The Contribution of γδ T cells to the Pathogenesis of EAE and MS

Multiple Sclerosis and Experimental Autoimmune Encephalomyelitis (EAE)

Multiple Sclerosis (MS) is an immune-mediated disease of the central nervous system (CNS) characterized by perivascular CD4⁺ T cell and mononuclear cell inflammation and results in subsequent primary demyelination of axonal tracks leading to progressive paralysis [1]. It affects approximately one million people worldwide between the ages of 17 and 65 years old and is twice as common in women as men [2, 3]. In 1994, the annual cost in the USA for the direct care and lost productivity due to MS was estimated at $6.8 billion and for a lifetime, $2.2 million per patient; current costs are certainly much higher [4]. Current treatments aim at general immunosuppression with plieotropic anti-inflammatory mediators. IFN-β, glatiramer acetate (GA), and IVIg have beneficial effects on the relapse rate, however the impact on disease progressions is not as impressive—recent therapeutic trials has been nicely reviewed elsewhere [5]. Clearly, research to develop more specific treatments for this debilitating disease is necessary.

Multiple mouse models have been created and used to study the autoimmune inflammatory response as well as mechanisms involved in demyelination and repair, the most common is experimental autoimmune encephalitis (EAE). Disease induction can be accomplished by the immunization of susceptible strains of mice with myelin peptides in the context of adjuvant or disease can be transferred by either transgenic T cells specific for myelin peptides or a polyclonal population of T cells induced by immunization with myelin peptides. The resulting disease can be acute, chronic or relapsing in nature but all are mediated by CD4⁺ cells and closely resemble human MS [6, 7].

γδ T Cells

γδ T cells were discovered when in addition to the well known α and β-TCR genes, novel rearranged TCR genes were found [8, 9]. They constitute 1–5% of the total blood lymphocytes and are more commonly found in the skin and mucosal tissues where they can constitute up to 50% of the T cells in these areas [10]. Their development is primarily thymic dependent but some subsets can develop in the periphery in a thymic-independent fashion. Detailed features describing the development of γδ T cells have been carefully reviewed elsewhere [11]. Due to the limited repertoire of the γδ TCR as well as the multiple functions depending on tissue distribution, TCR subset or local environment it has been proposed that γδ T cells could be considered both innate and adaptive cells [12]. Based on the number of V, D and J combinations possible for a γδ TCR it is predicted these cells could have a greater potential for TCR diversity than αβ T cells, however limited VJ rearrangements suggest otherwise [13]. The CDR regions are determined primarily by somatic generation whish results in TCRs with a high frequency and a limited repertoire similar to an innate cell. Unlike αβ TCRs, ligand recognition of γδ T cells does not require antigen processing and presentation on MHC class I or class II molecules. In addition, the exact nature of the ligands for γδ T cells and how the γδ TCR sees ligands are not entirely clear. One category of γδ T cell ligands appear to be constitutively expressed in host cells or microbial pathogens. They tend to be non-protein pyrophosphates and alkylamines or heat shock proteins. γδ T cells which recognize these ligands are common in the blood and lymphoid tissues. The second category of γδ T cell ligands are inducibly expressed proteins which may be restricted to particular cell types. They include MHC I-like or MHC I polypeptide related sequences such as MIC A, MIC B and Rae1. Interestingly, the expression of these proteins can be modified by stress, inflammation and infection, perhaps serving as early warning signals. The γδ T cells which respond to these types of ligands are common to epithelium. These data have been more extensively reviewed elsewhere [14].

Protective or Pathogenic?

The role of γδ T cells in EAE is rather controversial. Despite the use of common depleting antibodies or genetic deletion approaches, differential results occur when using different strains. The use of UC7-13D5, a monoclonal antibody against the γδ TCR, in B10.PL mice 3 days before immunization with spinal cord homogenate (SCH) and CFA results in an acceleration of disease onset as well as relapse compared to control Ig treated mice implying a protective function of γδ T cells [25]. A correlative increase in splenocyte proliferation and IFN-γ production in vitro occurred at early time points, further suggesting γδ T cells have a regulatory role. Interestingly, the return of γδ T cells in the spleen correlates with the relapse, suggesting at this phase, γδ T cells are pathogenic and potential mediating disease. What is unclear is whether the antibody depletes or removes the γδ cell population or results in activation due to crosslinking of the TCR. This antibody has been used in other models of autoimmunity with similar effects in early and later phases of disease. Treatment before disease induction resulted in the delay of onset and severity of collagen induced arthritis but treatment 40 days after disease induction exacerbated disease, however the use of a F(ab′)₂ fragments of the antibody did not cause long standing increases [26]. Therefore, it is possible that stimulating γδ T cell populations at different time points during disease has different outcomes. Similarly, depending on the activation status of the γδ T cell, crosslinking the TCR could result in activation and cytokine production if naïve, or activation- induced death if previously stimulated.

Similarly, γδ T cell-deficient mice with a genetic disruption of the δ gene of the TCR and are on the same B10.PL background develop a chronic disease compared to the monophasic acute disease course seen in the control animals [27, 28]. One mechanism of suppression suggested is due to Fas-FasL mediated killing of CNS antigen specific T cells [28].

Conversely, the same adoptive transfer model into γδ deficient recipient mice on the C57BL/6 (B6) background, in one case showed no change in disease compared to the WT controls suggesting γδ T cells do not play a significant role in the mediation or regulation of effector mechanisms in EAE [29]. Mice which exclusively contain MBP specific T cells on a RAG−/− background spontaneously develop EAE where as the same MBP transgenic mouse on a wild type background does not, suggesting a rag-dependent regulatory cell [30]. When the resistant strain was crossed to a γδ-deficient mouse, they remained resistant. Only when this strain was crossed to TCR-α and –β knockout mice did spontaneous EAE occur. Collectively these data from B6 strains suggest the γδ T cells do not have a protective function in preventing onset of spontaneous EAE.

To further complicate the picture, there are many data suggesting γδ T cells play a pathogenic role in CNS inflammation and autoimmunity. The use of both depleting anti-γδ TCR antibodies and γδ T cell deficient mice as well as both adoptive transfer and peptide immunization mediated models within a variety of mouse strains provide strong evidence for the pathogenic role of γδ T cells [31–35]. The relapsing-remitting model of adoptively transferred induced EAE in SJL mice closely represents the human disease of MS. Although γδ T cells remained in low numbers in the spleen at all stages of disease, γδ T cells in the CNS reached over 10% at the peak of the acute phase, fell during remission and rose to above 10% again during relapse [31]. Depletion of γδ T cells during various stages of disease results in decreased disease, suggesting γδ T cells play a critical role in the pathogenesis of EAE during both acute and chronic phases. Similarly, in both actively induced and adoptively transferred EAE in B6 mice which lack γδ T cells, EAE disease was significantly reduced [33]. Using these types of models, γδ T cells are seen in the CNS and co-localize with heat shock protein expression in plaques. The TCR repertoire is limited during acute phases but most V and J regions are represented during chronic disease. This pattern suggests an antigen specific expansion during early phases of disease and possible non-specific recruitment there after [23].

One explanation for the difference in data could be explained by the gene array data obtained from expanded human Vδ1 and Vδ2 γδ T cells which indicated a distinct pattern of expression proposing opposing functions for the different subtypes [19]. It is possible different strains of mice have varying ratio of opposing subsets so that depletion has diametric affects. The local environment created by different strains may also be critical for how γδ T cells develop or function. Although controversial, taken together, these data suggest γδ T cells have various roles in autoimmune mediated CNS inflammation which can depend on the phase of disease, the activation status of the cells themselves and the strain of mouse which may dictate γδ T cell subtypes or the local environment.

γδ T Cell Subsets

Similarly to αβ T cells, there are multiple subtypes of γδ T cells. These subtypes can be distinguished by the variable region usage for both the γ and the δ genes. They differ from one another in the time in which they appear in ontogeny, their thymic dependence, anatomical location, and TCR repertoires. Additionally they express unique combinations of adhesion molecules which may be responsible for the differences in migration and location. Very comprehensive reviews of the γδ T cell subsets have been written, therefore we will focus on the subsets implicated in EAE and human MS [10, 11, 14]. Briefly, murine γδ T cells are designated by their Vγ usage. Vγ5 γδ T cells are thymus dependent, primarily found in the skin and have been referred to as dendritic epidermal T cells (DETC). They express NKG2D and recognize Rae1 and HSP60. Vγ6 T cells are mostly found at mucosal areas such as the reproductive tract and tongue and are also thymus dependent for their development. The two main peripheral γδ T cells in the mouse are the Vγ1 and Vγ4 (also referred to as Vγ2) subsets. Vγ1 γδ T cells are typically tissue derived and found the spleen, LN and intestine. They develop in the adult thymus and have been shown to have autoreactivity and specificity for heat shock proteins. Vγ4 T cells are the circulatory γδ T cell and found in the blood, LN and spleen. Vγ7 are the major lymphocyte in the intestine and can develop independently of the thymus. These cells express the CD8α homodimer which may act as a coreceptor for the TCR.

The first study to investigate the molecular diversity of γδ T cells during EAE pathogenesis resulted in very interesting findings in which the repertoire of γδ T cells during early phases of disease was very limited where as during later chronic phases of disease most Vδ and Vγ transcripts were observed [23]. Although the LN contained γδ T cells expressing most Vγ transcripts at all stages of disease, the γδ T cells in the brain had a limited TCR repertoire in early stages consisting primarily of peripheral and mucosal γδ T cells with Vγ6 transcripts. Interestingly, this study found γδ transcripts in the naïve, unimmunized mouse brain and these were specifically Vγ1-4 expressing cells commonly found in the circulation and could have been due to insufficient perfusion. However, consistent with these findings, Vδ6 expressing γδ T cells were also found in naïve mice by an independent group [24]. These data suggest a non random selection and migration of γδ T cells from the periphery into the CNS as the majority of circulating γδ T cells express the Vγ4/Vδ5 TCR [36]. The antigen specificity of the Vγ6 T cells which specifically accumulate during early stages of disease is unknown but these cells have been identified at mucosal surfaces as well and the significance of this similarity is unclear. One could speculate that at mucosal surfaces the γδ TCR is recognizing microbial antigens where as in the CNS it recognizes a self antigen which mimics the bacterial protein and stimulates the γδ only under certain conditions such as inflammation [37].

Human γδ T cell subsets have been studied more extensively and summarized nicely elsewhere [11]. The two major human γδ T cell subsets are Vδ1 and Vδ2. Vδ1 γδ T cells are considered the tissue γδ T cells and recognize CD1c and MICA and MICB through NKG2D. The major human γδ T cell subset in the circulation is Vδ2⁺ expressing γδ T cells which can respond to heat shock proteins and non-protein components of Mycobacteria. Culturing γδ T cells from the CSF of MS patients resulted in an expansion of both Vδ1 and Vδ2 γδ T cells [20]. Interestingly, these subsets could only be expanded from patients with recent-onset disease but not from subjects with chronic MS, supporting the limited repertoire data from the murine EAE model discussed above. It is also possible that γδ T cells from a chronic patient have become anergic suggesting their role may be more important in the initiation of autoimmunity. Junctional region sequencing indicated the expanded γδ T cells were oligoclonal in nature which suggests they may have been stimulated and expanded in vivo by antigen. It is unclear what that antigen may be or what role the γδ T cells may play in acute, but not chronic disease. When lesions from MS patients were compared to other non-neurological control brains, Vδ2 expressing circulatory γδ T cells again were of significance in diseased brains [18]. In contrast to the murine model, these data show multiple clones within a lesion implying more than one antigenic stimulus. In order to determine if the limited γδ TCR repertoire was specific to MS or to brain inflammation in general, γδ TCR transcripts from MS lesions and lesions from other neurological diseases were compared [22]. Interestingly, the CSF from MS patients showed increased frequency of γδ T cells after PHA stimulation compared to both other neurologic disorder patients such as encephalitis, cerebellitis or migraine control patients. The majority of these γδ T cells from MS patients responded to HSP70 where as only 2 out of 30 control patients responded. As seen before, the γδ T cells in the periphery of MS subjects were primarily Vδ2 and this was similar to the repertoire in the control patients. These data further suggest a role for γδ T cells in MS pathogenesis and confirm the presence of a limited repertoire in the CNS.

In addition to variable region gene segments, NK receptor (NKR) expression differentiates γδ T cell subsets. NKR have been suggested to be important regulators of T cell function, for fine tuning the response to antigen and influencing the immune response by rapid secretion of inflammatory cytokines. This regulation is conferred by the combination of both activating and inhibitory signals from the NKR family [38]. The majority of circulating γδ T cells in normal human samples expressed one or more NKR, however the most frequently expressed NKR was CD94 [39]. CD94 displays a broad specificity in recognizing MHC class I molecules including HLA-A, B, C or G alleles [40, 41]. The role of CD94 and its possible inhibitory function on γδ T cells remains unknown but suggests that like NK cells, it may allow the cell to respond quickly to pathogens or infected cells, but to remain tolerant to MHC class I- expressing self cells, providing further support for the innate nature of γδ T cells. Interestingly, the circulating γδ T cells which expressed NKR-P1A, a NKR whose function is unknown, were Vδ2 expressing γδ T cells. The significance of this correlation is unknown, it may shed light on the function of either this subset of γδ T cells or NKRs in the future. This study emphasizes a critical difference of γδ and αβ T cells in that NKR expression on αβ T cells is rare but most circulating γδ T cells express one or more of these types of receptors.

Potentially the most intriguing, but least understood factor for defining subsets of γδ T cells is antigen recognition. Although it is relatively unknown what and how γδ T cells recognize antigen, it appears they do so in a very different fashion than αβ T cells [14]. The majority of circulating γδ T cells in human, Vδ2⁺ TCRs, can be stimulated by non-peptide antigens derived from Mycobacterium, but the mechanism of presentation is unknown [42]. Some human Vδ2 γδ T cells have been shown to recognize the MHC class I-like molecule CD1 [43–45]. CD1 molecules are a family of MHC class I-like proteins which are non-polymporphic and important for the presentation of lipid antigens [46]. Recognition of this molecule by this subset of γδ T cells suggests self restriction. γδ T cells which bear the Vδ1-encoded TCRs account for the vast majority of γδ T cells in the spleen and intestine and have been shown to recognize the MHC-encoded proteins, MICA and MICB through the NKG2D protein [11, 47, 48]. Interestingly, MICA and MICB probably do not present peptides and more likely function in innate immunity as important targets of γδ T cells to kill stressed cells [47, 49]. Closely related MHC class Ib molecules T10 and T22 are ligands for a large population of γδ T cells in unimmunized mice [50]. Using T22- specific tetramers, γδ T cells which recognize the MHC class Ib molecules were identified and characterized. The results of this study suggest that the V region usage correlates mainly with tissue origin rather than antigen specificity, but the significance of the self-restriction is still unclear. Whether the Vδ2 expressing receptor is important in antigen recognition or simply identifies the subtype of cell is unknown. Presently, more is understood about the mechanisms of ligand recognition than the ligands themselves.

In addition to subsets defined by the ligands and/or antigens they recognize, one study has determined opposing functions for two subsets of γδ T cells [19]. Using microarray analysis of Vδ1 and Vδ2 cells expanded from the peripheral blood of normal patients, it was determined that 4500 genes changed in both subsets following PMA/ionomycin stimulation and that 50% of the genes were subset specific suggesting polarization.

Taken together, multiple types of γδ T cells with distinct possible ligands, functions and potential methods of regulation exist. In the future it will be important to determine which subtype is critical in the pathogenesis of MS and which subtype may have regulatory properties as a potential therapeutic option for the treatment of MS.

γδ activity in the CNS

Considering the plieotropic roles of γδ T cells in immunity, it is no surprise they appear to have roles in the development, maintenance and resolution of EAE. The development and presence of inflammation in the CNS correlates with an increase in inflammatory cytokines. However, in the absence of γδ T cells, production of IL-1, IL-6, TNF, lymphotoxin (LT) and IFN-γ were all diminished at the height of disease [32]. IL-2, IL-5 and IL-10 production have also been shown to be decreased in γδ deficient mice [33]. These studies suggest γδ T cells have the capacity to regulate inflammation in the CNS but did not investigate which cytokine producing cell types were affected. It is possible the γδ T cells themselves could be responsible for the cytokine production in the control animals or they could have secondary affects on other effector T cells or innate cells such as macrophages or dendritic cells. It is also unclear whether the reduced response in the absence of γδ T cells is due to the inability to generate encephalitogenic T cells or the inability to produce cytokines. The mechanistic contribution of γδ T cells to inflammatory cytokine production in the CNS has yet to be determined.

In addition to regulating inflammatory cytokine production, the absence of γδ T cells results in a decrease in the overall number of infiltrating cells to the CNS. This could be due to the lack of migrational cues required for adequate entry to the CNS. Chemokines are the molecular signals which direct the movement of lymphocytes through their normal circulation as well as to sites of inflammation. These signals are carefully regulated and dependent on the activation status of the cell. Indeed, chemokines and chemokine receptor expression in the CNS is altered in γδ-depleted mice. During normal disease progression, the levels of multiple chemokines in the CNS increases during onset, are highest at peak acute and fall during remission. These include RANTES, eotaxin, macrophage-inflammatory protein (MIP)-1α, MIP-1β, MIP-2, inducible protein-10 (IP-10) and monocyte chemoattractant protein-1 (MCP-1) [51]. Interestingly, in γδ T cell- depleted mice, the mRNA for all these chemokines was reduced during the progression of disease, but at the peak of disease the levels were similar to controls. Similarly, CCR1 and CCR5 were reduced at onset, but not at the height of disease in the absence of γδ T cells. These data indicate the importance of γδ T cells in the development and initial recruitment of inflammatory infiltrates into the CNS, and suggest that at some point, other cells may be sufficient to promote chemokine production and recruitment to the CNS. It is not clear whether γδ T cells themselves are a source of these chemokines in the lesions or whether their expression is downstream from γδ T cell activation or some other γδ-T cell mediated effect. Nonetheless, this data supports the ability of γδ T cells to regulate the initial stages of lymphocyte recruitment and migration to the CNS during autoimmune inflammation. Collectively, these data strongly suggest γδ T cells contribute to the pathogenesis of EAE with the recruitment of inflammatory cells to the CNS and by augmenting and/or polarizing the inflammatory cytokine milieu of the infiltrates.

In addition to facilitating recruitment of inflammatory cells, γδ T cells have other functional outcomes to regulate the overall inflammatory response in the CNS. Antigen presenting cells (APC) are the key mediator between the innate and adaptive response [52]. It is these cells which translate pathogenic signals into antigen specific responses. Although dendritic cells (DC) are known as professional APC and are the most well described cell of this type, other cells in the CNS have the capacity to uptake antigen, process and present peptides to T cells which result in antigen specific inflammatory responses. Macrophages, microglia and in some cases B-cells can also provide this essential role in the immune response. The relationship between γδ T cells and APCs is very interesting and not well understood. γδ T cells may directly influence DC function. The overall immune response in γδ-deficient mice is decreased when immunized with ovalbumin (OVA) peptide, OVA protein or in con-A stimulated splenocytes [33]. DC enriched from mice which lack γδ T cells, produce significantly less cytokine in response to LPS. The impaired response to bacterial pathogen stimulation by the DC in the γδ-deficient mice suggest γδ T cells are critical in regulating APC function. The mechanism for this regulation remains unknown, however it is likely to require cell-cell contact and may not involve the regulation of costimulatory molecules such as CD40, B7-1 or B7-2, but does help regulate APC- produced IL-12 [34]. A subset of γδ T cells has been shown to recognize the non-classical class I molecule CD1 [45]. This interaction is described in more detail as it relates to γδ TCR antigen recognition in another section of this review, however CD1- restricted γδ T cells have been shown to mediate DC maturation, including IL-12 production, in a CD1- and TCR-dependent manner thus amplifying and polarizing the overall response [53, 54]. Interestingly, the various human CD1 isoforms have different additional requirements such as CD40L or IFN-γ in order to produce IL-12 upon CD1 ligation by the γδ TCR. The induction of maturation and IL-12 production are key requirements for the initiation of an antigen-specific response by αβ T cells. Thus, γδ T cells influence the adaptive response indirectly by providing the necessary signals for antigen presentation and cytokine production by DCs. Interestingly, the ability of γδ T cells to “prime” DC for antigen presentation and the initiation of inflammation still requires a pathogenic signal. This is important considering the γδ TCR specific component could be self antigen in the context of CD1 molecule. The requirement for a danger signal therefore, allows for a speedy activation of the adaptive response, but only in the presence of pathogen and not under normal healthy circumstances, thus maintaining self tolerance.

In addition to “licensing” APCs, γδ T cells themselves have been shown to have antigen presenting functions. Human Vδ2⁺ γδ T cells specifically recognize small non-peptide antigens, derived primarily from microbes or necrotic host cells [55, 56]. When stimulated with the prototypic TCR ligand, isopentenyl pyrophosphate (IPP), Vδ2⁺ T cells upregulated co-stimulatory molecules such as CD80, CD86, CD40 and MHCII at levels that were indistinguishable from LPS matured DCs and furthermore, induced CD8⁺ αβ T cell proliferation and cytolytic activity [57]. The ability of γδ T cells to rapidly expand in response to microbial infections in addition to processing and presenting antigen provides a novel mechanism to initiate a strong CD4⁺ or CD8⁺-mediated immune response for rapid eradication of pathogens. If activated under inflammatory conditions in which much tissue destruction occurs, γδ T cells could also serve as APCs for self antigens and promote autoimmunity. In addition to their indirect activation of cytotoxic CD8⁺ T cells through the maturation of APCs, γδ T cells display cytolytic activity themselves and can directly lyse oligodendroctyes [58]. Oligodendrocytes are the primary producers of myelin, the protective sheath around neurons which allows efficient conduction of electrical signals [59]. The ability of γδ T cells to destroy oligodendrocytes provides additional evidence for the pathogenic nature of these T cells in EAE. Collectively, γδ T cells play major roles in regulating the recruitment, activation and polarization of the CNS inflammatory response in addition to the direct destruction of CNS tissue during EAE and could prove to be an important therapeutic target for MS therapy.

Migration of γδ T Cells

One important mechanism of regulation when considering CNS inflammation is migrational signaling. It is somewhat controversial as to whether lymphocytes circulate regularly throughout the CNS, but once inflammation has begun chemokines and adhesion molecules promote their infiltration and organization resulting in tissue destruction [19, 60, 61]. In addition to producing multiple types of chemokines themselves and recruiting other lymphocytes and amplifying inflammation, γδ T cells are also regulated by chemokines and adhesion molecules. In humans, two major subtypes of γδ T cells have been proposed to be involved in MS pathogenesis and can be distinguished by not only the TCR subunits expressed but also chemokine receptor and adhesion molecule surface expression which may facilitate differential abilities to enter and/or localize in the CNS. The migratory pathways of these two subsets have been reviewed previously; therefore we will only briefly discuss the importance of these mechanisms to EAE and MS [62]. It is the patterns of chemokine receptors and adhesion molecules which dictate the specific locations in which the subsets reside. The Vδ2 T cell subset, which represents the majority of the peripheral blood γδ lymphocytes in humans, expresses CD161 and is capable of transendothelial migration [63]. This subtype typically is found in the blood and lymph nodes and increases in MS patients [64]. Its particular use of CD161 for localization may be of particular interest as a target to prevent migration or entry into the CNS and/or traverse across the blood brain barrier. Vδ2 T cells express a wide range of chemokine receptors including CCR1 and CCR5, which respond to RANTES, MIP-1α and MIP-1β [65, 66].

The human Vδ1 population expresses high amounts of CXCR4 and PECAM1 which may facilitate entry into areas such as the gut, skin and lung [67]. Although these cells may not be the major subtype found in early phases of EAE, their role in the maintenance or resolution of CNS inflammation cannot ruled out. Clearly, chemokines and their receptors are major regulators of γδ T cells to position the particular subsets at sites of antigen entry or in the periphery to peruse the blood for pathogens. The chemokines and receptors important for murine γδ T cell migration to the CNS have yet to be determined. The differential function of these subsets may be best suited for the type of antigen or pathogen it may encounter based on its environment. It seems as though both circulating and tissue-resident γδ T cells express a distinct pattern of receptors which allow them to migrate to areas of inflammation during onset and progression of autoimmune disease and this process is likely coordinated by both chemokines and cytokines which can modulate receptor expression.

γδ T cells and Th17 inflammation

The recent discovery of a new lineage of CD4⁺ T cells which are defined by IL-17 production has changed the way the development of inflammatory responses is studied. More importantly, Th17 cells have been show to exacerbate autoimmune diseases. Th17 differentiation and the factors regulating development of these cells has been reviewed elsewhere [68]. CD4⁺ T cells, however are not the exclusive producers of this cytokine; multiple cell type such as NK cells and γδ T cells also produce large quantities of IL-17 [69, 70]. IL-17 has been implicated in multiple autoimmune diseases including rheumatoid arthritis and MS lesions [71, 72]. IL-17 can act directly on stromal cells to produce inflammation and MS, and has been suggested to potentiate the migration of lymphocytes across the blood brain barrier [73, 74]. IL-17 production from γδ T cells was shown to be critical in the development of disease in murine collagen-induced arthritis [75]. This study identified the Vγ4⁺ γδ T cells as required for disease, suggesting IL-17 from this subset of γδ T cells may be important for CNS inflammation as well. Interestingly, the increase in γδ T cell numbers was antigen-independent and likely due to the response to CFA adjuvant as the numbers of γδ T cells in the arthritic lesions were similar with or without antigen provided the CFA had been injected. Therefore, the role of γδ-produced IL-17 in the pathogenesis of collagen induced arthritis disease must be further studied. Although it is clear γδ T cells are potent producers of IL-17, it is unknown whether activation by natural ligands will yield similar results or whether this subset of γδ T cells are the prime producers of IL-17 in the CNS.

Summary

γδ T cells clearly play a role in the development and potentially the resolution of demylinating CNS inflammation. Particular subtypes may have opposing roles providing an explanation for the controversial data surrounding whether these cells are pathogenic or protective. As multi-faceted cells, γδ T cells have the potential to influence all levels of inflammation through rapid production of inflammatory mediators, recruitment of inflammatory cells via chemokines, influencing T cell differentiation by cytokine production, and/or direct killing via production of cytotoxic mediators. Considering their recognition of self molecules, some subsets of γδ cells have the capacity for promoting autoimmunity. With the most recent data suggesting γδ T cells as primary producers of the pro-inflammatory autoimmune-associated cytokine, Th17, the potential pathologic role of these cells in EAE and multiple sclerosis requires more extensive investigation.