|Home | About | Journals | Submit | Contact Us | Français|
The cutaneous surface of a normal adult individual contains approximately 20 billion T cells, nearly twice the number present in the entire circulation. Recent studies have demonstrated a role for these cells in both normal immunity and in inflammatory skin diseases such as psoriasis. Regulatory T cells protect against autoimmune reactions to self antigens and assist in the resolution of cutaneous inflammation. However, they can also shield tumors from immune detection, allow latent infections to persist and can dysfunction under the conditions present in inflammatory skin diseases. Th17 T cells protect organisms against extracellular pathogens but also play a key role in the pathogenesis of psoriasis. Evidence suggests that effector memory T cells produced during immune reactions survive and persist long term within the skin, providing local and rapid protection against pathogen reexposure. This review summarizes the current understanding of how skin resident T cells contribute to normal and aberrant immunity in the skin.
T and B lymphocytes rearrange the DNA encoding their antigen receptors and therefore have the capacity to recognize any antigen. These cells also serve as the repositories for immunologic memory; lymphocytes generated during an immune response can persist for decades, providing rapid and specific responses to rechallenge. T and B cells thus provide highly flexible and long lasting immunity.
Naïve T cells are found primarily within the blood and lymph nodes. Expression of the homing addressins L-selectin and CCR7 supports the migration of naïve T cells into lymph nodes where they encounter dendritic cells (DC) bearing antigen derived from the peripheral tissues (Baekkevold et al., 2001; Warnock et al., 1998). When naïve T cells encounter their cognate antigen, they differentiate and gain effector functions including cytotoxicity and cytokine production. During this differentiation process, T cells are imprinted to express tissue specific homing addressins that affect their subsequent migration patterns. In mice, naïve transgenic T cells developed expression of the gut homing addressins α4β7 and CCR9 when they first encountered antigen in the gut draining lymph nodes but expressed the skin homing addressins CLA and CCR4 if they encountered antigen first in the skin draining lymph nodes (Campbell and Butcher, 2002; Campbell et al., 1999). Further studies demonstrated that both DC and lymph node stromal cells have the ability to imprint homing receptor expression on T cells (Edele et al., 2008; Mora et al., 2003; Mora et al., 2005). Effector T cells are thus programmed during differentiation to migrate to the tissue from which their cognate antigen was originally derived.
When confronted by an infectious pathogen, the immune system does two important things: it responds and it remembers. Two different types of T cells are generated during an immune response (Sallusto et al., 1999). Effector memory T cells (TEM) up-regulate expression of tissue homing addressins and develop effector functions. These cells predominate in the blood in the early stages of an immune response, migrate into peripheral tissues and effect clearance of the pathogen (Mackay et al., 1990; Whitton and Zhang, 1995). Following the peak of the immune response, most of these cells disappear from the blood and a second population, central memory T cells (TCM) predominates (Razvi et al., 1995). Like naïve T cells, TCM express the lymph node homing addressins L-selectin and CCR7 and generally lack addressins for peripheral tissues. TCM have lower levels of effector functions but can proliferate vigorously and develop into effector T cells when rechallenged with antigen (Sallusto et al., 2004).
Until fairly recently, it was believed that T cells only entered tissues such as the skin under conditions of active inflammation (Kupper and Fuhlbrigge, 2004). This review will discuss recent findings that skin and other tissues are stably colonized by long-lived populations of memory T cells. These tissue resident T cells provide long lasting, local and rapid responses to pathogen reexposure but can also contribute to inflammatory and autoimmune skin diseases.
It has been known for decades that T cells are present in non-inflamed human skin and it has been proposed that these cells may comprise a skin specific immune system (Bos et al., 1987). Comprehensive study of these T cells and an understanding of their true numbers has been difficult because mechanical or enzymatic dissociation of the skin produces very few cells. Despite this difficulty, studies using histologic methods or skin dissociation have successfully demonstrated that >95% of the T cells in normal skin are CD45RO memory T cells, <5% are naïve, most express CLA, 50% express CCR8 and a subset express CCR7 and CCR10 (Bos et al., 1989; Campbell et al., 2001; Homey et al., 2002; Schaerli et al., 2004).
T cells are arguably the most migratory cells in the body. T cells enter and travel through every human tissue with the likely exception of cortical bone. Unlike neutrophils and monocyte derived macrophages, T cell migration into the tissues is not a one way trip. T cells can migrate into peripheral tissues and then return to the circulation via migration through the lymph nodes. A simple method of isolating skin T cells that took advantage of their tendency to migrate towards chemokines produced by dermal fibroblasts extracted surprising numbers of T cells from normal human skin (Clark et al., 2006b). This observation led to the enumeration of T cells in skin and it was found that normal human skin contains about 1 million T cells/cm2 (Clark et al., 2006a). Extrapolation of this finding suggested that the skin of a normal adult contains approximately 20 billion T cells, nearly twice the number present in the entire blood volume. These T cells co-expressed high levels of the skin homing addressins CLA and CCR4, had a diverse T cell repertoire and were polarized to produce a variety of different cytokines in response to T cell stimulation (Clark et al., 2006a). 80% of T cells in skin lacked expression of CCR7/L-selectin, confirming their identity as effector memory T cells. T cells that did express CCR7 and L-selectin also co-expressed the skin addressins CLA and CCR4, suggesting that they were an intermediate phenotype capable of accessing both the skin and secondary lymphoid organs.
At the time of these studies, it was assumed that TEM primarily remained in the circulation until recruited into the tissues at sites of inflammation. Classical skin homing TEM (CLA+, CCR7/L-selectin−) cannot enter the lymph nodes and as a result, should be found either in the blood or the skin. Enumeration of CLA+ T cells in skin and blood demonstrated that >90% of CLA+ skin homing T cells were present in skin under resting, non-inflamed conditions and that less than 10% were in the peripheral circulation (Clark et al., 2006a). Thus, even in the absence of inflammatory stimuli, the vast majority of skin homing TEM were resident in skin and well placed to respond to local challenges.
In considering the treatment of inflammatory skin diseases, we and others have often focused on inhibiting the migration of T cells into skin. E-selectin is expressed by all postcapillary venules in skin and is upregulated with inflammation (Chong et al., 2004; Kupper and Fuhlbrigge, 2004). By binding to CLA expressed by skin homing T cells, E-selectin supports lymphocyte rolling, the first step of T cell entry into the skin (Berg et al., 1991; Kupper and Fuhlbrigge, 2004). It therefore came as a surprise to find that blockade of E-selectin, which should block migration of T cells into skin, was ineffective in the treatment of psoriasis, a T cell mediated inflammatory skin disease (Bhushan et al., 2002).
This result was made more intelligible by an elegant series of experiments in mice grafted with human psoriatic skin. Boyman et al. found that normal appearing, non-lesional skin from patients with psoriasis developed full blown psoriatic lesions when transplanted onto immunodeficient mice but skin taken from a normal patients did not (Boyman et al., 2004). Development of the psoriatic lesion was dependent on the activation and local proliferation of a population of autoreactive T cells transferred with the initial skin graft. These T cells were themselves stimulated into action by the local production of IFNα by plasmacytoid DC (pDC), likely produced in response to the trauma of transplantation (Nestle et al., 2005).
This series of experiments showed two important things about skin resident T cells. First, T cells present in normal appearing skin were able to give rise to a full psoriatic lesion in the absence of T cell recruitment from blood. Thus, T cells resident in even normal appearing skin can initiate full blown immune responses and although migration of T cells into the skin from the blood occurs in many inflammatory conditions, it may not always be required. In light of these experiments, the failure of E-selectin blockade to control psoriasis became understandable. Autoreactive T cells capable of initiating psoriasis reside within the normal appearing skin of patients with psoriasis. Preventing the entry of additional T cells into skin does not address the threat posed by the lymphocytes already present.
The second lesson from these experiments is that autoreactive T cells in the skin behave very differently depending on their local inflammatory environment. Autoreactive T cells exist in normal appearing skin from psoriatic patients but these cells remain quiescent despite living side by side with APC expressing and capable of presenting autoantigens. Only when skin APC are activated to produce IFNα do these autoreactive T cells proliferate, produce cytokines, and initiate an active psoriatic lesion. This finding highlights the importance of understanding how T cells are activated by innate immune cells within the skin and explains why TNF antagonists (etanercept, infliximab, adalimumab) and integrin antibodies that disrupt APC/T cell crosstalk (efalizumab, alefacept) are effective in psoriasis whereas E-selectin blockade is not (Bhushan et al., 2002; Griffiths, 2004).
Additional studies in mice and humans have confirmed that T cells can become activated, proliferate and carry out effector functions locally within the skin. As is described more fully below, HSV-specific T cells in mice and humans remained localized around latently infected nerves and prevented HSV reactivation (Wakim et al., 2008b; Zhu et al., 2007). In patients immunized with BCG and challenged with a PPD, effector T cells proliferated within the skin at the PPD site (Vukmanovic-Stejic et al., 2006). In summary, all elements necessary for a memory T cell response - T cells and APC – are resident within human skin and their interaction can give rise to full secondary immune responses within the skin.
The immune system is faced with the difficult problem of mounting immune responses to dangerous pathogens while maintaining tolerance to the body’s own tissues and to harmless or commensal organisms. Regulatory T cells (Tregs) are one of many mechanisms developed by the immune system to enforce tolerance to harmless and self antigens. Tregs are critical to the development and maintenance of self-tolerance and function by suppressing the activation, cytokine production and proliferation of other T cells (Sakaguchi, 2005). These cells are characterized by high expression of the transcription factor FOXP3 and by their ability to suppress T cell responses in vitro.
We have found that between 5–10% of the T cells resident in normal human skin are FOXP3+ Tregs and that these cells proliferate under conditions similar to those found in inflamed skin (Clark et al., 2008; Clark and Kupper, 2007). This suggests that local proliferation of Tregs in the skin may serve as a brake for cutaneous inflammation. Indeed, Tregs were observed to locally proliferate within human skin during DTH reactions and increased numbers of Tregs were found in the skin lesions of contact dermatitis and resolving fixed drug eruptions (Teraki and Shiohara, 2003; Vukmanovic-Stejic et al., 2008)
Recent experiments in mice support the idea that Tregs decrease inflammatory tone in the skin. Mice with mutations in FOXP3 lack Tregs and develop widespread and lethal autoimmunity, a condition that can be prevented by the transfer of wild type Tregs (Khattri et al., 2003). These mice were reconstituted with Tregs that lacked 1,3-fucosyltransferase VII (FuT7), an enzyme necessary for formation of E-selectin ligands (Malý et al., 1996). These Tregs functioned normally but had a selective inability to migrate into the skin. These reconstituted mice had decreased inflammation in the liver and lungs, consistent with normal homing of Tregs to these sites, but they had marked skin inflammation. These mice were kept in germ-free facilities and not exposed to pathogens or other inflammatory stimuli, suggesting that the activity of Tregs is needed to control inflammation even in normal, non-challenged skin. These experiments also showed that Tregs need to be present within the skin in order to exert their anti-inflammatory effects, as opposed to suppressing skin inflammation by acting in the skin draining lymph nodes. This work suggests that Tregs play a role in maintaining homeostasis in normal skin and that they function by locally suppressing the activity of other T cells resident in skin.
For each mechanism of immune tolerance, there are cancers and pathogens that co-opt it to escape immune detection. Tregs are expanded in patients with many types of cancer and are also often recruited into the tumors themselves (Baecher-Allan and Anderson, 2006). 50% of the T cells present in human squamous cell carcinomas of the skin (SCC) are FOXP3+ Tregs and increased Tregs have also been observed in basal cell carcinomas, primary melanoma and melanoma metastases (Ahmadzadeh et al., 2008; Kaporis et al., 2007; Mourmouras et al., 2007). Topical therapy of human SCC with imiquimod, a topical immunomodulator and TLR7 agonist, reduced the % and suppressive function of Tregs, suggesting that it may be useful in reversing Treg induced tumor tolerance (Clark et al., 2008). Tregs also enable latent infection and induce reactivation of cutaneous leishmaniasis and paracoccidiomycosis in mice and likely play a similar role in humans (Belkaid et al., 2002; Xu et al., 2003).
Tregs have the capacity to be potently anti-inflammatory but they are ineffective under certain conditions. Tregs can suppress the activity of T cells which have received a low to mid strength T cell receptor (TCR) signal but cannot suppress T cells that receive a high avidity TCR signal, such as those observed in memory responses to dangerous pathogens (Baecher-Allan et al., 2002). Tregs therefore act as the immunologic equivalent of a high band pass filter, suppressing T cell responses with TCR avidities below a certain threshold and allowing those with higher avidity signals to proceed. Local production of the cytokine IL-6 also renders T cells resistant to the suppressive effects of Tregs in mice and preliminary work suggests this is also true in humans (Goodman et al., 2007; Korn et al., 2007). Expression of the chemokine receptor CCR5 is important for Treg migration and suppressive ability; Tregs that lack CCR5 are less effective in mouse models of paracoccidioidomycosis, leishmaniasis and graft vs. host disease (Moreira et al., 2008; Wysocki et al., 2005; Yurchenko et al., 2006). Lastly, Tregs can be outnumbered. In a mouse model of sarcoma, it was the percentage of Tregs that determined if a tumor would be tolerated or destroyed by the immune system (Bui et al., 2006).
Considerable numbers of Tregs are present in the skin lesions of psoriasis but these cells have decreased suppressive activity (Sugiyama et al., 2005). IL-6 is increased in psoriatic skin and this cytokine may interfere with the ability of Tregs to suppress inflammation (Goodman et al., 2007; Grossman et al., 1989) Additionally, patients with psoriasis have fewer CCR5+ Tregs than normal individuals and the CCR5+ Tregs that are present in psoriatic patients have decreased function (Sugiyama et al., 2008). Thus it appears that the Tregs present within psoriatic lesions fail to suppress inflammation because they are less active intrinsically and because their activity is antagonized by the pro-inflammatory cytokine milieu in psoriasis.
Th17 cells are a separate lineage of T cells that produce the Th17 cytokines IL-17A, IL-17F, TNFα, IL-21 and IL-22 and depend upon IL-23 for their development, survival and proliferation (Harrington et al., 2005; Park et al., 2005). Th17 cells provide immunity against a variety of extracelluar pathogens, including bacteria such as Klebsiella pneumoniae and fungi such as Cryptococcus neoformans and Candida albicans (Happel et al., 2005; Huang et al., 2004; Kleinschek et al., 2006).
Th17 cells have also been implicated in a variety of inflammatory and autoimmune disorders. IL-17 is increased and Th17 cells are demonstrable in the synovial fluid and tissues of patients with rheumatoid arthritis (Aarvak et al., 1999; Chabaud et al., 1999; Kotake et al., 1999) and in the brain lesions and CSF of patients with multiple sclerosis (Lock et al., 2002; Matusevicius et al., 1999). Polymorphisms in the IL-23R, a receptor required for the development and survival of Th17 cells, are associated with susceptibility to ulcerative colitis and Crohn's disease (Duerr et al., 2006). Th17 cells also contribute to the pathology observed in mouse models of colitis, experimental autoimmune encephalomyelitis and arthritis (Ouyang et al., 2008).
Increasing evidence suggests that Th17 cells are also key players in the pathogenesis of psoriasis. DC and keratinocytes in the skin lesions of psoriasis produce increased amounts of IL-23, a cytokine that supports the development and proliferation of Th17 cells (Kryczek et al., 2008; Lee et al., 2004; Piskin et al., 2006; Wilson et al., 2007; Zaba et al., 2008). Treatment of patients with monoclonal antibodies against IL-12/23p40, a component shared between the IL-12 and IL-23 receptors, led to significant clinical improvements in psoriasis (Kimball et al., 2008; Krueger et al., 2007). Although this finding could not discriminate between causative roles for IL-23 vs. IL-12, a role for IL-23 in psoriasis is supported by genetic studies demonstrating that polymorphisms in the IL-23 receptor and other genes in the IL-23 signaling pathway are associated with psoriasis (Capon et al., 2007; Cargill et al., 2007; Nair et al., 2009; Nair et al., 2008). Th17 cells are demonstrable in psoriatic lesions and are found in higher numbers than in normal human skin (Kryczek et al., 2008; Lowes et al., 2008). Th17 cells produce the cytokine IL-22, a cytokine that induces human keratinocyte proliferation and acanthosis in vitro (Sa et al., 2007). Injection of IL-23 into the skin of mice induced dermal inflammation and epidermal acanthosis reminiscent of the changes seen in psoriasis and these effects that were found to be mediated through production of IL-22 (Chan et al., 2006; Zheng et al., 2007). These findings support a role for IL-23 in supporting the survival, proliferation and function of Th17 T cells that in turn contribute to psoriasis through their production of IL-22 and other inflammatory cytokines. These observations suggest that selective disruption of IL-23 signaling (for example, by targeting IL-23p19 subunit) should lead to reduced Th17 T cells and potentially to long lasting improvements in psoriasis.
TCM persist long term in the circulation following resolution of an immune response, a finding that led researchers to propose that TCM are the only T cells responsible for maintaining long term immunologic memory (Lanzavecchia and Sallusto, 2005). This contention is at odds with clear and elegant experiments in mice demonstrating that TEM do persist long term—not in the blood but within the peripheral tissues (Masopust et al., 2001; Reinhardt et al., 2001). In animal models, T cells generated during immune responses in the skin, gut and lungs persist within these tissues and provide protection against re-infection at these sites (Hogan et al., 2001b; Liang et al., 1994; Stittelaar et al., 2005; Xu et al., 2004). A series of elegant studies examining HSV infection of the skin in mice have provided some intriguing details about the nature of these cells. Following infection with HSV, CD8 antigen specific T cells accumulated in the skin near latently infected dorsal root ganglia and suppressed reactivation of the virus (van Lint et al., 2005; Wakim et al., 2008b). Re-infection with HSV led to local proliferation of these resident cells and also recruitment of additional antigen specific T cells from the circulation (Wakim et al., 2008a). There were therefore two populations of antigen specific T cells capable of responding to rechallenge: skin resident T cells that arose and were on site as a result of the primary infection and circulating antigen specific T cells that were newly recruited from the blood. Using GFP-expressing mice and serial transplantation of latently infected ganglia, this group dissected the contributions and characteristics of these two cell types (Gebhardt et al., 2009). Surprisingly, antigen specific tissue resident T cells generated during the primary immune response remained in the same location over 100 days after primary HSV infection. These T cells failed to reenter the circulation and did not even migrate into infected ganglia transplanted in direct approximation. Antigen specific T cells were present in highest numbers at the site of initial infection but were also demonstrable in other locations within the skin. These tissue resident T cells provided enhanced local protection against reinfection with HSV. Skin resident T cells lacked L-selectin and expressed higher levels of the activation antigen CD69 than T cells recruited from blood, a finding that has been replicated in humans (Clark et al., 2006a). These studies demonstrated that a functionally distinct, non-migratory population of skin resident T cells arose following viral infection, persisted long-term within the skin and provided effective protection against local reinfection. In humans, a similar population of HSV-specific T cells was found resident in the skin of military recruits more than two months after clearance of a primary HSV infection and these T cells increased in number during subclinical HSV reactivation. Antigen specific T cells also persisted long term locally in the skin at the site of PPD injection in BCG-vaccinated individuals (Vukmanovic-Stejic et al., 2006).
The finding that skin resident T cells are a sessile, nonmigratory population may help to explain some of the puzzling eruptions we see in dermatology. Fixed drug eruptions (FDE) are local skin lesions that occur following ingestion of a causative drug, spontaneously resolve once the drug is discontinued, then recur in the same location years or even decades later when the drug is taken again. Histologically, this eruption is a cytotoxic reaction against epidermal keratinocytes mediated by a CD8 T cells and indeed, a population of CD8 T cells producing IFNγ and TNFα is found within the epidermis of clinically resolved, hyperpigmented FDE lesions (Shiohara and Moriya, 1997; Teraki et al., 1994; Teraki and Shiohara, 2003). The fixed location of this eruption may reflect the fixed location of a population of drug reactive T cells resident in skin. Similarly, it is striking how psoriatic plaques remain in the same location long-term and tend to recur in the same locations following cessation of suppressive therapy. Given the evidence that autoreactive Th17 cells may drive this disease, it is tempting to speculate that each psoriatic plaque represents a population of autoreactive skin resident Th17 cells. However, changes in other cell types could also mediate this localized behavior, including fixed populations of pathogenic APC or possibly fixed changes in the blood vessels at these sites. Lastly, the different behavior of central memory and skin resident effector memory T cells may underlie the differing clinical manifestations we observe in cutaneous T cell lymphoma. In patients with stage IA mycosis fungoides (MF), malignant T cells are confined to stable patches and plaques on the skin and life expectancy is normal (Kim et al., 2003). In Sézary syndrome, the malignant T cells migrate throughout the entire skin surface, giving rise to erythroderma, and also colonize the blood and lymph nodes. Intriguing new data suggests that the malignant T cells in MF are skin resident effector memory cells, a population expected to remain in a fixed position. However, malignant T cells in Sézary syndrome bear markers suggestive of central memory T cells, a cell type that normally migrates through the blood, lymph nodes, and can also be found in low numbers in normal human skin (Campbell et al., 2009; Clark et al., 2006a). The fact that malignant central memory T cells give rise to diffuse erythroderma and malignant skin resident T cells give rise to fixed plaques of inflamed skin is an eloquent argument in favor of the mobile versus sessile nature of these two lymphocyte subsets.
Other tissues have their own populations of resident T cells that contribute local immune protection. In mice, a population of antigen-specific T cells remained resident in the lung several months after recovery from influenza or Sendai virus infection (Hogan et al., 2001a). It was the number of lung resident pathogen-specific T cells and not the number in the lymph nodes that correlated with protection against re-infection. The gut epithelium in mice is colonized by CD8 memory T cells with substantial lytic activity (Masopust et al., 2001). Murine intraepithelial CD8 cells bearing the αβ TCR and both αβ CD8 chains (comparable to conventional T cells) are scarce in the gut at birth but progressively increase throughout life as antigen-experienced T cells accumulate (Cheroutre, 2004; Cheroutre and Madakamutil, 2004). Gut resident T cells isolated from mice that have recovered from pathogens such as Toxoplasma gondii protect against infection when transferred into naïve animals (Lepage et al., 1998).
To conceptualize how TEM and TCM contribute to immunity, it may be helpful to describe their contributions to both naïve and memory T cell responses in the skin. During a primary immune response, when the immune system encounters an antigen for the first time, the antigen is taken up by skin resident DC and these cells migrate to the skin draining lymph nodes (Figure 1). Within the lymph nodes, DC present antigen to naïve T cells. Upon recognition of cognate antigen, naïve T cells undergo differentiation and polarization into skin homing TEM and TCM. TEM then migrate through the bloodstream and are distributed to all parts of the skin, although the highest numbers of these cells will be found at the site of pathogen exposure (Gebhardt et al., 2009). These cells effect clearance of the pathogen and then remain resident locally within the skin (Gebhardt et al., 2009). In the early stages of a primary immune response, proliferating T cells are also released from the skin draining lymph nodes and distribute to antigen-free lymph nodes draining other tissues (Liu et al., 2006). T cells continue to proliferate within these new lymph nodes and give rise to new populations of effector T cells that migrate to and take up residence in the gut, lungs and other peripheral tissues. In this way, immunization through the skin actually generates widespread systemic immunity via the generation of disparate populations of tissue resident TEM cells.
In a memory immune response, T cell responses can be divided into three distinct stages (Figure 2). First, local re-exposure to a pathogen leads to antigen uptake and local antigen presentation by tissue resident DC. These DC stimulate the proliferation and effector functions of antigen specific skin resident T cells located in the skin, leading to rapid neutralization of the pathogen (Wakim et al., 2008b). Second, local inflammation leads to up regulation of vascular adhesion receptors on the skin endothelium, leading to nonspecific recruitment of T cells from the circulation. Only a minority of these T cells will be antigen specific, yet the small numbers of antigen specific cells that do enter the skin under these conditions have been shown to contribute to immune responses (Wakim et al., 2008a). Third, migration of antigen laden DC to the skin draining lymph nodes will lead to stimulation of TCM and the subsequent production of large numbers of skin homing TEM. These TEM will then migrate through the bloodstream, enter areas of inflamed skin, and affect clearance of the pathogen.
T cells have the flexibility to respond to any antigen, to migrate to any tissue, and to produce a plethora of cytokines and effector functions fine tuned to efficiently eliminate pathogens and tumors. Although tissue resident T cells have been described in gut, lung and skin, it is likely that all tissue types have to some degree their own populations of resident T cells. By populating both the peripheral tissues and the lymph nodes with distinct types of memory T cells, the adaptive immune system manages to provide both flexibility and the ability to neutralize pathogens rapidly.
Human skin is populated by 20 billion T cells, charged with the responsibility of locally defending the skin against tumors and pathogens while maintaining tolerance to self antigens exposed during injury. The evolution of this system of local T cell deposition, combined with amount of energy it takes to generate and maintain these cells, make it clear that defending the skin is a high priority for the immune system. However, skin resident T cells can dysfunction; these cells can fail to suppress pathogenic inflammation, shield tumors and infections with T cells designed to limit autoimmunity and generate autoreactive T cells that can initiate psoriasis and other inflammatory skin diseases. Studies of the antigen specificity of skin resident T cells and how they are activated in skin are likely to identify ways that T cell function can be enhanced in tumors and inhibited in inflammatory skin disorders. Because immune reactions in skin can be visually observed, sampled and manipulated with topical medications, the skin provides an accessible site in which to study human immune responses. Novel therapies arising from an understanding of T biology in skin should be broadly applicable to pathogenic immune states in other tissues.
Research by the author described in this review was carried out with the support and advice of Dr. Thomas S. Kupper, Chairman of Dermatology at Brigham and Women’s Hospital. The authors would like to thank Adam Calarese and Dr. Michael Lichtman for providing helpful suggestions and editorial comments. The author is supported by NIH grants 1K08AI060890-01A1, a Translational Research Award from the Leukemia and Lymphoma Society, a New Investigator Award from the Scleroderma Foundation and a Clinical Investigator Award from the Damon Runyon Cancer Research Foundation.