Too Much of a Good Thing? Tim-3 and TCR Signaling in Exhaustion Robert L. Ferris, Binfeng Lu and Lawrence P. Kane This information is current as J Immunol 2014; 193:1525-1530; ; of September 26, 2021. doi: 10.4049/jimmunol.1400557 http://www.jimmunol.org/content/193/4/1525 Downloaded from References This article cites 61 articles, 23 of which you can access for free at: http://www.jimmunol.org/content/193/4/1525.full#ref-list-1

Why The JI? Submit online. http://www.jimmunol.org/ • Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on September 26, 2021 Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2014 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Th eJournal of Brief Reviews Immunology

Too Much of a Good Thing? Tim-3 and TCR Signaling in T Cell Exhaustion Robert L. Ferris,*,†,‡ Binfeng Lu,*,‡ and Lawrence P. Kane*,‡ T cell exhaustion is thought to be a natural mechanism and effector molecules, with IL-2, cytotoxicity and proliferation for limiting immune pathology, although it may be de- among the earliest affected functions and IFN-g among the sirable to circumvent this mechanism to help eliminate latest (1, 2). Exhausted T cells also may become “addicted” to viral reservoirs or tumors. Although there are no defin- AgR signals and lose responsiveness to the homeostatic cyto- itive markers, a fingerprint for exhausted T cells has kine IL-7, with the latter due in part to loss of CD127 (IL-7r been described that includes the transmembrane pro- a-chain) expression (2). Importantly, for possible therapeutic teins PD-1, LAG3, and Tim-3. However, apart from reversal, exhausted T cells also gain high-level and persistent, as the recruitment of tyrosine phosphatases to PD-1, little opposed to transient, expression of several , including Downloaded from is known about the biochemical mechanisms by which the transcription factor BLIMP-1 and the transmembrane these proteins contribute to the development or main- proteins PD-1, Tim-3, and LAG3 (1, 2). The latter proteins, tenance of exhaustion. Tim-3 contains no known motifs so-called “checkpoint” receptors, have attracted attention as possible dominant mediators of T cell exhaustion because Abs for the recruitment of inhibitory phosphatases, but it to these proteins or their ligands can sometimes rescue the may actually increase signaling downstream of TCR/ function of exhausted T cells (2–4). Because this topic has been http://www.jimmunol.org/ CD3, at least under acute conditions. Other studies covered extensively in other relatively recent reviews (1, 2), we showed that T cell exhaustion results from chronic stim- focus mainly on recent studies of Tim-3, which has attracted ulation that extends the effector phase of T cell activa- substantial preclinical attention of late as a novel therapeutic tion, at the expense of T cell memory. We suggest that target for reversal of T cell exhaustion. We also review what is Tim-3 may contribute to T cell exhaustion by enhancing known regarding signal transduction pathways implicated in TCR-signaling pathways. The Journal of Immunology, Tim-3 function. Finally, we discuss the role of TCR signaling 2014, 193: 1525–1530. in driving the development of exhaustion and how this might

be influenced by Tim-3. by guest on September 26, 2021 cell activation, including development of a robust Lessons from tumors memory response, is critical for the development of an The tumor microenvironment is known to be immunosup- efficient immune response to viral infection and can also T pressive as a result of inhibitory signals from cell surface and be instrumental in mounting an immune response to solid soluble mediators (5), although the precise strategies used by tumors. However, overly vigorous or sustained immune responses different tumors can vary by tissue and even from patient to can cause immune-mediated pathology that is detrimental to the patient. Although T cells specific to tumor Ags can be readily host. Such a problem is particularlyevidentwithvirusesthatcause isolated from solid tumors of patients and in mouse models, chronic infections (1). In these cases, the sustained presence of these cells often respond poorly to ex vivo stimulation. This T viral Ags appears to drive the formation of a state of Ag-specific cell dysfunction is thought to result, at least in part, from T cell exhaustion. Although this has the beneficial effect of exhaustion (or over-stimulation) of effector tumor-infiltrating limiting immune pathology, it can result in the establishment lymphocytes. T cell exhaustion is caused by chronic antigenic of a viral reservoir that may become reactivated under conditions stimulation and expression of inhibitory coreceptors and of physiological stress. T cell exhaustion can also be detrimental cytokines, among other factors (6). Based on the recent suc- when it impairs the ability of an adaptive immune response to cess of CTLA-4 Ab therapy (7), as well as accumulating data eliminate a tumor. from preclinical models, there is now considerable excitement Functionally, the development of T cell exhaustion is char- surrounding molecules whose targeting may allow for broad acterized by the gradual loss of expression of various cytokines enhancement of T cell responses against tumors. Solid tu-

*Department of Immunology, University of Pittsburgh School of Medicine, Pittsburgh, Address correspondence and reprint requests to Dr. Lawrence P. Kane, Department of PA 15261; †Department of Otolaryngology, University of Pittsburgh School of Medi- Immunology, University of Pittsburgh, 200 Lothrop Street, Pittsburgh, PA 15261. cine, Pittsburgh, PA 15261; and ‡Cancer Immunology Program, University of Pitts- E-mail address: [email protected] burgh Cancer Institute, University of Pittsburgh School of Medicine, Pittsburgh, PA Abbreviations used in this article: gal9, galectin-9; HCV, hepatitis C virus; mTORC1, 15261 mTOR complex 1; SH2, Src homology 2; TB, tuberculosis; Treg, regulatory T cell; WT, ORCID: 0000-0001-5198-516X (L.P.K.). wild-type. Received for publication February 28, 2014. Accepted for publication May 13, 2014. Copyright Ó 2014 by The American Association of Immunologists, Inc. 0022-1767/14/$16.00 This work was supported by National Institutes of Health Grants DE019727 and CA097190 (to R.L.F.), CA167229 and CA097190 (to B.L.), and AI109605 and AI073748 (to L.P.K.).

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1400557 1526 BRIEF REVIEWS: SIGNALING AND T CELL EXHAUSTION mor–infiltrating T cells often express high levels of one or ligands—galectin-9 (gal9), expressed on macrophages— more inhibitory or exhaustion-associated receptors, including stimulated the production of the cytokine IL-1b, enhancing PD-1, LAG3, and/or Tim-3. Indeed, consistent with Ag bacterial killing. This effect was mimicked by administration acting as a driver of exhaustion, a recent study on melanoma of Tim-3–Ig fusion , through its interaction with gal9. patients demonstrated that PD-1 can be used to prospectively A similar phenomenon was observed in human macrophages distinguish tumor-specific T cells at the tumor site (8). Tim-3 (32). At this point, the nature of the signal transmitted by expression on T cells is also seen in the context of nonsolid gal9 after Tim-3 binding is unknown. Still, this finding is of tumors. For example, upregulation of Tim-3 (possibly driven by possible relevance for the interpretation of multiple other IL-12) on effector T cells of patients with follicular B cell non- studies, because a soluble Tim-3–Ig fusion protein is often Hodgkin lymphoma was associated with poor outcomes (9). used as a blocking reagent in other settings. Returning to PD-1 has been extensively studied as a potential therapeutic T cells and human patients, active TB, as opposed to latent target, and recent clinical trial data suggest that mAbs to PD-1 infection, is associated with the upregulation of Tim-3 on or one of its ligands, PD-L1, are clinically effective against both CD4+ and CD8+ T cells (33). Surprisingly, however, certain solid tumors, including melanoma, as well as non-small these Tim-3+ T cells display more potent anti-TB responses, cell lung cancer, generally regarded as a nonimmunogenic contrary to what has been observed in chronic viral infection. tumor (10–12). mAbs specific for Tim-3 also were shown to Although the reasons for these differences are not clear, the promote rejection of solid tumors in murine models (13, 14), investigators speculated that they may be due to the somewhat + and mAbs to human Tim-3 can rescue the ex vivo function divergent phenotypes of Tim-3 TB-specific T cells compared Downloaded from of apparently exhausted T cells from tumor-bearing patients with Tim-3+ T cells found in HIV and HCV patients. For (15). In the former case, the efficacy of Tim-3 mAb therapy example, Tim-3+ T cells in TB patients were found to express appeared to result, at least in part, from effects on regulatory CD127 (IL-7r a-chain) (33), whereas this marker is well T cells (Tregs), which also can express Tim-3 (16). Strikingly, known to be lost in exhausted T cells (including those that are these Tim-3+ Tregs appear to be among the most potent at Tim-3+) during chronic viral infections of humans and mice inhibiting effector T cell function and express greater levels of (1, 2). http://www.jimmunol.org/ IL-10. In this study, monotherapy with Tim-3 Ab was not A recently published study using a newly generated Tim-3– sufficient to augment antitumor immunity in vivo, but rather knockout mouse model provides further support for a con- it cooperated with primary anti–PD-L1 treatment (13). High text-dependent positive role for Tim-3 in vivo, in this case in levels of Tim-3 also have been observed on tumor-infiltrating acute bacterial infection. Colgan and colleagues (34) first Tregs from human patients with non-small cell lung cancer, infected wild-type (WT) mice with Listeria monocytogenes a finding that correlated with poor clinical outcomes (17). and followed the expression of Tim-3. Tim-3 was robustly Thus, in the case of tumors, it clearly will be important to parse expressed on CD8+ T cells during this infection, with par-

out the effects of Tim-3 manipulation on effector/exhausted ticularly high expression on effector T cells. In mice with by guest on September 26, 2021 T cells from the effects on Tregs. In addition, the precise role a germline deletion of Tim-3, both primary and secondary played by Tim-3 may vary, depending on tumor type. T cell responses to L. monocytogenes were significantly im- paired. A similar, although less severe, defect was also seen Tim-3 in infectious disease: recent studies when Tim-3–deficient T cells were transferred to WT hosts. T cell exhaustion has been observed in mouse models (dis- Finally, evidence was provided that the decreased T cell cussed above), and in human disease, mainly in the context of responses in the absence of Tim-3 were due in part to the various chronic viral infections (18–25). For instance, a pop- compromised survival of the knockout T cells. It should be ulation of Tim-3+CD8+ T cells is readily identified in HIV noted that the Tim-3–deficient mice used in this study were patients (26), and the expression of Tim-3 correlates posi- derived from 129 strain embryonic stem cells and then tively with progression to AIDS and inversely with viral backcrossed to C57BL/6 mice for 10 generations. Because the control (27). Ex vivo stimulation of these HIV-associated, Tim is polymorphic between 129 and C57BL/6, the exhausted, CD8+ T cells in the presence of Tim-3 mAb can carryover of polymorphisms in Havcr1 (encoding Tim-1) and at least partially rescue their function (26), whereas anti- in other neighboring should be taken into account as retroviral therapy may reduce Tim-3 expression (28). Simi- potential confounding variables in this study. Nonetheless, larly, study of hepatitis C virus (HCV)-infected patients this report provides additional compelling evidence that the revealed the presence of dysfunctional CD4+ and CD8+ effects of Tim-3 expression during infection are complex T cells with high-level Tim-3 expression (29, 30). In both and may lead to enhanced, rather than diminished, T cell HIV and HCV, upregulation of Tim-3 was associated with responses. 2 the accumulation of central memory (CD45RA CCR7+) T cells (26, 29). Thus, a large body of evidence now supports Tim-3 signaling and T cell exhaustion a net negative impact of Tim-3 expression on T cell–depen- Surprisingly, at this point there is relatively sparse evidence to dent antiviral immune responses. support the model that Tim-3 directly mediates suppression The role of Tim-3 in tuberculosis (TB) has proven to be of T cell activation or cytokine secretion in a manner similar quite complex and points out that effects of Tim-3 expression to PD-1. Rather, despite evidence that Tim-3 manipulation can (or its therapeutic targeting) on infectious disease outcomes ameliorate T cell exhaustion, there are a number of observations may be disease and/or context specific. An elegant study by that do not fit with a simple narrative of direct inhibition Behar and colleagues (31) revealed a novel and unexpected of T cell activation by Tim-3. First is the fact that Tim-3+ function for Tim-3 in a murine model of TB. Thus, the T cells from HIV-infected patients display defective stimulation- investigators found that interaction of Tim-3 with one of its induced phosphorylation of Stat5, ERK1/2, and p38, but The Journal of Immunology 1527 actually possess higher basal phosphorylation of all of the same (Fig. 1A). However, the sequences around these tyrosines pathways (26). This is consistent with our own data obtained do not conform to any known inhibitory signaling motifs with ectopic expression of Tim-3 in T cell lines and primary (e.g., ITIM, ITSM), such as those found in PD-1. Analysis of T cells. Thus, ectopic expression of Tim-3 actually enhanced the cytoplasmic tail of murine Tim-3 with the Scansite T cell activation, leading to increased activation of NFAT/AP-1 algorithm (http://scansite3.mit.edu) revealed that several of and NF-kB transcriptional reporters, and it even enhanced these tyrosines conformed well to putative sites of phos- cytokine production from both T cell lines and primary T cells phorylation by multiple tyrosine kinases, particularly those of (35). the Src family (Fig. 1B). Importantly, the analogous tyrosines When considering how Tim-3 might signal to regulate the in human Tim-3 also scored as possible sites of phosphory- development or activation of exhausted T cells, it may be in- lation by the same classes of kinases (data not shown). We structive to consider the reported functions of Tim-3 in myeloid (35) and other investigators (43, 44) demonstrated that the lineage cells. Ligation of Tim-3 with specific Abs or one of the cytoplasmic tail of Tim-3 can be phosphorylated on multiple reported Tim-3 ligands can (under some circumstances) en- tyrosine residues. Although the identity of the kinases that hance the activation and function of various myeloid lineage cell phosphorylate the Tim-3 cytoplasmic tail have not been de- types (36). This is consistent with induction of the transcrip- finitively shown in vivo, our data are consistent with such tion factor NF-kB after Tim-3 mAb treatment of dendritic phosphorylation being carried out by the Src family kinases cells (37). In addition, Ab ligation of Tim-3 on murine bone Fyn and/or Lck, at least in T cells (35). However, it should be marrow–derived mast cells was shown to augment IgE/Ag- pointed out that the Tec family kinase Itk also has been Downloaded from induced cytokine production (38), a finding that we repro- implicated in the phosphorylation of Tim-3 (43). Tyrosine duced with multiple Abs to Tim-3 (B. Phong and L.P. Kane, phosphorylation often functions to recruit downstream sig- unpublished observations). Another recently described function naling proteins, particularly those containing Src homology 2 for Tim-3 in dendritic cells is to inhibit an innate response to (SH2) domains. Thus, we found that the SH2 domain of Fyn nucleic acids, an effect that involves interaction of Tim-3 with and one of the SH2 domains of the PI3K adaptor protein p85 the endogenous danger signal HMGB1 (39). This unusual could bind to a phosphorylated peptide corresponding to the http://www.jimmunol.org/ pathway seems to prevent the trafficking of exogenously ac- region around Y256 and Y263 of murine Tim-3 (35). Con- quired DNA into endosomes. The precise mechanism under- sistent with these findings, we also found that ectopic ex- lying this effect is unclear (e.g., whether it is mediated by active pression of Tim-3 enhanced the phosphorylation of both Tim-3 signaling or by passive blockade of HMGB1 function), PLC-g1 (which is dependent on Src and Syk family kinases) but it may have relevance for the effects of Tim-3 in T cell and ribosomal protein S6, which lies downstream of the responses against viral and/or tumor Ags. It should be noted PI3K/Akt/mTOR pathway (35). Some of these experiments that this is one of the few published studies to directly dem- were performed with the workhorse Jurkat T cell model (45),

onstrate that a particular Tim-3 Ab could block interaction which does have some limitations, including lack of PTEN by guest on September 26, 2021 with the ligand under study (in this case HMGB1). Several expression. Importantly, however, key findings of these studies studies also suggested a negative regulatory role for Tim-3 in were reproduced in the murine T cell clone D10, as well as in myeloid cells (40–42). Thus, as in the case of T cells, it appears primary murine T cells (35). Thus, at the levels of both signal that the precise effects of Tim-3 on myeloid cells depend upon transduction and function (cytokine production), acute up- the context in which Tim-3 ligation occurs. regulation of Tim-3 expression can enhance T cell activation. A Administration of certain Tim-3 Abs can rescue the function current challenge is deciphering how these observations fit into of exhausted T cells in both in vivo mouse models (13, 23) and in the current paradigm of Tim-3 function in vivo. vitro experiments with cells from patients with chronic viral A recent report suggested that interaction of Tim-3 with infections or tumors (15, 26). However, particularly in the case a chaperone protein known variously as Bat3, Bag6, or Scythe of the aforementioned in vivo experiments, it is not clear regulates suppression of T cell responses by Tim-3 (46). One whether the effects of the Tim-3 Abs are actually the direct of the more intriguing findings reported in this article was result of blocking the interaction of Tim-3 with one or more of that knockdown of Bat3 led to a dramatic upregulation of its ligands (e.g., gal9 and/or HMGB1 or other yet-to-be dis- Tim-3 and other (although not all) phenotypic and functional covered ligands). The presence of Tim-3 on other cells, in- markers of T cell exhaustion. The investigators also identified cluding APCs and Tregs, might also explain the function of at an interaction between Bat3 and a pool of Lck tyrosine kinase least some Tim-3 Abs in certain disease settings. In such cases, that was phosphorylated on its activation loop tyrosine one might imagine that either agonistic or antagonistic activities (Y394), but not the inhibitory tyrosine near the C terminus of such Abs could eventually lead to the observed downstream (Y505), suggesting a preferential interaction of Bat3 with functional effects. However, specific agonistic or antagonistic active Lck. Furthermore, this interaction appeared to be activity has not been directly ascribed yet to the most commonly inhibited by Ab ligation of Tim-3, which itself was also seen used Tim-3 Abs. One approach to this problem would be to to interact with Lck (46). The latter finding is consistent with compare the activities of Tim-3 mAbs in their native forms our own observation that Tim-3 could interact with Lck, as versus the effects of F(ab)9 fragments of the same Abs. This will well as with the related Src family tyrosine kinase Fyn (35). be an important point to clarify in future studies. Importantly, we confirmed that the T cell lines used in our signaling experiments express Bat3 (L.P. Kane, unpublished Tim-3 proximal signaling: tyrosine kinases and beyond observations). Whereas these results suggest a possible mecha- When we began studying signaling pathways linked to Tim-3, nism for inhibition of proximal TCR signaling by Tim-3 we were intrigued by the presence of multiple tyrosine residues and Bat3, this study did not address the effects of TCR signal- in the cytoplasmic tails of both murine and human Tim-3 ing on the Tim-3/Bat3 interaction. Thus, further investigation 1528 BRIEF REVIEWS: SIGNALING AND T CELL EXHAUSTION

FIGURE 1. Amino acid sequence of the Tim-3 cytoplasmic tail and possible sites of phosphorylation. Alignment of the murine and human Tim-3 cytoplasmic tail Downloaded from sequences (upper panel). Predicted sites of tyrosine phosphorylation are numbered, and other tyrosines are in bold type. Conserved charged residues (suggesting favorable sites for phosphorylation) upstream of Y256 and Y263 are underlined. Results of a Scansite search using the murine Tim-3 cytoplasmic domain and restricting results to cytoplasmictyrosinekinases(lower panel). Analogous tyrosines in human Tim-3 also were predicted to be phosphorylated by the same kinases (data not shown). will be necessary to determine the precise relationship between sistent with the above data, several groups demonstrated that these different pools of Tim-3 and Src family kinases, as well as limiting the activation of mTOR enhances the development of http://www.jimmunol.org/ the consequences on downstream-signaling pathways, in exhaus- T cell memory. This was shown through the use of the relatively ted T cells. specific mTORC1 inhibitor rapamycin (54, 55) or by reducing Another recent study examined the interaction of Tim-3 expression of the Raptor , which encodes a critical subunit of with transmembrane proteins expressed on T cells and re- the mTORC1 complex (54). ported that Tim-3 can associate with the transmembrane These previous studies on Akt and mTOR in T cell memory phosphatases CD45 and CD148, which the investigators are also intriguing in light of a recent report describing a human proposed might promote dephosphorylation of downstream immunodeficiency associated with activating mutations of the mediators of T cell activation (44). Furthermore, Tim-3, PI3K catalytic protein p110d, with consequent hyperactivation by guest on September 26, 2021 CD45, and CD148 were found to be recruited into the im- of downstream Akt and mTOR signaling (56). Of relevance for munological synapse. Such a mechanism could help to explain the discussion of T cell exhaustion, patients with these muta- the observed negative effects of Tim-3 on T cell activation tions present with an accumulation of terminally differentiated during exhaustion. It remains to be seen to what extent these T cells. These T cells are refractory to stimulation with mitogen various interactions affect Tim-3 function, including whether and display a deficient recall response to tetanus toxoid, con- the interaction of these phosphatases with Tim-3 modulates sistent with a defect in generation of stable memory T cell its interaction with Bat3 or Lck, as discussed above. responses. It should be noted that patients with the activating p110d mutations have a combined immunodeficiency because TCR signaling and the development of T cell exhaustion B cells are also affected. Strikingly, treatment of one of the T cell exhaustion appears to be maintained by transcriptional affected patients with the mTORC1 inhibitor rapamycin par- reprogramming (e.g., through BATF) and/or active negative tially restored the T cell compartment and improved clinical signaling through receptors likePD-1(2).Indeed,tyrosine outcomes (56). phosphatases like SHP1/2 were shown to mediate dominant The studies discussed above provide compelling evidence that suppression of TCR signaling after ligation of PD-1 (47). excessive and/or sustained activation of one or more signaling However, it is quite clear that development of T cell exhaustion pathways is critical for driving the development of T cell ex- is driven, at least in part, by high levels—and/or the sustained haustion. However, it is still not known whether such signals are presence—of cognate Ag (48, 49). Evidence (much of it still initiated solely from the TCR itself or whether other molecules indirect) has begun to emerge regarding how this process may be might also contribute. Given what is known about the multi- controlled by TCR-signaling pathways. Several studies have factorial regulation of T cell activation by costimulatory re- implicated Akt- and/or mTOR-dependent signaling in this ceptors, the latter seems more likely. Based on our previous checkpoint. Thus, a study from Kaech and colleagues (50) findings demonstrating the ability of Tim-3 to enhance TCR demonstrated that ectopic expression of Akt led to reduced signaling under acute conditions (35), we propose that Tim-3 memory T cell formation. In addition, Cantrell and colleagues might function, at least in part, to help drive T cell exhaustion (51) showed that although inhibition of Akt impairs the func- by enhancing TCR/CD28-dependent signaling (Fig. 2). In tion of CTL effector cells, it actually promotes the development such a scenario, upregulation of Tim-3 during an extended of memory cells. Similar conclusions were reached in a study by effector phase of T cell activation would act in a feed-forward Suresh and colleagues (52). Among the many downstream effects loop to enhance T cell activation signals and drive T cells even of Akt signaling is the activation of an mTOR-containing more toward an exhausted phenotype, at the expense of T cell complex known as mTOR complex 1 (mTORC1) (53). Con- memory (Fig. 2A). An additional nonexclusive possibility is The Journal of Immunology 1529

FIGURE 2. Speculative model for the development of T cell exhaustion. (A) Acute infection results in elimination of a pathogen (and its Ags) and allows for the Downloaded from formation of a pool of memory T cells (upper portion). This response may be accompanied by transient and/or low-level expression of Tim-3. Establishment of a chronic infection (or significant tumor burden) promotes the development of T cell exhaustion, marked in part by upregulation of Tim-3 (lower portion). This process may be accelerated by Tim-3–derived signals that enhance TCR/CD3 signaling in the short-term. In addition, these positive signals might contribute to the reversal of exhaustion by some Tim-3 mAbs. (B) Tim-3–derived positive signals that may contribute to the induction of T cell exhaustion or activation. These positive signals could contribute to the reversal of exhaustion by some Tim-3 Abs, although the latter could be the result of qualitatively different signals or blocking of an undefined inhibitory coreceptor. http://www.jimmunol.org/ that positive signals from Tim-3 augment the (suppressive) Disclosures function of Tregs, contributing to the overall immune-activating R.L.F. has received research funding from Bristol-Myers Squibb (related effects of Tim-3 “blocking” Abs. With respect to the specific to Tim-3), and consulted for Bristol-Myers Squibb and Astra-Zeneca/ signaling pathways downstream of Tim-3, the work described MedImmune. L.P.K. has consulted for Janssen Pharmaceuticals and Jounce above suggests that at least one such pathway is the PI3K/Akt/ Therapeutics on Tim-3. mTOR pathway, with upstream involvement of Src or Tec

kinases and downstream involvement of transcription factors, like References by guest on September 26, 2021 NFAT or NF-kB (Fig. 2B). Coming back to the role of signaling 1. Virgin, H. W., E. J. Wherry, and R. Ahmed. 2009. Redefining chronic viral in- fection. Cell 138: 30–50. in driving T cell exhaustion (discussed above), we propose that 2. Wherry, E. J. 2011. T cell exhaustion. Nat. Immunol. 12: 492–499. an accessory receptor like Tim-3 could help to drive this process, 3. Odorizzi, P. M., and E. J. Wherry. 2012. Inhibitory receptors on lymphocytes: insights from infections. J. Immunol. 188: 2957–2965. in part through enhanced PI3K/Akt/mTOR signaling. 4. Sakuishi, K., P. Jayaraman, S. M. Behar, A. C. Anderson, and V. K. Kuchroo. 2011. As discussed above, it will be critical to more thoroughly Emerging Tim-3 functions in antimicrobial and tumor immunity. Trends Immunol. define the biophysical and biochemical properties of the 32: 345–349. 5. Drake, C. G., E. Jaffee, and D. M. Pardoll. 2006. Mechanisms of immune evasion various Tim-3 Abs being used in both mouse and human by tumors. Adv. Immunol. 90: 51–81. studies. Thus, it is still a distinct possibility that some Abs act to 6. Schietinger, A., and P. D. Greenberg. 2014. Tolerance and exhaustion: defining mechanisms of T cell dysfunction. Trends Immunol. 35: 51–60. enhance signaling in T cells (or other cell types). In this regard, 7. Lipson, E. J., and C. G. Drake. 2011. Ipilimumab: an anti-CTLA-4 antibody for it is interesting to note that the efficacy of mAbs targeting metastatic melanoma. Clin. Cancer Res. 17: 6958–6962. 8. Gros, A., P. F. Robbins, X. Yao, Y. F. Li, S. Turcotte, E. Tran, J. R. Wunderlich, CTLA-4 for immunotherapy of melanoma, at least in mouse A. Mixon, S. Farid, M. E. Dudley, et al. 2014. PD-1 identifies the patient-specific CD8+ models, was just recently attributed to FcR-dependent de- tumor-reactive repertoire infiltrating human tumors. J. Clin. Invest. 124: 2246–2259. pletion of Treg (57, 58), despite two decades of study of 9. Yang, Z. Z., D. M. Grote, S. C. Ziesmer, T. Niki, M. Hirashima, A. J. Novak, T. E. Witzig, and S. M. Ansell. 2012. IL-12 upregulates TIM-3 expression and CTLA-4 biology. Finally, the unintended consequences of Ab induces T cell exhaustion in patients with follicular B cell non-Hodgkin lymphoma. manipulation of CD28 offer a cautionary tale with regard to J. Clin. Invest. 122: 1271–1282. 10. Topalian, S. L., F. S. Hodi, J. R. Brahmer, S. N. Gettinger, D. C. Smith, the clinical translation of a target that regulates T cell acti- D. F. McDermott, J. D. Powderly, R. D. Carvajal, J. A. Sosman, M. B. Atkins, et al. vation (59–62). 2012. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N. Engl. J. Med. 366: 2443–2454. 11.Brahmer,J.R.,S.S.Tykodi,L.Q.Chow,W.J.Hwu,S.L.Topalian,P.Hwu, Conclusions C. G. Drake, L. H. Camacho, J. Kauh, K. Odunsi, et al. 2012. Safety and activity of anti- Recent investigation into Tim-3 function has clearly elucidated PD-L1 antibody in patients with advanced cancer. N. Engl. J. Med. 366: 2455–2465. 12. Hamid, O., C. Robert, A. Daud, F. S. Hodi, W. J. Hwu, R. Kefford, J. D. Wolchok, its importance, in part due to its frequent upregulation during P. Hersey, R. W. Joseph, J. S. Weber, et al. 2013. Safety and tumor responses with antiviral T cell responses or in the tumor microenvironment. lambrolizumab (anti-PD-1) in melanoma. N. Engl. J. Med. 369: 134–144. Preclinical studies in mouse models are also encouraging. 13. Sakuishi, K., L. Apetoh, J. M. Sullivan, B. R. Blazar, V. K. Kuchroo, and A. C. Anderson. 2010. Targeting Tim-3 and PD-1 pathways to reverse T cell ex- However, these studies are generally restricted by a lack of haustion and restore anti-tumor immunity. J. Exp. Med. 207: 2187–2194. suitable reagents to clarify the significance and function of 14. Ngiow, S. F., B. von Scheidt, H. Akiba, H. Yagita, M. W. Teng, and M. J. Smyth. 2011. Anti-TIM3 antibody promotes T cell IFN-g-mediated antitumor immunity a putatively stimulatory signal on a population of exhausted and suppresses established tumors. Cancer Res. 71: 3540–3551. T cells. These gaps in the field are slowly being rectified, but 15. Fourcade, J., Z. Sun, M. Benallaoua, P. Guillaume, I. F. Luescher, C. Sander, J. M. Kirkwood, V. Kuchroo, and H. M. Zarour. 2010. Upregulation of Tim-3 and this must be accelerated because therapeutic manipulation in PD-1 expression is associated with tumor antigen-specific CD8+ T cell dysfunction patients is being seriously contemplated. in melanoma patients. J. Exp. Med. 207: 2175–2186. 1530 BRIEF REVIEWS: SIGNALING AND T CELL EXHAUSTION

16. Sakuishi, K., S. F. Ngiow, J. M. Sullivan, M. W. Teng, V. K. Kuchroo, 39. Chiba, S., M. Baghdadi, H. Akiba, H. Yoshiyama, I. Kinoshita, H. Dosaka-Akita, M. J. Smyth, and A. C. Anderson. 2013. TIM3(+)FOXP3(+) regulatory T cells are Y. Fujioka, Y. Ohba, J. V. Gorman, J. D. Colgan, et al. 2012. Tumor-infiltrating DCs tissue-specific promoters of T-cell dysfunction in cancer. OncoImmunology 2: suppress nucleic acid-mediated innate immune responses through interactions between e23849. the receptor TIM-3 and the alarmin HMGB1. Nat. Immunol. 13: 832–842. 17. Gao, X., Y. Zhu, G. Li, H. Huang, G. Zhang, F. Wang, J. Sun, Q. Yang, X. Zhang, 40. Zhang, Y., C. J. Ma, J. M. Wang, X. J. Ji, X. Y. Wu, Z. S. Jia, J. P. Moorman, and and B. Lu. 2012. TIM-3 expression characterizes regulatory T cells in tumor tissues Z. Q. Yao. 2011. Tim-3 negatively regulates IL-12 expression by monocytes in and is associated with lung cancer progression. PLoS ONE 7: e30676. HCV infection. PLoS ONE 6: e19664. 18. Allen, S. J., P. Hamrah, D. Gate, K. R. Mott, D. Mantopoulos, L. Zheng, T. Town, 41. Zhang, Y., C. J. Ma, J. M. Wang, X. J. Ji, X. Y. Wu, J. P. Moorman, and Z. Q. Yao. C. Jones, U. H. von Andrian, G. J. Freeman, et al. 2011. The role of LAT in in- 2012. Tim-3 regulates pro- and anti-inflammatory cytokine expression in human creased CD8+ T cell exhaustion in trigeminal ganglia of mice latently infected with CD14+ monocytes. J. Leukoc. Biol. 91: 189–196. herpes simplex virus 1. J. Virol. 85: 4184–4197. 42. Yang, X., X. Jiang, G. Chen, Y. Xiao, S. Geng, C. Kang, T. Zhou, Y. Li, X. Guo, 19. Angelosanto, J. M., and E. J. Wherry. 2010. Transcription factor regulation of CD8 H. Xiao, et al. 2013. T cell Ig -3 promotes homeostasis of sepsis by negatively + T-cell memory and exhaustion. Immunol. Rev. 236: 167–175. regulating the TLR response. J. Immunol. 190: 2068–2079. 20. Barber, D. L., E. J. Wherry, D. Masopust, B. Zhu, J. P. Allison, A. H. Sharpe, 43. van de Weyer, P. S., M. Muehlfeit, C. Klose, J. V. Bonventre, G. Walz, and G. J. Freeman, and R. Ahmed. 2006. Restoring function in exhausted CD8 T cells E. W. Kuehn. 2006. A highly conserved tyrosine of Tim-3 is phosphorylated upon during chronic viral infection. Nature 439: 682–687. stimulation by its ligand galectin-9. Biochem. Biophys. Res. Commun. 351: 571–576. 21. Blackburn, S. D., H. Shin, W. N. Haining, T. Zou, C. J. Workman, A. Polley, 44. Clayton, K. L., M. S. Haaland, M. B. Douglas-Vail, S. Mujib, G. M. Chew, M. R. Betts, G. J. Freeman, D. A. Vignali, and E. J. Wherry. 2009. Coregulation of L. C. Ndhlovu, and M. A. Ostrowski. 2014. T cell Ig and mucin domain-containing CD8+ T cell exhaustion by multiple inhibitory receptors during chronic viral in- protein 3 is recruited to the immune synapse, disrupts stable synapse formation, and fection. Nat. Immunol. 10: 29–37. associates with receptor phosphatases. J. Immunol. 192: 782–791. 22. Day, C. L., D. E. Kaufmann, P. Kiepiela, J. A. Brown, E. S. Moodley, S. Reddy, 45. Abraham, R. T., and A. Weiss. 2004. Jurkat T cells and development of the T-cell E. W. Mackey, J. D. Miller, A. J. Leslie, C. DePierres, et al. 2006. PD-1 expression receptor signalling paradigm. Nat. Rev. Immunol. 4: 301–308. on HIV-specific T cells is associated with T-cell exhaustion and disease progression. 46. Rangachari, M., C. Zhu, K. Sakuishi, S. Xiao, J. Karman, A. Chen, M. Angin, Nature 443: 350–354. A. Wakeham, E. A. Greenfield, R. A. Sobel, et al. 2012. Bat3 promotes T cell 23. Jin, H. T., A. C. Anderson, W. G. Tan, E. E. West, S. J. Ha, K. Araki, responses and autoimmunity by repressing Tim-3–mediated cell death and ex- G. J. Freeman, V. K. Kuchroo, and R. Ahmed. 2010. Cooperation of Tim-3 and haustion. Nat. Med. 18: 1394–1400. Downloaded from PD-1 in CD8 T-cell exhaustion during chronic viral infection. Proc. Natl. Acad. Sci. 47. Chemnitz, J. M., R. V. Parry, K. E. Nichols, C. H. June, and J. L. Riley. 2004. USA 107: 14733–14738. SHP-1 and SHP-2 associate with immunoreceptor tyrosine-based switch motif of 24. Nakamoto, N., H. Cho, A. Shaked, K. Olthoff, M. E. Valiga, M. Kaminski, programmed death 1 upon primary human T cell stimulation, but only receptor E. Gostick, D. A. Price, G. J. Freeman, E. J. Wherry, and K. M. Chang. 2009. ligation prevents T cell activation. J. Immunol. 173: 945–954. Synergistic reversal of intrahepatic HCV-specific CD8 T cell exhaustion by com- 48. Richter, K., T. Brocker, and A. Oxenius. 2012. Antigen amount dictates CD8+ T- bined PD-1/CTLA-4 blockade. PLoS Pathog. 5: e1000313. cell exhaustion during chronic viral infection irrespective of the type of antigen 25. Shin, H., and E. J. Wherry. 2007. CD8 T cell dysfunction during chronic viral presenting cell. Eur. J. Immunol. 42: 2290–2304.

infection. Curr. Opin. Immunol. 19: 408–415. 49. Mueller, S. N., and R. Ahmed. 2009. High antigen levels are the cause of T cell http://www.jimmunol.org/ 26. Jones, R. B., L. C. Ndhlovu, J. D. Barbour, P. M. Sheth, A. R. Jha, B. R. Long, exhaustion during chronic viral infection. Proc. Natl. Acad. Sci. USA 106: 8623– J. C. Wong, M. Satkunarajah, M. Schweneker, J. M. Chapman, et al. 2008. Tim-3 8628. expression defines a novel population of dysfunctional T cells with highly elevated 50. Hand, T. W., W. Cui, Y. W. Jung, E. Sefik, N. S. Joshi, A. Chandele, Y. Liu, and frequencies in progressive HIV-1 infection. J. Exp. Med. 205: 2763–2779. S. M. Kaech. 2010. Differential effects of STAT5 and PI3K/AKT signaling on 27. Kassu, A., R. A. Marcus, M. B. D’Souza, E. A. Kelly-McKnight, L. Golden-Mason, effector and memory CD8 T-cell survival. Proc. Natl. Acad. Sci. USA 107: 16601– R. Akkina, A. P. Fontenot, C. C. Wilson, and B. E. Palmer. 2010. Regulation of 16606. virus-specific CD4+ T cell function by multiple costimulatory receptors during 51. Macintyre, A. N., D. Finlay, G. Preston, L. V. Sinclair, C. M. Waugh, P. Tamas, chronic HIV infection. J. Immunol. 185: 3007–3018. C. Feijoo, K. Okkenhaug, and D. A. Cantrell. 2011. Protein kinase B controls 28. Kassu, A., R. A. Marcus, M. B. D’Souza, E. A. Kelly-McKnight, and B. E. Palmer. transcriptional programs that direct cytotoxic T cell fate but is dispensable for T cell 2011. Suppression of HIV replication by antiretroviral therapy reduces TIM-3 metabolism. Immunity 34: 224–236. expression on HIV-specific CD8(+) T cells. AIDS Res. Hum. Retroviruses 27: 1–3. 52. Kim, E. H., J. A. Sullivan, E. H. Plisch, M. M. Tejera, A. Jatzek, K. Y. Choi, and

29. Golden-Mason, L., B. E. Palmer, N. Kassam, L. Townshend-Bulson, S. Livingston, M. Suresh. 2012. Signal integration by Akt regulates CD8 T cell effector and by guest on September 26, 2021 B. J. McMahon, N. Castelblanco, V. Kuchroo, D. R. Gretch, and H. R. Rosen. memory differentiation. J. Immunol. 188: 4305–4314. 2009. Negative immune regulator Tim-3 is overexpressed on T cells in hepatitis C 53. Laplante, M., and D. M. Sabatini. 2012. mTOR signaling in growth control and virus infection and its blockade rescues dysfunctional CD4+ and CD8+ T cells. J. disease. Cell 149: 274–293. Virol. 83: 9122–9130. 54. Araki, K., A. P. Turner, V. O. Shaffer, S. Gangappa, S. A. Keller, M. F. Bachmann, 30. McMahan, R. H., L. Golden-Mason, M. I. Nishimura, B. J. McMahon, C. P. Larsen, and R. Ahmed. 2009. mTOR regulates memory CD8 T-cell differ- M. Kemper, T. M. Allen, D. R. Gretch, and H. R. Rosen. 2010. Tim-3 expression entiation. Nature 460: 108–112. on PD-1+ HCV-specific human CTLs is associated with viral persistence, and its 55. Rao, R. R., Q. Li, K. Odunsi, and P. A. Shrikant. 2010. The mTOR kinase blockade restores hepatocyte-directed in vitro cytotoxicity. J. Clin. Invest. 120: determines effector versus memory CD8+ T cell fate by regulating the expression of 4546–4557. transcription factors T-bet and Eomesodermin. Immunity 32: 67–78. 31. Jayaraman, P., I. Sada-Ovalle, S. Beladi, A. C. Anderson, V. Dardalhon, C. Hotta, 56. Lucas, C. L., H. S. Kuehn, F. Zhao, J. E. Niemela, E. K. Deenick, U. Palendira, V. K. Kuchroo, and S. M. Behar. 2010. Tim3 binding to galectin-9 stimulates D. T. Avery, L. Moens, J. L. Cannons, M. Biancalana, et al. 2014. Dominant- antimicrobial immunity. J. Exp. Med. 207: 2343–2354. activating germline mutations in the gene encoding the PI(3)K catalytic subunit 32. Sada-Ovalle, I., L. Cha´vez-Gala´n, L. Torre-Bouscoulet, L. Nava-Gamin˜o, p110d result in T cell senescence and human immunodeficiency. Nat. Immunol. 15: L. Barrera, P. Jayaraman, M. Torres-Rojas, M. A. Salazar-Lezama, and S. M. Behar. 88–97. 2012. The Tim3-galectin 9 pathway induces antibacterial activity in human mac- 57. Simpson, T. R., F. Li, W. Montalvo-Ortiz, M. A. Sepulveda, K. Bergerhoff, F. Arce, rophages infected with Mycobacterium tuberculosis. J. Immunol. 189: 5896–5902. C. Roddie, J. Y. Henry, H. Yagita, J. D. Wolchok, et al. 2013. Fc-dependent de- 33. Qiu, Y., J. Chen, H. Liao, Y. Zhang, H. Wang, S. Li, Y. Luo, D. Fang, G. Li, pletion of tumor-infiltrating regulatory T cells co-defines the efficacy of anti-CTLA- B. Zhou, et al. 2012. Tim-3-expressing CD4+ and CD8+ T cells in human tu- 4 therapy against melanoma. J. Exp. Med. 210: 1695–1710. berculosis (TB) exhibit polarized effector memory phenotypes and stronger anti-TB 58. Bulliard, Y., R. Jolicoeur, M. Windman, S. M. Rue, S. Ettenberg, D. A. Knee, effector functions. PLoS Pathog. 8: e1002984. N. S. Wilson, G. Dranoff, and J. L. Brogdon. 2013. Activating Fc g receptors 34. Gorman, J. V., G. Starbeck-Miller, N. L. Pham, G. L. Traver, P. B. Rothman, contribute to the antitumor activities of immunoregulatory receptor-targeting J. T. Harty, and J. D. Colgan. 2014. Tim-3 directly enhances CD8 T cell responses antibodies. J. Exp. Med. 210: 1685–1693. to acute Listeria monocytogenes infection. J. Immunol. 192: 3133–3142. 59. Eastwood, D., L. Findlay, S. Poole, C. Bird, M. Wadhwa, M. Moore, C. Burns, 35. Lee, J., E. W. Su, C. Zhu, S. Hainline, J. Phuah, J. A. Moroco, T. E. Smithgall, R. Thorpe, and R. Stebbings. 2010. Monoclonal antibody TGN1412 trial failure V. K. Kuchroo, and L. P. Kane. 2011. Phosphotyrosine-dependent coupling of explained by species differences in CD28 expression on CD4+ effector memory T- Tim-3 to T-cell receptor signaling pathways. Mol. Cell. Biol. 31: 3963–3974. cells. Br. J. Pharmacol. 161: 512–526. 36. Han, G., G. Chen, B. Shen, and Y. Li. 2013. Tim-3: An Activation Marker and 60. Horvath, C., L. Andrews, A. Baumann, L. Black, D. Blanset, J. Cavagnaro, Activation Limiter of Innate Immune Cells. Front. Immunol. 4: 449. K. L. Hastings, D. L. Hutto, T. K. MacLachlan, M. Milton, et al. 2012. Storm 37. Anderson, A. C., D. E. Anderson, L. Bregoli, W. D. Hastings, N. Kassam, C. Lei, forecasting: additional lessons from the CD28 superagonist TGN1412 trial. Nat. R. Chandwaskar, J. Karman, E. W. Su, M. Hirashima, et al. 2007. Promotion of Rev. Immunol. 12: 740–, author reply 740. tissue inflammation by the immune receptor Tim-3 expressed on innate immune 61. Hunig,€ T. 2012. The storm has cleared: lessons from the CD28 superagonist cells. Science 318: 1141–1143. TGN1412 trial. Nat. Rev. Immunol. 12: 317–318. 38. Nakae, S., M. Iikura, H. Suto, H. Akiba, D. T. Umetsu, R. H. Dekruyff, H. Saito, 62. Suntharalingam, G., M. R. Perry, S. Ward, S. J. Brett, A. Castello-Cortes, and S. J. Galli. 2007. TIM-1 and TIM-3 enhancement of Th2 cytokine production M. D. Brunner, and N. Panoskaltsis. 2006. Cytokine storm in a phase 1 trial of the by mast cells. Blood 110: 2565–2568. anti-CD28 monoclonal antibody TGN1412. N. Engl. J. Med. 355: 1018–1028.