Targeting ACLY sensitizes castration-resistant prostate cancer cells to AR antagonism by impinging on an ACLY-AMPK-AR feedback mechanism

Abstract

The androgen receptor (AR) plays a central role in prostate tumor growth. Inappropriate reactivation of the AR after androgen deprivation therapy promotes development of incurable castration-resistant prostate cancer (CRPC). In this study, we provide evidence that metabolic features of prostate cancer cells can be exploited to sensitize CRPC cells to AR antagonism. We identify a feedback loop between ATP-citrate lyase (ACLY)-dependent fatty acid synthesis, AMPK, and the AR in prostate cancer cells that could contribute to therapeutic resistance by maintaining AR levels. When combined with an AR antagonist, ACLY inhibition in CRPC cells promotes energetic stress and AMPK activation, resulting in further suppression of AR levels and target gene expression, inhibition of proliferation, and apoptosis. Supplying exogenous fatty acids can restore energetic homeostasis; however, this rescue does not occur through increased β-oxidation to support mitochondrial ATP production. Instead, concurrent inhibition of ACLY and AR may drive excess ATP consumption as cells attempt to cope with endoplasmic reticulum (ER) stress, which is prevented by fatty acid supplementation. Thus, fatty acid metabolism plays a key role in coordinating ER and energetic homeostasis in CRPC cells, thereby sustaining AR action and promoting proliferation. Consistent with a role for fatty acid metabolism in sustaining AR levels in prostate cancer in vivo, AR mRNA levels in human prostate tumors correlate positively with expression of ACLY and other fatty acid synthesis genes. The ACLY-AMPK-AR network can be exploited to sensitize CRPC cells to AR antagonism, suggesting novel therapeutic opportunities for prostate cancer.

Keywords: AMPK; ER stress; acetyl-CoA; fatty acid metabolism; prostate cancer.

Figure 1. ACLY inhibition sensitizes castration-resistant prostate cancer cells to AR antagonism under androgen depletion

A. C4-2 cells were cultured in androgen-depleted conditions for 24 hours and then treated with doses of enzalutamide (ENZ) ranging from 0-40 μM, in the presence or absence of 10 μM ACLY inhibitor (ACLYi; BMS-303141), for 72 hours. Cell number was quantitated relative to cell number at the start of inhibitor treatment and initial cell number set at 100. B-E. Indicated cell lines were cultured for 72 hours in androgen-depleted conditions, +/− ACLYi, ENZ, or Bicalutamide (Bical). Cell number was quantitated relative to starting cell number, set at 100. F, G. Cells were treated for 24 hours with indicated inhibitors in androgen-depleted conditions. Immunofluorescent imaging was used to detect the proliferation indicator Ki67 (F) or apoptotic indicator cleaved caspase 3 (CCP3) (G). Data was quantified and graphed from 5 fields. For all values, mean +/− SEM of triplicates is graphed; *, p<0.05; **, p<0.01, ***, p<0.001.

Figure 2. Combining ACLY and AR inhibitors suppresses AR target gene expression

A. C4-2 cells were cultured in androgen-depleted conditions for 24 hours and then treated with DHT (10 nM), +/− ACLYi (10 μM), +/− ENZ (20 μM) for an additional 24 hours. Gene expression was analyzed by Q-RT-PCR and normalized to 18S rRNA. Combo denotes ACLYi + ENZ. B. C4-2 cells were cultured in androgen-depleted conditions for 24 hours and then treated +/− ACLYi (10 μM), +/− ENZ (20 μM) for an additional 24 hours. Gene expression was analyzed by Q-RT-PCR and normalized to B2M. C. Cells were treated with ACLYi or ENZ for 48 hours in androgen-depleted conditions and AR levels analyzed by Western blot. D. C4-2 cells in androgen-depleted conditions were treated with vehicle, ACLYi alone, ENZ alone, or the inhibitor Combo. Genes differentially expressed in any two comparisons (FDR<0.0001) were clustered and are represented in the heat map. See Supplementary Table S1 for associated gene list and cluster ID. E. Gene set enrichment analysis (GSEA) was used to test whether the “DHT-Up” and “Androgen Independent” gene signatures, as defined by Decker et al [7], were enriched in Ctrl vs. Combo-treated cells. Both of these gene sets were enriched in control C4-2 cells as compared to Combo-treated C4-2 cells. Normalized enrichment score (NES) and p-values are indicated. See Supplementary Figure S2 for comparisons between Ctrl vs. ACLYi alone. For all bar graphs, mean +/− SEM of triplicates is graphed; *, p<0.05; **, p<0.01

Figure 3. Combining ACLY and AR inhibitors induces endoplasmic reticulum stress

A, B. RNA-sequencing data was analyzed using the HOMER tool [46]. GOTERM_BP and KEGG_PATHWAY analyses are shown for the purple cluster (downregulated genes, A) and blue cluster (upregulated genes, B). C. C4-2 cells were treated with indicated inhibitors for 24 hours under androgen deprivation and analyzed by Western blot. D. C4-2 cells were treated with ACLYi + ENZ under androgen deprivation for indicated times and analyzed by Western blot and Q-RT-PCR. E. C4-2 cells were treated for 1 or 24 hours with indicated inhibitors in standard serum (not androgen depleted) and analyzed by Western blot. F. AKT phosphorylation was analyzed by Western blot after treating C4-2 cells as indicated under androgen deprivation. G. C4-2 cells were pre-treated for 1 hour with rapamycin, then treated with Combo for an additional hour and analyzed by Western blot. For all bar graphs, mean +/− SEM of triplicates is graphed; *, p<0.05

Figure 4. ACLY and AR inhibition promotes AMPK activation without suppressing glucose uptake or oxidation

A. C4-2 cells were treated as indicated for 24 hours under androgen deprivation and analyzed by Western blot. B. C4-2 cells were treated with Combo over a time-course and analyzed by Western blot and Q-RT-PCR. C. Consumption of glucose and glutamine from media and production of lactate were analyzed from control and Combo-treated C4-2 cells using a YSI-7100 Bioanalyzer. D. [U-¹³C]-glucose labeling of C4-2 cells was conducted, +/− Combo, for 3 hours. Labeling was initiated at the same time as application of the inhibitors. Enrichment in TCA cycle intermediates was assessed. E. Oxygen consumption was measured in androgen-depleted C4-2 cells, treated +/− Combo for 24 hours. For all bar graphs, mean +/− SEM of triplicates is graphed; *, p<0.05.

Figure 5. Fatty acid supplementation restores energetic homeostasis in ACLY/AR inhibitor-treated cells, without directly supporting ATP production through β-oxidation

A. C4-2 cells in androgen-depleted conditions were treated with vehicle, Combo alone, or Combo plus indicated metabolites including fatty acids (75 μM, 50:50 Palmitic acid (PA): Oleic acid (OA), BSA conjugated), dimethyl-α-KG (1 mM), mevalonate (100 μM), Na-pyruvate (5 mM), acetate (500 μM) for 72 hours. After 72 hours, viable cells were counted using trypan blue exclusion and normalized to staring cell number, set to 100. B. Immunoblotting was carried out on C4-2 cell lysates treated with vehicle or Combo, +/− fatty acids for 1 hour or 24 hours. C. Gene expression in C4-2 cells treated with vehicle or Combo, +/− fatty acids for 24 hours. D. C4-2 cells were treated with Combo +/− fatty acids for 1 hour or 24 hours. AMP and ATP were measured by HPLC. Experiment was repeated 4 times, with relative AMP/ATP ratio in untreated cells for each experiment set to 1. 2-tailed paired t-test was conducted. E. To assess the relative contribution of glucose and fatty acids to TCA cycle intermediates in control and Combo-treated cells, labeling studies were conducted. Cells were labeled either with 10 mM [U-¹³C]-glucose + 75 μM unlabeled PA: OA or with 75 μM [U-¹³C]-PA: [U-¹³C]-OA + 10 mM unlabeled glucose. Cells were treated +/− Combo concurrent with initiation of labeling, and metabolite enrichment was analyzed after 6 hours. Unless otherwise indicated, mean +/− SEM of triplicates is graphed; *, p<0.05; **, p<0.01, ****, p<0.0001.

Figure 6. Exogenous fatty acids prevent ER stress and support cell proliferation and viability during treatment with ACLY and AR inhibitors

A. ENTPD5 gene expression analyzed by Q-RT-PCR after 24 hours Combo treatment in androgen-depleted C4-2 cells. B. siRNA targeting ENTPD5 or a non-silencing control siRNA were transfected into C4-2 cells, and after 96 hours, cells were treated +/− Combo for an additional 24 hours before analyzing by Western blot. C. LNCaP-Abl cells were treated with indicated inhibitors, +/− 75 μM fatty acids (50:50 PA:OA). D. Gene expression was analyzed by Q-RT-PCR at indicated time points, +/− Combo, +/− 75 μM fatty acids (50:50 PA:OA). E, F. At indicated time points following Combo treatment, +/− fatty acids, apoptotic marker CCP3 (E) or proliferation marker Ki67 (F) were assessed. For all bar graphs, mean +/− SEM of triplicates is graphed; *, p<0.05; **, p<0.01, ***, p<0.001. G. Using the www.cBioportal.org resource [81, 82], AR mRNA levels were tested for correlation with the indicated genes in the prostate cancer TCGA Prostate Adenocarcinoma (mRNA expression, RNA-Seq V2 RSEM) and the Metastatic Prostate Cancer, SU2C/PCF Dream Team (mRNA Expression/polyA, RNA-Seq RPKM) [16, 67]. The key indicates the strength of the correlation, as assessed by Pearson's r. For TCGA data, with n=333, all correlations are significant to p<0.0001. For the metastatic dataset, with n=150, ACLY and FASN correlations are significant to p<0.0001; ACACA correlation is significant to p<0.01.

Figure 7. Model: An ACLY-AMPK-AR network sustains ER homeostasis, AR levels, and cell survival in prostate cancer

A. In proliferating PTEN null prostate cancer cells, the demands of oncogenic signaling are met by metabolic rewiring including de novo fatty acid synthesis, which supports endoplasmic reticulum (ER) homeostasis and suppresses AMPK activation to promote proliferation. B. Upon inhibition of AR, AKT is hyper-activated, placing increased demand on the de novo fatty acid synthesis pathways to support ER homeostasis. If ACLY and hence fatty acid synthesis is also inhibited, unresolvable ER stress ensues. In the PTEN null context, this can trigger ENTPD5 activation, excessive ATP consumption, and AMPK activation. AR can then be further suppressed by AMPK and AKT, two negative regulators. The combination of AR suppression and pro-apoptotic ER stress may induce cell death. Thus, the ACLY-AMPK-AR feedback loop can be exploited to sensitize CRPC cells to AR antagonism.