Destabilization of Fatty Acid Synthase by Acetylation Inhibits De Novo Lipogenesis and Tumor Cell Growth

Abstract

Fatty acid synthase (FASN) is the terminal enzyme in de novo lipogenesis and plays a key role in cell proliferation. Pharmacologic inhibitors of FASN are being evaluated in clinical trials for treatment of cancer, obesity, and other diseases. Here, we report a previously unknown mechanism of FASN regulation involving its acetylation by KAT8 and its deacetylation by HDAC3. FASN acetylation promoted its degradation via the ubiquitin-proteasome pathway. FASN acetylation enhanced its association with the E3 ubiquitin ligase TRIM21. Acetylation destabilized FASN and resulted in decreased de novo lipogenesis and tumor cell growth. FASN acetylation was frequently reduced in human hepatocellular carcinoma samples, which correlated with increased HDAC3 expression and FASN protein levels. Our results suggest opportunities to target FASN acetylation as an anticancer strategy. Cancer Res; 76(23); 6924-36. ©2016 AACR.

Figure 1. Acetylation promotes FASN protein degradation

(A) HEK293T cells overexpressing FASN was treated without or with control solvent, NAM (5 mM, 6 hours) and/or TSA (0.5 mM, 16 hours). Acetylation levels of immunopurified FASN-Flag were determined by Western blot using a pan-anti-acetyllysine antibody (α-AcK). IP denotes immunoprecipitation. Relative FASN acetylation was normalized by Flag protein. (B) Endogenous FASN protein was immunoprecipitated with an anti-FASN antibody from the indicated cells treated with control solvent alcohol or TSA (0.5 mM, 16 hours). Acetylation levels of endogenous FASN were determined by Western blot. (C) Quantification of acetylation levels of endogenous FASN in the indicated cell lines in (B). Relative FASN acetylation ratios were calculated after normalizing against total FASN protein. (D) The indicated cells were treated with control solvent alcohol or TSA (0.5 mM, 16 hours), followed by Western blot to determine endogenous FASN protein levels. (E) Quantification of endogenous FASN protein levels in the indicated cell lines in (D). Relative FASN protein levels were normalized by β-actin. (F) The indicated cells were treated with control solvent alcohol or TSA (0.5 mM, 16 hours), followed by qRT-PCR to determine endogenous FASN mRNA expression. Shown are average values with standard deviation (S.D.) of triplicated experiments. *Denotes P < 0.05, and **denotes P < 0.01 for the indicated comparison; n.s. = not significant.

Figure 2. Acetylation promotes FASN protein polyubiquitylation and destabilization

(A) HCT116 cells were pre-treated with control solvent alcohol or TSA (0.5 mM, 4 hours), and then subjected to CHX treatment (10 μg/ml) for the indicated time course. Endogenous FASN protein level was analyzed by Western blot (left). Quantification of endogenous FASN protein levels are shown (right). (B) HEK293T cells were treated with MG132 (10 μM) for the indicated time course. Endogenous FASN protein level was analyzed by Western blot. Relative FASN protein levels were normalized by β-actin. (C) HEK293T cells were treated with control solvent alcohol or TSA (0.5 μM) in the absence or presence of MG132 (10 μM). Endogenous FASN protein level was analyzed by Western blot, normalized by β-actin protein (top). Please note that the protein levels of endogenous FASN or β-actin in lane 1/2 and lane 3 were determined on the same blot. Quantification of endogenous FASN protein levels are shown (bottom). Shown are average values with standard deviation (S.D.) of triplicated experiments. (D–E) Polyubiquitylation levels of affinity purified FASN-Flag proteins were detected by Western blot in transfected HEK293T cells under treatment with control solvent, NAM and/or TSA in the absence (D) or presence (E) of MG132 (10 μM, 6 hours). (F) Polyubiquitylation, acetylation, and the protein levels of endogenous FASN in HEK293T cells under treatment with control solvent alcohol or TSA (0.5 mM, 16 hours) were detected by Western blot.

Figure 3. KAT8 is a major acetyltransferase of FASN

(A–B) Interaction between the indicated Myc-tagged KAT proteins and endogenous FASN was detected by Western blot. (C) Increased amount of Myc-KAT8 was transiently overexpressed in HEK293T cells, and then acetylation levels of endogenous FASN proteins were determined. (D) Stable HEK293T cells with KAT8 knockdown were generated by retroviral infection, and the knockdown efficiency was determined by qRT-PCR (bottom). Acetylation levels of endogenous FASN were detected by Western blot (top). (E) Myc-KAT8 was transiently overexpressed in HEK293T cells, and then mRNA and protein levels of endogenous FASN were determined by qRT-PCR and western blot, respectively. Relative FASN protein levels were normalized by β-actin. (F) In stable HEK293T cells with KAT8 knockdown, mRNA and protein levels of endogenous FASN were determined by qRT-PCR and western blot, respectively. Shown are average values with standard deviation (S.D.) of triplicated experiments.

Figure 4. HDAC3 is a major acetyltransferase of FASN

(A) Flag-tagged FASN was overexpressed in HEK293T cells in combination with different HA-tagged HDACs as indicated. Flag-FASN protein was immunopurified, followed by Western blot to detect its interaction with HA-tagged HDACs. (B) In HEK293T cells, FASN protein was immunopurified with FASN antibody, followed by Western blot to detect HDAC3 interaction. (C) HA-tagged wild-type HDAC3 or its catalytically inactive mutant HDAC3^Y298H was expressed in HEK293T cells. FASN protein was immunopurified with FASN antibody, followed by Western blot to detect its acetylation. (D) Flag-tagged FASN and HA-tagged Ub plasmids were co-overexpressed in HEK293T cells, which were then subjected to increased concentrations of RGFP966 as indicated. FASN protein was immunopurified with Flag beads, followed by Western blot to detect FASN polyubiquitylation with a HA antibody. (E) HEK293T cells were subjected to increased concentrations of RGFP966 as indicated. Endogenous FASN was immunopurified, followed by Western blot to detect its acetylation and ubiquitylation levels. (F) HEK293T cells were subjected to increasing concentrations of RGFP966 or TSA as indicated. Endogenous FASN mRNA and protein levels were determined by qRT-PCR and Western blot, respectively. Shown are average values with standard deviation (S.D.) of triplicated experiments.

Figure 5. Acetylation promotes FASN interaction with ubiquitin E3 ligase TRIM21

(A) HA-TRIM21 was overexpressed in HEK293T cells, and the transfected cells were treated with control solvent alcohol or TSA (0.5 mM, 16 hours). FASN protein was immunopurified, followed by Western blot to detect its interaction with HA-TRIM21. (B) In HEK293T cells treated with control solvent alcohol or TSA (0.5 mM, 16 hours), FASN protein was immunopurified with FASN antibody, followed by Western blot to detect its interaction with endogenous TRIM21. (C) Myc-tagged wild-type or E3 ligase catalytic inactive mutant TRIM21^ΔRING was overexpressed in HEK293T cells. Endogenous FASN mRNA and protein levels were determined by qRT-PCR and Western-blot, respectively. Shown are average values with standard deviation (S.D.) of triplicated experiments. (D) In HEK293T cells, TRIM21 gene was deleted by the CRISPR-Cas9 system. TRIM21 knockout efficiency and endogenous FASN protein level were determined by Western blot with the indicated antibodies. Relative FASN protein levels were normalized by β-actin. (E) TRIM21 promotes FASN polyubiquitylation. FASN-Flag, Myc-TRIM21, and HA-Ub were co-overexpressed in HEK293T cells as indicated. The transfected cells were treated with control solvents or TSA (0.5 mM, 16 hours) or MG132 (10 μM, 6 hours). FASN-Flag was immunopurified with Flag beads, followed by Western blot to detect its polyubiquitylation with a HA antibody. (F) FASN-Flag and HA-Ub were co-overexpressed in TRIM21 knockout HEK293T cell pools and control cells. The transfected cells were treated with control solvents, TSA (0.5 mM, 16 hours) or RGFP966 (10 μM, 24 hours). FASN-Flag protein was immunopurified, followed by Western blot to detect FASN polyubiquitylation with a HA antibody.

Figure 6. Acetylation of FASN inhibits de novo lipogenesis and suppresses tumor cell growth

(A) HCT116 cells were treated with control solvents, TSA (0.5 mM, 16 hours) or RGFP966 (10 μM, 24 hours). FASN activity in the whole cell lysates were measured (right). FASN protein was immunopurified, followed by Western blot to detect its acetylation. Relative FASN protein levels were normalized by β-actin. (B) HCT116 cells stably expressing empty vector or FASN were treated as mentioned in (A). Lipid content was measured and quantified by employing a SRS microscope as described in ‘Method’. (C–F) HCT116 or ZR-75-30 cells stably expressing empty vector or FASN were seeded in 6-well plates, and were treated with control solvent DMSO or RGFP966 (10 μM) as indicated. Cell numbers were counted every day over a period of 5 days (C and E). At day 5, cells were harvested, and the acetylation and protein levels of FASN were determined by Western blot (D and F). Relative FASN protein levels were normalized by β-actin.

Figure 7. Acetylation of FASN is downregulated in human hepatocellular cancers

(A) In total, 17 pairs of tumor tissues (T) and adjacent normal tissues (N) were lysed. FASN protein was immunopurified with FASN antibody, followed by Western blot to detect its acetylation. Protein levels of FASN, HDAC3, and KAT8 were determined by direct Western blot. Relative FASN and HDAC3 protein levels were normalized by β-actin. Shown are 5 pairs of samples. See Figure S6 for the other 12 pairs of samples. (B–D) Quantification of relative FASN protein, FASN acetylation, and relative HDAC3 protein levels in the 17 pairs of samples tested. (E–F) Correlation between FASN protein levels and its acetylation levels (E) or HDAC3 protein levels (F) in the tested 17 pairs of samples. Shown are average values with standard deviation (S.D.). Statistical analyses were performed with a two-tailed paired t-test. *Denotes P < 0.05, **denotes P < 0.01, and ***denotes P < 0.001 for the indicated comparison. (G) Shown is a working model depicting how acetylation promotes FASN protein degradation through the ubiquitin-proteasome pathway, thereby inhibiting de novo lipogenesis.