The tumor microenvironment disarms CD8⁺ T lymphocyte function via a miR-26a-EZH2 axis

Abstract

One of the most important factors that limit the potency of CD8 C cytotoxic T lymphocyte (CTL) responses is the tumor microenvironment (TME). Here, we provide evidence that miR-26a is a negative regulator of CTL function in the TME. Specifically, we identified miR-26a as a crucial suppressor gene in CTLs from the TME, as we found that, miR-26a expression was elevated in CTLs to respond to TME secretome stimulation. CTLs from miR-26a-transgenic mice showed impaired IFNg and granzyme B production in response to their cognate antigen. Conversely, we found that miR-26a inhibition in CTLs could effectively increase the cytotoxicity and suppress tumor growth. Mechanically, we identified EZH2 as a direct target of miR-26a. miR-26a and EZH2 expression were found to be inversely correlated in CTLs, and the inhibition of EZH2 in CTLs impairs CTL function. These functional correlations were validated in a cohort of non-small cell lung cancer patients, indicating that the miR-26a-EZH2 axis is clinically relevant. Our findings suggested that miR-26a silencing as a novel strategy to improve the efficacy of CTL-based cancer immunotherapy.

Epigenetic regulation of cancer progression by EZH2: from biological insights to therapeutic potential

“…Inhibition of EZH2 in tumor-specific T cells increases the tumor burden and the metastatic potential in mice models of ovarian cancer [ 41 ]. Long et al identified that miR-26a expression is elevated in CTLs responding to tumor microenvironment secretome stimulation, miR-26a inhibits EZH2 to impair CTL function, indicating miR-26a-EZH2 axis as a novel target to improve the efficacy of CTL-based cancer immunotherapy [ 42 ].…”

EZH2 as a Regulator of CD8+ T Cell Fate and Function

“…Using a similar model, in which T cells are cultured in lyophilized conditioned media from tumor cells, Long et al recapitulated the increase in microRNA-26 to corroborate Zhao et al's findings. However, they suggested that unknown soluble factors within the TME mediate the induction of inhibitory microRNA expression (32,33). Using a humanized ovarian cancer model, DZNep-treated T cells were unable to control tumor metastasis to the same extent as control counterparts (33); similarly, CD8+ T cells transduced with a microRNA-26 decoy reduced tumor growth rate in an adoptive transfer melanoma model (32).…”

“…Ezh2 is negatively affected by factors present in the tumor microenvironment (TME) as evidenced by the reduction of Ezh2 transcription when T cells were activated in the presence of lyophilized tumor supernatant ( 32 , 33 ). Notably, one study observed that human CD8+EZH2+ T cells did not express KLRG1, TIM-3, or CD57, suggesting that EZH2 is primarily expressed in cycling or activated cells, not in senescent or anergic cells ( 15 , 33 , 34 ).…”

“…Phenotypic analysis of human CD8+EZH2+ cells suggested that this population had greater effector capacity and reduced sensitivity to apoptosis ( 33 ). CD8+EZH2+ T cells from healthy donors and ovarian cancer patients were polyfunctional, defined by their capability to coproduce IFNγ, TNFα, and granzyme B, and resistant to apoptosis, in part due to their elevated Bcl-2 expression ( 32 , 33 ). Consistent with this, inhibition of EZH2 via pharmacological inhibitors (DZNep or GSK126) or RNA interference (RNAi) reduced the frequency of polyfunctional (triple positive) human CD8+ T cells and increased the frequency of apoptotic cells in vitro ( 33 ).…”

“…However, they suggested that unknown soluble factors within the TME mediate the induction of inhibitory microRNA expression ( 32 , 33 ). Using a humanized ovarian cancer model, DZNep-treated T cells were unable to control tumor metastasis to the same extent as control counterparts ( 33 ); similarly, CD8+ T cells transduced with a microRNA-26 decoy reduced tumor growth rate in an adoptive transfer melanoma model ( 32 ). As effector CD8+ T cells are reliant on glucose, it is tempting to speculate that while Akt signaling induces glucose receptor expression, without substrate in the glucose-poor TME, these inhibitory microRNA's are induced to suppress T cell function via EZH2 inhibition ( 40 ).…”