Gillian Murdock

Confirmation of ATRT growth dependency on SIRT2 with rescue vector. A. Western Blot demonstrates shRNA mediated knockdown in BT16(MYC) cells and restoring SIRT2 level with ORF rescue vector for SIRT2. B. Membranes for Western Blot assay. C. Colony formation assay with colonies images and quantification. Genetic knockdown of SIRT2 decreases SIRT2 mRNA in BT16(MYC). Rescue vector for SIRT2 confirms growth dependency on SIRT2 in ATRT. ( **p<0.01). D. Real time cell assay with cell growth images and quantification. SIRT2 depletion inhibits BT16(MYC) proliferation as compared to shNULL ( *p<0.05). Rescue vector confirms growth dependency on SIRT2 in ATRT.

SIRT2 is essential for ATRT growth. A. Genetic knockdown SIRT2 decreases SIRT2 mRNA in BT16(MYC) and MAF737(TYR) cells as measured by RT qPCR assay ( **p<0.01). B. Western Blot of SIRT2 demonstrates shRNA mediated knockdown in BT16(MYC) and MAF737(TYR) cells. C. Membranes for Western Blot assay (Figure S2.B). D. Real time cell assay with cell growth images and quantification. SIRT2 depletion inhibits BT16(MYC) and MAF737(TYR) ATRT proliferation as compared to shNULL(*p<0.05).

An atypical teratoid rhabdoid tumor (ATRT) is a highly aggressive pediatric brain tumor driven by the loss of SMARCB1, which results in epigenetic dysregulation of the genome. SMARCB1 loss affects lineage commitment and differentiation by controlling gene expression. We hypothesized that additional epigenetic factors cooperate with SMARCB1 loss to control cell self-renewal and drive ATRT. We performed an unbiased epigenome-targeted screen to identify genes that cooperate with SMARCB1 and identified SIRT2 as a key regulator. Using in vitro pluripotency assays combined with in vivo single-cell RNA transcriptomics, we examined the impact of SIRT2 on differentiation of ATRT cells. We used a series of orthotopic murine models treated with SIRT2 inhibitors to examine the impact on survival and clinical applicability. We found that ATRT cells are highly dependent on SIRT2 for survival. Genetic or chemical inhibition led to decreased cell self-renewal and induction of differentiation in tumor spheres and in vivo models. We found that SIRT2 inhibition can restore gene expression programs lost because of SMARCB1 loss and reverse the differentiation block in ATRT in vivo. Finally, we showed the in vivo efficacy of a clinically relevant inhibitor demonstrating SIRT2 inhibition as a potential therapeutic strategy. We concluded that SIRT2 is a critical dependency in SMARCB1-deficient ATRT cells and acts by controlling the pluripotency–differentiation switch. Thus, SIRT2 inhibition is a promising therapeutic approach that warrants further investigation and clinical development.Implications:SIRT2 inhibition is a molecular vulnerability in SMARCB1-deleted tumors.

All hits in MAF737 ATRT from shRNA screen with fold change<0.8; p<0.05. One of the top hits is SIRT2 (fold change = 0.309, p<0.05).

SIRT2 inhibition suppressed BT16(MYC) ATRT growth in vivo. A. Bioluminescence images of orthotopic ATRT tumors treated with Vehicle (Control) or TM. B. Survival analysis of Vehicle vs TM treated animals (median 19 days vs 33 days, log rank test p<0.05, n=4). C. Immunohistochemical (IHC) stains of cerebellar tumors from BT16(MYC) injected mice are shown (40x magnification). FOXM1 accumulation in mice tumors 4 weeks TM treatment. D. Quantification of Image C. Plots show values from quantification of representative images.TM treatment decreased oncogenic marker FOXM1 (p<0.05). E. Ki67 accumulation in mice tumors 4 weeks TM treatment (40x). F. Quantification of Image E. Plots show values from quantification representative images. Stain accumulation from 5 fields of each slide were analyzed using ImageJ program.TM treatment decreased oncogenic marker KI67 (* p<0.05). G. MEF2A accumulation in mice tumors 4 weeks TM treatment (40x). H. Quantification of Image G (40x magnification). Plots show values from quantification representative images. Stain accumulation from 5 fields of each slide were analyzed using ImageJ program. TM treatment increased accumulation neuronal differentiation marker MEF2A (p<0.05).

Colony Formation Assay and SIRT2 expression in ATRT subgroups. A. CFA for hits genes in MAF737 ATRT from shRNA screen. Representative images of colony formations. B. Quantification of CFA. SIRT2 depletion results in a significant loss of ATRT cell viability (*p <0.001). C. Expression of SIRT2 RNA in patient cohorts in all ATRT subgroups as compared to normal cerebellum. Data from Children's Hospital Colorado.

H&E stained sections showed a high-grade neoplasm consisting of mildly pleomorphic tumor cells with large vesicular chromatin, occasional prominent nucleoli (arrowheads), increased mitoses and zonal necrosis (asterisk). Scattered cells showed rhabdoid features with cytoplasmic eosinophilic inclusions (arrow) and eccentrically placed nuclei.

Tenovin6 and TM treatment effect on different groups ATRT cells and Normal Human Astrocytes. A. Representative images of Colony formation assay following Tenovin6 treatment of BT16(MYC) and MAF737(TYR) . B. Quantification of Colony formation assay. Tenovin6 significantly inhibits clonogenic potential of BT16(MYC) and MAF737(TYR) cells with IC50 5uM and 0.3uM respectively (* p <0.05). C. Cell viability of NHA and ATRT subgroups. Cells treated with indicated doses of Tenovin6. Tenovin6 changes ATRT cells viability differently and slightly changes viability of NHA cells. Tenovin6 treatment with 5uM decreased on 12% cell viability of NHA, significantly changed cell viability of the CHLA04(SHH), CHB-ATRT1(MYC) and BT16(MYC) cells (on 92%; 89 and 50% respectively). D. IC50 of Tenovin6 treatment for different ATRT subgroups and NHA. E. Colony formation assay image of NHA. F. Cell viability of NHA and ATRT subgroups. Cells treated with indicated doses of TM. TM treatment with 20uM did not change cell viability of NHA, slightly changed cell viability of the CHLA04(SHH) (on 9%) and decreased on 50% number of viable cells of the BT16(MYC) and CHB- ATRT1(MYC) cells. G. IC50 of TM treatment for different ATRT subgroups and NHA. H. Colony formation assay image of NHA treated with TM.

SIRT2 inhibition leads to apoptosis in ATRT cells. ATRT cell lines were treated with their respective IC50 of Tenovin6 or TM for 72h and apoptosis was measured using Annexin V FITC reagent. Bar chart showing increased proportion of apoptotic cells after treatment. The representative results of three independent experiments are shown. A, B. Tenovn6 treatment increased apoptosis in BT16(MYC) and MAF737(TYR) cells ( p <0.01). C-F .TM treatment increased apoptosis in BT16(MYC) (p<0.05), MAF737(TYR) (p< 0.05) and CHB-ATRT1(MYC) (p<0.01) and SU-ATRT2(TYR) (p<0.01), a short-term culture of primary patient tumor cells.

SIRT2 therapy demonstrates anti-tumor effect in intracranial orthotopic xenograft models of ATRT. A. IVIS bioluminescence images and tumor volume quantification from Vehicle control and Tenovin6 treated cohorts in BT16(MYC) cells. Data are displayed as mean ± SEM. B. Axial T2-weighted turbo RARE MRI sequences of selected mice from Vehicle and Tenovin6 treated cohorts. Orange arrows indicate tumor at given time point. MRI volumetric analysis (right) demonstrates trend towards decreased tumor volume in the Tenovin6 treated cohort (n=3 each, Tenovin6 vs control Vehicle one-tailed t-test p<0.05).