Activation of SIRT1 Deacetylase As a Therapeutic Approach for Myelodysplastic Syndromes By Restoring TET2 Function

Myelodysplastic syndromes (MDS) which arise from a primitive CD34⁺ hematopoitic stem/progenitor cell (HSPC), are incurable by non-transplant therapy. Better understanding of MDS HSPC regulation is critical for developing targeted therapies against the fatal disease. Tet methylcytosine dioxygenase 2 (TET2), initiating DNA demethylation through converting 5-methylcytosine to 5-hydromethylcytosine (5hmC), is the most frequently mutated gene in MDS. TET2 alterations lead to DNA hypermethylation and dysregulated gene expression in HSPCs, resulting in enhanced self-renewal and skewed myeloid specific expansion. While TET2's biological function has been well studied, the knowledge of its regulation is limited. We had demonstrated the TET2 c-terminus acetylation was reversed by SIRT1 deacetylase which maintains normal HSPCs activity under stress. We also observed that SIRT1 protein in MDS CD34⁺CD38^- was expressed below normal levels, and that knockdown (KD) of SIRT1 enhanced in vivo growth of MDS-L cells, a patient derived MDS cell line, is associated with TET2 acetylation (Blood. 2016;128:1053). These results indicate an important role for SIRT1 in MDS HSPCs maintenance.

To further assess SIRT1 function in MDS, we ectopically expressed SIRT1 in primary MDS CD34⁺ cells enriched for HSPCs and MDS-L cells, respectively. Overexpression (OE) of SIRT1 resulted in inhibition of colony formation of MDS CD34⁺ cells (p<0.01, n=3) but not normal counterpart (p>0.05, n=3). SIRT1 OE also suppressed MDSL colony formation (p<0.01) and cell growth (p<0.01). SIRT1 OE significantly reduced MDS-L cell BM engraftment in NSGs mice relative to controls (human CD45⁺ cells: 5.7±0.8% of BM in SIRT1 OE group vs 31.4 ± 5.8% in control, p<0.01, n=7), resulting in delayed mortality of recipient mice (median survival: 52 days vs 35 days; p<0.05, n=5). Meanwhile, treatment of SIRT1 agonist SRT1720, reduced colony formation of primary MDS CD34⁺ cells in a dose dependent manner (p<0.01, n=9). Similar results were also achieved in SRT1720 treated MDS-L cells. Furthermore, TET2 was hyperacetylated in primary MDS cells (n=3) compared to normal counterpart (n=3). Next, we tested whether inhibition of MDS HSPCs following SIRT1 activation was related to TET2 activation via protein deacetylation. Indeed, enhanced TET2 activity marked by 5hmC was observed in MDS CD34⁺ cells upon SRT1720 treatment (P<0.05, n=3). TET2 KD diminished growth inhibition induced by SIRT1 OE or SRT1720, indicating a requirement of TET2 for SIRT1 function. Through mass spectrometry analysis, three conserved acetylated lysine residuals (K1472, K1473, K1478) were identified in TET2 catalytic domain (CD). Engineered lysines to glutamines (TET2 3KQ) mutant, mimicking acetylated TET2, produced lower 5hmC, and led to increased colony growth compared to MDS-L expressing TET2 WT (p<0.01). We further evaluated the effects of SRT1720 using NUP98-HOXD13 (NHD13)MDS mouse. This murine MDS is transferable and secondary recipients receiving primary NHD13 MDS cells succumb to MDS at 4 months, allowing generation of a cohort of MDS mice for preclinical studies of therapeutics against MDS. CD45.2⁺ BM cells from donor NHD13⁺ mice showed evidence of dysplasia were harvested, and transplanted into preconditioned congenic recipients expressing CD45.1. SRT1720 treatment (100mg/kg/day for 12 weeks, p.o.) resulted in partial reversal of MDS-like disease developed in recipients receiving MDS cells, including an increase in HGB, PLT, RBC and WBC (p<0.01,n=6). Importantly, significant reduction of NHD13⁺ CD45.2⁺ chimerism in LSK and other progenitors from BM were also observed (p<0.01, n=6). As expected, 5hmC level in lineage depleted cells from SRT1720 treated mice was increased (p<0.01). Secondary transplantation was performed to evaluate the effects of SRT1720 on MDS initiating cells. We observed that the mice receiving cells from SRT1720 treated donors showed reduced NHD13⁺ CD45.2⁺ cell long-term engraftment (p<0.01, n=5).

In summary, we demonstrated that genetic overexpression or pharmacological activation of SIRT1 drastically inhibited MDS maintenance, and that TET2 function was required for the inhibitory effect of SIRT1. These results support the hypothesis that SIRT1 deficiency promotes MDS cell expansion through TET2 inactivation and support the rationale for exploring SIRT1 activation as a promising approach to selectively target MDS stem cells.