원형탈모 T cell 매개 자가면역질환...

[문형철]

Autoimmun Rev. Author manuscript; available in PMC 2023 May 1.

Published in final edited form as:

Autoimmun Rev. 2022 May; 21(5): 103061.

Published online 2022 Feb 10. doi: 10.1016/j.autrev.2022.103061

PMCID: PMC9018517

NIHMSID: NIHMS1779400

PMID: 35151885

The current state of knowledge of the immune ecosystem in alopecia areata

Samuel J Connella,b and Ali Jabbaria,b,c

Author information Copyright and License information PMC Disclaimer

The publisher's final edited version of this article is available at Autoimmun Rev

Go to:

Abstract

Alopecia areata (AA) is an autoimmune disease that affects approximately 2% of the general population. Patients with AA most commonly present with one or more patches of hair loss on the scalp in defined circular areas. A fraction of patients progress to more severe forms of the disease, in some cases with involvement of all body surfaces. The healthy anagen stage hair follicle is considered an immune privileged site, described as an environment that suppresses inflammatory immune responses. However, in AA, this immune privileged state collapses and marks the hair follicle as a target for the immune system, resulting in peri- and intrafollicular infiltration by lymphocytes. The complexity of the inflammatory ecosystem of the immune response to the hair follicle, and the relationships between the cellular and soluble participants, in AA remains incompletely understood. Many studies have demonstrated the presence of various immune cells around diseased hair follicles; however, often little is known about their respective contributions to AA pathogenesis. Furthering our understanding of the mechanisms of disease in AA is essential for the novel identification of targeted therapeutics that are efficacious and have few unintended effects.

원형 탈모증(AA)은 

일반 인구의 약 2%가 앓고 있는 

자가 면역 질환입니다. 

원형 탈모증 환자는 

두피에 하나 이상의 원형 탈모 패치가 뚜렷한 원형 부위에 나타나는 것이 가장 흔합니다. 

일부 환자는 

더 심각한 형태의 탈모로 진행하며,

일부에서는 신체 표면 전체가 탈모에 관여하기도 합니다. 

건강한 휴지기 모낭은 

염증성 면역 반응을 억제하는 환경으로 설명되는 

면역 특권 부위로 간주됩니다. 

그러나

 AA에서는 이러한 면역 특권 상태가 무너지고 

모낭이 면역 체계의 표적으로 표시되어

 림프구에 의한 모낭 주위 및 모낭 내 침윤을 초래합니다. 

AA에서 모낭에 대한 면역 반응의 염증 생태계의 복잡성과 

세포와 용해성 참여자 간의 관계는 완전하게 이해되지 않았습니다. 

많은 연구에서 

병든 모낭 주변에 다양한 면역 세포가 존재한다는 사실이 입증되었지만, 

AA 발병에 대한 각 세포의 기여도에 대해서는 알려진 바가 거의 없습니다. 

탈모증의 질병 메커니즘에 대한 이해를 넓히는 것은 

효과적이고 의도하지 않은 부작용이 적은 

표적 치료제를 새롭게 발굴하는 데 필수적입니다.

Keywords: Alopecia areata, autoimmunity, hair follicle, immune privilege, lymphocytes, T cells

Go to:

1. Introduction

The incidence of autoimmune diseases has recently been increasing in the worldwide population [1]. Autoimmune diseases arise when there is a loss of immune tolerance towards self-antigens, resulting in immune responses targeting self-tissue. It is generally understood that interactions between genetic and environmental factors are associated with the risk of developing autoimmunity [2]. Alopecia areata (AA) is a common autoimmune disease of the hair follicle that has an approximate lifetime risk of 2.1% [3]. The mean onset of disease in patients is between 25–36 years of age and is associated with increased rates of psychological disorders, including depression and anxiety, and a diminished quality of life [4]. Treatment options range from local to systemic immunosuppressants and are often unsatisfactory for patients with longstanding and severe disease [5]. However, the complexity of the immune landscape in AA and difficulties in parsing the critical cell types and pathways has complicated therapeutic discovery. Identifying how specific cells and pathways may be contributing to the pathogenesis of AA will help pave the way to developing effective treatment options for patients and provide better outcomes.

With this review, we aim to compile a comprehensive overview of our current state of knowledge of the participants of inflammation, both cellular and soluble, in AA. In particular, we focus on factors thought to be contributing to the immunoregulatory environment of the immune privileged hair follicle, the mechanisms contributing to its collapse, and the immune milieu associated with autoimmunity to the hair follicle in AA. Strides into our understanding of disease pathogenesis have resulted in identification of novel treatments in AA in the last several years, and further identification and addressing gaps in our understanding of disease mechanisms increase the likelihood that further progress will be made.

최근 전 세계적으로

자가면역 질환의 발병률이 증가하고 있습니다[1].

자가면역질환은

자기 항원에 대한 면역 관용이 상실되어

자기 조직을 표적으로 삼는 면역 반응이 일어날 때 발생합니다.

일반적으로

유전적 요인과 환경적 요인 간의 상호작용이

자가면역 발병 위험과 관련이 있는 것으로 알려져 있습니다[2].

원형 탈모증(AA)은

모낭의 흔한 자가 면역 질환으로

평생 약 2.1%의 위험성을 가지고 있습니다[3].

환자의 평균 발병 연령은 25~36세이며,

우울증과 불안을 포함한 심리적 장애의 발생률 증가와

삶의 질 저하와 관련이 있습니다[4].

치료 옵션은

국소 면역 억제제부터

전신 면역 억제제까지 다양하며,

장기간 중증 질환을 앓고 있는 환자에게는 만족스럽지 않은 경우가 많습니다[5].

그러나

AA의 면역 환경이 복잡하고

중요한 세포 유형과 경로를 분석하는 데 어려움이 있기 때문에

치료법 발견이 복잡해졌습니다.

특정 세포와 경로가 AA의 발병에 어떻게 기여하는지 파악하면

환자를 위한 효과적인 치료 옵션을 개발하고

더 나은 결과를 제공하는 데 도움이 될 것입니다.

이 리뷰에서는

알코올성 지방간에서

세포와 수용성 염증의 참여자에 대한

현재 지식의 포괄적인 개요를 정리하고자 합니다.

특히 면역 특권 모낭의 면역 조절 환경,

모낭의 붕괴에 기여하는 메커니즘,

AA에서 모낭에 대한 자가면역과 관련된 면역 환경에 기여하는 것으로 생각되는 요인에

초점을 맞추고 있습니다. 질

병 발병 기전에 대한 이해의 진전으로 지난 몇 년 동안 새로운 치료법이 밝혀졌으며,

질병 메커니즘에 대한 이해의 격차를 더 파악하고 해결하면 더 많은 진전이 이루어질 가능성이 높아집니다.

Go to:

2. Clinical presentation of disease

AA manifests as non-scarring hair loss. The disease course is relatively unpredictable and can affect any hair-bearing location on the body. In most cases, hair loss develops in well-defined, circular areas on the scalp [6]. On the face, hair loss may emerge within the beard area as well as within the eyebrows and eyelashes [6]. The disease can progress to include total loss of scalp hair, termed alopecia totalis (AT), or complete loss of hair on the entire body, termed alopecia universalis (AU) [7]. Mild cases often undergo spontaneous remission, marked by hair regrowth. This usually occurs within the first year after the development of disease, and spontaneous relapse may occur at any time [4, 7, 8]. Descriptions of “overnight graying” have been used to describe AA due to preferential loss of pigmented hairs and relative sparring of non-pigmented for white/grey hairs [6], although other mechanisms may explain this observation [9]. Furthermore, emergence of non-pigmented hairs in lesions exhibiting regrowth is not uncommon [10]. The baseline hair color is not thought to impact the loss of hair [11]. Nail involvement is another characteristic seen in patients with AA. In particular, pitting and trachyonychia, or a sandpaper-like appearance, are most often described in association with AA [6, 7]. Although the morphology of lesions in AA usually has a distinctive appearance, the severity, location, and course are, in general, highly variable.

원형 탈모는

흉터가 남지 않는 탈모로 나타납니다.

질병의 진행 과정은 비교적 예측할 수 없으며

신체의 모든 모발이 있는 부위에 영향을 미칠 수 있습니다.

대부분의 경우 탈모는

두피의 잘 정의된 원형 부위에서 발생합니다[6].

얼굴에서는

눈썹과 속눈썹뿐만 아니라 수염 부위에서도 탈모가 나타날 수 있습니다[6].

이 질환은

두피 모발이 완전히 빠지는 전신성 탈모증(AT)이나

전신에 모발이 완전히 빠지는 전신성 탈모증(AU)으로 진행될 수 있습니다[7].

경증인 경우 모발이 다시 자라면서

자연적으로 완화되는 경우가 많습니다.

이는 보통 발병 후 첫 1년 이내에 발생하며

언제든지 자연적으로 재발할 수 있습니다[4, 7, 8].

다른 메커니즘이 이러한 관찰을 설명할 수 있지만[6],

색소성 모발의 우선적 손실과 흰색/회색 모발에 대한 비색소성 모발의 상대적 스파링으로 인해

"밤새 회색이 되는 것"에 대한 설명이

AA를 설명하는 데 사용되었습니다 [9].

또한,

재성장을 보이는 병변에서 색소가 없는 모발이 출현하는 경우도 드물지 않습니다 [10].

기준 모발 색은 모발 손실에 영향을 미치지 않는 것으로 생각됩니다 [11]. 손발톱 침범은 원형 탈모 환자에서 나타나는 또 다른 특징입니다. 특히, 손발톱이 사포처럼 보이는 구덩이와 조갑박리증은 AA와 관련하여 가장 자주 설명됩니다 [6, 7]. AA의 병변 형태는 일반적으로 독특한 외관을 가지고 있지만, 일반적으로 중증도, 위치 및 경과가 매우 다양합니다.

Go to:

3. Genetic risk factors

Numerous studies of various methodologies examining gene relationships with AA have been conducted. Many of the earlier studies used focused or targeted techniques to identify associations with specific loci or genes (Table I) [12–21]. Genome wide assessments have more recently been used to perform more comprehensive and unbiased analyses of susceptible genes in AA patients. The first genome wide association study (GWAS) was performed on a US population of patients [22] and identified eight susceptibility loci/variants. Several of these loci were found to be closely associated with genes implicated in autoimmunity and within T cell pathways including: IL2/IL21 on chromosome 4q27; CTLA4 and ICOS on chromosome 2q33.2, ULBP3 and ULBP6 on chromosome 6q25.1; IL2RA on chromosome 10p15.1, IKZF4 on chromosome 12q13; and closely bundled HLA-DRA/HLA-DQA1/HLA-DQA3/HLA-DQB2/BTNL2/NOTCH4/MICA on chromosome 6p21.32. STX17 on chromosome 9q31.1 and PRDX5 on chromosome 11q13 were also implicated. A follow-up GWAS performed on a European population reproduced five of the susceptibility loci/genes previously identified including: ULBP3/ULBP6, PRDX5, IL2/IL21, IKZF4/ERBB3, and STX17 [23]. Furthermore, this study found two additional loci, closely associated with the genes IL13 and KIAA0350/CLEC16A, with variants demonstrating increased susceptibility to AA. Two additional GWASs, performed by Forstbauer et al and Betz et al, examined patients from the US and Europe and identified SPATA5 on chromosome 4q27-q28, ACOXL/BCL2L11 on chromosome 2q13, and C11ofr30/LRCC32 on chromosome 11q13.5 as additional variants with increased susceptibility in AA patients [24, 25]. In most cases, it is currently unknown how these variants affect gene expression‎ and function, complicating translating these findings into insights into disease pathogenesis.

Table 1.

Single targeted gene studies with variants identified to be associated with AA.

GeneAA associated variants†

HLA	DQB1*03 (DQ3) 12 DRB1*1104 (DR11) 12 DQB1*0301 (DQ7) 12 DRB1*0401 (DR4) 12
IL1RN	IL-1RN*2 13
IL4	Intron 3 VNTR polymorphism 14
IL17RA	rs879577 SNP 15
IL17A	GG genotype in A7488G polymorphism 16
IL1RN/IL1L1	IL1RN+2018/IL1L1+4734 SNPs 17
AIRE	AIRE961G 18
FOXP3	rs2294020–3675(A) SNP 19
ICOSLG	rs378299–509(C) SNP 19
MICA	MICA*5.1 & MICA*6 20
MIF	MIF-173*C 21

Open in a separate window

†Nomenclature as per reference

VNTR= variable number tandem repeat, SNP= single nucleotide polymorphism

Go to:

4. Mouse models of AA

In 1994, Sundberg et al. reported that C3H/HeJ mice undergo spontaneous hair loss with features similar to that of human AA, including defined regions of hair loss undergoing a waxing and waning cycle, regrowth of depigmented hairs in the regions of alopecia, and mononuclear cell infiltration of the hair follicle bulb and epithelium [26]. Additionally, similarly to the infiltrate around human AA hair follicles, the hair follicle infiltrate of alopecic C3H/HeJ mice was comprised predominantly of CD8 and CD4 T cells. Spontaneous disease development affected approximately 15–20% of C3H/HeJ mice, with a strong preponderance for female mice. In addition, the onset of disease was relatively protracted in male mice; female mice developed disease as early as 2–3 month of age, whereas male mice first began exhibiting hair loss at 7 months of age.

The ability to experimentally induce AA in the C3H/HeJ strain has increased its usefulness as a laboratory model. Transfer of skin from previously-affected mice led to the induction of AA in recipient mice with substantially diminished latency [27]. Engraftment using skin from either the dorsal or ventral side of an affected mouse induced hair loss in as little as 7 weeks post-surgery and in all recipients by 11 weeks. Furthermore, grafted skin from unaffected hairy donors onto AA affected mice resulted in loss of hair within the graft by 21 days post-surgery.

Injection of lymphocytes isolated from AA affected mice has similarly been found to induce AA in previously unaffected recipient mice. Subcutaneous injections of unsorted skin draining lymph node (LN, SDLN) cells, sorted CD4 T cells from SDLNs, and sorted CD8 T cells from SDLNs were able to induce hair loss in recipient mice [28]. In this work, unseparated LN cells induced widespread hair loss in 80% of mice injected, whereas sorted CD8 T cells induced hair loss in about 3 to 5 weeks that was localized to the site of injection in 100% of mice, and CD4 T cells induced diffuse hair loss in 7 weeks in 43% of mice with limited involvement at the site of injection. Recently, Wang et al. reported a similar method of inducing disease using a stimulated population of unsorted lymphocytes derived from SDLNs of AA affected mice [29]. In this study, LN cells co-cultured in vitro with anti-CD3/anti-CD28-coated beads were intradermally injected into eight-week-old recipient C3H mice. Approximately 90% of recipient mice experienced hair loss between 2 to 4 weeks following transfer. It was noted that the route of injection was critically important for disease induction in this model, as subcutaneous injections were less effective at inducing hair loss than intradermal injections. This innovation additionally allowed for a greater disease-inducing capacity from individual donor mice, as the lymphocyte population from a single donor mouse could be used to induce disease in dozens of recipient mice.

Both graft- and cell-induction models are capable of recapitulating characteristics consistent with those observed in AA patients and C3H/HeJ mice with spontaneous disease. Examination of the follicles within lesional skin from mice of each model shows dystrophic hair follicles and perifollicular infiltration of CD4 and CD8 T cells, although the unmanipulated LN cell transfer study [27] solely reported mononuclear infiltration without further identification. The relatively simple implementation of these models makes them particularly attractive for use in the laboratory.

In addition to the C3H/HeJ mouse model, there are several reports of use of xenograft mouse models of AA. In one described model, skin biopsies from severe AA patients were transplanted onto severe combined immunodeficient (SCID) mice and observed for hair regrowth [30]. These grafts were then injected with T cells extracted from AA scalp skin biopsies that had been cocultured with irradiated patient peripheral blood mononuclear cells (PBMCs) and follicular homogenate. In over half of the grafts that received activated T cell injections, there was an induction of hair loss nine to ten weeks following injection, which was not observed in grafts that received injections of uncultured AA scalp T cells, unirradiated PBMCs, or no cells. An additional humanized mouse model of AA has more recently been reported in which skin from a healthy donor was grafted onto SCID mice [31]. After three months, grafts were injected subcutaneously with an in vitro activated PBMC population from the same healthy donor that had been bead-enriched for natural killer group 2 membrane D (NKG2D)+/CD56+ lymphocytes. In the grafts that received the activated enriched lymphocyte populations, hair loss was appreciated between three to five weeks following transfer.

Another murine model of AA was recently described in the C57BL/6J mouse strain [32]. In this model, Alli et al. surreptitiously discovered a T cell receptor (TCR) that, when used in a retrogenic bone marrow chimera model, could be used to generate a mouse model of autoimmune hair loss. Following transplantation of bone marrow cells transduced to express the TCR, recipient mice developed hair loss as early as six weeks post transplantation. These mice experienced waxing and waning of the extent of disease, with a median of 3 cycles before achieving complete loss of hair. The hair follicles in lesional skin showed intrafollicular CD3+ mononuclear infiltration of the anagen follicles, a similar feature of AA in humans. However, these mice exhibited discrepancies with human AA, including the presence of surface epidermal changes as well as scratching behaviors, both of which may be indicative of the presence of other factors not seen in human AA or the C3H AA model. In addition, fixing of the TCR in this model to that associated with CD8 T cells also complicates study of the role of other T cell types, including conventional CD4 T cells and regulatory T cells, in disease development.

Go to:

5. Immune privileged status of the hair follicle

5.1. Concept of immune privilege and relationship to hair follicle biology

The hair follicle is an immune privileged organ. Immune privilege (IP) is believed to be a state in which inflammation is limited and the recognition of foreign antigens is restricted [33]. Multiple organs have been described as being immune privileged sites including the eyes [34], brain [35], testes [36], fetus [37], liver [38], gut [39], and hair follicles [40]. Experimentally, the first description of immune privilege was made in 1948 by Peter Medawar, where he showed that, in previously unimmunized rabbits, skin grafted into the anterior chamber of the eye or left cerebral hemisphere would not be rejected, in contrast with grafts onto previously immunized rabbits [41]. He proposed that this discrepancy was due to a passive mechanism, where the grafts were rejected only once lymphatic drainage was present through vascularization of the grafts, allowing for an immune reaction to occur. This idea of immune privilege was first described specifically in the context of the hair follicle by Billingham, when he presented experiments in which skin from black guinea pigs grafted onto white guinea pigs [42]. Whereas the grafted skin failed to regain pigment, hairs from within the engrafted skin were retained and emerged as donor black. This observation is consistent with the concept that the follicular bulb in the grafted skin was associated with a protective environment that allowed the melanocytes to survive and avoid immediate destruction by the host immune system.

5.1. 면역 특권의 개념과 모낭 생물학과의 관계
모낭은

면역 특권을 가진 기관입니다.

면역 특권(IP)은

염증이 제한되고

외부 항원에 대한 인식이 제한되는 상태로 여겨집니다 [33].

눈 [34], 뇌 [35], 고환 [36], 태아 [37], 간 [38], 장 [39], 모낭 [40] 등

여러 장기가 면역 특권 부위로 설명되어 왔습니다.

실험적으로

면역 특권에 대한 최초의 설명은

1948년 피터 메다와(Peter Medawar)에 의해 이루어졌는데,

그는 이전에 면역을 받지 않은 토끼의 경우 눈의 전실이나 좌뇌 반구에 이식한 피부가

이전에 면역을 받은 토끼에 이식한 것과 달리 거부 반응을 보이지 않는다는 것을 보여주었습니다[41].

그는 이러한 불일치가

이식편의 혈관화를 통해 림프 배수가 일어나면

이식편이 거부되어 면역 반응이 일어날 수 있는 수동적인 메커니즘 때문이라고 제안했습니다.

이러한 면역 특권에 대한 개념은

흑인 기니피그의 피부를

백인 기니피그에 이식한 실험을 발표하면서

모낭의 맥락에서 처음으로 구체적으로 설명되었습니다[42].

이식된 피부는 색소를 회복하지 못했지만, 생착된 피부 내부의 모발은 유지되어 기증자의 검은색으로 나타났습니다. 이러한 관찰은 이식된 피부의 모낭구가 멜라닌 세포가 생존하고 숙주 면역 체계에 의한 즉각적인 파괴를 피할 수 있는 보호 환경과 관련이 있다는 개념과 일치합니다.

5.2. Mechanisms of hair follicle IP

Similar to other organs that exhibit IP, the hair follicle possesses an array of factors that lead to its ability to regulate immune cell responses and mitigate potential autoimmunity. There are low levels of major histocompatibility complex (MHC) class I expression‎ in the cells of the lower/proximal hair follicle [43–45], indicating that the presentation of antigens to autoreactive CD8+ T cells is decreased. There is also diminished expression‎ of components involved in the formation of the MHC class I complex, including transporter associated with antigen processing-2 (TAP2) and β2 microglobulin (β2m) [44, 45]. Furthermore, there are few MHC class II-presenting cells and few MHC class II-presenting intraepithelial dendritic cells of the hair follicle [45], indicative of reduced capacity to stimulate CD4 T cell responses in the steady state. In AA, collapse of immune privilege at the hair follicle is thought to be a critical event that is needed for the development of disease; this is characterized by an upregulation of MHC class I and class II molecules within and in association with the follicular epithelium, which is seen in human skin and mouse models of AA.

Studies using in vitro cultures of microdissected hair follicles and immunohistological analyses of skin samples have expanded our understanding of how the follicular epithelium and surrounding cells enforce hair follicle IP by identifying a handful of secreted molecules which can impede an immune response. Transforming growth factor beta 1 (TGFβ1), a molecule often associated with diminished immune responses, was identified to be expressed at lower levels around the follicles of AA patients as compared to healthy controls [46]. When added into culture with microdissected human hair follicles, TGFβ1 led to decreased MHC class I expression‎ that had been induced by interferon (IFN)-γ [44]. In addition, in in vitro hair follicle cultures, both alpha-melanocyte stimulating hormone (α-MSH), a neuropeptide known to induce melanogenesis [47] and highly expressed in the bulge region of the hair follicle in healthy skin [48], and insulin growth factor 1 (IGF1) also led to downregulated expression‎ of MHC class I in the follicular epithelium [44]. IGF1 was also shown in these studies to induce downregulation of β2m, TAP2, and MHC class II expression‎.

Macrophage migration inhibitory factor (MIF) may be contributing to hair follicle IP and its dysregulation may be contributing to AA pathogenesis. Two groups identified strong expression‎ of MIF in healthy control hair follicles [48, 49]. Additionally, it was found that MIF expression‎ is greatly reduced in lesional samples from AA patients [49]. Its published role as a suppressor of natural killer (NK) cell function [50], in combination with these data, supports a potential role in hair follicle IP.

Expression‎ of programmed death ligand-1 (PDL1), an inhibitory molecule that signals through PD-1 present on T cells to induce anergy, may be another mechanism by which the hair follicle regulates T cell activation. Isolated cells of the hair follicle including dermal sheath cup cells (DSCC) and dermal papilla (DP) cells express high levels of PDL1 when compared to non-follicular fibroblasts [51]. When allogenic T cells were placed into a coculture system with DSCC and DP cells, the T cells exhibited a diminished IFNγ response, decreased proliferation, and an increased expression‎ of caspase molecules, as compared to non-follicular fibroblast cells which do not express PDL1. These differences seen could be mitigated by using siRNA technology to knock down PDL1 expression‎. These data suggest that PDL1 expression‎ by cells of the hair follicle may be contributing to hair follicle IP in part by directly impacting T cell function. Increased expression‎ of PDL1 and PDL2 on fibroblasts in AA lesions in the C3H model has been described; how this relates to AA pathogenesis is unclear [52].

Fas-Fas ligand (FasL) interactions mediate cellular apoptosis and has been proposed as a immunoregulatory mechanism in the IP of the eye [53], raising questions of its potential role in hair follicle IP. In a skin graft-induced AA model, mice deficient in Fas and FasL showed a resistance to developing AA [54]. Additionally, when skin from Fas or FasL was grafted onto an AA affected mice, the grafts remained protected from losing hair. Whether the Fas/FasL pathway is a mechanism of hair follicle IP or is acting in another manner has yet to be determined. Fas is expressed by the cells of the hair follicle in mice with and without AA, and FasL positive cells in the perifollicular region are low in unaffected mice but are increased in AA mice [54]. Additionally, the hair follicles in AA mice show a higher number of apoptotic cells as compared to the unaffected hair follicles. This is suggestive that this pathway occurs during or following the collapse of hair follicle IP, as the perifollicular infiltration contains a higher number of FasL positive cells, which could then induce apoptosis of the hair follicle cells.

The collapse of these mechanisms leaves the hair follicle vulnerable to being targeted by the immune system, creating an environment in which AA can develop. How cells of the immune system and cytokines may be contributing to the pathogenesis of disease is explored below.

IP를 나타내는 다른 기관과 마찬가지로

모낭은 면역 세포 반응을 조절하고

잠재적인 자가 면역을 완화하는 능력을 가진 다양한 요인을 가지고 있습니다.

하부/근위 모낭의 세포에서 주요 조직 적합성 복합체(MHC) 클래스 I 발현 수준이 낮으며[43-45],

이는 자가 반응성 CD8+ T 세포에 대한 항원의 제시가 감소한다는 것을 나타냅니다.

또한 항원 처리와 관련된 수송체-2(TAP2) 및 β2 마이크로글로불린(β2m)을 포함하여 MHC 클래스 I 복합체 형성에 관여하는 성분의 발현이 감소합니다[44, 45]. 또한, 모낭의 MHC 클래스 II 제시 세포와 MHC 클래스 II 제시 상피 내 수지상 세포가 거의 없으며[45], 이는 정상 상태에서 CD4 T 세포 반응을 자극하는 능력이 감소했음을 나타냅니다. AA에서 모낭에서의 면역 특권의 붕괴는 질병의 발병에 필요한 중요한 사건으로 생각되며, 이는 모낭 상피 내에서 그리고 모낭 상피와 관련하여 MHC 클래스 I 및 클래스 II 분자의 상향 조절이 특징이며, 이는 인간 피부와 AA 마우스 모델에서 볼 수 있습니다.

미세 절개한 모낭의 시험관 배양과 피부 샘플의 면역 조직학적 분석을 이용한 연구를 통해 모낭 상피와 주변 세포가 면역 반응을 방해할 수 있는 소수의 분비 분자를 확인함으로써 모낭 IP를 강화하는 방법에 대한 이해의 폭이 넓어졌습니다. 면역 반응 감소와 관련이 있는 분자인 형질 전환 성장 인자 베타 1(TGFβ1)은 건강한 대조군에 비해 AA 환자의 모낭 주변에서 더 낮은 수준으로 발현되는 것으로 확인되었습니다[46]. 미세 절개한 인간 모낭을 배양에 첨가했을 때 TGFβ1은 인터페론(IFN)-γ에 의해 유도된 MHC 클래스 I 발현을 감소시켰습니다 [44]. 또한 시험관 내 모낭 배양에서 멜라닌 생성을 유도하는 것으로 알려져 있고[47] 건강한 피부에서 모낭의 돌출 부위에서 높게 발현되는 신경 펩타이드인 알파 멜라닌 세포 자극 호르몬(α-MSH)과[48] 인슐린 성장 인자 1(IGF1)도 모낭 상피에서 MHC 클래스 I의 발현을 하향 조절하는 것으로 나타났습니다[44]. 이러한 연구에서 IGF1은 또한 β2m, TAP2 및 MHC 클래스 II 발현의 하향 조절을 유도하는 것으로 나타났습니다.

대식세포 이동 억제 인자(MIF)가 모낭 IP에 기여할 수 있으며, 이 인자의 조절 장애가 AA 발병에 기여할 수 있습니다. 두 그룹에서 건강한 대조군 모낭에서 MIF의 강한 발현을 확인했습니다[48, 49]. 또한 AA 환자의 병변 샘플에서 MIF 발현이 크게 감소하는 것으로 밝혀졌습니다 [49]. 자연살해(NK) 세포 기능의 억제제로서 발표된 역할[50]은 이러한 데이터와 함께 모낭 IP에서의 잠재적 역할을 뒷받침합니다.

T 세포에 존재하는 PD-1을 통해 신호를 전달하여 항원을 유도하는 억제 분자인 프로그램된 사멸 리간드-1(PDL1)의 발현은 모낭이 T 세포 활성화를 조절하는 또 다른 메커니즘일 수 있습니다. 모낭의 분리된 세포인 피부 피막 컵 세포(DSCC)와 피부 유두(DP) 세포는 비모낭 섬유아세포와 비교했을 때 높은 수준의 PDL1을 발현합니다[51]. 동종 T 세포를 DSCC 및 DP 세포와 공배양 시스템에 넣었을 때, T 세포는 PDL1을 발현하지 않는 비모낭 섬유아세포 세포에 비해 IFNγ 반응이 감소하고 증식이 감소하며 카스파제 분자의 발현이 증가하는 것으로 나타났습니다. 이러한 차이는 siRNA 기술을 사용하여 PDL1 발현을 억제함으로써 완화될 수 있습니다. 이러한 데이터는 모낭 세포의 PDL1 발현이 부분적으로 T세포 기능에 직접적인 영향을 미침으로써 모낭 IP에 기여할 수 있음을 시사합니다. C3H 모델에서 AA 병변의 섬유아세포에서 PDL1 및 PDL2의 발현 증가가 설명되었지만 이것이 AA 발병과 어떤 관련이 있는지는 불분명합니다 [52].

Fas-Fas 리간드(FasL) 상호 작용은 세포 사멸을 매개하며 눈의 IP에서 면역 조절 메커니즘으로 제안되어 모낭 IP에서의 잠재적 역할에 대한 의문이 제기되었습니다 [53]. 피부 이식 유도 AA 모델에서 Fas와 FasL이 결핍된 마우스는 AA 발생에 대한 저항성을 보였습니다 [54]. 또한, AA에 걸린 마우스에 Fas 또는 FasL의 피부를 이식했을 때 이식된 피부는 탈모로부터 보호되었습니다. Fas/FasL 경로가 모낭 IP의 메커니즘인지 아니면 다른 방식으로 작용하는지는 아직 밝혀지지 않았습니다. Fas는 AA가 있는 마우스와 없는 마우스의 모낭 세포에서 발현되며, 모낭 주위 영역의 FasL 양성 세포는 영향을 받지 않은 마우스에서는 낮지만 AA 마우스에서는 증가합니다 [54]. 또한 AA 마우스의 모낭은 영향을 받지 않은 모낭에 비해 세포 사멸 세포 수가 더 많은 것으로 나타났습니다. 이는 모낭 주위 침윤에 더 많은 수의 FasL 양성 세포가 포함되어 모낭 세포의 세포 사멸을 유도할 수 있기 때문에 이 경로가 모낭 IP가 붕괴되는 동안 또는 붕괴 후에 발생한다는 것을 암시합니다.

이러한 메커니즘이 무너지면 모낭이 면역 체계의 표적이 되기 쉬워져 AA가 발생할 수 있는 환경이 조성됩니다. 면역 체계의 세포와 사이토카인이 질병의 발병에 어떻게 기여하는지 아래에서 살펴보세요.

Go to:

6. Cytokines, signaling pathways, and immune cells in AA

6.1. Interferon gamma (IFN-γ)

IFN-γ plays an important role in a range of processes including host defense against pathogens, inflammation, immune cell responses, and tumor immunity [55]. IFN-γ has been implicated as a critical contributor in the pathogenesis of AA. Using hair follicles microdissected from human scalp skin, Ito and colleagues found that low dose IFN-γ treatment induces a significant upregulation of MHC class I in the hair follicle epithelium in areas where they had previously seen low or absent expression‎ [44]. Additionally, IFN-γ co-culture led to upregulated expression‎ of TAP2 and β2m, machinery associated with and required for MHC class I presentation [44]. In this manner, IFN-γ is thought to be driving the collapse of the hair follicle IP, causing upregulation of MHC class I on the follicular epithelium and allowing for recognition by and stimulation of autoreactive T cells.

With an important role in hair follicle IP, multiple groups have asked what role IFN-γ may play in AA pathogenesis and examined AA patient samples for its presence. Microarray analyses conducted on human and C3H/HeJ AA skin identified increased IFN response gene expression‎ signatures in AA-affected lesional skin when compared to skin from non-diseased controls [56]. Studies assessing levels of circulating cytokines have reported higher IFN-γ levels in patients with AA as compared to healthy controls [57–60]. Additionally, the capacity of IFN-γ to induce AA or contribute to AA pathogenesis has been examined in the C3H/HeJ mouse model, with studies showing conflicting findings. In one set of studies, Gilhar and colleagues injected low dose IFN-γ or saline intravenously for three consecutive days and then weekly in C3H/HeJ mice that had been shaved to initiate an anagen hair cycle [61]. They found seven of nine mice who received injections of IFN-γ experienced hair loss, whereas none of the saline injected group did. In contrast, Sundberg et al. injected IFN-γ or phosphate buffered saline (PBS) subcutaneously three times in succession following depilation in recipient mice, and found that only 2 of 10 mice that received IFN-γ injections experienced hair loss; they concluded from this that mice did not exhibit an increased rate of hair loss compared to what would have been expected for spontaneous disease [62]. A more recent study found that subcutaneous injection of IFN-γ in combination with polyinosinic-polycytidylic acid was able to induce hair loss in approximately 80% of recipient mice; in contrast, mice injected with IFN-γ alone did not exhibit hair loss [63]. This last study suggests induction of AA by IFN-γ is enhanced by additional inflammatory signals. In in vivo models, the induction of AA in C3H mice using AA skin grafts could be prevented in the setting of IFNγ-deficiency [64] or antibody-mediated neutralization of IFN-γ [56]. In IFN-γ deficient mice and anti-IFN-γ antibody-treated mice, weak expression‎ of MHC class I was associated with the lack of disease development, indicating IFN-γ may be a vital player in the development of AA through the collapse of hair follicle IP and upregulation of MHC Class I, among other possible contributions.

6.2. Common gamma chain (γc) cytokines

Common gamma chain (γc) cytokines, so named by their shared use of the γc subunit as a part of their respective receptors play an important role in the biological functions and regulations of the adaptive immune system [65]. This family of cytokines, comprised of interleukin (IL) −2, IL-4, IL-7, IL-9, IL-15, and IL-21 are notable for providing critical survival signals and influencing the function of T and B cells. Analysis of lesional AA skin from both human and mice identified a genetic signature from this family of cytokines, supporting a role in AA pathogenesis [56].

Functionally, IL-2 is important in promoting CD8 T cell activity, driving naïve CD4 T cell differentiation, and maintaining T regulatory cells [66], and can therefore have context-dependent pro- or anti- inflammatory roles. Expression‎ of the IL-2 receptor (IL-2R) is recognized in two forms: the low affinity dimeric IL-2R and the high affinity trimeric IL-2R [66]. In lesional skin from AA patients, expression‎ of IL-2 and the IL-2Rα and β subunits were found to be upregulated [56, 67, 68]. Additionally, serum IL-2 levels have been reported to be increased in patients with AA as compared to healthy controls in two studies [69, 70], although a third study found that serum IL-2 levels in AA patients were significantly decreased compared to controls [71]. Furthermore, serum IL-2 levels were found to be the highest in patients with the lowest duration of disease (<1 year) [69].

Several studies have examined the effect of manipulating IL-2 or IL-2 receptor signaling on AA pathogenesis. Neutralization of IL-2 in C3H/HeJ mice that received skin grafts from AA donor mice prevented development of disease and resulted in diminished effector CD8 T cells in the skin [56]. These data are in line with data from mouse models exhibiting reduced IL-2 expression‎. Heterozygotic IL-2+/− mice, that exhibit reduced IL-2 expression‎ but do not suffer from the high mortality observed in homozygotic IL-2−/− mice, developed disease at a decreased rate as compared to IL-2+/+ controls [72]. Lesional skin from IL-2+/+ mice showed dense intrafollicular infiltrates of CD8 T cells and perifollicular infiltration by CD4 T cells, while lesional skin from IL-2+/− showed reduced numbers of CD8 and CD4 T cells. While these studies would indicate that IL-2 acts in a pro-inflammatory manner, small studies examining therapeutic use of IL-2, in order to stimulate the formation of an anti-inflammatory regulatory T cell population, in the clinical setting have been performed. In one pilot study, patients with severe AA that had failed to respond to prior systemic therapies were treated with low dose IL-2 injections [73]. From these treatments, four out of five patients responded with some amount of hair regrowth. Examination of the skin before and after treatment found that those patients who responded to treatment showed a decrease in CD8 T cell infiltration and an increase in the presence of T regulatory cells. However, in a follow-up multicenter randomized trial where patients with severe AA were treated with low dose IL-2 injections, IL-2 treatment failed to show an improvement in hair regrowth [74]. Further studies will be required to determine if IL-2- or IL-2 receptor-based therapies can serve as a treatment for AA.

IL-15 plays a critical role in the maintenance and survival of T cells as well as in the regulation of T cell cytotoxicity and NK cell development [75, 76]. In a manner unique among the γc cytokines, IL-15 binds to the IL-15Rα intracellularly and is shuttled to the membrane of the cell. In the context of this complex, IL-15 is presented in trans to cells expressing the IL-2Rβ/IL-15Rβ and γc chain heterodimer [77]. Similar to IL-2, messenger RNA (mRNA) of IL-15, and its receptor, were found to be upregulated in the skin of AA patients and AA affected mice [56, 68]. Furthermore, histological analysis of human and mouse AA skin shows upregulation of IL-15 and its receptor around the hair follicles when compared to that of healthy hair follicles [56]. Serum IL-15 levels have been examined across multiple studies and predominantly show increased levels in AA patients as compared to healthy controls [57, 78, 79], although one study failed to describe a difference in serum IL-15 levels between AA and control patients [80]. To determine if IL-15 contributed mechanistically to AA pathogenesis, Xing et al performed experiments in which IL-15 signaling was neutralized in mice following AA skin-grafting [56]. In contrast to AA skin-grafted mice treated with an isotype control antibody that robustly developed hair loss, AA skin-grafted mice that received IL-2Rβ/IL-15Rβ antibody treatments were protected from developing AA. In the context of this model of AA, IL-15 is therefore a necessary factor in AA disease emergence. In particular, IL-15 may be augmenting the function of a pathogenic population of CD8 T cells, discussed further below.

6.3. Janus kinase (JAK)-mediated signaling

Cytokines known to be critical in AA, such as IFN-γ and IL-15, bind to their receptors and activate JAKs, which go on to induce phosphorylation of signal transducer and activator of transcription (STAT) molecules and further transcription. The JAK family, made of JAK1, JAK2, JAK3 and Tyk2, mediate signaling of a wide variety of receptors for cytokines, hormones, and growth factors. For example, IFN-γ signals through JAK1 and JAK2, while IL-15 signals through JAK1 and JAK3. [81]. Spurred by recognition that numerous cytokine pathways that contribute to AA are mediated by JAK/STAT signaling, small molecule JAK inhibitors have been and continue to be studied as therapeutic treatment for AA as well as a variety of inflammatory and autoimmune diseases [82]. The US Food and Drug Administration (FDA) initially approved tofacitinib and ruxolitinib for the treatment of rheumatoid arthritis and myelofibrosis, respectively [83, 84]; these agents were soon thereafter explored for their efficacy in the treatment of AA. Murine AA skin graft recipients systemically treated with these inhibitors starting at the time of grafting were protected from developing hair loss, and mice with established disease demonstrated hair regrowth when these agents were used as a topical treatment [56]. In the same report, treatment of three human subjects with moderate to severe AA with ruxolitinib, an inhibitor of JAK1 and JAK2, orally twice daily resulted in near complete hair regrowth within 5 months of treatment. Other JAK inhibitors have been similarly assessed for their efficacy in the treatment of AA. A study looking at baricitinib, another FDA approved JAK1/2 inhibitor, demonstrated this agent similarly protected C3H AA-skin graft recipients from developing disease, successfully treated established disease in the mouse model, and reversed disease in a patient with AA [85]. Further clinical studies examining the effectiveness of JAK inhibitors in the treatment of moderate to severe AA, AU, and AT have emerged and are ongoing [86–92]. These studies have shown that the use of JAK inhibitors for the treatment of AA may be safe and effective; however, as is the case with immunosuppressive and biologic treatment for other cutaneous autoimmune diseases, JAK inhibitors do not induce a durable remission in patients with AA, as most patients appear to relapse upon discontinuation [91, 92]. Interestingly, AA patients have been shown to harbor a small but detectable population of clonally expanded T cells clones in scalp skin following treatment with tofacitinib [93], consistent with JAK inhibitors suppressing those critical effector functions of autoreactive T cells required for the disease but not completely eliminating them. More complete compilations of the use of specific JAK inhibitors in the treatment of human AA has been extensively reviewed by others [94, 95].

6.4. CD8+ cytotoxic T cells (CTL)

CD8 T cells are members of the adaptive immune system with cytotoxic capabilities and are found clustered around the bulb of the hair follicle in lesional AA skin [49]. An early murine AA study found that when CD8 T cells isolated from the SDLNs of AA mice were transferred subcutaneously into recipient mice, those mice exhibited localized hair loss around the injection site [28]. These data underscored that these cells are playing a role in disease, leading to future insights into disease mechanisms.

The first GWAS for AA identified the NKG2D ligands, UL16-binding protein (ULBP) 3 and ULBP6, as disease-associated genes, and complementary experiments showed that ULBP3 is highly expressed within the follicle of AA patients [22]. This work also demonstrated that CD8 T cells localized to the hair follicle in AA patients expressed NKG2D. Growing data support a central role for the NKG2D-expressing CD8 T cell population in disease pathogenesis [56]. NKG2D+ CD8 T cells are found in higher frequencies in the skin and SDLNs of affected mice compared with unaffected mice, where they are largely absent [56]. In an in vitro killing assay, CD8 T cells expressing NKG2D from AA mice demonstrated cytotoxicity towards NKG2D-ligand-expressing dermal sheath cells, and this effect could be partially blocked using antibodies specific for NKG2D. Furthermore, transfer of NKG2D+ CD8 T cells from the SDLNs of AA-affected mice was sufficient to induce disease in recipient mice following intradermal injection, whereas transfer of NKG2Dneg CD8 T cells from the same lymph nodes was not. These data provide compelling evidence that CD8 T cells expressing NKG2D are critical in autoimmune attack on the hair follicle in murine AA and AA patients. More work is required to identify the mechanisms behind how this cell population is contributing to AA.

As previously mentioned, the emergence of a pathogenic CD8 T cell population may be licensed at least in part from additional stimulatory input in the form of IL-15. IL-15 has been shown to contribute to the generation of lymphokine activated killer (LAK) cells, a heterogeneous population of cells with NK cell-like cytolytic functions [96]. Generation of these LAK cells has been examined through work focused on intraepithelial (IE) CD8 T cells, a population of cells found in the gut, which contribute to the pathogenesis of celiac disease (CD) [96]. Like NKG2D-expressing CD8 T cells in AA, patients with CD harbor a population of NKG2D-expressing CD8 T cells that localize to the intraepithelial region of the targeted tissue. In normal healthy intestines, IE CD8 T cells express low levels of NKG2D [96]. However, if those cells are stimulated with IL-15, they upregulate their expression‎ of NKG2D and their cytolytic capability. In active CD patients, where upregulation of IL-15 in the intestinal epithelium is a hallmark characteristic, the IE CD8 T cells express increased levels of NKG2D as compared to normal IE CD8 T cells [77, 96].

Additionally, single cell sequencing identified a CD8 T cell cluster within the skin and SDLNs of alopecic mice which expressed NKG2D, but also other genes associated with NK cells, like natural killer cell granule protein 7 (NKG7) [97]. This pattern of gene expression‎ is similar to that observed involving a CD8 T cell cluster identified from AA patients, and particularly enriched in lesional skin, in which NKG7 was also expressed [97]. In a recent publication, a group demonstrated that NKG7 expression‎ on CD8 T cells is involved in the secretion of cytotoxic granules by exocytosis and mediating inflammation [98]. Using a model of severe malaria, where antigen-specific CD8 T cells trafficked to the brain, CD8 T cells that were deficient in NKG7 had reduced expression‎ of activation markers, including CD11a, CD49d, and granzyme B, but also were not recruited to the site of inflammation. Furthermore, they found that NKG7 is colocalized with CD107a and contributes to the translocation of CD107a to the surface of the cell [98]. The role of NKG7 expression‎ on the putatively pathogenic CD8 T cell population in AA, however, has yet to be defined.

The trafficking of AA-inducing CD8 T cells to disease sites has begun to be examined. CD8 T cells expressing NKG2D in the SDLNs of AA-affected mice express CXC receptor (CXCR) 3, a chemokine receptor that is associated with a type I inflammatory response in CD4 and CD8 T cells, which are localized to the follicle in lesional skin [99, 100]. Expression‎ of CXCR3 on CD8 T cells is thought to contribute to trafficking of autoreactive CD8 T cells into the skin during disease development. Neutralization of CXCR3 blocked migration of the CD8 T cells into active AA skin, and AA skin graft mediated disease induction was greatly inhibited [99]. Expression‎ of the CXCR3 ligands, CXC ligand (CXCL) 9, CXCL10, and CXCL11, are upregulated in the skin of AA affected mice [99]. These IFN-γ-induced ligands exhibit increased expression‎ at the level of mRNA as early as 5 weeks post skin graft, prior to AA emergence in C3H mice [101]. Additionally, in the chronic phase of disease in AA patients, CXCR3+ CD8 T cells were present around the bulb of the hair follicle in high numbers [102]. The previously-mentioned single cell sequencing study provided complementary data confirm‎ing high CXCR3 expression‎ by the putatively-pathogenic CD8 T cell population with shared TCR sequences in skin and SDLNs [97].

Although it is unclear which effector mechanisms CD8 T cells are employing to effect disease, there is data indicating that their cytotoxic capabilities play a role. Of the CD8 T cells characterized by single cell sequencing, the cluster of CD8 T cells identified that expressed high levels of CXCR3 also expressed high levels of granzyme B and perforin [97]. Among AA patients treated with squaric acid dibutylester, nonresponding patients exhibited a higher percent of granzyme B producing CD8 T cells in the skin when compared to that of patients that responded with favorable outcomes. In C3H mice that underwent skin graft induction, granzyme B transcripts were found to be increased in the skin 5 weeks post graft, an early timepoint in disease induction, suggesting that granzyme B or granzyme B-expressing cells may have a role in the early stages of disease development [101]. Despite these data, mechanistic studies supporting a role for cytotoxicity as a central mechanism of AA pathogenesis are lacking.

6.5. CD4 T helper cells

CD4 T cells, also called helper T cells, have been identified as perifollicular infiltrating immune cells of the hair follicle in AA patients [103], and there is growing appreciation of the varied composition of CD4 T cell types present (Figure 1). In the C3H mouse model, subcutaneous injection of isolated CD4 T cells from SDLNs of affected mice were capable of inducing systemic hair loss [28]. Furthermore, a potential effector role of CD4 T cells in inducing hair loss was also suggested in the Dundee experimental bald rat (DEBR) model of AA, as these rats experience lesions of hair loss that mimic that of human AA [104]. In this model, lesional hair follicles exhibited a mononuclear cell infiltration that consisted of perifollicular CD4 T cells and intrafollicular CD8 T cells and macrophages [104, 105]. Temporary depletion of CD4 T cells using monoclonal antibodies in this model led to partial hair regrowth, with resumption of hair loss when the CD4 T cell population began re-accumulating [106]. However, how CD4 T cells are contributing to AA, including which effector mechanisms are being employed to induce disease, have yet to be defined.

Figure 1.

T cell populations in AA.

T cell populations present in human AA skin. CD4 and CD8 T cells are present in the peri- and intra-follicular regions of the hair follicle, respectively, in AA. CD8 T cells expressing NKG2D and CXCR3 have been appreciated as effector cells contributing to the development of AA. Other studies have suggested that the various CD4 T helper subsets (Th1, Th2, and Th17) are present around the hair follicle in human AA, identified in part by expression‎ of their respective chemokine receptors. Whether these CD4+ T cell subsets have a pathogenic role in disease needs to be elucidated further. Additionally, further work to determine the role of regulatory T cells is needed.

6.5.1. T-helper type 1 (Th1) 

Th1 cells are an effector helper subtype that produce of IFN-γ, IL-2, and tumor necrosis factor (TNF)-β [107]. T-box protein expressed in T cells (T-bet) is regarded as the master transcription factor for Th1 cells. Their primary function is thought to be the elimination of intracellular pathogens, but they have also been associated with organ-specific autoimmune diseases, including multiple sclerosis [108].

Serologic analysis of AA patients shows an upregulation of Th1 axis cytokines (IFN-γ, IL-2, and IL-12) and chemokines (CXCL10, CC receptor ligand (CCL) 2, and CCL3) in circulation [69, 78]. During acute phase disease (<6 months and patchy), patients have increased numbers of circulating and skin infiltrating CXCR3+ CD4 T cells, suggesting that these cells could be contributing to the early stages of disease pathogenesis [102]. Immunohistochemical analysis of skin biopsies showed CD4 T cells expressing CC receptor (CCR) 5, a marker associated with Th1 cells [109], infiltrating the intra- and peri-bulbar regions of the hair follicles, suggesting a potential role in AA [110]. Transcriptomic analyses demonstrate an upregulation of a Th1/IFN-associated genes CXCL9, CXCL10, and STAT1 in lesional AA when compared to either nonlesional AA or healthy control skin, as well as in nonlesional samples compared to healthy controls [68]. Additionally, expression‎ of a similar set of markers were examined in samples from nonlesional and lesional scalp skin before and after treatment with methylprednisolone [67]. At baseline, lesional skin samples showed an upregulation of CXCL9, CXCL10, and IFN-γ when compared with non-lesional skin. Following treatment, there was decreased expression‎ of CXCL9; other Th1 markers such as IFN-γ and CXCL10 trended lower but did not reach statistical significance, and markers of other T helper subsets, such as IL-13, were also decreased following treatment. Although it appears the presence of Th1 cells correlates with disease, further studies are needed to determine their role in AA.

6.5.2. T-regulatory cells (Treg) 

Tregs play an important role in maintaining immunologic tolerance, whose function is to aid in the suppression of an immune response. Regulated by Forkhead box P3 (FoxP3), Tregs are often defined in two ways: (1) natural Tregs (nTregs), which are produced in the thymus as FoxP3 expressing cells or (2) induced Tregs (iTregs), which are naïve CD4 T cells that have been induced to differentiate in the periphery under certain circumstances [107]. Their immunoregulatory function raises the question of whether their dysfunction contributes to autoimmune diseases including AA, and they have been investigated as to whether they may be exploited as a potential therapeutic agent [111]. Histological examination of scalp biopsies from AA patients as well as patients with other cutaneous diseases, specifically lichen planopilaris, inflamed seborrheic keratosis, and non-specific folliculitis, showed a lower frequency of CD4+ FoxP3+ and CD25+ FoxP3+ Tregs in AA lesions compared to these other entities [112]. This finding is consistent with other work that found few FoxP3+ cells localized around the hair follicle in samples from AA patients with varied clinical presentations (multiple patchy, AT, and acute, diffuse, and total alopecia (ADTA)) [113]. Additionally, the percentage of lymphocytes that are FoxP3+/CD4+ found in peripheral blood are higher in healthy control patients when compared to that of AA patients; similarly, the percentage of lymphocytes that are FoxP3+ found in scalp skin samples were higher in healthy control samples compared to that of AA patients [114]. Somewhat paradoxically, the percentages of either FoxP3+ CD4 T cells in the circulation or FoxP3+ cells in the scalp lesions, of total lymphocytes, were found to be higher in lesions from patients with more severe disease when compared to those from patients with mild forms of disease, according to the same study. In the C3H/HeJ model, mice that received grafts from AA donors two weeks prior had fewer skin infiltrating CD4+CD25+ cells when compared to mice that received grafts from normal hairy mice [115]. The spleens and SDLNs of these mice also contained a smaller frequency of these Treg cells. In a study examining disease induction by way of transfer of different lymph node populations from AA mice, transfer of CD4+CD25+ lymphocytes at a 1:2 ratio with either CD25− CD4+ or CD8+ T cells decreased the rate of disease development when compared to mice that received either CD25− CD4+ or CD8+ T cells alone [28], indicative that these cells can play an immunosuppressive role in the context of AA. However, a more rigorous assessment of the mechanisms by which they are impacting AA development, are needed.

6.5.3. T-helper type 17 (Th17) 

Published data indicates that Th17 cells and the cytokines they produce are present in AA, but their roles in disease pathogenesis, if any, is unknown. Among many reports, serum IL-17A levels are found to be elevated in AA patients when compared to healthy controls [57, 71, 116–118]. Furthermore, both protein and RNA expression‎ of IL-17A are increased in lesional scalp tissue of AA patients [71, 118]. Other Th17-associated cytokines, including IL-21 and IL-22, have been assessed in AA patients. In a few reports, levels of both cytokines were found to be increased in the serum and lesional skin of patients [71, 116, 118]; however, in another study, no difference in serum IL-22 levels among AA and controls patients was observed [118]. RNA expression‎ of the two subunits of IL-23 (p40, shared with IL-12, and IL-23p19), which plays a role in the differentiation of naïve CD4 T cells into the Th17 phenotype, were found to be upregulated in lesional skin of patients [68]. This upregulation was also reported at the protein level in the skin and serum of AA-affected mice [119]. However, neutralization of IL-23, using an antibody specific for the p40 subunit shared with IL-12, did not prevent mice from developing disease, suggesting that IL-23 does not play an active role in disease pathogenesis [119]. Lesional mRNA levels of both IL-17A and IL-22 were found to be positively correlated with disease subtype in patients, with the lowest levels in the healthy control group and highest in those with AU and AT [71]. However, a relationship between serum levels of IL-17A and disease severity is not consistently observed among reports. In one study, serum IL-17 levels were found to be correlated with both clinical type and disease severity [116], but that result was not found in a handful of other studies [57, 117, 118].

Several examinations focused on Th17 cells, rather than IL-17 or Th17-associated cytokines, have been conducted. Histological examination of AA affected hair follicles has shown an increase in infiltrating IL-17 producing lymphocytes as compared to the hair follicles from healthy controls [113, 114, 120]. One study, using CCR6 as an identifying marker, found intra- and peri-bulbar infiltration of Th17 cells in AA, with some cells localized around the bulge of the follicle, an area not typically associated with inflammation in AA [110]. Furthermore, a second study found a higher frequency of IL-17 producing lymphocytes around the hair follicle in patients with less severe disease (patchy subtype) as compared to more severe forms (AT) [113]. Additionally, Th17 cells have been found to be increased in the circulation of AA patients [114]. These cells were more abundant in patients with acute disease (< 6 months) when compared to patients with chronic disease (> 6 months). Whether there are Th17 or IL-17 producing lymphocytes present in the C3H/HeJ mouse model of AA is not known; however, in the MOG TCR mouse model, the pathogenic T cells produced IL-17 during disease development, and these levels correlated with disease progression [32].

Secukinumab, a FDA approved IL-17A inhibitor that is effective at treating plaque psoriasis, has been tested in AA patients for efficacy [121]. In a double-blind, randomized, prospective pilot study, 11 patients were enrolled to examine the potential for secukinumab as an AA treatment [122]. Of the patients enrolled, seven were placed on 300mg secukinumab treatment and four on a placebo treatment. There was a high attrition rate in the study, with 2 patients who received the drug and 1 in the placebo group remaining until the final endpoint, complicating drawing rigorous conclusions from the study. One of the two patients receiving the drug had an improvement in their Severity of Alopecia Tool (SALT) score, a clinical assessment metric that correlates with percent of scalp involvement, while the other had no change; the single patient in the placebo group had a worsening of their score. Multiple case reports of patients being given secukinumab for their psoriasis have shown conflicting results. In one report, after six weeks of treatment, a patient with a 25-year history of AU showed hair regrowth [123]. This patient exhibited hair loss four months following the cessation of treatment. In contrast, two other reports suggested that targeting the IL-17 pathways for psoriasis may have induced AA in the patients. In one report, a patient presented with hair loss in the beard area, which spread to the scalp, six months after starting secukinumab treatment for psoriasis [124]. In the second report, two patients with psoriasis vulgaris underwent treatment, the first receiving secukinumab and the second receiving brodalumab, an IL-17R antagonist; both showed development of AA within months after starting treatment [125]. Taken together, the current data on if and how Th17 cells and their cytokines are affecting the hair follicle and disease progression remain unclear.

6.5.4. T-helper type 2 (Th2) 

Recent studies have suggested the presence of a Th2 signature in AA. Transcriptomic analysis of lesional and nonlesional samples from AA patients as well as control patient samples has shown an increase in molecules associated with a Th2 profile, including IL-13, CCL5, CCL13, CCL18, and CCL26 [68]. In a comparison of AA and control scalp skin sections, IL-5 and IL-10 mRNA expression‎ were found to be increased in the deep dermis, defined as below the level of the sebaceous glands and inclusive of where hair bulbs would typically be found, despite the absence of differences in the expression‎ of these cytokines in the upper dermis of AA and control samples [126]. Serologic analysis of AA demonstrated increased levels of the Th2 cytokine IL-13 and Th2-associated chemokines CCL13, CCL17, CCL22, and CCL26 when compared with healthy controls patients [78]. In both acute (<6 months) and chronic (>6 months) phases of disease, CD4 T cells expressing CCR4, a chemokine receptor expressed by Th2 cells, were found at higher frequencies in PBMCs from AA patients when compared to that of healthy controls [102].

A recent double-blind multicenter study looked at the efficacy of dupilumab, a monoclonal antibody that binds the alpha subunit of the IL-4 receptor, in the treatment of AA [127]. Dupilumab is an effective treatment option in patients with type 2 diseases, like atopic dermatitis and asthma, suggesting it may have positive effects in AA, as AA is commonly associated with atopic diseases [128–130]. In total, 60 patients were enrolled in this trial with the criteria of having AA longer than 6 months and with greater than 30% scalp hair loss. These patients were enrolled into two arms at a 2:1 ratio, one being treatment with dupilumab for the duration of the study (drug), and the second being given placebo for the first 24 weeks followed by dupilumab for the remainder of the study (placebo). At 24 weeks, those in the drug group showed a positive change in SALT score while those in the placebo group showed a negative change. Between week 24 and 48, patients in the drug group continued to show a positive change, although those in the placebo group also began to show a positive change in SALT score. At 48 weeks, 32.5% of the patients who had received the treatment showed at least a 30% improvement in SALT score. Additionally, patients with high serum IgE or familial atopy were more likely to respond to the drug. Further studies are required to determine whether the Th2 profile that can be seen in the skin of AA patients is a contributing factor to the disease itself, or whether it is having an indirect effect and whether targeting Th2 cytokines may be an efficacious therapeutic strategy.

6.6. γδ T cells

Observations of healthy hair follicles have identified few γδ T cells to be present; however, in AA skin, the number of γδ T cells is significantly higher when compared to healthy skin, and they appear to be localized to the bulb and suprabulbar epithelium [45, 131]. Because γδ T cells share some functional and phenotypic characteristics with conventional αβ T cells, including expression‎ of NKG2D and secretion of cytotoxic molecules and pro-inflammatory cytokines, there is biological plausibility for their participation in AA pathogenesis [132]. The γδ T cells identified around healthy follicles were γδ1+, expressed skin homing receptors (CXCR3, CXCR4, and CCR2), and displayed a non-activated phenotype (NKG2DdimCD69−) [131]. In contrast, γδ T cells found in AA skin, which were identified in both lesional and nonlesional skin, expressed NKG2D and were capable of producing IFN-γ. Paradoxically, however, there were a higher number of IFN-γ producing γδ T cells present in nonlesional skin when compared to lesional skin. How these cells may be contributing to AA requires further study.

6.7. Dendritic cells

Numerous studies have identified the presence of dendritic cells (DCs) around hair follicles of AA patients [67, 68, 126, 133, 134]. Studies have reported the presence of CD11c+ myeloid DCs [67, 68], plasmacytoid DCs (pDC) [133], and CD1a+ Langerhans cells [67, 126] in the DC compartment of AA lesions. Furthermore, unbiased single cell RNA-seq performed on murine AA skin identified antigen presenting cell (APC) signatures consistent with monocyte derived DCs (CCR2+ CD64+) among other cell types [97]. Adoptive transfer of CD11c+ cells isolated from the SDLNs of AA affected C3H mice was not sufficient at inducing disease in recipient mice [28]; a definitive role in AA pathogenesis therefore has yet to be defined.

Following activation, DCs are able to induce T cell responses through secretion of cytokines and presentation of antigens through MHC class I and II [135]. IL-12 and IL-23 are cytokines produced by DCs known to skew or maintain polarized T cell responses towards Th1 and Th17 differentiation, respectively. Analysis of AA skin by real-time PCR (RT-PCR) showed an increased expression‎ of the shared IL-12/23 p40 subunit in lesional skin as compared to nonlesional skin [68], and its expression‎ decreased following corticosteroid treatment [67]. In addition, increased protein levels of IL-12/23p40 were detected in the skin of AA-affected mice when compared with unaffected controls [119]. Targeting the IL-12 and IL-23 receptor pathways as a potential treatment for AA, however, has demonstrated conflicting results. In multiple case reports and case series, treatment with ustekinumab, a FDA approved IL-12/23p40 blocker for treating psoriasis, was effective at inducing hair regrowth in patients with AA [136–138]. In contrast, a separate case series of patients treated with ustekinumab failed to respond to treatment [119]. These latter data were in line with similar, aforementioned results found in the C3H AA model, in which treatment with an IL-12/23p40 neutralizing antibody failed to prevent the onset of disease. While these types of clinical reports, including case reports and case series, are often helpful in identifying the potential for efficacy, the high spontaneous resolution rate, especially in patients with patchy-type disease necessitates use of larger, randomized, prospective, well-controlled clinical trials to determine efficacy.

The role of type I interferon, a product of pDCs, has been examined in human tissue and mouse AA models. In human samples, a type I interferon signature has been suggested in AA with studies reporting increased myxovirus-resistance protein 1 (MxA) staining localized to the bulb of inflammatory AA lesions, which were defined as histologically demonstrating peribulbar inflammation, when compared to noninflammatory AA lesions, in which peribulbar inflammation was not appreciated [133, 139]. In vitro treatment of vibrissae microdissected from mice with IFN-α induced upregulation of MHC class I and CXCL10 on the outer root sheath and matrix cells of the follicles [134]. Immunostaining of murine skin for pDCs and IFN-α+ pDCs demonstrated an increase in the number of these populations localized around the bulbar region in samples from spontaneous AA mice compared with that from mice that did not develop AA [134]. Somewhat paradoxically, nonlesional skin from spontaneous AA mice showed even higher numbers of these cell types than lesional skin. Furthermore, mice with hair loss induced by transfer of in vitro-expanded T cells from donor AA mice showed increased numbers of skin-infiltrating pDCs in nonlesional samples as compared to lesional samples, potentially suggesting that these cells may play a role in the initiation phase of disease development. In addition, induction of AA was observed when activated pDCs were intradermally injected in three out of three recipient C3H mice. While these studies seem to indicate a role for IFN-α and pDCs in the induction of AA, larger mechanistic studies are likely needed to conclusively dissect the contributions of pDCs in AA pathogenesis.

6.8. NK and NKT cells

Two of the loci linked with AA are associated with MICA and ULBP3, both of which encode ligands of NKG2D that have previously shown to be highly expressed in lesional AA skin [20, 22, 49]. This raises the possibility that NK and NKT cells, both of which may express NKG2D, may participate in the pathogenesis of AA. A classic function of NK cells is to address and clear MHC class I-negative cells. Although low or absent expression‎ of MHC class I has been demonstrated in the inferior region of hair follicles, they do not appear to be normally targeted by NK cells, and NK cells are not thought to commonly localize around these structures [45, 49]. Prior work indicates that healthy hair follicles express increased MIF and diminished levels of MICA, mechanisms that would prevent targeting by NK cells. In AA, however, infiltration of CD56+NKG2D+ NK cells has been demonstrated in the perifollicular area in lesional AA skin [49], and hair follicles from AA patients exhibit increased MICA expression‎ and downregulated MIF expression‎ [49]. Further, increased frequencies of NK cells, determined by expression‎ of CD16, have been reported in the peripheral blood of patients with AU when compared with normal controls [140]. Following corticosteroid treatment, the frequency of these cells in the circulation were reduced back to levels found in healthy controls. From these studies, however, it is unclear whether the presence of these cells in circulation serves as a biomarker for disease or whether they are contributors to AA pathogenesis.

Experiments in murine AA, however, suggest that NK cells may be playing an immunoprotective role in the hair follicle. Depletion of cutaneous NK cells using an anti-asialo-GM1 treatment during the 4 weeks following AA skin grafting appeared to result in an acceleration of disease, as compared to the mice receiving an isotype control treatment [141]. The mechanism surrounding this is not well understood, but are consistent with reports in other diseases of NK cells having an anti-inflammatory function [142, 143]. Indeed, NK cells have been documented to lyse T cells as well as secrete immunosuppressive cytokines, including IL-10 or TGFβ [144–146], which may inhibit the formation of autoreactive T cells [147]. It is therefore yet unclear if they promote or act to prevent the development of AA.

Using a human-xenograft mouse model of AA, data reported by Ghraieb et al suggested that invariant NKT (iNKT) cells may also have a protective role in AA pathogenesis [148]. Injection of in vitro-stimulated, bead-enriched NKT cells cultured with alpha-Galactosylceramide (α-GalCer), or direct injection of α-GalCer, into donor healthy human scalp skin transplanted onto mice was sufficient to prevent development of PBMC-induced hair loss in the graft. Additionally, depletion or neutralization of NKT cells in PBMC cultures prior to injection in grafts resulted in hair loss. Furthermore, the ability of α-GalCer-treatment on PBMCs to prevent disease onset in skin grafts was lost when the skin grafts were treated with IL-10 blocking antibodies administered daily, indicated that IL-10 is required for NKT cell-mediated protection of hair loss. Interestingly, in vitro proliferation of NKG2D+ CD8 T cells was suppressed when cocultured with IL-10 producing iNKT cells, raising the possibility that iNKT cells may be directly suppressing the pathogenic CD8 T cell population in AA. Further studies are needed to identify the role and extent to which of iNKT cells affect autoimmunity to the hair follicle in the native setting and their potential for therapeutic purposes.

6.9. Mast cells

The contributions of mast cells in the pathogenesis of AA remains unknown. Examinations of lesional skin samples from AA patients indicated there are a higher number of mast cells localized around the hair follicle, specifically in the mesenchymal, perivascular, and perifollicular regions, as compared to healthy control skin [46, 126]. As expected, these cells expressed tryptase, a marker of degranulation and activation, suggesting that the increased number of mast cells may be contributing to the inflammatory response around the hair follicle [46, 126]. Additionally, in AA lesions, mast cells are found to be in close proximity to CD8 T cells and strongly express MHC class I, supporting a role in modulating the CD8 T cell response [46, 126]. Indeed, mast cells located around the hair follicle expressed the co-stimulatory molecules OX40L, CD30L, and 4–1BBL, lacked expression‎ of TGFβ1 and failed to produce IL-10, further supporting a role that they may be directly activating surrounding T cells [46]. In the AA skin-induced C3H mouse model, an increased number of mast cells and an increase in mast cells in close proximity with CD8 T cells following disease development was observed. In a human xenograft mouse model, grafted scalp skin induced to lose hair exhibited an increased number of perifollicular mast cells as compared to the uninduced skin grafts. Additionally, an increase in the number of mast cells that were interacting with CD8 T cells in these AA lesions was observed [46]. While in total these data raise the possibility that mast cells may be participating in AA pathogenesis, experimental data from models that directly alter mast cell numbers or function are lacking.

6.10. Autoantibodies and B cell involvement

Early studies examined the role of autoantibodies in AA [149–151]. Two studies found that a greater frequency of AA patients had autoantibodies targeting the outer root sheath and matrix of the hair follicle and were predominantly reacting with different hair follicle antigens of 44, 47, 50, 52, and 57 kDa [149, 150]. To examine a potential mechanistic role of autoantibodies in the development of AA, a xenograft AA model was used wherein skin from AA or control patients was grafted onto nude mice and injected with patient serum [152]. Immunohistological analysis found IgG and C3 deposits around the hair follicles of both AA and healthy control skin that was grafted onto the mice. The introduction of AA patient serum, and the autoantibodies present, had no effect on hair regrowth within grafts of AA skin, as compared with otherwise unmanipulated conditions in this model. In addition, adoptive transfer experiments in C3H mice indicated that splenic B cells from AA mice were not able to induce disease in recipients [28]. In total, the data favor that autoantibodies to the hair follicle appear to be an epiphenomenon and not directly involved in the disease process. The role of B cells as APCs, however, has not been studied.

6.11. Autoantigens and T cell receptor

The identity of a common autoantigen in AA has not been well-defined. The observation that depigmented hair is often spared in clinical disease has led some to hypothesize that melanocyte-derived antigens could be the target of autoimmune attack [153]. T cells isolated from lesional skin of patients with AA have been noted to prevent hair regrowth after stimulation with melanocyte peptides in a mouse xenograft model. Histologically, these grafts showed similar features to that of AA lesions, with dystrophic follicles, infiltration of CD4 and CD8 T cells, and expression‎ of MHC class I and II in the epithelium [153]. In addition, a decreased number of follicular melanocytes have been reported in AA patients, as detected by immunohistochemical staining on scalp samples from AA patients for melanocyte-specific protein melanoma antigen recognized by T cells 1 (MART1) [154]. Alternatively, trichohyalin (THH/TCHH) has been proposed as a potential hair follicle antigen involved in the AA immune response [155]. Immunoprecipitation experiments using serum from AA or control patients found that 100% of AA patients exhibited antibodies to THH, as compared to only 50% control patients. In silico prediction models have also been used to identify candidate hair follicle peptides with high affinity to HLA-A*0201 [156]. Protein epitopes identified to be more highly stabilized onto HLA when compared to their positive control influence peptide included specific TCHH-derived polypeptides, glycoprotein 100 (GP100), and MART1. Pools of mixed TCHH-derived polypeptides were capable of inducing greater CTL activation in PBMCs from AA patients as compared to controls. In addition. using an ELISPOT assay, two additional melanocyte specific antigens, tyrosinase (TYR) and tyrosinase related protein 2 (TYRP2) induced a higher frequency of activation in PBMCs from AA patients. Stimulating PBMCs with a mixed group of TCHH polypeptides showed an increased number of IFN-γ+CD8+ cells from AA samples, while TYR stimulation showed non-specific cell activation as both AA and control samples showed higher numbers of IFN-γ+CD8+ cells. Identification of the critical target epitopes would greatly further our understanding of the immune response of AA.

The identification of a clonal T cell population, with a defined T cell receptor, capable of inducing disease would also support the presence of a critical antigen in AA. Sequencing of the T cell receptor repertoire in AA affected mice identified matched sequences of NKG2D+ CD8 T cells that were isolated between the SDLNs and lesional skin [93]. Furthermore, paired single cell RNA-sequencing and TCR sequencing identified a population of CD8 T cells with a pathogenic signature and a shared/overlapping TCR repertoire unique to murine AA skin and lymph nodes, indicating that a single population of NKG2D-expressing CD8 T cells travels between these tissues and effects disease [97]. At this time, these data are mostly correlative and indicative of biomarkers of disease rather than causation; further work is therefore needed.

Go to:

7. Conclusions

Great advancements in our understanding of the immunology of AA have been made in recent years. A variety of immune cells have been described to be present in AA skin and the hair follicle microenvironment, and transfer studies in in vivo models and mechanistic studies have furthered our understanding of AA pathogenesis. Ultimately, these latter types of studies may shed light on the sequence of immunological events that leads to loss of tolerance to hair follicle and its targeting by the immune system, resulting in the emergence of clinical disease in AA. Although newer therapeutic options are being refined, there is a need for a targeted, efficacious treatment with minimal untoward effects, such as immunosuppression or susceptibility to opportunistic pathogens. Identifying how certain cell populations or cytokines are contributing to AA pathogenesis has great potential in pinpointing the next therapeutic target.

Highlights:

• Alopecia areata is characterized by immune infiltration of the hair follicle leading to hair loss.

• Mouse models have aided in our understanding of mechanisms in the immune response of the hair follicle.

• Collapse of immune privilege at the hair follicle is a critical event for AA development.

• There is evidence that many different immune cells may be present around the hair follicle in AA.

• Further studies are needed to dissect the contribution of these various immune cells to the development of AA.

Go to:

Acknowledgements

We thank Teresa Ruggle of the University of Iowa Design Center for help in designing the figure for this manuscript.

Funding

This work was supported by the Department of Veterans Affairs (VA Merit Award I01 BX004907 to AJ), the National Institutes of Health (R01 AR077194 and K08 AR069111 to AJ), the University of Iowa Predoctoral Training Program in Immunology (T32AI007485), and the University of Iowa Department of Dermatology.

Go to:

Abbreviations:

AA	alopecia areata
AT	alopecia totalis
AU	alopecia universalis
GWAS	genome wide association study
SDLN	skin draining lymph node
SCID	severe combined immunodeficient
PBMC	peripheral blood mononuclear cells
NKG2D	natural killer group 2 membrane D
TCR	T cell receptor
IP	immune privilege
MHC	major histocompatibility complex
TAP	transporter associated with antigen processing
β2M	β2 microglobulin
TFGβ	transforming growth factor beta
IFN	interferon
α-MSH	alpha-melanocyte stimulating hormone
IGF1	insulin growth factor 1
MIF	macrophage migration inhibitory factor
NK	natural killer
PDL1	programmed death ligand 1
DSCC	dermal sheath cup cells
DP	dermal papilla
FasL	Fas ligand
PBS	phosphate buffered saline
IL	interleukin
mRNA	messenger RNA
JAK	janus kinase
STAT	signal transducer and activator of transcription
FDA	Food and Drug Administration
ULBP	UL16-binding protein
LAK	lymphokine activated killer
IE	intraepithelial
CD	Celiac disease
NKG7	natural killer granule protein 7
CXCR	CXC receptor
CXCL	CXC ligand
DEBR	Dundee experimental bald rat
TNF	tumor necrosis factor
Tbet	T-box protein expressed in T cells
CCL	CC receptor ligand
CCR	CC receptor
nTreg	natural Treg
iTreg	induced Treg
ADTA	acute, diffuse, and total alopecia
SALT	Severity of Alopecia Tool
DC	dendritic cell
pDC	plasmacytoid dendritic cell
RT-PCR	real-time PCR
MxA	myxovirus-resistance protein 1
MICA	major histocompatibility complex class I chain-related gene A
iNKT	invariant NKT
MART1	melanoma antigen recognized by T cells 1
TCHH/THH	trichohyalin
GP100	glycoprotein 100
TYR	tyrosinase
TYRP2	tyrosinase related protein 2

Go to:

Footnotes

Declaration of interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Go to:

References

[1] Wang L, Wang F-S and Gershwin ME, Human autoimmune diseases: a comprehensive update, J. Intern. Med. 278 (2015) 369–395. [PubMed] [Google Scholar]

[2] Ellis JA, Kemp AS and Ponsonby A-L, Gene–environment interaction in autoimmune disease, Expert Rev. Mol. Med. 16 (2014). [PubMed] [Google Scholar]

[3] Mirzoyev SA, Schrum AG, Davis MDP and Torgerson RR, Lifetime Incidence Risk of Alopecia Areata Estimated at 2.1% by Rochester Epidemiology Project, 1990–2009, J. Invest. Dermatol. 134 (2014) 1141–1142. [PMC free article] [PubMed] [Google Scholar]

[4] Villasante Fricke A and Miteva M, Epidemiology and burden of alopecia areata: a systematic review, Clin. Cosmet. Investig. Dermatol. (2015) 397–403. [PMC free article] [PubMed] [Google Scholar]

[5] Strazzulla LC, Wang EHC, Avila L, Lo Sicco K, Brinster N, Christiano AM and Shapiro J, Alopecia areata: An appraisal of new treatment approaches and overview of current therapies, J. Am. Acad. Dermatol. 78 (2018) 15–24. [PubMed] [Google Scholar]