Withaferin A Inhibits Fatty Acid Synthesis in Rat Mammary Tumors

Withaferin A (WA) is a small molecule derived from the root of a medicinal plant (Withania somnifera) that is still used in Ayurvedic medicine formulations in India and surrounding countries (1). Clinical trials have investigated the effects of Withania somnifera on various conditions, including maximum oxygen consumption, neurodegenerative diseases, anxiety, cognitive dysfunction etc. (2–5). Preclinical studies have determined the effects of Withania somnifera on different solid tumors, including breast cancer (6, 7). The root extract of Withania somnifera contains multiple compounds, but WA is the predominant anticancer phytochemical in this plant (8, 9). For example, although WA treatment significantly inhibited proliferation of MCF-7 and MDA-MB-231 human breast cancer cells, such an effect was not observed with withanone or withanolide A that are also present in Withania somnifera (9). Mechanistic studies have revealed G₂–M phase and mitotic arrest and induction of apoptosis as well as autophagy in breast cancer cells after WA treatment (9–13). At the molecular level, the mitotic arrest by WA treatment in breast cancer cells was accompanied by downregulation as well as covalent binding at cysteine 303 of β-tubulin (9). The apoptosis induction in breast cancer cells following WA treatment was triggered by reactive oxygen species-mediated activation of Bax and Bak (12). The SV40-immortalized embryonic fibroblasts derived from Bax and Bak double knockout mice were more resistant to WA-induced apoptosis compared with fibroblasts derived from wild-type mice (12). On the other hand, inhibition of autophagy did not have a meaningful impact on growth inhibition by WA treatment at least in human breast cancer cells (13). WA is known to inhibit other oncogenic transcription factors, including estrogen receptor-α, signal transducer and activator of transcription 3 (STAT3), forkhead box (Fox) Q1, and FoxO3a (14–17). A much higher concentration of WA is required to induce apoptosis in the normal human mammary epithelial cell line MCF-10A when compared with breast cancer cells (14).

Our laboratory was the first to investigate in vivo effects of WA treatment using rodent models of breast cancer (14, 18, 19). Intraperitoneal administration of 4-mg WA/kg body weight 5 times/wk reduced the growth of MDA-MB-231 xenografts by about 45% (14). Administration of 100-μg WA/mouse (three times/wk) for 28 weeks significantly decreased macroscopic mammary tumor size, microscopic mammary tumor area, and the incidence of pulmonary metastasis in MMTV-neu transgenic mice (18). The area of invasive cancer was lower by about 95% in the WA-treated MMTV-neu mice when compared with the solvent-treated control mice (18). In a rat model of luminal-type breast cancer induced by a single injection of N-methyl-N-nitrosourea (MNU), WA administration (4 or 8 mg/kg body weight, five times/wk) decreased the incidence, multiplicity, and tumor burden (19). As an example, the wet tumor weight in the 8 mg/kg group was decreased by about 68% compared with control rats (19). In each of the rodent study, WA treatment did not cause weight loss or any other toxicity (14, 18, 19). WA-mediated inhibition of breast cancer growth in mice and rats was associated with a significant decrease in proliferation marker like proliferating cell nuclear antigen as well as apoptosis induction (14, 18, 19). Inhibition of breast cancer development in MMTV-neu mice and MNU-rats was accompanied by accumulation of mitotic cells (19).

Recently, we performed RNA-seq analysis to identify novel mechanistic targets of WA (20). The RNA-seq data revealed downregulation of fatty acid synthesis genes. The present study was undertaken to determine the effect of WA treatment on fatty acid synthesis using human breast cancer cell lines (MCF-7 and MDA-MB-231) and plasma and mammary tumor tissues of control and WA-treated MNU-rats.

The use of rats was approved by the University of Pittsburgh Institutional Animal Care and Use Committee (protocol numbers 21100115, 20128342, and 20128342).

WA (purity 95.8%) was purchased from ChromaDex, dissolved in dimethyl sulfoxide (DMSO), and stored at −80°C. Cell culture media were from MediaTech. FBS was obtained from Atlanta Biologicals. Antibiotic mixture and other cell culture reagents were purchased from Life Technologies-Thermo Fisher Scientific. The antibodies were purchased from the following vendors: anti–acetyl-CoA carboxylase 1 (ACC1) antibody was from Proteintech Group; antibody against fatty acid synthase (FASN) was from Cell Signaling Technology; anti-ATP citrate lyase (ACLY), anti-carnitine palmitoyltransferase 1A (CPT1A), and anti-sterol regulatory element–binding protein 1 (SREBP1) antibodies were from Abcam; Alexa fluor 488–conjugated goat anti-rabbit antibody was from Life Technologies; fluorescein isothiocyanate-conjugated anti-CD3 antibody and allophycocyanin-conjugated anti-CD161 antibody were from BioLegend; BV711-conjugated anti-CD4 antibody was from BD Biosciences; and anti-granzyme B antibody and anti-perforin 1 antibody for IHC were from Lifespan Biosciences. Kits for determination of total free fatty acids and total phospholipids, and anti-ACC1 antibody for IHC were purchased from Sigma-Aldrich.

The MCF-7 and MDA-MB-231 cell lines were procured from the ATCC and each cell line was maintained as suggested by the supplier. These cell lines were last authenticated by us in March of 2017.

The RNA-seq data presented in this study have been submitted to the Gene Expression Omnibus of NCBI (GSE158085).

The expression of ACLY, ACC1 (also known as ACACA), FASN or CPT1A in normal breast as well as in different subtypes of breast cancer was determined from The Cancer Genome Atlas (TCGA) dataset using the following link from the University of California Santa Cruz (https://tcga.xenahubs.net).

The MCF-7 and MDA-MB-231 cells were plated in triplicate on coverslips in 24-well plates. After overnight incubation, the cells were treated with DMSO (control) or WA for 24 hours. The cells were then fixed and permeabilized with 2% paraformaldehyde and 0.5% Triton X-100, respectively. After blocking with PBS containing 0.5% BSA and 0.15% glycine, the cells were treated overnight at 4°C with the anti-ACLY, anti-ACC1 or anti-FASN antibody followed by incubation with Alexa fluor 488–conjugated goat anti-rabbit antibody for 1 hour. Corrected total cell fluorescence was determined using ImageJ software. For microscopic analysis of CPT1A, the cells were treated with DMSO or WA in triplicate on coverslips in 24-well plates. The cells were then stained with 100 nmol/L MitoTracker Red in complete medium at 37°C for 1 hour followed by washing and fixation of the cells with 4% formaldehyde in complete medium at 37°C for 15 minutes. The cells were permeabilized with 0.5% Triton X-100 for 5 minutes and incubated for 1 hour with blocking solution consisting of PBS, 0.5% BSA, and 0.15% glycine, and then stained with anti-CPT1A antibody at 4°C overnight followed by Alexa fluor 488–conjugated goat anti-mouse secondary antibody for 1 hour at room temperature.

Thirty female Sprague Dawley rats (20 days old) were purchased from Charles River Laboratories. On 21 days of age, the rats were randomly divided into two groups. The MNU group of rats (n = 15) received single intraperitoneal injection of 50 mg/kg of MNU in 0.9% sodium chloride solution, whereas the control group of rats (n = 15) received single injection of 0.9% sodium chloride solution (IACUC protocol number: 21100115). All rats were fed with regular 5P76-Prolab Isopro RMH 3000 diet and sacrificed after 10 weeks post-injection. Thirteen rats from the MNU group developed mammary tumors at the end of study. No tumors were found in the control rats. After sacrifice, blood was collected from all rats to prepare plasma. Mammary tumors from MNU group of rats and normal mammary glands from the control rats were collected and processed for the determination of fatty acid levels and western blot analysis. In a separate study (IACUC protocol number: 20128342), a total of 20 MNU-treated rats were divided into two groups and the rats of the control group were treated intraperitoneally with vehicle, consisting of 10% DMSO, 40% Cremophor EL/ethanol (3:1), and 50% PBS, five times/wk for 7 weeks. The rats in the WA treatment groups were treated with vehicle containing 4 mg WA/kg body weight five times/wk for 7 weeks. This study was used to determine the in vivo effect of WA administration on plasma levels of granzyme B and perforin 1 and tumor expression of Granzyme B and perforin 1 as well as immunophenotyping of tumor and spleen tissues. Effect of WA administration on expression of fatty acid–metabolizing proteins (ACLY, ACC1, FASN, and CPT1A) and SREBP1 as well as levels of total free fatty acids and total phospholipids were determined using the fresh frozen plasma and tumor tissues as well as paraffin-embedded tumor tissue from our previously published study (ref. 19; IACUC protocol number: 14064037).

The lysates from cultured cells and tissues were prepared as described by us previously (18, 19, 21). Western blotting was performed as described by us previously (18, 21). Western blots were stripped and re-probed with anti–β-actin antibody to correct for protein loading. Immunoreactive bands were visualized by enhanced chemiluminescence method and quantified by densitometric scanning using UN-SCAN-IT gel analysis software (Version 7.1, Silk Scientific Corporation).

Levels of total free fatty acids (C8 and longer) and total phospholipids in the plasma or tissue supernatants of normal mammary glands and mammary tumors were determined using commercially available kits as described by us previously (22).

Mammary tumor sections from our previous study (19) were de-paraffinized, hydrated, and immersed in boiling citrate retrieval buffer solution (pH 6.0) for 20–30 minutes followed by treatment with 0.3% hydrogen peroxide in 100% methanol for 20 minutes at room temperature. Sections were exposed to blocking buffer (PBS consisting of 5% normal goat serum) for 1 hour at room temperature followed by overnight incubation with anti-ACLY (1:200 dilution), anti-ACC1 (1:150 dilution), anti-FASN (1:250 dilution) or anti-CPT1A (1:50 dilution) antibody for 24 hour incubation with anti-granzyme B (1:200 dilution) and anti-perforin (1:44 dilution) antibody in humidified chambers at 4°C. Sections were washed with PBS, incubated with horseradish peroxidase–conjugated secondary antibody (1:150∼400) for 2 hours at room temperature. Color was developed by incubation with 3,3′-diaminobenzidine tetrahydrochloride. Sections were counterstained with hematoxylin and examined under Leica microscope equipped with DFC 450C digital camera. The H-score was determined as described previously by us previously (18).

Plasma level of granzyme B and perforin 1 was measured using ELISA kits commercially available from LifeSpan BioSciences by following supplier's instructions.

Tumor tissues from rats were cutoff into small pieces, minced with a scalpel, and incubated in digesting medium containing collagenase/hyaluronidase/DNase at 37°C for 30 minutes with gentle stirring. Tissue homogenates were sieved through a 70-μm cell strainer to obtain single cell suspension. Red blood cells lysis buffer (BD Biosciences) was added to remove these cells from single-cell suspensions. The cells were counted and incubated with desired fluorochrome-conjugated antibodies and then processed for multicolor flow cytometry using Cytek Aurora flow cytometer (Cytek Biosciences).

Harvested spleen tissues from rats were placed in 70-μm cell strainer, minced using 1-mL syringe, and then flushed with cold PBS. Red blood cells were removed. The splenocytes were counted and stained with desired fluorochrome-conjugated antibody and then processed for flow cytometry using Cytek Aurora flow cytometer.

The MDA-MB-231 cell line was transiently transfected with empty vector (pcDNA3; hereafter abbreviated as EV) or pcDNA3.1–2xFLAG-SREBP1C plasmid (hereafter abbreviated as SREBP1) using FuGENE HD transfection reagent (Promega). The pcDNA3.1–2xFLAG-SREBP-1c was a gift from Timothy Osborne (Addgene plasmid # 26802; ref. 23). The cells were treated for 24 hours with either DMSO or 2 μmol/L of WA.

Total RNA from MCF-7, MDA-MB-231, and EV and SREBP1 MDA-MB231–overexpressing cells was isolated using RNeasy kit and cDNA was synthesized using Superscript Reverse Transcriptase (Invitrogen-Life Technologies) with oligo dT₂₀ primer. The RT-PCR was performed using 2x SYBR green qPCR Kit (Thermo Fisher Scientific). The primers used in this study are listed in Supplementary Table S1. The PCR conditions were as follows: 95°C for 10 minutes followed by 40 cycles at 95°C for 15 seconds, 60°C for 1 minute, and 72°C for 30 seconds. Relative gene expression was calculated using the method of Livak and Schmittgen (24).

GraphPad Prism (version 8.0.0) was used to perform statistical analyses. Statistical significance of difference for two sample comparison was determined by unpaired Student t test. One-way ANOVA followed by Dunnett's test or Bonferroni's correction was used for multiple group comparisons. A P value of <0.05 was considered statistically significant.

Data would be made available upon written request to the corresponding author.

As summarized in Supplementary Fig. S1A, citrate is the precursor of fatty acid synthesis. Citrate is converted to acetyl-CoA through catalytic mediation of ACLY. Acetyl-CoA is further converted to malonyl-CoA by a reaction catalyzed by ACC1. One molecule of acetyl-CoA and seven molecules of malonyl-CoA are then used to synthesize saturated fatty acids (palmitic acid, myristic acid, and stearic acid) by the FASN complex. The CPT1A is a mitochondrial protein that is responsible for formation of acyl carnitines. This reaction is critical for mitochondrial uptake of fatty acids for β-oxidation. The RNA-seq data from our previously published study (20) revealed downregulation of ACLY, ACC1, and/or FASN in MCF-7 and/or MDA-MB-231 cells after WA treatment (Supplementary Fig. S1B). The RNA-seq data were validated by qRT-PCR for the mRNA expression of ACLY, ACC1, FASN, and CPT1A in MCF-7 and MDA-MB-231 cells after WA treatment (Supplementary Fig. S2).

In this study, we used WA concentrations of 1 or 2 μmol/L based on pharmacokinetic considerations. The maximal plasma concentration of WA in rats was about 4.9 μmol/L after intravenous administration of 5-mg WA/kg body weight (25). The plasma half-life in rats after intravenous treatment with 5-mg WA/kg body weight was about 4.5 hours (25). A lower steady-state concentration of WA was observed in our chemoprevention study in rats because the blood samples were collected only 1 hour after last injection (19). Figure 1A depicts confocal microscopic images of ACLY and ACC1 proteins in control and WA-treated MCF-7 and MDA-MB-231 cells. The expression of ACLY protein was decreased by 59.2% and 39.6% in MCF-7 and MDA-MB-231 cells, respectively upon treatment with 2 μmol/L WA (Fig. 1B). A respective decrease of 54.4% and 56% in protein level of ACC1 was observed in MCF-7 and MDA-MB-231 cells following treatment with 2 μmol/L WA (Fig. 1B). The effects of WA treatment on protein expression of FASN and CPT1A are shown in Fig. 2A. Expressions of both FASN and CPT1A proteins were also decreased significantly upon WA treatment in MCF-7 and MDA-MB-231 cells (Fig. 2B). These results indicated inhibitory effect of WA treatment on protein levels of fatty acid synthesis enzyme proteins.

The expression of ACLY was significantly higher in luminal-type, Her2⁺, and basal-like subtypes of breast cancer when compared with normal mammary gland (Supplementary Fig. S3). The expression of ACC1 was significantly lower in basal-like breast cancer in comparison with normal mammary gland (Supplementary Fig. S3). Only Her2⁺ mammary tumors exhibited upregulation of FASN when compared with normal mammary gland. Finally, the expression of CPT1A was significantly higher in Luminal B type breast cancer (Supplementary Fig. S3).

Initially, we conducted a study to determine expression of fatty acid synthesis pathway proteins in normal mammary glands and MNU-induced breast tumors of rats. The expression of ACLY, FASN, and CPT1A protein was very low or undetectable in normal mammary glands (Fig. 3A). On the other hand, expression of FASN and CPT1A proteins was higher by about 4.98 and 7.8-fold, respectively, in mammary tumors when compared with normal mammary glands (Fig. 3B). Total free fatty acid levels in mammary tumors and plasma were significantly higher compared with those of normal mammary glands (Fig. 3C). Similarly, the plasma and mammary tumors from MNU-treated rats had significantly elevated levels of total phospholipids in comparison with normal mammary tissues (Fig. 3D). These results indicated elevation of fatty acid synthesis in MNU-induced mammary tumors in rats when compared with normal mammary tissues.

Next, we determined the expression of fatty acid metabolism proteins using mammary tumors from control and WA-treated MNU-rats (Fig. 4A). A decrease in expression of each protein was observed in the WA treatment group compared with the control group but the difference was statistically significant only for ACC1 and FASN (Fig. 4B). The WA treatment resulted in a significant decrease in circulating levels of total and free fatty acids and total phospholipids (Fig. 4C). These results indicated in vivo inhibition of fatty acid synthesis by WA treatment.

Because lipids are known to influence tumor microenvironment and immune cells (26), we determined the effect of WA treatment on markers of T cells (Granzyme B and perforin 1) and fraction of CD3⁺CD4⁺ T cells and CD161⁺ natural killer (NK) cells. The level of Granzyme B was higher in both plasma and the mammary tumors of WA-treated rats in comparison with vehicle-treated control rats, but the difference did not reach significance due to large data scatter (Fig. 5A and B). Therefore, the CD8⁺ T cells were not quantitated. The fraction of CD3⁺CD4⁺ T cells did not differ between control and the WA treatments groups (Fig. 5C). However, the fraction of CD161⁺ NK cells was significantly higher in the spleen of WA-treated rats when compared with controls (Fig. 5C). These results indicated that inhibition of mammary tumor development by WA treatment was accompanied by an increase in fraction of CD161⁺ NK cells at least in the spleen. Mammary tumor infiltration of CD161⁺ NK cells was also modestly higher in the WA treatment group when compared with the control group but the difference was not significant

The SREBP1 is known to play an important role in regulation of fatty acid synthesis genes (27). The expression of precursor and mature SREBP1 was decreased upon treatment of MDA-MB-231 cells with WA (Fig. 6A). The level of SREBP1 protein was also lower in the mammary tumors of WA-treated rats than that of control rats but the difference was not significant (Fig. 6B). Overexpression of SREBP1 (Fig. 6C) conferred partial protection against WA-mediated downregulation of ACLY and ACC1 (Fig. 6D). On the other hand, SREBP1 overexpression has no meaningful impact on FASN expression.

Metabolic reprogramming, including increased glycolysis and fatty acid synthesis, is a hallmark of cancer (28). Although increased glycolysis provides energy, the fatty acids are used for β-oxidation in the mitochondria to not only produce energy but also in cell signaling (e.g., palmitylation and myristylation of oncogenic proteins) or incorporation into complex lipids for membrane integrity (29, 30). Studies have implicated fatty acid synthesis enzyme proteins in breast cancer progression (31–35). The expression of ACLY protein was higher in breast tumors in comparison with adjacent normal mammary tissue (31). Knockdown of ACLY protein in MCF-7 cells decreased cell viability due to apoptosis induction (31). The enzymatic activity of ACLY was higher by about 150-fold in breast cancer compared with normal mammary tissue (32). Another study showed physical interaction of ACLY with post-translationally cleaved low molecular weight cyclin E thereby promoting breast cancer growth (33). Both ACC1 and FASN proteins were expressed at higher level in breast carcinoma in situ than in histologically normal mammary tissue (34). The BRCA1 physically interacts with phosphorylated ACC1 to control lipid metabolism in breast cancer cells (35). Even though studies overwhelmingly support an oncogenic role of ACLY, ACC1, and FASN, their inhibitors are not available for cancer therapy (36). Prior efforts in clinical development of FASN inhibitors have been hampered due to a variety of reasons, including stability (Cerulenin), poor bioavailability (Orlistat), and side effects (appetite suppression and weight loss through direct activation of CPT1A; ref. 36). We have shown previously that inhibition of breast cancer development in MMTV-neu mice is accompanied by suppression of glycolysis (18). The present study is the first to demonstrate inhibition of fatty acid synthesis by WA treatment in vivo in a rodent model of breast cancer. Therefore, plasma and tumor levels of total free fatty acids or immunohistochemical analyses of ACLY, ACC1, and/or FASN protein expression may be integrated as biomarkers of WA efficacy in future clinical trials. In this context, a placebo-controlled (n = 13) clinical study with daily oral administration of one 400-mg Withania somnifera capsule (three times/d) for 1 month in patients with schizophrenia (n = 12) showed a significant decrease (P < 0.05) in circulating levels of triglycerides (37). Because triglycerides are a by-product of fatty acid synthesis, we are quite optimistic for suppression of circulating and possibly tumor levels of total free fatty acids by WA administration in breast cancer patients.

Progression of cancer, including mammary carcinoma, is regulated by cells of innate immunity (e.g., NK cells) and adaptive immunity (CD4⁺ and CD8⁺ T cells) in the tumor microenvironment (38). Tumor-infiltrating lymphocytes like CD8⁺ cytotoxic T cells serve as prognostic indicators for response to chemotherapy and for survival (38). On the other hand, T_reg and myeloid-derived suppressor cells cause immune evasion (39, 40). The level of myeloid-derived suppressor cells in the blood correlates with tumor burden, stage, and with poor prognosis in multiple cancer types (39). A meta-analysis showed a negative effect of FoxP3⁺ T_reg on overall survival (41). Previous studies have shown immunomodulatory effects of Withania somnifera as well as WA. The WA treatment was shown to inhibit mitogen induced T- and B-cell proliferation in vitro in association with upregulation of T-cell markers CD25, CD69, CD71, and CD54 and B-cell activation markers CD80, CD86, and MHC-II (42). Function of myeloid-derived suppressor cells was inhibited by WA (43). Dietary administration of Withania somnifera increased NK cell function in a laying hen model of spontaneous ovarian cancer (44). Our results also showed an increase in CD161⁺ NK cells in the spleen by WA administration to tumor bearing rats. The mammary tumor level of CD161⁺ NK cells was higher in the WA treatment group, but the difference was insignificant.

We found that overexpression of SREBP1 only partially attenuates WA-mediated downregulation of fatty acid synthesis genes. It is possible that additional mechanisms are involved in WA-mediated suppression of fatty acid genes. For example, we have shown previously that c-Myc is another regulator of fatty acid synthesis at least in prostate cancer (22). The c-Myc was recruited at the promoter of ACLY, ACC1, and FASN (22). It is possible that WA treatment suppresses c-Myc to inhibit fatty acid synthesis in breast cancer. The STAT3 was also shown to regulate fatty acid metabolism in breast cancer (45). However, further work is necessary to determine the role of c-Myc or STAT3 in WA-mediated downregulation of fatty acid synthesis enzyme proteins.

The key conclusions from the present study include: (i) fatty acid synthesis and expression of key enzymes of this pathway are elevated in mammary tumors resulting from MNU injection and this model may be useful for future screening of fatty acid synthesis inhibitors; (ii) WA treatment inhibits fatty acid synthesis in vivo by downregulating expression of fatty acid synthesis proteins; and (iii) inhibition of breast cancer development by WA administration in the MNU-rat model is accompanied by a statistically significant increase in CD161⁺ NK cells in the spleen.

No disclosures were reported.

The funders had no role in the design of the study, data collection, analysis or interpretation of the data, article preparation or decision to submit the article for publication.

K.B. Singh: Formal analysis, investigation, writing–original draft. E.-R. Hahm: Investigation, writing–review and editing. S.-H. Kim: Investigation, writing–review and editing. S.V. Singh: Conceptualization, resources, supervision, funding acquisition, writing–review and editing.

This study was supported in part by the National Cancer Institute grants R01 CA219180 and CA142604 (to S.V. Singh). This study used the UPMC Hillman Cancer Center Core Facilities, including the Animal Facility and Tissue and Research Pathology Facility.

The publication costs of this article were defrayed in part by the payment of publication fees. Therefore, and solely to indicate this fact, this article is hereby marked “advertisement” in accordance with 18 USC section 1734.