Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Immunol. Author manuscript; available in PMC 2011 February 15.
Published in final edited form as:
PMCID: PMC2887735

Mechanisms of spatial and temporal development of autoimmune vitiligo in tyrosinase specific TCR transgenic mice1


Generalized vitiligo is thought to have an autoimmune etiology, and has been correlated with the presence of CD8 T cells specific for melanocyte differentiation antigen. However, limited animal models for the disease have hampered its understanding. Thus, we generated TCR transgenic mice that recognize an epitope of the melanocyte protein, tyrosinase. These animals develop vitiligo with strikingly similar characteristics to the human disease. Vitiligo develops temporally and spatially, with juvenile lesions forming bilaterally in head and facial areas, and only arising later in the body of adult animals. Vitiligo is entirely dependent on CD8 T cells, while CD4 T cells exert a negative regulatory effect. Importantly, CD8 T cells can be pervasively present in the skin in the steady state without inducing vitiligo in most areas. This points to developmental differences in melanocyte susceptibility and/or immunological effector mechanisms over time, or in different body locations. Disease is strongly dependent on both IFNγ and CXCR3, while dependence on CCR5 is more limited, and both CCR4 and perforin are dispensable. Genetic ablation of CXCR3 or IFNγ also resulted in scarce CD8 T cell infiltration into the skin. Our results identify unexpected complexity in vitiligo development and point towards possible therapeutic interventions.

Keywords: Human, Antigens/peptides/epitopes, MHC, Antigen presentation/processing, Tumor immunity


Non-segmental or generalized vitiligo is an acquired disorder characterized by a chronic and progressive loss of functional melanocytes from the epidermis and follicular reservoir. It affects about 0.5% of the world population with nearly half of patients presenting before 20 years of age (1-3). The disease is characterized by the development of small depigmented macules that enlarge and coalesce into larger patches. Lesions are often symmetrical and localized near eyes, nose, mouth, nipples, umbilicus and genitalia. An autoimmune etiology has been suggested based on detection of skin-homing T cells specific for melanocyte proteins in the circulation and infiltrates from lesions (4-9). Studies have documented the presence of anti-melanocyte antibodies (10, 11) with the capacity to promote melanocyte destruction (12). An autoimmune etiology is further supported by the observation of vitiligo development following melanoma immunotherapy (13-17) and the effectiveness of immunomodulatory agents in the treatment of vitiligo (18).

The available evidence in humans is consistent with a role for infiltrating lymphocytes in the destruction of perilesional melanocytes leading to vitiligo. Perilesional CD4 and CD8 T cells expressing TNFα and IFNγ, and the cytotoxic molecules granzyme and perforin, have been detected in patient skin (19-22). In a skin explant model, perilesional CD8+ T cells were shown to produce IL-17, IFNγ and TNFα and granzyme B upon activation with melanocyte-specific peptide antigens, and to induce melanocyte apoptosis (8). However, specific molecules were not assessed for involvement in melanocyte killing, and the mechanisms of melanocyte destruction remain unclear. In a murine transgenic model, melanocyte destruction by vitiligo-inducing CD4+ T cells was shown to be dependent upon expression of Fas ligand but not perforin (23). However, the cell expressing Fas ligand was not identified. Along with T cell infiltrates, macrophages have also been found to be abundant in vitiligo skin (24), but their role in melanocyte destruction has not been evaluated.

Mechanistic investigations into the factors controlling the development of vitiligo and the basis of melanocyte destruction have been limited by a lack of appropriate animal models. Accordingly, we developed a murine model based on an epitope derived from tyrosinase that is recognized by CD8 T cells (25-27). We used transgenic C57BL/6 mice that express a recombinant class I MHC molecule, AAD, which contains the peptide-binding region of human HLA-A*0201 linked to the CD8-binding domain of murine H-2Dd. AAD+ C57BL/6 mice are tyrosinase+, and endogenously process and present the murine homolog of a previously identified HLA-A*0201 restricted tyrosinase epitope, Tyr369 (25, 28). Most characterized murine mutations in pigmentation genes are substitutions that alter protein folding or function, but not expression, and thus are likely to continue to engender immunological tolerance. However, the c38R145L albino mutation represents a complete deletion of the tyrosinase structural gene (29), which we backcrossed into C57BL/6 mice (25). Immune responses to Tyr369 in AAD+ albinos were robust, while those in AAD+ tyrosinase+ mice were usually undetectable (25, 30, 31). We subsequently created a TCR transgenic mouse using a Tyr369 + AAD-specific receptor, called FH, isolated from these non-tolerant mice.

We previously showed that FH cells do not undergo central tolerance, despite expression of low levels of tyrosinase mRNA in the thymus (32, 33), and the FH TCR is expressed on most peripheral CD8 T cells of both albino and tyrosinase+ AAD+ mice. However, FH cells transferred into AAD+ tyrosinase+ recipients undergo activation and deletion based on encounter with a population of LN stromal cells that express tyrosinase directly. In the present work, we have explored another important characteristic of FH mice: their development of vitiligo. We show that vitiligo in FH mice displays a spatial and temporal development similar to that observed in the human disease, and explore the immunological basis for its occurrence. Our results reveal an unexpected spatial and temporal complexity in vitiligo development, and suggest that depigmentation processes occurring in different areas of the body may be mechanistically distinct.



FH transgenic mice were generated using TCR genes isolated from a T cell clone specific for a peptide corresponding to amino acids 369–377 of tyrosinase (Tyr369)5, and derived from an AAD+ albino mouse (32). Chemokine receptor and effector molecule deficient FH mice were produced by crossing FH+ AAD+ parentals to mice deficient for CXCR3-/-, CCR4 (generous gift from Eugene Butcher, Stanford University, Stanford, CA), CCR5, perforin, and IFNγ (all from The Jackson Laboratory, Bar Harbor, ME). Animals were maintained in pathogen-free facilities at the University of Virginia and procedures were approved by the University of Virginia Institutional Animal Care and Use Committee.

Depletion of T cells

For depletion of CD4+, CD8+ and CD25+ T cells in vivo, FH mice were injected i.p. with 100 μg/mouse of anti-CD4 (clone GK1.5), anti-CD8 (clone 2.43) or anti-CD25 antibody (clone PC61) within 24 hours of birth and 7 days later. Treated mice received additional injections of 300 μg/mouse (anti-CD4 or anti-CD8) or 500 μg/mouse (anti-CD25) weekly for 13 additional weeks. Rat IgG (100-500 μg/mouse) was used as a control. In some FH mice, animals received one injection of anti-CD25 antibody (500 μg/mouse) beginning at 20 weeks of age and weekly thereafter through 27 weeks.


Ears and back skin from 2-day- and 49-day-old FH mice were embedded in Tissue-Tek® OCT frozen at -80°C and cut into 5 μm thick sections. Slides were fixed in cold acetone, incubated in 0.3% hydrogen peroxide to block endogenous peroxidase activity, and blocked with 10% normal donkey serum (Jackson ImmunoResearch). Slides were then incubated with mouse anti-CD4 (clone GK1.5), anti-CD8 (BD Biosciences), anti-F4/80 (eBiosciences) or Rat IgG2a isotype control (eBiosciences) diluted in 10% normal donkey serum, followed by biotinylated donkey anti-rat Ab (Jackson ImmunoResearch) and then streptavidin-horseradish peroxidase. Antibody reactivity was detected with VECTOR® NovaRed™ substrate (Vector Laboratories) according to the manufacturer’s instructions. The slides were counterstained with hematoxylin, dehydrated with alcohol, cleared with xylene and cover slips were mounted using CytoSeal XYL mounting solution (Fisher Scientific).

Induction of vitiligo by adoptive transfer of purified FH CD8+ T cells

IFNγ+/+ and IFNγ-/- FH CD8+ T cells were isolated from spleen and lymph nodes by positive selection on microbeads (Miltenyi Biotech) according to manufacturer’s instructions. Purified CD8+ T cells (3×106) were injected i.v. into sublethally irradiated (600 rads) AAD+ non-transgenic (nonTg) C57BL/6 mice (8-10 weeks of age). Recipient mice also received i.p. injections of IL-2 (1500 CU) at the time of T cell transfer and every other day up to 10 days.


Spontaneous vitiligo development in FH mice

Despite the occurrence of peripheral self-tolerance after adoptive transfer of FH T cells into AAD+ tyrosinase+ recipients, all AAD+ tyrosinase+ mice that express FH genetically show progressive depigmentation of epidermis and hair follicles. No depigmentation was observed in AADneg FH mice, despite positive selection of FH T cells (32), demonstrating that the target antigen is Tyr369 + AAD. We developed a semiquantitative grading system based upon the natural progression of depigmentation, which shows distinct juvenile and adult patterns (Table 1). Juvenile vitiligo begins with uniform bilateral depigmentation of the ear epidermis by postnatal day 3, when such pigmentation is normally first visible (Fig. 1A, B). A ring of epidermal depigmentation around each eye is observed as early as 5d after birth and is complete in all mice by 21d. Depigmentation of the muzzle usually begins 1-2d after eye rings form. Tail depigmentation can begin in differing sites along the tail with the most common appearance near the midpoint at 21d, progressing to the tip, and is completed by 36d. Depigmentation in these areas is regular, complete, and all mice have progressed to a maximum juvenile pattern vitiligo score by 6 weeks of age (Fig. 1B). At 7 weeks of age, FH mice begin to display reduced pigmentation of hair follicles of the body (Fig. 1A, C). Initial alterations include both “salt/pepper” patches where both fully pigmented and reduced pigmented hairs are intermixed, and white patches of complete depigmentation (Fig. 1A). “Salt/pepper” areas may progress to a more uniform gray coloration and then develop into depigmented, white patches. Some, but not all, white patches expand over time, and may ultimately result in substantial depigmentation over the entire body with the exception of the top of head. Adult pattern vitiligo progression ceases at 20-21 weeks of age (Fig. 1C). Vitiligo development in the FH mouse did not have a gender preference and both sexes were used in subsequent studies.

Figure 1
Spatial and temporal development of vitiligo in FH mice. (A) Progression of depigmentation in FH mice as shown at the indicated ages; shown with an unaffected littermate nontransgenic control (top right panel). Note ears and eye ring depigmentation at ...

CD4 and CD8 T cell contributions to juvenile and adult vitiligo in FH mice

Peripheral lymphoid organs of tyrosinase+ FH mice are populated with large numbers of CD8 T cells expressing this transgenic receptor (32). The numbers of FH cells are somewhat lower in tyrosinase+ FH mice than their albino counterparts, reflecting deletional tolerance (32). However, FH CD8 T cells in both types of mice are >90% CD62L+, >80% CD127+, < 3% CD69 and <1% CD25 (not shown). Thus, there is no evidence for significant numbers of effector or memory cells in tyrosinase+ FH mice.

To evaluate the roles of CD4 and CD8 T cells in the development of vitiligo in the FH mice, we administered anti-CD4 and anti-CD8 depleting mAbs weekly beginning 12-24h after birth and continuing to 15 weeks of age. Depletion of CD8+ T cells resulted in complete abrogation of vitiligo development (Fig. 2A). However, depletion of CD4+ T cells accelerated the development of both juvenile and adult vitiligo, and increased the end-stage severity of the adult form (Fig. 2A). Because of the well-established role of CD4+CD25+ T regulatory (Treg) cells in controlling autoimmunity, we next administered anti-CD25 mAb weekly from birth to 15 weeks of age. Vitiligo development was again significantly accelerated, with kinetics and severity comparable to those of anti-CD4 treated mice (Fig. 2B, left panel). Interestingly, although vitiligo progression in untreated FH mice stabilized by 20-21 weeks of age, half of these anti-CD25-treated animals continued to depigment for up to 30 weeks (data not shown). In addition, administration of anti-CD25 mAb weekly from 20-27 weeks of age led to a resumption of vitiligo progression (Fig. 2B, right panel). Collectively, these results indicate that FH CD8 T cells are sufficient to cause juvenile and adult vitiligo in the absence of CD4 TH cells, and that CD4+CD25+ T cells are involved as negative regulators of both forms of the disease.

Figure 2
Role of CD4 and CD8 T cells in FH vitiligo development. (A) FH mice were intraperitoneally injected with anti-CD4 (open circles, n=8), anti-CD8 (shaded circles, n=6) or IgG isotype control (boxes, n=8) weekly from birth through 15 weeks of age and assessed ...

Vitiligo develops in close association with CD8 T cell infiltration

The sites of juvenile depigmentation represent those in which melanocytes are located in the epidermis of mice, while sites of adult depigmentation are generally in the hair follicles. To gain further insight into the basis for this difference, we evaluated the pattern of vitiligo development in adult mice that had been treated with anti-CD8 for 15 weeks, after cessation of Ab treatment. Depigmentation was not evident until 10 weeks after the last Ab treatment, and then occurred with a similar time course at most sites associated with both juvenile and adult patterns of the disease (Fig. 2C). Interesting however, no depigmentation was evident in the ears. As will be shown in Fig. 7 below, this reflects the fact that CD8 T cells are found in significant numbers in juvenile, but not adult ear skin. More generally, these results suggests that the progressive, location-dependent development of vitiligo in young AAD+ FH mice reflects differences in the ability of T cells to access melanocytes in different locations in the body, and/or in susceptibility of melanocytes in different body locations to immune destruction.

Figure 7
Differential dependence of skin and ear cellular infiltrates upon CXCR3 and IFNγ. CD8 T cells (top and middle panels) and F4/80+ cells (bottom panels) from 2 day old juvenile (left panels) and 49 days adult (right panels) skin of the indicated ...

To determine whether the timing of depigmentation is associated with differences in the extent of T cell infiltration into the skin or with a secondary process, 15-week-old FH mice were sacrificed and skin sections examined by immunohistochemistry for the presence of CD8 and CD4 T cells. A relatively low number of CD8 T cells were associated with nonTg hair follicles, and a similar number were found in follicles of non-lesional skin areas of FH mice (Fig. 3). However, there was a substantial increase in the number of CD8 T cells associated with hair follicles in perilesional (grey bordering fully depigmented) areas. The number of CD8 T cells found associated with follicles of lesional (fully depigmented) areas was also significantly elevated, albeit to a lesser extent. In contrast, there was a mild, but not statistically significant increase in the numbers of CD4 T cells in these same sites (Fig. 3B). Thus, vitiligo develops in close association with an increase in CD8 T cell infiltration into sites containing melanocytes, but CD8 T cells may also be present without evidence of melanocyte destruction or depigmentation.

Figure 3
Vitiligo is associated with increased infiltration of CD8, but not CD4 cells, into perilesional skin. (A) CD8 T cells were identified by immunohistochemical staining of nonlesional, perilesional (gray bordering fully depigmented), and lesional (fully ...

Development of vitiligo depends on CXCR3 and CCR5, but not CCR4

To gain further insight into the factors that might control infiltration of FH T cells into the skin, we evaluated the role of different chemokine receptors in the development of juvenile and adult vitiligo. We produced AAD+ FH mice that carried knockouts in CCR4, CCR5, or CXCR3. These receptors were chosen due to their reported importance in the trafficking of T cells to the skin (CCR4) (34) and to sites of inflammation (CCR5 and CXCR3) (35-38). The development of both juvenile and adult vitiligo in FH CCR4-/- mice was regionally and kinetically indistinguishable from that of FH CCR4+/+ animals (Fig. 4A). However, juvenile pattern vitiligo development was significantly impaired in both FH CCR5-/- (Fig. 4B) and FH CXCR3-/- mice (Fig. 4C). In a majority of these animals, eye rings, muzzle, and tails remained partially pigmented over the entire 25-week observation period (Fig. 4, ,5),5), although depigmentation of ears was unaffected (Fig 5). Interestingly, adult pattern vitiligo developed normally in FH CCR5-/- mice (Fig. 4B, ,5),5), but was substantially diminished in FH CXCR3-/- mice over the entire period of 6-21 weeks (Fig. 4C, ,5).5). These results indicate that both CCR5 and CXCR3 contribute to the development of vitiligo in FH mice, but their impact depends upon the body site. These results also suggest that the development of vitiligo in these different sites depends on distinct mechanisms.

Figure 4
Development of vitiligo depends upon CCR5 and CXCR3, but not CCR4. (A) CCR4+/+ (boxes, n=10), CCR4+/- (open circles, n=15) and CCR4-/- (shaded circles, n=17) FH mice were scored for vitiligo through 21 weeks. (B) CCR5+/+ (boxes, n=10), CCR5+/- (open circles, ...
Figure 5
Representative vitiligo appearance in FH mice lacking CCR4, CCR5 or CXCR3. Left panels depict juvenile pattern vitiligo at 21 weeks of age. Right panels depict adult pattern vitiligo at 18 weeks of age with the exception of the CCR4-/- FH mice, which ...

Development of vitiligo depends on IFNγ but not perforin

To gain insight into the effector mechanism(s) of melanocyte destruction mediated by FH cells, we also produced AAD+ FH mice carrying knockouts of the genes for either perforin or IFNγ. Surprisingly, lack of perforin had only a minor, albeit statistically significant, effect on the development of juvenile pattern vitiligo, and no impact on the development of adult pattern vitiligo (Fig. 6A). Strikingly, the absence of IFNγ led to a nearly complete absence of juvenile pattern vitiligo and a dramatic reduction in the adult form over the 21 week observation period (Fig. 6B). The only evidence of juvenile pattern vitiligo was partial depigmentation (gray coloration) of the ears, which contrasts to the full depigmentation seen in CCR5-/- and CXCR3-/- FH mice. Since many aspects of the immune response might be influenced by the lack of IFNγ, we evaluated the ability of FH CD8 T cells from IFNγ-/- mice to induce vitiligo after adoptive transfer into sublethally irradiated mice (25). Significant depigmentation beyond that normally induced by irradiation was observed upon transfer of IFNγ competent FH CD8 T cells, but not after transfer of IFNγ-/- FH CD8 T cells (Fig. 6C). These results demonstrate that melanocyte destruction resulting in vitiligo depends on the ability of FH CD8 T cells to express IFNγ, but not perforin.

Figure 6
Development of vitiligo depends upon expression of IFNγ but not perforin. (A) Perforin+/+ (n=7), perforin+/- (n=7) and perforin-/- (n=8) FH mice were scored for juvenile and adult vitiligo. **, P<0.008 compared with Perforin+/+ FH mice. ...

Influence of CXCR3 and IFNγ on the infiltration of CD8 T cells and macrophages into juvenile and adult skin

CXCR3 has previously been associated with recruitment of CD4 and CD8 T cells to sites of inflammation, including the skin (35). IFNγ induces a multiplicity of effects, including expression of CXCR3 by T cells (39), the expression of its ligands in peripheral tissue (40), increased expression of ICAM-1 on endothelial cell surfaces (41), and macrophage activation (42). To gain insight into the factors leading to diminished vitiligo in FH CXCR3-/- and FH IFNγ-/- mice, we examined the infiltration of CD8 T cells into non-lesional juvenile and adult skin (including dermis, epidermis and hair follicles). Both knockouts reduced the number of CD8 T cells in juvenile skin to undetectable levels, and significantly diminished their representation in adult skin (Fig. 7, top panels). The modestly higher representation in the skin of adult CXCR3-/- compared to adult IFN-/- mice is consistent with the higher level of residual depigmentation seen in CXCR3-/- mice (compare Fig 4 and and6).6). These results extend the work shown in Fig. 3, and suggest that vitiligo depends upon the steady-state infiltration of CD8 T cells into non-lesional as well as perilesional areas.

Interestingly, infiltration of CD8 T cells into the juvenile ears of FH CXCR3-/- and FH IFNγ-/- mice was comparable to that of normal FH mice (Fig. 7, middle panels). This is in accord with the complete ear depigmentation that occurs in CXCR3-/- mice. However, it is quite surprising in light of the diminished ear depigmentation seen in IFN-/- mice, and suggests that IFNγ plays a more direct role in promoting melanocyte destruction, at least in this site. In addition, CD8 T cell infiltration into adult ears was uniformly undetectable. This is consistent with the observation that ears in chronically CD8-depleted adult animals do not depigment when CD8 T cells are allowed to reconstitute (Fig 2C). It also suggests that CD8 T cell infiltration of both juvenile and adult ear tissue is a mechanistically distinct process from that occurring in other vitiligo sites.

Macrophages have previously been implicated as effector cells in tissue destruction associated with other autoimmune diseases (43-45) and macrophage infiltration parallels infiltration of T cells in lesional vitiligo skin (24). While macrophages do not express CXCR3 (46), they are activated by IFNγ (42). Hence, we examined the macrophage accumulation in juvenile and adult skin of both CXCR3-/- and IFNγ-/- FH mice. Neither knockout influenced the representation of F4/80+ macrophages in juvenile skin (Fig. 7, bottom panels), suggesting that the development of juvenile vitiligo does not depend on increased infiltration of macrophages into skin. However, ablation of IFNγ eliminated an age-related increase in the number of macrophages evident in the skin of normal adult FH mice, while the number in the skin of adult CXCR3-/- FH mice was significantly diminished. The correlation between infiltration of macrophages and CD8 T cells in adult skin of these different strains suggests that trafficking of CD8 T cells into skin may promote recruitment of macrophages through secretion of IFNγ, as well as their activation. Since macrophages do not express CXCR3 (46), this would presumably occur through induction of additional chemokine receptors or their ligands.


Generalized vitiligo in humans has been correlated with the presence of CD8 T cells specific for melanocyte differentiation antigen in perilesional skin and blood (4-9). In the present study, we have characterized a murine CD8 T cell dependent autoimmune vitiligo model with features resembling the generalized human disease. Vitiligo in this model is slowly progressive, with juvenile lesions forming bilaterally in head and facial areas, and only arising later in the body of adult animals. Development of vitiligo is entirely dependent on CD8 T cells, while CD4 T cells exert a negative regulatory effect. Thus, it differs significantly from another model involving TCR transgenic expressing CD8 T cells specific for another melanocyte differentiation protein, gp100. This TCR was derived from a gp100+ mouse, and vitiligo development depends on both CD4 T cells and genetic deficiency of CTLA-4 (47). It also differs from another TCR transgenic model involving CD4 T cells specific for TRP-1, in which vitiligo fails to develop because a relatively low number of TCR transgenic T cells are efficiently deleted (48). However, disease can be induced upon transfer into sublethally irradiated recipients, as we demonstrate here. Our model is most similar to one involving mice expressing lysozyme under the control of the TRP-2 promoter in conjunction with a class II MHC-restricted transgenic TCR, in which disease develops spontaneously with what appears to be a similar temporal and spatial progression, but does not initiate for 3 weeks (23).

The ability to quantify disease progression over time and in different locations allowed inquiry into the factors contributing to depigmentation. Our work demonstrates a strong direct association of vitiligo with a local increase in CD8 T cells in actively depigmenting perilesional areas relative to adjacent non-pigmented or fully depigmented areas. In addition, the impaired development of vitiligo in CXCR3-/- mice was associated with greatly diminished infiltration of CD8 T cells into skin. CD8 effectors normally acquire CXCR3, and this chemokine receptor has been shown to mediate infiltration into a variety of inflamed sites (35-38). These results are consistent with a model in which the temporal and spatial pattern of depigmentation develops in association with localized inflammatory processes that result in the expression of CXCR3 ligands, and lead to the recruitment of CD8 T cells into the skin. However, this simple model does not readily account for other observations in these mice. One exceptional site is the ears, which depigment normally in juvenile CCR5-/- and CXCR3-/- FH mice. Ablation of these chemokine receptors does not alter the normal representation of CD8 T cells in the ears of these mice, suggesting that alternate receptors are employed. In keeping with this, adult animals allowed to regain CD8 T cells after sustained depletion from birth show minimal CD8 infiltration and no evidence of ear depigmentation, suggesting that the receptors or ligands mediating ear infiltration are developmentally regulated.

The rapid development of vitiligo in facial and head areas of juveniles demonstrates the presence of immunological effector mechanisms with the potential to cause complete depigmentation. However, there is also a significant infiltration of CD8 T cells into nonlesional skin and hair follicles in juveniles, yet these latter areas do not begin to depigment for 2 months. Conversely, body hair and facial areas depigment simultaneously in adult animals allowed to regain CD8 T cells after sustained depletion from birth. It has been suggested that the development of vitiligo reflects a differential susceptibility of melanocytes to immune mediated destruction (49). Thus, it is possible that body hair melanocytes in juveniles are less susceptible to immune destruction than those of the head and body areas. Interestingly, juvenile pattern depigmentation is partially blocked by ablation of either CCR5 or CXCR3, while adult pattern depigmentation is only sensitive to ablation of CXCR3. The strong dependence of CD8 T cell infiltration into juvenile skin on CXCR3 suggests that CCR5 plays a different role. CCR5 signaling enhances the activation of both CD8 T cells (50) and macrophages (51, 52). This suggests that there are differences in the quality or activation state of immunological effectors infiltrating these distinct body areas as well. While further work will be required to distinguish these possibilities, our results point to factors over and above simple CD8 T cell infiltration that control vitiligo development.

Our work also provides some insight into the mechanisms responsible for melanocyte destruction. It was recently shown that perilesional CD8 T cells directly destroyed melanocytes in a skin explant model through an apoptotic mechanism (8), but the mode of induction was not defined. In a transgenic model involving CD4 T cell recognition of melanocyte expressed antigen, it was shown that vitiligo development was partially dependent upon expression of Fas-ligand (23). We found that perforin was not necessary for normal lesion development, with the exception of incomplete depigmentation of epidermis surrounding the eyes and the muzzle, although we cannot exclude a redundant contribution of this pathway to melanocyte loss. However, we found IFNγ to be essential. This is consistent with earlier work showing that transgenic expression of IFNγ in murine skin results in accumulation of T lymphocytes and macrophages into the epidermis, reduced numbers of melanocytes in the hair follicles and hypopigmentation, and in some animals, alopecia (53).

The mechanism of action of IFNγ in relation to vitiligo development is likely to be multifaceted. IFNγ expression can enhance CD8 T cell infiltration into skin in multiple ways, including upregulation of CXCR3 (39), the induction of CXCR3 ligands in peripheral tissue (40), and increased expression of ICAM-1 on endothelial cell surfaces (54). The reduced number of CD8 T cells infiltrating adult skin of both CXCR3-/- and IFNγ-/- mice is consistent with a model in which activation of resident CD8 T cells in the skin leads to local expression of IFNγ, and enhanced expression of ICAM-1 and CXCL9, 10, and 11. This would lead to the recruitment of additional CD8 T cells. However, the diminished CD8 T cell representation is evident in nonlesional skin of both juveniles and adults. This suggests that there are at least two distinct IFNγ dependent processes occurring in these mice: one leading to steady-state infiltration of CD8 T cells into skin, and another leading to augmented infiltration, perhaps in association with more fulminant effector function. Only this latter process leads to vitiligo development. An important, but unresolved issue, is whether this latter process is due to local activation of CD8 T cells already present in skin, or to activation of CD8 T cells in the draining LN that then enter the skin in a CXCR3 and IFNγ –dependent manner. In keeping with this, most T cells in FH AAD+ tyrosinase+ mice show no evidence of ongoing activation (unpublished), presumably because of limited availability of antigen or APC. Thus, we hypothesize that the development of vitiligo is also dependent on changes in the presentation of tyrosinase that lead to immunogenic rather than tolerogenic T cell activation.

Our data also suggests that IFNγ may play a more direct role in melanocyte destruction apart from its influence on CD8 T cell infiltration. First, vitiligo induced by adoptive transfer of FH T cells into sublethally irradiated recipients is diminished when those cells do not express IFNγ. More directly, CD8 T cell infiltration into juvenile ears was not influenced in either CXCR3-/- or IFNγ-/- mice. However, while vitiligo developed normally in the ears of CXCR3-/- mice, it was significantly impaired in those lacking IFNγ. In several models, IFNγ has been shown to directly induce apoptosis (55, 56), and to enhance expression of Fas (57-59). IFNγ also supports the recruitment and activation of macrophages (42). Activated macrophages produce TNFα, which has previously been shown to induce apoptosis of hair bulb melanocytes (60). Thus, we propose that the contribution of IFNγ to vitiligo development is two fold: it promotes the trafficking of melanocyte-specific CD8 T cells and macrophages to the skin and participates directly or indirectly in melanocyte destruction.

Overall, this report has demonstrated that autoimmune vitiligo is a complex, temporally and spatially determined process. Importantly, CD8 T cells can be pervasively present in the skin in the steady state without inducing vitiligo in most areas. This points to developmental differences in melanocyte susceptibility and/or immunological effector mechanisms over time, or in different body locations. We show that vitiligo development is strongly dependent on both IFNγ and CXCR3, while dependence on CCR5 is more limited, and both CCR4 and perforin are dispensable. Our results suggest that antagonists against IFNγ or CXCR3 during lesion expansion periods may serve to limit T cell influx to the skin and preserve pigmentation.


We thank Janet V. Gorman for excellent technical assistance.


1This work was supported by USPHS grants AI059996, AI068836, and CA78400 to VHE. RKG was supported by USPHS training grant AI07486 and fellowship CA121916. LAN was supported by USPHS training grants A107486 and GM08136.

5nonTg, non-transgenic; Tyr369, peptide corresponding to amino acids 369–377 of tyrosinase; Treg, regulatory T cell

Disclosures The authors have no financial conflict of interest.


1. Handa S, Kaur I. Vitiligo: clinical findings in 1436 patients. J Dermatol. 1999;26:653–657. [PubMed]
2. Huggins RH, Schwartz RA, Janniger CK. Vitiligo. Acta Dermatovenerol Alp Panonica Adriat. 2005;14:137–142. [PubMed]
3. Taieb A, Picardo M. Clinical practice. Vitiligo. N Engl J Med. 2009;360:160–169. [PubMed]
4. Ogg GS, Rod DP, Romero P, Chen JL, Cerundolo V. High frequency of skin-homing melanocyte-specific cytotoxic T lymphocytes in autoimmune vitiligo. J Exp Med. 1998;188:1203–1208. [PMC free article] [PubMed]
5. Palermo B, Campanelli R, Garbelli S, Mantovani S, Lantelme E, Brazzelli V, Ardigo M, Borroni G, Martinetti M, Badulli C, Necker A, Giachino C. Specific cytotoxic T lymphocyte responses against Melan-A/MART1, tyrosinase and gp100 in vitiligo by the use of major histocompatibility complex/peptide tetramers: the role of cellular immunity in the etiopathogenesis of vitiligo. J Invest Dermatol. 2001;117:326–332. [PubMed]
6. Wankowicz-Kalinska A, Le Poole C, van den Wijngaard R, Storkus WJ, Das PK. Melanocyte-specific immune response in melanoma and vitiligo: two faces of the same coin? Pigment Cell Res. 2003;16:254–260. [PubMed]
7. Wankowicz-Kalinska A, van den Wijngaard RM, Tigges BJ, Westerhof W, Ogg GS, Cerundolo V, Storkus WJ, Das PK. Immunopolarization of CD4+ and CD8+ T cells to Type-1-like is associated with melanocyte loss in human vitiligo. Lab Invest. 2003;83:683–695. [PubMed]
8. van den Boorn JG, Konijnenberg D, Dellemijn TA, d van V, Bos JD, Melief CJ, Vyth-Dreese FA, Luiten RM. Autoimmune destruction of skin melanocytes by perilesional T cells from vitiligo patients. J Invest Dermatol. 2009;129:2220–2232. [PubMed]
9. Le Poole IC, Luiten RM. Autoimmune etiology of generalized vitiligo. Curr Dir Autoimmun. 2008;10:227–243. [PubMed]
10. Kemp EH, Waterman EA, Gawkrodger DJ, Watson PF, Weetman AP. Identification of epitopes on tyrosinase which are recognized by autoantibodies from patients with vitiligo. J Invest Dermatol. 1999;113:267–271. [PubMed]
11. Kemp EH, Waterman EA, Hawes BE, O’Neill K, Gottumukkala RV, Gawkrodger DJ, Weetman AP, Watson PF. The melanin-concentrating hormone receptor 1, a novel target of autoantibody responses in vitiligo. J Clin Invest. 2002;109:923–930. [PMC free article] [PubMed]
12. Gilhar A, Zelickson B, Ulman Y, Etzioni A. In vivo destruction of melanocytes by the IgG fraction of serum from patients with vitiligo. J Invest Dermatol. 1995;105:683–686. [PubMed]
13. Sakai C, Kawakami Y, Law LW, Furumura M, Hearing VJ. Melanosomal proteins as melanoma-specific immune targets. Melanoma Res. 1997;3:83–95. [PubMed]
14. Yee C, Thompson JA, Roche P, Byrd DR, Lee PP, Piepkorn M, Kenyon K, Davis MM, Riddell SR, Greenberg PD. Melanocyte destruction after antigenspecific immunotherapy of melanoma: direct evidence of T cell-mediated vitiligo. J Exp Med. 2000;192:1637–1644. [PMC free article] [PubMed]
15. Dudley ME, Wunderlich JR, Robbins PF, Yang JC, Hwu P, Schwartzentruber DJ, Topalian SL, Sherry R, Restifo NP, Hubicki AM, Robinson MR, Raffeld M, Duray P, Seipp CA, Rogers-Freezer L, Morton KE, Mavroukakis SA, White DE, Rosenberg SA. Cancer regression and autoimmunity in patients after clonal repopulation with antitumor lymphocytes. Science. 2002;298:850–854. [PMC free article] [PubMed]
16. Phan GQ, Yang JC, Sherry RM, Hwu P, Topalian SL, Schwartzentruber DJ, Restifo NP, Haworth LR, Seipp CA, Freezer LJ, Morton KE, Mavroukakis SA, Duray PH, Steinberg SM, Allison JP, Davis TA, Rosenberg SA. Cancer regression and autoimmunity induced by cytotoxic T lymphocyte-associated antigen 4 blockade in patients with metastatic melanoma. Proc Natl Acad Sci U S A. 2003;100:8372–8377. [PubMed]
17. Luiten RM, Kueter EW, Mooi W, Gallee MP, Rankin EM, Gerritsen WR, Clift SM, Nooijen WJ, Weder P, van de Kasteele WF, Sein J, van den Berk PC, Nieweg OE, Berns AM, Spits H, de Gast GC. Immunogenicity, including vitiligo, and feasibility of vaccination with autologous GM-CSF-transduced tumor cells in metastatic melanoma patients. J Clin Oncol. 2005;23:8978–8991. [PubMed]
18. Njoo MD, Spuls PI, Bos JD, Westerhof W, Bossuyt PM. Nonsurgical repigmentation therapies in vitiligo. Meta-analysis of the literature. Arch Dermatol. 1998;134:1532–1540. [PubMed]
19. Ongenae K, Van Geel N, Naeyaert JM. Evidence for an autoimmune pathogenesis of vitiligo. Pigment Cell Res. 2003;16:90–100. [PubMed]
20. Abdel-Naser MB, Kruger-Krasagakes S, Krasagakis K, Gollnick H, Abdel-Fattah A, Orfanos CE. Further evidence for involvement of both cell mediated and humoral immunity in generalized vitiligo. Pigment Cell Res. 1994;7:1–8. [PubMed]
21. al Badri AM, Foulis AK, Todd PM, Gariouch JJ, Gudgeon JE, Stewart DG, Gracie JA, Goudie RB. Abnormal expression of MHC class II and ICAM-1 by melanocytes in vitiligo. J Pathol. 1993;169:203–206. [PubMed]
22. Horn TD, Abanmi A. Analysis of the lymphocytic infiltrate in a case of vitiligo. Am J Dermatopathol. 1997;19:400–402. [PubMed]
23. Lambe T, Leung JC, Bouriez-Jones T, Silver K, Makinen K, Crockford TL, Ferry H, Forrester JV, Cornall RJ. CD4 T cell-dependent autoimmunity against a melanocyte neoantigen induces spontaneous vitiligo and depends upon Fas-Fas ligand interactions. J Immunol. 2006;177:3055–3062. [PubMed]
24. Le Poole IC, van den Wijngaard RM, Westerhof W, Das PK. Presence of T cells and macrophages in inflammatory vitiligo skin parallels melanocyte disappearance. Am J Pathol. 1996;148:1219–1228. [PubMed]
25. Colella TA, Bullock TNJ, Russell LB, Mullins DW, Overwijk W, Luckey CJ, Pierce RA, Restifo NP, Engelhard VH. Self-tolerance to the murine homologue of a tyrosinase-derived melanoma antigen: implications for tumor immunotherapy. J Exp Med. 2000;191:1221–1231. [PMC free article] [PubMed]
26. Engelhard VH, Bullock TN, Colella TA, Mullins DW. Direct identification of human tumor-associated peptide antigens and a preclinical model to evaluate their use. Cancer J Sci Am. 2000;6(Suppl 3):S272–S280. [PubMed]
27. Mullins DW, Bullock TN, Colella TA, Robila VV, Engelhard VH. Immune responses to the HLA-A*0201-restricted epitopes of tyrosinase and glycoprotein 100 enable control of melanoma outgrowth in HLA-A*0201- transgenic mice. J Immunol. 2001;167:4853–4860. [PubMed]
28. Skipper JCA, Hendrickson RC, Gulden PH, Brichard V, Van Pel A, Chen Y, Shabanowitz J, Wolfel T, Slingluff CL, Boon T, Hunt DF, Engelhard VH. An HLA-A2 restricted tyrosinase antigen on melanoma cells results from post-translational modification and suggests a novel processing pathway for membrane proteins. J Exp Med. 1996;183:527–534. [PMC free article] [PubMed]
29. Rinchik EM, Stoye JP, Frankel WN, Coffin J, Kwon BS, Russell LB. Molecular analysis of viable spontaneous and radiation-induced albino (c)-locus mutations in the mouse. Mut Res. 1993;286:199–207. [PubMed]
30. Slingluff CL, Jr, Chianese-Bullock KA, Bullock TN, Grosh WW, Mullins DW, Nichols L, Olson W, Petroni G, Smolkin M, Engelhard VH. Immunity to melanoma antigens: from self-tolerance to immunotherapy. Adv Immunol. 2006;90:243–295. [PubMed]
31. Engelhard VH, Bullock TN, Colella TA, Sheasley SL, Mullins DW. Antigens derived from melanocyte differentiation proteins: self-tolerance, autoimmunity, and use for cancer immunotherapy. Immunol Rev. 2002;188:136–146. [PubMed]
32. Nichols LA, Chen Y, Colella TA, Bennet CL, Clausen BE, Engelhard VH. Deletional self-tolerance to a melancyte/melanoma antigen derived from tyrosinase is mediated by a radio-resistant cell in peripheral and mesenteric lymph nodes. J Immunol. 2007;179:993–1003. [PubMed]
33. Derbinski J, Schulte A, Kyewski B, Klein L. Promiscuous gene expression in medullary thymic epithelial cells mirrors the peripheral self. Nat Immunol. 2001;2:1032–1039. [PubMed]
34. Campbell JJ, Haraldsen G, Pan J, Rottman J, Qin S, Ponath P, Andrew DP, Warnke R, Ruffing N, Kassam N, Wu L, Butcher EC. The chemokine receptor CCR4 in vascular recognition by cutaneous but not intestinal memory T cells. Nature. 1999;400:776–780. [PubMed]
35. Mohan K, Cordeiro E, Vaci M, McMaster C, Issekutz TB. CXCR3 is required for migration to dermal inflammation by normal and in vivo activated T cells: differential requirements by CD4 and CD8 memory subsets. Eur J Immunol. 2005;35:1702–1711. [PubMed]
36. Akashi S, Sho M, Kashizuka H, Hamada K, Ikeda N, Kuzumoto Y, Tsurui Y, Nomi T, Mizuno T, Kanehiro H, Hisanaga M, Ko S, Nakajima Y. A novel small-molecule compound targeting CCR5 and CXCR3 prevents acute and chronic allograft rejection. Transplantation. 2005;80:378–384. [PubMed]
37. Mora JR, von Andrian UH. T-cell homing specificity and plasticity: new concepts and future challenges. Trends Immunol. 2006;27:235–243. [PubMed]
38. Liu L, Callahan MK, Huang D, Ransohoff RM. Chemokine receptor CXCR3: an unexpected enigma. Curr Top Dev Biol. 2005;68:149–181. [PubMed]
39. Nakajima C, Mukai T, Yamaguchi N, Morimoto Y, Park WR, Iwasaki M, Gao P, Ono S, Fujiwara H, Hamaoka T. Induction of the chemokine receptor CXCR3 on TCR-stimulated T cells: dependence on the release from persistent TCR-triggering and requirement for IFN-gamma stimulation. Eur J Immunol. 2002;32:1792–1801. [PubMed]
40. Moser B, Wolf M, Walz A, Loetscher P. Chemokines: multiple levels of leukocyte migration control. Trends Immunol. 2004;25:75–84. [PubMed]
41. Hubbard AK, Rothlein R. Intercellular adhesion molecule-1 (ICAM-1) expression and cell signaling cascades. Free Radic Biol Med. 2000;28:1379–1386. [PubMed]
42. Schroder K, Hertzog PJ, Ravasi T, Hume DA. Interferon-gamma: an overview of signals, mechanisms and functions. J Leukoc Biol. 2004;75:163–189. [PubMed]
43. Willcox A, Richardson SJ, Bone AJ, Foulis AK, Morgan NG. Analysis of islet inflammation in human type 1 diabetes. Clin Exp Immunol. 2009;155:173–181. [PubMed]
44. Drexler SK, Kong PL, Wales J, Foxwell BM. Cell signalling in macrophages, the principal innate immune effector cells of rheumatoid arthritis. Arthritis Res Ther. 2008;10:216. [PMC free article] [PubMed]
45. Lassmann H. Models of multiple sclerosis: new insights into pathophysiology and repair. Curr Opin Neurol. 2008;21:242–247. [PubMed]
46. Mantovani A, Sica A, Sozzani S, Allavena P, Vecchi A, Locati M. The chemokine system in diverse forms of macrophage activation and polarization. Trends Immunol. 2004;25:677–686. [PubMed]
47. Gattinoni L, Ranganathan A, Surman DR, Palmer DC, Antony PA, Theoret MR, Heimann DM, Rosenberg SA, Restifo NP. CTLA-4 dysregulation of self/tumor-reactive CD8+ T cell function is CD4+ T cell-dependent. Blood. 2006;108:3818–3823. [PubMed]
48. Muranski P, Boni A, Antony PA, Cassard L, Irvine KR, Kaiser A, Paulos CM, Palmer DC, Touloukian CE, Ptak K, Gattinoni L, Wrzesinski C, Hinrichs CS, Kerstann KW, Feigenbaum L, Chan CC, Restifo NP. Tumor-specific Th17- polarized cells eradicate large established melanoma. Blood. 2008;112:362–373. [PubMed]
49. Westerhof W, d’Ischia M. Vitiligo puzzle: the pieces fall in place. Pigment Cell Res. 2007;20:345–359. [PubMed]
50. Contento RL, Molon B, Boularan C, Pozzan T, Manes S, Marullo S, Viola A. CXCR4-CCR5: a couple modulating T cell functions. Proc Natl Acad Sci U S A. 2008;105:10101–10106. [PubMed]
51. Qiu L, Ding L, Huang J, Wang D, Zhang J, Guo B. Induction of copper/zinc-superoxide dismutase by CCL5/CCR5 activation causes tumour necrosis factor-alpha and reactive oxygen species production in macrophages. Immunol. 2008;128:325–334. [PubMed]
52. Lee C, Liu QH, Tomkowicz B, Yi Y, Freedman BD, Collman RG. Macrophage activation through. J Leukoc Biol. 2003;74:676–682. [PubMed]
53. Carroll JM, Crompton T, Seery JP, Watt FM. Transgenic mice expressing IFN-gamma in the epidermis have eczema, hair hypopigmentation, and hair loss. J Invest Dermatol. 1997;108:412–422. [PubMed]
54. van den Wijngaard R, Wankowicz-Kalinska A, Le Poole C, Tigges B, Westerhof W, Das P. Local immune response in skin of generalized vitiligo patients. Destruction of melanocytes is associated with the prominent presence of CLA+ T cells at the perilesional site. Lab Invest. 2000;80:1299–1309. [PubMed]
55. Kano A, Watanabe Y, Takeda N, Aizawa S, Akaike T. Analysis of IFNgamma-induced cell cycle arrest and cell death in hepatocytes. J Biochem. 1997;121:677–683. [PubMed]
56. Fulda S, Debatin KM. IFNgamma sensitizes for apoptosis by upregulating caspase-8 expression through the Stat1 pathway. Oncogene. 2002;21:2295–2308. [PubMed]
57. Cnop M, Welsh N, Jonas JC, Jorns A, Lenzen S, Eizirik DL. Mechanisms of pancreatic beta-cell death in type 1 and type 2 diabetes: many differences, few similarities. Diabetes. 2005;54(Suppl 2):S97–107. [PubMed]
58. Ruemmele FM, Russo P, Beaulieu J, Dionne S, Levy E, Lentze MJ, Seidman EG. Susceptibility to FAS-induced apoptosis in human nontumoral enterocytes: role of costimulatory factors. J Cell Physiol. 1999;181:45–54. [PubMed]
59. Bernassola F, Scheuerpflug C, Herr I, Krammer PH, Debatin KM, Melino G. Induction of apoptosis by IFNgamma in human neuroblastoma cell lines through the CD95/CD95L autocrine circuit. Cell Death Differ. 1999;6:652–660. [PubMed]
60. Soma T, Ogo M, Suzuki J, Takahashi T, Hibino T. Analysis of apoptotic cell death in human hair follicles in vivo and in vitro. J Invest Dermatol. 1998;111:948–954. [PubMed]