential gene expression‎ of ADMSCs in healthy people and patients with diabetes. According to the KEGG_pathway annotation classification, the phyper function in R software was used for enrichment analysis, the P value was calculated, and then false discovery rate (FDR) was performed on the P value to obtain a Q value. Generally, a Q value of ≤0.05 was regarded as significant enrichment. GSEA was used to analyze significant differences in gene expression‎ between inflammatory and apoptosis-related pathways. Expression‎ cluster heatmap was used to analyze the expression‎ of genes associated with angiogenesis.

      Measurement of NO release

      The NO-releasing profile was determined by the Griess kit assay. In brief, 50 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were dissolved in phosphate-buffered saline (PBS) buffer (pH 7.4), and enzymes were added to the solutions at a concentration of 0.005 mg/ml. At each predetermined time interval, 50 ml of solution was transferred into a 96-well plate, and 50 ml of Griess I and 50 ml of Griess II were added thereafter. The azo compound of purple color was formed, and the absorbance was measured at a wavelength of 540 nm using an iMark microplate reader (Bio-Rad, USA).

      Cell cultureMesenchymal stem cells

      MSCs derived from human umbilical cord were obtained from Health-Biotech, maintained in Dulbecco's modified Eagle’s medium (DMEM)/F12 (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, USA) and 1% penicillin- streptomycin solution (Gibco, USA), and placed in a cell culture chamber containing 5% CO2 at 37°C.

      Human embryonic kidney 293T cells

      Human embryonic kidney (HEK) 293T cells were obtained from the American Type Culture Collection (ATCC), maintained in high-glucose DMEM (Gibco, USA) with 10% FBS (Gibco, USA) and 1% penicillin-streptomycin solution (Gibco, USA), and placed in a cell culture chamber containing 5% CO2 at 37°C.

      Construction of eMSCs

      The coding sequence of mutant β-galactosidase (A4-β-GalH363A) can be obtained from the previous publication (21). The lentivirus packaging system containing A4-β-GalH363A sequence and Rluc-RFP sequence was constructed by Wuhan Miaolingbio Co. Ltd. The constructed lentivirus plasmid containing the target gene and the package gene (psPAX2 and pMD2.G) was transfected into HEK 293T cells through Lipo2000, and the supernatant was collected to obtain the virus solution. After removing impurities, the virus solution was mixed 1:1 with fresh MSC medium, and polybrene (10 μg/ml) was added. MSCs were infected with virus through incubation in the mixture medium. The infection efficiency was observed under an inverted fluorescence microscope, and the expression‎ of target protein was determined by Western blotting.

      Cell immunofluorescence staining

      eMSCs were inoculated in 24-well plates. Cells were fixed with 4% paraformaldehyde and blocked in 4% bovine serum albumin in PBS for 30 min at room temperature. Then, the cells were incubated with primary antibodies overnight at 4°C. The bound primary antibodies were displayed by incubation with the secondary antibodies for 2 hours at room temperature. Cells were counterstained with DAPI (4′,6-diamidino-2-phenylindole)–containing Fluoromount-G and visualized with a fluorescence microscope (Axio Imager Z1). Antibodies used include anti–β-galactosidase (1:100, A1863, Abclonal) and anti-RFP (1:100, PA1-986, Invitrogen).

      Western blot

      eMSCs were collected, and total protein was extracted using radioimmunoprecipitation assay (RIPA) lysate containing protease inhibitor (Solarbio, China). Cytoplasmic protein and nucleoprotein were extracted using a nucleoprotein extraction kit containing protease inhibitors (Solarbio, China). The protein concentration was quantified using a BCA protein assay kit (Solarbio, China). The samples were diluted with 4× SDS–polyacrylamide gel electrophoresis (PAGE) loading buffer and boiled in boiling water for 8 min. Then, 30 μg of protein was isolated by 10% SDS-PAGE electrophoresis. The isolated proteins were transferred to an Immobilon-P Transfer membrane (Millipore, USA) and incubated with the primary antibody overnight at 4°C and then with the secondary antibody at room temperature for 2 hours. The bands were detected with chemiluminescent horseradish peroxidase substrate (Millipore, USA). Signals were generated by using an enhanced chemiluminescence (ECL) reagent (Millipore, USA) and were captured by using the Tanon-5200 Chemiluminescence Imaging System (Tanon, China). The antibodies used included anti–β-galactosidase (1:1000, A1863, Abclonal), anti–His-tag (1:1000, 12698S, Cell Signaling Technology), anti–glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (1:1000, AC001, Abclonal), anti–histone H3 (1:1000, 4499S, Cell Signaling Technology), anti–β-actin (1:1000, UM4001, Utibody), anti-BCL2 (1:1000, WL01556, Wanleibio), anti-Bax (1:1000, 50599-2-lg, Proteintech), anti-Bad (1:1000, WL02140, Wanleibio), and anti-caspase/cleaved caspase3 (1:1000, WL02117, Wanleibio).

      6-OMeGal-Ph-NO uptake by eMSCs

      eMSCs were inoculated in 12-well plates, and 50 μM substrate (6-OMeGal-Ph-NO) was added per well. At predetermined time points (0, 1, 3, 6, and 12 hours), an appropriate amount of culture medium was collected, and an excess of A4-β-GalH363A was added immediately to fully catalyze the decomposition of the remaining substrate at room temperature. The amount of 6-OMeGal-Ph-NO substrate in the culture medium was determined by Griess kit assay.

      Real-time imaging of intracellular NO

      The fluorescence emission associated with NO in the cytosol was detected by an electron-multiplying charge-coupled device (DU-897D-CS0-BV; Andor, Belfast, UK) connected to an inverted fluorescence microscope (Axio Observer D1; Carl Zeiss, Oberkochen, Germany). Intracellular NO imaging was performed using an NO fluorescence probe, DAF-AM DA (Beyotime, China), according to the manufacturer’s instruction. eMSCs were inoculated in small confocal dishes in advance, and the experiment was conducted when the cell density reached 80%. First, the medium was collected for later use. After two gentle washes with PBS, 5 μM DAF-AM DA solution was incubated at 37°C for 30 min in the dark. Then, cells were gently washed with PBS twice. The previously collected medium was added anew, and cell imaging was performed immediately. At the 488-nm excitation wavelength, pictures were taken every 5 s. After stable shooting for 2 min, 100 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were added. The filming for cell fluorescence was continued for 4 min. The final fluorescence intensity was determined without the background fluorescence value. The proportion of change in fluorescence intensity of each cell in the visual field was calculated.

      Intracellular NO detection

      Intracellular NO radicals (NO•) were detected using EPR as described (51, 52). In brief, sodium DETC (4.5 mg) and FeSO4•7H2O were dissolved in two separate volumes (10 μl) of deoxygenated Krebs/Hepes solution. Equal volumes of these parent solutions were rapidly mixed and aspirated into Eppendorf combi tips. The 0.5 mM Fe•(DETC)2 colloid solution had a yellow-brownish color with a slight opalescence in light. No aggregate formation was observed, at least during the first 30 min. eMSCs were rinsed with modified Krebs/Hepes buffer and incubated with freshly prepared NO•-specific spin trap Fe•(DETC)2 colloid (0.5 mM) for 30 min. Meanwhile, 100 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were added to the buffer. Gently collected cell suspensions were snap-frozen in liquid nitrogen. Ethyl acetate (200 μl) was added, and the cells were ultrasonically broken to extract DETC2-Fe-NO. The ethyl acetate extract was concentrated with nitrogen and transferred to a 50-μl capillary, and then the X-band EPR was measured at room temperature. The following acquisition parameters were used: modulation frequency, 100 kHz; microwave power, 10 mW; modulation amplitude, 2 G; number of scans, 60. The double-integrated area of the EPR spectra was calibrated into concentrations of DETC2-Fe-NO using TEMPO (2,2,5,5-tetramethyl piperidine 1-oxyl) as a standard. EPR spectral simulation was conducted by the WINSIM program.

      Extracellular NO detection

      eMSCs were treated with β-Gal-NO or 6-OMeGal-Ph-NO (30 μM). The production of NO in the medium of each group was detected 6 hours after incubation with NO-sensitive near-infrared fluorescence probe (5 μM). The NO production of medium in different groups was compared by the relative fluorescence intensity under the excitation at 750 nm (emission at 800 nm).

      Cell apoptosis detection

      To test the protective effect of NO delivery on cellular oxidative stress stimulation, 30 μM NO substrate (6-OMeGal-Ph-NO) was added to the medium in advance. Then, H2O2 with different concentrations (100, 200, 400, and 600 μM) was added to stimulate the lentivirus-infected eMSCs. BLI was performed immediately after addition of the luciferase substrate coelenterin to eval‎uate cell apoptosis. Additionally, eMSCs treated with 200 mM H2O2 were stimulated for 24 hours to induce cell apoptosis. An Annexin V/PI assay kit (Solarbio) was used to detect eMSC apoptosis.

      BLI detection of cell retention

      BLI and luciferase substrates were used in mice to eval‎uate the retention of NO-eMSCsGluc/RFP in cardiac orthotopic transplantation. The mice after eMSC injections were anesthetized with 1.5% isoflurane and injected with coelenterin through the caudal vein at 150 mg/kg. After injection, the mice were immediately placed in a BLI system to detect cell retention in the myocardium.

      Animals

      C57BL/6 mice (male, 8 weeks old) and Sprague-Dawley rats (male, 8 weeks old) were purchased from SPF (Beijing) Biotechnology Co. Ltd., Beijing, China. Animals were randomly grouped for treated and untreated controls. All experiments and animal procedures were approved by the Animal Experiments Ethical Committee of Nankai University and carried out in conformity with the Guide for Care and Use of Laboratory Animals.

      MI in mice and rats

      Surgical induction of MI was performed on C57BL/6 mice (male, 8 weeks old) as previously described with some modifications. Briefly, mice were anesthetized with 2% isoflurane, followed by fixation to a heating pad (37°C) at supine position, and then ventilated with a mechanical ventilation system (Hallowell EMC Microvent I, USA) set at 110 breaths per minute with a tidal volume of 0.2 to 0.3 ml. The third intercostal space over the left chest and the heart was exposed. After left thoracotomy, the left anterior descending coronary artery was ligated with a 6-0 silk ligature. The left main descending coronary artery was sutured and tied with a slip knot at a site approximately 3 mm from its origin. Cardiac ischemia was confirmed by the presence of myocardial blanching. Immediately, eMSCs encapsulated with HA hydrogel were injected into the myocardium of mice through three-point injection around the infarct zone. The chest cavity was closed to restore negative pressure and prevent pneumothorax.

      Similar MI surgery was performed on Sprague-Dawley rats (male, 8 weeks old) first. Briefly, rats were anesthetized via intraperitoneal injection of 10% chloral hydrate (350 mg/kg), followed by fixation to a heating pad (37°C) at supine position. Then, they were ventilated with a mechanical ventilation system (Hallowell EMC Microvent I, USA) set at 110 breaths per minute with a tidal volume of 6 ml. The third intercostal space over the left chest and the heart was exposed. After left thoracotomy, the left anterior descending coronary artery was ligated with a 6-0 silk ligature. The left main descending coronary artery was sutured and tied with a slip knot at a site approximately 3 mm from its origin. Cardiac ischemia was confirmed by the presence of myocardial blanching. The chest cavity was closed to restore negative pressure and prevent pneumothorax. Three days after surgery, secondary thoracotomy was performed, and eMSCs were injected into the myocardium around the infarct zone. The chest cavity was closed to restore negative pressure and prevent pneumothorax.

      Mice and rats in the AMI group only received MI without eMSC injection, while sham-operated mice only experienced thoracotomy without MI surgery or eMSC injection.

      At 1, 3, 5, and 7 days after myocardial injection of eMSCs, the prodrug was injected through the tail vein. Mice were injected with 100 μl of prodrug (1 mg/ml) each time, and rats were injected with 200 μl of prodrug (1 mg/ml) each time.

      TTC staining

      Two days after surgery, a thoracotomy was performed. The heart was quickly excised after quick freezing for 15 min and sliced at 1 mm thickness. Afterward, the sections were incubated with 1.5% TTC (Sigma-Aldrich) solution at 37°C in an incubator for 15 to 30 min and then with a 4% formaldehyde solution for 2 hours. The normal myocardial tissue was red, while the ischemic myocardium was white. The size of the infarcted myocardium, which was white or pale, was measured by ImageJ software.

      Cardiac function assessment

      Transthoracic echocardiography was performed with the Vevo 2100 Imaging System (Fuji Film Visual Sonics Inc., Canada) equipped with an MS-250/400 imaging transducer. The baseline cardiac function of mice and rats was measured at 3 days before surgery. Cardiac function was analyzed at days 1 and 28 after MI surgery with different treatments, as reported previously. Mice or rats were slightly anesthetized in a box with isoflurane. Their limbs were fixed in a supine position on the echo mat, and the chest hair was removed by depilating cream. Then, mice or rats were anesthetized by inhalation of isoflurane (0.5 to 1%) mixed with oxygen to maintain the heart rate at approximately 500 to 600, and M-mode echocardiography was performed. The left ventricular internal diameter at end-diastole (LVIDd) and systole (LVIDs) were obtained by measuring the long axis and the short axis. Accordingly, the cardiac parameters LV-EF, LV-FS, LV-EDV, and LV end-systole volume (LV-ESV) were determined. The echocardiography measurement was carried out in a double-blind manner.

      Histological analysis

      At the indicated time points, mice and rats were anesthetized via intraperitoneal injection of chloral hydrate, and a thoracotomy was performed. The hearts were fixed with trans-cardiac perfusion of saline and immersed in 4% paraformaldehyde over 24 hours. The heart tissue samples were dehydrated with gradient alcohol and xylene, embedded in paraffin blocks, and cut into sections in 5 μm thickness.

      The paraffin-embedded sections were stained with Masson trichrome, H&E, and Sirius Red following a standard protocol. Immunofluorescence staining was performed on paraffin-embedded sections of the heart tissue samples. After deparaffinization and heat-mediated antigen retrieval‎ in citrate solution, the samples were washed with PBS three times and incubated with blocking serum, which was used to avoid nonspecific binding, at room temperature for 30 min. The sections were incubated with specific antibodies diluted in goat serum at 4°C overnight. On the second day, the sections were rewarmed at room temperature for 1 hour and washed with PBS three times. Afterward, the sections were incubated with Alexa Fluor–coupled secondary antibodies for 2 hours at room temperature. After washing with PBS, the sections were counterstained with DAPI-containing Fluoromount-G (SouthernBiotech, USA) and coverslipped. The antibodies used included anti–α-SMA (1:100, ab5694, Abcam), anti-vWF (1:100, ab6694, Abcam), anti–α-actinin (1:100, ab9475, Abcam), anti-Connexin43 (1:100, ab11370, Abcam), anti-CD68 (1:100, ab125212, Abcam), WGA (1:500, FL-1021, Novus Biologicals), anti-iNOS (1:100, ab178945, Abcam), and anti-CD206 (1:100, ab64693, Abcam).

      Macrophage isolation and detection

      Three days before euthanasia, mice were intraperitoneally injected with 2 ml of 4% thioglycolate. Three days later, the mice were sacrificed by cervical dislocation and immersed in 75% alcohol and then transferred to an ultraclean workbench. The mouse limb was fixed in the supine position, and the mouse abdominal wall was carefully cut open with the peritoneal. PBS [1% penicillin-streptomycin (PS)] was injected intraperitoneally to collect the cell suspension, which was centrifuged at 2000 rpm for 10 min. After discarding the supernatant, the cells were incubated with anti-F4/80/TNF-α and anti-F4/80/CD206 antibodies. FlowJo software was used to analyze the results of flow cytometry.

      Quantitative real-time PCR

      Total RNA samples from the cells were prepared using TRIeasy Total RNA Extraction Reagent (Yeasen, China) according to the manufacturer’s instructions. Heart tissue samples were collected at the indicated time points after MI surgery.

      The tissue samples were dissected at the border zone of the left ventricle and frozen in liquid nitrogen immediately. Afterward, the total RNA was extracted with TRIzol reagent, as mentioned before. The concentration of the RNA was measured with a NanoDrop spectrophotometer (NanoDrop Technologies, USA). The complementary cDNA was synthesized using a first-strand cDNA synthesis kit (Yeasen, China). Quantitative real-time PCR was performed on a CFX96 real-time PCR system (Bio-Rad, Hercules, USA) with an SYBR Green–based real-time detection system (Yeasen, China). The relative gene expression‎ of mRNA was expressed as 2−(△△CT) and normalized to GAPDH as an internal control. Each reaction was performed in triplicate to obtain an average value, and the changes in relative gene expression‎ normalized to the internal control levels were determined. The highly purified primers used in this experiment were commercially synthesized (Sango, China). The sequences of the primers used in this experiment are summarized in the Supplementary Materials.

      Statistics

      All data are presented as the mean ± SEM from at least three independent experiments. Comparisons between two groups were performed by Student’s t test, and comparisons among more than two groups were performed by one-way or two-way analysis of variance (ANOVA). Statistical analyses were performed with GraphPad Prism software 7.0, and a statistical significance level of less than 0.05 was accepted.

      Acknowledgments

      Funding: This study is supported by the National Key R&D Program of China (2018YFE0200503), the National Natural Science Foundation of China (nos. 81925021, 82330066, 81921004, and U2004126), and the Tianjin Natural Science Foundation (21JCZDJC00240).

      Author contributions: Q.Z. and Z.L. conceived the original concept and initiated this project. Q.Z., Z.L., and F.G. designed the experiment and supervised the entire project. S.W. collected human adipose mesenchymal stem cells. M.Q. synthesized all NO prodrugs and probes. P.L. prepared engineered enzymes. T.H., G.J., and Q.X.L. established mouse and rat MI models. T.H. and G.J. performed histological analysis. G.J. and W.D. carried out in vitro cell experiments. S.D. carried out NO cell imaging under the supervision of L.P. T.H., G.J., and M.Q. analyzed data under the supervision of Q.Z. H.H. helped with lentivirus packaging and cell infection. W.G. and T.L. helped in establishing animal MI models. Y.W., J.H., J.C., and J.T. helped with data collection. T.H. and Q.Z. wrote the paper with input from other authors.

      Competing interests: The authors declare that they have no competing interests.

      Data and materials availability: All data needed to eval‎uate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Transcriptome sequencing dataset is available at https://doi.org/10.5061/dryad.tqjq2bw5b.

      Supplementary MaterialsThis PDF file includes:

      Supplementary Text

      Figs. S1 to S9

      Table S1

      • Download
      • 1.94 MB

      REFERENCES AND NOTES

      1

      M. F. Pittenger, A. M. Mackay, S. C. Beck, R. K. Jaiswal, R. Douglas, J. D. Mosca, M. A. Moorman, D. W. Simonetti, S. Craig, D. R. Marshak, Multilineage potential of adult human mesenchymal stem cells. Science 284, 143–147 (1999).

      GO TO REFERENCE

      Crossref

      PubMed

      Web of Science

      Google Scholar

      2

      A. Uccelli, L. Moretta, V. Pistoia, Mesenchymal stem cells in health and disease. Nat. Rev. Immunol. 8, 726–736 (2008).

      GO TO REFERENCE

      Crossref

      PubMed

      Web of Science

      Google Scholar

      3

      Y. Wang, J. Fang, B. Liu, C. Shao, Y. Shi, Reciprocal regulation of mesenchymal stem cells and immune responses. Cell Stem Cell 29, 1515–1530 (2022).

      GO TO REFERENCE

      Crossref

      PubMed

      Web of Science

      Google Scholar

      4

      Y. Wang, X. Chen, W. Cao, Y. Shi, Plasticity of mesenchymal stem cells in immunomodulation: Pathological and therapeutic implications. Nat. Immunol. 15, 1009–1016 (2014).

      GO TO REFERENCE

      Crossref

      PubMed

      Web of Science

      Google Scholar

      Show all references

      (0)eLetters

      eLetters is a forum for ongoing peer review. eLetters are not edited, proofread, or indexed, but they are screened. eLetters should provide substantive and scholarly commentary on the article. Neither embedded figures nor equations with special characters can be submitted, and we discourage the use of figures and equations within eLetters in general. If a figure or equation is essential, please include within the text of the eLetter a link to the figure, equation, or full text with special characters at a public repository with versioning, such as Zenodo. Please read our Terms of Service before submitting an eLetter.

      Log In to Submit a Response

      No eLetters have been published for this article yet.

      Recommended articles from TrendMD

      • In vivo priming of human mesenchymal stem cells with hepatocyte growth factor–engineered mesenchymal stem cells promotes therapeutic potential for cardiac repai...

        Bong-Woo Park, Sci Adv, 2020

      • Shattering barriers toward clinically meaningful MSC therapies

        Oren Levy, Sci Adv, 2020

      • Nanovesicles derived from iron oxide nanoparticles–incorporated mesenchymal stem cells for cardiac repair

        Ju-Ro Lee, Sci Adv, 2020

      • Exendin-4 gene modification and microscaffold encapsulation promote self-persistence and antidiabetic activity of MSCs

        Yuanyuan Zhang, Sci Adv, 2021

      • Human gingival mesenchymal stem cells retain their growth and immunomodulatory characteristics independent of donor age

        Jay R. Dave, Sci Adv, 2022

      • The secretion profile of mesenchymal stem cells and potential applications in treating human diseases

        Yuyi Han, Signal Transduction and Targeted Therapy, 2022

      • Mesenchymal Stem Cell Therapy for Ischemic Heart Disease: Systematic Review and Meta-analysis

        Hyunsuk Jeong, International Journal of Stem Cells

      • Mesenchymal Stem Cell and MicroRNA Therapy of Musculoskeletal Diseases

        International Journal of Stem Cells

      • Excessive branched-chain amino acid accumulation restricts mesenchymal stem cell-based therapy efficacy in myocardial infarction

        Fuyang Zhang, Signal Transduction and Targeted Therapy, 2022

        Intracellular delivery of nitric oxide enhances the therapeutic efficacy of mesenchymal stem cells for myocardial infarction

        Tian Hao https://orcid.org/0009-0007-8657-0424, Guangbo Ji https://orcid.org/0009-0000-7079-8896, Meng Qian https://orcid.org/0009-0006-9135-4593, Qiu Xuan Li, Haoyan Huang, Shiyu Deng, Pei Liu, Weiliang Deng https://orcid.org/0009-0001-8402-7280, Yongzhen Wei, [...], and Qiang Zhao https://orcid.org/0000-0003-4656-6002 +9 authorsAuthors Info & Affiliations

        Science Advances

        29 Nov 2023

        Vol 9, Issue 48

        DOI: 10.1126/sciadv.adi9967

        4,3612

        Metrics

        Total Downloads4,361

        • Last 6 Months1,270
        • Last 12 Months3,995

        Total Citations2

        • Last 6 Months1
        • Last 12 Months2

        NotificationsBookmark

        • • Abstract
          • INTRODUCTION
          • RESULTS
          • DISCUSSION
          • MATERIALS AND METHODS
          • Acknowledgments
          • Supplementary Materials
          • REFERENCES AND NOTES
          • eLetters (0)
        • • Information & Authors
          • Metrics & Citations
          • View Options
          • References
          • Media
          • Tables
          • Share

        Abstract

        Cell therapy by autologous mesenchymal stem cells (MSCs) is a clinically acceptable strategy for treating various diseases. Unfortunately, the therapeutic efficacy is largely affected by the low quality of MSCs collected from patients. Here, we showed that the gene expression‎ of MSCs from patients with diabetes was differentially regulated compared to that of MSCs from healthy controls. Then, MSCs were genetically engineered to catalyze an NO prodrug to release NO intracellularly. Compared to extracellular NO conversion, intracellular NO delivery effectively prolonged survival and enhanced the paracrine function of MSCs, as demonstrated by in vitro and in vivo assays. The enhanced therapeutic efficacy of engineered MSCs combined with intracellular NO delivery was further confirmed in mouse and rat models of myocardial infarction, and a clinically relevant cell administration paradigm through secondary thoracotomy has been attempted.

        SIGN UP FOR THE AWARD-WINNING SCIENCEADVISER NEWSLETTER

        The latest news, commentary, and research, free to your inbox daily

        Sign up

        INTRODUCTION

        Mesenchymal stem cells (MSCs) are pluripotent stem cells with high self-renewal abilities and multidirectional differentiation potential (1, 2). They are extensively distributed throughout the body and serve a variety of purposes, including tissue regeneration (3), immunoregulation (4), and angiogenesis (5). However, an important challenge for stem cell therapy is the low survival rate of stem cells after transplantation, which is associated with nonspecific homing of cells and ischemia/hypoxia at the injury site (6). In addition, transplanted stem cells cannot fully exert their paracrine effects in pathological environments, which seriously limits the clinical application of stem cells.

        The proangiogenic function of MSCs is a key factor contributing to the treatment of ischemic diseases. Generally, MSCs can stimulate local angiogenesis in ischemic tissue by secreting cytokines, including vascular endothelial growth factor (VEGF), transforming growth factor β (TGFβ), and basic fibroblast growth factor (bFGF), to induce endothelial cells to form tubular structures (7). In addition, MSCs can recruit pericytes and smooth muscle cells to promote neovascularization (8, 9). Recently, MSCs have been reported to participate in the construction of blood vessels by directly differentiating into endothelial or smooth muscle cells (10).

        As an important signaling molecule, nitric oxide (NO) plays a crucial role in the maintenance of vascular homeostasis by inhibiting thrombus formation and promoting angiogenesis (11, 12). Recently, increasing attention has been given to the regulation of stem cells by NO due to its multiple biological functions (13, 14). NO can regulate the proliferation of stem cells by regulating the activities of cyclin and mitotic receptors, as well as their downstream pathways (15, 16). On the other hand, NO can regulate the expression‎ of angiogenic cytokines and immunomodulatory factors to improve the paracrine performance of stem cells (17). Additionally, recent studies have demonstrated that NO can regulate the differentiation behavior of stem cells through the phosphatidylinositol 3-kinase (PI3K)/AKT, guanosine 3′,5′-monophosphate (cGMP), and other signaling pathways (18).

        As a result, NO-releasing biomaterials have been used as delivery carriers for stem cells to enhance their survival and regulate paracrine functions (19). However, NO, which is a gaseous molecule, easily diffuses and has a high level of instability. Furthermore, the physiologic function of NO is dose dependent, and an overdose of NO often leads to notable cytotoxicity (20). Thus, optimizing the beneficial effects of NO to strengthen the therapeutic efficacy of stem cells by tuning their release profile should be taken into account.

        In our previous study, based on the chemical biology principle of “bump-and-hole,” we designed and prepared an enzyme-prodrug delivery system and achieved targeted delivery of NO at the lesion site in two different ischemic disease models (21). Here, MSCs were further modified by gene transfection to express a catalytic enzyme (A4-β-GalH363A). The engineered MSCs (eMSCs) were transplanted using an injectable hyaluronic acid (HA) hydrogel as the carrier, while the NO prodrug was injected through the tail vein to achieve controlled release of NO catalyzed by the enzymes in eMSCs. The therapeutic efficacy of MSCs combined with exogenous NO delivery was eval‎uated in mouse and rat models of myocardial infarction (MI) with an emphasis on comparing the therapeutic efficacy of two different NO administration methods (intracellular or extracellular), and the underlying mechanism of their myocardial protective effect was further explored.

        RESULTSThe gene expression‎ of MSCs collected from patients is differentially regulated

        In the clinic, the therapeutic efficacy of autologous stem cell transplantation is largely affected by the low quality (including cell survival and paracrine function) of stem cells collected from patients due to chronic diseases. Diabetes mellitus is a chronic medical condition that can lead to a variety of complications, and these complications can affect various parts of the body, such as the kidney, lower limb, and heart. Diabetes predisposes affected individuals to a spectrum of cardiovascular complications, and one of the most debilitating in terms of prognosis is heart failure (22).

        Accordingly, although autologous MSCs have been widely accepted as a promising strategy for treating various complications associated with diabetes, the therapeutic efficacy is largely affected by the quality of stem cells collected from the patients themselves. The gene levels in adipose-derived MSCs (ADMSCs) from patients with diabetes and healthy individuals were first compared by transcriptome sequencing. The heatmap shows that multiple genes in MSCs from patients with diabetes were up- or down-regulated compared to healthy controls (Fig. 1A). Gene ontology (GO) enrichment analysis revealed that the most differentially up-regulated genes were related to tumor necrosis factor (TNF) signaling pathways (Fig. 1B). Subsequently, we performed gene set enrichment analysis (GSEA) based on the RNA-sequencing results. GSEA revealed that inflammatory target genes were highly enriched in MSCs from patients with diabetes, with a normalized enrichment score (NES) of 1.65 (P < 0.01) (Fig. 1C). Apoptosis-related genes were also highly enriched, with an NES of 1.77 (P < 0.01) (Fig. 1D). In addition, through enrichment analysis, we found that proangiogenic genes in MSCs were greatly down-regulated in patients with diabetes (Fig. 1E).

        

        Fig. 1. Transcriptome sequencing of adipose mesenchymal stem cells (MSCs) collected from patients with diabetes and healthy controls.

        (A) Heatmap showing the differentially expressed genes of MSCs from patients with diabetes and healthy controls (n = 3). (B) Gene ontology (GO) analysis of the up-regulated transcriptome of MSCs from patients with diabetes. (C and D) Gene set enrichment analysis (GSEA) was performed to determine the enrichment of inflammation (C) and apoptosis (D) target genes in the diabetic group. (E) Heatmap showing angiogenesis-related genes in the two groups (n = 3). (F) Schematic illustration demonstrating the difference in MSCs between patients with diabetes and healthy controls at the gene level.

        Open in viewer

        eMSCs are constructed to express mutant β-galactosidase

        Here, we first designed and constructed plasmids carrying the mutant β-galactosidase (A4-β-GalH363A) target gene and luciferase–red fluorescent protein (RFP) dual reporter genes, which could be further used for in vivo imaging. eMSCs expressing A4-β-GalH363A were constructed by infecting MSCs with lentiviruses obtained from human embryonic kidney 293T cells (Fig. 2A). Immunofluorescence staining for RFP confirmed that A4-β-GalH363A was successfully expressed by eMSCs (Fig. 2B). The subcellular fraction and intracellular distribution of enzymes expressed by the eMSCs was determined by Western blotting (Fig. 2C). In contrast to natural β-galactosidases, which are widely distributed within cells, A4-β-GalH363A was mainly confined to the nucleus of eMSCs.

        

        Fig. 2. Intracellular expression‎ and localization of A4-β-GalH363A.

        (A) Schematic diagram of lentivirus packaging and mesenchymal stem cell (MSC) infection. (B) Immunofluorescence staining of β-Gal and A4-β-GalH363A in eMSCs. Scale bar, 50 μm. (C) The distribution of two different enzymes in engineered MSCs (eMSCs) was analyzed by Western blotting.

        Open in viewer

        Intracellular release of NO is realized via decomposition of the 6-OMeGal-Ph-NO prodrug

        Since the mutant β-galactosidase was confined to the nucleus in eMSCs, we designed a prodrug by introducing a lipid-soluble self-decomposition chain into 6-OMeGal-NO to improve its oil and water distribution coefficient; therefore, the resultant NO donor 6-OMeGal-Ph-NO could penetrate the cell membrane and decompose and release NO under the catalysis of A4-β-GalH363A (Fig. 3A and Supplementary Materials). An in vitro release assay showed that the 6-OMeGal-Ph-NO prodrug was efficiently recognized and converted by A4-β-GalH363A with a cumulative release ratio of approximately 97.3%, while nearly no release was observed in the presence of wild-type β-galactosidase (Fig. 3B). To confirm‎ intracellular NO release, eMSCs were preincubated with an NO fluorescent probe (DAF-AM DA) and then treated with different NO prodrugs. The changes in fluorescence signals were examined by time-delay cell imaging (Fig. 3C). The results indicated that the fluorescence intensity continuously increased in the group that was treated with 6-OMeGal-Ph-NO, indicating conversion into NO (Fig. 3D). In contrast, no detectable changes were identified in the β-Gal-NO group because the high water solubility restricted its entry into eMSCs. Next, intracellular and extracellular release of NO in eMSCs was assessed (Fig. 3E). The quantity of intracellular NO was measured by electron paramagnetic resonance (EPR) using ferrous N-diethyl dithiocarbamate (DETC2-Fe) as the spin-trapping reagent. The resultant NO adduct (DETC2-Fe-NO) exhibited a characteristic triplet EPR signal (aN = 13.06 G, giso = 2.041) at room temperature. Quantitative analysis showed that the NO level was significantly (P < 0.001 or 0.0001) higher in eMSCs treated with 6-OMeGal-Ph-NO than in the β-Gal-NO and control groups (Fig. 3F). Furthermore, in the group treated with β-Gal-NO, the release of NO was mainly catalyzed by β-galactosidase that translocated from the cytoplasm in eMSCs, and the extracellular release profile of β-Gal-NO was confirmed by detecting the NO level in the cell culture medium with the NO-sensitive near-infrared fluorescence probe (23); it was significantly (P < 0.01) higher than that in the 6-OMeGal-Ph-NO and control groups (Fig. 3G). To determine the uptake of the NO prodrug by eMSCs, we incubated 6-OMeGal-Ph-NO with eMSCs, and the concentration in the culture medium was determined at different time points. The results reflected that approximately 45% of 6-OMeGal-Ph-NO was incorporated into the eMSCs within 12 hours (fig. S1).

        

        Fig. 3. Intracellular generation of nitric oxide (NO) from the NO prodrug under the catalysis of A4-β-GalH363A expressed by engineered mesenchymal stem cells (eMSCs).

        (A) Synthesis of two NO prodrugs with different enzyme response abilities and cellular permeabilities. (B) In vitro release profile of NO from the NO prodrug (6-OMeGal-Ph-NO) in the presence of β-Gal or A4-β-GalH363A. (C) Schematic illustration of intracellular NO imaging by using an NO fluorescence probe (DAF-AM DA). (D) Representative time-lapse images of NO generation from two different prodrugs in eMSCs and quantification of the fluorescence intensity (n = 6). ***P < 0.001, ****P < 0.0001 versus 6-OMeGal-Ph-NO group. (E) Schematic illustration showing the detection of intracellular and extracellular NO generation differentially. (F) Representative electron paramagnetic resonance (EPR) spectra and quantification of intracellular NO generation by measuring the DETC2-Fe-NO complex using 2,2,5,5-tetramethyl piperidine 1-oxyl (TEMPO) as a standard (n = 3). (G) Relative quantification of NO production in the medium determined using the near-infrared fluorescence probe (n = 4). Data are expressed as the mean ± SEM. Significant differences were detected by one-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

        Expand for more

        Open in viewer

        Intracellular delivery of NO inhibits apoptosis and enhances the proangiogenic activity of eMSCs

        Overproduction of reactive oxygen species due to cellular oxidative stress has been accepted as an important factor contributing to apoptosis in transplanted cells in ischemic tissue. Therefore, we assessed the protective effect of exogenously administered NO on the survival of eMSCs with H2O2-induced oxidative stress and focused on comparing the protection provided by intracellular and extracellular NO administration. The results showed that H2O2 stimulated apoptosis, and delivery of NO via extracellular and intracellular strategies significantly (P < 0.01 or 0.001) reduced apoptosis in eMSCs stimulated by oxidative stress, and the highest fluorescence signal was observed in response to intracellular NO delivery (Fig. 4A).

        

        Fig. 4. Intracellular delivery of nitric oxide (NO) inhibits apoptosis of engineered mesenchymal stem cells (eMSCs).

        (A) Bioluminescence imaging (BLI) was used to detect the effect of NO delivery on cell apoptosis stimulated by different concentrations of H2O2, and the fluorescence signals were further quantified (n = 3). **P < 0.01, ***P < 0.001, ****P < 0.0001 versus control group; #P < 0.05 versus β-Gal-NO group. (B) Flow cytometry assay of cell viability and apoptosis of eMSCs after H2O2 stimulation, and quantification of mean percent values of apoptotic cells (n = 3). (C) The expression‎ of apoptosis-related protein (BCL2, Bax, Bad, caspase3, and cleaved caspase3) by eMSCs was detected after H2O2 stimulation by Western blots (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001. (D) Representative BLI photographs reflecting the retention of eMSCs with and without intracellular NO delivery after in vivo transplantation as well as the quantitative analysis of signals (n = 3). Data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01.

        Open in viewer

        The survival of eMSCs after treatment with various NO delivery strategies was compared by using flow cytometry. Intracellular NO delivery had the most remarkable antiapoptotic effect on cells stimulated by oxidative stress (P < 0.001) (Fig. 4B). The expression‎ of apoptosis-related genes by eMSCs followed a similar trend; NO delivery (extracellular and intracellular) effectively promoted the expression‎ of the antiapoptotic gene BCL2 in eMSCs after oxidative stress stimulation, while the expression‎ levels of proapoptotic genes were reduced accordingly. Intracellular NO delivery via administration of the 6-OMeGal-Ph-NO prodrug exerted a more pronounced antiapoptotic effect at both the gene and protein levels (Fig. 4C and fig. S2A), which may be because the intracellular generation of NO directly activated the antioxidant system in cells to resist oxidative stress damage and inhibit further apoptosis.

        The expression‎ of proangiogenic genes, including ANGPT1, ANGPT2, FGF2, VEGFA, and KDR, in eMSCs was further detected by reverse transcription polymerase chain reaction (RT-PCR). The results showed that the expression‎ level of proangiogenic genes was significantly (P < 0.05, 0.01, or 0.001) higher in eMSCs treated with the 6-OMeGal-Ph-NO prodrug than in the other groups, indicating the enhanced proangiogenic functions of eMSCs after intracellular NO delivery (fig. S2B).

        We further eval‎uated the effect of NO delivery on the in vivo retention of eMSCs after orthotopic transplantation in the myocardial tissue of mice. As shown in Fig. 4D, intracellular delivery of NO effectively prolonged the retention of eMSCs within the myocardium, and an evident bioluminescence imaging (BLI) signal corresponding to the retention of eMSCs was observed 7 days after transplantation compared to the counterpart without administration of the NO prodrug. To further eval‎uate the translational potential of eMSCs in clinical settings, we used MSCs derived from diabetic patients and conducted a series of assays related to cell survival and paracrine function. The findings indicated that intracellular delivery of NO also confers advantages in the attenuation of cell apoptosis under stress conditions, thereby prolonging the in vivo retention of eMSCs (fig. S3).

        Intracellular delivery of NO ameliorates myocardial injury in MI mice after treatment with eMSCs

        The therapeutic efficacy of MSCs combined with exogenous NO was further eval‎uated in a mouse MI model (Fig. 5A and fig. S4). The inflammatory response in the early stage (3 days) was first detected by hematoxylin-eosin (H&E) staining and CD68 immunofluorescence staining. The results demonstrated that severe inflammatory cell infiltration occurred in the injured myocardium of MI mice, and it was effectively alleviated after eMSC treatment. More prominent restoration in the injured myocardium was observed after further administration of NO (Fig. 5, B and C), confirm‎ing the inhibitory effect on inflammation after MI provided by the combination of MSCs and NO. Moreover, this inhibitory effect was more significant (P < 0.01 or 0.0001) in the group with intracellular NO delivery than in the group with extracellular NO delivery.

        

        Fig. 5. Intracellular delivery of nitric oxide (NO) ameliorates myocardial injury in myocardial infarction (MI) mice after treatment with engineered mesenchymal stem cells (eMSCs) via regulation of the inflammation and stimulation of angiogenesis.

        (A) Experimental schedule for the treatment of MI in a mouse model. (B) Hematoxylin-eosin (H&E) staining was performed to detect inflammatory cell infiltration in the early stage of MI (n = 6). Scale bar, 100 μm. (C) Representative images of CD68 immunofluorescence staining (green) and quantification of CD68+ macrophages in injured myocardium (n = 6). Scale bar, 25 μm. (D and E) Flow cytometry was performed to detect peritoneal macrophage polarization 7 days after surgery followed by different treatments. TNFα- and CD206-positive ascites macrophages (markers of M1 and M2 macrophage phenotypes, respectively) were quantified accordingly (n = 3). (F) Representative images of α–Smooth muscle actin (α-SMA) immunofluorescence staining and the quantification of α-SMA+ arterioles (n = 6). Scale bar, 100 μm. (G) Representative images of von Willebrand factor (vWF) immunofluorescence staining and the quantification of vWF+ capillaries (n = 6). Scale bar, 100 μm. Data are expressed as the mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

        Expand for more

        Open in viewer

        It has been widely accepted that a higher proportion of M2-type macrophages is more conducive to the repair of tissue damage. Thus, peritoneal macrophages were extracted at 7 days after surgery and examined by flow cytometry to assess the polarization of macrophages in MI mice that received different treatments. MI modeling leads to a marked increase in the polarization of macrophages toward the M1 phenotype (TNF-α positive); however, the ratio of M1-type macrophages was moderately reduced after treatment with eMSCs. Further administration of NO via the intracellular delivery method significantly (P < 0.001 or 0.01) inhibited the polarization of macrophages toward the M1 phenotype while increasing the proportion of M2-type macrophages compared to the control and eMSC groups (Fig. 5, D and E). Additionally, we conducted immunofluorescence staining in heart section (fig. S5). The results revealed that intracellular delivery of NO also induces the polarization of macrophages into the M2 phenotype within the heart.

        In vitro studies demonstrated that exogenous NO could improve the proangiogenic capacity of eMSCs. Here, we further explored the influence of the combined delivery of exogenous NO and eMSCs on the reconstruction of the vascular network at the site of infarction. α–Smooth muscle actin (α-SMA)–positive arterioles and von Willebrand factor (vWF)–positive small vessels in MI mice after the different treatments were detected by immunofluorescence staining (Fig. 5, F and G). Treatment with eMSCs efficiently promoted angiogenesis in the injured myocardium, and more prominent enhancement was observed in response to further treatment with intracellular NO. This finding was further supported by the expression‎ of angiogenesis-related genes in the border zone of the infarcted heart (fig. S6).

        Intracellular delivery of NO improves heart function and inhibits adverse myocardial remodeling in MI mice after treatment with eMSCs

        Ultrasound and histological analyses were performed to eval‎uate the long-term recovery of cardiac function after MI. Cardiac injury was first eval‎uated by 2,3,5-triphenyltetrazolium chloride (TTC) staining (Fig. 6A). Treatment with eMSCs moderately repressed MI compared to the acute myocardial infarction (AMI) group, but a more pronounced inhibitory effect was observed in the group with further intracellular NO delivery. Left ventricular function was assessed by echocardiography at different time points. As shown in Fig. 6B, after 1 day of MI, the left ventricle in each group was markedly enlarged, cardiac function decreased rapidly, and deterioration of heart function continued for 28 days without detectable restoration in the AMI group. However, eMSC treatment could restore left ventricular systolic function and reduce ventricular dilation, as shown by the increase in left ventricular ejection fraction (LV-EF) and fraction shortening (LV-FS), as well as the decrease in left ventricular end-diastolic diameter (LVIDd) and left ventricular end-diastolic volume (LV-EDV) to a certain extent. In the group treated with eMSCs and intracellular NO delivery (NO-eMSCs), LV-EF and LV-FS were effectively recovered, while LVIDd and LV-EDV were significantly (P < 0.001 or 0.0001) enhanced compared to the AMI group.

        

        Fig. 6. Intracellular delivery of nitric oxide (NO) improves heart function and reduces adverse cardiac remodeling in myocardial infarction (MI) mice after treatment with engineered mesenchymal stem cells (eMSCs).

        (A) Representative images of 2,3,5-triphenyltetrazolium chloride (TTC) staining and quantification of the infarct area (n = 3). Scale bar, 2 mm. (B) Cardiac echo measurement was performed at different time points after surgery, and cardiac function indicators of left ventricular ejection fraction (LV-EF), left ventricular fractional shortening (LV-FS), left ventricular internal diameter at end diastole (LVIDd), and left ventricular end-diastolic volume (LV-EDV) were eval‎uated accordingly (n = 6). ***P < 0.001, ****P < 0.0001 versus acute myocardial infarction (AMI) group. (C) Masson’s trichrome staining was performed, and the infarct size was quantified accordingly (n = 6). (D) Collagen deposition in the hearts was detected by Sirius Red staining (n = 6). Scale bar, 100 μm. (E) Representative images of wheat germ agglutinin (WGA) immunofluorescence staining and quantification of the cross-sectional area of cardiomyocytes (n = 6). Scale bar, 50 μm. (F) Representative images of immunofluorescence staining (red) for the gap junction protein (Cx43) and the quantification of the intensity of red fluorescence to the whole area of images (n = 6). Scale bar, 25 μm. Data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

        Expand for more

        Open in viewer

        Masson staining was used to detect the degree of myocardial fibrosis after MI. Severe myocardial fibrosis was observed in MI mice compared to the sham group (Fig. 6C). In addition, the ventricular wall became thinner, which was closely related to impaired left ventricular systolic function after MI, as demonstrated by echocardiography. However, these adverse cardiac remodeling events were ameliorated after eMSC treatment and accompanied by a lowered degree of myocardial fibrosis and a thickened ventricular wall in contrast to the AMI group. Notably, treatment with eMSCs and intracellular NO delivery exerted the most prominent inhibitory effect on myocardial fibrosis after MI (Fig. 6C).

        Collagen deposition in MI mice was assessed by Sirius staining (Fig. 6D), and the results showed that MI resulted in severe collagen deposition in the injured myocardium compared to the sham group. It was effectively reduced after eMSC treatment, and the inhibitory effect of eMSC plus intracellular NO delivery was significantly higher than that in the other two groups (P < 0.001 or 0.0001).

        Next, wheat germ agglutinin (WGA) staining was carried out to eval‎uate myocardial cell hypertrophy 28 days after MI (Fig. 6E). The cross-sectional area of cardiomyocytes was increased in MI mice in contrast to the sham operation group due to compensatory hypertrophy in the heart to maintain the normal rate of cardiac ejection. Hypertrophy was significantly (P < 0.05) mitigated after treatment with eMSCs, especially in the presence of exogenous NO (P < 0.001), indicating an ideal therapeutic effect on inhibiting myocardial cell hypertrophy and adverse ventricular remodeling by the combination of eMSCs and NO.

        Gap junctions (GJs) are the main connections between cardiomyocytes in the heart, and Cx43 is the main GJ protein in ventricular muscle in the heart (24). Studies have shown that the absence of Cx43 leads to the occurrence of cardiac ventricular arrhythmia, which can develop into heart failure (25). After 28 days of MI, immunofluorescence staining for Cx43 revealed abundant and uniform distribution of GJ proteins in the sham group, whereas MI injury resulted in a marked decrease in the expression‎ of Cx43 (Fig. 6F). Despite the moderate inhibitory effect provided by the administration of eMSCs, further delivery of NO via the intracellular method significantly enhanced (P < 0.001 or 0.01) the expression‎ of Cx43 compared to that in the AMI or eMSC groups.

        Intracellular delivery of NO enhances the therapeutic efficacy of eMSCs in a rat MI model

        Although the outcome in a mouse model supported the beneficial effect of NO via intracellular delivery on enhancing the therapeutic efficacy of MSCs for MI, immediate administration of stem cells after MI is different from the clinical treatment of MI due to the limitation of the administration paradigm in mouse models. In addition, 3 to 7 days after MI is the outbreak period of the inflammatory response. For this reason, we established a rat model of MI and conducted secondary thoracotomy 3 days after surgery (Fig. 7A), and eMSCs were delivered via an injectable HA hydrogel as the carrier (Fig. 7B). Lactate dehydrogenase (LDH), a crucial marker for assessing the extent of myocardial damage, exhibited an initial elevation within 2 to 48 hours following the onset of MI, reaching its zenith between 2 and 5 days after MI. We collected blood samples from the orbital venous plexus of rats 5 days after MI to measure serum LDH levels (fig. S7). The findings revealed a sharp increase in serum LDH levels due to MI. However, treatment with NO-eMSCs significantly reduced serum LDH levels, indicating an attenuation of cardiac injury.

        

        Fig. 7. Intracellular delivery of nitric oxide (NO) enhances the therapeutic efficacy of engineered mesenchymal stem cells (eMSCs) in a rat myocardial infarction (MI) model.

        (A) Experimental schedule for the treatment of rat MI. (B) Representative images showing the second thoracotomy in rats after MI. (C) Cardiac echo measurement was performed at different time points after surgery, and cardiac function indicators of left ventricular ejection fraction (LV-EF), left ventricular fractional shortening (LV-FS), left ventricular internal diameter at end diastole (LVIDd), and left ventricular end-diastolic volume (LV-EDV) were eval‎uated accordingly (n = 6). **P < 0.01, ***P < 0.001 versus acute myocardial infarction (AMI) group. (D) Representative images of Masson’s trichrome staining and quantification of the infarct size and infarct thickness (n = 6). **P < 0.01, ***P < 0.001, ****P < 0.0001 versus AMI group. (E) Representative images of wheat germ agglutinin (WGA) immunofluorescence staining and quantification of the cross-sectional area of cardiomyocytes (n = 6). Scale bar, 50 μm. (F) Representative images of α–Smooth muscle actin (α-SMA) immunofluorescence staining and quantification of α-SMA+ arterioles (n = 6). Scale bar, 100 μm. Data are expressed as the mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

        Open in viewer

        Echocardiography was performed to detect heart function at different time points after MI (Fig. 7C). At 4 weeks after surgery, the anterior wall of the left ventricle was completely infarcted, and the ventricular cavity was increased dramatically. Quantitative analysis further indicated that both EF and FS were decreased markedly after surgery. Treatment with MSCs could restore the systolic function of the heart, and the highest recovery rate (60 to 80%) was observed in the group treated with eMSCs and intracellular NO delivery. Ventricular dilatation was also effectively restored after combined treatment with eMSCs and NO, which was consistent with the results obtained in the mouse model.

        Histological analyses showed that treatment with eMSCs combined with NO efficiently reduced the degree of fibrosis (Fig. 7D) and collagen deposition (fig. S8A). Furthermore, it alleviated infarct size after MI and restored ventricular cavity morphology of the heart with a significant enhancement in the thickness of the infarcted left ventricular wall and the interventricular septum (IVS), as eval‎uated by Masson (Fig. 7D) and H&E staining (fig. S9). Twenty-eight days after MI, WGA staining also showed that the combination of eMSCs and NO significantly (P < 0.01) inhibited cardiomyocyte hypertrophy in contrast to the AMI group, thus inhibiting further myocardial systolic dysfunction (Fig. 7E). The expression‎ of the GJ protein Cx43 (fig. S8B) followed a similar trend to that in mouse models; that is, acute MI led to a marked decrease in the distribution of Cx43 in the myocardium, and treatment with eMSCs up-regulated the expression‎ of Cx43 to a certain extent. Further delivery of NO via the intracellular method produced a more significant (P < 0.05 or 0.01) effect on the up-regulation of Cx43 in the injured myocardium compared to the eMSC and AMI groups, thereby facilitating the connection between cardiomyocytes and further inhibiting the occurrence of arrhythmias and heart failure.

        Tissue repair after MI is often closely related to angiogenesis, which begins at the infarct border and extends to the center of the infarction (26). Therefore, we further compared the proangiogenic effect of eMSCs with and without NO delivery on the damaged margin of the infarcted myocardium 28 days after MI in rats (Fig. 7F and fig. S8C). The combination of eMSCs and NO remarkably promoted angiogenesis and reconstruction of the vascular network compared to the group treated with eMSCs only, which is beneficial to the repair of myocardial injury after MI.

        DISCUSSION

        Cell therapy based on MSCs has proven to be a powerful solution for treating degenerative diseases and tissue damage (27–29). Despite the advantages of autologous stem cells over allogeneic stem cells, including the absence of immune rejection, the decreased survival and impaired paracrine functions of stem cells collected from patients with chronic diseases greatly limit their clinical use (7, 30, 31). Here, we first showed that MSCs collected from patients with diabetes exhibited marked up-regulation of apoptosis- and inflammation-related genes and down-regulation of proangiogenic genes, affecting the efficacy of cell therapy. As a result, genetic engineering strategies are often required to enhance the therapeutic efficacy of autologous stem cells. A recent study revealed that eMSCs, engineered to express PD-L1 on their surface and secrete CTLA4-Ig (immunoglobulin) as an extracellular factor, exhibited immunoprotective properties, which improved the outcome of both syngeneic and allogeneic islet transplantation in diabetic mice (32).

        NO is involved in a variety of physiological processes. Studies have shown that as an important signaling molecule, NO plays a pivotal role in regulating stem cell behavior (33–35), including cell survival, migration, differentiation, and paracrine behavior. These factors affect the interaction of stem cells with other cells and the tissue microenvironment. Previously, different types of NO-releasing biomaterials, such as injectable hydrogels, have been prepared by us and other groups (36–40), and further studies have shown that the combination of NO and MSCs is more effective in treating various diseases than MSC therapy alone. In addition, it has been reported that pretreatment of MSCs with NO-releasing biomaterials could enhance the therapeutic efficacy of MSCs and their secreted exosomes because of their enhanced proangiogenic functions (41).

        Due to the spatiotemporal characteristics of NO (42), precise delivery of NO in a site-specific and controllable manner holds great importance in the regulatory effect of exogenously administered NO. In addition to the controlled release rate, the site at which NO is generated is also a key factor due to the relative half-life and limited diffusion distance (43, 44). It is reasonable to speculate that intracellular and extracellular NO delivery may lead to different outcomes when regulating the survival and function of MSCs. In our previous work, an enzyme-prodrug delivery system was designed based on a bump-and-hole strategy (21). The mutant galactosidase (A4-β-GalH363A) enables the targeted delivery of NO, thus reducing the side effects due to the unspecific decomposition of the NO prodrug and enhancing the therapeutic efficacy. Here, we transfected a plasmid expressing mutant galactosidase into MSCs and successfully constructed eMSCs. The enzyme expressed by MSCs could catalyze the decomposition of the 6-OMe-galactose–protected NO prodrug and release NO intracellularly.

        Western blotting and fluorescence imaging demonstrated that the expression‎ of the engineered enzyme was confined to the nucleus of MSCs, while wild-type β-galactosidase was widely distributed in the cytoplasm, including the lysosome and perinuclear region (45, 46). Since the corresponding prodrug for wild-type β-galactosidase is highly hydrophilic, it fails to enter MSCs and releases NO extracellularly by enzymes that translocate from the cell. In contrast, the prodrug for mutant galactosidase is cell penetrating because of the modified molecular structure; therefore, it can enter MSCs and release NO intracellularly under the catalysis of the corresponding enzyme expressed by the cells. Accordingly, two different NO delivery paradigms were successfully developed in this study and further confirmed by a series of eval‎uations, including cell imaging and electronic paramagnetic resonance. Further in vitro and in vivo assays indicated that in contrast to extracellular NO delivery, intracellular administration of NO enhanced cell survival and the paracrine effects of MSCs, including inhibiting apoptosis and supporting angiogenesis.

        Next, we established a mouse MI model to systematically eval‎uate the therapeutic efficacy of MSCs combined with exogenous NO. The results showed that intracellular delivery of NO prolonged the retention of eMSCs after myocardial orthotopic transplantation. In addition, the combination of eMSCs and intracellular NO delivery improved cardiac function after MI and reduced adverse ventricular remodeling compared to the group treated with MSCs only. Additionally, it could effectively restore the reconstruction of the blood vessel network and further promote the repair of the infarcted myocardium.

        To gain further insight into the translational potential of the combinatory therapeutic strategy developed in this study, a rat model of MI was established, and MSCs were administered by a second thoracotomy after 3 days to mimic the clinical use of MSCs for the treatment of MI (47, 48). Clinically, acute MI is typically due to the rupture of coronary atherosclerotic plaque and the formation of thrombus, which causes coronary artery obstruction. After the acute phase of MI, adverse ventricular remodeling further affects the prognosis of patients, which is specifically characterized as a decrease in ventricular wall thickness and myocardial tension in the MI area, myocardial hypertrophy in the noninfarction area, and a change in the morphology of the ventricular cavity, thus leading to arrhythmia and further development into heart failure. The efficacy of MSCs in managing arrhythmias remains a topic of ongoing debate. Some researchers argue that MSCs do not appear to reduce or prevent arrhythmias, with the antiarrhythmic or proarrhythmic potential of MSCs primarily relying on paracrine factors (49). Conversely, other studies suggest that MSCs themselves may play a role in the post-MI recovery process (50). In our study, we observed an evident up-regulation of Cx43 expression‎ after NO-eMSC treatment, which is a potential target associated with antiarrhythmic effects. Further investigation is still required to comprehensively explore the antiarrhythmic potential of NO-eMSCs. In line with the enhanced therapeutic efficacy in the mouse model, intracellular delivery of NO showed enormous advantages in the rat MI model by inhibiting apoptosis and enhancing the paracrine function of MSCs.

        In summary, we first showed that survival and paracrine function were reduced in MSCs collected from patients with diabetes, which could greatly affect therapeutic efficacy. Accordingly, eMSCs were successfully constructed, and the mutant β-galactosidase expressed by the cells enabled the intracellular generation of NO via the conversion of an exogenous NO prodrug. In vitro and in vivo assays indicated that intracellular delivery of NO effectively enhanced the survival of transplanted MSCs and promoted the paracrine function of MSCs, which was further confirmed by the enhanced therapeutic efficacy in mouse and rat models of MI compared to the group treated with MSCs only. This synergistic strategy provides an option for the treatment of MI by autologous MSCs in the clinic.

        MATERIALS AND METHODSRNA sequencing analysis

        RNA sequencing was performed by the BGI (Shenzhen, China). Briefly, RNA from the ADMSCs of healthy people and patients with diabetes was extracted using TRIzol reagent (Yeasen, China). RNA samples were sequenced on the BGISEQ platform. The raw data containing low-quality reads, adaptor sequences, and high levels of N bases were filtered before analysis. Then, the clean reads were mapped to the reference genome using HISAT, and Bowtie2 was used to align the clean reads to the reference genes. The reference genome source is National Center for Biotechnology Information (NCBI), and the reference genome version is GCF_000001405.39_GRCh38.p13. The expression‎ levels of genes were quantified to identify differentially expressed genes by RNA-Seq by expectation maximization (RSEM). The analyses of hierarchical clustering and heatmap were performed using the online Dr. Tom system (biosys.bgi.com) to compare differential gene expression‎ of ADMSCs in healthy people and patients with diabetes. According to the KEGG_pathway annotation classification, the phyper function in R software was used for enrichment analysis, the P value was calculated, and then false discovery rate (FDR) was performed on the P value to obtain a Q value. Generally, a Q value of ≤0.05 was regarded as significant enrichment. GSEA was used to analyze significant differences in gene expression‎ between inflammatory and apoptosis-related pathways. Expression‎ cluster heatmap was used to analyze the expression‎ of genes associated with angiogenesis.

        Measurement of NO release

        The NO-releasing profile was determined by the Griess kit assay. In brief, 50 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were dissolved in phosphate-buffered saline (PBS) buffer (pH 7.4), and enzymes were added to the solutions at a concentration of 0.005 mg/ml. At each predetermined time interval, 50 ml of solution was transferred into a 96-well plate, and 50 ml of Griess I and 50 ml of Griess II were added thereafter. The azo compound of purple color was formed, and the absorbance was measured at a wavelength of 540 nm using an iMark microplate reader (Bio-Rad, USA).

        Cell cultureMesenchymal stem cells

        MSCs derived from human umbilical cord were obtained from Health-Biotech, maintained in Dulbecco's modified Eagle’s medium (DMEM)/F12 (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, USA) and 1% penicillin- streptomycin solution (Gibco, USA), and placed in a cell culture chamber containing 5% CO2 at 37°C.

        Human embryonic kidney 293T cells

        Human embryonic kidney (HEK) 293T cells were obtained from the American Type Culture Collection (ATCC), maintained in high-glucose DMEM (Gibco, USA) with 10% FBS (Gibco, USA) and 1% penicillin-streptomycin solution (Gibco, USA), and placed in a cell culture chamber containing 5% CO2 at 37°C.

        Construction of eMSCs

        The coding sequence of mutant β-galactosidase (A4-β-GalH363A) can be obtained from the previous publication (21). The lentivirus packaging system containing A4-β-GalH363A sequence and Rluc-RFP sequence was constructed by Wuhan Miaolingbio Co. Ltd. The constructed lentivirus plasmid containing the target gene and the package gene (psPAX2 and pMD2.G) was transfected into HEK 293T cells through Lipo2000, and the supernatant was collected to obtain the virus solution. After removing impurities, the virus solution was mixed 1:1 with fresh MSC medium, and polybrene (10 μg/ml) was added. MSCs were infected with virus through incubation in the mixture medium. The infection efficiency was observed under an inverted fluorescence microscope, and the expression‎ of target protein was determined by Western blotting.

        Cell immunofluorescence staining

        eMSCs were inoculated in 24-well plates. Cells were fixed with 4% paraformaldehyde and blocked in 4% bovine serum albumin in PBS for 30 min at room temperature. Then, the cells were incubated with primary antibodies overnight at 4°C. The bound primary antibodies were displayed by incubation with the secondary antibodies for 2 hours at room temperature. Cells were counterstained with DAPI (4′,6-diamidino-2-phenylindole)–containing Fluoromount-G and visualized with a fluorescence microscope (Axio Imager Z1). Antibodies used include anti–β-galactosidase (1:100, A1863, Abclonal) and anti-RFP (1:100, PA1-986, Invitrogen).

        Western blot

        eMSCs were collected, and total protein was extracted using radioimmunoprecipitation assay (RIPA) lysate containing protease inhibitor (Solarbio, China). Cytoplasmic protein and nucleoprotein were extracted using a nucleoprotein extraction kit containing protease inhibitors (Solarbio, China). The protein concentration was quantified using a BCA protein assay kit (Solarbio, China). The samples were diluted with 4× SDS–polyacrylamide gel electrophoresis (PAGE) loading buffer and boiled in boiling water for 8 min. Then, 30 μg of protein was isolated by 10% SDS-PAGE electrophoresis. The isolated proteins were transferred to an Immobilon-P Transfer membrane (Millipore, USA) and incubated with the primary antibody overnight at 4°C and then with the secondary antibody at room temperature for 2 hours. The bands were detected with chemiluminescent horseradish peroxidase substrate (Millipore, USA). Signals were generated by using an enhanced chemiluminescence (ECL) reagent (Millipore, USA) and were captured by using the Tanon-5200 Chemiluminescence Imaging System (Tanon, China). The antibodies used included anti–β-galactosidase (1:1000, A1863, Abclonal), anti–His-tag (1:1000, 12698S, Cell Signaling Technology), anti–glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (1:1000, AC001, Abclonal), anti–histone H3 (1:1000, 4499S, Cell Signaling Technology), anti–β-actin (1:1000, UM4001, Utibody), anti-BCL2 (1:1000, WL01556, Wanleibio), anti-Bax (1:1000, 50599-2-lg, Proteintech), anti-Bad (1:1000, WL02140, Wanleibio), and anti-caspase/cleaved caspase3 (1:1000, WL02117, Wanleibio).

        6-OMeGal-Ph-NO uptake by eMSCs

        eMSCs were inoculated in 12-well plates, and 50 μM substrate (6-OMeGal-Ph-NO) was added per well. At predetermined time points (0, 1, 3, 6, and 12 hours), an appropriate amount of culture medium was collected, and an excess of A4-β-GalH363A was added immediately to fully catalyze the decomposition of the remaining substrate at room temperature. The amount of 6-OMeGal-Ph-NO substrate in the culture medium was determined by Griess kit assay.

        Real-time imaging of intracellular NO

        The fluorescence emission associated with NO in the cytosol was detected by an electron-multiplying charge-coupled device (DU-897D-CS0-BV; Andor, Belfast, UK) connected to an inverted fluorescence microscope (Axio Observer D1; Carl Zeiss, Oberkochen, Germany). Intracellular NO imaging was performed using an NO fluorescence probe, DAF-AM DA (Beyotime, China), according to the manufacturer’s instruction. eMSCs were inoculated in small confocal dishes in advance, and the experiment was conducted when the cell density reached 80%. First, the medium was collected for later use. After two gentle washes with PBS, 5 μM DAF-AM DA solution was incubated at 37°C for 30 min in the dark. Then, cells were gently washed with PBS twice. The previously collected medium was added anew, and cell imaging was performed immediately. At the 488-nm excitation wavelength, pictures were taken every 5 s. After stable shooting for 2 min, 100 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were added. The filming for cell fluorescence was continued for 4 min. The final fluorescence intensity was determined without the background fluorescence value. The proportion of change in fluorescence intensity of each cell in the visual field was calculated.

        Intracellular NO detection

        Intracellular NO radicals (NO•) were detected using EPR as described (51, 52). In brief, sodium DETC (4.5 mg) and FeSO4•7H2O were dissolved in two separate volumes (10 μl) of deoxygenated Krebs/Hepes solution. Equal volumes of these parent solutions were rapidly mixed and aspirated into Eppendorf combi tips. The 0.5 mM Fe•(DETC)2 colloid solution had a yellow-brownish color with a slight opalescence in light. No aggregate formation was observed, at least during the first 30 min. eMSCs were rinsed with modified Krebs/Hepes buffer and incubated with freshly prepared NO•-specific spin trap Fe•(DETC)2 colloid (0.5 mM) for 30 min. Meanwhile, 100 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were added to the buffer. Gently collected cell suspensions were snap-frozen in liquid nitrogen. Ethyl acetate (200 μl) was added, and the cells were ultrasonically broken to extract DETC2-Fe-NO. The ethyl acetate extract was concentrated with nitrogen and transferred to a 50-μl capillary, and then the X-band EPR was measured at room temperature. The following acquisition parameters were used: modulation frequency, 100 kHz; microwave power, 10 mW; modulation amplitude, 2 G; number of scans, 60. The double-integrated area of the EPR spectra was calibrated into concentrations of DETC2-Fe-NO using TEMPO (2,2,5,5-tetramethyl piperidine 1-oxyl) as a standard. EPR spectral simulation was conducted by the WINSIM program.

        Extracellular NO detection

        eMSCs were treated with β-Gal-NO or 6-OMeGal-Ph-NO (30 μM). The production of NO in the medium of each group was detected 6 hours after incubation with NO-sensitive near-infrared fluorescence probe (5 μM). The NO production of medium in different groups was compared by the relative fluorescence intensity under the excitation at 750 nm (emission at 800 nm).

        Cell apoptosis detection

        To test the protective effect of NO delivery on cellular oxidative stress stimulation, 30 μM NO substrate (6-OMeGal-Ph-NO) was added to the medium in advance. Then, H2O2 with different concentrations (100, 200, 400, and 600 μM) was added to stimulate the lentivirus-infected eMSCs. BLI was performed immediately after addition of the luciferase substrate coelenterin to eval‎uate cell apoptosis. Additionally, eMSCs treated with 200 mM H2O2 were stimulated for 24 hours to induce cell apoptosis. An Annexin V/PI assay kit (Solarbio) was used to detect eMSC apoptosis.

        BLI detection of cell retention

        BLI and luciferase substrates were used in mice to eval‎uate the retention of NO-eMSCsGluc/RFP in cardiac orthotopic transplantation. The mice after eMSC injections were anesthetized with 1.5% isoflurane and injected with coelenterin through the caudal vein at 150 mg/kg. After injection, the mice were immediately placed in a BLI system to detect cell retention in the myocardium.

        Animals

        C57BL/6 mice (male, 8 weeks old) and Sprague-Dawley rats (male, 8 weeks old) were purchased from SPF (Beijing) Biotechnology Co. Ltd., Beijing, China. Animals were randomly grouped for treated and untreated controls. All experiments and animal procedures were approved by the Animal Experiments Ethical Committee of Nankai University and carried out in conformity with the Guide for Care and Use of Laboratory Animals.

        MI in mice and rats

        Surgical induction of MI was performed on C57BL/6 mice (male, 8 weeks old) as previously described with some modifications. Briefly, mice were anesthetized with 2% isoflurane, followed by fixation to a heating pad (37°C) at supine position, and then ventilated with a mechanical ventilation system (Hallowell EMC Microvent I, USA) set at 110 breaths per minute with a tidal volume of 0.2 to 0.3 ml. The third intercostal space over the left chest and the heart was exposed. After left thoracotomy, the left anterior descending coronary artery was ligated with a 6-0 silk ligature. The left main descending coronary artery was sutured and tied with a slip knot at a site approximately 3 mm from its origin. Cardiac ischemia was confirmed by the presence of myocardial blanching. Immediately, eMSCs encapsulated with HA hydrogel were injected into the myocardium of mice through three-point injection around the infarct zone. The chest cavity was closed to restore negative pressure and prevent pneumothorax.

        Similar MI surgery was performed on Sprague-Dawley rats (male, 8 weeks old) first. Briefly, rats were anesthetized via intraperitoneal injection of 10% chloral hydrate (350 mg/kg), followed by fixation to a heating pad (37°C) at supine position. Then, they were ventilated with a mechanical ventilation system (Hallowell EMC Microvent I, USA) set at 110 breaths per minute with a tidal volume of 6 ml. The third intercostal space over the left chest and the heart was exposed. After left thoracotomy, the left anterior descending coronary artery was ligated with a 6-0 silk ligature. The left main descending coronary artery was sutured and tied with a slip knot at a site approximately 3 mm from its origin. Cardiac ischemia was confirmed by the presence of myocardial blanching. The chest cavity was closed to restore negative pressure and prevent pneumothorax. Three days after surgery, secondary thoracotomy was performed, and eMSCs were injected into the myocardium around the infarct zone. The chest cavity was closed to restore negative pressure and prevent pneumothorax.

        Mice and rats in the AMI group only received MI without eMSC injection, while sham-operated mice only experienced thoracotomy without MI surgery or eMSC injection.

        At 1, 3, 5, and 7 days after myocardial injection of eMSCs, the prodrug was injected through the tail vein. Mice were injected with 100 μl of prodrug (1 mg/ml) each time, and rats were injected with 200 μl of prodrug (1 mg/ml) each time.

        TTC staining

        Two days after surgery, a thoracotomy was performed. The heart was quickly excised after quick freezing for 15 min and sliced at 1 mm thickness. Afterward, the sections were incubated with 1.5% TTC (Sigma-Aldrich) solution at 37°C in an incubator for 15 to 30 min and then with a 4% formaldehyde solution for 2 hours. The normal myocardial tissue was red, while the ischemic myocardium was white. The size of the infarcted myocardium, which was white or pale, was measured by ImageJ software.

        Cardiac function assessment

        Transthoracic echocardiography was performed with the Vevo 2100 Imaging System (Fuji Film Visual Sonics Inc., Canada) equipped with an MS-250/400 imaging transducer. The baseline cardiac function of mice and rats was measured at 3 days before surgery. Cardiac function was analyzed at days 1 and 28 after MI surgery with different treatments, as reported previously. Mice or rats were slightly anesthetized in a box with isoflurane. Their limbs were fixed in a supine position on the echo mat, and the chest hair was removed by depilating cream. Then, mice or rats were anesthetized by inhalation of isoflurane (0.5 to 1%) mixed with oxygen to maintain the heart rate at approximately 500 to 600, and M-mode echocardiography was performed. The left ventricular internal diameter at end-diastole (LVIDd) and systole (LVIDs) were obtained by measuring the long axis and the short axis. Accordingly, the cardiac parameters LV-EF, LV-FS, LV-EDV, and LV end-systole volume (LV-ESV) were determined. The echocardiography measurement was carried out in a double-blind manner.

        Histological analysis

        At the indicated time points, mice and rats were anesthetized via intraperitoneal injection of chloral hydrate, and a thoracotomy was performed. The hearts were fixed with trans-cardiac perfusion of saline and immersed in 4% paraformaldehyde over 24 hours. The heart tissue samples were dehydrated with gradient alcohol and xylene, embedded in paraffin blocks, and cut into sections in 5 μm thickness.

        The paraffin-embedded sections were stained with Masson trichrome, H&E, and Sirius Red following a standard protocol. Immunofluorescence staining was performed on paraffin-embedded sections of the heart tissue samples. After deparaffinization and heat-mediated antigen retrieval‎ in citrate solution, the samples were washed with PBS three times and incubated with blocking serum, which was used to avoid nonspecific binding, at room temperature for 30 min. The sections were incubated with specific antibodies diluted in goat serum at 4°C overnight. On the second day, the sections were rewarmed at room temperature for 1 hour and washed with PBS three times. Afterward, the sections were incubated with Alexa Fluor–coupled secondary antibodies for 2 hours at room temperature. After washing with PBS, the sections were counterstained with DAPI-containing Fluoromount-G (SouthernBiotech, USA) and coverslipped. The antibodies used included anti–α-SMA (1:100, ab5694, Abcam), anti-vWF (1:100, ab6694, Abcam), anti–α-actinin (1:100, ab9475, Abcam), anti-Connexin43 (1:100, ab11370, Abcam), anti-CD68 (1:100, ab125212, Abcam), WGA (1:500, FL-1021, Novus Biologicals), anti-iNOS (1:100, ab178945, Abcam), and anti-CD206 (1:100, ab64693, Abcam).

        Macrophage isolation and detection

        Three days before euthanasia, mice were intraperitoneally injected with 2 ml of 4% thioglycolate. Three days later, the mice were sacrificed by cervical dislocation and immersed in 75% alcohol and then transferred to an ultraclean workbench. The mouse limb was fixed in the supine position, and the mouse abdominal wall was carefully cut open with the peritoneal. PBS [1% penicillin-streptomycin (PS)] was injected intraperitoneally to collect the cell suspension, which was centrifuged at 2000 rpm for 10 min. After discarding the supernatant, the cells were incubated with anti-F4/80/TNF-α and anti-F4/80/CD206 antibodies. FlowJo software was used to analyze the results of flow cytometry.

        Quantitative real-time PCR

        Total RNA samples from the cells were prepared using TRIeasy Total RNA Extraction Reagent (Yeasen, China) according to the manufacturer’s instructions. Heart tissue samples were collected at the indicated time points after MI surgery.

        The tissue samples were dissected at the border zone of the left ventricle and frozen in liquid nitrogen immediately. Afterward, the total RNA was extracted with TRIzol reagent, as mentioned before. The concentration of the RNA was measured with a NanoDrop spectrophotometer (NanoDrop Technologies, USA). The complementary cDNA was synthesized using a first-strand cDNA synthesis kit (Yeasen, China). Quantitative real-time PCR was performed on a CFX96 real-time PCR system (Bio-Rad, Hercules, USA) with an SYBR Green–based real-time detection system (Yeasen, China). The relative gene expression‎ of mRNA was expressed as 2−(△△CT) and normalized to GAPDH as an internal control. Each reaction was performed in triplicate to obtain an average value, and the changes in relative gene expression‎ normalized to the internal control levels were determined. The highly purified primers used in this experiment were commercially synthesized (Sango, China). The sequences of the primers used in this experiment are summarized in the Supplementary Materials.

        Statistics

        All data are presented as the mean ± SEM from at least three independent experiments. Comparisons between two groups were performed by Student’s t test, and comparisons among more than two groups were performed by one-way or two-way analysis of variance (ANOVA). Statistical analyses were performed with GraphPad Prism software 7.0, and a statistical significance level of less than 0.05 was accepted.

        Acknowledgments

        Funding: This study is supported by the National Key R&D Program of China (2018YFE0200503), the National Natural Science Foundation of China (nos. 81925021, 82330066, 81921004, and U2004126), and the Tianjin Natural Science Foundation (21JCZDJC00240).

        Author contributions: Q.Z. and Z.L. conceived the original concept and initiated this project. Q.Z., Z.L., and F.G. designed the experiment and supervised the entire project. S.W. collected human adipose mesenchymal stem cells. M.Q. synthesized all NO prodrugs and probes. P.L. prepared engineered enzymes. T.H., G.J., and Q.X.L. established mouse and rat MI models. T.H. and G.J. performed histological analysis. G.J. and W.D. carried out in vitro cell experiments. S.D. carried out NO cell imaging under the supervision of L.P. T.H., G.J., and M.Q. analyzed data under the supervision of Q.Z. H.H. helped with lentivirus packaging and cell infection. W.G. and T.L. helped in establishing animal MI models. Y.W., J.H., J.C., and J.T. helped with data collection. T.H. and Q.Z. wrote the paper with input from other authors.

        Competing interests: The authors declare that they have no competing interests.

        Data and materials availability: All data needed to eval‎uate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Transcriptome sequencing dataset is available at https://doi.org/10.5061/dryad.tqjq2bw5b.

        Supplementary MaterialsThis PDF file includes:

        Supplementary Text

        Figs. S1 to S9

        Table S1

        • Download
        • 1.94 MB

        REFERENCES AND NOTES

        1

        M. F. Pittenger, A. M. Mackay, S. C. Beck, R. K. Jaiswal, R. Douglas, J. D. Mosca, M. A. Moorman, D. W. Simonetti, S. Craig, D. R. Marshak, Multilineage potential of adult human mesenchymal stem cells. Science 284, 143–147 (1999).

        GO TO REFERENCE

        Crossref

        PubMed

        Web of Science

        Google Scholar

        2

        A. Uccelli, L. Moretta, V. Pistoia, Mesenchymal stem cells in health and disease. Nat. Rev. Immunol. 8, 726–736 (2008).

        GO TO REFERENCE

        Crossref

        PubMed

        Web of Science

        Google Scholar

        3

        Y. Wang, J. Fang, B. Liu, C. Shao, Y. Shi, Reciprocal regulation of mesenchymal stem cells and immune responses. Cell Stem Cell 29, 1515–1530 (2022).

        GO TO REFERENCE

        Crossref

        PubMed

        Web of Science

        Google Scholar

        4

        Y. Wang, X. Chen, W. Cao, Y. Shi, Plasticity of mesenchymal stem cells in immunomodulation: Pathological and therapeutic implications. Nat. Immunol. 15, 1009–1016 (2014).

        GO TO REFERENCE

        Crossref

        PubMed

        Web of Science

        Google Scholar

        Show all references

        (0)eLetters

        eLetters is a forum for ongoing peer review. eLetters are not edited, proofread, or indexed, but they are screened. eLetters should provide substantive and scholarly commentary on the article. Neither embedded figures nor equations with special characters can be submitted, and we discourage the use of figures and equations within eLetters in general. If a figure or equation is essential, please include within the text of the eLetter a link to the figure, equation, or full text with special characters at a public repository with versioning, such as Zenodo. Please read our Terms of Service before submitting an eLetter.

        Log In to Submit a Response

        No eLetters have been published for this article yet.

        Recommended articles from TrendMD

        • In vivo priming of human mesenchymal stem cells with hepatocyte growth factor–engineered mesenchymal stem cells promotes therapeutic potential for cardiac repai...

          Bong-Woo Park, Sci Adv, 2020

        • Shattering barriers toward clinically meaningful MSC therapies

          Oren Levy, Sci Adv, 2020

        • Nanovesicles derived from iron oxide nanoparticles–incorporated mesenchymal stem cells for cardiac repair

          Ju-Ro Lee, Sci Adv, 2020

        • Exendin-4 gene modification and microscaffold encapsulation promote self-persistence and antidiabetic activity of MSCs

          Yuanyuan Zhang, Sci Adv, 2021

        • Human gingival mesenchymal stem cells retain their growth and immunomodulatory characteristics independent of donor age

          Jay R. Dave, Sci Adv, 2022

        • The secretion profile of mesenchymal stem cells and potential applications in treating human diseases

          Yuyi Han, Signal Transduction and Targeted Therapy, 2022

        • Mesenchymal Stem Cell Therapy for Ischemic Heart Disease: Systematic Review and Meta-analysis

          Hyunsuk Jeong, International Journal of Stem Cells

        • Mesenchymal Stem Cell and MicroRNA Therapy of Musculoskeletal Diseases

          International Journal of Stem Cells

        • Excessive branched-chain amino acid accumulation restricts mesenchymal stem cell-based therapy efficacy in myocardial infarction

          Fuyang Zhang, Signal Transduction and Targeted Therapy, 2022

          Intracellular delivery of nitric oxide enhances the therapeutic efficacy of mesenchymal stem cells for myocardial infarction

          Tian Hao https://orcid.org/0009-0007-8657-0424, Guangbo Ji https://orcid.org/0009-0000-7079-8896, Meng Qian https://orcid.org/0009-0006-9135-4593, Qiu Xuan Li, Haoyan Huang, Shiyu Deng, Pei Liu, Weiliang Deng https://orcid.org/0009-0001-8402-7280, Yongzhen Wei, [...], and Qiang Zhao https://orcid.org/0000-0003-4656-6002 +9 authorsAuthors Info & Affiliations

          Science Advances

          29 Nov 2023

          Vol 9, Issue 48

          DOI: 10.1126/sciadv.adi9967

          4,3612

          Metrics

          Total Downloads4,361

          • Last 6 Months1,270
          • Last 12 Months3,995

          Total Citations2

          • Last 6 Months1
          • Last 12 Months2

          NotificationsBookmark

          • • Abstract
            • INTRODUCTION
            • RESULTS
            • DISCUSSION
            • MATERIALS AND METHODS
            • Acknowledgments
            • Supplementary Materials
            • REFERENCES AND NOTES
            • eLetters (0)
          • • Information & Authors
            • Metrics & Citations
            • View Options
            • References
            • Media
            • Tables
            • Share

          Abstract

          Cell therapy by autologous mesenchymal stem cells (MSCs) is a clinically acceptable strategy for treating various diseases. Unfortunately, the therapeutic efficacy is largely affected by the low quality of MSCs collected from patients. Here, we showed that the gene expression‎ of MSCs from patients with diabetes was differentially regulated compared to that of MSCs from healthy controls. Then, MSCs were genetically engineered to catalyze an NO prodrug to release NO intracellularly. Compared to extracellular NO conversion, intracellular NO delivery effectively prolonged survival and enhanced the paracrine function of MSCs, as demonstrated by in vitro and in vivo assays. The enhanced therapeutic efficacy of engineered MSCs combined with intracellular NO delivery was further confirmed in mouse and rat models of myocardial infarction, and a clinically relevant cell administration paradigm through secondary thoracotomy has been attempted.

          SIGN UP FOR THE AWARD-WINNING SCIENCEADVISER NEWSLETTER

          The latest news, commentary, and research, free to your inbox daily

          Sign up

          INTRODUCTION

          Mesenchymal stem cells (MSCs) are pluripotent stem cells with high self-renewal abilities and multidirectional differentiation potential (1, 2). They are extensively distributed throughout the body and serve a variety of purposes, including tissue regeneration (3), immunoregulation (4), and angiogenesis (5). However, an important challenge for stem cell therapy is the low survival rate of stem cells after transplantation, which is associated with nonspecific homing of cells and ischemia/hypoxia at the injury site (6). In addition, transplanted stem cells cannot fully exert their paracrine effects in pathological environments, which seriously limits the clinical application of stem cells.

          The proangiogenic function of MSCs is a key factor contributing to the treatment of ischemic diseases. Generally, MSCs can stimulate local angiogenesis in ischemic tissue by secreting cytokines, including vascular endothelial growth factor (VEGF), transforming growth factor β (TGFβ), and basic fibroblast growth factor (bFGF), to induce endothelial cells to form tubular structures (7). In addition, MSCs can recruit pericytes and smooth muscle cells to promote neovascularization (8, 9). Recently, MSCs have been reported to participate in the construction of blood vessels by directly differentiating into endothelial or smooth muscle cells (10).

          As an important signaling molecule, nitric oxide (NO) plays a crucial role in the maintenance of vascular homeostasis by inhibiting thrombus formation and promoting angiogenesis (11, 12). Recently, increasing attention has been given to the regulation of stem cells by NO due to its multiple biological functions (13, 14). NO can regulate the proliferation of stem cells by regulating the activities of cyclin and mitotic receptors, as well as their downstream pathways (15, 16). On the other hand, NO can regulate the expression‎ of angiogenic cytokines and immunomodulatory factors to improve the paracrine performance of stem cells (17). Additionally, recent studies have demonstrated that NO can regulate the differentiation behavior of stem cells through the phosphatidylinositol 3-kinase (PI3K)/AKT, guanosine 3′,5′-monophosphate (cGMP), and other signaling pathways (18).

          As a result, NO-releasing biomaterials have been used as delivery carriers for stem cells to enhance their survival and regulate paracrine functions (19). However, NO, which is a gaseous molecule, easily diffuses and has a high level of instability. Furthermore, the physiologic function of NO is dose dependent, and an overdose of NO often leads to notable cytotoxicity (20). Thus, optimizing the beneficial effects of NO to strengthen the therapeutic efficacy of stem cells by tuning their release profile should be taken into account.

          In our previous study, based on the chemical biology principle of “bump-and-hole,” we designed and prepared an enzyme-prodrug delivery system and achieved targeted delivery of NO at the lesion site in two different ischemic disease models (21). Here, MSCs were further modified by gene transfection to express a catalytic enzyme (A4-β-GalH363A). The engineered MSCs (eMSCs) were transplanted using an injectable hyaluronic acid (HA) hydrogel as the carrier, while the NO prodrug was injected through the tail vein to achieve controlled release of NO catalyzed by the enzymes in eMSCs. The therapeutic efficacy of MSCs combined with exogenous NO delivery was eval‎uated in mouse and rat models of myocardial infarction (MI) with an emphasis on comparing the therapeutic efficacy of two different NO administration methods (intracellular or extracellular), and the underlying mechanism of their myocardial protective effect was further explored.

          RESULTSThe gene expression‎ of MSCs collected from patients is differentially regulated

          In the clinic, the therapeutic efficacy of autologous stem cell transplantation is largely affected by the low quality (including cell survival and paracrine function) of stem cells collected from patients due to chronic diseases. Diabetes mellitus is a chronic medical condition that can lead to a variety of complications, and these complications can affect various parts of the body, such as the kidney, lower limb, and heart. Diabetes predisposes affected individuals to a spectrum of cardiovascular complications, and one of the most debilitating in terms of prognosis is heart failure (22).

          Accordingly, although autologous MSCs have been widely accepted as a promising strategy for treating various complications associated with diabetes, the therapeutic efficacy is largely affected by the quality of stem cells collected from the patients themselves. The gene levels in adipose-derived MSCs (ADMSCs) from patients with diabetes and healthy individuals were first compared by transcriptome sequencing. The heatmap shows that multiple genes in MSCs from patients with diabetes were up- or down-regulated compared to healthy controls (Fig. 1A). Gene ontology (GO) enrichment analysis revealed that the most differentially up-regulated genes were related to tumor necrosis factor (TNF) signaling pathways (Fig. 1B). Subsequently, we performed gene set enrichment analysis (GSEA) based on the RNA-sequencing results. GSEA revealed that inflammatory target genes were highly enriched in MSCs from patients with diabetes, with a normalized enrichment score (NES) of 1.65 (P < 0.01) (Fig. 1C). Apoptosis-related genes were also highly enriched, with an NES of 1.77 (P < 0.01) (Fig. 1D). In addition, through enrichment analysis, we found that proangiogenic genes in MSCs were greatly down-regulated in patients with diabetes (Fig. 1E).

          

          Fig. 1. Transcriptome sequencing of adipose mesenchymal stem cells (MSCs) collected from patients with diabetes and healthy controls.

          (A) Heatmap showing the differentially expressed genes of MSCs from patients with diabetes and healthy controls (n = 3). (B) Gene ontology (GO) analysis of the up-regulated transcriptome of MSCs from patients with diabetes. (C and D) Gene set enrichment analysis (GSEA) was performed to determine the enrichment of inflammation (C) and apoptosis (D) target genes in the diabetic group. (E) Heatmap showing angiogenesis-related genes in the two groups (n = 3). (F) Schematic illustration demonstrating the difference in MSCs between patients with diabetes and healthy controls at the gene level.

          Open in viewer

          eMSCs are constructed to express mutant β-galactosidase

          Here, we first designed and constructed plasmids carrying the mutant β-galactosidase (A4-β-GalH363A) target gene and luciferase–red fluorescent protein (RFP) dual reporter genes, which could be further used for in vivo imaging. eMSCs expressing A4-β-GalH363A were constructed by infecting MSCs with lentiviruses obtained from human embryonic kidney 293T cells (Fig. 2A). Immunofluorescence staining for RFP confirmed that A4-β-GalH363A was successfully expressed by eMSCs (Fig. 2B). The subcellular fraction and intracellular distribution of enzymes expressed by the eMSCs was determined by Western blotting (Fig. 2C). In contrast to natural β-galactosidases, which are widely distributed within cells, A4-β-GalH363A was mainly confined to the nucleus of eMSCs.

          

          Fig. 2. Intracellular expression‎ and localization of A4-β-GalH363A.

          (A) Schematic diagram of lentivirus packaging and mesenchymal stem cell (MSC) infection. (B) Immunofluorescence staining of β-Gal and A4-β-GalH363A in eMSCs. Scale bar, 50 μm. (C) The distribution of two different enzymes in engineered MSCs (eMSCs) was analyzed by Western blotting.

          Open in viewer

          Intracellular release of NO is realized via decomposition of the 6-OMeGal-Ph-NO prodrug

          Since the mutant β-galactosidase was confined to the nucleus in eMSCs, we designed a prodrug by introducing a lipid-soluble self-decomposition chain into 6-OMeGal-NO to improve its oil and water distribution coefficient; therefore, the resultant NO donor 6-OMeGal-Ph-NO could penetrate the cell membrane and decompose and release NO under the catalysis of A4-β-GalH363A (Fig. 3A and Supplementary Materials). An in vitro release assay showed that the 6-OMeGal-Ph-NO prodrug was efficiently recognized and converted by A4-β-GalH363A with a cumulative release ratio of approximately 97.3%, while nearly no release was observed in the presence of wild-type β-galactosidase (Fig. 3B). To confirm‎ intracellular NO release, eMSCs were preincubated with an NO fluorescent probe (DAF-AM DA) and then treated with different NO prodrugs. The changes in fluorescence signals were examined by time-delay cell imaging (Fig. 3C). The results indicated that the fluorescence intensity continuously increased in the group that was treated with 6-OMeGal-Ph-NO, indicating conversion into NO (Fig. 3D). In contrast, no detectable changes were identified in the β-Gal-NO group because the high water solubility restricted its entry into eMSCs. Next, intracellular and extracellular release of NO in eMSCs was assessed (Fig. 3E). The quantity of intracellular NO was measured by electron paramagnetic resonance (EPR) using ferrous N-diethyl dithiocarbamate (DETC2-Fe) as the spin-trapping reagent. The resultant NO adduct (DETC2-Fe-NO) exhibited a characteristic triplet EPR signal (aN = 13.06 G, giso = 2.041) at room temperature. Quantitative analysis showed that the NO level was significantly (P < 0.001 or 0.0001) higher in eMSCs treated with 6-OMeGal-Ph-NO than in the β-Gal-NO and control groups (Fig. 3F). Furthermore, in the group treated with β-Gal-NO, the release of NO was mainly catalyzed by β-galactosidase that translocated from the cytoplasm in eMSCs, and the extracellular release profile of β-Gal-NO was confirmed by detecting the NO level in the cell culture medium with the NO-sensitive near-infrared fluorescence probe (23); it was significantly (P < 0.01) higher than that in the 6-OMeGal-Ph-NO and control groups (Fig. 3G). To determine the uptake of the NO prodrug by eMSCs, we incubated 6-OMeGal-Ph-NO with eMSCs, and the concentration in the culture medium was determined at different time points. The results reflected that approximately 45% of 6-OMeGal-Ph-NO was incorporated into the eMSCs within 12 hours (fig. S1).

          

          Fig. 3. Intracellular generation of nitric oxide (NO) from the NO prodrug under the catalysis of A4-β-GalH363A expressed by engineered mesenchymal stem cells (eMSCs).

          (A) Synthesis of two NO prodrugs with different enzyme response abilities and cellular permeabilities. (B) In vitro release profile of NO from the NO prodrug (6-OMeGal-Ph-NO) in the presence of β-Gal or A4-β-GalH363A. (C) Schematic illustration of intracellular NO imaging by using an NO fluorescence probe (DAF-AM DA). (D) Representative time-lapse images of NO generation from two different prodrugs in eMSCs and quantification of the fluorescence intensity (n = 6). ***P < 0.001, ****P < 0.0001 versus 6-OMeGal-Ph-NO group. (E) Schematic illustration showing the detection of intracellular and extracellular NO generation differentially. (F) Representative electron paramagnetic resonance (EPR) spectra and quantification of intracellular NO generation by measuring the DETC2-Fe-NO complex using 2,2,5,5-tetramethyl piperidine 1-oxyl (TEMPO) as a standard (n = 3). (G) Relative quantification of NO production in the medium determined using the near-infrared fluorescence probe (n = 4). Data are expressed as the mean ± SEM. Significant differences were detected by one-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

          Expand for more

          Open in viewer

          Intracellular delivery of NO inhibits apoptosis and enhances the proangiogenic activity of eMSCs

          Overproduction of reactive oxygen species due to cellular oxidative stress has been accepted as an important factor contributing to apoptosis in transplanted cells in ischemic tissue. Therefore, we assessed the protective effect of exogenously administered NO on the survival of eMSCs with H2O2-induced oxidative stress and focused on comparing the protection provided by intracellular and extracellular NO administration. The results showed that H2O2 stimulated apoptosis, and delivery of NO via extracellular and intracellular strategies significantly (P < 0.01 or 0.001) reduced apoptosis in eMSCs stimulated by oxidative stress, and the highest fluorescence signal was observed in response to intracellular NO delivery (Fig. 4A).

          

          Fig. 4. Intracellular delivery of nitric oxide (NO) inhibits apoptosis of engineered mesenchymal stem cells (eMSCs).

          (A) Bioluminescence imaging (BLI) was used to detect the effect of NO delivery on cell apoptosis stimulated by different concentrations of H2O2, and the fluorescence signals were further quantified (n = 3). **P < 0.01, ***P < 0.001, ****P < 0.0001 versus control group; #P < 0.05 versus β-Gal-NO group. (B) Flow cytometry assay of cell viability and apoptosis of eMSCs after H2O2 stimulation, and quantification of mean percent values of apoptotic cells (n = 3). (C) The expression‎ of apoptosis-related protein (BCL2, Bax, Bad, caspase3, and cleaved caspase3) by eMSCs was detected after H2O2 stimulation by Western blots (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001. (D) Representative BLI photographs reflecting the retention of eMSCs with and without intracellular NO delivery after in vivo transplantation as well as the quantitative analysis of signals (n = 3). Data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01.

          Open in viewer

          The survival of eMSCs after treatment with various NO delivery strategies was compared by using flow cytometry. Intracellular NO delivery had the most remarkable antiapoptotic effect on cells stimulated by oxidative stress (P < 0.001) (Fig. 4B). The expression‎ of apoptosis-related genes by eMSCs followed a similar trend; NO delivery (extracellular and intracellular) effectively promoted the expression‎ of the antiapoptotic gene BCL2 in eMSCs after oxidative stress stimulation, while the expression‎ levels of proapoptotic genes were reduced accordingly. Intracellular NO delivery via administration of the 6-OMeGal-Ph-NO prodrug exerted a more pronounced antiapoptotic effect at both the gene and protein levels (Fig. 4C and fig. S2A), which may be because the intracellular generation of NO directly activated the antioxidant system in cells to resist oxidative stress damage and inhibit further apoptosis.

          The expression‎ of proangiogenic genes, including ANGPT1, ANGPT2, FGF2, VEGFA, and KDR, in eMSCs was further detected by reverse transcription polymerase chain reaction (RT-PCR). The results showed that the expression‎ level of proangiogenic genes was significantly (P < 0.05, 0.01, or 0.001) higher in eMSCs treated with the 6-OMeGal-Ph-NO prodrug than in the other groups, indicating the enhanced proangiogenic functions of eMSCs after intracellular NO delivery (fig. S2B).

          We further eval‎uated the effect of NO delivery on the in vivo retention of eMSCs after orthotopic transplantation in the myocardial tissue of mice. As shown in Fig. 4D, intracellular delivery of NO effectively prolonged the retention of eMSCs within the myocardium, and an evident bioluminescence imaging (BLI) signal corresponding to the retention of eMSCs was observed 7 days after transplantation compared to the counterpart without administration of the NO prodrug. To further eval‎uate the translational potential of eMSCs in clinical settings, we used MSCs derived from diabetic patients and conducted a series of assays related to cell survival and paracrine function. The findings indicated that intracellular delivery of NO also confers advantages in the attenuation of cell apoptosis under stress conditions, thereby prolonging the in vivo retention of eMSCs (fig. S3).

          Intracellular delivery of NO ameliorates myocardial injury in MI mice after treatment with eMSCs

          The therapeutic efficacy of MSCs combined with exogenous NO was further eval‎uated in a mouse MI model (Fig. 5A and fig. S4). The inflammatory response in the early stage (3 days) was first detected by hematoxylin-eosin (H&E) staining and CD68 immunofluorescence staining. The results demonstrated that severe inflammatory cell infiltration occurred in the injured myocardium of MI mice, and it was effectively alleviated after eMSC treatment. More prominent restoration in the injured myocardium was observed after further administration of NO (Fig. 5, B and C), confirm‎ing the inhibitory effect on inflammation after MI provided by the combination of MSCs and NO. Moreover, this inhibitory effect was more significant (P < 0.01 or 0.0001) in the group with intracellular NO delivery than in the group with extracellular NO delivery.

          

          Fig. 5. Intracellular delivery of nitric oxide (NO) ameliorates myocardial injury in myocardial infarction (MI) mice after treatment with engineered mesenchymal stem cells (eMSCs) via regulation of the inflammation and stimulation of angiogenesis.

          (A) Experimental schedule for the treatment of MI in a mouse model. (B) Hematoxylin-eosin (H&E) staining was performed to detect inflammatory cell infiltration in the early stage of MI (n = 6). Scale bar, 100 μm. (C) Representative images of CD68 immunofluorescence staining (green) and quantification of CD68+ macrophages in injured myocardium (n = 6). Scale bar, 25 μm. (D and E) Flow cytometry was performed to detect peritoneal macrophage polarization 7 days after surgery followed by different treatments. TNFα- and CD206-positive ascites macrophages (markers of M1 and M2 macrophage phenotypes, respectively) were quantified accordingly (n = 3). (F) Representative images of α–Smooth muscle actin (α-SMA) immunofluorescence staining and the quantification of α-SMA+ arterioles (n = 6). Scale bar, 100 μm. (G) Representative images of von Willebrand factor (vWF) immunofluorescence staining and the quantification of vWF+ capillaries (n = 6). Scale bar, 100 μm. Data are expressed as the mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

          Expand for more

          Open in viewer

          It has been widely accepted that a higher proportion of M2-type macrophages is more conducive to the repair of tissue damage. Thus, peritoneal macrophages were extracted at 7 days after surgery and examined by flow cytometry to assess the polarization of macrophages in MI mice that received different treatments. MI modeling leads to a marked increase in the polarization of macrophages toward the M1 phenotype (TNF-α positive); however, the ratio of M1-type macrophages was moderately reduced after treatment with eMSCs. Further administration of NO via the intracellular delivery method significantly (P < 0.001 or 0.01) inhibited the polarization of macrophages toward the M1 phenotype while increasing the proportion of M2-type macrophages compared to the control and eMSC groups (Fig. 5, D and E). Additionally, we conducted immunofluorescence staining in heart section (fig. S5). The results revealed that intracellular delivery of NO also induces the polarization of macrophages into the M2 phenotype within the heart.

          In vitro studies demonstrated that exogenous NO could improve the proangiogenic capacity of eMSCs. Here, we further explored the influence of the combined delivery of exogenous NO and eMSCs on the reconstruction of the vascular network at the site of infarction. α–Smooth muscle actin (α-SMA)–positive arterioles and von Willebrand factor (vWF)–positive small vessels in MI mice after the different treatments were detected by immunofluorescence staining (Fig. 5, F and G). Treatment with eMSCs efficiently promoted angiogenesis in the injured myocardium, and more prominent enhancement was observed in response to further treatment with intracellular NO. This finding was further supported by the expression‎ of angiogenesis-related genes in the border zone of the infarcted heart (fig. S6).

          Intracellular delivery of NO improves heart function and inhibits adverse myocardial remodeling in MI mice after treatment with eMSCs

          Ultrasound and histological analyses were performed to eval‎uate the long-term recovery of cardiac function after MI. Cardiac injury was first eval‎uated by 2,3,5-triphenyltetrazolium chloride (TTC) staining (Fig. 6A). Treatment with eMSCs moderately repressed MI compared to the acute myocardial infarction (AMI) group, but a more pronounced inhibitory effect was observed in the group with further intracellular NO delivery. Left ventricular function was assessed by echocardiography at different time points. As shown in Fig. 6B, after 1 day of MI, the left ventricle in each group was markedly enlarged, cardiac function decreased rapidly, and deterioration of heart function continued for 28 days without detectable restoration in the AMI group. However, eMSC treatment could restore left ventricular systolic function and reduce ventricular dilation, as shown by the increase in left ventricular ejection fraction (LV-EF) and fraction shortening (LV-FS), as well as the decrease in left ventricular end-diastolic diameter (LVIDd) and left ventricular end-diastolic volume (LV-EDV) to a certain extent. In the group treated with eMSCs and intracellular NO delivery (NO-eMSCs), LV-EF and LV-FS were effectively recovered, while LVIDd and LV-EDV were significantly (P < 0.001 or 0.0001) enhanced compared to the AMI group.

          

          Fig. 6. Intracellular delivery of nitric oxide (NO) improves heart function and reduces adverse cardiac remodeling in myocardial infarction (MI) mice after treatment with engineered mesenchymal stem cells (eMSCs).

          (A) Representative images of 2,3,5-triphenyltetrazolium chloride (TTC) staining and quantification of the infarct area (n = 3). Scale bar, 2 mm. (B) Cardiac echo measurement was performed at different time points after surgery, and cardiac function indicators of left ventricular ejection fraction (LV-EF), left ventricular fractional shortening (LV-FS), left ventricular internal diameter at end diastole (LVIDd), and left ventricular end-diastolic volume (LV-EDV) were eval‎uated accordingly (n = 6). ***P < 0.001, ****P < 0.0001 versus acute myocardial infarction (AMI) group. (C) Masson’s trichrome staining was performed, and the infarct size was quantified accordingly (n = 6). (D) Collagen deposition in the hearts was detected by Sirius Red staining (n = 6). Scale bar, 100 μm. (E) Representative images of wheat germ agglutinin (WGA) immunofluorescence staining and quantification of the cross-sectional area of cardiomyocytes (n = 6). Scale bar, 50 μm. (F) Representative images of immunofluorescence staining (red) for the gap junction protein (Cx43) and the quantification of the intensity of red fluorescence to the whole area of images (n = 6). Scale bar, 25 μm. Data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

          Expand for more

          Open in viewer

          Masson staining was used to detect the degree of myocardial fibrosis after MI. Severe myocardial fibrosis was observed in MI mice compared to the sham group (Fig. 6C). In addition, the ventricular wall became thinner, which was closely related to impaired left ventricular systolic function after MI, as demonstrated by echocardiography. However, these adverse cardiac remodeling events were ameliorated after eMSC treatment and accompanied by a lowered degree of myocardial fibrosis and a thickened ventricular wall in contrast to the AMI group. Notably, treatment with eMSCs and intracellular NO delivery exerted the most prominent inhibitory effect on myocardial fibrosis after MI (Fig. 6C).

          Collagen deposition in MI mice was assessed by Sirius staining (Fig. 6D), and the results showed that MI resulted in severe collagen deposition in the injured myocardium compared to the sham group. It was effectively reduced after eMSC treatment, and the inhibitory effect of eMSC plus intracellular NO delivery was significantly higher than that in the other two groups (P < 0.001 or 0.0001).

          Next, wheat germ agglutinin (WGA) staining was carried out to eval‎uate myocardial cell hypertrophy 28 days after MI (Fig. 6E). The cross-sectional area of cardiomyocytes was increased in MI mice in contrast to the sham operation group due to compensatory hypertrophy in the heart to maintain the normal rate of cardiac ejection. Hypertrophy was significantly (P < 0.05) mitigated after treatment with eMSCs, especially in the presence of exogenous NO (P < 0.001), indicating an ideal therapeutic effect on inhibiting myocardial cell hypertrophy and adverse ventricular remodeling by the combination of eMSCs and NO.

          Gap junctions (GJs) are the main connections between cardiomyocytes in the heart, and Cx43 is the main GJ protein in ventricular muscle in the heart (24). Studies have shown that the absence of Cx43 leads to the occurrence of cardiac ventricular arrhythmia, which can develop into heart failure (25). After 28 days of MI, immunofluorescence staining for Cx43 revealed abundant and uniform distribution of GJ proteins in the sham group, whereas MI injury resulted in a marked decrease in the expression‎ of Cx43 (Fig. 6F). Despite the moderate inhibitory effect provided by the administration of eMSCs, further delivery of NO via the intracellular method significantly enhanced (P < 0.001 or 0.01) the expression‎ of Cx43 compared to that in the AMI or eMSC groups.

          Intracellular delivery of NO enhances the therapeutic efficacy of eMSCs in a rat MI model

          Although the outcome in a mouse model supported the beneficial effect of NO via intracellular delivery on enhancing the therapeutic efficacy of MSCs for MI, immediate administration of stem cells after MI is different from the clinical treatment of MI due to the limitation of the administration paradigm in mouse models. In addition, 3 to 7 days after MI is the outbreak period of the inflammatory response. For this reason, we established a rat model of MI and conducted secondary thoracotomy 3 days after surgery (Fig. 7A), and eMSCs were delivered via an injectable HA hydrogel as the carrier (Fig. 7B). Lactate dehydrogenase (LDH), a crucial marker for assessing the extent of myocardial damage, exhibited an initial elevation within 2 to 48 hours following the onset of MI, reaching its zenith between 2 and 5 days after MI. We collected blood samples from the orbital venous plexus of rats 5 days after MI to measure serum LDH levels (fig. S7). The findings revealed a sharp increase in serum LDH levels due to MI. However, treatment with NO-eMSCs significantly reduced serum LDH levels, indicating an attenuation of cardiac injury.

          

          Fig. 7. Intracellular delivery of nitric oxide (NO) enhances the therapeutic efficacy of engineered mesenchymal stem cells (eMSCs) in a rat myocardial infarction (MI) model.

          (A) Experimental schedule for the treatment of rat MI. (B) Representative images showing the second thoracotomy in rats after MI. (C) Cardiac echo measurement was performed at different time points after surgery, and cardiac function indicators of left ventricular ejection fraction (LV-EF), left ventricular fractional shortening (LV-FS), left ventricular internal diameter at end diastole (LVIDd), and left ventricular end-diastolic volume (LV-EDV) were eval‎uated accordingly (n = 6). **P < 0.01, ***P < 0.001 versus acute myocardial infarction (AMI) group. (D) Representative images of Masson’s trichrome staining and quantification of the infarct size and infarct thickness (n = 6). **P < 0.01, ***P < 0.001, ****P < 0.0001 versus AMI group. (E) Representative images of wheat germ agglutinin (WGA) immunofluorescence staining and quantification of the cross-sectional area of cardiomyocytes (n = 6). Scale bar, 50 μm. (F) Representative images of α–Smooth muscle actin (α-SMA) immunofluorescence staining and quantification of α-SMA+ arterioles (n = 6). Scale bar, 100 μm. Data are expressed as the mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

          Open in viewer

          Echocardiography was performed to detect heart function at different time points after MI (Fig. 7C). At 4 weeks after surgery, the anterior wall of the left ventricle was completely infarcted, and the ventricular cavity was increased dramatically. Quantitative analysis further indicated that both EF and FS were decreased markedly after surgery. Treatment with MSCs could restore the systolic function of the heart, and the highest recovery rate (60 to 80%) was observed in the group treated with eMSCs and intracellular NO delivery. Ventricular dilatation was also effectively restored after combined treatment with eMSCs and NO, which was consistent with the results obtained in the mouse model.

          Histological analyses showed that treatment with eMSCs combined with NO efficiently reduced the degree of fibrosis (Fig. 7D) and collagen deposition (fig. S8A). Furthermore, it alleviated infarct size after MI and restored ventricular cavity morphology of the heart with a significant enhancement in the thickness of the infarcted left ventricular wall and the interventricular septum (IVS), as eval‎uated by Masson (Fig. 7D) and H&E staining (fig. S9). Twenty-eight days after MI, WGA staining also showed that the combination of eMSCs and NO significantly (P < 0.01) inhibited cardiomyocyte hypertrophy in contrast to the AMI group, thus inhibiting further myocardial systolic dysfunction (Fig. 7E). The expression‎ of the GJ protein Cx43 (fig. S8B) followed a similar trend to that in mouse models; that is, acute MI led to a marked decrease in the distribution of Cx43 in the myocardium, and treatment with eMSCs up-regulated the expression‎ of Cx43 to a certain extent. Further delivery of NO via the intracellular method produced a more significant (P < 0.05 or 0.01) effect on the up-regulation of Cx43 in the injured myocardium compared to the eMSC and AMI groups, thereby facilitating the connection between cardiomyocytes and further inhibiting the occurrence of arrhythmias and heart failure.

          Tissue repair after MI is often closely related to angiogenesis, which begins at the infarct border and extends to the center of the infarction (26). Therefore, we further compared the proangiogenic effect of eMSCs with and without NO delivery on the damaged margin of the infarcted myocardium 28 days after MI in rats (Fig. 7F and fig. S8C). The combination of eMSCs and NO remarkably promoted angiogenesis and reconstruction of the vascular network compared to the group treated with eMSCs only, which is beneficial to the repair of myocardial injury after MI.

          DISCUSSION

          Cell therapy based on MSCs has proven to be a powerful solution for treating degenerative diseases and tissue damage (27–29). Despite the advantages of autologous stem cells over allogeneic stem cells, including the absence of immune rejection, the decreased survival and impaired paracrine functions of stem cells collected from patients with chronic diseases greatly limit their clinical use (7, 30, 31). Here, we first showed that MSCs collected from patients with diabetes exhibited marked up-regulation of apoptosis- and inflammation-related genes and down-regulation of proangiogenic genes, affecting the efficacy of cell therapy. As a result, genetic engineering strategies are often required to enhance the therapeutic efficacy of autologous stem cells. A recent study revealed that eMSCs, engineered to express PD-L1 on their surface and secrete CTLA4-Ig (immunoglobulin) as an extracellular factor, exhibited immunoprotective properties, which improved the outcome of both syngeneic and allogeneic islet transplantation in diabetic mice (32).

          NO is involved in a variety of physiological processes. Studies have shown that as an important signaling molecule, NO plays a pivotal role in regulating stem cell behavior (33–35), including cell survival, migration, differentiation, and paracrine behavior. These factors affect the interaction of stem cells with other cells and the tissue microenvironment. Previously, different types of NO-releasing biomaterials, such as injectable hydrogels, have been prepared by us and other groups (36–40), and further studies have shown that the combination of NO and MSCs is more effective in treating various diseases than MSC therapy alone. In addition, it has been reported that pretreatment of MSCs with NO-releasing biomaterials could enhance the therapeutic efficacy of MSCs and their secreted exosomes because of their enhanced proangiogenic functions (41).

          Due to the spatiotemporal characteristics of NO (42), precise delivery of NO in a site-specific and controllable manner holds great importance in the regulatory effect of exogenously administered NO. In addition to the controlled release rate, the site at which NO is generated is also a key factor due to the relative half-life and limited diffusion distance (43, 44). It is reasonable to speculate that intracellular and extracellular NO delivery may lead to different outcomes when regulating the survival and function of MSCs. In our previous work, an enzyme-prodrug delivery system was designed based on a bump-and-hole strategy (21). The mutant galactosidase (A4-β-GalH363A) enables the targeted delivery of NO, thus reducing the side effects due to the unspecific decomposition of the NO prodrug and enhancing the therapeutic efficacy. Here, we transfected a plasmid expressing mutant galactosidase into MSCs and successfully constructed eMSCs. The enzyme expressed by MSCs could catalyze the decomposition of the 6-OMe-galactose–protected NO prodrug and release NO intracellularly.

          Western blotting and fluorescence imaging demonstrated that the expression‎ of the engineered enzyme was confined to the nucleus of MSCs, while wild-type β-galactosidase was widely distributed in the cytoplasm, including the lysosome and perinuclear region (45, 46). Since the corresponding prodrug for wild-type β-galactosidase is highly hydrophilic, it fails to enter MSCs and releases NO extracellularly by enzymes that translocate from the cell. In contrast, the prodrug for mutant galactosidase is cell penetrating because of the modified molecular structure; therefore, it can enter MSCs and release NO intracellularly under the catalysis of the corresponding enzyme expressed by the cells. Accordingly, two different NO delivery paradigms were successfully developed in this study and further confirmed by a series of eval‎uations, including cell imaging and electronic paramagnetic resonance. Further in vitro and in vivo assays indicated that in contrast to extracellular NO delivery, intracellular administration of NO enhanced cell survival and the paracrine effects of MSCs, including inhibiting apoptosis and supporting angiogenesis.

          Next, we established a mouse MI model to systematically eval‎uate the therapeutic efficacy of MSCs combined with exogenous NO. The results showed that intracellular delivery of NO prolonged the retention of eMSCs after myocardial orthotopic transplantation. In addition, the combination of eMSCs and intracellular NO delivery improved cardiac function after MI and reduced adverse ventricular remodeling compared to the group treated with MSCs only. Additionally, it could effectively restore the reconstruction of the blood vessel network and further promote the repair of the infarcted myocardium.

          To gain further insight into the translational potential of the combinatory therapeutic strategy developed in this study, a rat model of MI was established, and MSCs were administered by a second thoracotomy after 3 days to mimic the clinical use of MSCs for the treatment of MI (47, 48). Clinically, acute MI is typically due to the rupture of coronary atherosclerotic plaque and the formation of thrombus, which causes coronary artery obstruction. After the acute phase of MI, adverse ventricular remodeling further affects the prognosis of patients, which is specifically characterized as a decrease in ventricular wall thickness and myocardial tension in the MI area, myocardial hypertrophy in the noninfarction area, and a change in the morphology of the ventricular cavity, thus leading to arrhythmia and further development into heart failure. The efficacy of MSCs in managing arrhythmias remains a topic of ongoing debate. Some researchers argue that MSCs do not appear to reduce or prevent arrhythmias, with the antiarrhythmic or proarrhythmic potential of MSCs primarily relying on paracrine factors (49). Conversely, other studies suggest that MSCs themselves may play a role in the post-MI recovery process (50). In our study, we observed an evident up-regulation of Cx43 expression‎ after NO-eMSC treatment, which is a potential target associated with antiarrhythmic effects. Further investigation is still required to comprehensively explore the antiarrhythmic potential of NO-eMSCs. In line with the enhanced therapeutic efficacy in the mouse model, intracellular delivery of NO showed enormous advantages in the rat MI model by inhibiting apoptosis and enhancing the paracrine function of MSCs.

          In summary, we first showed that survival and paracrine function were reduced in MSCs collected from patients with diabetes, which could greatly affect therapeutic efficacy. Accordingly, eMSCs were successfully constructed, and the mutant β-galactosidase expressed by the cells enabled the intracellular generation of NO via the conversion of an exogenous NO prodrug. In vitro and in vivo assays indicated that intracellular delivery of NO effectively enhanced the survival of transplanted MSCs and promoted the paracrine function of MSCs, which was further confirmed by the enhanced therapeutic efficacy in mouse and rat models of MI compared to the group treated with MSCs only. This synergistic strategy provides an option for the treatment of MI by autologous MSCs in the clinic.

          MATERIALS AND METHODSRNA sequencing analysis

          RNA sequencing was performed by the BGI (Shenzhen, China). Briefly, RNA from the ADMSCs of healthy people and patients with diabetes was extracted using TRIzol reagent (Yeasen, China). RNA samples were sequenced on the BGISEQ platform. The raw data containing low-quality reads, adaptor sequences, and high levels of N bases were filtered before analysis. Then, the clean reads were mapped to the reference genome using HISAT, and Bowtie2 was used to align the clean reads to the reference genes. The reference genome source is National Center for Biotechnology Information (NCBI), and the reference genome version is GCF_000001405.39_GRCh38.p13. The expression‎ levels of genes were quantified to identify differentially expressed genes by RNA-Seq by expectation maximization (RSEM). The analyses of hierarchical clustering and heatmap were performed using the online Dr. Tom system (biosys.bgi.com) to compare differential gene expression‎ of ADMSCs in healthy people and patients with diabetes. According to the KEGG_pathway annotation classification, the phyper function in R software was used for enrichment analysis, the P value was calculated, and then false discovery rate (FDR) was performed on the P value to obtain a Q value. Generally, a Q value of ≤0.05 was regarded as significant enrichment. GSEA was used to analyze significant differences in gene expression‎ between inflammatory and apoptosis-related pathways. Expression‎ cluster heatmap was used to analyze the expression‎ of genes associated with angiogenesis.

          Measurement of NO release

          The NO-releasing profile was determined by the Griess kit assay. In brief, 50 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were dissolved in phosphate-buffered saline (PBS) buffer (pH 7.4), and enzymes were added to the solutions at a concentration of 0.005 mg/ml. At each predetermined time interval, 50 ml of solution was transferred into a 96-well plate, and 50 ml of Griess I and 50 ml of Griess II were added thereafter. The azo compound of purple color was formed, and the absorbance was measured at a wavelength of 540 nm using an iMark microplate reader (Bio-Rad, USA).

          Cell cultureMesenchymal stem cells

          MSCs derived from human umbilical cord were obtained from Health-Biotech, maintained in Dulbecco's modified Eagle’s medium (DMEM)/F12 (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, USA) and 1% penicillin- streptomycin solution (Gibco, USA), and placed in a cell culture chamber containing 5% CO2 at 37°C.

          Human embryonic kidney 293T cells

          Human embryonic kidney (HEK) 293T cells were obtained from the American Type Culture Collection (ATCC), maintained in high-glucose DMEM (Gibco, USA) with 10% FBS (Gibco, USA) and 1% penicillin-streptomycin solution (Gibco, USA), and placed in a cell culture chamber containing 5% CO2 at 37°C.

          Construction of eMSCs

          The coding sequence of mutant β-galactosidase (A4-β-GalH363A) can be obtained from the previous publication (21). The lentivirus packaging system containing A4-β-GalH363A sequence and Rluc-RFP sequence was constructed by Wuhan Miaolingbio Co. Ltd. The constructed lentivirus plasmid containing the target gene and the package gene (psPAX2 and pMD2.G) was transfected into HEK 293T cells through Lipo2000, and the supernatant was collected to obtain the virus solution. After removing impurities, the virus solution was mixed 1:1 with fresh MSC medium, and polybrene (10 μg/ml) was added. MSCs were infected with virus through incubation in the mixture medium. The infection efficiency was observed under an inverted fluorescence microscope, and the expression‎ of target protein was determined by Western blotting.

          Cell immunofluorescence staining

          eMSCs were inoculated in 24-well plates. Cells were fixed with 4% paraformaldehyde and blocked in 4% bovine serum albumin in PBS for 30 min at room temperature. Then, the cells were incubated with primary antibodies overnight at 4°C. The bound primary antibodies were displayed by incubation with the secondary antibodies for 2 hours at room temperature. Cells were counterstained with DAPI (4′,6-diamidino-2-phenylindole)–containing Fluoromount-G and visualized with a fluorescence microscope (Axio Imager Z1). Antibodies used include anti–β-galactosidase (1:100, A1863, Abclonal) and anti-RFP (1:100, PA1-986, Invitrogen).

          Western blot

          eMSCs were collected, and total protein was extracted using radioimmunoprecipitation assay (RIPA) lysate containing protease inhibitor (Solarbio, China). Cytoplasmic protein and nucleoprotein were extracted using a nucleoprotein extraction kit containing protease inhibitors (Solarbio, China). The protein concentration was quantified using a BCA protein assay kit (Solarbio, China). The samples were diluted with 4× SDS–polyacrylamide gel electrophoresis (PAGE) loading buffer and boiled in boiling water for 8 min. Then, 30 μg of protein was isolated by 10% SDS-PAGE electrophoresis. The isolated proteins were transferred to an Immobilon-P Transfer membrane (Millipore, USA) and incubated with the primary antibody overnight at 4°C and then with the secondary antibody at room temperature for 2 hours. The bands were detected with chemiluminescent horseradish peroxidase substrate (Millipore, USA). Signals were generated by using an enhanced chemiluminescence (ECL) reagent (Millipore, USA) and were captured by using the Tanon-5200 Chemiluminescence Imaging System (Tanon, China). The antibodies used included anti–β-galactosidase (1:1000, A1863, Abclonal), anti–His-tag (1:1000, 12698S, Cell Signaling Technology), anti–glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (1:1000, AC001, Abclonal), anti–histone H3 (1:1000, 4499S, Cell Signaling Technology), anti–β-actin (1:1000, UM4001, Utibody), anti-BCL2 (1:1000, WL01556, Wanleibio), anti-Bax (1:1000, 50599-2-lg, Proteintech), anti-Bad (1:1000, WL02140, Wanleibio), and anti-caspase/cleaved caspase3 (1:1000, WL02117, Wanleibio).

          6-OMeGal-Ph-NO uptake by eMSCs

          eMSCs were inoculated in 12-well plates, and 50 μM substrate (6-OMeGal-Ph-NO) was added per well. At predetermined time points (0, 1, 3, 6, and 12 hours), an appropriate amount of culture medium was collected, and an excess of A4-β-GalH363A was added immediately to fully catalyze the decomposition of the remaining substrate at room temperature. The amount of 6-OMeGal-Ph-NO substrate in the culture medium was determined by Griess kit assay.

          Real-time imaging of intracellular NO

          The fluorescence emission associated with NO in the cytosol was detected by an electron-multiplying charge-coupled device (DU-897D-CS0-BV; Andor, Belfast, UK) connected to an inverted fluorescence microscope (Axio Observer D1; Carl Zeiss, Oberkochen, Germany). Intracellular NO imaging was performed using an NO fluorescence probe, DAF-AM DA (Beyotime, China), according to the manufacturer’s instruction. eMSCs were inoculated in small confocal dishes in advance, and the experiment was conducted when the cell density reached 80%. First, the medium was collected for later use. After two gentle washes with PBS, 5 μM DAF-AM DA solution was incubated at 37°C for 30 min in the dark. Then, cells were gently washed with PBS twice. The previously collected medium was added anew, and cell imaging was performed immediately. At the 488-nm excitation wavelength, pictures were taken every 5 s. After stable shooting for 2 min, 100 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were added. The filming for cell fluorescence was continued for 4 min. The final fluorescence intensity was determined without the background fluorescence value. The proportion of change in fluorescence intensity of each cell in the visual field was calculated.

          Intracellular NO detection

          Intracellular NO radicals (NO•) were detected using EPR as described (51, 52). In brief, sodium DETC (4.5 mg) and FeSO4•7H2O were dissolved in two separate volumes (10 μl) of deoxygenated Krebs/Hepes solution. Equal volumes of these parent solutions were rapidly mixed and aspirated into Eppendorf combi tips. The 0.5 mM Fe•(DETC)2 colloid solution had a yellow-brownish color with a slight opalescence in light. No aggregate formation was observed, at least during the first 30 min. eMSCs were rinsed with modified Krebs/Hepes buffer and incubated with freshly prepared NO•-specific spin trap Fe•(DETC)2 colloid (0.5 mM) for 30 min. Meanwhile, 100 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were added to the buffer. Gently collected cell suspensions were snap-frozen in liquid nitrogen. Ethyl acetate (200 μl) was added, and the cells were ultrasonically broken to extract DETC2-Fe-NO. The ethyl acetate extract was concentrated with nitrogen and transferred to a 50-μl capillary, and then the X-band EPR was measured at room temperature. The following acquisition parameters were used: modulation frequency, 100 kHz; microwave power, 10 mW; modulation amplitude, 2 G; number of scans, 60. The double-integrated area of the EPR spectra was calibrated into concentrations of DETC2-Fe-NO using TEMPO (2,2,5,5-tetramethyl piperidine 1-oxyl) as a standard. EPR spectral simulation was conducted by the WINSIM program.

          Extracellular NO detection

          eMSCs were treated with β-Gal-NO or 6-OMeGal-Ph-NO (30 μM). The production of NO in the medium of each group was detected 6 hours after incubation with NO-sensitive near-infrared fluorescence probe (5 μM). The NO production of medium in different groups was compared by the relative fluorescence intensity under the excitation at 750 nm (emission at 800 nm).

          Cell apoptosis detection

          To test the protective effect of NO delivery on cellular oxidative stress stimulation, 30 μM NO substrate (6-OMeGal-Ph-NO) was added to the medium in advance. Then, H2O2 with different concentrations (100, 200, 400, and 600 μM) was added to stimulate the lentivirus-infected eMSCs. BLI was performed immediately after addition of the luciferase substrate coelenterin to eval‎uate cell apoptosis. Additionally, eMSCs treated with 200 mM H2O2 were stimulated for 24 hours to induce cell apoptosis. An Annexin V/PI assay kit (Solarbio) was used to detect eMSC apoptosis.

          BLI detection of cell retention

          BLI and luciferase substrates were used in mice to eval‎uate the retention of NO-eMSCsGluc/RFP in cardiac orthotopic transplantation. The mice after eMSC injections were anesthetized with 1.5% isoflurane and injected with coelenterin through the caudal vein at 150 mg/kg. After injection, the mice were immediately placed in a BLI system to detect cell retention in the myocardium.

          Animals

          C57BL/6 mice (male, 8 weeks old) and Sprague-Dawley rats (male, 8 weeks old) were purchased from SPF (Beijing) Biotechnology Co. Ltd., Beijing, China. Animals were randomly grouped for treated and untreated controls. All experiments and animal procedures were approved by the Animal Experiments Ethical Committee of Nankai University and carried out in conformity with the Guide for Care and Use of Laboratory Animals.

          MI in mice and rats

          Surgical induction of MI was performed on C57BL/6 mice (male, 8 weeks old) as previously described with some modifications. Briefly, mice were anesthetized with 2% isoflurane, followed by fixation to a heating pad (37°C) at supine position, and then ventilated with a mechanical ventilation system (Hallowell EMC Microvent I, USA) set at 110 breaths per minute with a tidal volume of 0.2 to 0.3 ml. The third intercostal space over the left chest and the heart was exposed. After left thoracotomy, the left anterior descending coronary artery was ligated with a 6-0 silk ligature. The left main descending coronary artery was sutured and tied with a slip knot at a site approximately 3 mm from its origin. Cardiac ischemia was confirmed by the presence of myocardial blanching. Immediately, eMSCs encapsulated with HA hydrogel were injected into the myocardium of mice through three-point injection around the infarct zone. The chest cavity was closed to restore negative pressure and prevent pneumothorax.

          Similar MI surgery was performed on Sprague-Dawley rats (male, 8 weeks old) first. Briefly, rats were anesthetized via intraperitoneal injection of 10% chloral hydrate (350 mg/kg), followed by fixation to a heating pad (37°C) at supine position. Then, they were ventilated with a mechanical ventilation system (Hallowell EMC Microvent I, USA) set at 110 breaths per minute with a tidal volume of 6 ml. The third intercostal space over the left chest and the heart was exposed. After left thoracotomy, the left anterior descending coronary artery was ligated with a 6-0 silk ligature. The left main descending coronary artery was sutured and tied with a slip knot at a site approximately 3 mm from its origin. Cardiac ischemia was confirmed by the presence of myocardial blanching. The chest cavity was closed to restore negative pressure and prevent pneumothorax. Three days after surgery, secondary thoracotomy was performed, and eMSCs were injected into the myocardium around the infarct zone. The chest cavity was closed to restore negative pressure and prevent pneumothorax.

          Mice and rats in the AMI group only received MI without eMSC injection, while sham-operated mice only experienced thoracotomy without MI surgery or eMSC injection.

          At 1, 3, 5, and 7 days after myocardial injection of eMSCs, the prodrug was injected through the tail vein. Mice were injected with 100 μl of prodrug (1 mg/ml) each time, and rats were injected with 200 μl of prodrug (1 mg/ml) each time.

          TTC staining

          Two days after surgery, a thoracotomy was performed. The heart was quickly excised after quick freezing for 15 min and sliced at 1 mm thickness. Afterward, the sections were incubated with 1.5% TTC (Sigma-Aldrich) solution at 37°C in an incubator for 15 to 30 min and then with a 4% formaldehyde solution for 2 hours. The normal myocardial tissue was red, while the ischemic myocardium was white. The size of the infarcted myocardium, which was white or pale, was measured by ImageJ software.

          Cardiac function assessment

          Transthoracic echocardiography was performed with the Vevo 2100 Imaging System (Fuji Film Visual Sonics Inc., Canada) equipped with an MS-250/400 imaging transducer. The baseline cardiac function of mice and rats was measured at 3 days before surgery. Cardiac function was analyzed at days 1 and 28 after MI surgery with different treatments, as reported previously. Mice or rats were slightly anesthetized in a box with isoflurane. Their limbs were fixed in a supine position on the echo mat, and the chest hair was removed by depilating cream. Then, mice or rats were anesthetized by inhalation of isoflurane (0.5 to 1%) mixed with oxygen to maintain the heart rate at approximately 500 to 600, and M-mode echocardiography was performed. The left ventricular internal diameter at end-diastole (LVIDd) and systole (LVIDs) were obtained by measuring the long axis and the short axis. Accordingly, the cardiac parameters LV-EF, LV-FS, LV-EDV, and LV end-systole volume (LV-ESV) were determined. The echocardiography measurement was carried out in a double-blind manner.

          Histological analysis

          At the indicated time points, mice and rats were anesthetized via intraperitoneal injection of chloral hydrate, and a thoracotomy was performed. The hearts were fixed with trans-cardiac perfusion of saline and immersed in 4% paraformaldehyde over 24 hours. The heart tissue samples were dehydrated with gradient alcohol and xylene, embedded in paraffin blocks, and cut into sections in 5 μm thickness.

          The paraffin-embedded sections were stained with Masson trichrome, H&E, and Sirius Red following a standard protocol. Immunofluorescence staining was performed on paraffin-embedded sections of the heart tissue samples. After deparaffinization and heat-mediated antigen retrieval‎ in citrate solution, the samples were washed with PBS three times and incubated with blocking serum, which was used to avoid nonspecific binding, at room temperature for 30 min. The sections were incubated with specific antibodies diluted in goat serum at 4°C overnight. On the second day, the sections were rewarmed at room temperature for 1 hour and washed with PBS three times. Afterward, the sections were incubated with Alexa Fluor–coupled secondary antibodies for 2 hours at room temperature. After washing with PBS, the sections were counterstained with DAPI-containing Fluoromount-G (SouthernBiotech, USA) and coverslipped. The antibodies used included anti–α-SMA (1:100, ab5694, Abcam), anti-vWF (1:100, ab6694, Abcam), anti–α-actinin (1:100, ab9475, Abcam), anti-Connexin43 (1:100, ab11370, Abcam), anti-CD68 (1:100, ab125212, Abcam), WGA (1:500, FL-1021, Novus Biologicals), anti-iNOS (1:100, ab178945, Abcam), and anti-CD206 (1:100, ab64693, Abcam).

          Macrophage isolation and detection

          Three days before euthanasia, mice were intraperitoneally injected with 2 ml of 4% thioglycolate. Three days later, the mice were sacrificed by cervical dislocation and immersed in 75% alcohol and then transferred to an ultraclean workbench. The mouse limb was fixed in the supine position, and the mouse abdominal wall was carefully cut open with the peritoneal. PBS [1% penicillin-streptomycin (PS)] was injected intraperitoneally to collect the cell suspension, which was centrifuged at 2000 rpm for 10 min. After discarding the supernatant, the cells were incubated with anti-F4/80/TNF-α and anti-F4/80/CD206 antibodies. FlowJo software was used to analyze the results of flow cytometry.

          Quantitative real-time PCR

          Total RNA samples from the cells were prepared using TRIeasy Total RNA Extraction Reagent (Yeasen, China) according to the manufacturer’s instructions. Heart tissue samples were collected at the indicated time points after MI surgery.

          The tissue samples were dissected at the border zone of the left ventricle and frozen in liquid nitrogen immediately. Afterward, the total RNA was extracted with TRIzol reagent, as mentioned before. The concentration of the RNA was measured with a NanoDrop spectrophotometer (NanoDrop Technologies, USA). The complementary cDNA was synthesized using a first-strand cDNA synthesis kit (Yeasen, China). Quantitative real-time PCR was performed on a CFX96 real-time PCR system (Bio-Rad, Hercules, USA) with an SYBR Green–based real-time detection system (Yeasen, China). The relative gene expression‎ of mRNA was expressed as 2−(△△CT) and normalized to GAPDH as an internal control. Each reaction was performed in triplicate to obtain an average value, and the changes in relative gene expression‎ normalized to the internal control levels were determined. The highly purified primers used in this experiment were commercially synthesized (Sango, China). The sequences of the primers used in this experiment are summarized in the Supplementary Materials.

          Statistics

          All data are presented as the mean ± SEM from at least three independent experiments. Comparisons between two groups were performed by Student’s t test, and comparisons among more than two groups were performed by one-way or two-way analysis of variance (ANOVA). Statistical analyses were performed with GraphPad Prism software 7.0, and a statistical significance level of less than 0.05 was accepted.

          Acknowledgments

          Funding: This study is supported by the National Key R&D Program of China (2018YFE0200503), the National Natural Science Foundation of China (nos. 81925021, 82330066, 81921004, and U2004126), and the Tianjin Natural Science Foundation (21JCZDJC00240).

          Author contributions: Q.Z. and Z.L. conceived the original concept and initiated this project. Q.Z., Z.L., and F.G. designed the experiment and supervised the entire project. S.W. collected human adipose mesenchymal stem cells. M.Q. synthesized all NO prodrugs and probes. P.L. prepared engineered enzymes. T.H., G.J., and Q.X.L. established mouse and rat MI models. T.H. and G.J. performed histological analysis. G.J. and W.D. carried out in vitro cell experiments. S.D. carried out NO cell imaging under the supervision of L.P. T.H., G.J., and M.Q. analyzed data under the supervision of Q.Z. H.H. helped with lentivirus packaging and cell infection. W.G. and T.L. helped in establishing animal MI models. Y.W., J.H., J.C., and J.T. helped with data collection. T.H. and Q.Z. wrote the paper with input from other authors.

          Competing interests: The authors declare that they have no competing interests.

          Data and materials availability: All data needed to eval‎uate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Transcriptome sequencing dataset is available at https://doi.org/10.5061/dryad.tqjq2bw5b.

          Supplementary MaterialsThis PDF file includes:

          Supplementary Text

          Figs. S1 to S9

          Table S1

          • Download
          • 1.94 MB

          REFERENCES AND NOTES

          1

          M. F. Pittenger, A. M. Mackay, S. C. Beck, R. K. Jaiswal, R. Douglas, J. D. Mosca, M. A. Moorman, D. W. Simonetti, S. Craig, D. R. Marshak, Multilineage potential of adult human mesenchymal stem cells. Science 284, 143–147 (1999).

          GO TO REFERENCE

          Crossref

          PubMed

          Web of Science

          Google Scholar

          2

          A. Uccelli, L. Moretta, V. Pistoia, Mesenchymal stem cells in health and disease. Nat. Rev. Immunol. 8, 726–736 (2008).

          GO TO REFERENCE

          Crossref

          PubMed

          Web of Science

          Google Scholar

          3

          Y. Wang, J. Fang, B. Liu, C. Shao, Y. Shi, Reciprocal regulation of mesenchymal stem cells and immune responses. Cell Stem Cell 29, 1515–1530 (2022).

          GO TO REFERENCE

          Crossref

          PubMed

          Web of Science

          Google Scholar

          4

          Y. Wang, X. Chen, W. Cao, Y. Shi, Plasticity of mesenchymal stem cells in immunomodulation: Pathological and therapeutic implications. Nat. Immunol. 15, 1009–1016 (2014).

          GO TO REFERENCE

          Crossref

          PubMed

          Web of Science

          Google Scholar

          Show all references

          (0)eLetters

          eLetters is a forum for ongoing peer review. eLetters are not edited, proofread, or indexed, but they are screened. eLetters should provide substantive and scholarly commentary on the article. Neither embedded figures nor equations with special characters can be submitted, and we discourage the use of figures and equations within eLetters in general. If a figure or equation is essential, please include within the text of the eLetter a link to the figure, equation, or full text with special characters at a public repository with versioning, such as Zenodo. Please read our Terms of Service before submitting an eLetter.

          Log In to Submit a Response

          No eLetters have been published for this article yet.

          Recommended articles from TrendMD

          • In vivo priming of human mesenchymal stem cells with hepatocyte growth factor–engineered mesenchymal stem cells promotes therapeutic potential for cardiac repai...

            Bong-Woo Park, Sci Adv, 2020

          • Shattering barriers toward clinically meaningful MSC therapies

            Oren Levy, Sci Adv, 2020

          • Nanovesicles derived from iron oxide nanoparticles–incorporated mesenchymal stem cells for cardiac repair

            Ju-Ro Lee, Sci Adv, 2020

          • Exendin-4 gene modification and microscaffold encapsulation promote self-persistence and antidiabetic activity of MSCs

            Yuanyuan Zhang, Sci Adv, 2021

          • Human gingival mesenchymal stem cells retain their growth and immunomodulatory characteristics independent of donor age

            Jay R. Dave, Sci Adv, 2022

          • The secretion profile of mesenchymal stem cells and potential applications in treating human diseases

            Yuyi Han, Signal Transduction and Targeted Therapy, 2022

          • Mesenchymal Stem Cell Therapy for Ischemic Heart Disease: Systematic Review and Meta-analysis

            Hyunsuk Jeong, International Journal of Stem Cells

          • Mesenchymal Stem Cell and MicroRNA Therapy of Musculoskeletal Diseases

            International Journal of Stem Cells

          • Excessive branched-chain amino acid accumulation restricts mesenchymal stem cell-based therapy efficacy in myocardial infarction

            Fuyang Zhang, Signal Transduction and Targeted Therapy, 2022

            Intracellular delivery of nitric oxide enhances the therapeutic efficacy of mesenchymal stem cells for myocardial infarction

            Tian Hao https://orcid.org/0009-0007-8657-0424, Guangbo Ji https://orcid.org/0009-0000-7079-8896, Meng Qian https://orcid.org/0009-0006-9135-4593, Qiu Xuan Li, Haoyan Huang, Shiyu Deng, Pei Liu, Weiliang Deng https://orcid.org/0009-0001-8402-7280, Yongzhen Wei, [...], and Qiang Zhao https://orcid.org/0000-0003-4656-6002 +9 authorsAuthors Info & Affiliations

            Science Advances

            29 Nov 2023

            Vol 9, Issue 48

            DOI: 10.1126/sciadv.adi9967

            4,3612

            Metrics

            Total Downloads4,361

            • Last 6 Months1,270
            • Last 12 Months3,995

            Total Citations2

            • Last 6 Months1
            • Last 12 Months2

            NotificationsBookmark

            • • Abstract
              • INTRODUCTION
              • RESULTS
              • DISCUSSION
              • MATERIALS AND METHODS
              • Acknowledgments
              • Supplementary Materials
              • REFERENCES AND NOTES
              • eLetters (0)
            • • Information & Authors
              • Metrics & Citations
              • View Options
              • References
              • Media
              • Tables
              • Share

            Abstract

            Cell therapy by autologous mesenchymal stem cells (MSCs) is a clinically acceptable strategy for treating various diseases. Unfortunately, the therapeutic efficacy is largely affected by the low quality of MSCs collected from patients. Here, we showed that the gene expression‎ of MSCs from patients with diabetes was differentially regulated compared to that of MSCs from healthy controls. Then, MSCs were genetically engineered to catalyze an NO prodrug to release NO intracellularly. Compared to extracellular NO conversion, intracellular NO delivery effectively prolonged survival and enhanced the paracrine function of MSCs, as demonstrated by in vitro and in vivo assays. The enhanced therapeutic efficacy of engineered MSCs combined with intracellular NO delivery was further confirmed in mouse and rat models of myocardial infarction, and a clinically relevant cell administration paradigm through secondary thoracotomy has been attempted.

            SIGN UP FOR THE AWARD-WINNING SCIENCEADVISER NEWSLETTER

            The latest news, commentary, and research, free to your inbox daily

            Sign up

            INTRODUCTION

            Mesenchymal stem cells (MSCs) are pluripotent stem cells with high self-renewal abilities and multidirectional differentiation potential (1, 2). They are extensively distributed throughout the body and serve a variety of purposes, including tissue regeneration (3), immunoregulation (4), and angiogenesis (5). However, an important challenge for stem cell therapy is the low survival rate of stem cells after transplantation, which is associated with nonspecific homing of cells and ischemia/hypoxia at the injury site (6). In addition, transplanted stem cells cannot fully exert their paracrine effects in pathological environments, which seriously limits the clinical application of stem cells.

            The proangiogenic function of MSCs is a key factor contributing to the treatment of ischemic diseases. Generally, MSCs can stimulate local angiogenesis in ischemic tissue by secreting cytokines, including vascular endothelial growth factor (VEGF), transforming growth factor β (TGFβ), and basic fibroblast growth factor (bFGF), to induce endothelial cells to form tubular structures (7). In addition, MSCs can recruit pericytes and smooth muscle cells to promote neovascularization (8, 9). Recently, MSCs have been reported to participate in the construction of blood vessels by directly differentiating into endothelial or smooth muscle cells (10).

            As an important signaling molecule, nitric oxide (NO) plays a crucial role in the maintenance of vascular homeostasis by inhibiting thrombus formation and promoting angiogenesis (11, 12). Recently, increasing attention has been given to the regulation of stem cells by NO due to its multiple biological functions (13, 14). NO can regulate the proliferation of stem cells by regulating the activities of cyclin and mitotic receptors, as well as their downstream pathways (15, 16). On the other hand, NO can regulate the expression‎ of angiogenic cytokines and immunomodulatory factors to improve the paracrine performance of stem cells (17). Additionally, recent studies have demonstrated that NO can regulate the differentiation behavior of stem cells through the phosphatidylinositol 3-kinase (PI3K)/AKT, guanosine 3′,5′-monophosphate (cGMP), and other signaling pathways (18).

            As a result, NO-releasing biomaterials have been used as delivery carriers for stem cells to enhance their survival and regulate paracrine functions (19). However, NO, which is a gaseous molecule, easily diffuses and has a high level of instability. Furthermore, the physiologic function of NO is dose dependent, and an overdose of NO often leads to notable cytotoxicity (20). Thus, optimizing the beneficial effects of NO to strengthen the therapeutic efficacy of stem cells by tuning their release profile should be taken into account.

            In our previous study, based on the chemical biology principle of “bump-and-hole,” we designed and prepared an enzyme-prodrug delivery system and achieved targeted delivery of NO at the lesion site in two different ischemic disease models (21). Here, MSCs were further modified by gene transfection to express a catalytic enzyme (A4-β-GalH363A). The engineered MSCs (eMSCs) were transplanted using an injectable hyaluronic acid (HA) hydrogel as the carrier, while the NO prodrug was injected through the tail vein to achieve controlled release of NO catalyzed by the enzymes in eMSCs. The therapeutic efficacy of MSCs combined with exogenous NO delivery was eval‎uated in mouse and rat models of myocardial infarction (MI) with an emphasis on comparing the therapeutic efficacy of two different NO administration methods (intracellular or extracellular), and the underlying mechanism of their myocardial protective effect was further explored.

            RESULTSThe gene expression‎ of MSCs collected from patients is differentially regulated

            In the clinic, the therapeutic efficacy of autologous stem cell transplantation is largely affected by the low quality (including cell survival and paracrine function) of stem cells collected from patients due to chronic diseases. Diabetes mellitus is a chronic medical condition that can lead to a variety of complications, and these complications can affect various parts of the body, such as the kidney, lower limb, and heart. Diabetes predisposes affected individuals to a spectrum of cardiovascular complications, and one of the most debilitating in terms of prognosis is heart failure (22).

            Accordingly, although autologous MSCs have been widely accepted as a promising strategy for treating various complications associated with diabetes, the therapeutic efficacy is largely affected by the quality of stem cells collected from the patients themselves. The gene levels in adipose-derived MSCs (ADMSCs) from patients with diabetes and healthy individuals were first compared by transcriptome sequencing. The heatmap shows that multiple genes in MSCs from patients with diabetes were up- or down-regulated compared to healthy controls (Fig. 1A). Gene ontology (GO) enrichment analysis revealed that the most differentially up-regulated genes were related to tumor necrosis factor (TNF) signaling pathways (Fig. 1B). Subsequently, we performed gene set enrichment analysis (GSEA) based on the RNA-sequencing results. GSEA revealed that inflammatory target genes were highly enriched in MSCs from patients with diabetes, with a normalized enrichment score (NES) of 1.65 (P < 0.01) (Fig. 1C). Apoptosis-related genes were also highly enriched, with an NES of 1.77 (P < 0.01) (Fig. 1D). In addition, through enrichment analysis, we found that proangiogenic genes in MSCs were greatly down-regulated in patients with diabetes (Fig. 1E).

            

            Fig. 1. Transcriptome sequencing of adipose mesenchymal stem cells (MSCs) collected from patients with diabetes and healthy controls.

            (A) Heatmap showing the differentially expressed genes of MSCs from patients with diabetes and healthy controls (n = 3). (B) Gene ontology (GO) analysis of the up-regulated transcriptome of MSCs from patients with diabetes. (C and D) Gene set enrichment analysis (GSEA) was performed to determine the enrichment of inflammation (C) and apoptosis (D) target genes in the diabetic group. (E) Heatmap showing angiogenesis-related genes in the two groups (n = 3). (F) Schematic illustration demonstrating the difference in MSCs between patients with diabetes and healthy controls at the gene level.

            Open in viewer

            eMSCs are constructed to express mutant β-galactosidase

            Here, we first designed and constructed plasmids carrying the mutant β-galactosidase (A4-β-GalH363A) target gene and luciferase–red fluorescent protein (RFP) dual reporter genes, which could be further used for in vivo imaging. eMSCs expressing A4-β-GalH363A were constructed by infecting MSCs with lentiviruses obtained from human embryonic kidney 293T cells (Fig. 2A). Immunofluorescence staining for RFP confirmed that A4-β-GalH363A was successfully expressed by eMSCs (Fig. 2B). The subcellular fraction and intracellular distribution of enzymes expressed by the eMSCs was determined by Western blotting (Fig. 2C). In contrast to natural β-galactosidases, which are widely distributed within cells, A4-β-GalH363A was mainly confined to the nucleus of eMSCs.

            

            Fig. 2. Intracellular expression‎ and localization of A4-β-GalH363A.

            (A) Schematic diagram of lentivirus packaging and mesenchymal stem cell (MSC) infection. (B) Immunofluorescence staining of β-Gal and A4-β-GalH363A in eMSCs. Scale bar, 50 μm. (C) The distribution of two different enzymes in engineered MSCs (eMSCs) was analyzed by Western blotting.

            Open in viewer

            Intracellular release of NO is realized via decomposition of the 6-OMeGal-Ph-NO prodrug

            Since the mutant β-galactosidase was confined to the nucleus in eMSCs, we designed a prodrug by introducing a lipid-soluble self-decomposition chain into 6-OMeGal-NO to improve its oil and water distribution coefficient; therefore, the resultant NO donor 6-OMeGal-Ph-NO could penetrate the cell membrane and decompose and release NO under the catalysis of A4-β-GalH363A (Fig. 3A and Supplementary Materials). An in vitro release assay showed that the 6-OMeGal-Ph-NO prodrug was efficiently recognized and converted by A4-β-GalH363A with a cumulative release ratio of approximately 97.3%, while nearly no release was observed in the presence of wild-type β-galactosidase (Fig. 3B). To confirm‎ intracellular NO release, eMSCs were preincubated with an NO fluorescent probe (DAF-AM DA) and then treated with different NO prodrugs. The changes in fluorescence signals were examined by time-delay cell imaging (Fig. 3C). The results indicated that the fluorescence intensity continuously increased in the group that was treated with 6-OMeGal-Ph-NO, indicating conversion into NO (Fig. 3D). In contrast, no detectable changes were identified in the β-Gal-NO group because the high water solubility restricted its entry into eMSCs. Next, intracellular and extracellular release of NO in eMSCs was assessed (Fig. 3E). The quantity of intracellular NO was measured by electron paramagnetic resonance (EPR) using ferrous N-diethyl dithiocarbamate (DETC2-Fe) as the spin-trapping reagent. The resultant NO adduct (DETC2-Fe-NO) exhibited a characteristic triplet EPR signal (aN = 13.06 G, giso = 2.041) at room temperature. Quantitative analysis showed that the NO level was significantly (P < 0.001 or 0.0001) higher in eMSCs treated with 6-OMeGal-Ph-NO than in the β-Gal-NO and control groups (Fig. 3F). Furthermore, in the group treated with β-Gal-NO, the release of NO was mainly catalyzed by β-galactosidase that translocated from the cytoplasm in eMSCs, and the extracellular release profile of β-Gal-NO was confirmed by detecting the NO level in the cell culture medium with the NO-sensitive near-infrared fluorescence probe (23); it was significantly (P < 0.01) higher than that in the 6-OMeGal-Ph-NO and control groups (Fig. 3G). To determine the uptake of the NO prodrug by eMSCs, we incubated 6-OMeGal-Ph-NO with eMSCs, and the concentration in the culture medium was determined at different time points. The results reflected that approximately 45% of 6-OMeGal-Ph-NO was incorporated into the eMSCs within 12 hours (fig. S1).

            

            Fig. 3. Intracellular generation of nitric oxide (NO) from the NO prodrug under the catalysis of A4-β-GalH363A expressed by engineered mesenchymal stem cells (eMSCs).

            (A) Synthesis of two NO prodrugs with different enzyme response abilities and cellular permeabilities. (B) In vitro release profile of NO from the NO prodrug (6-OMeGal-Ph-NO) in the presence of β-Gal or A4-β-GalH363A. (C) Schematic illustration of intracellular NO imaging by using an NO fluorescence probe (DAF-AM DA). (D) Representative time-lapse images of NO generation from two different prodrugs in eMSCs and quantification of the fluorescence intensity (n = 6). ***P < 0.001, ****P < 0.0001 versus 6-OMeGal-Ph-NO group. (E) Schematic illustration showing the detection of intracellular and extracellular NO generation differentially. (F) Representative electron paramagnetic resonance (EPR) spectra and quantification of intracellular NO generation by measuring the DETC2-Fe-NO complex using 2,2,5,5-tetramethyl piperidine 1-oxyl (TEMPO) as a standard (n = 3). (G) Relative quantification of NO production in the medium determined using the near-infrared fluorescence probe (n = 4). Data are expressed as the mean ± SEM. Significant differences were detected by one-way analysis of variance (ANOVA) with Tukey’s multiple comparisons test, *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001.

            Expand for more

            Open in viewer

            Intracellular delivery of NO inhibits apoptosis and enhances the proangiogenic activity of eMSCs

            Overproduction of reactive oxygen species due to cellular oxidative stress has been accepted as an important factor contributing to apoptosis in transplanted cells in ischemic tissue. Therefore, we assessed the protective effect of exogenously administered NO on the survival of eMSCs with H2O2-induced oxidative stress and focused on comparing the protection provided by intracellular and extracellular NO administration. The results showed that H2O2 stimulated apoptosis, and delivery of NO via extracellular and intracellular strategies significantly (P < 0.01 or 0.001) reduced apoptosis in eMSCs stimulated by oxidative stress, and the highest fluorescence signal was observed in response to intracellular NO delivery (Fig. 4A).

            

            Fig. 4. Intracellular delivery of nitric oxide (NO) inhibits apoptosis of engineered mesenchymal stem cells (eMSCs).

            (A) Bioluminescence imaging (BLI) was used to detect the effect of NO delivery on cell apoptosis stimulated by different concentrations of H2O2, and the fluorescence signals were further quantified (n = 3). **P < 0.01, ***P < 0.001, ****P < 0.0001 versus control group; #P < 0.05 versus β-Gal-NO group. (B) Flow cytometry assay of cell viability and apoptosis of eMSCs after H2O2 stimulation, and quantification of mean percent values of apoptotic cells (n = 3). (C) The expression‎ of apoptosis-related protein (BCL2, Bax, Bad, caspase3, and cleaved caspase3) by eMSCs was detected after H2O2 stimulation by Western blots (n = 3). *P < 0.05, **P < 0.01, ***P < 0.001. (D) Representative BLI photographs reflecting the retention of eMSCs with and without intracellular NO delivery after in vivo transplantation as well as the quantitative analysis of signals (n = 3). Data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01.

            Open in viewer

            The survival of eMSCs after treatment with various NO delivery strategies was compared by using flow cytometry. Intracellular NO delivery had the most remarkable antiapoptotic effect on cells stimulated by oxidative stress (P < 0.001) (Fig. 4B). The expression‎ of apoptosis-related genes by eMSCs followed a similar trend; NO delivery (extracellular and intracellular) effectively promoted the expression‎ of the antiapoptotic gene BCL2 in eMSCs after oxidative stress stimulation, while the expression‎ levels of proapoptotic genes were reduced accordingly. Intracellular NO delivery via administration of the 6-OMeGal-Ph-NO prodrug exerted a more pronounced antiapoptotic effect at both the gene and protein levels (Fig. 4C and fig. S2A), which may be because the intracellular generation of NO directly activated the antioxidant system in cells to resist oxidative stress damage and inhibit further apoptosis.

            The expression‎ of proangiogenic genes, including ANGPT1, ANGPT2, FGF2, VEGFA, and KDR, in eMSCs was further detected by reverse transcription polymerase chain reaction (RT-PCR). The results showed that the expression‎ level of proangiogenic genes was significantly (P < 0.05, 0.01, or 0.001) higher in eMSCs treated with the 6-OMeGal-Ph-NO prodrug than in the other groups, indicating the enhanced proangiogenic functions of eMSCs after intracellular NO delivery (fig. S2B).

            We further eval‎uated the effect of NO delivery on the in vivo retention of eMSCs after orthotopic transplantation in the myocardial tissue of mice. As shown in Fig. 4D, intracellular delivery of NO effectively prolonged the retention of eMSCs within the myocardium, and an evident bioluminescence imaging (BLI) signal corresponding to the retention of eMSCs was observed 7 days after transplantation compared to the counterpart without administration of the NO prodrug. To further eval‎uate the translational potential of eMSCs in clinical settings, we used MSCs derived from diabetic patients and conducted a series of assays related to cell survival and paracrine function. The findings indicated that intracellular delivery of NO also confers advantages in the attenuation of cell apoptosis under stress conditions, thereby prolonging the in vivo retention of eMSCs (fig. S3).

            Intracellular delivery of NO ameliorates myocardial injury in MI mice after treatment with eMSCs

            The therapeutic efficacy of MSCs combined with exogenous NO was further eval‎uated in a mouse MI model (Fig. 5A and fig. S4). The inflammatory response in the early stage (3 days) was first detected by hematoxylin-eosin (H&E) staining and CD68 immunofluorescence staining. The results demonstrated that severe inflammatory cell infiltration occurred in the injured myocardium of MI mice, and it was effectively alleviated after eMSC treatment. More prominent restoration in the injured myocardium was observed after further administration of NO (Fig. 5, B and C), confirm‎ing the inhibitory effect on inflammation after MI provided by the combination of MSCs and NO. Moreover, this inhibitory effect was more significant (P < 0.01 or 0.0001) in the group with intracellular NO delivery than in the group with extracellular NO delivery.

            

            Fig. 5. Intracellular delivery of nitric oxide (NO) ameliorates myocardial injury in myocardial infarction (MI) mice after treatment with engineered mesenchymal stem cells (eMSCs) via regulation of the inflammation and stimulation of angiogenesis.

            (A) Experimental schedule for the treatment of MI in a mouse model. (B) Hematoxylin-eosin (H&E) staining was performed to detect inflammatory cell infiltration in the early stage of MI (n = 6). Scale bar, 100 μm. (C) Representative images of CD68 immunofluorescence staining (green) and quantification of CD68+ macrophages in injured myocardium (n = 6). Scale bar, 25 μm. (D and E) Flow cytometry was performed to detect peritoneal macrophage polarization 7 days after surgery followed by different treatments. TNFα- and CD206-positive ascites macrophages (markers of M1 and M2 macrophage phenotypes, respectively) were quantified accordingly (n = 3). (F) Representative images of α–Smooth muscle actin (α-SMA) immunofluorescence staining and the quantification of α-SMA+ arterioles (n = 6). Scale bar, 100 μm. (G) Representative images of von Willebrand factor (vWF) immunofluorescence staining and the quantification of vWF+ capillaries (n = 6). Scale bar, 100 μm. Data are expressed as the mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

            Expand for more

            Open in viewer

            It has been widely accepted that a higher proportion of M2-type macrophages is more conducive to the repair of tissue damage. Thus, peritoneal macrophages were extracted at 7 days after surgery and examined by flow cytometry to assess the polarization of macrophages in MI mice that received different treatments. MI modeling leads to a marked increase in the polarization of macrophages toward the M1 phenotype (TNF-α positive); however, the ratio of M1-type macrophages was moderately reduced after treatment with eMSCs. Further administration of NO via the intracellular delivery method significantly (P < 0.001 or 0.01) inhibited the polarization of macrophages toward the M1 phenotype while increasing the proportion of M2-type macrophages compared to the control and eMSC groups (Fig. 5, D and E). Additionally, we conducted immunofluorescence staining in heart section (fig. S5). The results revealed that intracellular delivery of NO also induces the polarization of macrophages into the M2 phenotype within the heart.

            In vitro studies demonstrated that exogenous NO could improve the proangiogenic capacity of eMSCs. Here, we further explored the influence of the combined delivery of exogenous NO and eMSCs on the reconstruction of the vascular network at the site of infarction. α–Smooth muscle actin (α-SMA)–positive arterioles and von Willebrand factor (vWF)–positive small vessels in MI mice after the different treatments were detected by immunofluorescence staining (Fig. 5, F and G). Treatment with eMSCs efficiently promoted angiogenesis in the injured myocardium, and more prominent enhancement was observed in response to further treatment with intracellular NO. This finding was further supported by the expression‎ of angiogenesis-related genes in the border zone of the infarcted heart (fig. S6).

            Intracellular delivery of NO improves heart function and inhibits adverse myocardial remodeling in MI mice after treatment with eMSCs

            Ultrasound and histological analyses were performed to eval‎uate the long-term recovery of cardiac function after MI. Cardiac injury was first eval‎uated by 2,3,5-triphenyltetrazolium chloride (TTC) staining (Fig. 6A). Treatment with eMSCs moderately repressed MI compared to the acute myocardial infarction (AMI) group, but a more pronounced inhibitory effect was observed in the group with further intracellular NO delivery. Left ventricular function was assessed by echocardiography at different time points. As shown in Fig. 6B, after 1 day of MI, the left ventricle in each group was markedly enlarged, cardiac function decreased rapidly, and deterioration of heart function continued for 28 days without detectable restoration in the AMI group. However, eMSC treatment could restore left ventricular systolic function and reduce ventricular dilation, as shown by the increase in left ventricular ejection fraction (LV-EF) and fraction shortening (LV-FS), as well as the decrease in left ventricular end-diastolic diameter (LVIDd) and left ventricular end-diastolic volume (LV-EDV) to a certain extent. In the group treated with eMSCs and intracellular NO delivery (NO-eMSCs), LV-EF and LV-FS were effectively recovered, while LVIDd and LV-EDV were significantly (P < 0.001 or 0.0001) enhanced compared to the AMI group.

            

            Fig. 6. Intracellular delivery of nitric oxide (NO) improves heart function and reduces adverse cardiac remodeling in myocardial infarction (MI) mice after treatment with engineered mesenchymal stem cells (eMSCs).

            (A) Representative images of 2,3,5-triphenyltetrazolium chloride (TTC) staining and quantification of the infarct area (n = 3). Scale bar, 2 mm. (B) Cardiac echo measurement was performed at different time points after surgery, and cardiac function indicators of left ventricular ejection fraction (LV-EF), left ventricular fractional shortening (LV-FS), left ventricular internal diameter at end diastole (LVIDd), and left ventricular end-diastolic volume (LV-EDV) were eval‎uated accordingly (n = 6). ***P < 0.001, ****P < 0.0001 versus acute myocardial infarction (AMI) group. (C) Masson’s trichrome staining was performed, and the infarct size was quantified accordingly (n = 6). (D) Collagen deposition in the hearts was detected by Sirius Red staining (n = 6). Scale bar, 100 μm. (E) Representative images of wheat germ agglutinin (WGA) immunofluorescence staining and quantification of the cross-sectional area of cardiomyocytes (n = 6). Scale bar, 50 μm. (F) Representative images of immunofluorescence staining (red) for the gap junction protein (Cx43) and the quantification of the intensity of red fluorescence to the whole area of images (n = 6). Scale bar, 25 μm. Data are expressed as the mean ± SEM. *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001.

            Expand for more

            Open in viewer

            Masson staining was used to detect the degree of myocardial fibrosis after MI. Severe myocardial fibrosis was observed in MI mice compared to the sham group (Fig. 6C). In addition, the ventricular wall became thinner, which was closely related to impaired left ventricular systolic function after MI, as demonstrated by echocardiography. However, these adverse cardiac remodeling events were ameliorated after eMSC treatment and accompanied by a lowered degree of myocardial fibrosis and a thickened ventricular wall in contrast to the AMI group. Notably, treatment with eMSCs and intracellular NO delivery exerted the most prominent inhibitory effect on myocardial fibrosis after MI (Fig. 6C).

            Collagen deposition in MI mice was assessed by Sirius staining (Fig. 6D), and the results showed that MI resulted in severe collagen deposition in the injured myocardium compared to the sham group. It was effectively reduced after eMSC treatment, and the inhibitory effect of eMSC plus intracellular NO delivery was significantly higher than that in the other two groups (P < 0.001 or 0.0001).

            Next, wheat germ agglutinin (WGA) staining was carried out to eval‎uate myocardial cell hypertrophy 28 days after MI (Fig. 6E). The cross-sectional area of cardiomyocytes was increased in MI mice in contrast to the sham operation group due to compensatory hypertrophy in the heart to maintain the normal rate of cardiac ejection. Hypertrophy was significantly (P < 0.05) mitigated after treatment with eMSCs, especially in the presence of exogenous NO (P < 0.001), indicating an ideal therapeutic effect on inhibiting myocardial cell hypertrophy and adverse ventricular remodeling by the combination of eMSCs and NO.

            Gap junctions (GJs) are the main connections between cardiomyocytes in the heart, and Cx43 is the main GJ protein in ventricular muscle in the heart (24). Studies have shown that the absence of Cx43 leads to the occurrence of cardiac ventricular arrhythmia, which can develop into heart failure (25). After 28 days of MI, immunofluorescence staining for Cx43 revealed abundant and uniform distribution of GJ proteins in the sham group, whereas MI injury resulted in a marked decrease in the expression‎ of Cx43 (Fig. 6F). Despite the moderate inhibitory effect provided by the administration of eMSCs, further delivery of NO via the intracellular method significantly enhanced (P < 0.001 or 0.01) the expression‎ of Cx43 compared to that in the AMI or eMSC groups.

            Intracellular delivery of NO enhances the therapeutic efficacy of eMSCs in a rat MI model

            Although the outcome in a mouse model supported the beneficial effect of NO via intracellular delivery on enhancing the therapeutic efficacy of MSCs for MI, immediate administration of stem cells after MI is different from the clinical treatment of MI due to the limitation of the administration paradigm in mouse models. In addition, 3 to 7 days after MI is the outbreak period of the inflammatory response. For this reason, we established a rat model of MI and conducted secondary thoracotomy 3 days after surgery (Fig. 7A), and eMSCs were delivered via an injectable HA hydrogel as the carrier (Fig. 7B). Lactate dehydrogenase (LDH), a crucial marker for assessing the extent of myocardial damage, exhibited an initial elevation within 2 to 48 hours following the onset of MI, reaching its zenith between 2 and 5 days after MI. We collected blood samples from the orbital venous plexus of rats 5 days after MI to measure serum LDH levels (fig. S7). The findings revealed a sharp increase in serum LDH levels due to MI. However, treatment with NO-eMSCs significantly reduced serum LDH levels, indicating an attenuation of cardiac injury.

            

            Fig. 7. Intracellular delivery of nitric oxide (NO) enhances the therapeutic efficacy of engineered mesenchymal stem cells (eMSCs) in a rat myocardial infarction (MI) model.

            (A) Experimental schedule for the treatment of rat MI. (B) Representative images showing the second thoracotomy in rats after MI. (C) Cardiac echo measurement was performed at different time points after surgery, and cardiac function indicators of left ventricular ejection fraction (LV-EF), left ventricular fractional shortening (LV-FS), left ventricular internal diameter at end diastole (LVIDd), and left ventricular end-diastolic volume (LV-EDV) were eval‎uated accordingly (n = 6). **P < 0.01, ***P < 0.001 versus acute myocardial infarction (AMI) group. (D) Representative images of Masson’s trichrome staining and quantification of the infarct size and infarct thickness (n = 6). **P < 0.01, ***P < 0.001, ****P < 0.0001 versus AMI group. (E) Representative images of wheat germ agglutinin (WGA) immunofluorescence staining and quantification of the cross-sectional area of cardiomyocytes (n = 6). Scale bar, 50 μm. (F) Representative images of α–Smooth muscle actin (α-SMA) immunofluorescence staining and quantification of α-SMA+ arterioles (n = 6). Scale bar, 100 μm. Data are expressed as the mean ± SEM. **P < 0.01, ***P < 0.001, ****P < 0.0001.

            Open in viewer

            Echocardiography was performed to detect heart function at different time points after MI (Fig. 7C). At 4 weeks after surgery, the anterior wall of the left ventricle was completely infarcted, and the ventricular cavity was increased dramatically. Quantitative analysis further indicated that both EF and FS were decreased markedly after surgery. Treatment with MSCs could restore the systolic function of the heart, and the highest recovery rate (60 to 80%) was observed in the group treated with eMSCs and intracellular NO delivery. Ventricular dilatation was also effectively restored after combined treatment with eMSCs and NO, which was consistent with the results obtained in the mouse model.

            Histological analyses showed that treatment with eMSCs combined with NO efficiently reduced the degree of fibrosis (Fig. 7D) and collagen deposition (fig. S8A). Furthermore, it alleviated infarct size after MI and restored ventricular cavity morphology of the heart with a significant enhancement in the thickness of the infarcted left ventricular wall and the interventricular septum (IVS), as eval‎uated by Masson (Fig. 7D) and H&E staining (fig. S9). Twenty-eight days after MI, WGA staining also showed that the combination of eMSCs and NO significantly (P < 0.01) inhibited cardiomyocyte hypertrophy in contrast to the AMI group, thus inhibiting further myocardial systolic dysfunction (Fig. 7E). The expression‎ of the GJ protein Cx43 (fig. S8B) followed a similar trend to that in mouse models; that is, acute MI led to a marked decrease in the distribution of Cx43 in the myocardium, and treatment with eMSCs up-regulated the expression‎ of Cx43 to a certain extent. Further delivery of NO via the intracellular method produced a more significant (P < 0.05 or 0.01) effect on the up-regulation of Cx43 in the injured myocardium compared to the eMSC and AMI groups, thereby facilitating the connection between cardiomyocytes and further inhibiting the occurrence of arrhythmias and heart failure.

            Tissue repair after MI is often closely related to angiogenesis, which begins at the infarct border and extends to the center of the infarction (26). Therefore, we further compared the proangiogenic effect of eMSCs with and without NO delivery on the damaged margin of the infarcted myocardium 28 days after MI in rats (Fig. 7F and fig. S8C). The combination of eMSCs and NO remarkably promoted angiogenesis and reconstruction of the vascular network compared to the group treated with eMSCs only, which is beneficial to the repair of myocardial injury after MI.

            DISCUSSION

            Cell therapy based on MSCs has proven to be a powerful solution for treating degenerative diseases and tissue damage (27–29). Despite the advantages of autologous stem cells over allogeneic stem cells, including the absence of immune rejection, the decreased survival and impaired paracrine functions of stem cells collected from patients with chronic diseases greatly limit their clinical use (7, 30, 31). Here, we first showed that MSCs collected from patients with diabetes exhibited marked up-regulation of apoptosis- and inflammation-related genes and down-regulation of proangiogenic genes, affecting the efficacy of cell therapy. As a result, genetic engineering strategies are often required to enhance the therapeutic efficacy of autologous stem cells. A recent study revealed that eMSCs, engineered to express PD-L1 on their surface and secrete CTLA4-Ig (immunoglobulin) as an extracellular factor, exhibited immunoprotective properties, which improved the outcome of both syngeneic and allogeneic islet transplantation in diabetic mice (32).

            NO is involved in a variety of physiological processes. Studies have shown that as an important signaling molecule, NO plays a pivotal role in regulating stem cell behavior (33–35), including cell survival, migration, differentiation, and paracrine behavior. These factors affect the interaction of stem cells with other cells and the tissue microenvironment. Previously, different types of NO-releasing biomaterials, such as injectable hydrogels, have been prepared by us and other groups (36–40), and further studies have shown that the combination of NO and MSCs is more effective in treating various diseases than MSC therapy alone. In addition, it has been reported that pretreatment of MSCs with NO-releasing biomaterials could enhance the therapeutic efficacy of MSCs and their secreted exosomes because of their enhanced proangiogenic functions (41).

            Due to the spatiotemporal characteristics of NO (42), precise delivery of NO in a site-specific and controllable manner holds great importance in the regulatory effect of exogenously administered NO. In addition to the controlled release rate, the site at which NO is generated is also a key factor due to the relative half-life and limited diffusion distance (43, 44). It is reasonable to speculate that intracellular and extracellular NO delivery may lead to different outcomes when regulating the survival and function of MSCs. In our previous work, an enzyme-prodrug delivery system was designed based on a bump-and-hole strategy (21). The mutant galactosidase (A4-β-GalH363A) enables the targeted delivery of NO, thus reducing the side effects due to the unspecific decomposition of the NO prodrug and enhancing the therapeutic efficacy. Here, we transfected a plasmid expressing mutant galactosidase into MSCs and successfully constructed eMSCs. The enzyme expressed by MSCs could catalyze the decomposition of the 6-OMe-galactose–protected NO prodrug and release NO intracellularly.

            Western blotting and fluorescence imaging demonstrated that the expression‎ of the engineered enzyme was confined to the nucleus of MSCs, while wild-type β-galactosidase was widely distributed in the cytoplasm, including the lysosome and perinuclear region (45, 46). Since the corresponding prodrug for wild-type β-galactosidase is highly hydrophilic, it fails to enter MSCs and releases NO extracellularly by enzymes that translocate from the cell. In contrast, the prodrug for mutant galactosidase is cell penetrating because of the modified molecular structure; therefore, it can enter MSCs and release NO intracellularly under the catalysis of the corresponding enzyme expressed by the cells. Accordingly, two different NO delivery paradigms were successfully developed in this study and further confirmed by a series of eval‎uations, including cell imaging and electronic paramagnetic resonance. Further in vitro and in vivo assays indicated that in contrast to extracellular NO delivery, intracellular administration of NO enhanced cell survival and the paracrine effects of MSCs, including inhibiting apoptosis and supporting angiogenesis.

            Next, we established a mouse MI model to systematically eval‎uate the therapeutic efficacy of MSCs combined with exogenous NO. The results showed that intracellular delivery of NO prolonged the retention of eMSCs after myocardial orthotopic transplantation. In addition, the combination of eMSCs and intracellular NO delivery improved cardiac function after MI and reduced adverse ventricular remodeling compared to the group treated with MSCs only. Additionally, it could effectively restore the reconstruction of the blood vessel network and further promote the repair of the infarcted myocardium.

            To gain further insight into the translational potential of the combinatory therapeutic strategy developed in this study, a rat model of MI was established, and MSCs were administered by a second thoracotomy after 3 days to mimic the clinical use of MSCs for the treatment of MI (47, 48). Clinically, acute MI is typically due to the rupture of coronary atherosclerotic plaque and the formation of thrombus, which causes coronary artery obstruction. After the acute phase of MI, adverse ventricular remodeling further affects the prognosis of patients, which is specifically characterized as a decrease in ventricular wall thickness and myocardial tension in the MI area, myocardial hypertrophy in the noninfarction area, and a change in the morphology of the ventricular cavity, thus leading to arrhythmia and further development into heart failure. The efficacy of MSCs in managing arrhythmias remains a topic of ongoing debate. Some researchers argue that MSCs do not appear to reduce or prevent arrhythmias, with the antiarrhythmic or proarrhythmic potential of MSCs primarily relying on paracrine factors (49). Conversely, other studies suggest that MSCs themselves may play a role in the post-MI recovery process (50). In our study, we observed an evident up-regulation of Cx43 expression‎ after NO-eMSC treatment, which is a potential target associated with antiarrhythmic effects. Further investigation is still required to comprehensively explore the antiarrhythmic potential of NO-eMSCs. In line with the enhanced therapeutic efficacy in the mouse model, intracellular delivery of NO showed enormous advantages in the rat MI model by inhibiting apoptosis and enhancing the paracrine function of MSCs.

            In summary, we first showed that survival and paracrine function were reduced in MSCs collected from patients with diabetes, which could greatly affect therapeutic efficacy. Accordingly, eMSCs were successfully constructed, and the mutant β-galactosidase expressed by the cells enabled the intracellular generation of NO via the conversion of an exogenous NO prodrug. In vitro and in vivo assays indicated that intracellular delivery of NO effectively enhanced the survival of transplanted MSCs and promoted the paracrine function of MSCs, which was further confirmed by the enhanced therapeutic efficacy in mouse and rat models of MI compared to the group treated with MSCs only. This synergistic strategy provides an option for the treatment of MI by autologous MSCs in the clinic.

            MATERIALS AND METHODSRNA sequencing analysis

            RNA sequencing was performed by the BGI (Shenzhen, China). Briefly, RNA from the ADMSCs of healthy people and patients with diabetes was extracted using TRIzol reagent (Yeasen, China). RNA samples were sequenced on the BGISEQ platform. The raw data containing low-quality reads, adaptor sequences, and high levels of N bases were filtered before analysis. Then, the clean reads were mapped to the reference genome using HISAT, and Bowtie2 was used to align the clean reads to the reference genes. The reference genome source is National Center for Biotechnology Information (NCBI), and the reference genome version is GCF_000001405.39_GRCh38.p13. The expression‎ levels of genes were quantified to identify differentially expressed genes by RNA-Seq by expectation maximization (RSEM). The analyses of hierarchical clustering and heatmap were performed using the online Dr. Tom system (biosys.bgi.com) to compare differential gene expression‎ of ADMSCs in healthy people and patients with diabetes. According to the KEGG_pathway annotation classification, the phyper function in R software was used for enrichment analysis, the P value was calculated, and then false discovery rate (FDR) was performed on the P value to obtain a Q value. Generally, a Q value of ≤0.05 was regarded as significant enrichment. GSEA was used to analyze significant differences in gene expression‎ between inflammatory and apoptosis-related pathways. Expression‎ cluster heatmap was used to analyze the expression‎ of genes associated with angiogenesis.

            Measurement of NO release

            The NO-releasing profile was determined by the Griess kit assay. In brief, 50 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were dissolved in phosphate-buffered saline (PBS) buffer (pH 7.4), and enzymes were added to the solutions at a concentration of 0.005 mg/ml. At each predetermined time interval, 50 ml of solution was transferred into a 96-well plate, and 50 ml of Griess I and 50 ml of Griess II were added thereafter. The azo compound of purple color was formed, and the absorbance was measured at a wavelength of 540 nm using an iMark microplate reader (Bio-Rad, USA).

            Cell cultureMesenchymal stem cells

            MSCs derived from human umbilical cord were obtained from Health-Biotech, maintained in Dulbecco's modified Eagle’s medium (DMEM)/F12 (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) (Gibco, USA) and 1% penicillin- streptomycin solution (Gibco, USA), and placed in a cell culture chamber containing 5% CO2 at 37°C.

            Human embryonic kidney 293T cells

            Human embryonic kidney (HEK) 293T cells were obtained from the American Type Culture Collection (ATCC), maintained in high-glucose DMEM (Gibco, USA) with 10% FBS (Gibco, USA) and 1% penicillin-streptomycin solution (Gibco, USA), and placed in a cell culture chamber containing 5% CO2 at 37°C.

            Construction of eMSCs

            The coding sequence of mutant β-galactosidase (A4-β-GalH363A) can be obtained from the previous publication (21). The lentivirus packaging system containing A4-β-GalH363A sequence and Rluc-RFP sequence was constructed by Wuhan Miaolingbio Co. Ltd. The constructed lentivirus plasmid containing the target gene and the package gene (psPAX2 and pMD2.G) was transfected into HEK 293T cells through Lipo2000, and the supernatant was collected to obtain the virus solution. After removing impurities, the virus solution was mixed 1:1 with fresh MSC medium, and polybrene (10 μg/ml) was added. MSCs were infected with virus through incubation in the mixture medium. The infection efficiency was observed under an inverted fluorescence microscope, and the expression‎ of target protein was determined by Western blotting.

            Cell immunofluorescence staining

            eMSCs were inoculated in 24-well plates. Cells were fixed with 4% paraformaldehyde and blocked in 4% bovine serum albumin in PBS for 30 min at room temperature. Then, the cells were incubated with primary antibodies overnight at 4°C. The bound primary antibodies were displayed by incubation with the secondary antibodies for 2 hours at room temperature. Cells were counterstained with DAPI (4′,6-diamidino-2-phenylindole)–containing Fluoromount-G and visualized with a fluorescence microscope (Axio Imager Z1). Antibodies used include anti–β-galactosidase (1:100, A1863, Abclonal) and anti-RFP (1:100, PA1-986, Invitrogen).

            Western blot

            eMSCs were collected, and total protein was extracted using radioimmunoprecipitation assay (RIPA) lysate containing protease inhibitor (Solarbio, China). Cytoplasmic protein and nucleoprotein were extracted using a nucleoprotein extraction kit containing protease inhibitors (Solarbio, China). The protein concentration was quantified using a BCA protein assay kit (Solarbio, China). The samples were diluted with 4× SDS–polyacrylamide gel electrophoresis (PAGE) loading buffer and boiled in boiling water for 8 min. Then, 30 μg of protein was isolated by 10% SDS-PAGE electrophoresis. The isolated proteins were transferred to an Immobilon-P Transfer membrane (Millipore, USA) and incubated with the primary antibody overnight at 4°C and then with the secondary antibody at room temperature for 2 hours. The bands were detected with chemiluminescent horseradish peroxidase substrate (Millipore, USA). Signals were generated by using an enhanced chemiluminescence (ECL) reagent (Millipore, USA) and were captured by using the Tanon-5200 Chemiluminescence Imaging System (Tanon, China). The antibodies used included anti–β-galactosidase (1:1000, A1863, Abclonal), anti–His-tag (1:1000, 12698S, Cell Signaling Technology), anti–glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (1:1000, AC001, Abclonal), anti–histone H3 (1:1000, 4499S, Cell Signaling Technology), anti–β-actin (1:1000, UM4001, Utibody), anti-BCL2 (1:1000, WL01556, Wanleibio), anti-Bax (1:1000, 50599-2-lg, Proteintech), anti-Bad (1:1000, WL02140, Wanleibio), and anti-caspase/cleaved caspase3 (1:1000, WL02117, Wanleibio).

            6-OMeGal-Ph-NO uptake by eMSCs

            eMSCs were inoculated in 12-well plates, and 50 μM substrate (6-OMeGal-Ph-NO) was added per well. At predetermined time points (0, 1, 3, 6, and 12 hours), an appropriate amount of culture medium was collected, and an excess of A4-β-GalH363A was added immediately to fully catalyze the decomposition of the remaining substrate at room temperature. The amount of 6-OMeGal-Ph-NO substrate in the culture medium was determined by Griess kit assay.

            Real-time imaging of intracellular NO

            The fluorescence emission associated with NO in the cytosol was detected by an electron-multiplying charge-coupled device (DU-897D-CS0-BV; Andor, Belfast, UK) connected to an inverted fluorescence microscope (Axio Observer D1; Carl Zeiss, Oberkochen, Germany). Intracellular NO imaging was performed using an NO fluorescence probe, DAF-AM DA (Beyotime, China), according to the manufacturer’s instruction. eMSCs were inoculated in small confocal dishes in advance, and the experiment was conducted when the cell density reached 80%. First, the medium was collected for later use. After two gentle washes with PBS, 5 μM DAF-AM DA solution was incubated at 37°C for 30 min in the dark. Then, cells were gently washed with PBS twice. The previously collected medium was added anew, and cell imaging was performed immediately. At the 488-nm excitation wavelength, pictures were taken every 5 s. After stable shooting for 2 min, 100 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were added. The filming for cell fluorescence was continued for 4 min. The final fluorescence intensity was determined without the background fluorescence value. The proportion of change in fluorescence intensity of each cell in the visual field was calculated.

            Intracellular NO detection

            Intracellular NO radicals (NO•) were detected using EPR as described (51, 52). In brief, sodium DETC (4.5 mg) and FeSO4•7H2O were dissolved in two separate volumes (10 μl) of deoxygenated Krebs/Hepes solution. Equal volumes of these parent solutions were rapidly mixed and aspirated into Eppendorf combi tips. The 0.5 mM Fe•(DETC)2 colloid solution had a yellow-brownish color with a slight opalescence in light. No aggregate formation was observed, at least during the first 30 min. eMSCs were rinsed with modified Krebs/Hepes buffer and incubated with freshly prepared NO•-specific spin trap Fe•(DETC)2 colloid (0.5 mM) for 30 min. Meanwhile, 100 μM NO prodrugs (β-Gal-NO/6-OMeGal-Ph-NO) were added to the buffer. Gently collected cell suspensions were snap-frozen in liquid nitrogen. Ethyl acetate (200 μl) was added, and the cells were ultrasonically broken to extract DETC2-Fe-NO. The ethyl acetate extract was concentrated with nitrogen and transferred to a 50-μl capillary, and then the X-band EPR was measured at room temperature. The following acquisition parameters were used: modulation frequency, 100 kHz; microwave power, 10 mW; modulation amplitude, 2 G; number of scans, 60. The double-integrated area of the EPR spectra was calibrated into concentrations of DETC2-Fe-NO using TEMPO (2,2,5,5-tetramethyl piperidine 1-oxyl) as a standard. EPR spectral simulation was conducted by the WINSIM program.

            Extracellular NO detection

            eMSCs were treated with β-Gal-NO or 6-OMeGal-Ph-NO (30 μM). The production of NO in the medium of each group was detected 6 hours after incubation with NO-sensitive near-infrared fluorescence probe (5 μM). The NO production of medium in different groups was compared by the relative fluorescence intensity under the excitation at 750 nm (emission at 800 nm).

            Cell apoptosis detection

            To test the protective effect of NO delivery on cellular oxidative stress stimulation, 30 μM NO substrate (6-OMeGal-Ph-NO) was added to the medium in advance. Then, H2O2 with different concentrations (100, 200, 400, and 600 μM) was added to stimulate the lentivirus-infected eMSCs. BLI was performed immediately after addition of the luciferase substrate coelenterin to eval‎uate cell apoptosis. Additionally, eMSCs treated with 200 mM H2O2 were stimulated for 24 hours to induce cell apoptosis. An Annexin V/PI assay kit (Solarbio) was used to detect eMSC apoptosis.

            BLI detection of cell retention

            BLI and luciferase substrates were used in mice to eval‎uate the retention of NO-eMSCsGluc/RFP in cardiac orthotopic transplantation. The mice after eMSC injections were anesthetized with 1.5% isoflurane and injected with coelenterin through the caudal vein at 150 mg/kg. After injection, the mice were immediately placed in a BLI system to detect cell retention in the myocardium.

            Animals

            C57BL/6 mice (male, 8 weeks old) and Sprague-Dawley rats (male, 8 weeks old) were purchased from SPF (Beijing) Biotechnology Co. Ltd., Beijing, China. Animals were randomly grouped for treated and untreated controls. All experiments and animal procedures were approved by the Animal Experiments Ethical Committee of Nankai University and carried out in conformity with the Guide for Care and Use of Laboratory Animals.

            MI in mice and rats

            Surgical induction of MI was performed on C57BL/6 mice (male, 8 weeks old) as previously described with some modifications. Briefly, mice were anesthetized with 2% isoflurane, followed by fixation to a heating pad (37°C) at supine position, and then ventilated with a mechanical ventilation system (Hallowell EMC Microvent I, USA) set at 110 breaths per minute with a tidal volume of 0.2 to 0.3 ml. The third intercostal space over the left chest and the heart was exposed. After left thoracotomy, the left anterior descending coronary artery was ligated with a 6-0 silk ligature. The left main descending coronary artery was sutured and tied with a slip knot at a site approximately 3 mm from its origin. Cardiac ischemia was confirmed by the presence of myocardial blanching. Immediately, eMSCs encapsulated with HA hydrogel were injected into the myocardium of mice through three-point injection around the infarct zone. The chest cavity was closed to restore negative pressure and prevent pneumothorax.

            Similar MI surgery was performed on Sprague-Dawley rats (male, 8 weeks old) first. Briefly, rats were anesthetized via intraperitoneal injection of 10% chloral hydrate (350 mg/kg), followed by fixation to a heating pad (37°C) at supine position. Then, they were ventilated with a mechanical ventilation system (Hallowell EMC Microvent I, USA) set at 110 breaths per minute with a tidal volume of 6 ml. The third intercostal space over the left chest and the heart was exposed. After left thoracotomy, the left anterior descending coronary artery was ligated with a 6-0 silk ligature. The left main descending coronary artery was sutured and tied with a slip knot at a site approximately 3 mm from its origin. Cardiac ischemia was confirmed by the presence of myocardial blanching. The chest cavity was closed to restore negative pressure and prevent pneumothorax. Three days after surgery, secondary thoracotomy was performed, and eMSCs were injected into the myocardium around the infarct zone. The chest cavity was closed to restore negative pressure and prevent pneumothorax.

            Mice and rats in the AMI group only received MI without eMSC injection, while sham-operated mice only experienced thoracotomy without MI surgery or eMSC injection.

            At 1, 3, 5, and 7 days after myocardial injection of eMSCs, the prodrug was injected through the tail vein. Mice were injected with 100 μl of prodrug (1 mg/ml) each time, and rats were injected with 200 μl of prodrug (1 mg/ml) each time.

            TTC staining

            Two days after surgery, a thoracotomy was performed. The heart was quickly excised after quick freezing for 15 min and sliced at 1 mm thickness. Afterward, the sections were incubated with 1.5% TTC (Sigma-Aldrich) solution at 37°C in an incubator for 15 to 30 min and then with a 4% formaldehyde solution for 2 hours. The normal myocardial tissue was red, while the ischemic myocardium was white. The size of the infarcted myocardium, which was white or pale, was measured by ImageJ software.

            Cardiac function assessment

            Transthoracic echocardiography was performed with the Vevo 2100 Imaging System (Fuji Film Visual Sonics Inc., Canada) equipped with an MS-250/400 imaging transducer. The baseline cardiac function of mice and rats was measured at 3 days before surgery. Cardiac function was analyzed at days 1 and 28 after MI surgery with different treatments, as reported previously. Mice or rats were slightly anesthetized in a box with isoflurane. Their limbs were fixed in a supine position on the echo mat, and the chest hair was removed by depilating cream. Then, mice or rats were anesthetized by inhalation of isoflurane (0.5 to 1%) mixed with oxygen to maintain the heart rate at approximately 500 to 600, and M-mode echocardiography was performed. The left ventricular internal diameter at end-diastole (LVIDd) and systole (LVIDs) were obtained by measuring the long axis and the short axis. Accordingly, the cardiac parameters LV-EF, LV-FS, LV-EDV, and LV end-systole volume (LV-ESV) were determined. The echocardiography measurement was carried out in a double-blind manner.

            Histological analysis

            At the indicated time points, mice and rats were anesthetized via intraperitoneal injection of chloral hydrate, and a thoracotomy was performed. The hearts were fixed with trans-cardiac perfusion of saline and immersed in 4% paraformaldehyde over 24 hours. The heart tissue samples were dehydrated with gradient alcohol and xylene, embedded in paraffin blocks, and cut into sections in 5 μm thickness.

            The paraffin-embedded sections were stained with Masson trichrome, H&E, and Sirius Red following a standard protocol. Immunofluorescence staining was performed on paraffin-embedded sections of the heart tissue samples. After deparaffinization and heat-mediated antigen retrieval‎ in citrate solution, the samples were washed with PBS three times and incubated with blocking serum, which was used to avoid nonspecific binding, at room temperature for 30 min. The sections were incubated with specific antibodies diluted in goat serum at 4°C overnight. On the second day, the sections were rewarmed at room temperature for 1 hour and washed with PBS three times. Afterward, the sections were incubated with Alexa Fluor–coupled secondary antibodies for 2 hours at room temperature. After washing with PBS, the sections were counterstained with DAPI-containing Fluoromount-G (SouthernBiotech, USA) and coverslipped. The antibodies used included anti–α-SMA (1:100, ab5694, Abcam), anti-vWF (1:100, ab6694, Abcam), anti–α-actinin (1:100, ab9475, Abcam), anti-Connexin43 (1:100, ab11370, Abcam), anti-CD68 (1:100, ab125212, Abcam), WGA (1:500, FL-1021, Novus Biologicals), anti-iNOS (1:100, ab178945, Abcam), and anti-CD206 (1:100, ab64693, Abcam).

            Macrophage isolation and detection

            Three days before euthanasia, mice were intraperitoneally injected with 2 ml of 4% thioglycolate. Three days later, the mice were sacrificed by cervical dislocation and immersed in 75% alcohol and then transferred to an ultraclean workbench. The mouse limb was fixed in the supine position, and the mouse abdominal wall was carefully cut open with the peritoneal. PBS [1% penicillin-streptomycin (PS)] was injected intraperitoneally to collect the cell suspension, which was centrifuged at 2000 rpm for 10 min. After discarding the supernatant, the cells were incubated with anti-F4/80/TNF-α and anti-F4/80/CD206 antibodies. FlowJo software was used to analyze the results of flow cytometry.

            Quantitative real-time PCR

            Total RNA samples from the cells were prepared using TRIeasy Total RNA Extraction Reagent (Yeasen, China) according to the manufacturer’s instructions. Heart tissue samples were collected at the indicated time points after MI surgery.

            The tissue samples were dissected at the border zone of the left ventricle and frozen in liquid nitrogen immediately. Afterward, the total RNA was extracted with TRIzol reagent, as mentioned before. The concentration of the RNA was measured with a NanoDrop spectrophotometer (NanoDrop Technologies, USA). The complementary cDNA was synthesized using a first-strand cDNA synthesis kit (Yeasen, China). Quantitative real-time PCR was performed on a CFX96 real-time PCR system (Bio-Rad, Hercules, USA) with an SYBR Green–based real-time detection system (Yeasen, China). The relative gene expression‎ of mRNA was expressed as 2−(△△CT) and normalized to GAPDH as an internal control. Each reaction was performed in triplicate to obtain an average value, and the changes in relative gene expression‎ normalized to the internal control levels were determined. The highly purified primers used in this experiment were commercially synthesized (Sango, China). The sequences of the primers used in this experiment are summarized in the Supplementary Materials.

            Statistics

            All data are presented as the mean ± SEM from at least three independent experiments. Comparisons between two groups were performed by Student’s t test, and comparisons among more than two groups were performed by one-way or two-way analysis of variance (ANOVA). Statistical analyses were performed with GraphPad Prism software 7.0, and a statistical significance level of less than 0.05 was accepted.

            Acknowledgments

            Funding: This study is supported by the National Key R&D Program of China (2018YFE0200503), the National Natural Science Foundation of China (nos. 81925021, 82330066, 81921004, and U2004126), and the Tianjin Natural Science Foundation (21JCZDJC00240).

            Author contributions: Q.Z. and Z.L. conceived the original concept and initiated this project. Q.Z., Z.L., and F.G. designed the experiment and supervised the entire project. S.W. collected human adipose mesenchymal stem cells. M.Q. synthesized all NO prodrugs and probes. P.L. prepared engineered enzymes. T.H., G.J., and Q.X.L. established mouse and rat MI models. T.H. and G.J. performed histological analysis. G.J. and W.D. carried out in vitro cell experiments. S.D. carried out NO cell imaging under the supervision of L.P. T.H., G.J., and M.Q. analyzed data under the supervision of Q.Z. H.H. helped with lentivirus packaging and cell infection. W.G. and T.L. helped in establishing animal MI models. Y.W., J.H., J.C., and J.T. helped with data collection. T.H. and Q.Z. wrote the paper with input from other authors.

            Competing interests: The authors declare that they have no competing interests.

            Data and materials availability: All data needed to eval‎uate the conclusions in the paper are present in the paper and/or the Supplementary Materials. Transcriptome sequencing dataset is available at https://doi.org/10.5061/dryad.tqjq2bw5b.

            Supplementary MaterialsThis PDF file includes:

            Supplementary Text

            Figs. S1 to S9

            Table S1

            • Download
            • 1.94 MB

            REFERENCES AND NOTES

            1

            M. F. Pittenger, A. M. Mackay, S. C. Beck, R. K. Jaiswal, R. Douglas, J. D. Mosca, M. A. Moorman, D. W. Simonetti, S. Craig, D. R. Marshak, Multilineage potential of adult human mesenchymal stem cells. Science 284, 143–147 (1999).

            GO TO REFERENCE

            Crossref

            PubMed

            Web of Science

            Google Scholar

            2

            A. Uccelli, L. Moretta, V. Pistoia, Mesenchymal stem cells in health and disease. Nat. Rev. Immunol. 8, 726–736 (2008).

            GO TO REFERENCE

            Crossref

            PubMed

            Web of Science

            Google Scholar

            3

            Y. Wang, J. Fang, B. Liu, C. Shao, Y. Shi, Reciprocal regulation of mesenchymal stem cells and immune responses. Cell Stem Cell 29, 1515–1530 (2022).

            GO TO REFERENCE

            Crossref

            PubMed

            Web of Science

            Google Scholar

            4

            Y. Wang, X. Chen, W. Cao, Y. Shi, Plasticity of mesenchymal stem cells in immunomodulation: Pathological and therapeutic implications. Nat. Immunol. 15, 1009–1016 (2014).

            GO TO REFERENCE

            Crossref

            PubMed

            Web of Science

            Google Scholar

            Show all references

            (0)eLetters

            eLetters is a forum for ongoing peer review. eLetters are not edited, proofread, or indexed, but they are screened. eLetters should provide substantive and scholarly commentary on the article. Neither embedded figures nor equations with special characters can be submitted, and we discourage the use of figures and equations within eLetters in general. If a figure or equation is essential, please include within the text of the eLetter a link to the figure, equation, or full text with special characters at a public repository with versioning, such as Zenodo. Please read our Terms of Service before submitting an eLetter.

            Log In to Submit a Response

            No eLetters have been published for this article yet.

            Recommended articles from TrendMD

            • In vivo priming of human mesenchymal stem cells with hepatocyte growth factor–engineered mesenchymal stem cells promotes therapeutic potential for cardiac repai...

              Bong-Woo Park, Sci Adv, 2020

            • Shattering barriers toward clinically meaningful MSC therapies

              Oren Levy, Sci Adv, 2020

            • Nanovesicles derived from iron oxide nanoparticles–incorporated mesenchymal stem cells for cardiac repair

              Ju-Ro Lee, Sci Adv, 2020

            • Exendin-4 gene modification and microscaffold encapsulation promote self-persistence and antidiabetic activity of MSCs

              Yuanyuan Zhang, Sci Adv, 2021

            • Human gingival mesenchymal stem cells retain their growth and immunomodulatory characteristics independent of donor age

              Jay R. Dave, Sci Adv, 2022

            • The secretion profile of mesenchymal stem cells and potential applications in treating human diseases

              Yuyi Han, Signal Transduction and Targeted Therapy, 2022

            • Mesenchymal Stem Cell Therapy for Ischemic Heart Disease: Systematic Review and Meta-analysis

              Hyunsuk Jeong, International Journal of Stem Cells

            • Mesenchymal Stem Cell and MicroRNA Therapy of Musculoskeletal Diseases

              International Journal of Stem Cells

            • Excessive branched-chain amino acid accumulation restricts mesenchymal stem cell-based therapy efficacy in myocardial infarction

              Fuyang Zhang, Signal Transduction and Targeted Therapy, 2022