TH1 effector CD4 T cells rely on IFN-γ production to induce alopecia areata

Samuel J Connell; Sydney Crotts; Ryan Reis; Maddison M Lensing; Payton Kahl; Nicholas Henderson; Otgonzaya Ayush; Zhaowen Zhu; Luana S Ortolan; Audrey C Ko; Erin M Shriver; Keith D Carter; John T Harty; Joan M Goverman; Ali Jabbari

doi:10.1126/sciadv.adz2257

. 2025 Oct 31;11(44):eadz2257. doi: 10.1126/sciadv.adz2257

T_H1 effector CD4 T cells rely on IFN-γ production to induce alopecia areata

Samuel J Connell ^1,², Sydney Crotts ², Ryan Reis ^2,^3,⁴, Maddison M Lensing ^1,², Payton Kahl ², Nicholas Henderson ², Otgonzaya Ayush ², Zhaowen Zhu ², Luana S Ortolan ², Audrey C Ko ⁵, Erin M Shriver ⁵, Keith D Carter ⁵, John T Harty ^1,⁶, Joan M Goverman ⁷, Ali Jabbari ^1,^2,^3,^8,^*

PMCID: PMC12577704 PMID: 41171930

Abstract

Alopecia areata (AA) is an autoimmune disease defined by hair loss and peribulbar infiltrate of CD8 and CD4 T cells. Prior studies have focused on the role of CD8 T cells in the development of AA. Multiple roles for CD4 T cell help have been demonstrated for support of CD8 T cell responses; however, the role of CD4 T cells in AA remains unclear. Here, we demonstrate that CD4 T cells from the skin-draining lymph nodes (SDLNs) of AA mice transferred disease to recipient mice. These cells exhibited a T helper type 1 (T_H1) effector transcriptional and phenotypic profile, and their pathogenic activity required endogenous CD8 T cells and host IFN-γ responsiveness. Targeted deletion of CD4 T cell–mediated production of IFN-γ abrogated the ability of this cell population to transfer disease. Together, these data provide mechanistic insights into pathways driving AA development, strengthening our understanding of the disease and inviting studies into exploring alternative therapeutic strategies for human patients.

CD4 T cells contribute to the development of alopecia areata through the production of IFN-γ.

INTRODUCTION

Alopecia areata (AA) is an autoimmune disease of the hair follicle with a lifetime incidence rate of ~2% (1–3). Clinical presentation of AA can range from small, localized patches of hair loss to widespread hair loss of the whole body (4, 5). Recently, the Food and Drug Administration approved the use of Janus kinase (JAK) inhibitors for the treatment of AA. However, this family of drugs has shown incomplete efficacy in clinical trials, and severe side effects including opportunistic infections, cancer, and thrombosis have been described with the use of JAK inhibitors (6–10). Further, following discontinuation of treatment, many patients often undergo a relapse of disease and lose hair. These caveats support the need to continue the investigation into the immune mechanisms involved in AA.

The hair follicle is considered an immune privileged (IP) site. Contributing factors to this state include minimal expression of major histocompatibility complex (MHC) molecules in the inferior portion of the follicle, a lack of immune cell infiltration, and the presence and action of immunoregulatory factors (11). AA is hypothesized to occur due to the collapse of IP of the hair follicle, indicated by an up-regulation of the expression of MHC molecules and danger signals involving the epithelium of the hair follicle. This loss of IP is thought to be followed by infiltration into the hair follicle environment of autoreactive immune cells, which coordinate and mediate attack of the hair follicle, resulting in nonscarring hair loss (11–13). The immune cell infiltrate is largely composed of T cells, including both CD4⁺ and CD8⁺ T cells, which produce the key effector molecule interferon-γ (IFN-γ). IFN-γ has been implicated in the loss of hair follicle IP (14) and shown to be critical in the development of AA in mouse models (15, 16). Prior work has highlighted the pathogenic role of cytotoxic CD8 T cells expressing NKG2D in AA pathogenesis (16, 17). Examination of lesional scalp tissue from patients with AA have demonstrated that CD4 T cells outnumber CD8 T cells in the T cell infiltrate (18–20); however, the functional contributions of CD4 T cells to the development of AA are not well understood.

Prior work using the C3H/HeJ AA mouse model indicated that transfer of cells derived from the skin-draining lymph nodes (SDLNs) of AA mice is sufficient to induce disease in recipient mice (21, 22); subsequent studies supported that sorted CD8 T cells are able to transfer disease (16, 23, 24). CD4 T cells from AA donor mice have been shown to induce disease in recipient mice when transferred subcutaneously (21). However, the specific contributions of CD4 T cells in the pathogenesis of AA have not yet been defined. CD4 T cell help is thought to be essential for CD8 T cell responses by supporting their expansion, activation, adoption of effector functions, formation of memory populations, and robust secondary responses following antigen restimulation (25–30). Given the presence of CD4 T cells in AA lesional patient samples and their potential to induce disease in the murine model, our aim was to identify the mechanisms used and factors required for CD4 T cell–mediated effects in AA.

Here, we demonstrate that CD4 T cells from AA mice have a robust ability to induce disease when adoptively transferred into recipient mice. Focused examination of CD4 T cells in AA SDLNs revealed that this population exhibited a heightened activated T helper type 1 (T_H1) effector gene profile when compared to CD4 T cells from unaffected (UA) mice. In addition, CD4 T cell–mediated disease induction was dependent on the presence of endogenous CD8 T cells and host IFN-γ signaling. Single-cell RNA sequencing (scRNA-seq) data of human scalp tissue revealed that CD4 T cells in patients with AA had an analogous, T_H1 gene–enriched profile. Disruption of CD4 T cell–mediated production of IFN-γ abolished the ability of CD4 T cells to transfer disease. In sum, we found that T_H1-associated effector T cells serve a pivotal role in the development of AA.

RESULTS

SDLNs from AA mice contain CD8 and CD4 T cells that can contribute to disease induction

It has been previously demonstrated that lymphocytes from the SDLN of mice with AA induce disease in adoptive transfer recipient mice (21, 22). We made use of a model in which lymphocytes from AA SDLNs that underwent in vitro expansion (22) induced disease when adoptively transferred into recipient mice (Fig. 1A). We were interested in defining the composition of the transferred population to gain insights into which specific components may be needed to induce disease. Flow cytometric analysis showed that the cultured cell population was composed of both CD4⁺ and CD8⁺ T cells (Fig. 1B), supporting a potential role for either cell type in the induction of AA. An examination of human biopsy specimens demonstrated a robust population of T cells around the hair follicle in patients with AA that was not observed in healthy control samples (Fig. 1C). In addition, the T cell population in AA skin was found to be composed of predominantly CD4⁺ T cells as well as a smaller population of CD8⁺ T cells (Fig. 1D), similar to what others have observed (19, 20). We then performed a transcriptomic analysis of skin samples from healthy control and patients with AA. We generated a robust scRNA-seq dataset by compiling data from multiple studies (31–33), which were then integrated in a manner to mitigate batch effects using established protocols. The dataset derived from 11 AA patient samples and 9 healthy control samples consisting of 86,110 total cells, composed of 48,197 from patients with AA and 37,913 from control patients. Uniform manifold approximation and projection (UMAP) dimensionality reduction resulted in the identification of clusters representing various immune cell populations [CD4 T cells, CD8 T cells, natural killer (NK)/γδ T cells, myeloid cells, dendritic cells, B cells, and mast cells], skin cell populations (keratinocytes, fibroblasts, and melanocytes), and other various cell types (smooth muscle cells, vascular endothelial cells, lymphatic endothelial cells, and mitotic) (Fig. 1E and fig. S1). We performed differential abundance testing using MiloR to interrogate differences in cellular abundance for the broad cell types seen in AA versus control skin (34). MiloR was applied to our harmonized dataset, and, using a 10% false discovery rate (FDR), we identified 5914 neighborhoods, with 942 showing an increased abundance in AA skin (shown in red) (Fig. 1F). We observed significant enrichment of multiple cell types in AA samples including both CD8 T cells and CD4 T cells (Fig. 1G). Together, these data support that both CD8 and CD4 T cells participate in the immune infiltration of lesional AA skin, suggesting that both T cell populations could have a role in the development of disease.

Fig. 1. — (A) Kaplan-Meier disease curve following SDLN cell transfer from AA and UA donors (n = 5). **P < 0.01, log-rank test. (B) Representative flow cytometry of CD4 and CD8 expression in AA SDLNs following 6 days of in vitro expansion (gated on live, CD3⁺ cells). (C) Immunohistochemical staining of CD3 expression in human scalp tissue from control and AA samples. Scale bars, 100 μm. (D) Proportion of CD4 and CD8 T cells in CD3⁺ cells from AA human scalp tissue by flow cytometry. **P < 0.01, nonparametric paired t test. (E) UMAP of combined scRNA-seq from human AA and control skin. (F) Neighborhood (Nhood) graph from Milo differential abundance testing. Node layout is based on UMAP in (E). Nodes represent neighborhoods, colored by log fold change (FC; FDR 10%). (G) Beeswarm plot of the log-transformed FCs in abundance of cells in AA and control skin. Colored nodes denote significant neighborhoods (FDR, 10%). (H to K) Expanded CD8 T cells from SDLNs of UA and AA donor mice were transferred into recipient mice (n = 8). (H) Representative ventral images of recipient mice and (I) Kaplan-Meier disease curve. ****P < 0.0001, log-rank test. (J) Representative flow cytometry and (K) frequency of NKG2D⁺CD8⁺ T cells in the SDLNs (mean ± SD). ***P < 0.001, Mann-Whitney test. (L to O) Expanded CD4 T cells from SDLNs of UA and AA donor mice were transferred into recipient mice (n = 6 UA, n = 5 AA). (L) Representative images of ventral surface of recipient mice and (M) Kaplan-Meier disease curve. **P < 0.01, log-rank test. (N) Representative flow cytometry and (O) frequency of NKG2D⁺CD8⁺ T cells in the SDLNs (mean ± SD). *P < 0.05, Mann-Whitney test. Mice were observed twice weekly for hair loss. Data are representative of at least two independent experiments.

We next investigated the potential of CD8 and CD4 T cells to contribute to the induction of disease in the mouse model. CD8 T cells were magnetically sorted from the SDLNs of UA and AA donor mice. After in vitro expansion and transfer to recipient mice, we found that CD8 T cells derived from AA mice induced disease, whereas CD8 T cells from previously UA mice were largely unable (Fig. 1, H and I, and fig. S2A). This is consistent with data from other groups that have shown transfer of CD8 T cells from the SDLNs of AA mice, but not from UA mice, can induce disease (16, 21, 24). Flow cytometric analysis of the SDLNs of these mice demonstrates that the CD8 T cell–mediated induction is associated with the emergence of NKG2D⁺CD8⁺ T cells (Fig. 1, J and K). We next focused on the inductive ability of CD4 T cells. CD4 T cells derived from AA mice had a robust capacity to transfer disease, whereas CD4 T cells derived from UA mice did not (Fig. 1, L and M, and fig. S2B). In addition, recipients that developed disease following the CD4 T cell transfer demonstrated the development of NKG2D⁺CD8⁺ T cells in the SDLNs (Fig. 1, N and O). We found that disease induction occurred at a faster rate when a larger number of CD8 T cells (fig. S3, A and B) or CD4 T cells (fig. S4, A and B) were transferred. Together, these results demonstrate that CD4 or CD8 T cells from AA mice each have the ability to induce AA.

AA inductive capacity is restricted to CD4 T cells from the SDLNs

Lymphadenopathy and splenomegaly are features seen in C3H/HeJ mice with AA, but it is unclear which populations are expanded in the setting of AA in the SDLNs and spleen (16). We found that mice with AA had an increased number of cells in the spleen (Fig. 2A). Further analysis demonstrated that there was an increase in the number of multiple immune cell populations including CD11c⁺ cells, αβ T cells, and B cells (fig. S5A). Within the αβ T cell compartment, we found an increase in the number of CD4 T cells, including both conventional CD4 T (Tconv) cells and regulatory CD4 T cells (fig. S5B).

Fig. 2. — (A) Total cell counts from spleens of UA (n = 5) and AA (n = 7) mice (mean ± SD). **P < 0.01, Mann-Whitney test. CD4 T cells were isolated from the spleens and SDLN of the same AA donor mice and transferred into recipient mice (n = 4). (B) Representative images of ventral surface of recipient mice. (C) Kaplan-Meier disease curve for mice. **P < 0.01, log-rank test. (D) Percent total body hair loss of mice following disease induction (mean ± SD). *P < 0.05, Mann-Whitney test. (E) Representative flow cytometric plots and (F) frequency of NKG2D⁺CD8⁺ T cells in SDLNs of recipient mice (mean ± SD). *P < 0.05, Mann-Whitney test. (G) Representative flow cytometric plots and (H) frequency of NKG2D⁺CD8⁺ T cells in skin of recipient mice (mean ± SD). *P < 0.05, Mann-Whitney test. Mice were observed twice weekly for hair loss. Data are representative of two independent experiments.

Our group and others have demonstrated that SDLN-derived T cells transfer disease into recipient mice (21, 22). However, it is not clear whether circulating T cells are capable of transferring disease. CD4 T cells from the SDLNs and spleens were sorted, expanded in vitro, and transferred into recipient mice. We found that SDLN-derived CD4 T cells readily transferred disease, whereas splenic-derived CD4 T cells from the same donors did not have the same capacity to transfer disease (Fig. 2, B to D). Mice that developed AA after SDLN-derived CD4 T cell transfer exhibited the development of NKG2D⁺CD8⁺ T cells in SDLNs (Fig. 2, E and F) and skin (Fig. 2, G and H). These data demonstrate that pathogenic CD4 T cells capable of inducing AA are highly enriched in the regional lymph nodes but not circulating systemically.

CD4 T cells in draining lymph nodes of AA mice exhibit an enriched T_H1 effector phenotype when compared to UA mice

To gain a better understanding of the molecular mechanisms that contribute to the pathogenicity of CD4 T cells in the SDLNs, CD4⁺ T cells were isolated from the SDLNs of UA and AA mice, and RNA-seq was performed. When evaluated through principal components analysis (PCA), we observed a distinction between the two groups of mice (Fig. 3A). We identified 124 genes that were significantly up-regulated in the AA CD4 T cells compared to UA CD4 T cells (Fig. 3B). Furthermore, we assessed our data for differences in gene expression across multiple immune-related pathways including effector molecules, transcription factors (TFs), cytokine signaling, and chemokine signaling. In CD4 T cells from AA mice, we observed an increase in expression of several genes related to activation including Cd44, Pdcd1, Icos, and Ctla4 (Fig. 3C). We also found several genes with increased expression relating to T_H1 differentiation and function including Stat4, Tbx21, Stat1, Ifng, Il18r1, Il12rb1, and Cxcr3 (Fig. 3C). We next used the decoupleR package to infer TF activity. We compared the inferred activity between UA and AA CD4 T cells (Fig. 3D) and found an enriched activity in TFs relating to proinflammatory signaling including NFKB1 and RELA, as well as T_H1 differentiation including TBX21 (T-bet) and SP1, a positive regulator of T-bet activity (35). On the basis of these findings, we next performed a gene set enrichment analysis (GSEA) of the genes with differential expression and observed a significant enrichment (enrichment score 1.81 and adjusted P < 0.001) of genes related to IFN-γ production (Fig. 3E), the defining characteristic of T_H1 cells. Together, these data provide evidence for an enhanced T_H1 response in the CD4 T cells of AA mice.

Fig. 3. — Bulk CD4 T cells were FACS (fluorescence-activated cell sorting) sorted (live, TCRβ⁺, and CD4⁺) from the SDLNs of age matched UA and AA mice (n = 4) and processed for RNA-seq. (A) PCA plot of UA and AA samples. (B) Volcano plot highlighting 124 up-regulated genes in AA CD4 T cells. Log₂FC > 1, −log₁₀ P > 5. (C) Heatmaps showing differential gene expression (DEG) by z-score for immune-related pathways. (D) Bar chart of top 25 differentially active TFs ranked by absolute score. Higher activity scores in AA CD4 T cells or UA CD4 T cells are indicated in red and blue, respectively. TFs related to proinflammatory signaling or T_H1 differentiation are highlighted in red. (E) GSEA of IFN-γ production pathway in AA CD4 T cells versus UA CD4 T cells. NES, normalized enrichment score; abs, absolute.

Given the enhanced T_H1 signature in AA CD4 T cells, we next sought to investigate the phenotypic differences between CD4 T cells of AA and UA mice. The SDLNs of AA mice exhibited an increase in the total number of CD4 T cells compared to UA mice (Fig. 4A).

AA mice showed an overall increase in the number of IFN-γ–producing CD4 T cells in the SDLNs when compared with UA mice (Fig. 4B), despite no change in their frequency among CD4 T cells compared to UA mice (fig. S6A). We found no differences in the total number or frequency of CD4 T cells producing interleukin-17 (IL-17A; Fig. 4B and fig. S6B), a T_H17 cytokine thought to play a pivotal role in other inflammatory skin diseases, such as psoriasis (36, 37). We used spectral flow cytometry to analyze the immunophenotype of the CD4 T cell populations and performed an unsupervised clustering approach, t-distributed stochastic neighbor embedding (tSNE), to compare the Tconv cell compartment among AA and UA mice (Fig. 4C and fig. S7). We found a minor but distinct population of cells with a greater density among the AA population (Fig. 4, C and D). Within that population, two markers of T cell activation, CD44 and CD11a, demonstrated the highest difference in expression, indicating that this population has been previously activated and has likely undergone antigen-mediated stimulation (red circles) (Fig. 4D). Using standard flow cytometry gating, we confirmed an increase in the expression of CD44 and CD11a among our Tconv populations in the SDLNs (Fig. 4, E and F). Furthermore, these markers define a heterogeneous population with varying expression of other markers of T cell activation, including CXCR3 and CD28 (Fig. 4D). To investigate whether this population may be responsible for the induction of disease, we sorted this population based on distinctive markers (CD25⁻CD44⁺CD62L⁻) and as well as a control population (CD25⁻CD44⁻CD62L⁺) of Tconv cells from the same AA mice (Fig. 4G). We found that this effector phenotype CD4 T cell population demonstrated greater potential to produce IFN-γ following in vitro expansion and stimulation (Fig. 4H). Following transfer, the effector phenotype CD4 T cells were able to induce disease in our recipient mice, whereas the control CD4 T cells were unable (Fig. 4I). Collectively, these data indicate that the CD4 T cells of AA mice exhibit a T_H1 T effector transcriptional and phenotypic profile and that these activated cells have the capacity to induce AA.

CD4 T cell–mediated induction of AA requires endogenous CD8 T cells

CD8 T cells have been shown to play an important role in the development of AA (16, 24, 32), and our data indicate that NKG2D⁺ CD8 T cell effectors emerge in mice with CD4 T cell–initiated disease. We sought to determine whether CD4 T cells were required for CD8 T cell–mediated AA induction and, conversely, whether CD8 T cells were required for CD4 T cell–mediated AA. To address this, we magnetically enriched CD8 T cells from AA SDLNs before their transfer into recipient mice. One group of recipient mice were depleted of CD4 T cells by antibody administration and compared to recipient mice that were given an isotype control antibody (Fig. 5A). We found that mice treated with a CD4-targeting depleting antibody went on to develop AA similarly to those that were injected with an isotype control antibody (Fig. 5, B to D). We next magnetically sorted CD4 T cells from AA SDLNs and transferred this population into recipient mice to induce AA. The endogenous CD8 T cells were depleted from one group of mice using an anti-CD8β antibody, while the other group received an isotype control antibody (Fig. 5E). We found that mice lacking an endogenous CD8 T cell population did not develop disease (Fig. 5, F to H). These results indicate that CD4 T cells require the presence of CD8 T cells to induce the development of AA. Therefore, in CD4 T cell–mediated AA, the actions of CD4 T cells likely work upstream of, or are dependent on, the pathogenic actions of CD8 T cells.

Fig. 5. — (A to D) In vitro–activated CD8 T cells isolated from the SDLNs of AA mice were intradermally injected into recipient mice receiving 400 μg of intraperitoneal injections of isotype control or αCD4 antibodies (n = 14 isotype control, n = 15 αCD4). Mice received antibody treatment twice weekly for the duration of the experiment. (A) Schematic showing experimental design. (B) Representative images of ventral surface of mice in each group. (C) Kaplan-Meier disease curve for development of hair loss. Log-rank test. (D) Percent total body hair loss of mice following disease induction (mean ± SD). Mann-Whitney test. (E to H) In vitro–activated CD4 T cells isolated from the SDLNs of AA mice were intradermally injected into recipient mice receiving 400 μg of intraperitoneal injections of isotype control or αCD8β antibodies (n = 10). Mice were injected twice weekly for the duration of the experiment. (E) Schematic showing experimental design. (F) Representative images of ventral surface of mice in each group. (G) Kaplan-Meier disease curve showing development of hair loss. **P < 0.01, log-rank test. (H) Percent total body hair loss of mice following disease induction (mean ± SD). **P < 0.01, Mann-Whitney test. Mice were observed twice weekly for hair loss. Data represent two to three combined experiments.

Host IFN-γ signaling is required for the induction of AA by CD4 T cells

IFN-γ is known to be important in the development of AA both by promoting MHC class I expression in the follicular epithelium and by inducing the production of chemokines that are important in the migration of T cells (14, 16, 38). We sought to determine whether endogenous response to IFN-γ is required for CD4 T cell–mediated AA induction. Expanded CD4 T cells from wild-type (WT) AA mice were transferred into either WT or IFN-γR^−/− host mice. We found that IFN-γR^−/− mice were fully prevented from developing disease (Fig. 6, A and B), despite no gross baseline deficiency in CD4 or CD8 T cells (fig. S8A). We did not observe infiltration of CD8 T cells or up-regulation of MHC class I throughout the hair follicles in IFN-γR^−/− mice (Fig. 6C), and no aberrations were seen at the baseline in the absence of adoptive transfer (fig. S8B). In addition, we found NKG2D-expressing CD8 T cells in the SDLNs (Fig. 6, D and E) and the skin (Fig. 6, F and G) of WT mice that went on to develop AA, which were largely absent in IFN-γR^−/− mice. Together, these data demonstrate that host IFN-γ signaling is important for the collapse of immune privilege, the development of NKG2D⁺ CD8 T cells, and the development of disease in CD4 T cell–mediated AA induction.

Fig. 6. — CD4 T cells were isolated from the SDLNs of WT AA donor mice and underwent in vitro expansion. Activated CD4 T cells were intradermally injected into WT or IFN-γ receptor knockout (IFN-γR^−/−) mice (n = 9). (A) Representative ventral images of mice. (B) Kaplan-Meier disease curve. ***P < 0.001, log-rank test. (C) Representative immunofluorescent images showing skin sections from WT and IFN-γR^−/− mice, stained with anti-CD8α and anti-MHC class I antibodies. Scale bar, 100 μm. (D) Representative flow cytometric analysis and (E) bar graph of NKG2D⁺ CD8 T cells in the SDLNs (mean ± SD). *P < 0.05, Mann-Whitney test. (F) Representative flow cytometric analysis and (G) bar graph of NKG2D⁺ CD8 T cells in the skin (mean ± SD). *P < 0.05, Mann-Whitney test. Mice were observed twice weekly for hair loss. Data represent two combined experiments.

Effector CD4 T cells share similar transcriptional characteristics in murine and human AA

Given that the CD4 T cells of AA mice exhibit an effector T_H1 phenotype, we sought to confirm this finding in human patients. We assembled and interrogated scRNA-seq data, performed an unsupervised reclustering of the T and NK cell cluster, and identified four cell populations (CD8 effector T cells, CD4 effector T cells, CD4 regulatory T cells, and NK/γδ T cells) (Fig. 7A and fig. S9) (31–33). We observed that all four of these populations were more highly represented in AA samples when compared to the healthy control samples. Focusing on the CD4 effector T cell cluster, we performed differential gene expression (DEG) analysis and found an up-regulation of genes associated with immune cell regulation, antigen presentation, and inflammation (Fig. 7B). IFNG was found to be among the top differentially expressed genes in AA CD4 effector T cells (Fig. 7B). Gene set scoring using the UCell package and two separately generated T_H1-related datasets (GSE22886 and Yasumizu et al. (39)] demonstrated an increased T_H1 score among the CD4 effector T cells in AA samples compared to the control samples, suggesting that AA CD4 T cells exhibit a greater T_H1 signature (Fig. 7C) (39, 40). In addition, we performed a third gene set scoring analysis using a smaller, more specific list of genes that are often associated with T_H1 differentiation and function. We observed a similar increase in T_H1 gene scoring among the CD4 effector T cells in AA samples compared to control samples (Fig. 7D). Collectively, these data support that CD4 effector cells share similar T_H1 features among human samples and our mouse model.

Fig. 7. — (A) Reclustered UMAP of the CD4 effector T cell, CD8 effector T cell, CD4 regulatory T cell (T_reg), and NK/γδ T cell populations from the data in Fig. 1E. (B) Top 10 genes up-regulated in the AA-associated CD4 effector T cells as compared to control-associated cells. Table shows up-regulation of genes associated with immune cell regulation, antigen presentation, and inflammation. (C) Violin plots of T_H1 gene set scores as calculated by UCell using MsigDB GSE22886 and dataset from Yasumizu *et al.*, adapted from table S4 in (39). ****P < 0.0001, Student’s t test. MsigDB, Molecular Signatures Database. (D) Violin plot of T_H1 gene set scores using a defined set of genes. Heatmap shows mean normalized expression for genes included in gene set. ****P < 0.0001, Student’s t test.

CD4 T cell–derived IFN-γ is responsible for CD4 T cell–mediated AA development

To specifically investigate the role of IFN-γ produced from CD4 T cells in AA development, we generated CD4CreERT2-Ifng^flox/flox C3H/HeJ mice (IFN-γ^ΔCD4), whereby IFN-γ can be efficiently deleted via a tamoxifen-inducible system (Fig. 8A) (41). AA donor mice were generated through CD8 T cell transfers to ensure that the CD4 T cell population that we would later sort from the recipient mice was made up of the genotype of the recipient. IFN-γ^ΔCD4 and IFN-γ^WT (Ifng^flox/flox C3H/HeJ mice that did not harbor the Cre-ERT2 fusion transgene) mice that developed AA were treated with tamoxifen, and loss of IFN-γ production was confirmed for CD4 T cells from IFN-γ^ΔCD4 mice despite maintained IFN-γ production by CD8 T cells (Fig. 8B). Next, CD4 T cells were magnetically sorted from the SDLNs of IFN-γ^ΔCD4 and IFN-γ^WT mice with AA and expanded in vitro. We confirmed that IFN-γ^ΔCD4 T cells were largely devoid of IFN-γ production (Fig. 8C). To determine the importance of CD4 T cell–derived IFN-γ in AA development, we transferred IFN-γ^WT and IFN-γ^ΔCD4 T cells into separate groups of recipient mice. We observed that IFN-γ^ΔCD4 T cells were unable to induce disease in recipient mice, whereas IFN-γ^WT CD4 T cell recipients developed AA (Fig. 8, D and E). In addition, mice that received IFN-γ^WT CD4 T cells went on to develop NKG2D⁺CD8⁺ T cells in the SDLN (Fig. 8F), and this population was absent in IFN-γ^ΔCD4 CD4 T cell recipients. Together, these data demonstrate that IFN-γ produced by CD4 T cells is critical for CD4 T cell–mediated disease induction.

Fig. 8. — (A) Experimental schematic. AA donor mice (IFN-γ^WT and IFN-γ^ΔCD4) were treated with tamoxifen as indicated in Materials and Methods. Isolated CD4 T cells from the SDLNs were pooled and activated in vitro and transferred into recipient mice. (B) Peripheral blood leukocytes were collected from mice and stimulated for 4 hours with PMA and ionomycin to assess for IFN-γ production (n = 6) (mean ± SD). **P < 0.01, Mann-Whitney test. (C) IFN-γ production from pooled SDLN CD4 T cells was assessed pre– and post–in vitro culture. (D) Representative images of ventral surface of mice in each group. (E) Kaplan-Meier disease curve (n = 4 IFN-γ^WT and n = 6 IFN-γ^ΔCD4). *P < 0.05, log-rank test. (F) Representative flow cytometric plots and summary graphs of NKG2D⁺ CD8 T cells in the SDLNs of mice (mean ± SD). *P < 0.05, Mann-Whitney test. Mice were observed twice weekly for hair loss. Data are representative of two independent experiments.

DISCUSSION

AA is characterized as a T cell–mediated autoimmune disease of the hair follicle. Analysis of tissue samples from patients with AA has demonstrated an increased presence of both CD8 and CD4 T cells around the hair follicles (19, 20). Much of the previous work has primarily focused on CD8 T cells and defined them as a pivotal effector cell responsible for hair loss (16, 24, 32). However, AA lesions show that CD4 T cells numerically outnumber CD8 T cells (18–20), yet no role for CD4 T cells has been clearly defined. Here, we leveraged multiple techniques, including scRNA-seq, flow cytometry, and immunohistochemistry, to analyze tissue samples from patients with AA. We found that T cells infiltrated the hair follicle in AA samples and that the skin infiltrate was made up of proportionally more CD4 T cells compared to CD8 T cells. Further, we demonstrated that CD4 T cells from the SDLNs of AA mice are capable of inducing disease when adoptively transferred into recipient mice. We found that these cells express a T_H1 effector profile with increased expression of the effector molecules CD44 and CD11a, and this ability to induce disease was dependent on the production of IFN-γ from donor cells and IFN-γ receptor signaling within the host. Together, these data support a previously underappreciated role and capacity for disease induction for T_H1 CD4 T cells in the development of AA.

Previous studies have shown that SDLN-derived lymphocytes from AA mice could transfer disease (21, 22). Here, we similarly show that SDLN lymphocytes from AA mice transferred disease, in contrast to SDLN lymphocytes from UA mice. Our finding that both CD4 and CD8 T cells may be involved in this model led us to further investigate the ability of CD8 and CD4 T cells to induce disease. Our data showed that CD8 T cells from the SDLNs of AA mice could transfer disease that could affect the entire body surface area of recipient mice, whereas UA CD8 T cells did not, consistent with findings from other groups (16, 23, 24). A previous study by McElwee et al. (21) reported that CD8 T cell transfer led to only localized hair loss at the injection site, differing from our observation that CD8 T cells induced total body hair loss in recipient mice. Furthermore, we demonstrated that CD4 T cells from AA mice had a robust ability to also induce systemic disease, consistent with findings in the prior study (21). That study, however, observed disease development in 43% of recipient mice, whereas our study demonstrated disease development in 75 to 100% of mice; this difference could in part be accounted for by differences in the route of cell transfer. CD4 T cell recipient mice exhibited the emergence of NKG2D⁺CD8⁺ T cells in the SDLNs, a population of cells that is associated with disease pathogenesis (16, 17, 24). These data support that CD4 T cells have the capacity to initiate and contribute to the development of disease.

It was unclear, however, whether CD4 T cells capable of inducing AA were restricted to the SDLNs or may be found systemically. Splenomegaly has been previously described in AA mice (16). We found that the expanded number of cells in the spleens of AA mice is made up of an increase in multiple different immune cell populations, including αβ T cells. Within the T cell compartment, only CD4 T cells appeared to be expanded, leading us to question whether these CD4 T cells could also transfer disease. In other mouse models of peripheral autoimmunity, splenic-derived T cells have been used to induce disease in the target organs, including colitis, type 1 diabetes, and Sjögren’s disease (42, 43). In our model, splenic-derived CD4 T cells were not able to transfer disease similarly to SDLN-derived CD4 T cells, demonstrating that the splenic CD4 T cell pool does not harbor the same relative abundance of pathogenic CD4 T cell populations that are found in the lymph nodes.

Many studies have contributed to our understanding of the transcriptional and phenotypic profiles of CD8 T cells in AA, especially with regard to NKG2D-expressing CD8 T cells, which are thought to be primary effector cells in AA (16, 24, 31, 32, 44). Fewer studies, though, have explored these profiles in the CD4 T cell population, leaving a gap in our understanding of the role of CD4 T cells in AA (31, 32, 44). Using RNA-seq and spectral flow cytometry, we found that CD4 T cells in the SDLNs of AA mice exhibited a T_H1 effector profile distinct from CD4 T cells from UA mice. We used tSNE analysis of our spectral flow cytometric data as an approach to identifying markers expressed by the Tconv cells in AA. We found a population of cells that was more represented in the AA samples, marked by the expression of CD11a and CD44, indicative of a previously activated, effector-like phenotype. We found that effector CD4 T cells (CD25⁻CD44⁺CD62L⁻) were potent producers of IFN-γ and transferred disease.

A variety of disease models indicates that CD4 T cell helper functions are important for a robust and memory-generating CD8 T cell response. In a model using glia-tropic mouse hepatitis virus, depletion of CD4 T cells early in the immune response compromised CD8 T cell function and decreased viral clearance (45). In addition, other models of bacterial or viral infections have demonstrated that, during CD8 T cell priming, the absence of CD4 T cell help results in an inadequate expansion of the effector CD8 T cells either during the primary or subsequent challenges (26–28, 30, 46). These data support that CD4 T cell help is important in the early activation stages of a CD8 T cell response, which could be in part by licensing dendritic cells through the CD40-CD40L pathway (25, 47–49). Previously, using scRNA-seq, our group found that antigen-presenting cells in AA skin showed an increase gene signature for CD40 signaling, indicating one potential mechanism by how CD4 T cells could be contributing to AA (31). In addition, our prior scRNA-seq data demonstrated an enriched IL-2/signal transducer and activator of transcription 5 gene signature in AA mice (31), and targeting of IL-2 in vivo through genetic models with reduced IL-2 expression and antibody-mediated neutralization inhibited the development of AA (16, 50). Activated CD4 T cells are a source of IL-2 which has been shown to help promote CD8 T cell expansion and function (51, 52) and may therefore be another potential mechanism by which CD4 T cell help could be contributing to AA development. Together, these data led us to question how CD4 and CD8 T cells cooperate in AA and what role CD4 T cells may have in helping in the induction of NKG2D-expressing CD8 T cells. We found that CD8 T cells can transfer disease whether or not CD4 T cells are present. However, CD4 T cells lacked the ability to transfer disease if CD8 T cells were not present. This suggests that, in AA development, CD4 T cells may have a role in the priming of CD8 T cell responses that are needed for disease to emerge. A prior study showed that, in patients in an acute phase of disease (reported as less than 6-month duration and between 1 and 20 lesions), there were more CD4 T cells present in the skin compared to CD8 T cells, potentially supporting a role for CD4 T cell involvement during early stages of disease (20). Recently, a study that used the skin graft model of AA demonstrated that depleting CD4 T cells lead to a delay in AA induction (32, 53). This study further examined whether CD4 depletion would have a therapeutic effect in mice with established disease. This recent study found that depletion of CD4 T cells was unable to successfully reverse hair loss in mice with established disease; these results are in line with our data but contrasts with prior published data using C3H/HeJ mice with spontaneous hair loss that found modest hair regrowth in mice after CD4 depletion (54). Furthermore, it has been demonstrated that depletion of CD8 T cells protects AA skin graft recipients from developing disease (32). Together, these data offer support for the idea that CD4 T cells are playing a role in the activation of CD8 T cells in AA and invite future studies to help dissect what specific mechanisms CD4 T cells use, potentially including the CD40-CD40L pathway or IL-2 production.

IFN-γ has been implicated as a driver of the collapse of IP of the hair follicle and a critical cytokine in AA pathogenesis (14, 16). Furthermore, IFN-γ stimulation of the hair follicles induced the expression of the proinflammatory chemokines CXCL9 and CXCL10, which have been shown to be important in the migration of T cells to the hair follicle and in AA development (38). Previous studies have specifically targeted IFN-γ to investigate the role in AA induction; IFN-γ–deficient mice were protected from developing disease using the skin graft model of AA (15), and, in another study, antibody-mediated neutralization of IFN-γ blocked AA induction in AA skin graft–recipient mice (16). We investigated the role of IFN-γ signaling in mice following CD4 T cell–mediated induction using mice exhibiting genetic ablation of the IFN-γR.

Before cell transfer, IFN-γR^−/− mice demonstrated an increased number of T cells in the SDLNs compared to control mice but showed no differences in the CD8 T cell infiltration or MHC class I expression in the skin. Following the transfer of WT CD4 T cells, IFN-γR^−/− mice were protected from developing AA. Our data demonstrate that host responsiveness to IFN-γ is critical to disease development following CD4 T cell–mediated AA induction. Together, these data implicate IFN-γ as an essential cytokine in AA pathogenesis in a variety of models. It is still unclear, however, which cell populations are responding to IFN-γ. The hair follicle expresses the IFN-γ receptor and, following stimulation with IFN-γ, up-regulates MHC class I molecules, a hallmark of the IP collapse seen in AA (14, 55, 56). Furthermore, CD8 T cells express the IFN-γ receptor, and signaling through this pathway is important for a variety of cellular functions. Following lymphocytic choriomeningitis viral infection, IFN-γ receptor–deficient CD8 T cells showed a reduced expansion as compared to IFN-γ receptor–sufficient cells (29). In another study of graft rejection, using intravital imaging, CD8 T cells stimulated with IFN-γ moved faster through the skin and traveled a greater distance (57). In addition, these cells were also more cytotoxic when challenged with keratinocytes loaded with relevant peptides. These data support that the hair follicle, CD8 T cells, or both may be responding to IFN-γ in AA to engender disease development.

We more closely investigated the CD4 effector T cell population in the skin of human patients using scRNA-seq. Our data indicate CD4 effector T cells are more highly represented in AA samples when compared to healthy control samples, and these cells express a T_H1 transcriptional signature, with IFNG as one of the top differentially expressed genes in CD4 T cells from AA lesions. A recent study of murine skin found that the Ifng gene is also among the top differentially expressed genes in AA CD4 T cells (32). These data, along with our mouse SDLN RNA-seq data, suggest that IFN-γ may be a critical part of the involvement of CD4 T cells in AA. Although CD4⁺ T cells are more prominent than CD8⁺ T cells in human AA lesions, the opposite is often observed in the C3H/HeJ mouse model (32). However, despite this difference in cellular composition, the shared T_H1-like profile across species and tissues supports a translationally relevant role for T_H1 cells in AA pathogenesis. T_H1 cells have often been associated with development of other autoimmune diseases and models, including Sjögren’s disease, colitis, and experimental autoimmune encephalomyelitis (58–60). In a recent study of Sjögren’s disease, deletion of IFN-γ production from CD4 T cells eliminated their ability to induce glandular disease (58). In AA, we hypothesized that CD4 T cell–derived IFN-γ may be an essential function of CD4 T cells in inducing disease. Deleting IFN-γ production from CD4 T cells eliminated their ability to transfer disease. These data are supported in part by our earlier findings (Fig. 4H), indicating that effector T cells produced greater amounts of IFN-γ compared to their naïve counterparts. Together, these data support that T_H1 cells can contribute to the development of AA.

Overall, our study describes a pivotal role for CD4 T cells in AA. We found that CD4 T cells from AA mice have a distinct T_H1 transcriptional and phenotypic profile, which is indicative that a specific population of cells can induce disease. These data support further investigation into CD4 T cells in human patients to better understand the characteristics that may help define populations important in disease pathogenesis. In addition, our data demonstrate the potential of CD4 T cells to be important in the early stages of disease development, prompting future studies to better understand the kinetics of CD4 T cells in AA and define potential mechanisms that may be useful for therapeutic intervention.

MATERIALS AND METHODS

Mice

Female C3H/HeJ mice were purchased from the Jackson Laboratory (stock #000659). Mice were kept in pathogen-free conditions at the University of Iowa Animal Facility. IFN-γR^−/− mice on the C3HeB/FeJ background were a gift from the Goverman laboratory (Department of Immunology, University of Washington) and were backcrossed for seven generations onto the C3H/HeJ background (61). B6(129X1)-Tg(Cd4-cre/ERT2)11Gnri/J (CD4CreERT2) mice were purchased from the Jackson Laboratory (stock #022356) and backcrossed onto the C3H/HeJ background for at least 11 generations. Ifng^flox/flox mice on the C57BL/6 background were a gift from the Harty laboratory (Department of Pathology, University of Iowa) and were backcrossed for 10 generations onto the C3H/HeJ background (41). CD4CreERT2 and Ifng^flox/flox mice were bred to produce CD4CreERT2-Ifng^flox/flox mice, and Cre genotypes were confirmed by polymerase chain reaction. All animal procedures were approved (protocol #2101942) and done in accordance with the University of Iowa Institutional Animal Care and Use Committee (IACUC) and the Department of Veteran’s Affairs IACUC.

Study participants

Healthy control patients and patients with AA were recruited from the Department of Dermatology at University of Iowa Health Care, and written informed consent was obtained in compliance with the guidelines of the institutional review board–approved protocol #201706728 and the Declaration of Helsinki.

T cell isolation for in vitro culture and induction of AA

SDLNs (inguinal, brachial, axillary, and cervical) and whole spleens were isolated from UA and/or AA-affected mice. Single-cell suspensions were made by mechanically dissociating tissue through a 70-μm cell strainer. Splenocytes were depleted of erythrocytes by RBC Lysis Buffer (Thermo Fisher Scientific, 00-4333-57).

From the single-cell suspensions, CD4 or CD8 T cells were isolated using magnetic bead separation per the manufacturer’s instructions [magnetic-activated cell sorting (MACS), Miltenyi Biotec, CD4 (L3T4) 130-117-043 and CD8α (Ly-2) 130-117-044]. Enriched T cells were then placed into culture in complete RPMI 1640 media containing 10% fetal bovine serum (FBS) (Thermo Fisher Scientific), 2 mM l-glutamine (Thermo Fisher Scientific), 10 mM Hepes (Thermo Fisher Scientific), 1× antibiotic-antimycotic [penicillin (100 U/ml), streptomycin (100 μg/ml), and amphotericin B (250 ng/ml)] (Thermo Fisher Scientific), 1 mM Na pyruvate (Thermo Fisher Scientific), and 50 μm 2-mercaptoethanol (Avantor). Culture media was supplemented with IL-2 (30 U/ml), IL-7 (12.5 ng/ml), IL-15 (25 ng/ml) (Cytek, BioLegend, Cell Signaling, and Thermo Fisher Scientific), and mouse T-activator CD3/CD28 Dynabeads (Thermo Fisher Scientific). Cells were activated and expanded for 6 to 8 days at 37°C and 5% CO₂. In some experiments, bulk SDLN cells were placed into culture in complete RPMI media containing 10% FBS, 2 mM GlutaMAX (Thermo Fisher Scientific), and penicillin-streptomycin (100 U/ml; Thermo Fisher Scientific). Cells were expanded for 6 to 8 days followed by magnetic enrichment of CD4 T cells by MACS separation. Purity of T cells was assessed postculture, and cultures with greater than 98% purity were used.

To induce AA in C3H/HeJ mice, cells were collected, and CD3/CD28 Dynabeads were magnetically separated. The cells were then washed in sterile phosphate-buffered saline (PBS) and resuspended to the desired concentration in sterile saline or PBS. Cells were transferred intradermally into the caudal dorsal skin of 10- to 14-week-old UA C3H/HeJ mice (22). For experiments using CD8 T cells, cells were injected between 1.11 and 10 × 10⁶ per mouse, and for experiments using bulk CD4 T cells, cells were injected between 3.33 and 30 × 10⁶ per mouse. For experiments using naïve and effector CD4 T cells, cells were injected at 1.5 × 10⁶ per mouse. Mice were checked twice weekly for the development of AA on the ventral surface. Hair loss was quantified using Fiji software by calculating percent total body hair loss and subtracting percent areas of partial or total body hair loss.

In vivo depleting antibodies

For in vivo depletion studies, antibody treatments were started 1 day before the transfer of lymphocytes. Anti-CD4 (BioXCell, GK1.5, BE0003), anti-CD8β (Lyt 3.2) (BioXCell, 53-5.8, BE0223), or isotype control immunoglobulin G (BioXCell) was administered by intraperitoneal injection (400 μg) two times weekly for the duration of the experiments.

Preparation of tissue cell suspensions

SDLNs or whole spleens were mechanically dissociated and filtered through a 70-μm cell strainer. Splenocytes were depleted of erythrocytes by RBC Lysis Buffer (Thermo Fisher Scientific) and washed in PBS. To prepare skin single-cell suspension, hair was removed from the mice using hair clippers. Skin was collected, defatted, and finely minced using scissors and digested for 60 min at 37°C with collagenase IV (1.5 mg/ml; Worthington) and deoxyribonuclease I (DNase I) (0.025 mg/ml; Roche) in RPMI 1640 with 3% FBS in a shaker. Digested skin was filtered through a 70-μm cell strainer and washed twice with cold RPMI 1640 with 10% FBS before staining.

Human skin samples were collected, defatted, and finely minced using scissors and digested for 60 min at 37°C with Liberase TL (350 mg/ml; Sigma-Aldrich) and DNase I (75 mg/ml; Worthington). Digested samples were then mechanically dissociated and filtered through a 70-μm cell strainer and washed twice with PBS containing 2% FBS before staining.

Flow cytometry and fluorescence-activated cell sorting

Single-cell suspensions were stained with Live Dead Blue or Zombie ultraviolet (Thermo Fisher Scientific and BioLegend) for 30 min at 4°C before surface markers. Surface proteins were then labeled with fluorophore-conjugated antibodies (tables S1 and S2) for 20 to 30 min at 4°C in PBS with 2% FBS. In some experiments, Ghost Dye antibodies (Cytek Biosciences) were included into the surface marker cocktail to detect viable cells. For intracellular staining of TFs and/or cytokines, cells were fixed with the FoxP3/Transcription Factor Staining Buffer Kit (Cytek Biosciences, TNB-0607) according to the manufacturer’s instructions and then incubated with fluorophore-conjugated antibodies. For cytokine staining, cells were stimulated for 4 to 5 hours with phorbol 12-myristate 13-acetate (PMA) and ionomycin (BioLegend, 423302) with brefeldin A (BioLegend, 420601), a nonspecific activator of T cells that bypasses the T cell receptor, and washed two times in PBS with 2% FBS before staining. Flow cytometry data were acquired using a BD LSRII (BD Biosciences) or Cytek Aurora (Cytek Biosciences) and analyzed with FlowJo V10 software (Treestar).

For cell sorting, cells were incubated with fluorophore-conjugated antibodies in PBS with 2% FBS for 30 min at 4°C. The samples were then washed and resuspended in PBS supplemented with 2% FBS and 10 mM Hepes (Thermo Fisher Scientific). Cells were sorted on either the BD Aria Fusions sorter (Becton Dickinson) or the Cytek Aurora CS (Cytek Biosciences). Viable cells were gated on forward and side scatter and Hoechst 33458 staining.

FlowJo software was used for tSNE analysis. For the analysis, cells were gated on forward and side scatter (FSC-A/SSC-A), single cells (FSC-H/FSC-A), live, CD3⁺, CD4⁺, and FoxP3⁻ to get Tconv cells (fig. S7). Tconv cells were down sampled to 10,000 events, concatenated from all samples to perform tSNE analysis (62, 63), and then separated by disease state. Heatmaps of individual markers were assessed on the total concatenated sample.

Immunostaining

Mouse skin was embedded in optimal cutting temperature compound, and 30-μm sections were cut. Sections were fixed in cold (−20°C) acetone for 10 min, washed with wash buffer (1× tris-buffered saline containing Ca and 0.09% NaN₃) for 10 min, and blocked with wash buffer containing 5% bovine serum albumin [Research Products International (RPI)], and 5% goat serum (Sigma-Aldrich) for 60 min at 37°C. After blocking, the sections were stained with rat anti-CD8α (Clone 53–6.7, Thermo Fisher Scientific, 14–0081-85) and biotin mouse anti-H2K^k (clone 36-7-5, BioLegend, 114903) overnight at 4°C at a 1:100 dilution in blocking buffer. Following the primary antibody, sections were stained with fluorescently tagged secondary antibodies, anti-rat Alexa Fluor 568 (1:600; Thermo Fisher Scientific, A-11077) and Alexa Fluor 488 anti-biotin (1:400; Jackson ImmunoResearch, 200-542-211). For detection of nuclei, sections were stained using 4′,6-diamidino-2-phenylindole (DAPI; 1 μg/ml; Cell Signaling, 4083) for 5 min at room temperature. Antifade mountant with DAPI (Thermo Fisher Scientific, P36962) was used as the mounting medium. Immunofluorescent images were captured on a Zeiss LSM 880 confocal microscope, and images were compiled using Fiji.

Human scalp tissue was fixed in 10% neutral buffered formalin and embedded in paraffin. Paraffin sections were cut at 5 μm, and CD3 staining was performed by the University of Iowa Comparative Pathology Laboratory. Images were captured using an Olympus BX63 microscope.

Tamoxifen treatment

Tamoxifen (Sigma-Aldrich, T5648) was suspended in corn oil and 10% ethanol, and treatments were performed by intraperitoneal injection. Three to four doses of 4 mg were administered every other day into Cre-positive (IFN-γ^ΔCD4) and Cre-negative (IFN-γ^WT) mice. Excision of IFN-γ in the CD4 T cells was confirmed by intracellular cytokine staining of peripheral blood after PMA/ionomycin stimulation before collection of SDLNs. Production of IFN-γ from CD8 T cells was also examined to confirm CD4 T cell–specific deletion. Deletion of IFN-γ was also confirmed in SDLNs before and after the in vitro expansion of the CD4 T cells.

Bulk RNA-seq

Mice were induced to develop AA by full-thickness skin grafts (53) and allowed to develop disease. For cell isolation, SDLNs were removed from aged-matched AA and UA mice. Single-cell suspensions were made, and live TCRβ⁺CD4⁺ T cells were sorted using a FACS Aria Fusion (Becton Dickinson). Cells were washed and frozen before RNA isolation. Libraries were generated using the TruSeq Stranded mRNA kit (Illumina, 20020595), following the manufacturer’s protocol with 100 ng of RNA input. Libraries were analyzed for quality using Qubit (Thermo Fisher Scientific, Q32854) and size distribution using Agilent TapeStation. Samples were pooled and sequenced on NovaSeq S4 200 cycle flowcell (Illumina) according to the manufacturer’s recommendations to achieve 20 million paired reads per sample at MedGenome Inc. Reads were aligned to the reference genome using the Spliced Transcripts Alignment to a Reference (STAR) aligner (version 2.7.10a), and a genome index was generated using the GRCm39 primary assembly (FASTA) and the corresponding GENCODE vM30 annotation [gene transfer format (GTF)]. Gene-level counts were quantified during alignment using the --quantMode GeneCounts option in STAR. Gene counts were imported into R (version 4.4) and consolidated into a single expression matrix for all eight samples. Lowly expressed genes were filtered out by retaining genes with a total count greater than 30 across all samples and fewer than 6 zero-count samples. Transcript-to-gene mappings were performed using the GENCODE vM30 annotation file (tx2gene.gencode.vM30.csv), and gene-level counts were aggregated by summing counts across transcript isoforms. Differential expression analysis was performed using the DESeq2 package (version 1.46.0). A DESeqDataSet object was created from the filtered count matrix and sample metadata, and the DESeq function was applied to normalize counts and fit a negative binomial model. Variance-stabilizing transformation (VST) was applied to the normalized counts for visualization and downstream analyses. PCA was performed on the VST-transformed counts using the top 500 most variable genes, and 95% confidence ellipses were added to visualize group separation using the ggplot2 package. A volcano plot was generated to visualize DEG using the EnhancedVolcano package. Genes with significant differential expression [adjusted P < −log₁₀(5) and |log₂ fold change (FC)| > 1] were highlighted in red. Heatmaps were generated for selected marker genes and grouped by functional categories. Expression levels for each gene were extracted from the VST-transformed matrix, centered, and scaled across all samples. Rows and columns were clustered on the basis of the scaled expression values (z-score), and boxes were colored according to this z-score. Plots were generated using the ggplot2 package. TF activity was inferred using the decoupleR package and the CollecTRI network. Significant differentially expressed genes (adjusted P < 0.05) between the AA and UA were used as input, with the stat value from DESeq2 differential expression analysis serving as the weight. The top 25 differentially inferred TFs were extracted for visualization. GSEA was performed using the clusterProfiler package on the differentially expressed genes. The Gene Ontology database GO:0032609 (IFN-γ production) was used for analysis.

Human scRNA-seq data analysis

Previously published human scRNA-seq data were acquired from the Gene Expression Omnibus (GEO) database using accessions GSE145095 (31), GSE233906 (32), and GSE212450 (33). For each dataset, the filtered barcode, feature, and matrix files were downloaded and loaded into R. scDblFinder was used to remove doublets from each sample before being merged into a single Seurat object (min.cells = 5, min.features = 500). This Seurat object was filtered to include percent.mt < 20, nCount_RNA < 60000, nCount_RNA > 800, and db.class == “singlet.” Variable features were identified using default parameters, and any mitochondrial, ribosomal, cell cycle, T cell receptor, B cell receptor, and other select genes that have been noted to cause issue were removed. The data were normalized by log transformation and scaled before being integrated using PCA and the Harmony package. The harmonized data were used to generate a UMAP using the FindNeighbors, FindClusters, and RunUMAP functions. The UMAP was generated using a resolution of 0.3, and clusters were labeled on the basis of the expression profiles and DEG results. Differential abundance testing was performed using the miloR package. Three samples were excluded from this analysis due to immune cell flow sorting or poor cell acquisition (>90% keratinocyte abundance). The remaining samples were converted into a milo object and applied to the harmonized data. A graph was built, and neighborhoods were created using k = 20, d = 50, prop = 0.1, and refined = T. Differential testing was done using disease state (AA versus control), and the results were overlayed on the existing UMAP. The logFC abundances were colored for all neighborhoods (red = AA, blue = control) below an FDR of 10%. The neighborhoods were then annotated according to the most abundant cell type in each neighborhood, and the differential abundance testing results were plotted as a beeswarm plot, grouped by the cell types identified (FDR, 10%). To analyze the T cells, lymphoid cells were isolated from the dataset, processed similarly to the full dataset, and labeled according to the major subsets identified. The CD4 effector T cell population was further isolated, and DEG analysis was performed on this population, comparing AA versus control cells. Genes with a pct.1 greater than 0.1 and adjusted P value less than or equal to 0.05 were filtered out, and the data were sorted by log₂ FC and presented as the top 10 highest log₂ FC. The UCell package was used to score the normalized expression of the CD4 effector T cell populations using different T_H1 gene signature lists. Two of the gene signature lists were collected from (i) the GEO database using accession #GSE22886 (naïve CD4 T cell versus 48H Act T_H1 DN) and (ii) two nonnegative matrix factorization (NMF) populations expressing T_H1 gene signatures, NMF 0 and 11, from a prior study (39, 40). A selective gene list was generated using genes that have been used to define T_H1 cells in Fig. 7D. The UCell scores for each of the three gene signature lists were used to generate violin plots. A heatmap was made of the selective gene list showing the average expression of each gene compared between AA and control CD4 effector T cells colored according to the absolute expression level.

Statistical analysis

For biological (nonomics) analysis, data were analyzed using GraphPad Prism (version 10) software. Two groups were compared using an unpaired nonparametric t test (Mann-Whitney test) or a paired nonparametric t test (Wilcoxon signed-rank test). A Kruskal-Wallis test with Dunn’s multiple comparisons test was used for mean difference comparison from multiple groups. Log-rank (Mantel-Cox) tests were used to analyze the hair loss curve. Data are presented as means ± SD, and significance is indicated as follows: *P < 0.05, **P < 0.01, ***P < 0.001, and ****P < 0.0001. All experiments were repeated at least two times. All bioinformatic analyses were performed using R. For the scRNA-seq data, a Student’s t test was used to compare the CD4 effector T cells in AA versus control samples in Fig. 7.

Acknowledgments

The data presented herein were obtained at the Flow Cytometry Facility, which is a Carver College of Medicine/Holden Comprehensive Cancer Center core research facility at the University of Iowa. The facility is funded through user fees and the financial support of the Carver College of Medicine, Holden Comprehensive Cancer Center, and Iowa City Veteran’s Administration Medical Center. We would like to acknowledge use of the University of Iowa Central Microscopy Research Facility, a core resource supported by the University of Iowa Vice President for Research, and the Carver College of Medicine.

Funding:

The following grants are acknowledged for support of the project: National Institutes of Health grant K08AR069111 (A.J.), National Institutes of Health grant R01AR077194 (A.J.), National Institutes of Health grant T32AI007485 (S.J.C.), and Department of Veterans Affairs grant I01BX004907 (A.J., Biomedical Laboratory Research and Development). Research reported in this publication was supported by the National Center for Research Resources of the National Institutes of Health under award number 1 S10 OD034193-01.

Author contributions:

Conceptualization: S.J.C. and A.J. Methodology: S.J.C., S.C., R.R., J.M.G., and A.J. Investigation: S.J.C., S.C., R.R., M.M.L., P.K., N.H., O.A., Z.Z., L.S.O., A.C.K., and E.M.S. Formal analysis: S.J.C. and R.R. Visualization: S.J.C., R.R., and A.J. Validation: S.J.C., R.R., and A.J. Data curation: R.R. Software: R.R. Resources: R.R., A.C.K., E.M.S., K.D.C., J.T.H., J.M.G., and A.J. Funding acquisition: A.J. Supervision: A.J. Project administration: A.J. Writing—original draft: S.J.C. and A.J. Writing—review and editing: S.J.C., S.C., R.R., O.A., L.S.O., J.T.H., J.M.G., and A.J.

Competing interests:

The authors declare that they have no competing interests.

Data and materials availability:

All data needed to evaluate the conclusions in the paper are present in the paper and/or the Supplementary Materials. All RNA-seq data for SDLN CD4⁺ T cells have been deposited in NCBI’s GEO and are accessible through GEO accession number GSE295313.

Supplementary Materials

This PDF file includes:

Figs. S1 to S9

Tables S1 and S2

sciadv.adz2257_sm.pdf^{(1.5MB, pdf)}

REFERENCES AND NOTES

1.Mirzoyev S. A., Schrum A. G., Davis M. D. P., Torgerson R. R., Lifetime incidence risk of alopecia areata estimated at 2.1% by Rochester Epidemiology Project, 1990-2009. J. Invest. Dermatol. 134, 1141–1142 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Safavi K. H., Muller S. A., Suman V. J., Moshell A. N., Melton L. J. III, Incidence of alopecia areata in Olmsted County, Minnesota, 1975 through 1989. Mayo Clin. Proc. 70, 628–633 (1995). [DOI] [PubMed] [Google Scholar]
3.Mostaghimi A., Gao W., Ray M., Bartolome L., Wang T., Carley C., Done N., Swallow E., Trends in prevalence and incidence of alopecia areata, alopecia totalis, and alopecia universalis among adults and children in a US employer-sponsored insured population. JAMA Dermatol. 159, 411–418 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
4.Finner A. M., Alopecia areata: Clinical presentation, diagnosis, and unusual cases. Dermatol. Ther. 24, 348–354 (2011). [DOI] [PubMed] [Google Scholar]
5.Gilhar A., Paus R., Kalish R. S., Lymphocytes, neuropeptides, and genes involved in alopecia areata. J. Clin. Invest. 117, 2019–2027 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Hoisnard L., Lebrun-Vignes B., Maury S., Mahevas M., El Karoui K., Roy L., Zarour A., Michel M., Cohen J. L., Amiot A., Claudepierre P., Wolkenstein P., Grimbert P., Sbidian E., Adverse events associated with JAK inhibitors in 126,815 reports from the WHO pharmacovigilance database. Sci. Rep. 12, 7140 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Ytterberg S. R., Bhatt D. L., Connell C. A., Cardiovascular and cancer risk with tofacitinib in rheumatoid arthritis. Reply. N. Engl. J. Med. 386, 1768 (2022). [DOI] [PubMed] [Google Scholar]
8.King B., Ohyama M., Kwon O., Zlotogorski A., Ko J., Mesinkovska N. A., Hordinsky M., Dutronc Y., Wu W. S., McCollam J., Chiasserini C., Yu G., Stanley S., Holzwarth K., DeLozier A. M., Sinclair R., BRAVE-AA Investigators , Two phase 3 trials of baricitinib for alopecia areata. N. Engl. J. Med. 386, 1687–1699 (2022). [DOI] [PubMed] [Google Scholar]
9.King B., Senna M. M., Mesinkovska N. A., Lynde C., Zirwas M., Maari C., Prajapati V. H., Sapra S., Brzewski P., Osman L., Hanna S., Wiseman M. C., Hamilton C., Cassella J., Efficacy and safety of deuruxolitinib, an oral selective Janus kinase inhibitor, in adults with alopecia areata: Results from the Phase 3 randomized, controlled trial (THRIVE-AA1). J. Am. Acad. Dermatol. 91, 880–888 (2024). [DOI] [PubMed] [Google Scholar]
10.King B., Zhang X., Harcha W. G., Szepietowski J. C., Shapiro J., Lynde C., Mesinkovska N. A., Zwillich S. H., Napatalung L., Wajsbrot D., Fayyad R., Freyman A., Mitra D., Purohit V., Sinclair R., Wolk R., Efficacy and safety of ritlecitinib in adults and adolescents with alopecia areata: A randomised, double-blind, multicentre, phase 2b-3 trial. Lancet 401, 1518–1529 (2023). [DOI] [PubMed] [Google Scholar]
11.Bertolini M., McElwee K., Gilhar A., Bulfone-Paus S., Paus R., Hair follicle immune privilege and its collapse in alopecia areata. Exp. Dermatol. 29, 703–725 (2020). [DOI] [PubMed] [Google Scholar]
12.Connell S. J., Jabbari A., The current state of knowledge of the immune ecosystem in alopecia areata. Autoimmun. Rev. 21, 103061 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
13.Lensing M., Jabbari A., An overview of JAK/STAT pathways and JAK inhibition in alopecia areata. Front. Immunol. 13, 955035 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
14.Ito T., Ito N., Bettermann A., Tokura Y., Takigawa M., Paus R., Collapse and restoration of MHC class-I-dependent immune privilege: Exploiting the human hair follicle as a model. Am. J. Pathol. 164, 623–634 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]
15.Freyschmidt-Paul P., McElwee K. J., Hoffmann R., Sundberg J. P., Vitacolonna M., Kissling S., Zoller M., Interferon-gamma-deficient mice are resistant to the development of alopecia areata. Br. J. Dermatol. 155, 515–521 (2006). [DOI] [PubMed] [Google Scholar]
16.Xing L. Z., Dai Z. P., Jabbari A., Cerise J. E., Higgins C. A., Gong W. J., de Jong A., Harel S., DeStefano G. M., Rothman L., Singh P., Petukhova L., Mackay-Wiggan J., Christiano A. M., Clynes R., Alopecia areata is driven by cytotoxic T lymphocytes and is reversed by JAK inhibition. Nat. Med. 20, 1043–1049 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Petukhova L., Duvic M., Hordinsky M., Norris D., Price V., Shimomura Y., Kim H., Singh P., Lee A., Chen W. V., Meyer K. C., Paus R., Jahoda C. A. B., Amos C. I., Gregersen P. K., Christiano A. M., Genome-wide association study in alopecia areata implicates both innate and adaptive immunity. Nature 466, 113–117 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
18.Todes-Taylor N., Turner R., Wood G. S., Stratte P. T., Morhenn V. B., T cell subpopulations in alopecia areata. J. Am. Acad. Dermatol. 11, 216–223 (1984). [DOI] [PubMed] [Google Scholar]
19.Ito T., Ito N., Saatoff M., Hashizume H., Fukamizu H., Nickoloff B. J., Takigawa M., Paus R., Maintenance of hair follicle immune privilege is linked to prevention of NK cell attack. J. Invest. Dermatol. 128, 1196–1206 (2008). [DOI] [PubMed] [Google Scholar]
20.Ito T., Hashizume H., Shimauchi T., Funakoshi A., Ito N., Fukamizu H., Takigawa M., Tokura Y., CXCL10 produced from hair follicles induces Th1 and Tc1 cell infiltration in the acute phase of alopecia areata followed by sustained Tc1 accumulation in the chronic phase. J. Dermatol. Sci. 69, 140–147 (2013). [DOI] [PubMed] [Google Scholar]
21.McElwee K. J., Freyschmidt-Paul P., Hoffmann R., Kissling S., Hummel S., Vitacolonna M., Zoller M., Transfer of CD8⁺ cells induces localized hair loss whereas CD4⁺/CD25⁻ cells promote systemic alopecia areata and CD4⁺/CD25⁺ cells blockade disease onset in the C3H/HeJ mouse model. J. Invest. Dermatol. 124, 947–957 (2005). [DOI] [PubMed] [Google Scholar]
22.Wang E. H. C., Khosravi-Maharlooei M., Jalili R. B., Yu R., Ghahary A., Shapiro J., McElwee K. J., Transfer of alopecia areata to C3H/HeJ mice using cultured lymph node-derived cells. J. Invest. Dermatol. 135, 2530–2532 (2015). [DOI] [PubMed] [Google Scholar]
23.Dai Z., Wang E. H. C., Petukhova L., Chang Y., Lee E. Y., Christiano A. M., Blockade of IL-7 signaling suppresses inflammatory responses and reverses alopecia areata in C3H/HeJ mice. Sci. Adv. 7, eabd1866 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Seok J., Cho S. D., Lee J., Choi Y., Kim S. Y., Lee S. M., Kim S. H., Jeong S., Jeon M., Lee H., Kim A. R., Choi B., Ha S. J., Jung I., Yoon K. J., Park J. E., Kim J. H., Kim B. J., Shin E. C., Park S. H., A virtual memory CD8⁺ T cell-originated subset causes alopecia areata through innate-like cytotoxicity. Nat. Immunol. 24, 1308–1317 (2023). [DOI] [PubMed] [Google Scholar]
25.Schoenberger S. P., Toes R. E., van der Voort E. I., Offringa R., Melief C. J., T-cell help for cytotoxic T lymphocytes is mediated by CD40-CD40L interactions. Nature 393, 480–483 (1998). [DOI] [PubMed] [Google Scholar]
26.Sun J. C., Bevan M. J., Defective CD8 T cell memory following acute infection without CD4 T cell help. Science 300, 339–342 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Shedlock D. J., Shen H., Requirement for CD4 T cell help in generating functional CD8 T cell memory. Science 300, 337–339 (2003). [DOI] [PubMed] [Google Scholar]
28.Janssen E. M., Lemmens E. E., Wolfe T., Christen U., von Herrath M. G., Schoenberger S. P., CD4+ T cells are required for secondary expansion and memory in CD8+ T lymphocytes. Nature 421, 852–856 (2003). [DOI] [PubMed] [Google Scholar]
29.Whitmire J. K., Tan J. T., Whitton J. L., Interferon-gamma acts directly on CD8+ T cells to increase their abundance during virus infection. J. Exp. Med. 201, 1053–1059 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Kumamoto Y., Mattei L. M., Sellers S., Payne G. W., Iwasaki A., CD4+ T cells support cytotoxic T lymphocyte priming by controlling lymph node input. Proc. Natl. Acad. Sci. U.S.A. 108, 8749–8754 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]
31.Borcherding N., Crotts S. B., Ortolan L. S., Henderson N., Bormann N. L., Jabbari A., A transcriptomic map of murine and human alopecia areata. JCI Insight 5, e137424 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
32.Lee E. Y., Dai Z., Jaiswal A., Wang E. H. C., Anandasabapathy N., Christiano A. M., Functional interrogation of lymphocyte subsets in alopecia areata using single-cell RNA sequencing. Proc. Natl. Acad. Sci. U.S.A. 120, e2305764120 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
33.Ober-Reynolds B., Wang C., Ko J. M., Rios E. J., Aasi S. Z., Davis M. M., Oro A. E., Greenleaf W. J., Integrated single-cell chromatin and transcriptomic analyses of human scalp identify gene-regulatory programs and critical cell types for hair and skin diseases. Nat. Genet. 55, 1288–1300 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Dann E., Henderson N. C., Teichmann S. A., Morgan M. D., Marioni J. C., Differential abundance testing on single-cell data using k-nearest neighbor graphs. Nat. Biotechnol. 40, 245–253 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Yu J., Wei M., Boyd Z., Lehmann E. B., Trotta R., Mao H., Liu S., Becknell B., Jaung M. S., Jarjoura D., Marcucci G., Wu L. C., Caligiuri M. A., Transcriptional control of human T-BET expression: The role of Sp1. Eur. J. Immunol. 37, 2549–2561 (2007). [DOI] [PubMed] [Google Scholar]
36.Ha H. L., Wang H., Pisitkun P., Kim J. C., Tassi I., Tang W., Morasso M. I., Udey M. C., Siebenlist U., IL-17 drives psoriatic inflammation via distinct, target cell-specific mechanisms. Proc. Natl. Acad. Sci. U.S.A. 111, E3422–E3431 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
37.van der Fits L., Mourits S., Voerman J. S., Kant M., Boon L., Laman J. D., Cornelissen F., Mus A. M., Florencia E., Prens E. P., Lubberts E., Imiquimod-induced psoriasis-like skin inflammation in mice is mediated via the IL-23/IL-17 axis. J. Immunol. 182, 5836–5845 (2009). [DOI] [PubMed] [Google Scholar]
38.Dai Z. P., Xing L. Z., Cerise J., Wang E. H. C., Jabbari A., de Jong A., Petukhova L., Christiano A. M., Clynes R., CXCR3 blockade inhibits T cell migration into the skin and prevents development of alopecia areata. J. Immunol. 197, 1089–1099 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Yasumizu Y., Takeuchi D., Morimoto R., Takeshima Y., Okuno T., Kinoshita M., Morita T., Kato Y., Wang M., Motooka D., Okuzaki D., Nakamura Y., Mikami N., Arai M., Zhang X., Kumanogoh A., Mochizuki H., Ohkura N., Sakaguchi S., Single-cell transcriptome landscape of circulating CD4⁺ T cell populations in autoimmune diseases. Cell Genom. 4, 100473 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
40.Abbas A. R., Baldwin D., Ma Y., Ouyang W., Gurney A., Martin F., Fong S., van Lookeren Campagne M., Godowski P., Williams P. M., Chan A. C., Clark H. F., Immune response in silico (IRIS): Immune-specific genes identified from a compendium of microarray expression data. Genes Immun. 6, 319–331 (2005). [DOI] [PubMed] [Google Scholar]
41.Drewry L. L., Pewe L. L., Hancox L. S., Van de Wall S., Harty J. T., CD4 T cell-dependent and -independent roles for IFN-γ in blood-stage malaria. J. Immunol. 210, 1305–1313 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Ciecko A. E., Foda B., Barr J. Y., Ramanathan S., Atkinson M. A., Serreze D. V., Geurts A. M., Lieberman S. M., Chen Y.-G., Interleukin-27 is essential for type 1 diabetes development and Sjogren syndrome-like inflammation. Cell Rep. 29, 3073–3086.e5 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Ostanin D. V., Bao J., Koboziev I., Gray L., Robinson-Jackson S. A., Kosloski-Davidson M., Price V. H., Grisham M. B., T cell transfer model of chronic colitis: Concepts, considerations, and tricks of the trade. Am. J. Physiol. Gastrointest. Liver Physiol. 296, G135–G146 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
44.Hashimoto K., Yamada Y., Fujikawa M., Sekiguchi K., Uratsuji H., Mori S., Watanabe H., Matsumoto T., Altered T cell subpopulations and serum anti-TYRP2 and tyrosinase antibodies in the acute and chronic phase of alopecia areata in the C3H/HeJ mouse model. J. Dermatol. Sci. 104, 21–29 (2021). [DOI] [PubMed] [Google Scholar]
45.Phares T. W., Stohlman S. A., Hwang M., Min B., Hinton D. R., Bergmann C. C., CD4 T cells promote CD8 T cell immunity at the priming and effector site during viral encephalitis. J. Virol. 86, 2416–2427 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
46.Novy P., Quigley M., Huang X., Yang Y., CD4 T cells are required for CD8 T cell survival during both primary and memory recall responses. J. Immunol. 179, 8243–8251 (2007). [DOI] [PubMed] [Google Scholar]
47.Ballesteros-Tato A., Leon B., Lund F. E., Randall T. D., CD4+ T helper cells use CD154-CD40 interactions to counteract T reg cell-mediated suppression of CD8+ T cell responses to influenza. J. Exp. Med. 210, 1591–1601 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
48.Bennett S. R., Carbone F. R., Karamalis F., Flavell R. A., Miller J. F., Heath W. R., Help for cytotoxic-T-cell responses is mediated by CD40 signalling. Nature 393, 478–480 (1998). [DOI] [PubMed] [Google Scholar]
49.Ferris S. T., Durai V., Wu R., Theisen D. J., Ward J. P., Bern M. D., Davidson J. T. IV, Bagadia P., Liu T., Briseno C. G., Li L., Gillanders W. E., Wu G. F., Yokoyama W. M., Murphy T. L., Schreiber R. D., Murphy K. M., cDC1 prime and are licensed by CD4⁺ T cells to induce anti-tumour immunity. Nature 584, 624–629 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
50.Freyschmidt-Paul P., McElwee K. J., Hoffmann R., Sundberg J. P., Kissling S., Hummel S., Vitacolonna M., Kopp-Schneider A., Zoller M., Reduced expression of interleukin-2 decreases the frequency of alopecia areata onset in C3H/HeJ mice. J. Invest. Dermatol. 125, 945–951 (2005). [DOI] [PubMed] [Google Scholar]
51.Lai Y. P., Lin C. C., Liao W. J., Tang C. Y., Chen S. C., CD4+ T cell-derived IL-2 signals during early priming advances primary CD8+ T cell responses. PLOS ONE 4, e7766 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]
52.Toumi R., Yuzefpolskiy Y., Vegaraju A., Xiao H., Smith K. A., Sarkar S., Kalia V., Autocrine and paracrine IL-2 signals collaborate to regulate distinct phases of CD8 T cell memory. Cell Rep. 39, 110632 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
53.McElwee K. J., Boggess D., King L. E. Jr., Sundberg J. P., Experimental induction of alopecia areata-like hair loss in C3H/HeJ mice using full-thickness skin grafts. J. Invest. Dermatol. 111, 797–803 (1998). [DOI] [PubMed] [Google Scholar]
54.Carroll J. M., McElwee K. J., King L. E., Byrne M. C., Sundberg J. P., Gene array profiling and immunomodulation studies define a cell-mediated immune response underlying the pathogenesis of alopecia areata in a mouse model and humans. J. Invest. Dermatol. 119, 392–402 (2002). [DOI] [PubMed] [Google Scholar]
55.Ito T., Ito N., Saathoff M., Bettermann A., Takigawa M., Paus R., Interferon-gamma is a potent inducer of catagen-like changes in cultured human anagen hair follicles. Br. J. Dermatol. 152, 623–631 (2005). [DOI] [PubMed] [Google Scholar]
56.Paus R., Bertolini M., The role of hair follicle immune privilege collapse in alopecia areata: Status and perspectives. J. Investig. Dermatol. Symp. Proc. 16, S25–S27 (2013). [DOI] [PubMed] [Google Scholar]
57.Bhat P., Leggatt G., Waterhouse N., Frazer I. H., Interferon-γ derived from cytotoxic lymphocytes directly enhances their motility and cytotoxicity. Cell Death Dis. 8, e2836 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
58.Wang Y.-H., Li W., McDermott M., Son G. Y., Maiti G., Zhou F., Tao A. Y., Raphael D., Moreira A. L., Shen B., Vaeth M., Nadorp B., Chakravarti S., Lacruz R. S., Feske S., IFN-γ-producing T_H1 cells and dysfunctional regulatory T cells contribute to the pathogenesis of Sjogren's disease. Sci. Transl. Med. 16, eado4856 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
59.Domingues H. S., Mues M., Lassmann H., Wekerle H., Krishnamoorthy G., Functional and pathogenic differences of Th1 and Th17 cells in experimental autoimmune encephalomyelitis. PLOS ONE 5, e15531 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]
60.Harbour S. N., Maynard C. L., Zindl C. L., Schoeb T. R., Weaver C. T., Th17 cells give rise to Th1 cells that are required for the pathogenesis of colitis. Proc. Natl. Acad. Sci. U.S.A. 112, 7061–7066 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
61.Simmons S. B., Liggitt D., Goverman J. M., Cytokine-regulated neutrophil recruitment is required for brain but not spinal cord inflammation during experimental autoimmune encephalomyelitis. J. Immunol. 193, 555–563 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
62.Linderman G. C., Rachh M., Hoskins J. G., Steinerberger S., Kluger Y., Fast interpolation-based t-SNE for improved visualization of single-cell RNA-seq data. Nat. Methods 16, 243–245 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
63.Belkina A. C., Ciccolella C. O., Anno R., Halpert R., Spidlen J., Snyder-Cappione J. E., Automated optimized parameters for T-distributed stochastic neighbor embedding improve visualization and analysis of large datasets. Nat. Commun. 10, 5415 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figs. S1 to S9

Tables S1 and S2

sciadv.adz2257_sm.pdf^{(1.5MB, pdf)}

Data Availability Statement

[R1] 1.Mirzoyev S. A., Schrum A. G., Davis M. D. P., Torgerson R. R., Lifetime incidence risk of alopecia areata estimated at 2.1% by Rochester Epidemiology Project, 1990-2009. J. Invest. Dermatol. 134, 1141–1142 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R2] 2.Safavi K. H., Muller S. A., Suman V. J., Moshell A. N., Melton L. J. III, Incidence of alopecia areata in Olmsted County, Minnesota, 1975 through 1989. Mayo Clin. Proc. 70, 628–633 (1995). [DOI] [PubMed] [Google Scholar]

[R3] 3.Mostaghimi A., Gao W., Ray M., Bartolome L., Wang T., Carley C., Done N., Swallow E., Trends in prevalence and incidence of alopecia areata, alopecia totalis, and alopecia universalis among adults and children in a US employer-sponsored insured population. JAMA Dermatol. 159, 411–418 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R4] 4.Finner A. M., Alopecia areata: Clinical presentation, diagnosis, and unusual cases. Dermatol. Ther. 24, 348–354 (2011). [DOI] [PubMed] [Google Scholar]

[R5] 5.Gilhar A., Paus R., Kalish R. S., Lymphocytes, neuropeptides, and genes involved in alopecia areata. J. Clin. Invest. 117, 2019–2027 (2007). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R6] 6.Hoisnard L., Lebrun-Vignes B., Maury S., Mahevas M., El Karoui K., Roy L., Zarour A., Michel M., Cohen J. L., Amiot A., Claudepierre P., Wolkenstein P., Grimbert P., Sbidian E., Adverse events associated with JAK inhibitors in 126,815 reports from the WHO pharmacovigilance database. Sci. Rep. 12, 7140 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R7] 7.Ytterberg S. R., Bhatt D. L., Connell C. A., Cardiovascular and cancer risk with tofacitinib in rheumatoid arthritis. Reply. N. Engl. J. Med. 386, 1768 (2022). [DOI] [PubMed] [Google Scholar]

[R8] 8.King B., Ohyama M., Kwon O., Zlotogorski A., Ko J., Mesinkovska N. A., Hordinsky M., Dutronc Y., Wu W. S., McCollam J., Chiasserini C., Yu G., Stanley S., Holzwarth K., DeLozier A. M., Sinclair R., BRAVE-AA Investigators , Two phase 3 trials of baricitinib for alopecia areata. N. Engl. J. Med. 386, 1687–1699 (2022). [DOI] [PubMed] [Google Scholar]

[R9] 9.King B., Senna M. M., Mesinkovska N. A., Lynde C., Zirwas M., Maari C., Prajapati V. H., Sapra S., Brzewski P., Osman L., Hanna S., Wiseman M. C., Hamilton C., Cassella J., Efficacy and safety of deuruxolitinib, an oral selective Janus kinase inhibitor, in adults with alopecia areata: Results from the Phase 3 randomized, controlled trial (THRIVE-AA1). J. Am. Acad. Dermatol. 91, 880–888 (2024). [DOI] [PubMed] [Google Scholar]

[R10] 10.King B., Zhang X., Harcha W. G., Szepietowski J. C., Shapiro J., Lynde C., Mesinkovska N. A., Zwillich S. H., Napatalung L., Wajsbrot D., Fayyad R., Freyman A., Mitra D., Purohit V., Sinclair R., Wolk R., Efficacy and safety of ritlecitinib in adults and adolescents with alopecia areata: A randomised, double-blind, multicentre, phase 2b-3 trial. Lancet 401, 1518–1529 (2023). [DOI] [PubMed] [Google Scholar]

[R11] 11.Bertolini M., McElwee K., Gilhar A., Bulfone-Paus S., Paus R., Hair follicle immune privilege and its collapse in alopecia areata. Exp. Dermatol. 29, 703–725 (2020). [DOI] [PubMed] [Google Scholar]

[R12] 12.Connell S. J., Jabbari A., The current state of knowledge of the immune ecosystem in alopecia areata. Autoimmun. Rev. 21, 103061 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R13] 13.Lensing M., Jabbari A., An overview of JAK/STAT pathways and JAK inhibition in alopecia areata. Front. Immunol. 13, 955035 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R14] 14.Ito T., Ito N., Bettermann A., Tokura Y., Takigawa M., Paus R., Collapse and restoration of MHC class-I-dependent immune privilege: Exploiting the human hair follicle as a model. Am. J. Pathol. 164, 623–634 (2004). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R15] 15.Freyschmidt-Paul P., McElwee K. J., Hoffmann R., Sundberg J. P., Vitacolonna M., Kissling S., Zoller M., Interferon-gamma-deficient mice are resistant to the development of alopecia areata. Br. J. Dermatol. 155, 515–521 (2006). [DOI] [PubMed] [Google Scholar]

[R16] 16.Xing L. Z., Dai Z. P., Jabbari A., Cerise J. E., Higgins C. A., Gong W. J., de Jong A., Harel S., DeStefano G. M., Rothman L., Singh P., Petukhova L., Mackay-Wiggan J., Christiano A. M., Clynes R., Alopecia areata is driven by cytotoxic T lymphocytes and is reversed by JAK inhibition. Nat. Med. 20, 1043–1049 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R17] 17.Petukhova L., Duvic M., Hordinsky M., Norris D., Price V., Shimomura Y., Kim H., Singh P., Lee A., Chen W. V., Meyer K. C., Paus R., Jahoda C. A. B., Amos C. I., Gregersen P. K., Christiano A. M., Genome-wide association study in alopecia areata implicates both innate and adaptive immunity. Nature 466, 113–117 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R18] 18.Todes-Taylor N., Turner R., Wood G. S., Stratte P. T., Morhenn V. B., T cell subpopulations in alopecia areata. J. Am. Acad. Dermatol. 11, 216–223 (1984). [DOI] [PubMed] [Google Scholar]

[R19] 19.Ito T., Ito N., Saatoff M., Hashizume H., Fukamizu H., Nickoloff B. J., Takigawa M., Paus R., Maintenance of hair follicle immune privilege is linked to prevention of NK cell attack. J. Invest. Dermatol. 128, 1196–1206 (2008). [DOI] [PubMed] [Google Scholar]

[R20] 20.Ito T., Hashizume H., Shimauchi T., Funakoshi A., Ito N., Fukamizu H., Takigawa M., Tokura Y., CXCL10 produced from hair follicles induces Th1 and Tc1 cell infiltration in the acute phase of alopecia areata followed by sustained Tc1 accumulation in the chronic phase. J. Dermatol. Sci. 69, 140–147 (2013). [DOI] [PubMed] [Google Scholar]

[R21] 21.McElwee K. J., Freyschmidt-Paul P., Hoffmann R., Kissling S., Hummel S., Vitacolonna M., Zoller M., Transfer of CD8⁺ cells induces localized hair loss whereas CD4⁺/CD25⁻ cells promote systemic alopecia areata and CD4⁺/CD25⁺ cells blockade disease onset in the C3H/HeJ mouse model. J. Invest. Dermatol. 124, 947–957 (2005). [DOI] [PubMed] [Google Scholar]

[R22] 22.Wang E. H. C., Khosravi-Maharlooei M., Jalili R. B., Yu R., Ghahary A., Shapiro J., McElwee K. J., Transfer of alopecia areata to C3H/HeJ mice using cultured lymph node-derived cells. J. Invest. Dermatol. 135, 2530–2532 (2015). [DOI] [PubMed] [Google Scholar]

[R23] 23.Dai Z., Wang E. H. C., Petukhova L., Chang Y., Lee E. Y., Christiano A. M., Blockade of IL-7 signaling suppresses inflammatory responses and reverses alopecia areata in C3H/HeJ mice. Sci. Adv. 7, eabd1866 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R24] 24.Seok J., Cho S. D., Lee J., Choi Y., Kim S. Y., Lee S. M., Kim S. H., Jeong S., Jeon M., Lee H., Kim A. R., Choi B., Ha S. J., Jung I., Yoon K. J., Park J. E., Kim J. H., Kim B. J., Shin E. C., Park S. H., A virtual memory CD8⁺ T cell-originated subset causes alopecia areata through innate-like cytotoxicity. Nat. Immunol. 24, 1308–1317 (2023). [DOI] [PubMed] [Google Scholar]

[R25] 25.Schoenberger S. P., Toes R. E., van der Voort E. I., Offringa R., Melief C. J., T-cell help for cytotoxic T lymphocytes is mediated by CD40-CD40L interactions. Nature 393, 480–483 (1998). [DOI] [PubMed] [Google Scholar]

[R26] 26.Sun J. C., Bevan M. J., Defective CD8 T cell memory following acute infection without CD4 T cell help. Science 300, 339–342 (2003). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R27] 27.Shedlock D. J., Shen H., Requirement for CD4 T cell help in generating functional CD8 T cell memory. Science 300, 337–339 (2003). [DOI] [PubMed] [Google Scholar]

[R28] 28.Janssen E. M., Lemmens E. E., Wolfe T., Christen U., von Herrath M. G., Schoenberger S. P., CD4+ T cells are required for secondary expansion and memory in CD8+ T lymphocytes. Nature 421, 852–856 (2003). [DOI] [PubMed] [Google Scholar]

[R29] 29.Whitmire J. K., Tan J. T., Whitton J. L., Interferon-gamma acts directly on CD8+ T cells to increase their abundance during virus infection. J. Exp. Med. 201, 1053–1059 (2005). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R30] 30.Kumamoto Y., Mattei L. M., Sellers S., Payne G. W., Iwasaki A., CD4+ T cells support cytotoxic T lymphocyte priming by controlling lymph node input. Proc. Natl. Acad. Sci. U.S.A. 108, 8749–8754 (2011). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R31] 31.Borcherding N., Crotts S. B., Ortolan L. S., Henderson N., Bormann N. L., Jabbari A., A transcriptomic map of murine and human alopecia areata. JCI Insight 5, e137424 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R32] 32.Lee E. Y., Dai Z., Jaiswal A., Wang E. H. C., Anandasabapathy N., Christiano A. M., Functional interrogation of lymphocyte subsets in alopecia areata using single-cell RNA sequencing. Proc. Natl. Acad. Sci. U.S.A. 120, e2305764120 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R33] 33.Ober-Reynolds B., Wang C., Ko J. M., Rios E. J., Aasi S. Z., Davis M. M., Oro A. E., Greenleaf W. J., Integrated single-cell chromatin and transcriptomic analyses of human scalp identify gene-regulatory programs and critical cell types for hair and skin diseases. Nat. Genet. 55, 1288–1300 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R34] 34.Dann E., Henderson N. C., Teichmann S. A., Morgan M. D., Marioni J. C., Differential abundance testing on single-cell data using k-nearest neighbor graphs. Nat. Biotechnol. 40, 245–253 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R35] 35.Yu J., Wei M., Boyd Z., Lehmann E. B., Trotta R., Mao H., Liu S., Becknell B., Jaung M. S., Jarjoura D., Marcucci G., Wu L. C., Caligiuri M. A., Transcriptional control of human T-BET expression: The role of Sp1. Eur. J. Immunol. 37, 2549–2561 (2007). [DOI] [PubMed] [Google Scholar]

[R36] 36.Ha H. L., Wang H., Pisitkun P., Kim J. C., Tassi I., Tang W., Morasso M. I., Udey M. C., Siebenlist U., IL-17 drives psoriatic inflammation via distinct, target cell-specific mechanisms. Proc. Natl. Acad. Sci. U.S.A. 111, E3422–E3431 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R37] 37.van der Fits L., Mourits S., Voerman J. S., Kant M., Boon L., Laman J. D., Cornelissen F., Mus A. M., Florencia E., Prens E. P., Lubberts E., Imiquimod-induced psoriasis-like skin inflammation in mice is mediated via the IL-23/IL-17 axis. J. Immunol. 182, 5836–5845 (2009). [DOI] [PubMed] [Google Scholar]

[R38] 38.Dai Z. P., Xing L. Z., Cerise J., Wang E. H. C., Jabbari A., de Jong A., Petukhova L., Christiano A. M., Clynes R., CXCR3 blockade inhibits T cell migration into the skin and prevents development of alopecia areata. J. Immunol. 197, 1089–1099 (2016). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R39] 39.Yasumizu Y., Takeuchi D., Morimoto R., Takeshima Y., Okuno T., Kinoshita M., Morita T., Kato Y., Wang M., Motooka D., Okuzaki D., Nakamura Y., Mikami N., Arai M., Zhang X., Kumanogoh A., Mochizuki H., Ohkura N., Sakaguchi S., Single-cell transcriptome landscape of circulating CD4⁺ T cell populations in autoimmune diseases. Cell Genom. 4, 100473 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R40] 40.Abbas A. R., Baldwin D., Ma Y., Ouyang W., Gurney A., Martin F., Fong S., van Lookeren Campagne M., Godowski P., Williams P. M., Chan A. C., Clark H. F., Immune response in silico (IRIS): Immune-specific genes identified from a compendium of microarray expression data. Genes Immun. 6, 319–331 (2005). [DOI] [PubMed] [Google Scholar]

[R41] 41.Drewry L. L., Pewe L. L., Hancox L. S., Van de Wall S., Harty J. T., CD4 T cell-dependent and -independent roles for IFN-γ in blood-stage malaria. J. Immunol. 210, 1305–1313 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R42] 42.Ciecko A. E., Foda B., Barr J. Y., Ramanathan S., Atkinson M. A., Serreze D. V., Geurts A. M., Lieberman S. M., Chen Y.-G., Interleukin-27 is essential for type 1 diabetes development and Sjogren syndrome-like inflammation. Cell Rep. 29, 3073–3086.e5 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R43] 43.Ostanin D. V., Bao J., Koboziev I., Gray L., Robinson-Jackson S. A., Kosloski-Davidson M., Price V. H., Grisham M. B., T cell transfer model of chronic colitis: Concepts, considerations, and tricks of the trade. Am. J. Physiol. Gastrointest. Liver Physiol. 296, G135–G146 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R44] 44.Hashimoto K., Yamada Y., Fujikawa M., Sekiguchi K., Uratsuji H., Mori S., Watanabe H., Matsumoto T., Altered T cell subpopulations and serum anti-TYRP2 and tyrosinase antibodies in the acute and chronic phase of alopecia areata in the C3H/HeJ mouse model. J. Dermatol. Sci. 104, 21–29 (2021). [DOI] [PubMed] [Google Scholar]

[R45] 45.Phares T. W., Stohlman S. A., Hwang M., Min B., Hinton D. R., Bergmann C. C., CD4 T cells promote CD8 T cell immunity at the priming and effector site during viral encephalitis. J. Virol. 86, 2416–2427 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R46] 46.Novy P., Quigley M., Huang X., Yang Y., CD4 T cells are required for CD8 T cell survival during both primary and memory recall responses. J. Immunol. 179, 8243–8251 (2007). [DOI] [PubMed] [Google Scholar]

[R47] 47.Ballesteros-Tato A., Leon B., Lund F. E., Randall T. D., CD4+ T helper cells use CD154-CD40 interactions to counteract T reg cell-mediated suppression of CD8+ T cell responses to influenza. J. Exp. Med. 210, 1591–1601 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R48] 48.Bennett S. R., Carbone F. R., Karamalis F., Flavell R. A., Miller J. F., Heath W. R., Help for cytotoxic-T-cell responses is mediated by CD40 signalling. Nature 393, 478–480 (1998). [DOI] [PubMed] [Google Scholar]

[R49] 49.Ferris S. T., Durai V., Wu R., Theisen D. J., Ward J. P., Bern M. D., Davidson J. T. IV, Bagadia P., Liu T., Briseno C. G., Li L., Gillanders W. E., Wu G. F., Yokoyama W. M., Murphy T. L., Schreiber R. D., Murphy K. M., cDC1 prime and are licensed by CD4⁺ T cells to induce anti-tumour immunity. Nature 584, 624–629 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R50] 50.Freyschmidt-Paul P., McElwee K. J., Hoffmann R., Sundberg J. P., Kissling S., Hummel S., Vitacolonna M., Kopp-Schneider A., Zoller M., Reduced expression of interleukin-2 decreases the frequency of alopecia areata onset in C3H/HeJ mice. J. Invest. Dermatol. 125, 945–951 (2005). [DOI] [PubMed] [Google Scholar]

[R51] 51.Lai Y. P., Lin C. C., Liao W. J., Tang C. Y., Chen S. C., CD4+ T cell-derived IL-2 signals during early priming advances primary CD8+ T cell responses. PLOS ONE 4, e7766 (2009). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R52] 52.Toumi R., Yuzefpolskiy Y., Vegaraju A., Xiao H., Smith K. A., Sarkar S., Kalia V., Autocrine and paracrine IL-2 signals collaborate to regulate distinct phases of CD8 T cell memory. Cell Rep. 39, 110632 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R53] 53.McElwee K. J., Boggess D., King L. E. Jr., Sundberg J. P., Experimental induction of alopecia areata-like hair loss in C3H/HeJ mice using full-thickness skin grafts. J. Invest. Dermatol. 111, 797–803 (1998). [DOI] [PubMed] [Google Scholar]

[R54] 54.Carroll J. M., McElwee K. J., King L. E., Byrne M. C., Sundberg J. P., Gene array profiling and immunomodulation studies define a cell-mediated immune response underlying the pathogenesis of alopecia areata in a mouse model and humans. J. Invest. Dermatol. 119, 392–402 (2002). [DOI] [PubMed] [Google Scholar]

[R55] 55.Ito T., Ito N., Saathoff M., Bettermann A., Takigawa M., Paus R., Interferon-gamma is a potent inducer of catagen-like changes in cultured human anagen hair follicles. Br. J. Dermatol. 152, 623–631 (2005). [DOI] [PubMed] [Google Scholar]

[R56] 56.Paus R., Bertolini M., The role of hair follicle immune privilege collapse in alopecia areata: Status and perspectives. J. Investig. Dermatol. Symp. Proc. 16, S25–S27 (2013). [DOI] [PubMed] [Google Scholar]

[R57] 57.Bhat P., Leggatt G., Waterhouse N., Frazer I. H., Interferon-γ derived from cytotoxic lymphocytes directly enhances their motility and cytotoxicity. Cell Death Dis. 8, e2836 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R58] 58.Wang Y.-H., Li W., McDermott M., Son G. Y., Maiti G., Zhou F., Tao A. Y., Raphael D., Moreira A. L., Shen B., Vaeth M., Nadorp B., Chakravarti S., Lacruz R. S., Feske S., IFN-γ-producing T_H1 cells and dysfunctional regulatory T cells contribute to the pathogenesis of Sjogren's disease. Sci. Transl. Med. 16, eado4856 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R59] 59.Domingues H. S., Mues M., Lassmann H., Wekerle H., Krishnamoorthy G., Functional and pathogenic differences of Th1 and Th17 cells in experimental autoimmune encephalomyelitis. PLOS ONE 5, e15531 (2010). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R60] 60.Harbour S. N., Maynard C. L., Zindl C. L., Schoeb T. R., Weaver C. T., Th17 cells give rise to Th1 cells that are required for the pathogenesis of colitis. Proc. Natl. Acad. Sci. U.S.A. 112, 7061–7066 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R61] 61.Simmons S. B., Liggitt D., Goverman J. M., Cytokine-regulated neutrophil recruitment is required for brain but not spinal cord inflammation during experimental autoimmune encephalomyelitis. J. Immunol. 193, 555–563 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R62] 62.Linderman G. C., Rachh M., Hoskins J. G., Steinerberger S., Kluger Y., Fast interpolation-based t-SNE for improved visualization of single-cell RNA-seq data. Nat. Methods 16, 243–245 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

[R63] 63.Belkina A. C., Ciccolella C. O., Anno R., Halpert R., Spidlen J., Snyder-Cappione J. E., Automated optimized parameters for T-distributed stochastic neighbor embedding improve visualization and analysis of large datasets. Nat. Commun. 10, 5415 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

TH1 effector CD4 T cells rely on IFN-γ production to induce alopecia areata

Samuel J Connell

Sydney Crotts

Ryan Reis

Maddison M Lensing

Payton Kahl

Nicholas Henderson

Otgonzaya Ayush

Zhaowen Zhu

Luana S Ortolan

Audrey C Ko

Erin M Shriver

Keith D Carter

John T Harty

Joan M Goverman

Ali Jabbari

Roles

Abstract

INTRODUCTION

RESULTS

SDLNs from AA mice contain CD8 and CD4 T cells that can contribute to disease induction

Fig. 1. CD8 and CD4 T cells from SDLNs of AA mice are sufficient to transfer disease.

AA inductive capacity is restricted to CD4 T cells from the SDLNs

Fig. 2. Induction of AA is restricted to the CD4 T cells in the SDLNs.

CD4 T cells in draining lymph nodes of AA mice exhibit an enriched TH1 effector phenotype when compared to UA mice

Fig. 3. Enriched TH1 effector gene signature in CD4 T cells from AA SDLNs.

Fig. 4. TH1 effector CD4 T cells in AA mice can induce disease.

CD4 T cell–mediated induction of AA requires endogenous CD8 T cells

Fig. 5. Endogenous CD8 T cells are required for CD4 T cell–mediated disease induction.

Host IFN-γ signaling is required for the induction of AA by CD4 T cells

Fig. 6. Host IFN-γ signaling is required for CD4 T cell induction of AA.

Effector CD4 T cells share similar transcriptional characteristics in murine and human AA

Fig. 7. Human AA CD4 T cells express a TH1 signature.

CD4 T cell–derived IFN-γ is responsible for CD4 T cell–mediated AA development

Fig. 8. CD4 T cell–derived IFN-γ is critical for the development of AA.

DISCUSSION

MATERIALS AND METHODS

Mice

Study participants

T cell isolation for in vitro culture and induction of AA

In vivo depleting antibodies

Preparation of tissue cell suspensions

Flow cytometry and fluorescence-activated cell sorting

Immunostaining

Tamoxifen treatment

Bulk RNA-seq

Human scRNA-seq data analysis

Statistical analysis

Acknowledgments

Funding:

Author contributions:

Competing interests:

Data and materials availability:

Supplementary Materials

This PDF file includes:

REFERENCES AND NOTES

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases

T_H1 effector CD4 T cells rely on IFN-γ production to induce alopecia areata

CD4 T cells in draining lymph nodes of AA mice exhibit an enriched T_H1 effector phenotype when compared to UA mice

Fig. 3. Enriched T_H1 effector gene signature in CD4 T cells from AA SDLNs.

Fig. 4. T_H1 effector CD4 T cells in AA mice can induce disease.

Fig. 7. Human AA CD4 T cells express a T_H1 signature.