Page 19 - 2021_05-Haematologica-web

T-cell dysfunction in CLL
IL-4 and IL-21 in the CLL TME are likely produced by T- follicular-helper (Tfh) cells, a distinct subset of CD4+ cells that localize to the LN and expresses PD-1, CXCR5 and CD40 on its cell surface. Tfh cells have overlapping func- tions with other Th cells. The chromatin of its key tran- scription factor BCL-6, shows to be accessible in multiple Th subsets,81 likely accounting for plasticity between Tfh and other Th cells.82 However, Tfh cells also display unique epigenetic features such as enhancer activity on a specific conserved noncoding sequences (CNS) of the IL4 locus, which distinguishes them from Th2 cells.
Tregs, which contribute to immunosuppression in CLL, highly express forkhead box P3 (FOXP3) and are charac- terized by a demethylated region within the FOXP3 gene.79 The core epigenetic program of thymic-derived Tregs is relatively stable but non-Treg cells can also acquire a Treg phenotype after leaving the thymus, known as inducible Tregs (iTregs). Conversion of Treg to Th-17 and vice versa can occur due to inflammatory signals or metabolic alterations of the cellular microenviron- ment.82 Thymic-derived Tregs can be distinguished from iTregs by the epigenetic program on specific CNS within the FOXP3 gene.
Currently, the CAR T-cell production process for CD4+ and CD8+ T cells is identical but as described above, these cells can have opposing roles in CLL. Monitoring the CAR T-cell phenotype before and after infusion is highly rele- vant. Tfh cells, that support CLL cell proliferation in the LN, and Tregs, that provide immune suppression, are par- ticularly detrimental and profiling epigenetic marks such as the ones described above can help to understand the behavior and plasticity of CAR T cells upon infusion and interaction with their target cells.
As we have discussed previously, CD8+ T cells of CLL patients show signs of exhaustion especially by the pres- ence of inhibitory receptors. Exhausted CD8+ T cells are recognized as a separate lineage within CD8+ differentia- tion,51 and although similarities exist to effector and mem- ory cells, many studies have shown extensive epigenetic remodeling in exhausted T-cells.65,66 The key characteristic of T-cell exhaustion, persistent PD-1 expression, is tightly regulated by DNA methylation and chromatin struc- ture.83,84 Within these PD-1 expressing cells we can define several intermediate states with different expression pat- terns and functional capacities. Beltra et al.66 describe four subsets within exhausted CD8+ T cells and show that robust chromatin remodeling occurred during differentia- tion into terminal exhaustion, enforcing different levels of functionality in the subsets. Due to the ability of TOX to bind and recruit a diverse set of chromatin remodeling complexes,57 this transcription factor plays a crucial role in establishing the chromatin landscape of exhausted T cells. Since LN and PB compartments are accessible in the TCL1 murine model, it would be interesting to apply Ly108 and CD69 phenotyping and investigate the exhausted T-cell subsets proposed by Beltra et al.66 in this model of CLL.
Progressive loss of T-cell function during differentiation towards terminal exhaustion has implications for immunotherapy. Pauken et al.23 showed that upon PD-1 blockade therapy, T cells regained some effector function but retained their exhausted epigenetic profile. For this reason, researchers are exploring combinations of epige- netic drugs with ICB to reverse the exhausted epigenetic program and achieve a more potent and long-lasting immune response against the tumor cells;85 a strategy that
might have potential for CLL treatment especially since ICB has not been successful in CLL patients yet.
Metabolism has recently received widespread attention in studying anti-tumor immunity, and chromatin-modify- ing enzymes are highly dependent on metabolites and metabolic cofactors to manage the epigenetic program within a cell.86 Metabolic alterations in the TME can there- fore hamper an adequate immune response through epi- genetic mechanisms.87 When T cells encounter their cog- nate antigen, they switch from oxidative phosphorylation (OXPHOS) to a more glycolysis-based metabolism. A key player in the switch to glycolysis is Sirtuin 6 (SIRT6)88 which belongs to a subclass of histone deacetylases (HDAC) that depend on NAD+ for removal of an acetyl- group.86 It functions as a gene repressor together with hypoxia-inducible factor 1-α (HIF1α), a transcription fac- tor that activates glycolytic genes such as glucose trans- porter-1 and hexokinase-2;89 genes that CLL T cells fail to upregulate after activation.25,90
In addition to diminished glucose metabolism of T cells in the TME, mitochondrial function and fitness of tumor- infiltrating and exhausted T cells is also impaired.91,92 Scharping et al.91 showed that loss of mitochondrial func- tion is due to a defect in mitochondrial biogenesis, which is regulated by peroxisome proliferator-activated receptor- γ co-activator-1α (PGC1α). PGC1α is a transcriptional coactivator and it recruits both histone acetylases and the SWI/SNF chromatin remodeling complex to the genome to activate gene expression.93 Expression of PGC1α itself was shown to be reduced by PD-1 signaling in a chronic infection model, which could explain the relationship between T-cell exhaustion and reduced mitochondrial biogenesis.94 Altogether, PGC1α presents a potential target to improve metabolic function and anti-tumor immunity of T cells.95
Diminished oxygen availability (hypoxia) in the TME arises from an imbalance between increased oxygen con- sumption and inadequate oxygen supply, and leads to ele- vated levels of reactive oxygen species (ROS). A recent study showed that the phosphatase PAC1 (also known as DUSP2) mediated ROS-induced CD8+ T-cell exhaustion and reduced anti-tumor immunity. PAC1 recruits the Mi- 2b nucleosome-remodeling-deacetylase complex, an ATP- dependent chromatin remodeling complex, to the chro- matin of T-cell effector genes. This results in a global reduction in chromatin accessibility and consequently decreased expression of these genes, establishing the dys- functional T-cell state.96
Even though metabolic rewiring is a hallmark of cancer, CLL does not necessarily follow the same paradigm. Due to circulation in and out of the LN and contact with oxygenat- ed blood, the CLL TME might not be nutrient-deprived or hypoxic, although studies on this are limited. Research shows heightened mitochondrial respiration and increased levels of ROS within CLL cells while glycolysis and the pro- duction of lactate is not enhanced.97 Immunosuppression in CLL is most prominent in the LN28 and CD8+ T-cell dysfunc- tion in CLL seems to be instructed in part by metabolic dys- regulation.25,90 However, it is not clear if the metabolic dys- regulation is initiated in the TME and whether it mediates dysfunction through epigenetic mechanisms.
The required metabolic switch from OXPHOS to gly- colysis is impaired in CLL T cells.90 The HDAC SIRT6 reg- ulates glucose homeostasis as co-repressor with HIF1α. Even though it is unknown whether SIRT6 is increased in
haematologica | 2021; 106(5)
1239