Self-reactive, innate-like T cells enhance cytotoxicity and immunosurveillance

Summary

Checkpoint blockade and adoptive T cell therapy are efficacious in some patients, but many do not respond to these immunotherapies. A publication in Nature by Chou et al. elucidates the characteristics of a new cell population with greater cytotoxicity, improved tumor homing, and reduced exhaustion when compared to conventional cytotoxic T cells.

CD8⁺ cytotoxic T cells play a key role in cancer immunosurveillance by binding neoplastic cells, secreting perforin and granzymes into their cytoplasm, and subsequently triggering their apoptosis. A delicate balance of this immune response is essential, as under-activation of CD8⁺ T cells results in immunosuppression, and hyperactivation can lead to inflammatory tissue damage and autoimmunity[1]. In order to avoid overactivation, activated T cells express programmed death-1 (PD-1), an inhibitory receptor. The binding of PD-1 to one of its ligands, PD-L1 or PD-L2, on the surface of antigen-presenting cells unleashes a downstream signaling pathway that culminates in reduced T cell activation and proliferation, accumulation of reactive oxygen species due to a glycolysis-to-β-oxidation metabolic shift, and eventual apoptosis[2]. Tumor cells take advantage of this mechanism through transcriptional upregulation of PD-L1/2, and the increased engagement of PD-1 with its ligands promotes a state of T cell exhaustion, in which T cell activity is dysregulated to promote immunosuppression and tumor growth[3].

Immunotherapies, such as immune checkpoint inhibitors (ICIs) and adoptive T cell transfers, harness the immunosurveillance characteristics of CD8⁺ T cells and improve their ability to target and eliminate neoplastic cells. ICIs target inhibitory receptors such as PD-1, aiming to reduce immunosuppression and T cell exhaustion. While monoclonal antibodies targeting PD-1/PD-L1 have experienced clinical success as checkpoint inhibitors, response rates remain low. Likewise, CAR T cell therapy and other adoptive cell transfers, in which CD8⁺ T cells with modified T cell receptors (TCRs) are expanded and reinfused into the patient, have proved ineffective due to downregulation or mutation of major histocompatibility complexes (MHC), T cell exhaustion, or the inability of T cells to infiltrate the tumor microenvironment[1].

In a recent study in Nature, Chou et al. [4] reveals a new population of cells, αβTCR⁺ innate-like T cells (αβILTCKs), which may have greater efficacy in cancer immunotherapies compared to conventional CD8⁺ T cells. The authors expand on previous findings by distinguishing the transcriptome of these cells from that of other CD8⁺ T cells, revealing their reactivity to self-antigens, and elucidating the relationship of IL-15 expression with αβILTCK activity[4, 5].

Chou et al. conducted single-cell RNA sequencing of TCRβ⁺CD8α⁺ cells from mouse breast tumor tissue and identified five clusters of distinct cell types characterized by diverse transcriptomes. Notably, the newly-defined αβILTCKs were marked by upregulated expression of Klrb1c, which encodes NK1.1, and were clustered separately from conventional, activated T cells, which are characterized by high expression of the Pdcd1 gene encoding PD-1. Given the role of PD-1 in T cell exhaustion, the limited expression of PD-1 among αβILTCKs signifies that these cells are less susceptible to exhaustion compared to conventional PD-1⁺ T cells. Likewise, their enhanced expression of NK1.1 grants these cells significant advantages. NK1.1 and CD161, its human homolog, are commonly expressed on NK and NKT cells but are also found among subpopulations of CD4⁺ and CD8⁺ T cells. CD8⁺ T cells expressing these receptors have been shown to upregulate the expression of cytotoxic granzymes. Adoptive transfer of CD8⁺NK1.1⁺ cells into murine lung and pancreatic ductal adenocarcinoma models has demonstrated reduced metastasis and increased survival compared to conventional NK1.1^- T cells[6]. Collectively, these findings suggest that αβILTCKs are less susceptible to exhaustion and more cytotoxic compared to conventional T cells (Figure 1).

Figure 1. Conventional PD-1⁺ cytotoxic T cells versus innate-like NK1.1⁺ cytotoxic T cells.

Over time, tumor cells upregulate expression of PD-L1 and PD-L2 and secrete greater levels of IL-15. In conventional PD-1⁺ cytotoxic T cells, continuous activation results in enhanced expression of PD-1, increased PD-1 interaction with its ligands, and, subsequently, reduced T cell activity and proliferation, greater T cell exhaustion, and lack of tumor infiltration. In contrast, innate-like NK1.1⁺ T cells exhibit enhanced activation and expansion and greater expression of cytotoxic molecules as a result of IL-15 release within the tumor microenvironment. Consequently, these cells display increased cytotoxic activity and tumor infiltration compared to their conventional counterparts. Figure created with BioRender.com. MHC: major histocompatibility complex, TCR: T cell receptor, PD-1: programmed death-1, PDL1/PD-L2: programmed death ligand-1 or 2, NK1.1: killer cell lectin-like receptor subfamily B, IL-15: interleukin-15.

During typical T cell development, CD4^-CD8^- T lymphocyte progenitors migrate from the bone marrow to the thymus, where they undergo TCR development and positive selection to give rise to mature CD8⁺ or CD4⁺ T cells that recognize MHC I or II proteins, respectively. This event is followed by negative selection, which induces apoptosis in T cells with high affinity for self-antigens to avoid autoimmunity[7]. αβILTCKs demonstrate self-reactivity and thus appear to have thwarted the conventional negative selection process. Typical CD8⁺ T cells that target cancer cells recognize specific neoantigens generated by mutations within the cancer cell, as these are recognized as “non-self” by the immune system[8]. Accordingly, Chou et al. utilized a TCR reporter assay and found that no TCRs derived from PD-1⁺ T cells reacted above the established baseline level to cancer cells from multiple mice, reflecting their TCR specificity. In contrast, 78.8% of TCRs derived from NK1.1⁺αβILTCKs demonstrated significant reactivity against these heterologous cells. These results indicate that αβILTCKs recognize a broad array of unmutated antigens rather than specific neoantigens.

The ability of αβILTCKs to target unmutated antigens signifies that their activity is not dependent on expression of a specific neoantigen, which may enable them to benefit patients refractory to traditional target-driven immunotherapies. However, it also raises the question of why these cells do not trigger autoimmunity. The authors reveal a potential mechanism through their investigation into the relationship between IL-15 and αβILTCK activity. A comparison of healthy and malignant mammary tissue in mice revealed significantly greater IL-15 expression within the malignant tissue. In accordance with previous findings, αβILTCK levels positively correlated with IL-15 levels in tissue from colon carcinoma patients, suggesting a dependence of αβILTCKs on IL-15. Chou et al. analyzed the tumor-infiltrating abilities of αβILTCKs in mice with deletion of Il15 in malignant tissues and found that, compared to controls, the mutated mice exhibited reduced tumor-infiltrating αβILTCKs, decreased expression of NK1.1 among αβILTCKs, and increased tumor growth. IL-15 alone was sufficient for NK1.1 upregulation and PD-1 downregulation in αβILTCK progenitors. Cumulatively, these results indicate that secretion of IL-15 from cancerous cells is necessary and sufficient for αβILTCK activity. These findings also present a potential mechanism for the avoidance of autoimmunity by αβILTCKs, as their dependence on IL-15 secreted from cancer cells implies that they would preferentially home to tumors rather than targeting healthy tissue.

The reduced exhaustion and improved cytotoxic activity of αβILTCKs, as well as their ability to home to tumor tissue without the need to target specific neoantigens, situates this new cell population as a breakthrough finding that can be harnessed as a future target for immunotherapies. Importantly, these cells could improve outcomes for patients who are unresponsive or refractory to checkpoint inhibitors or conventional neoantigen-based adoptive T cell therapies. While these preclinical studies suggest limited off-target reactivity, clinical trials will be necessary to ensure that the αβILTCKs have sufficient specificity for cancerous cells and do not initiate autoimmunity and damage to healthy tissue. This study also highlights the potential utility of the IL-15 signaling axis, which could be targeted by future cancer therapeutics to enhance αβILTCK anti-cancer activity. Yet the implications of this new cell population transcend cancer treatment, as αβILTCK adoptive transfer could be applied to other disease states with elevated IL-15 expression, as well. IL-15 levels are known to be elevated in Hepatitis B and C infection as well as in Celiac Disease and Inflammatory Bowel Disease[9, 10]. Further research into IL-15 expression in disease may provide insights into additional therapeutic avenues for αβILTCKs.