말초 신경의 탐구!!

[문형철]

• Research
• Open access
• Published: 13 May 2021

Identification of sensory and motor nerve fascicles by immunofluorescence staining after peripheral nerve injury

• Xijie Zhou, 
• Jian Du, 
• Liming Qing, 
• Thomas Mee, 
• Xiang Xu, 
• Zhuoran Wang, 
• Cynthia Xu & 
• Xiaofeng Jia 

Journal of Translational Medicine volume 19, Article number: 207 (2021) Cite this article

• 7066 Accesses

• 3 Altmetric

• Metricsdetails

Abstract

Background

Inappropriate matching of motor and sensory fibers after nerve repair or nerve grafting can lead to failure of nerve recovery. Identification of motor and sensory fibers is important for the development of new approaches that facilitate neural regeneration and the next generation of nerve signal-controlled neuro-prosthetic limbs with sensory feedback technology. Only a few methods have been reported to differentiate sensory and motor nerve fascicles, and the reliability of these techniques is unknown. Immunofluorescence staining is one of the most commonly used methods to distinguish sensory and motor nerve fibers, however, its accuracy remains unknown.

Methods

In this study, we aim to determine the efficacy of popular immunofluorescence markers for motor and sensory nerve fibers. We harvested the facial (primarily motor fascicles) and sural (primarily sensory fascicles) nerves in rats, and examined the immunofluorescent staining expressions of motor markers (choline acetyltransferase (ChAT), tyrosine kinase (TrkA)), and sensory markers [neurofilament protein 200 kDa (NF-200), calcitonin gene-related peptide (CGRP) and Transient receptor potential vanillic acid subtype 1 (TRPV1)]. Three methods, including the average area percentage, the mean gray value, and the axon count, were used to quantify the positive expression‎ of nerve markers in the immunofluorescence images.

Results

Our results suggest the mean gray value method is the most reliable method. The mean gray value of immunofluorescence in ChAT (63.0 ± 0.76%) and TRKA (47.6 ± 0.43%) on the motor fascicles was significantly higher than that on the sensory fascicles (ChAT: 49.2 ± 0.72%, P < 0.001; and TRKA: 29.1 ± 0.85%, P < 0.001). Additionally, the mean gray values of TRPV1 (51.5 ± 0.83%), NF-200 (61.5 ± 0.62%) and CGRP (37.7 ± 1.22%) on the motor fascicles were significantly lower than that on the sensory fascicles respectively (71.9 ± 2.32%, 69.3 ± 0.46%, and 54.3 ± 1.04%) (P < 0.001). The most accurate cutpoint occurred using CHAT/CRCP ratio, where a value of 0.855 had 100% sensitivity and 100% specificity to identify motor and sensory nerve with an area under the ROC curve of 1.000 (P < 0.001).

Conclusions

A combination of ChAT and CGRP is suggested to distinguish motor and sensory nerve fibers.

초록

배경
신경 재건 또는 신경 이식 후 운동 신경 섬유와 감각 신경 섬유의 부적절한 매칭은 

신경 회복 실패로 이어질 수 있습니다. 

운동 신경 섬유와 감각 신경 섬유의 식별은 

신경 재생 촉진 및 감각 피드백 기술을 갖춘 차세대 신경 신호 제어형 신경 보철 사지 개발을 

위한 새로운 접근 방식 개발에 중요합니다. 

감각과 운동 신경 속을 구분하는 방법은 

아직 몇 가지밖에 보고되지 않았으며, 

이러한 기술의 신뢰성은 알려지지 않았습니다. 

면역 형광 염색은 

감각과 운동 신경 섬유를 구분하는 가장 널리 사용되는 방법 중 하나이지만, 

그 정확성은 여전히 불확실합니다.

방법
본 연구에서는 운동 및 감각 신경 섬유에 대한 인기 있는 면역 형광 표지자의 효능을 평가하는 것을 목표로 합니다. 쥐의 얼굴 신경(주로 운동 신경 속)과 발목 신경(주로 감각 신경 속)을 채취하고, 운동 마커(콜린 아세틸트랜스퍼레이스 (ChAT), 티로신 키나제 (TrkA))와 감각 마커 [신경 필라멘트 단백질 200 kDa (NF-200), 칼시토닌 유전자 관련 펩티드(CGRP) 및 일시적 수용체 잠재력 반일산 아세트산 서브타입 1(TRPV1)]의 면역형광 염색 표현을 조사했습니다. 면역형광 이미지에서 신경 마커의 양성 표현을 정량화하기 위해 평균 면적 비율, 평균 회색 값, 축삭 수 세 가지 방법을 사용했습니다.

결과
우리의 결과는 평균 회색 값 방법이 가장 신뢰할 수 있는 방법임을 시사합니다. 운동 신경 속의 ChAT (63.0 ± 0.76%) 및 TRKA (47.6 ± 0.43%)의 면역형광 평균 회색 값은 감각 신경 속의 값보다 유의미하게 높았습니다 (ChAT: 49.2 ± 0.72%, P < 0.001; TRKA: 29.1 ± 0.85%, P < 0.001). 또한 운동 신경 속의 TRPV1 (51.5 ± 0.83%), NF-200 (61.5 ± 0.62%) 및 CGRP (37.7 ± 1.22%)의 평균 회색 값은 감각 신경 속의 값보다 각각 유의미하게 낮았습니다 (71.9 ± 2.32%, 69.3 ± 0.46%, 및 54. 3 ± 1.04%)보다 유의미하게 낮았습니다 (P < 0.001). 가장 정확한 구분점은 CHAT/CRCP 비율을 사용했을 때 발생했으며, 0.855의 값은 운동 신경과 감각 신경을 구분하는 데 100%의 민감도와 100%의 특이도를 보였으며, ROC 곡선 아래 면적은 1.000이었습니다 (P < 0.001).

결론
ChAT와 CGRP의 조합이 운동 및 감각 신경 섬유를 구분하는 데 권장됩니다.

Background

Peripheral nerve injury can lead to the loss of motor, sensory and autonomic nerve function in the body's ganglion segment, which seriously affects patients’ quality of life [1]. Mismatched nerve fascicle repair after peripheral nerve injury may result in partial or complete loss of function [2]. Despite several decades of progress in research and surgical techniques, surgeons still rely on experience to estimate the characteristics of damaged motor or sensory nerve stumps and perform a differentiated fascicular repair. Thus, a satisfactory recovery is often difficult to achieve [3]. We have shown a 3-dimensional, printed scaffold repair technology promoting neural regeneration with bifurcating sensory and motor pathways after complex peripheral nerve injuries [4]. Therefore, the ability to identify and differentiate motor and sensory fascicles is greatly beneficial to the development of new approaches to facilitate neural regeneration. In addition, the ability to identify motor and sensory fascicles is crucial to the development of next generation nerve signal-controlled neuro-prosthetic limbs with sensory feedback technology, which is connected to residual peripheral nerves through the neural interface via intrafascicular electrodes as we reported [5,6,7]. It’s particularly important to reliably distinguish motor and sensory nerve fascicles to properly transmit signals via microelectrode as motor order or sensory feedback via the regenerated nerves [7].

Currently, there are four reported methods to distinguish sensory and motor nerve fascicles: anatomical, electrophysiological, infrared spectrum, and enzymohistochemical staining [8,9,10,11]. The anatomical method has been widely used in the past decades, but surgeons can only rely on their own experience to estimate the type of nerve fascicle, and may cause partial or total loss of nerve function after nerve transplantation [12]. Therefore, the anatomical technique alone is difficult to effectively gauge the type of fascicle. Electrophysiological methods such as evoked potentials were used to distinguish sensory from motor nerve fibers. While we have extensive experience with the electrophysiological study of nerve injury [5, 6, 13, 14], this method can only distinguish the main nerve branch; it is unable to show the composition of the nerve fiber. Moreover, there is a need for larger sample sizes to study the electrophysical technique due to the high variance of measurements seen in current studies [15]. Infrared spectrum identification requires a variety of equipment, complex calculations, and many interference factors, which limits its application to a great extent [10]. Enzymohistochemistry staining, including immunofluorescence staining and immunohistochemical staining, is currently one of the most frequently used methods. However, it is difficult to distinguish the results using immunohistochemistry after labeling with a multi substrate color system. In addition, the color of the substrate and the thickness of the slice will affect the final result, and the chromogenic substrate is an enzymatic reaction that easily saturates the substrate, thus limiting the semi quantitative analysis [16]. Compared with immunohistochemistry, immunofluorescence can carry out a reaction with multiple markers. Antibody-coupled fluorescence can also increase the resolution [16].

Although immunofluorescence is widely used, the efficacy/accuracy of popular immunofluorescence markers for motor and sensory nerve fibers remains unclear. In addition, the results of quantitative processing methods for fluorescent images are different without a proper comparison. Although almost all peripheral nerves are a mixture of motor and sensory nerves, the facial nerve is mainly composed of motor nerve fibers and the sural nerve is mainly composed of sensory nerve fibers [17, 18].

We hypothesize that preferable markers will be eval‎uated and/or an optimal combination will be selected to identify and differentiate the motor and sensory axons using immunofluorescence staining in this side by side comparison in the facial and sural nerve. In this study, we selected the relatively pure facial and sural nerves as either the motor or sensory fascicles for staining, and examined the expressions of available and commonly used motor markers [choline acetyltransferase (ChAT), tyrosine kinase receptor A (TrkA)] and sensory markers [neurofilament protein 200 kDa (NF-200), calcitonin gene-related peptide (CGRP) and Transient receptor potential vanillic acid subtype1 (TRPV1)] using immunofluorescence staining aiming to eval‎uate the best differentiative approach. These markers have been widely used to label motor or sensory nerve fascicles experimentally [19, 20], however their quantified preferential staining in peripheral nerves after peripheral nerve injury have not been studied.

배경

말초 신경 손상은

신체의 신경절 부위에서 운동, 감각 및 자율 신경 기능의 상실을 초래하며,

이는 환자의 삶의 질에 심각한 영향을 미칩니다 [1].

말초 신경 손상 후 신경 속의 신경 속 묶음(fascicle)을 부적절하게 수복하면

부분적 또는 완전한 기능 상실이 발생할 수 있습니다 [2].

수십 년간의 연구와 수술 기술의 발전에도 불구하고,

외과의사들은 손상된 운동 또는 감각 신경의 잔여 부위 특성을 추정하고

차별화된 신경 속 묶음 수복을 수행하기 위해 여전히 경험에 의존하고 있습니다.

따라서

만족스러운 회복을 달성하는 것은 종종 어렵습니다 [3].

우리는 복잡한 말초 신경 손상 후 분기된 감각 및 운동 경로를 통해

신경 재생 촉진 효과를 보이는 3차원 인쇄 스캐폴드 복원 기술을 제시했습니다 [4].

따라서

운동 및 감각 신경 속을 식별하고 구분하는 능력은

신경 재생 촉진을 위한 새로운 접근법 개발에 크게 기여합니다.

또한,

운동 및 감각 신경 속을 식별하는 능력은 신경 인터페이스를 통해

잔여 말초 신경과 연결된 감각 피드백 기술을 갖춘 차세대 신경 신호 제어 신경 의수 개발에 필수적입니다.

이는 우리가 보고한 바와 같이 [5,6,7]

신경 속 내 전극을 통해 신호를 전달하는 방식입니다.

특히,

재생된 신경 통해 운동 명령이나 감각 피드백으로 신호를 정확히 전달하기 위해

운동 및 감각 신경 속을 신뢰성 있게 구분하는 것이 중요합니다 [7].

현재 감각 및 운동 신경 속을 구분하는 네 가지 방법이 보고되었습니다:

해부학적, 전기생리학적, 적외선 스펙트럼, 및 효소조직화학 염색법 [8,9,10,11]. 해부학적 방법은

지난 수십 년간 널리 사용되었지만,

외과의사는 자신의 경험에만 의존하여 신경 속의 유형을 추정해야 하며,

신경 이식 후 신경 기능의 부분적 또는 완전한 손실을 초래할 수 있습니다 [12].

따라서

해부학적 기술만으로는 신경 속의 유형을 효과적으로 판단하기 어렵습니다.

감각 신경과 운동 신경 섬유를 구분하기 위해

유발 전위와 같은 전기생리학적 방법이 사용되었습니다.

우리는

신경 손상의 전기생리학적 연구에 풍부한 경험을 보유하고 있습니다[5, 6, 13, 14],

하지만 이 방법은

주요 신경 분지를 구분할 수 있을 뿐,

신경 섬유의 구성은 보여주지 못합니다.

또한 현재 연구에서 측정 결과의 높은 변동성[15]으로 인해

전기생리학적 기술 연구를 위해 더 큰 표본 크기가 필요합니다.

적외선 스펙트럼 식별은

다양한 장비, 복잡한 계산, 많은 간섭 요인이 필요하여 적용 범위가 크게 제한됩니다[10].

면역형광 염색과 면역조직화학 염색을 포함한 효소조직화학 염색은

현재 가장 자주 사용되는 방법 중 하나입니다.

그러나

다중 기질 색소 시스템을 사용한 면역조직화학 염색 후 결과를 구분하기 어렵습니다.

또한 기질의 색상과 슬라이스 두께가 최종 결과에 영향을 미치며,

발색 기질은 기질을 쉽게 포화시키는 효소 반응이기 때문에

반정량적 분석을 제한합니다 [16].

면역조직화학에 비해 면역형광은

다중 마커와의 반응을 수행할 수 있습니다.

항체 결합 형광은 해상도를 높일 수 있습니다 [16].

면역형광법은 널리 사용되지만,

운동 및 감각 신경 섬유에 대한 인기 있는

면역형광 표지자의 효능/정확도는 여전히 명확하지 않습니다.

또한, 형광 이미지 정량 처리 방법의 결과는 적절한 비교 없이 서로 다를 수 있습니다.

주변 신경의 대부분은

운동 신경과 감각 신경의 혼합물이지만,

안면 신경은 주로 운동 신경 섬유로 구성되어 있으며,

sural nerve은 주로 감각 신경 섬유로 구성되어 있습니다 [17, 18].

Although almost all peripheral nerves are a mixture of motor and sensory nerves,

the facial nerve is mainly composed of motor nerve fibers and

the sural nerve is mainly composed of sensory nerve fibers [17, 18].

본 연구에서는

안면 신경과 Sural 신경에서 면역형광 염색을 통해

운동 및 감각 축삭을 식별하고 구분하기 위해 선호되는 표지자를 평가하고/또는

최적의 조합을 선택하는 것을 목표로 합니다.

본 연구에서는 염색을 위해

상대적으로 순수한 얼굴 신경과 수근 신경을 운동 또는 감각 신경 속으로 선택했으며,

이용 가능하고 일반적으로 사용되는

운동 마커 [콜린 아세틸트랜스퍼레이스 (ChAT),

티로신 키나제 수용체 A (TrkA)] 감각 마커 [신경필라멘트 단백질 200 kDa (NF-200),

칼시토닌 유전자 관련 펩티드 (CGRP) 및 일시적 수용체 잠재력 반일산 아미노산 서브타입 1 (TRPV1)]를

면역형광 염색을 통해 평가하여 가장 우수한 구분 방법을 평가했습니다.

이 마커들은 실험적으로

운동 또는 감각 신경 속을 표시하는 데 널리 사용되어 왔습니다 [19, 20],

그러나

말초 신경 손상 후 말초 신경에서의 정량화된 선호적 염색은 연구되지 않았습니다.

Materials and methods

Experimental procedures

From 5 fresh euthanized nude rats, weighing 200 g to 250 g, we obtained healthy facial nerve and sural nerve specimens. To expose the facial nerve trunk and its branches, a 1 cm skin incision on a horizontal plane was performed from the inferior margin of the auricular, extending anteriorly. Facial nerve specimens were obtained by transecting at the stylomastoid foramen and 5 mm distal to the skin incision, thus extracting a 10 mm nerve segment. A 1-cm long sural nerve segment was dissected from the sciatic nerve. We obtained both sides (10 facial nerve specimens and 10 sural nerve specimens) in all rats, and randomly selected one side of each nerve type for the following studies. Differentially expressed proteins could be observed in proximal and distal nerve segments due to different cells and extracellular matrix of proximal and distal nerve segments [21]. In our study, all the nerve sections were collected from the similar position of nerves. Experimental protocols were approved by the IACUC of University of Maryland School of Medicine Animal Care and Use Committee.

The nerve specimens were immersion-fixed in the 4% Paraformaldehyde (PFA) fixative for 24 h and then placed into 30% sucrose 0.1-M phosphate-buffered saline (PBS) for at least 48 h. The sural and facial nerves were frozen, and then 10 µm thick transverse serial sections were obtained using a freezing microtome (Leica, Germany) at − 20 °C. All sections were stored at − 20 °C. Three slides were randomly selected from each specimen for quantitative immunofluorescence staining in five samples total.

Immunofluorescence analysis

Standard Immunofluorescence procedures were followed [22]. The facial nerve and sural nerve sections were incubated with the primary antibody against ChAT (rabbit, diluted 1: 100; Millipore), NF-200 (rabbit, diluted 1: 200; Sigma), CGRP (mouse, diluted 1: 200; Abcam), TRKA (rabbit, diluted 1: 200; Abcam), TRPV1 (mouse, diluted 1: 200; Abcam) followed by 24 h at 4 °C. Following incubation with the primary antibody, the specimens were then washed again three times (5 min/wash) in PBS and incubated in the secondary antiserum solution. Antigens of facial nerve sections were observed by using Invitrogen Alexa Fluor-594 donkey anti-rabbit secondary antibody, Invitrogen Alexa Fluor 488-conjugated goat anti-rabbit and Abcam Alexa Fluor 488-conjugated goat anti-mouse (diluted 1: 500; USA) 2 h at 37 °C. Autofluorescence (negative control) that incubated with secondary antibody only were used during each staining.

After being washed, all sections were covered and mounted with ProLong™ Gold Antifade Mountant with DAPI (4′,6-diamino-2-phenylindole, blue, Invitrogen, USA). Micrographs were obtained by using a Leica DMi8 microscope camera equipment (Leica Microsystems). All fluorescent images were analyzed in Image J (National Institutes of Health, USA), as we previously described [23, 24]. Each data point is from five experimental animals. Three tissue sections from selected nerves were collected from each animal in a blinded fashion. Quantification was conducted by counting the axons with immunoreactivity in 3 randomized microscopic fields in each section. For the immunofluorescence analysis, area percentage and mean gray value were analyzed using Image J software (v1.8.0, NIH, USA). After converting the image to black and white with 8-bit type, the threshold of the image was adjusted to best cover the axon area (Image-Adjust-Threshold-Apply); then the area of the axon was measured and recorded (Analyze-Measure) after randomly selecting three fixed areas (50 μm * 50 μm) on each image. The area percentage, mean gray value, and axon count were automatically calculated using image J.

Statistics

An independent t-test was employed to determine statistical differences of quantified immunofluorescence area and mean gray value between two groups using SPSS version 22.0 (IBM Corp., Armonk, NY, USA). Image J software was used to determine the immunofluorescence area and mean gray value (Version 1.8.0). A receiver operating characteristic (ROC) curve was constructed to determine the cut-points of the mean gray value of ChAT and CGRP and the ChAT/CGRP ratio, with optimal sensitivity and specificity to identify motor and sensory fascicles. All values were expressed as a mean ± SEM. Statistical significance was set at P < 0.05.

Results

The expressions of motor markers in motor and sensory fascicles

ChAT is an enzyme synthesized within motor axons, and ChAT immunofluorescence is dominantly expressed in motor fascicles [25]. Our study showed that the area percentage of ChAT-labeled axons (red staining) in the facial nerve (7.9 ± 0.36%, Fig. 1a) was significantly higher than the sural nerve (6.3 ± 0.62%, Fig. 1b, P < 0.001). The mean gray value and axon count of ChAT was also significantly higher in the facial nerve (63.0 ± 0.76%, 20.47 ± 0.4%), compared to the sural nerve (49.2 ± 0.72%, 10.47 ± 0.4, P < 0.001) (Fig. 1c, d).

Fig. 1

ChAT immunofluorescence staining of the facial nerve and sural nerve. a representative image of staining with ChAT (red) and DAPI (blue). Scale bar is 25 μm. Quantification of b area percentage; c mean gray value; and d axon count of positive ChAT staining

Full size image

TrkA was reported to detect the function of extraocular motoneurons and spinal motor neurons after axotomy or in neurodegenerative diseases [26]. Our results showed that TrkA is mainly expressed on the cell membrane and strongly expressed in the facial nerve (Fig. 2a). The average immunofluorescence area was 21.6 ± 0.84%, significantly higher than that in sural nerve (7.6 ± 0.72%, P < 0.001) (Fig. 2b). The results were also confirmed with mean gray values of TrkA (47.6 ± 0.43% vs. 29.1 ± 0.85%, P < 0.001, Fig. 2c) in the facial nerve and sural nerve.

Fig. 2

TRKA immunofluorescence staining of the facial nerve and sural nerve. a representative image of staining with TRKA (green) and DAPI (blue). Scale bar is 25 μm. Quantification of b area percentage and c mean gray value

Full size image

The expressions of sensory markers in motor and sensory fascicles

CGRP is mainly expressed in primary afferent neurons and plays an important role in the repair of axonal injury [27]. From Fig. 3a we observed more positive expression‎ of CGRP in the sural nerve than in the facial nerve. The area percentage was nearly 2 times higher when comparing the fluorescence staining in the sural nerve with the facial nerve (26.7 ± 2.68% vs. 14.2 ± 2.85%, P < 0.001, Fig. 3b). The mean gray value also indicated higher CGRP expression‎ in the sural nerve (54.3 ± 1.04%) than in facial nerve (37.7 ± 1.22%, P < 0.001, Fig. 3c). The axon count was significantly higher in sural nerve compared with that of the facial nerve (21.53 ± 0.6 vs. 12.13 ± 0.4, P < 0.001, Fig. 3d).

Fig. 3

CGRP immunofluorescence staining of the facial nerve and sural nerve. a representative image of staining with CGRP (green) and DAPI (blue). Scale bar is 25 μm. Quantification of b area percentage; c mean gray value; and d axon count of positive CGRP staining

Full size image

TRPV1 is sensitized in the process of inflammation and injury and expressed in peptidergic and nonpeptidergic nociceptors [28, 29]. We found positive expression‎ of TRPV1in both the sural and the facial nerve fibers (Fig. 4a). There was no significant difference in the average area percentage between the two groups (11.0 ± 1.05% vs. 8.8 ± 1.26%, P > 0.05, Fig. 4b). Both TRPV1 mean gray values (51.5 ± 0.83% and 71.9 ± 2.32%, P < 0.001, Fig. 4c) and axon count (11.27 ± 0.6 and 18.73 ± 0.6 for the facial nerve and sural nerve, P < 0.001, Fig. 4d) were significantly higher in the sural nerve than in the facial nerve fibers.

Fig. 4

TRPV1 immunofluorescence staining of the facial nerve and sural nerve. a representative image of staining with TRPV1 (green) and DAPI (blue). Scale bar is 25 μm. Quantification of b area percentage; c mean gray value; and d axon count of positive TRPV1 staining

Full size image

NF-200 is an axon-specific intermediate filament found in peripheral nerves and is used as a marker for sensory myelinated fibers in previous studies [30, 31]. Our results showed NF-200 was positive in both sural and facial nerve fibers (Fig. 5a). The area percentage results showed that NF-200 had a significantly higher positive rate in sural nerve fibers (18.9 ± 1.08%) compared to facial nerve fibers. NF 200 was expressed in facial nerve fibers with the rate of 10.1 ± 0.90% (Fig. 5b). Furthermore, the mean gray value analysis showed the rate is 69.3 ± 0.46% in the sural nerve and 61.5 ± 0.62% in the facial nerve (Fig. 5c). Similarly, a higher stained axon count was observed in the sural nerve (30.20 ± 0.7) than that observed in the facial nerve (21.53 ± 0.5, P < 0.001) (Fig. 5d).

Fig. 5

NF-200 immunofluorescence staining of the facial nerve and sural nerve. a representative image of staining with NF-200 (green) and DAPI (blue). Scale bar is 25 μm. Quantification of b area percentage; c mean gray value; and d axon count of positive NF-200 staining

Full size image

The optimal combination of target markers to differentiate motor and sensory fascicles

Based on the above results, we selected ChAT and CGRP as the optimal combination of target markers to distinguish motor and sensory nerve fibers and performed double immunofluorescence staining (Fig. 6a). The quantified mean gray value based on the positive ChAT in facial nerve was 57.0 ± 1.48%, which was significantly higher than CGRP in facial nerve 52.4 ± 0.76% (P < 0.001) (Fig. 6b). Furthermore, the mean gray value of CGRP in the sural nerve (43.1 ± 0.77%) was significantly higher than that of ChAT in the sural nerve (58.6 ± 0.65%, P < 0.001) (Fig. 6c). There is a significant difference (P < 0.001) by the mean gray value comparison between ChAT and CGRP, which showed the ratio of 1.1 ± 0.03% and 0.7 ± 0.02% for the facial nerve and sural nerve, respectively (Fig. 6d).

Fig. 6

Double immunofluorescence staining results of ChAT and CGRP in facial and sural nerve. a representative image of staining with ChAT (red), CGRP (green), and DAPI (blue). Scale bar is 25 μm. Quantification of b area percentage; c mean gray value; and d axon count of positive ChAT/CGRP staining. Mean gray value cut-points of e CHAT to identify motor nerve fascicles and f CGRP to identify sensory nerve fascicles were determined using ROC curve methodology with sensitivity and specificity. g The most accurate cut-point occurred using CHAT/CRCP: a cutpoint of 0.855 yielded 100% sensitivity and 100% specificity identifying motor and sensory nerve fascicles with an area under the ROC curve of 1.000 (P < 0.001)

Full size image

Accurate cut-points of the mean gray values were determined using ROC curve methodology. A cut-point of 50.5 for CHAT had 86.7% sensitivity and 100% specificity identifying motor nerve fascicles with an area under the ROC curve of 0.982 (P < 0.001) (Fig. 6e), and a cut-point of 54.5 for CGRP had 100% sensitivity and 80% specificity identifying sensory nerve fascicles with an area under the ROC curve of 0.940 (P < 0.001) (Fig. 6f). The most accurate cut-point occurred using CHAT/CRCP ratio, where a value of 0.855 had 100% sensitivity and 100% specificity to identify motor and sensory nerve with an area under the ROC curve of 1.000 (P < 0.001) (Fig. 6g). All results indicated that ChAT was highly expressed in motor facial nerve fibers, while CGRP is more dominant in sensory sural nerve fibers, and a combination of ChAT and CGRP is an optimal combination to distinguish motor and sensory nerve fibers.

Discussion

Identification of motor and sensory fibers is important not only to develop new approaches to facilitate neural regeneration, but also crucial to develop the future generation of nerve signal-controlled neuro-prosthetic limbs with close-loop sensory feedback. The use of neuroprostheses to fully replace amputated limbs or recover their function after injury is still challenging [32, 33]. To achieve a natural control of the limb movement, the selective function of motor and sensory nerves and their fascicles should be properly selected for the neuroprosthesis [34]. Such selectivity is an unsolved challenge. Nerve mapping in amputees or in peripheral nerve injuries is not possible using intraoperative neuromonitoring techniques [35]. Thus, the surgeon must exclusively rely on the nerve topography for the implantation. Here our results showed the simple and useful immunofluorescent approach to distinguish motor and sensory nerves, by which micro-suture marked nerve stumps could be identified and prepared for interfacing at second stage. It is a simple tool to identify the motor and sensory nerves after trauma and provides useful information for the next step of the operation. Thus, it can be a complementary approach, which might help in the planning of the interfacing procedure and move the field one step closer to a comprehensive solution for the application of neuro-prosthetic limbs in a clinical setting.

We selected five commonly used biomarkers that have been used for identification of either motor or sensory axons and eval‎uated immunofluorescent staining in side-by-side comparisons, controlling for time and other conditions. Our study shows that nerve fibers can be distinguished effectively by using ChAT for motor nerve fibers and CGRP for sensory nerve fibers, and ChAT and CGRP were the optimal combination of markers for motor and sensory fascicles, respectively. We used three methods, including the average area percentage, the mean gray value, and the axon count, to quantify the positive expression‎ of nerve markers in the immunofluorescence images, and showed the mean gray value method is a more stable method. Since other, available solutions are scarce, this method will provide a convenient, fast, and reliable methodology to distinguish primarily motor fascicles from primarily sensory fascicles after peripheral nerve injury and will have a great translational value in the clinic. This technique enables a differentiated fascicle repair that will greatly improve the success rate of nerve repair and functional recovery after surgery. It will also be a very important addition for future neural prostheses, which makes neural motor signal control and sensory signal feedback technology feasible.

We have screened nearly all popular immunofluorescence markers in the literature and examined a panel of the five most effective ones including motor markers (ChAT and TRKA) and sensory markers (CGRP, TRPV1, and NF-200) in the sural nerve (primarily sensory nerve fibers) and the facial nerve (primarily motor nerve fibers). Although these markers have been suggested to be preferably used in either motor or sensory axons, the side-by-side comparison has never been assessed before. Since there is no available tool to identify the fascicles correctly as motor or sensory, the sensitivity and specificity of these markers in motor or sensory fascicles cannot be calculated. Thus, in our study we define it as efficacy of preferential staining, which refers to whether the antibody can show a positive result in the desired tissue, and how much the expression‎ of the positive result is the antibody. In this respect, all the antibodies we selected were expressed to varying degrees on both nerve fibers, but not specifically enough. With the strong positive antibodies that were expressed, the maximum positive area was about 30% when the average area percentage was used for data analysis. The axon count is up to 10–30% of the positively stained area in the whole selected microscopic fields, which is still high, as the space outside the axon contains a large amount of myelin sheath that cannot be shown in immunofluorescence without properly specialized antibodies. Using mean gray value, the positive rate reaches 60–80%, thus the mean gray value methodology has a higher efficacy of preferential staining and is a more stable method.

To select the best analysis methods to distinguish between the motor and sensory fibers of injured nerves, we eval‎uated the efficacy through different analysis methods including fluorescence area percentage, mean gray value, and axon count, which have never been systemically assessed in peripheral nerves yet. When using the average area percentage, the image J software randomly selects areas and calculates highly expressed axons, but ignores low grayscale and weakly expressed nerve fibers. The results show that the percentage obtained by the average area percentage is significantly lower than the mean gray value. Compared with the abovementioned methods, the axons with weak fluorescence can also be detected when the mean gray value is applied. Using axon counting is a generally accepted method, which can be popularly used in the slides of 100× under an optical microscope. However, it should be noted that when using a 40× magnified image, an axon with a smaller diameter may not be found. That reduces the resolution of some axons, making the accuracy of all axon counting unstable [36]. In addition, the method is based on the number of circular axons, which ignores compressed or oblique sub-areas in the nerve tissue sections that can easily lead to inaccurate axon count [37]. Therefore, we recommend using the mean gray value to quantitatively detect the immunofluorescence results, because of its high efficacy of preferential staining and relative stability.

ChAT, as a primary motor fascicles marker, is widely reported to be highly expressed in human facial motor neurons and rat facial motor neurons [38, 39], with high concentrations in motor neurons in both the central nervous system (CNS) and peripheral nerve axons [40, 41]. A key feature of motor neuron development and function is the expression‎ of the acetylcholine biosynthetic enzyme, ChAT [42, 43]. Yuan et al. used ChAT as an enzymatic marker in immunohistochemistry staining to detect motor axons and used NF200 to detect sensory axons and showed that the sciatic nerve contains both motor and sensory axons [44]. Castaneda and Wang et al. showed the same in their studies [45, 46]. However, these studies only show that ChAT and NF200 can be expressed positively in motor nerve fibers and sensory nerve fibers, without quantitative elaboration on the accuracy and efficacy of preferential staining of this method. The other marker, TrKA is mainly expressed on the cell membrane and strongly in the facial nerve in our study. Carrizosa et al. chose TrkA positive expression‎ to detect the function of spinal motor neurons and extraocular motoneurons [26]. In the CNS, within the striatal motor neurons and the septal/diagonal band complex, TrkA neurons (> 99 and > 95%, respectively) co-expressed ChAT [47]. Recently, Han et. al showed that TrkA was expressed in primary sensory fibers mainly concentrated in Dorsal root ganglion (DRG) [48]. TrkA’s expression‎ in peripheral sensory nerve fascicles is unknown. Our study showed TrkA is not obviously expressed in the sural nerve. ChAT is a protease that is scattered in the cells and the different branches of the facial nerve may have variable compositions. Therefore, although our results show that the average overall density is not very high, it is reasonable and sufficient to distinguish between two different types of nerve fibers by using this method. Thus, comparing all other markers eval‎uated, we recommend using ChAT to detect motor nerve fibers because they have the highest efficacy of preferential staining and accuracy.

NF-200 was used as a marker for sensory myelinated fibers and the regeneration of sensory nerve axons [30]. However, NF-200, an intermediate filament protein, is a cytoskeletal structure that can be found in any mature axon [49]. NF200 was not only expressed in the majority of myelinated DRG neurons and peripheral nerves [50], but also is used to identify large sensory neurofilaments [29, 51]. TRPV1 is widely distributed in nociceptors and receives external stimulus signals [52]. It covers a large spectrum of pain qualities, from chemical to thermal, as a peripheral pain-modulating target [53]. This ion channel that is abundantly expressed in the peripheral sensory system [53]. It is worth noting that the concentration of NGF in surrounding tissues might affect the activity of TRPV1 and up-regulate its expression‎ [54]. TRPV1 is a mechanosensitive receptor that is commonly found throughout sensory C fibers; when activated, it releases the neurotransmitter CGRP [55]. Typically used as a marker for peptidergic sensory nerves, CGRP plays a pivotal role in trigeminal system, pain and temperature sensation [56]. Furthermore, it is expressed in other sensory neurons as a marker of neurons essential for heat responses in peripheral nerves [57, 58]. CGRP is preferably expressed on sensory nerve fibers, and its role is to maintain the release of neurotrophic factors, activation of protein kinases, opening of cation channels, and amplify pain signals [55]. CGRP may be expressed more during nerve regeneration after peripheral nerve injury, however it has not been studied due to lack of proper control. Our study showed for the first time that CGRP is more reliable to distinguish the sensory nerve fascicles from the motor nerve fascicles.

Although the activity of the ChAT enzyme is affected by different tissues and time points, it has been reported that the activity of ChAT is highest following axotomy and gradually declines over time [59]. We experimented with ChAT within 1 day after nerve harvest, in hopes of maximally retaining the enzymatic activity. Sensory neurons are considerably more “plastic” with respect to specification than motor neurons [60]. Studies suggest that sensory neurons are intrinsically specified with respect to their peripheral targets [61] and subclasses of sensory neurons show different integrin expression‎ [60]. Thus, we chose three different sensory markers for comparison and selected the best one. While these markers may be expressed differently in CNS or other organs, this study examines only its efficacy of preferential staining after peripheral nerve injury. Overall, this work offers a meaningful view of the development of immunostaining which satisfies the criteria of rapidity, simplicity, cost-effectiveness, high efficacy of preferential staining and reproducibility for the identification of nerve fascicles.

논의

운동 신경 섬유와 감각 신경 섬유의 식별은 신경 재생 촉진을 위한 새로운 접근법을 개발하는 데 중요할 뿐만 아니라, 폐쇄 루프 감각 피드백을 갖춘 신경 신호 제어형 신경 의수 개발에도 필수적입니다. 신경 의수를 사용하여 절단된 사지를 완전히 대체하거나 부상 후 기능을 회복하는 것은 여전히 도전적인 과제입니다 [32, 33]. 사지 운동의 자연스러운 제어를 위해 신경보철 장치에 적용될 운동 및 감각 신경과 그 신경다발의 선택적 기능을 적절히 선택하는 것이 필수적입니다 [34]. 이러한 선택성은 아직 해결되지 않은 과제입니다. 절단 환자나 말초 신경 손상에서 신경 매핑은 수술 중 신경 모니터링 기술로 수행할 수 없습니다 [35]. 따라서 외과의사는 신경 이식 시 신경의 해부학적 분포에만 의존해야 합니다. 본 연구에서는 운동 신경과 감각 신경을 구분하는 단순하고 유용한 면역 형광 접근법을 제시했습니다. 이 방법을 통해 미세 봉합으로 표시된 신경 잔여부를 식별하고 2단계 인터페이스 준비가 가능했습니다. 이는 외상 후 운동 신경과 감각 신경을 식별하는 단순한 도구이며, 수술의 다음 단계에 유용한 정보를 제공합니다. 따라서 이는 인터페이스 절차 계획에 도움을 주고 임상 환경에서 신경 의수 적용을 위한 포괄적인 해결책에 한 걸음 더 다가가는 보완적 접근법이 될 수 있습니다.

우리는 운동 또는 감각 축삭을 식별하기 위해 일반적으로 사용되는 5개의 생물표지자를 선정하고 시간 및 기타 조건을 통제하여 면역형광 염색을 병행 비교 평가했습니다. 본 연구 결과, ChAT는 운동 신경 섬유를, CGRP는 감각 신경 섬유를 효과적으로 구분할 수 있으며, ChAT와 CGRP는 각각 운동 및 감각 신경 다발에 대한 최적의 마커 조합으로 확인되었습니다. 우리는 면역형광 이미지에서 신경 마커의 양성 발현을 정량화하기 위해 평균 면적 비율, 평균 회색 값, 축삭 수 세 가지 방법을 사용했으며, 평균 회색 값 방법이 더 안정적인 방법임을 보여주었습니다. 다른 가용한 해결책이 희박한 상황에서 이 방법은 말초 신경 손상 후 주로 운동 신경 속과 주로 감각 신경 속을 구분하는 데 편리하고 빠르며 신뢰할 수 있는 방법을 제공할 것이며, 임상에서 큰 번역적 가치를 가질 것입니다. 이 기술은 신경 수리 성공률과 수술 후 기능 회복을 크게 개선할 수 있는 차별화된 신경 속 수리를 가능하게 합니다. 또한 신경 운동 신호 제어와 감각 신호 피드백 기술을 실현 가능하게 하는 미래 신경 보철 장치에 매우 중요한 추가 요소로 작용할 것입니다.

우리는 문헌에 보고된 대부분의 인기 있는 면역형광 표지자를 스크리닝하고, 주로 감각 신경 섬유로 구성된 sural 신경과 주로 운동 신경 섬유로 구성된 facial 신경에서 가장 효과적인 5가지 표지자(운동 표지자: ChAT 및 TRKA, 감각 표지자: CGRP, TRPV1, 및 NF-200)를 포함한 패널을 평가했습니다. 이 마커들은 운동 또는 감각 축삭 중 하나에 선호적으로 사용되도록 제안되었지만, 직접적인 비교는 이전에 수행된 적이 없습니다. 운동 또는 감각 축삭으로 정확히 식별할 수 있는 도구가 없기 때문에, 이 마커들의 운동 또는 감각 축삭에서의 민감도와 특이도를 계산할 수 없습니다. 따라서 본 연구에서는 이를 '선호적 염색의 효능'으로 정의합니다. 이는 항체가 원하는 조직에서 양성 결과를 나타내는지, 그리고 양성 결과의 표현 정도가 항체에 의해 결정되는지를 의미합니다. 이 점에서 선택한 모든 항체는 두 신경 섬유에서 다양한 정도로 표현되었지만, 특정성이 충분하지 않았습니다. 강한 양성 반응을 보이는 항체에서 평균 면적 비율을 데이터 분석에 사용했을 때 최대 양성 면적은 약 30%였습니다. 축삭 수는 전체 선택된 현미경 필드에서 양성 염색된 면적의 10–30%에 달하며, 이는 여전히 높은 수치입니다. 이는 축삭 외부의 공간에 면역 형광으로 적절히 특이화된 항체 없이 표시되지 않는 많은 양의 마이엘린 층이 존재하기 때문입니다. 평균 회색 값을 사용하면 양성률이 60–80%에 달하며, 따라서 평균 회색 값 방법은 선택적 염색 효율성이 더 높고 안정적인 방법입니다.

손상된 신경의 운동 신경과 감각 신경 섬유를 구분하기 위해 가장 적합한 분석 방법을 선정하기 위해, 주변 신경에서 아직 체계적으로 평가되지 않은 형광 면적 비율, 평균 회색 값, 축삭 수 등 다양한 분석 방법을 통해 효능을 평가했습니다. 평균 면적 비율을 사용할 경우, Image J 소프트웨어는 임의로 영역을 선택해 고발현 축삭을 계산하지만, 회색 값이 낮거나 약하게 발현된 신경 섬유를 무시합니다. 결과는 평균 면적 비율로 얻은 비율이 평균 회색 값보다 유의미하게 낮음을 보여줍니다. 앞서 언급된 방법과 비교할 때, 평균 회색 값을 적용하면 약한 형광을 가진 축삭도 탐지 가능합니다. 축삭 계수는 광학 현미경 하에서 100× 확대 슬라이드에 널리 사용되는 일반적으로 인정된 방법입니다. 그러나 40배 확대 이미지를 사용할 경우 직경이 작은 축삭이 발견되지 않을 수 있습니다. 이는 일부 축삭의 해상도를 저하시켜 전체 축삭 계수의 정확성을 불안정하게 만듭니다 [36]. 또한 이 방법은 원형 축삭의 수에 기반하기 때문에 신경 조직 절편 내 압축되거나 경사된 부분 영역을 무시하여 축삭 계수 오류가 발생하기 쉽습니다 [37]. 따라서, 선택적 염색 효율이 높고 상대적으로 안정적인 특성 때문에 면역 형광 결과를 정량적으로 검출하기 위해 평균 회색 값을 사용하는 것을 권장합니다.

ChAT는 주요 운동 신경 속의 표지자로, 인간 얼굴 운동 신경 세포와 쥐 얼굴 운동 신경 세포에서 높은 발현이 널리 보고되었습니다 [38, 39], 중추 신경계(CNS)와 말초 신경 축삭 모두에서 운동 신경 세포에 높은 농도로 존재합니다 [40, 41]. 운동 신경 발달과 기능의 핵심 특징은 아세틸콜린 생합성 효소인 ChAT의 발현입니다 [42, 43]. Yuan 등[44]은 면역조직화학 염색에서 ChAT를 효소 마커로 사용하여 운동 축삭을 검출하고 NF200을 감각 축삭 검출에 사용해坐골 신경에 운동 및 감각 축삭이 모두 존재함을 보여주었습니다. Castaneda와 Wang 등도 동일한 결과를 보고했습니다 [45, 46]. 그러나 이 연구들은 ChAT와 NF200이 운동 신경 섬유와 감각 신경 섬유에서 양성이 표현될 수 있음을 보여줄 뿐, 이 방법의 선택적 염색의 정확성과 효능에 대한 정량적 분석은 포함하지 않았습니다. 다른 표지자인 TrKA는 본 연구에서 세포막에 주로 표현되며, 얼굴 신경에서 강하게 표현됩니다. Carrizosa 등[26]은 척추 운동 신경세포와 안구 운동 신경세포의 기능을 검출하기 위해 TrkA 양성 발현을 선택했습니다. 중추 신경계(CNS)에서 스트라이얼 운동 신경세포와 세프탈/디아고널 밴드 복합체 내의 TrkA 신경세포(각각 >99% 및 >95%)는 ChAT와 공발현되었습니다[47]. 최근 Han 등[48]은 TrkA가 주로 Dorsal root ganglion (DRG)에 집중된 일차 감각 신경 섬유에서 발현된다는 것을 보여주었습니다. TrkA의 말초 감각 신경 속의 발현은 알려지지 않았습니다. 본 연구에서는 TrkA가 sural nerve에서 명확히 발현되지 않았습니다. ChAT는 세포와 안면 신경의 다양한 분지에 산재해 있으며, 분지별 구성비가 다를 수 있습니다. 따라서, 우리 연구 결과 평균 전체 밀도가 매우 높지 않더라도 이 방법을 사용하여 두 가지 다른 유형의 신경 섬유를 구분하는 것은 합리적이고 충분합니다. 따라서 평가된 모든 마커를 비교할 때, ChAT를 운동 신경 섬유를 탐지하는 데 권장합니다. 이는 선택적 염색 효율과 정확도가 가장 높기 때문입니다.

NF-200은 감각성 마이엘린화 섬유와 감각 신경 축삭의 재생을 표시하는 마커로 사용되었습니다 [30]. 그러나 NF-200은 중간 필라멘트 단백질로, 성숙한 축삭 어디서나 발견되는 세포 골격 구조입니다 [49]. NF200은 미엘린화된 DRG 신경세포와 말초 신경의 대부분에서 발현되었을 뿐만 아니라 [50], 큰 감각 신경 필라멘트를 식별하는 데도 사용됩니다 [29, 51]. TRPV1은 통각 수용체에 널리 분포되어 외부 자극 신호를 수신합니다 [52]. 이는 화학적에서 열적까지 다양한 통증 특성을 포함하는 말초 통증 조절 표적으로 작용합니다 [53]. 이 이온 채널은 말초 감각 시스템에 풍부하게 발현됩니다 [53]. 주목할 점은 주변 조직의 NGF 농도가 TRPV1의 활성을 영향을 미치고 그 발현을 증가시킬 수 있다는 점입니다 [54]. TRPV1은 감각 C 섬유 전반에 널리 분포된 기계감각 수용체로, 활성화 시 신경전달물질 CGRP를 방출합니다 [55]. CGRP는 펩타이드성 감각 신경의 표지자로 널리 사용되며, 삼차 신경계, 통증 및 온도 감각에 핵심적인 역할을 합니다 [56]. 또한 말초 신경에서 열 반응에 필수적인 신경세포의 표지자로 다른 감각 신경세포에서도 발현됩니다 [57, 58]. CGRP는 감각 신경 섬유에 주로 발현되며, 신경 영양 인자의 방출 유지, 단백질 키나제 활성화, 양이온 채널 개방, 통증 신호 증폭 등의 역할을 합니다 [55]. 말초 신경 손상 후 신경 재생 과정에서 CGRP 발현이 증가할 수 있으나, 적절한 대조군 부족으로 인해 연구되지 않았습니다. 우리 연구는 처음으로 CGRP가 감각 신경 속과 운동 신경 속을 구분하는 데 더 신뢰할 수 있음을 보여주었습니다.

ChAT 효소의 활성은 조직과 시간점에 따라 영향을 받지만, ChAT 활성은 축삭 절단 후 가장 높고 시간이 지남에 따라 점차 감소한다는 보고가 있습니다 [59]. 우리는 신경 채취 후 1일 이내에 ChAT를 실험하여 효소 활성을 최대한 유지하기 위해 노력했습니다. 감각 신경은 운동 신경에 비해 특화 측면에서 훨씬 더 '가소성'이 높습니다 [60]. 연구 결과, 감각 신경은 주변 목표물에 대해 내재적으로 특화되어 있으며 [61], 감각 신경의 하위 분류는 다른 인테그린 발현을 보입니다 [60]. 따라서 비교를 위해 세 가지 다른 감각 마커를 선택하고 가장 적합한 것을 선정했습니다. 이러한 마커는 중추 신경계나 다른 장기에서 다르게 발현될 수 있지만, 본 연구는 말초 신경 손상 후 선택적 염색의 효능만을 평가했습니다. 전체적으로 이 연구는 신경 속의 신경 속을 식별하기 위해 신속성, 단순성, 비용 효율성, 선택적 염색의 높은 효능 및 재현성을 충족하는 면역 염색 기술의 발전에 의미 있는 통찰을 제공합니다.

Conclusion

In this study, immunofluorescence staining was used to identify motor nerve fibers and sensory nerve fibers with five popular markers, and to eval‎uate their efficacy by different analysis methods. It is suggested that ChAT and CGRP are the optimal combination to differentiate the motor and sensory nerve fascicles by using the mean gray value method. This technique provides a more convenient and reliable method to enable a differentiated nerve fascicle repair that will greatly improve the functional recovery after nerve repair and promote neural regeneration. It helps determine the effectiveness of nerve regeneration achieved by, for example, motor or sensory path specific growth factor guided regeneration and motor or sensory path specific rehabilitation using electrical stimulation. It will also benefit the development of future neural signal-controlled prosthesis with sensory signal feedback technology.

결론
본 연구에서는 5가지 인기 있는 표지자를 사용하여 운동 신경 섬유와 감각 신경 섬유를 식별하고, 다양한 분석 방법을 통해 그 효능을 평가했습니다. 평균 회색 값 방법을 사용해 운동 신경과 감각 신경 속을 구분하는 데 ChAT와 CGRP가 최적의 조합임을 제안합니다. 이 기술은 신경 복원 후 기능 회복을 크게 개선하고 신경 재생을 촉진하기 위해 분화된 신경 속의 복원을 가능하게 하는 더 편리하고 신뢰할 수 있는 방법을 제공합니다. 이는 예를 들어 운동 또는 감각 경로 특이적 성장 인자 유도 재생이나 전기 자극을 이용한 운동 또는 감각 경로 특이적 재활을 통해 달성된 신경 재생의 효과를 판단하는 데 도움을 줍니다. 또한 감각 신호 피드백 기술을 갖춘 미래의 신경 신호 제어형 의수 개발에도 기여할 것입니다.

Availability of data and materials

Not applicable.

Abbreviations

ChAT:

Choline acetyltransferase

CGRP:

Calcitonin gene-related peptide

CNS:

Central nervous system

DAPI:

4′,6-Diamino-2-phenylindole

DRG:

Dorsal root ganglion

NF-200:

Neurofilament protein 200 kDa

PBS:

Phosphate-buffered saline

TrkA:

Tyrosine kinase receptor A

TRPV1:

Transient receptor potential vanillic acid subtype1

References

1. Navarro X, Geuna S, Grothe C, Haastert-Talini K. Introduction: thematic papers issue on peripheral nerve regeneration and repair. Anat Rec (Hoboken). 2018;301:1614–7.

   Article Google Scholar 

2. Wang ML, Rivlin M, Graham JG, Beredjiklian PK. Peripheral nerve injury, scarring, and recovery. Connect Tissue Res. 2019;60:3–9.

   Article PubMed Google Scholar 

3. Zhao W, Sun SX, Xu JJ, Chen HY, Cao XJ, Guan XH. Electrochemical identification of the property of peripheral nerve fiber based on a biocompatible polymer film via in situ incorporating gold nanoparticles. Anal Chem. 2008;80:3769–76.

   Article CAS PubMed Google Scholar