Fig 2: ACSS2 Controls Acetate Incorporation Into Lipogenic Acetyl-CoA(A) Schematic of lipogenic AcCoA production.(B) Steady-state 13C labeling (percent) of lipogenic AcCoA from 90 μM U-13C-acetate (Ac) in normoxia or hypoxia (1% O2).(C) Steady-state 13C labeling (percent) of lipogenic AcCoA from U-13C-glucose (Gluc), U-13C-glutamine (Gln), and 90 μM U-13C-acetate.(D) Western blot of ACSS2 from cells transfected with scrambled RNA (SCR) or two independent ACSS2 siRNAs. The normalized ratio is relative to the SCR control.(E) Steady-state 13C labeling (percent) of lipogenic AcCoA from 90 μM U-13C-Ac in hypoxic SCR or ACSS2 siRNA-treated cells.(F) Steady-state 13C labeling (percent) of lipogenic AcCoA from U-13C-Gluc and U-13C-Gln in hypoxic SCR or ACSS2 siRNA-treated cells. The medium contained 90 μM 12C-acetate.(G) ACSS2 western blot from multiple human cancer cell lines under hypoxia (48 hr). Tubulin was used as the loading control.(H) Steady-state 13C labeling (percent) of lipogenic AcCoA from 500 μM U-13C-Acetate.(B–F) Experiments were done in MDA-MB-468 cells.All data are mean ± SD (n = 3); ∗p < 0.05, ∗∗p < 0.01. See also Figure S1.

Fig 3: Substrate use during fractionation follows defined metabolic pathways. A) Schematic overview of labeling approach to test for post-harvest metabolism. Cells were cultured in unlabeled media and 13C-labeled substrate was introduced in the fractionation buffer before fractionation processing and quenching of fractionations in trichloroacetic acid (TCA). B) Incorporation of 15N213C5-glutamine or 13C5-α-ketoglutarate into the mitochondrial Krebs cycle results in labeling of 4 acyl carbons of succinyl-CoA. The enzymatic pathway for incorporation of 13C5-α-ketoglutarate into succinyl-CoA M4 in the cytosol is not defined. C) Hepatocellular carcinoma (HepG2) cells were harvested in fractionation buffer containing either 0.5 mM 15N213C5-glutamine or 0.5 mM 13C5-α-ketoglutarate and % molar enrichment of acyl labeling was calculated by normalization to unlabeled control samples D) Acyl-CoA synthetase short-chain family member 2 (ACSS2) catalyzes the conversion of acetate to acetyl-CoA. HepG2 cells were harvested in fractionation buffer containing 0.1 mM 13C2-acetate and % molar enrichment of acetyl-CoA isotopologues was calculated by normalization to control samples with unlabeled acetate in the fractionation buffer for mito/cyto (E) and nuclear (F) fractionation protocols. G) Isotopologue enrichment was analyzed in de-proteinated HepG2 cell extracts resuspended in fractionation buffer for mito/cyto or nuclear protocols containing 0.1 mM 13C2-acetate or unlabeled acetate and incubated on ice for 1 h before extraction. H) ACLY KO, control and ACSS2 KO MEF cells were harvested in fractionation buffer containing 0.1 mM 13C2-acetate and acetyl-CoA M2 enrichment calculated by normalization to control samples harvested in fractionation buffer containing 0.1 mM unlabeled acetate. I) Western blot analysis comparing ACSS2 expression levels in ACLY KO, control and ACSS2 KO MEF cells with equal protein loading. Tubulin and histone H3 serve as loading controls and fraction-specific markers for non-nuclear and nuclear fractions, respectively. The fractionation buffers for both these methods are hypotonic sugar solutions, with the most notable difference being the presence of dilute detergent (0.1% NP-40) in the nuclear fractionation buffer. This detergent destroys the mitochondrial membrane such that the non-nuclear fraction is expected to contain soluble mitochondrial contents. Representative experiments are shown with bars representing the mean of n = 3 distinct replicate samples (individual symbols) and error bars indicating standard deviation. For comparison between two groups, datasets were analyzed by two-tailed student's t-test with statistical significance defined as p < 0.05 (*), p < 0.01 (**), p < 0.001 (***).

Fig 4: Acss2 protein levels are markedly reduced in ED Acss2 liver.(A) Immunohistochemical and (B) immunoblot assays (two independent samples per group; normalized to protein amount and compared to immunoblot of α-tubulin) reveal reduced staining of Acss2 protein in liver of ED Acss2 mice with some immunoreactivity noted around the portal vein (PV), hepatic artery (HA), and bile duct (BD). (C) Acss2 mRNA levels in ED liver are reduced to approximately 50% of levels found in WT mice (mean with SEM, n = 8 mice per group, *P = 0.016).

Fig 5: Depletion of ACSS1 and ACSS2 increases cell death, reduces proliferation, and reduces cell viability. (A-H) MCF7 parental and MCF7 LTED cells were transfected with control siRNA or siRNA against ACSS1 or ACSS2 and then collected for analysis after 5 days. (A) Cells were counted, and relative cell number was plotted. (B) Cell viability was assessed using trypan blue staining to detect non-viable cells. (C) Cells were collected for cell cycle analysis and the percentage of sub-G1 cells plotted. (D) mRNA expression was analyzed by qPCR for the indicated genes. (E) Cells were counted, and relative cell number was plotted. (F) Cell viability was assessed using trypan blue staining to detect non-viable cells. (G) Cells were collected for cell cycle analysis and the percentage of sub-G1 cells plotted. (H) mRNA expression was analyzed by qPCR for the indicated genes. Averages from three separate experiments are presented with standard error of the mean for A, B, C, E, F, and G. Significance of P < 0.05 is indicated by one asterisk; significance of P < 0.01 is indicated by two asterisks; significance of P < 0.001 is indicated by three asterisks; and significance of P < 0.0001 is indicated by four asterisks.