BAtCHing stem-like T cells during exhaustion

Abstract

Long-term pathogen and tumor control as well as checkpoint immunotherapies rely on ‘stem-like’ CD8⁺ T cells. New results uncover BACH2 as a key regulator of this subpopulation and solve an important piece of the puzzle.

Exhausted CD8⁺ T (T_ex) cells represent a distinct hypofunctional cellular lineage that arises in the context of high and sustained antigen loads and prevents the eradication of persistent pathogens and tumors in mice and humans. T_ex cells are characterized by a loss of effector functions, high expression of inhibitory receptors such as programmed cell death 1 (PD-1) and a unique transcriptional and epigenetic configuration¹. Among the cellular subsets that constitute the heterogeneous T_ex cell population, the TCF-1-expressing ‘stem-like’ population (T_ex^steam) contains self-renewing progenitors that give rise to more terminally differentiated T_ex cell subsets (T_ex^steam), enabling long-term maintenance of the T_ex cell pool and protective responses following PD-1 or PD-1 ligand (PD-L1) checkpoint immunotherapies¹. In this issue of Nature Immunology, Yao and colleagues² reveal that the transcriptional repressor BACH2 establishes the epigenetic and transcriptional landscape of T_ex^steam cells after chronic viral infection.

Although T_ex^steam cells are known to use transcriptional regulators that overlap with those expressed in memory CD8⁺ T cells after acute infection or immunization, they also exhibit transcriptional and epigenetic features unique to T_ex cells¹. Importantly, recent studies have indicated that T_ex^steam cells arise within a few days after infection with a rapidly spreading persistent pathogen and that TCF-1-mediated suppression of the competing terminal effector-like fate (T_eff-like) is essential for the early establishment of T_ex^stem cells^3,4. Given that transcription factors (TFs) typically act as part of complex molecular circuits, an important question still remains: which are the key nodes that promote the T_ex^stem cell fate and its unique epigenetic and transcriptional network early after infection?

In their study, Yao et al. performeda meta-analysis of epigenetic and transcriptional data from T_ex^steam cells and more terminally differentiated CD8⁺ T cells from mice that had been infected for one week with a persistent variant of lymphocytic choriomeningitis virus (LCMV). This approach identified several TFs whose DNA binding motifs were highly enriched in genomic regions with differential H3K27ac content or differential chromatin accessibility and could thus represent potential regulators of T_ex^steam cell differentiation. Among them, the genes encoding the transcriptional repressor BACH2 as well as the TFs TCF-1 and Jun also exhibited higher expression in T_ex^steam cells. Analogous to what has been described during acute viral infections, wherein the repression of memory genes allows the development of effector T cell responses^5,6, the aforementioned analysis suggests that BACH2 fosters T_ex^steam cells by suppressing T_ex^steam genes.

To test this hypothesis, Yao et al. performed a series of elegant gain- and loss-of-function experiments. Through overexpression of BACH2 in LCMV-specific CD8⁺ T cells, the authors observed a net increase in T_ex^steam cell frequency and numbers at both early and late time points after chronic infection. Two different gene-targeting approaches specific for Bach2 corroborated these observations, showing that BACH2 deficiency reduced the number of T_ex^steam cells in infected mice. Thus, together witha recent study⁴, the results by Yao et al. revealed BACH2 as a key regulator in the establishment of the T_ex^steam cell lineage early after infection.

The authors also observed that BACH2 overexpression in T_ex^steam cells was associated with reduced expression of an apoptotis gene signature and upregulation of the prosurvival factor Bcl-2 (Fig. 1). This phenotype was accompanied by BACH2-mediated induction of the TF c-MYB, which is known to drive Bcl-2 expression⁷. Importantly, the effect of BACH2 on T_ex^stem cells was not limited to prosurvival signals. Instead, additional studies revealed BACH2 had profound effects on T_ex^stem cell lineage commitment, functionality and tissue location.

Fig. 1 |. BACH2 promotes CD8⁺ T cell differentiation toward T_ex^stem cells while antagonizing T_ex^term cell fates via epigenetic and transcriptional mechanisms in the early stage of chronic viral infection.

BACH2 promotes the upregulation of the prosurvival factor Bcl-2 in T_ex^stem cells and modulates the expression of tissue-homing factors (promoting the upregulation of CD62L and CC R7 and the downregulation of CC R2 and CXCR6), which favor the localization of T_ex^stem cells to the splenic white pulp (grey background cells, left) and away from the red pulp (red background cells, right) and non-lymphoid organs, where T_ex^term cells are enriched. Furthermore, BACH2 suppresses the expression of effector molecules, such as granzyme B (GrzB), and inhibitory receptors, such as PD-1, and attenuates signaling downstream of the TCR (left, blunt arrows). Mechanistically, BACH2 mediates the repression of T_ex^term cell differentiation by restricting chromatin accessibility to the regulatory regions of the TFs RUNX3 and BATF (middle, above) and suppressing the expression of genes encoding BLIMP-1 (Prdm1) and BATF (middle, below).

Yao and collaborators discovered that BACH2 promotes the expression of transcriptional signatures characteristic of T_ex^stem cells while suppressing T_ex^stem-associated genes within the T_ex^stem cell subset. This included BACH2-mediated upregulation of TFs such as Lef1, encoding LEF1, which often acts together with TCF-1 (ref.⁸), raising the possibility that BACH2 and TCF-1 may be part of a feed-forward transcriptional loop that polarizes recently activated CD8⁺ T cells toward the T_ex^stem phenotype and restrains their differentiation into T_ex^stem cells. Intriguingly, T cell antigen receptor (TCR) signaling has been shown to repress BACH2 activity via post-translational modification⁹, which seems at odds with increased T_ex^stem cell differentiation in the presence of high amounts of antigen early after infection⁴. A possible explanation is that PD-1 and PD-L1 expression, which are potently induced in the presence of high antigen loads, in turn counteract TCR signaling¹ and the resulting downstream suppression of BACH-2 activity, thereby favoring T_ex^stem cell differentiation.

Interestingly, the authors also demonstrated that BACH2 suppresses the expression of several coinhibitory receptors, including PD-1, as well as effector and cytolytic genes, such as granzymes, within T_ex^stem cells (Fig. 1). This observation is consistent with BACH2-mediated downregulation of a T_ex^stem transcriptional signature within T_ex^stem cells. The authors further propose that reductions in coinhibitory receptor expression can be explained by BACH2-dependent attenuation of the signaling cascade downstream of the TCR. Similarly, BACH2 suppresses TCR-driven transcriptional changes during memory T cell differentiation⁹, suggesting that BACH2 could promote T_ex^stem cell differentiation by attenuating TCR signaling and, perhaps, by restricting the activity of TCR-induced TFs, such as IRF4 or NFAT, that promote T_ex^term cell differentiation¹⁰.

T_ex^stem cells are known to preferentially localize to the white pulp of lymphoid tissues¹. Consistent with BACH2 enforcement of the stem-like program in T_ex^stem cells, the authors found that BACH2 enhances the expression of lymphoid-tissue-homing molecules such as CD62L and CCR7 (Fig. 1). This finding, and the higher proportion of T_ex^stem cells among virus-specific-CD8⁺ T cells that overexpress BACH2, explains their almost complete restriction to the spleen white pulp as well as their significant reductionin non-lymphoid tissues. It would be interesting to characterize the cellular niche and the environmental cues (or lack thereof) that may nurture T_ex^stem cells within the white pulp of secondary lymphoid tissues and how these may influence BACH2 expression and/or activity. Of particular interest may be cytokines, such as interleukin-27, and type I interferons that are known to have opposing effects in modulating T_ex^stem cells^11,12

Overall, the study by Yao et al. demonstrates a crucial role for BACH2 in T_ex^stem lineage commitment, but how does it do this? BACH2 acts as an epigenetic modulator; it increases chromatin accessibility at loci associated with T_ex^stem cell fate while inducing a closed chromatin state at T_ex^stem-associated regions (Fig. 1). For example, BACH2 made chromatin inaccessible at regions containing binding motifs for the TFs RUNX3 and BATF, and the latter has been shown to oppose BACH2 activity and the generation of T_ex^stem cells early after infection⁴. Importantly, the authors used double-knockout approaches to demonstrate that, in the absence of RUNX3, the T_ex^stem cell fate was not affected by BACH2 deficiency, suggesting that RUNX3 acts downstream of BACH2. Yaoet al. also identified BACH2-binding motifs in open chromatin regions of the genes encoding the TFs BLIMP-1 and, again, BATF. As described above for RUNX3, the authors showed that, in the absence of BLIMP-1, the proportions of T_ex^stem cells were unchanged by BACH2 deficiency. Thus, the study by Yao et al. proposes a model whereby BACH2 establishes the optimal epigenetic and transcriptional landscape for T_ex^stem cell commitment by antagonizing TFs that promote T_ex^stem cells via two different mechanisms: (1) directly suppressing their expression, as in the case of BLIMP-1 and BATF, and (2) restricting access to their regulatory regions, as in the case of RUNX3 as well as BATF (Fig. 1). Given that TCR stimulation promotes RUNX3 activity and RUNX3 drives BLIMP-1 expression after acute infection¹³, this model is also consistent with the aforementioned pivotal role for BACH2 in promoting T_ex^stem cell fate by taming TCR signaling and downstream gene expression. Notably, RUNX3 has been shown to suppress TCF-1 and BACH2 expression¹³, suggesting the possibility that RUNX3 may favor commitment to T_ex^stem cell differentiation by silencing BACH2 or TCF-1, which would otherwise promote stemness.

The findings presented by Yao and collaborators constitute an important advancement in the understanding of the regulation of the T_ex^stem cell fate, as they help to explain how these cells arise and establish their identity upon infection with a persistent pathogen. This study has strong implications for chronic infections and, potentially, tumors, as immunotherapies are known to rely on T_ex^stem cells¹. Furthermore, the identification of BACH2 and other regulatory factors that establish T_ex^stem cells within a week after infection^3,4 suggest that targeting T_ex cell regulation may also aid in the control of some acute infections with pathogens that overwhelm the immune system and exhibit delayed clearance. It would be important, however, to assesshow BACH2-mediated manipulation of the balance between T_ex^stem, T_ex^stem and T_eff-like cells at different times after infection or tumor initiation may impact both host defense and immunopathology.

It is notable that BACH2 also promotes a memory fate in T cells after acute viral infection⁹. This degree of conservation suggests strong evolutionary pressureto establish a molecular circuit that ensures the generation of stem-likecells in response to different stimuliand in different contexts. Given that CD8⁺ T cells dictate the host–pathogen/tumor equilibrium, the generationand maintenance of a fraction of the population as stem-cell-like may have evolved to fulfill the dual needs to sustain protective responses long-term while restraining excessive differentiation into highly pathogenic cellular states that can compromise host fitness during rapidly spreading and/or protracted infections. The findings by Yao et al. suggest that BACH2 could play a crucial role in this delicate balancing act.