One of the key characteristics of γδ T cells is their abundance in peripheral tissues. Thus, murine γδ T cells typically constitute less than 2% of the T-cell pool present in secondary lymphoid organs (lymph nodes, spleen), but account for approximately 50% (small intestine) up to 95% (skin) of the respective intra-epithelial lymphocyte (IEL) compartments []. The establishment of these γδ IEL populations occurs very early in ontogeny (late embryo or newborn) and depends on γδ T-cell migration from the thymus to the peripheral tissues. This process is orchestrated by particular chemokines and their receptors: namely, skin homing is controlled by CC chemokine ligand (CCL)27/ CC chemokine receptor (CCR)10, whereas, gut homing is directed by CCL25/ CCR9 [] (Fig. 1).
Chemokines, such as CCL27 and CCL25, that coordinate leukocyte trafficking during tissue development and maintenance (under steady-state) are termed “homeostatic.” These contrast with “inflammatory” chemokines, which are induced during immune responses and recruit leukocytes to the sites of inflammation (triggered by the host reaction to infectious microorganisms or allergens, for example). In the case of γδ T cells, Penido et al. had previously shown that CCL2 (the ligand for CCR2) played such a role by attracting substantial numbers of lymphoid γδ T cells to sites of infection [] and allergic inflammation []. In this issue of the European Journal of Immunology, a further paper from this group [] now shows that “homeostatic” CCL25 is also important in a model of allergic pleurisy through recruitment of a subset of γδ T cells specialized in the production of interleukin-17 (IL-17).
IL-17-producing γδ (γδ17) cells have recently emerged as key players in immune responses to infection and tumors, as well as in autoimmunity [[6-9]]. γδ17 cells are generated in the thymus starting already in the embryonic stages [[10, 11]] and are contained within the CD27− compartment of γδ T cells, both in the thymus and in the periphery []. Peripheral γδ T cells respond very rapidly to challenge, as tested in murine models of infection [] and autoimmunity []. In both cases, γδ17 cell expansion seems to be highly dependent on innate inflammatory stimuli, such as IL-1β and IL-23 [[8, 12]]. Interestingly, the chemokine receptor CCR6 has been shown to constitute a selective marker for murine γδ17 cells [], and more generally for IL-17+ T cells in humans [].
While considerable mechanistic insight into γδ17 cell development (in the thymus) and expansion (in peripheral lymphoid organs) has been gained, we still know very little about recruitment of these cells to inflammatory sites. In their new report, Costa et al. [] provide strong evidence for the migration of γδ17 cells (expressing CCR9) in response to the intra-pleural accumulation of CCL25 following OVA challenge. They observed γδ17 cell recruitment both in the allergic pleurisy model and upon intra-pleural injection of purified CCL25. Conversely, anti-CCL25 antibody treatment inhibited γδ17 cell migration into the allergic pleura. Importantly, this effect was specific to the γδ17 subset, since both αβ T cells and the remaining IL-17− γδ T cells were still recruited despite CCL25 neutralization. Consistent with this, the exogenous provision of CCL25 attracted γδ17 cells but not IFN-γ+ or IL-4+ γδ T cells, and this was associated with a selective increase in IL-17 levels in the mouse pleura [].
In this commentary, I discuss three topics that contextualize the new findings by Costa et al. []: first, the homeostatic versus inflammatory functions of CCL25; second, the migration of γδ T cells to sites of inflammation; and, third, the potential roles of IL-17 (and γδ17 cells) in allergic reactions.
CCL25-dependent T-cell migration in homeostatic versus inflammatory conditions
CCL25, also known as thymus-expressed chemokine (TECK), controls the migration of CCR9+ T-cell precursors to and within the thymus, and thus is important at early stages of T-cell development. The CCL25/CCR9 axis seems to work synergistically with CCL21/CCR7, and the genetic ablation of both chemokine receptors leads to impaired development of the first “γδ T-cell wave” in mouse ontogeny (at embryonic day 15–17 of gestation) [], a subset that populates the epidermis with invariant Vγ5Vδ1 T cells. These cells are also known as “dendritic epidermal T cells” and constitute an important first line of defense against injury or transformation [[1, 6]].
Interestingly, although the migration of Vγ5Vδ1 T cells from the thymus to the skin does not depend on CCR9, but rather CCR10 (Fig. 1), the main “tissue address” conferred by CCR9 expression is actually the small intestine, where the epithelium produces large amounts of CCL25. An important study by Svensson et al. showed that CCL25 neutralization prevented the recruitment of effector T lymphocytes to the small intestinal epithelium []. However, CCR9-deficient animals displayed a specific impairment in γδ, but not αβ, IEL numbers [[17, 18]], suggesting a lack of alternative (compensatory) mechanisms to CCL25/CCR9-dependent migration of γδ T cells during gut homing.
While the role of CCL25 as a homeostatic chemokine is undisputed, various reports have unraveled its involvement in gut inflammation. In inflammatory bowel diseases, such as Crohn's disease, CCL25 is expressed in aberrant patterns, which are associated with the altered distribution of CCR9+ lymphocytes with regard to the small intestine, the colon, and the blood [[19, 20]]. Furthermore, the neutralization of CCL25 has been shown to ameliorate inflammation in early stages of a murine model of ileitis [].
The report by Costa et al. [] discloses a new inflammatory function for CCL25 outside the gut. In the pleura, which is a nonmucosal tissue, CCL25 can also be produced by epithelial (mesothelial) cells upon OVA challenge in vivo (or IL-4 treatment in vitro) []. This then triggers the recruitment of CCR9+ γδ17 cells from secondary lymphoid organs into the inflamed pleura via the peripheral blood.
A unifying mechanism of CCL25-dependent T-cell migration is the dependence on the α4β7 integrin. Both in inflammatory pleurisy [] and in homeostatic gut homing [], α4β7 neutralization abolished the chemotactic effect of CCL25. Therefore, the coexpression of CCR9 and α4β7 enables (γδ) T cells to migrate to CCL25 gradients, and infiltrate tissues on adhesion and transendothelial crossing mediated by α4β7 interactions with MadCAM-1 and VCAM-1.
Chemokine-mediated γδ T-cell migration to sites of inflammation
The inflammatory roles of CCL25 are part of a more global program of chemokine-dependent migration by which γδ T cells respond to breaks in homeostasis, such as challenge with microorganisms or allergens (Fig. 1). Before their current study in this issue of the European Journal of Immunology on CCL25 [], Penido et al. had demonstrated a crucial role for CCL2, a ligand for CCR2, in attracting lymphoid γδ T cells to the inflamed pleura []. They obtained similar results with infectious stimuli, namely Mycobacterium bovis Bacillus Calmette-Guérin (BCG) or bacterial endotoxin (LPS) []. Interestingly, although the β-chemokines CCL3 and CCL5 were also induced in allergic pleurisy (and can mediate γδ T-cell chemotaxis in vitro), neutralization experiments showed they were not essential for γδ T-cell recruitment. Thus, only CCL2/CCR2 played a nonredundant role in γδ T-cell migration to the inflammatory site in vivo [].
An important difference between these previous studies on CCL2/CCR2 [[3, 4]] and the current one on CCL25/CCR9 [] is that, while the former showed an impact on total γδ T-cell numbers, the latter only affects specifically the γδ17 subset. Thus, whereas CCR2 seems to be a determinant for γδ T-cell migration as a whole, CCR9 may be selective for γδ17 cells. Future studies should explore this possibility in other models of inflammation, and also investigate chemokine-dependent mechanisms that may selectively control the recruitment of IFN-γ+ γδ T cells, which are particularly relevant in the context of antitumor immune responses [].
It is also essential to gain a deeper understanding of the roles played by inflammatory chemokines in physiological responses of human γδ T cells, which have been shown to migrate in vitro toward CCL2, CCL3, CCL4, CCL5, CXL10, and CXCL12 []. Intriguingly, a study by Brandes et al. has suggested that human γδ T cells switch, on activation, from an “inflammatory migration profile” (CCR2+, CCR5+, CXCR3+) to a “lymph node homing profile” (CCR7+), which facilitates their cross-talk with components of the adaptive immune system (B and αβ T cells) [].
With regard to the physiological role of human γδ T cells, these cells are mostly comprised of Vγ9Vδ2 cells, which are the dominating subset in the peripheral blood, and Vδ1 cells, which are more abundant in epithelial tissues. Cells from these two subsets isolated from HIV patients have been shown to migrate in vitro preferentially toward CXCL10 (for Vγ9Vδ2 cells) and CXCL12 (for Vδ1 cells) []; however, it should also be noted that γδ T cells are themselves important sources of chemokines, especially the β-chemokines CCL3, CCL4, and CCL5, and therefore can influence immune responses. In fact, both Vγ9Vδ2 and Vδ1 can produce high levels of these CCR5 ligands and thus suppress HIV-1 replication in vitro [[25, 26]].
What is the role of IL-17 and IL-17-producing (γδ) T cells in allergy?
Focusing on allergic reactions, extensive work by Born and colleagues on pulmonary γδ T cells revealed an interesting dichotomy between a Vγ1+ subset that promoted, and a Vγ4+ subset that suppressed, airway hyper reactivity (AHR) []. None of these effects have, however, been linked to IL-17, even though γδ17 cells have been independently found in close association with the inflamed airway epithelium []. The current paper by Costa et al. [] thus raises the question of the relevance of γδ17 cells, and more generally of IL-17, in allergic reactions.
Various studies have found elevated levels of IL-17, and increased numbers of Th17 cells, in the lung and in the peripheral blood of allergic asthmatics []. However, results in animal models have been complicated by differences in types of allergens, experimental protocols, and mouse strains []. IL-17 seems to be required for granulocytic infiltration in mouse models that employ airway sensitization to allergen (instead of allergen injection) []. In fact, allergic airway sensitization produced Th17-cell recruitment to the lung, together with neutrophilia and AHR. This contrasted with the strong Th2 response and eosinophilia, but no AHR, obtained with intra-peritoneal injection of OVA []. Furthermore, the adoptive transfer of antigen-specific Th17 cells resulted in neutrophilia and AHR upon subsequent antigen challenge []. Interestingly, this Th17-dependent AHR was not sensitive to steroids, contrary to the AHR induced by the adoptive transfer of antigen-specific Th2 cells []. Corticosteroids, which inhibit Th2 responses, are the gold standard treatment for asthma, but this treatment fails in some patients that maintain elevated Th2 and Th17 cells and neutrophils, leading to morbidity in severe allergic asthma [].
Although these studies [[29-31]] have suggested a pathogenic role for Th17 cells and IL-17 in allergic asthma, a previous report [] had paradoxically suggested a protective role for this cytokine. The administration of IL-17 in the chronic allergic phase inhibited ongoing Th2 responses, eosinophilia, and bronchial hyper reactivity []. This is also consistent with a more recent study by Murdoch and Lloyd [] showing that γδ17 cells promote the resolution of airway inflammation and remodeling. Most interestingly, γδ T cells, which were the main source of IL-17 in the allergic lung, inhibited Th2 responses and reduced AHR when adoptively transferred at the peak of acute inflammation []. Importantly, IL-17-deficient γδ T cells failed to resolve inflammation. These data strongly suggest a regulatory role for γδ17 cells in allergic reactions.
In the study by Costa et al. [], it is noteworthy that, even though CCL25 neutralization impaired γδ17 cell migration, this did not translate into a reduction of the number of total γδ T cells in the inflamed pleura []. This implies that the vast majority of infiltrated γδ T cells were actually not producing IL-17. On the other hand, the intra-pleural levels of IL-17 were approximately fivefold lower than those of IL-4, and over tenfold lower than those of IFN-γ. This highlights the importance of understanding the specific contributions of IL-17, γδ17, and Th17 cells in allergic reactions mostly dominated by Th2- and Th1-associated cytokines.
Finally, even if the role of γδ17 cells in animal models of allergy is becoming clearer, it remains critical to determine whether an equivalent population participates in human allergic reactions. IL-17-producing γδ T cells are still poorly characterized in human pathology, possibly because they only accumulate—likely in an age-dependent fashion—in extremely inflammatory sites []. In allergy, as in infection or tumor immunology, integrating the rapidly evolving knowledge on murine γδ17 cells with new and definitive findings in humans will be a major challenge for γδ T-cell researchers.