Low ketolytic enzyme levels in tumors predict ketogenic diet responses in cancer cell lines in vitro and in vivo

Abstract

The ketogenic diet (KD) is a high-fat, very-low-carbohydrate diet that triggers a fasting state by decreasing glucose and increasing ketone bodies, such as β-hydroxybutyrate (βHB). In experimental models and clinical trials, the KD has shown anti-tumor effects, possibly by reducing energy supplies to cells, which damage the tumor microenvironment and inhibit tumor growth. Here, we determined expression levels of genes encoding the ketolytic enzymes 3-hydroxybutyrate dehydrogenase 1 (BDH1) and succinyl-CoA: 3-oxoacid CoA transferase 1 (OXCT1) in 33 human cancer cell lines. We then selected two representative lines, HeLa and PANC-1, for in vivo examination of KD sensitivity in tumors with high or low expression, respectively, of these two enzymes. In mice with HeLa xenografts, the KD increased tumor growth and mouse survival decreased, possibly because these tumors actively consumed ketone bodies as an energy source. Conversely, the KD significantly inhibited growth of PANC-1 xenograft tumors. βHB added to each cell culture significantly increased proliferation of HeLa cells, but not PANCI-1 cells. Downregulation of both BDH1 and OXCT1 rendered HeLa cells sensitive to the KD in vitro and in vivo. Tumors with low ketolytic enzyme expression may be unable to metabolize ketone bodies, thus predicting a better response to KD therapy. The ketogenic diet (KD) is a high-fat, very-low-carbohydrate diet that triggers a fasting state by decreasing glucose and increasing ketone bodies, such as β-hydroxybutyrate (βHB). In experimental models and clinical trials, the KD has shown anti-tumor effects, possibly by reducing energy supplies to cells, which damage the tumor microenvironment and inhibit tumor growth. Here, we determined expression levels of genes encoding the ketolytic enzymes 3-hydroxybutyrate dehydrogenase 1 (BDH1) and succinyl-CoA: 3-oxoacid CoA transferase 1 (OXCT1) in 33 human cancer cell lines. We then selected two representative lines, HeLa and PANC-1, for in vivo examination of KD sensitivity in tumors with high or low expression, respectively, of these two enzymes. In mice with HeLa xenografts, the KD increased tumor growth and mouse survival decreased, possibly because these tumors actively consumed ketone bodies as an energy source. Conversely, the KD significantly inhibited growth of PANC-1 xenograft tumors. βHB added to each cell culture significantly increased proliferation of HeLa cells, but not PANCI-1 cells. Downregulation of both BDH1 and OXCT1 rendered HeLa cells sensitive to the KD in vitro and in vivo. Tumors with low ketolytic enzyme expression may be unable to metabolize ketone bodies, thus predicting a better response to KD therapy. ERRATUMJournal of Lipid ResearchVol. 59Issue 7PreviewRegarding “Low ketolytic enzyme levels in tumors predict ketogenic diet responses in cancer cell lines in vitro and in vivo” (J. Lipid Res. 59: 625–634): Full-Text PDF Open Access Cancer is one of the most serious health problems in the world. The wide use of traditional cancer therapies, such as surgery, chemotherapy, or radiotherapy, is limited by indications, contraindications, side effects, and other considerations. The discovery of the Warburg effect provided a new direction for cancer treatment. Since then, cancer metabolism and nutrition have gradually become a hot topic of theoretical and clinical research (1.Mikawa T. Leonart M.E.L. Takaori-Kondo A. Inagaki N. Yokode M. Kondoh H. Dysregulated glycolysis as an oncogenic event.Cell. Mol. Life Sci. 2015; 72: 1881-1892Crossref PubMed Scopus (59) Google Scholar). Cancer cells ferment glucose to lactate even in the presence of sufficient oxygen levels. This so-called aerobic glycolysis, a distinct characteristic of cancer cells, can be caused by gene mutations also found in mitochondrial disorders (2.Ju H-Q. Zhan G. Huang A. Sun Y. Wen S. Yang J. Lu W-H. Xu R-H. Li J. Li Y. et al.ITD mutation in FLT3 tyrosine kinase promotes Warburg effect and renders therapeutic sensitivity to glycolytic inhibition.Leukemia. 2017; 31: 2143-2150Crossref PubMed Scopus (59) Google Scholar, 3.Gururaja Rao S. Mitochondrial changes in cancer.Handb. Exp. Pharmacol. 2017; 240: 211-227Crossref PubMed Scopus (14) Google Scholar). Thus, reducing carbohydrate supplementation may selectively cut off the energy supply of tumor cells, damage the tumor microenvironment, and thereby inhibit tumor growth (4.Simone B.A. Champ C.E. Rosenberg A.L. Berger A.C. Monti D.A. Dicker A.P. Simone N.L. Selectively starving cancer cells through dietary manipulation: methods and clinical implications.Future Oncol. 2013; 9: 959-976Crossref PubMed Scopus (43) Google Scholar). The ketogenic diet (KD) is a high-fat, very-low-carbohydrate, and adequate protein diet that simulates a fasting metabolic state by decreasing glucose level while simultaneously elevating the levels of ketone bodies, β-hydroxybutyrate (βHB), and acetoacetate (AcAc). The initial clinical application of KD was to treat epilepsy (5.Wilder R.M. The effects of ketonemia on course of epilepsy. 1921; 2: 307-308Google Scholar). Numerous previous studies have indicated that this dietary restriction also reduces tumor volume and growth rate (6.Morscher R.J. Aminzadeh-Gohari S. Feichtinger R.G. Mayr J.A. Lang R. Neureiter D. Sperl W. Kofler B. Inhibition of neuroblastoma tumor growth by ketogenic diet and/or calorie restriction in a CD1-Nu mouse model.PLoS One. 2015; 10: e0129802Crossref PubMed Scopus (65) Google Scholar), improves patients' quality of life (7.Shukla S.K. Gebregiworgis T. Purohit V. Chaika N.V. Gunda V. Radhakrishnan P. Mehla K. Pipinos I.I. Powers R. Yu F. et al.Metabolic reprogramming induced by ketone bodies diminishes pancreatic cancer cachexia.Cancer Metab. 2014; 2: 18-36Crossref PubMed Google Scholar), prolongs survival (8.Poff A.M. Ari C. Seyfried T.N. D'Agostino D.P. The ketogenic diet and hyperbaric oxygen therapy prolong survival in mice with systemic metastatic cancer.PLoS One. 2013; 8: e65522Crossref PubMed Scopus (118) Google Scholar), and enhances the sensitivity to radiotherapy and chemotherapy (9.Allen B.G. Bhatia S.K. Buatti J.M. Brandt K.E. Lindholm K.E. Button A.M. Szweda L.I. Smith B.J. Spitz D.R. Fath M.A. Ketogenic diets enhance oxidative stress and radio-chemo-therapy responses in lung cancer xenografts.Clin. Cancer Res. 2013; 19: 3905-3913Crossref PubMed Scopus (144) Google Scholar, 10.Klement R.J. Sweeney R.A. Impact of a ketogenic diet intervention during radiotherapy on body composition: I. Initial clinical experience with six prospectively studied patients.BMC Res. Notes. 2016; 9: 143-155Crossref PubMed Scopus (61) Google Scholar). In recent years, with more data becoming available on KD therapy and metabolic phenotypes of cancer cells, this diet has become popular among researchers and cancer patients due to its efficacy and potential anti-tumor effects. KD effects have also been shown to increase oxidative stress, suppress inflammation, enhance immune response, and affect various signaling pathways (11.Allen B.G. Bhatia S.K. Anderson C.M. Eichenberger-Gilmore J.M. Sibenaller Z.A. Mapuskar K.A. Schoenfeld J.D. Buatti J.M. Spitz D.R. Fath M.A. Ketogenic diets as an adjuvant cancer therapy: history and potential mechanism.Redox Biol. 2014; 2: 963-970Crossref PubMed Scopus (172) Google Scholar, 12.Boison D. New insights into the mechanisms of the ketogenic diet.Curr. Opin. Neurol. 2017; 30: 187-192Crossref PubMed Scopus (145) Google Scholar, 13.Newman J.C. Verdin E. B-hydroxybutyrate: much more than a metabolite.Diabetes Res. Clin. Pract. 2014; 106: 173-181Abstract Full Text Full Text PDF PubMed Scopus (193) Google Scholar, 14.Vidali S. Aminzadeh S. Lambert B. Rutherford T. Sperl W. Kofler B. Feichtinger R.G. Mitochondria: the ketogenic diet - a metabolism-based therapy.Int. J. Biochem. Cell Biol. 2015; 63: 55-59Crossref PubMed Scopus (109) Google Scholar). However, in some studies, very different results have been reported when KD was examined as adjuvant therapy in mouse models (15.Maurer G.D. Brucker D.P. Bahr O. Harter P.N. Hattingen E. Walenta S. Mueller-Klieser W. Steinbach J.P. Rieger J. Differential utilization of ketone bodies by neurons and glioma cell lines: a rationale for ketogenic diet as experimental glioma therapy.BMC Cancer. 2011; 11: 315-331Crossref PubMed Scopus (145) Google Scholar, 16.Mavropoulos J.C. Buschemeyer 3rd, W.C. Tewari A.K. Rokhfeld D. Pollak M. Zhao Y. Febbo P.G. Cohen P. Hwang D. Devi G. et al.The effects of varying dietary carbohydrate and fat content on survival in a murine LNCaP prostate cancer xenograft model.Cancer Prev. Res. (Phila.). 2009; 2: 557-565Crossref PubMed Scopus (77) Google Scholar, 17.Wheatley K.E. Williams E.A. Smith N.C.P. Dillard A. Park E.Y. Nunez N.P. Hursting S.D. Lane M.A. Low-carbohydrate diet versus caloric restriction: effects on weight loss, hormones, and colon tumor growth in obese mice.Nutr. Cancer. 2008; 60: 61-68Crossref PubMed Scopus (29) Google Scholar). In view of the completely opposite KD effects on cancer in different studies, the factors that influence the effectiveness of KD therapy need to be properly established. It is well known that ketone bodies are an important energy source during a KD. Whether cancer cells can utilize ketone bodies effectively is critically important to cell proliferation. Utilization of ketone bodies as energy source requires the presence of four enzymes: 3-hydroxybutyrate dehydrogenase 1/2 (BDH1/2), and succinyl-CoA:3-oxoacid CoA transferase 1/2 (OXCT1/2). Thus, we hypothesized that tumors with very low expression of these enzymes may have a better response to KD therapy. We investigated BDH1/2 and OXCT1/2 expression in several cancer cell lines and tested this hypothesis with two representative cell lines that had low and high expression levels of these key ketone body catabolism enzymes in vivo and in vitro. All cell lines used in this study were purchased from Cell Bank, Chinese Academy of Sciences (Shanghai, China). Cells were cultured in DMEM or RPMI-1640 Medium with 10% FBS, 1% penicillin/streptomycin (Gibco, Grand Island, NY) and washed with PBS (Hyclone, Los Angeles, CA) every 48 h. All cells were cultured at 37°C in the atmosphere of 95% air and 5% carbon dioxide. For proliferation experiments, 5 × 104 cells were planted per well in 6-well plates in DMEM or in low glucose (LG; 3 mM) medium with or without supplementation of 5 or 10 mM DL-β-hydroxybutyric acid sodium salt (Sigma-Aldrich, St Louis, MO). The medium with no βHB contained an equal volume of deionized water. βHB at the indicated concentration was added into the medium 5 h after first-day plating. Cell numbers were recorded at 12, 24, 48, 72, and 96 h by cell counting or using a Handheld Automated Cell Counter (Millipore, Billerica, MA). All wells were visually inspected to make sure that all cells were collected before counting. BDH1 and OXCT1 siRNA target sequences and a scrambled control sequence were designed and cloned into iLenti siRNA vectors (ABM, Zhenjiang, China) that carried green fluorescent protein (GFP) and puromycin resistance genes by using convergent promoters U6 and H1. The RNA interference target sequences are shown in Table 1 (four mixed target sequences for each). For lentivirus infection assay, cells were seeded on 6-well plates at a density of 2 × 105/well. The next day, the cells were infected with lentivirus at a multiplicity of infection value of 10. GFP fluorescence signal was examined 72 h after the infection to ensure infection efficiency. In addition, 72 h after the infection, cells infected with the lentivirus were selected using 2.5 mg/ml of puromycin. quantative RT-PCR and Western blot analysiswere used to explore interferential efficiency.TABLE 1The RNA interference target Open in a new RNA was using CA) was the was by was using an The in Table 2 were by (Shanghai, China). All genes were in the as was by the fluorescence was determined at were as at for by for 5 at and at were in a volume of All were in expression levels were by using the or Open in a new cells were washed with PBS and in 1 mM mM sodium and was by at at were with OXCT1 of and and were with and using tumor were in 10% in and cut into by using For all were to the The BDH1 and OXCT1 were at and was carried using and was used as a the were examined using a All on was by the and at the For tumor mice China) were in a a of 2 × HeLa cells in PBS or of 1 × PANC-1 cells in PBS was into the of the growth was recorded every 3 from 2 after by two using the × × the mice were into two a diet and a ketogenic diet (KD) KD was at a of which was on Thus, the calorie of the was equal to the and the of the diets are shown in Table volume was for using body weight was during the was from and glucose and βHB levels were using the and We also other as to explore The for the mice was as every mouse from systemic of or that was a of or be and were used for The and were to the in mouse in tumor body and gene expression levels in xenograft tumors were by the in and KD Cell proliferation was by by the when All were with a level of levels of BDH1/2 and OXCT1/2 in 33 cancer cell lines were determined by shown in these key ketolytic enzymes were in all cell lines, but expression levels significantly in different lines. However, BDH1 and OXCT1 may a more than and For experiments, we PANC-1 and HeLa cells with low and high expression levels of both BDH1 and OXCT1 We also the expression of the BDH1 and OXCT1 in PANC-1 and HeLa cells by Western The protein levels were with expression levels this we hypothesized that with high glycolysis in cancer cells may be by a level of ketolytic key enzymes. Thus, we examined glucose 1 glucose 3 However, no and ketone body metabolism was we found that the proliferation of PANC-1 and HeLa cells was significantly during in medium for 96 h with that in high glucose mM) medium In the presence of 5 mM or 10 mM βHB in a rate of proliferation of PANC-1 cells was not at In proliferation of HeLa cells cultured for 72 h in medium was significantly than that of cells cultured in medium and for 5 mM and 10 mM we also added 10 mM βHB into the control which had no influence on the proliferation of HeLa cells with control medium In the next of experiments, by using RNA we BDH1 and OXCT1 or simultaneously in HeLa cells. The infection was at a multiplicity of infection value of 10 as more than of HeLa cells GFP signal after lentivirus infection and by puromycin was by and Western BDH1 and OXCT1 expression levels were to of in control cells BDH1 and OXCT1 protein levels were also significantly that BDH1 and OXCT1 expression levels in HeLa cells were by lentivirus a of these experiments, we scrambled and HeLa cells. with the four of HeLa cells were cultured and to cell proliferation shown in the proliferation of HeLa cells in different with or without βHB was to that of HeLa cells that were not infected with lentivirus shown in and proliferation of and HeLa cells was increased in the presence of 5 mM or 10 mM βHB in medium proliferation of HeLa cells was not by βHB supplementation and to that in medium without βHB We next examined the use of KD in vivo to results with the in vitro data We found that KD was with significantly growth and weight of PANC-1 xenograft tumors in to the in the and KD significantly the survival of mice with PANC-1 xenograft tumors In KD was with tumor growth rate and tumor weight in mice with HeLa xenograft tumors but the effect not and the survival of mice with HeLa xenograft tumors that a KD was significantly than that of mice in In addition, we mouse body weight We found that the body weight of mice was significantly than that of mice during the 2 However, after 2 this the two was not KD also glucose and βHB significantly we BDH1 and OXCT1 expression levels in tumor by and shown in and was no increased expression of BDH1 and OXCT1 and protein in PANC-1 xenograft tumors in the KD with levels in mice from the However, in mice with HeLa cell xenograft KD was with significantly BDH1 and OXCT1 expression levels as by and and also a in for these the two We also examined lung cancer cells with low expression of BDH1 and KD not inhibit growth of tumors from cell in We the expression of BDH1 and OXCT1 in tumor and found that these two genes were significantly we the expression levels of BDH1 and OXCT1 in PANC-1, and HeLa cell in the KD BDH1 and OXCT1 expression levels in cell were than the in HeLa cells but significantly than the in PANC-1 We also that KD tumor growth in mice with from HeLa cells with BDH1 and The in tumor volume mice with HeLa cell in the KD and at the was However, KD not significantly inhibit tumor growth of from or HeLa cells and survival were all to in mice with HeLa cell as not We also expression levels of BDH1 and OXCT1 and protein and the effect of RNA interference in the In the we investigated the the expression of the key ketolytic enzymes BDH1 and OXCT1 in tumors and response to KD. We that xenograft tumors that from cancer cells with low expression levels of BDH1 and OXCT1 are more to KD because to metabolize ketone The Warburg effect is one of the most and metabolic phenotypes by most cancer cells A. M.A. R. of metabolic intervention in the of Oncol. 2017; PubMed Scopus Google Scholar). metabolic of tumors and using therapeutic may of the cancer metabolic caused by gene mutations or mitochondrial can cancer cells from effectively ketone bodies as a source of energy T.N. Cancer as a mitochondrial metabolic Cell Biol. 2015; PubMed Scopus Google Scholar). a tumor is unable to utilize ketone bodies, the use of KD may be an therapy for of such tumors. However, of this the of the to cancer with KD therapy. In medium with βHB not PANC-1 cell growth in vitro proliferation of HeLa cells was significantly increased by βHB results were in xenograft in mice HeLa cells not ketone bodies as to proliferation when BDH1 and the ketolytic were simultaneously However, previous studies have shown ketone bodies and may be much more than important signaling as well J.C. Verdin E. bodies as signaling Metab. 2014; Full Text Full Text PDF PubMed Scopus Google Scholar, S. R. Zhao Y. S. G. E. et of tumor Metab. 2017; Full Text Full Text PDF PubMed Scopus Google Scholar). It was βHB the proliferation of HeLa cells through some signaling pathways of its in cell of control medium with 10 mM βHB not affect proliferation of HeLa cells. This that the of βHB in HeLa cells is to as an energy source. results that cancer cells with very low expression levels of BDH1 and OXCT1 may be sensitive to KD therapy. In to this in a recent clinical et K.A. and glycolytic expression in for ketogenic diet Metab. 2013; 10: PubMed Scopus Google found expression of ketolytic enzymes BDH1 and in hypothesized that patients with low or very low expression of BDH1 and OXCT1 in may better to KD therapy. in mitochondrial and βHB are to and P. of ketone bodies in and Metab. 2017; Full Text Full Text PDF PubMed Scopus Google Scholar). The is to the mitochondrial Thus, in an BDH1 of βHB into and then is into in a by can be also into In or HeLa cells had limited proliferation in medium with which that the to βHB as energy source. However, KD not inhibit growth of xenograft tumors from or HeLa cells in to a we have not found a for the We that these to some of ketone body We also determined the levels of and that a in ketone body and levels in most cell lines were than of BDH1 and OXCT1 studies have shown that a in the of the and is a in cells by an enzyme with a in Full Text Full Text PDF PubMed Scopus Google Scholar, A. acid promotes and in Biol. 2014; PubMed Scopus Google Scholar). Thus, during catabolism of the ketone BDH1 and OXCT1 may a more than the and Cancer cells gene expression or metabolic pathways to proliferation T. R.A. The Warburg effect as an of cancer cells to in energy One. 2017; PubMed Scopus Google Scholar, J. P. The tumor systemic metabolic 2017; PubMed Scopus Google Scholar). In we determined BDH1 and OXCT1 expression levels in tumors. we not increased expression of BDH1 and OXCT1 in tumors from PANC-1 cell The of expression of these two key ketone catabolism enzymes that was not in xenograft tumors. In levels of BDH1 and OXCT1 in HeLa cell such as of key ketone catabolism may be one of the for of KD therapy on lung cancer cell metabolic can be also in cells that ketone bodies for energy supply and cancer D. Li T. R. Li K. S. G. W. et to for nutrition Res. 2016; PubMed Scopus Google Scholar). the potential by which gene expression is may enhance anti-tumor effects of KD therapy. In this we also that KD and mouse body mice had significantly ketone body a which was also in other studies using a KD in tumor mice R.J. Champ C.E. C. effects of ketogenic diets in a One. 2016; 11: PubMed Scopus Google Scholar). It may the efficacy of KD therapy mice of body weight with during the the mouse body weight in the KD was than that in the at every the during the was not results were with the data of several previous studies T.N. R. P. of glucose and ketone bodies in the metabolic control of experimental J. Cancer. PubMed Scopus Google Scholar, A.M. Ari C. P. Seyfried T.N. D'Agostino D.P. supplementation tumor cell and prolongs survival of mice with metastatic J. Cancer. 2014; PubMed Scopus Google Scholar). weight loss, KD was or with high energy content in most studies C. B. B. N. R. A. of human cancer cells in mice is by a ketogenic diet with and Cancer. 2008; 8: PubMed Scopus Google Scholar). However, different nutrition may significantly influence the efficacy of KD. In this we KD with the of In or KD may have an even better effect than a However, not the fat well M. Y. S. J. C. H. D. H. and of the ketogenic diet to - of with PubMed Scopus Google Scholar). In addition, the KD even increased tumor growth and the survival mice with HeLa cells. This that HeLa xenograft tumors actively ketone bodies as an energy source which to growth of tumor and energy in results that expression of the key ketolytic enzymes BDH1 and OXCT1 is one of the factors that the effect of a KD on Tumors with low expression of BDH1 and OXCT1 may better to KD therapy than with high expression of these BDH1 and OXCT1 may be very to the sensitivity of a cancer to a KD. The use of BDH1 and OXCT1 may enhance KD and the anti-tumor of KD therapy. studies are to other tumors from cancer cells with different expression levels of BDH1 and OXCT1 sensitivity to KD. the by which KD therapy BDH1 and OXCT1 expression be is to other sensitive can also affect KD with acetoacetate 3-hydroxybutyrate dehydrogenase 1 green fluorescent protein glucose 1 glucose 3 ketogenic diet low glucose succinyl-CoA:3-oxoacid CoA transferase 1 diet