Alopecia areata (AA) is postulated to be an organ-speci c

Transcription

1 ORIGINAL ARTICLE Autoimmunity: Alopecia Areata Maria Hordinsky and Marna Ericson Department of Dermatology, University of Minnesota, Minneapolis, Minnesota, USA Strong direct and indirect evidence supports an autoimmune etiology for alopecia areata. T lymphocytes that have been shown to be oligoclonal and autoreactive are predominantly present in the peribulbar in ammatory in ltrate. Alopecia areata frequently occurs in association with other autoimmune diseases, such as thyroiditis and vitiligo, and autoantibodies to follicular components have been detected. Finally, the use of immune modulating drugs, including corticosteroids and contact sensitizers such as dyphencyprone, can be bene cial in the management of this disease. Recent studies have demonstrated that alopecia areata scalp skin grafted onto nude mice with severe combined immunode ciency grow hair and that in ltrating lymphocytes in the graft are lost. It is now also possible to induce alopecia areata in human scalp explants on these mice by injecting T lymphocytes with scalp homogenate. Neuropeptides produced by cutaneous nerves are known to modify immune reactivity and, in all likelihood, a ect the alopecia areata process. Future studies may show that modulation of neuropeptide expression is associated with hair regrowth. Likewise, testing the e cacy of the newly developed immunomodulatory agents in patients with alopecia areata may lead to the introduction of novel therapies for this immunemediated disease of the hair follicle. JInvestigDermatol Symp Proc 9:73 ^78, 2004 ALOPECIA AREATA^CLINICAL PRESENTATIONS Alopecia areata (AA) is postulated to be an organ-speci c autoimmune disease. The incidence of AA in the United States (Minnesota) is 20.2/100,000 person-years (Muller and Winckelmann, 1963), and the lifetime risk is estimated to be approximately 1.7%. AA is characterized by nonscarring hair loss that may be patchy (areata) or extensive. Loss of all scalp hair is described as alopecia totalis (AT); the term alopecia universalis (AU) is frequently used when all body hair is lost (Fig 1). Guidelines have been established to assess disease extent and take into consideration that extent of hair loss may vary from one region of the body to another (Olsen and Hordinsky, 1999). Nail abnormalities are common in patients with AA, and abnormalities in other ectodermal derived appendages, such as sebaceous and sweat glands, have been suggested (Hordinsky, 2001). AA has been described as occurring in association with many diseases, several of which are considered autoimmune. The two main associations are with thyroid disease and vitiligo. Muller and Winkelman examined 736 patients with AA; 8% reported thyroid disorder, whereas the controls had an incidence of 2% (Muller, 1963). These investigators found the incidence of vitiligo Manuscript accepted for publication October 18, 2002 Correspondence and reprint requests to: Maria Hordinsky, Department of Dermatology, University of Minnesota, 420 Delaware St. SE, Minneapolis, Minnesota 55455, USA. hordi001@tc.umn.edu Abbreviations: AA, Alopecia Areata; AT, Alopecia Totalis; AU, Alopecia Universalis; (APS), Autoimmune Polyglandular Syndrome; CGRP, Calcitonin Gene-Related Peptide; DEBR, Dundee Experimental Bald Rat; UEAI, Eulex europeaus agglutinin I; HLA, Human Lymphocyte Antigen; IL-10, Interleukin 10; ICAM, Intracellular Adhesion Molecule; MHC, Major Histocompatibility Complex; PGP, Protein Gene Product; SCID, Severe Combined Immunode ciency; SP, Substance P; TNF-a, Tumor Necrosis Factor a. to be 3%^8% in patients with AA; the prevalence of vitiligo in the United States is considered to be about 1% (Majumder and Das, 1988). A very strong disease association with autosomal recessive disease autoimmune polyglandular syndrome (APS-1, chronic hypoparathyroidism-mucocutaneous candidiasis-autoimmune adrenal insu ciency) was recently reported. Also, AA commonly occurs in patients with Down s syndrome or Turner s syndrome and, in some studies, atopies (allergic rhinitis, asthma, and atopic dermatitis) have been found in more than 40% of AA patients whereas their prevalence in the general population is estimated to be around 20%. One of the interesting observations made about disease associations of AA is the decreased incidence of Type I (insulin-dependent) diabetes in AA patients and an increased incidence in their relatives. This has suggested to some that AA may have a protective e ect against Type I diabetes in predisposed individuals (Wang et al,1994). ANIMAL MODELS IN AA There are two rodent models for AA: a subpopulation of the C3H/HeJ mouse strain and the Dundee experimental bald rat (DEBR). In both models, the a ected surface provides a test site for studying the pathophysiology of AA and for testing the mechanism of action, safety, and e cacy of new therapies (McElwee et al, 1997; Freyschmidt-Paul et al, 1999). C3H/HeJ mice produce a normal coat of agouti hair, but as early as four months after birth hair loss can develop on the ventral surface and, more focally, on the dorsal surface. The clinical and histopathologic features of this hair loss are similar to those in human AA (Sundberg et al,1994; Sundberg and King, 1996). The DEBR rat is a hooded rat strain that also demonstrates hair loss similar to that observed in human AA. Rats develop a normal coat of hair from birth up until four months, but then up to 30% of males and 80% females experience hair loss on their heads, which can spread to the anks (Michie, 1991; Zhang and Oliver, 1994; McElwee et al, 1996) /04/$ Copyright r 2004 by The Society for Investigative Dermatology, Inc. 73

2 74 HORDINSKY AND ERICSON JID SYMPOSIUM PROCEEDINGS Figure 1. Alopecia areata. Left: Patchy disease activity consistent with the diagnosis of alopecia areata. Right: Extensive scalp hair loss characteristic of alopecia totalis. Figures 2 through 4. Miniaturized scalp hair follicles in 4-mm scalp biopsy specimens taken from patients with extensive alopecia areata of greater than two years duration. Images depict nerves (PGP 9.5) and neuropeptide, substance p (SP), or calcitonin gene^related peptide (CGRP) expression before and after a 21-day treatment period with capsaicin. Each image is a projection of multiple optical sections captured with a laser scanning confocal microscope (MRC-1000 Confocal Imaging System, Bio- Rad, Hercules, CA). AA AND THE MAJOR HISTOCOMPATIBILITY COMPLEX (MHC) Autoimmune diseases are commonly associated with an increase in certain HLA-region genes or haplotypes. Because the predisposition to autoimmunity has been associated with HLA- D alleles and AA is hypothesized to be a T cell^mediated autoimmune disease, several studies have looked for an association between HLA and AA (Aita and Christiano, 2001). HLA class I molecules are expressed on virtually all nucleated cells and platelets and present antigens to CD8 þ T cells. HLA class II molecules have three main subclasses (DR, DQ, and DP); they are found on speci c immune cells, including B cells, activated T cells, macrophages, keratinocytes, and dendritic cell and present peptides to CD4 þ T cells. Because class II molecules are associated with antigen presentation, many studies have focused on this area of the HLA molecule. The nomenclature of HLA molecules has changed over the years. The World Health Organization HLA Nomenclature committee now speci es a four-digit code to de ne each allele of each HLA locus at the DNA level (e.g., DRB1 1101), and twodigit codes to de ne the previous, lower-resolution serological equivalents (e.g., DRw11). The association of AA with HLA-DR and HLA-DQ antigens suggests a role for CD4 þ T cells in this disease, as MHC class II molecules present peptides to CD4 þ cells. Recent transplantation studies indicate that CD8 þ cells are also involved in AA, implicating MHC class I HLA-A,B,C molecules, which are associated with the presentation of peptides to CD8 þ T cells, in addition to MHC class II molecules. A genetic di erence between patients who have extensive AA (AT or AU) and those who patchy AA has been identi ed. Both patient groups have been found to have a positive association with DQB1 03. The HLA alleles DRB and HLA- DQB are now also viewed as markers for more severe, long-standing disease (Welsh et al, 1994; Columbe et al, 1995). Presently, other genes in addition to the HLA region are under investigation for possible genetic associations. AA may occur concurrently or sequentially in both monozygotic and fraternal twins. A concordance rate of 55% has been reported in monozygotic twins and 0% in fraternal twins. This leaves much room for the role of the environment in AA pathophysiology (Jackow et al,1998). HISTOLOGY AND IMMUNOCYTOCHEMICAL FINDINGS IN AA The histology of AA is characterized by the presence of peribulbar and intrabulbar mononuclear in ltrates, degenerative changes in the hair matrix, decreased numbers of terminal anagen follicles, increased numbers of terminal catagen and telogen follicles, increased numbers of follicular stelae, increased numbers of miniaturized vellus hair follicles, and pigment incontinence of hair bulbs (Whiting, 2001). Normal hair follicle epithelium typically does not express the HLA class I antigens A, B, and C, which suggests to some that the hair follicle is an immune privileged site (Christoph et al, 2000). In AA, in contrast, HLA-A,B,C as well as HLA-DR antigens are expressed by follicular epithelium, suggesting that the expression of these antigens may result in the presentation of follicular autoantigens and loss of immune privilege. Paus and colleagues have hypothesized that the induction of MHC class I (HLA-A,B,C) antigens in AA permits a response by melanocyte reactive CD8 þ T cells (Paus et al, 1994). The hypothesis is that CD8 þ cells induce HLA-DR expression on a ected hair follicles by the production of interferon-g with the subsequent recruitment of CD4 þ cells. Supporting this hypothesis is the work of Kalish and colleagues describing the presence of CD4 þ autoreactive T cells in the in ltrate of AA (Kalish et al, 1992). Adhesion molecules, such as intracellular adhesion molecule 1 (ICAM-1), E-selectin, and others, which are important in the homing of lymphocytes to sites of in ammation, have also been identi ed in microvascular endothelial cells in the perifollicular region of a ected hair follicles. B CELL FUNCTION Circulating antibodies to follicular structures have been found in both humans and animal models of AA, but they have not been found to be pathogenic in either (Tobin et al, 1997). It has been suggested that these autoantibodies are a marker for CD4 þ T cell recognition of hair follicles but do not play a primary role

3 VOL. 9, NO. 1 JANUARY 2004 ALOPECIA AREATA: AUTOIMMUNITY 75 in AA pathogenesis (Kalish and Gilhar, 2001). Circulating antibodies to follicular structures have also been reported in normal controls. EVIDENCE THATAA IS MEDIATED BY T LYMPHOCYTES: REVIEW OF THE HUMAN SCALP EXPLANT EXPERIMENTS A series of experiments have demonstrated the transfer of AA to human scalp explants on SCID mice. In the rst published experiments, Gilhar and colleagues transferred AA scalp explants from patients to mice. Once transplanted, hairs grew in the grafts and at day 40 the grafts were injected with (1) autologous peripheral blood mononuclear cells, (2) autologous T cells isolated from the AA scalp biopsy specimens, or (3) scalp T cells cultured with hair follicle homogenate. T lymphocytes that had been cultured with hair follicle homogenate were capable of inducing changes of histologic AA in the grafts. Clinically, hair loss was documented, and perifollicular in ltrates of T cells, as well as HLA-DR and ICAM-1 expression on the follicular epithelium, were observed (Gilhar et al, 1998). The results of this work showed it to be possible to induce AA in the human scalp explant/scid mouse system by injection of activated lesional T cells (Gilhar, 1998; Gilhar et al, 1999). To determine the role of CD4 þ and CD8 þ T lymphocytes in the pathogenesis of AA, Gilhar and colleagues separated CD4 þ and CD8 þ T cells using magnetic beads, and injected not only unseparated T cells but also mixed CD4 þ and CD8 þ T cells, CD4 þ cells alone, and CD8 þ T cells alone. Injection of unseparated T cells and mixed CD4 þ and CD8 þ T cells resulted in signi cant hair loss, whereas injection of puri ed CD4 þ or CD8 þ T cells alone did not result in reproducible hair loss. The current hypothesis is that both CD4 þ and CD8 þ T cells have a role in the pathogenesis of AA, the CD8 þ cells acting as the e ector cells with help from the CD4 þ T cells (Gilhar et al, 2001). The autoantigen(s) in AA remains to be identi ed, but results of current studies suggest that it may be melanocyte derived. Support for this hypothesis comes from clinical observations that pigmented hair bers are preferentially lost and that vitiligo is commonly associated with AA. In addition, hair bulb melanocytes in AA demonstrate both histologic and ultrastructural abnormalties (Tobin et al, 1990). To test the hypothesis that melanocyte-associated antigens can function as autoantigens and induce hair loss in AA, HLA-A2^ positive patients with AA were selected for studies of HLA-A2- restricted melanocyte peptide epitopes. ScalpTcells were cultured with autologous antigen- presenting cells and either hair follicle homogenate as the positive control or melanocyte T cell epitopes. Cells were then transferred to autologous scalp explants on SCID mice. Histologic changes and hair regrowth were monitored. Melanocyte peptide^activated T cells signi cantly reduced the number of regrowing hair bers, and injected scalp grafts demonstrated histologic and immunocytochemical features of AA. The most consistent peptide autoantigens were the Gp100-derived G9-209 and G9-280 peptides, as well as MART-1 (Gilhar et al, 2001). Additional work remains to be done to identify the range of melanocyte-associated autoantigens in di erent HLA backgrounds, as well as to put this knowledge to therapeutic use (Gilhar et al, 2001). Recently completed studies of CD69 expression suggest that T cells in scalp biopsies taken from patients with extensive AA are speci c for and activated by an antigen in the skin, and that T cell responses in such skin are tightly, albeit aberrantly, regulated via mechanisms of peripheral T cell tolerance ). 1 Whether the antigen these cells are reacting is a melanocyte peptide remains to be determined. 1 Steiner LP, Deeths MJ, Ericson ME, Hordinsky M: T cells in extensive alopecia areata are partially tolerant. Poster presentation, annual meeting of the SID, NERVES AND AA Neuropeptides are a heterogeneous group of biologically active peptides present in neurons of both the central and the peripheral nervous system (Wallengren, 1997). These peptides are involved with the transmission of signals not only between nerve cells but also with cells of the immune system (Rossi and Johansson, 1998). Two neuropeptides that have been studied extensively in AA are Substance P (SP) and calcitonin gene^related peptide (CGRP) (Hordinsky and Ericson, 1996). CGRP released from cutaneous nerves can induce mast cell degranulation and subsequent release of immunosuppressive tumor necrosis factor a (TNF-a) and interleukin 10 (IL-10). CGRP can also interact with keratinocyte factors to promote melanization (Toyoda et al, 1999). Patients with AA have been found to have low serum levels of CGRP and an exaggerated vasodilatory response to local CGRP injection. (Daly, 1998). Results of immunocytochemical studies combined with laser scanning confocal microscopy have demonstrated that perifollicular innervation is arranged in a basket-weave network around the miniaturized AA anagen follicles in patients with long-standing (42 years duration) extensive AA. A prominent hair follicle nerve plexus, which appears as a stockade of nerves around the bulge region of small miniaturized follicles, has also been described in some patients with long-standing AA. Modulating the peripheral nervous system in the management of AA has been explored. Capsaicin, derived from chile peppers, is known to excite subsets of sensory neurons associated with pain and thermoreception, and it is also known to release the neuropeptides SP and CGRP. In a cream formulation, capsaicin is frequently used as a topical medication to treat painful syndromes. When applied to normal skin, it elicits a sensation of burning pain by selectively activating sensory neurons that send information about noxious stimuli to the central nervous system. Capsaicin releases SP from sensory nerve bers and, after repeated application, depletes neurons of SP and results in nonpermanent injury to epidermal nerves (Simone et al,1998). In a small pilot study, we wanted to ascertain how perifollicular nerves in scalp biopsy specimens taken from patients with long-standing extensive AA respond to the topical application of capsaicin cream 0.075% (Zostrix-HP). Two adult female patients participated in this study after signing an informed consent approved by the University of Minnesota Institutional Review Board, Human Subjects Committee. Each subject applied 0.075% capsaicin cream to her entire bald scalp for three weeks. Both treated patients experienced a burning pain sensation, which improved over the 21-day treatment period but never completely disappeared. Both patients also experienced vellus hair regrowth by day 21. Four-millimeter punch biopsies were taken at day 0, day 1, and day 21. Specimens were multilabeled with antibodies to panneuronal protein gene product 9.5 (PGP 9.5) and cyanine 3.18, substance p or CGRP, and Cyanine 5.18, as well as the vascular marker Ulex europeans agglutinin (UEA I) conjugated to uorescein. In-focus images of well-de ned optical sections (0.5^1.0 microns) were captured by laser scanning confocal microscopy. Computer reconstruction of captured images yielded three-dimensional views of the architecture, perifollicular nerves and blood vessels, and the expression of the neuropeptides SP and CGRP. Analysis of 21-day scalp biopsy samples compared to baseline revealed (1) a qualitative decrease in the number of nerves staining with PGP 9.5 and SP in the epidermis, (2) small extra-neuronal globules staining with antibody to SP, and (3) an abundance of SP expression in the stockade region of the miniaturized follicle (Figs 2^4). The results indicate that cutaneous innervation is altered in AA. Though epidermal nerves have been reported to degenerate with intradermal injection and topical application of capsaicin, both patients experienced undiminished scalp sensation such as burning with drug application throughout the study. It is possible that the expression and targeting of the vanilloid receptor, VRI

4 76 HORDINSKY AND ERICSON JID SYMPOSIUM PROCEEDINGS Figure 2. Pre-treatmentöPGP 9.5 and CGRP expression. (a) PGP 9.5-positive subepidermal nerves extend from this plexus into the epidermis, and perifollicular nerves are present in a basketweave pattern. (b) CGRP-positive nerves can be found in the subepidermal nerve plexus and epidermis and as a network of nerves around a miniaturized hair follicle. (c) Colocalization of PGP 9.5 (green) and CGRP (red) in several subepidermal and perifollicular nerves is present. Each picture is a projection of 36 optical sections taken at 2-micron intervals. Scale bar ¼100 microns. Figure 3. Pre-treatmentöPGP9.5 and SP expression. (a) PGP 9.5-positive epidermal and perifollicular nerves, as well nerves highlighting eccrine glands and the arrector pili muscle, are seen. (b) SP staining nerves and globule-like extra neuronal structures are present in the dermis. (c) Colocalization of PGP 9.5 (green) and SP (red) reveals some SP-containing nerve bers throughout the scalp skin, but the majority of expression is within the vasculature (data not shown). Each picture is a projection of 46 optical sections taken at 1-micron intervals. Scale bar ¼100 microns. Figure 4. Post-treatmentöPGP9.5 and SP expression. (a) PGP 9.5-positive nerves are less prominent in the epidermis; the subepidermal plexus appears intact. The perifollicular nerve network is dominated by the PGP-positive staining nerve stockade surrounding a miniaturized follicle. (b) SP expression is signi cant in the area of the stockade, and there is signi cantly less SP expression elsewhere. (c) Colocalization of PGP 9.5 (green) and SP (red) demonstrate that the expression of PGP 9.5 is separate from SP expression in the stockade region. Each picture is a projection of 45 optical sections taken at 1-micron intervals. Scale bar ¼100 microns. (the receptor for capsaicin and the transducer of noxious thermal stimuli), is di erent in patients with long-standing extensive AA. The SP staining within the vasculature (data not shown) may re- ect lymphocytes, macrophages, or eosinophils, all of which have receptors for SP. The abundance of SP in the stockade region and the vellus hair regrowth support the need for additional studies to understand the role of the peripheral nervous system in the transmission of signals not only between nerve cells but also with cells of the immune system and the hair follicle in AA. IMMUNOMODULATING AGENTS FOR THE TREATMENT OF AA Because CD8 þ cells have been implicated as e ector cells in AA, testing of therapies that interfere with CD8 þ activity appear reasonable (Gilhar et al, 2001). Agents that e ect CD4 þ T cells may also be bene cial in the management of AA, but the ideal agent will probably be speci c for both CD4 þ and CD8 þ T cells. Antibodies that block antigen presentation or costimulation molecules such as CTLA- 4Ig and anti-cd11a also have the potential to be successful therapeutic agents (Gilhar et al, 2001). Other therapies that may be bene cial include agents active against TH1 cytokines, such as interleukin-2, interferon-g, and TNF-a, or agents that promote TH2 responses, such as interleukin-10 (Kalish, 2001). In a six-month trial for the therapy of longstanding AA with Aldara 5% cream, increased anagen^telogen and terminal^vellus ratios in scalp biopsy specimens were documented. 2 THE ALOPECIA AREATA REGISTRY The opportunity to establish a Registry for AA became a reality with grant support from the National Institutes of Health and the National Institute of Arthritis, Metabolism, and Skin Diseases. 2 Steiner LP, Boeck CM, Ericson ME, Deeths MJ, Whiting DA, Hordinsky M: E ect of Aldara 5% Cream on Anagen:Telogen and Terminal: Vellus Ratios in Extensive Alopecia Areata. Poster Presentation, Fourth International Alopecia Areata Research Workshop, Washington DC, November, 2002.

5 VOL. 9, NO. 1 JANUARY 2004 ALOPECIA AREATA: AUTOIMMUNITY 77 Figure 5. Flow diagram of the Alopecia Areata Registry. The goal of the protocol is to develop a registry from which subsequent analyses and selective patient sampling can be conducted. One goal is to search for associations between speci c genetic markers, such as HLA-linked loci, and the development of AA. The overall goals are to understand the genetic control of autoimmunity in AA, to better understand the complex biology of the cycling hair follicle, and to use this knowledge to devise safe and e ective treatments (Protocol Laboratory- 0391). The operation of the registry is outlined in Fig 5. SUMMARY The development of AA is most likely linked to many factors. One, genetic susceptibility, is related in part to HLA antigen^presenting molecules, alterations in neuropeptide expression, and an alteration in the immune privileged state of the hair follicle. AA appears to be a TH1 autoimmune condition mediated by both CD4 þ and CD8 þ T cells. Though the autoantigen is not known, it may well be melanocyte associated. Future therapeutic investigations should include testing immunomodulatory agents currently being developed and in clinical trials, particularly for the treatment of psoriasis, as well examining compounds that alter the peripheral nervous system. The establishment of the Alopecia Areata Registry will provide the patient material needed to conduct the many genetic studies that remain to be done in this disease. We thank Christine Baker for her assistance with the preparation of this manuscript. This work was supported in part by grants from the National Alopecia Areata Foundation and 3M Pharmaceuticals. REFERENCES Aita VM, Christiano AM: The genetics of alopecia areata. Dermatol Ther 14:329^339, 2001 Christoph T, Muller-Rover S, Audring H, et al: The human hair follicle immune system. Cellular composition and immune privilege. Br J Dermatol 142:862^ 873, 2000 Columbe BW, Price VH, Khoury EL, et al: HLA class II alleles in long-standing totalis/alopecia universalis and long-standing patchy alopecia areata di erentiate these two clinical groups. J Invest Dermatol 104:54^55, 1995 DalyTJ: Alopecia areata has low plasma levels of the vasodilator/immunomodulatory calcitonin gene-related protein. Arch Dermatol 13:1164^1165, 1998 Freyschmidt-Paul P, Sundberg JP, Happle R, McElwee KJ, Metz S, Boggess D, Ho - mann R: Successful treatment of alopecia areata-like hair loss with the contact sensitizer squaric acid dibutylester (SADBE) in C3H/HeJ mice. JInvestDermatol 113:61^68, 1999 Gilhar A, Landau M, Assy B, Shalaginov R: Mediation of alopecia areata by cooperation between CD4 þ and CD8 Lymphocytes. Arch Dermatol 138:916^922, 2001 Gilhar A, Landau M, Assy B, Shalaginov R, Sera movich S, Kalish RS: Melanocyte-associated T cell epitopes can function as autoantigens for transfer of alopecia areata to human scalp explants on Prkdc scid mice. JInvestDermatol 117:1357^1362, 2001 Gilhar A, Shalaginov R, Assy B, Sera movich S, Kalish RS: Alopecia areata is a T- lymphocyte mediated autoimmune disease: lesional human T-lymphocytes transfer alopecia areata to human skin grafts on SCID mice. J Investig Dermatol Symp Proc 4:20^210, 1999 Gilhar A, UllmannY, Berkutzki T, Assy B, Kalish R: Autoimmune hair loss (alopecia areata) transferred by T lymphocytes to human scalp explants on SCID mice. JClinInvest101:62^67, 1998 Hordinsky M: Clinical presentations of alopecia areata. Dermatol Ther 14:291^296, 2001 Hordinsky M, Ericson ME: Relationship between follicular nerve supply and alopecia. Dermatol Clin 14:651^660, 1996 Jackow C, Pu er N, Hordinsky M, Nelson J,Tarrand J, Duvic M: Alopecia areata and cytomegalovirus infection in twins: gene versus environment? JAmAcadDermatol 38:418^415, 1998 Kalish RS, Gilhar A: The immunology of alopecia areata and potential application to novel therapies. DermatolTher 14:322^328, 2001 Kalish RS, Johnson KL, Hordinsky MK: Autoreactive T-cells are variably enriched relative to peripheral blood in the scalp lesions of alopecia areata. Arch Dermatol 128:1072^1077, 1992 Majumder PP, Das SK LICC: A genetic model for vitiligo. Am J Hum Gen 43: 119^125, 1988 McElwee KJ, Pickett P, Oliver RF: The DEBR rat, alopecia areata and autoantibodies to the hair follicle. Br J Dermatol 134:55^63, 1996 McElwee KJ, Rushton DH, Trachy R, Oliver RE: Topical FK506: A potent immunotherapy for alopecia areata? Studies using the Dundee experimental bald rat model. Br J Dermatol 137:491^497, 1997 Michie HJ, Jahoda CAB, Oliver RF, Johnson BE: The DEBR rat: An animal model of human alopecia areata. Br J Dermatol 125:94^100, 1991 Muller SA, Winckelmann RK: Alopecia areata. Arch Dermatol 88:290^297, 1963 Olsen E, Hordinsky M, McDonald-Hull SP, Price V, Roberts J, Shapiro J, Stenn K: Alopecia areata investigational guidelines. JAmAcadDermatol40:242^246, 1999 Paus R, Slominski A, Czarnetzki BM: Is alopecia areata an autoimmune-response against melanogenesis-related proteins, exposed by abnormal MHC Class I expression in the anagen hair bulb? Yale J Biol Med 66:541^554, 1994 Rossi R, Johansson O: Cutaneous innervation and the role of neuronal peptides in cutaneous in ammation: A mini-review. Eur J Dermatol 8:299^306, 1998 Simone DA, Nolano M, Johnson T,Wendelschafer-Crabb G, Kennedy WR: Intradermal injection of capsaicin in humans produces degeneration and subsequent reinnervation of epidermal nerve bers: Correlation with sensory function. J Neurosci 18:8947^8959, 1998 Sundberg JP, Cordy WR, King LE Jr: Alopecia areata in aging C3H/HeJ mice. J Invest Dermatol 102:847^856, 1994 Sundberg JP, King LE Jr: Mouse models for the study of human hair loss. Dermatol Clin 14:619^632, 1996 Tobin DJ, Fenton DA, Kendall MD: Ultrastructural observations on the hair bulb melanocytes and melanosomes in acute alopecia areata. J Invest Dermatol 94:803^817, 1990

6 78 HORDINSKY AND ERICSON JID SYMPOSIUM PROCEEDINGS Tobin DJ, Hann SK, Song MS, Bystryn JC: Hair follicle structures targeted by antibodies in patients with alopecia areata. Arch Dermatol 133:57^61, 1997 Tobin DJ, Sundberg JP, King LE Jr, Boggess D, Bystryn JC: Autoantibodies to hair follicles in C3HeJ mice with alopecia areata-like hair loss. J Invest Dermatol 109:329^333, 1997 Toyoda M, LuoY, Makino T, Matsui C, Morohashi M: Calcitonin gene-related peptide upregulates melanogenesis and enhances melanocyte dendrictiy via induction of keratinocyte-derived melanotrophic factors. J Investig Dermatol Symp Proc 4:116^125, 1999 Wallengren J: Vasoactive peptides in the skin. J Investig Dermatol Symp Proc 2:49^55, 1997 Wang SJ, Shohat T, Vadheim C, Shellow W, Edwards J, Rotter JI: Increased risk for Type I (insulin-dependent) diabetes in relatives of patients with alopecia areata (AA). Am J Med Gen 51:234^239, 1994 Welsh EA, Clark HH, Zane S, et al: Human leukocyte antigen-dqb01 03 alleles are signi cantly associated with alopecia areata. J Invest Dermatol 103:758^763, 1994 Whiting DA: The histopathology of alopecia areata in vertical and horizontal sections. DermatolTher 14:297^305, 2001 Zhang JG, Oliver RF: Immunohistological study of the development of the cellular in ltrate in the pelage follicles of the DEBR model for alopecia areata. Br JDermatol130:405^414, 1994