高选择性HDAC6抑制剂：它们是免疫调节的答案吗？

Pan-HDAC inhibitors have been approved for the treatment of hematologic malignancies, but their use in solid tumors remains limited due to poor efficacy and significant side effects although direct and immune mediated antitumor effects can be triggered. To overcome these limitations, attention has been focused on the development of selective HDACi. In particular, the generation of selective HDAC6 inhibitors (HDAC6i) has gained interest, since its involvement in tumor progression and immune regulation. HDAC6 knockout mice are viable and fertile, suggesting that selective inhibition could offer therapeutic benefits with minimal toxicity. Among HDAC6i, hydroxamic acid-based compounds such as ITF3756 and ACY-1083 have shown high potency and selectivity, although off-target effects resulting from the binding to other HDACs or non-HDAC targets cannot be ruled out at high concentrations. More recently, difluoromethyl-1,3,4-oxadiazole (DFMO) -based HDAC6i have emerged, offering unprecedented selectivity, prolonged residence time and lack of hydroxamic acid-related limitations. Our group synthesized various DFMO-based HDAC6i and investigated their immunomodulatory potential by evaluating PD-L1 expression and cytokine production in activated human monocytes and CD3 T cells. However, only ITF3756 significantly reduced PD-L1 expression and modulated cytokine release, including TNF-alpha, IL-1beta, IL-10, IFN-gamma, and IL-17. DFMO compounds did not exhibit significant effects on these immune responses. These findings suggest that specific HDAC6 inhibition may fail to reproduce the immunomodulatory effects observed with less selective inhibitors or in HDAC6 knockout models. To further explore this discrepancy, we tested PROTAC-DFMO compounds designed to degrade HDAC6, using either Cereblon (CRBN) or Von Hippel-Lindau (VHL) recruiters. Both PROTACs effectively and selectively degraded HDAC6 and increased alpha-tubulin acetylation. However, only the CRBN-linked PROTAC reduced cytokine production in monocytes, an effect that we determined being attributable to the pomalidomide recruiter itself. In contrast, VHL-linked PROTACs had limited impact. None of the PROTACs significantly modulated cytokine production in T cells. These results indicate that HDAC6 degradation alone is not sufficient to modulate certain immune cell functions, suggesting that broader HDAC inhibition may be required to achieve effective immunomodulatory activity. Furthermore, our results point-out that the effects of degraders in an immunological context may be influenced by their recruiter moiety, rather than being solely driven by target degradation. Overall, this work highlights the complexity of HDAC6's contribution to immune regulation and suggests that while high selectivity may reduce toxicity, it could limit therapeutic efficacy in immuno-oncology contexts.