Low levels of pyruvate induced by a positive feedback loop protects cholangiocarcinoma cells from apoptosis
- 272 Downloads
Cancer cells avidly consume glucose and convert it to lactate, resulting in a low pyruvate level. This phenomenon is known as the Warburg effect, and is important for cell proliferation. Although cMyc has often been described as an oncoprotein that preferentially contributes to the Warburg effect and tumor proliferation, mechanisms of action remain unclear. Histone deacetylase 3 (HDAC3) regulates gene expression by removing acetyl groups from lysine residues, as well as has an oncogenic role in apoptosis and contributes to the proliferation of many cancer cells including cholangiocarcinoma (CCA). HDAC inhibitors display antitumor activity in many cancer cell lines. Cancer cells maintain low levels of pyruvate to prevent inhibition of HDAC but the mechanisms remain elusive. The purpose of our study was to explore the role of cMyc in regulating pyruvate metabolism, as well as to investigate whether the inhibitory effect of pyruvate on HDAC3 could hold promise in the treatment of cancer cells.
We studied pyruvate levels in CCA cell lines using metabolite analysis, and analyzed the relationship of pyruvate levels and cell proliferation with cell viability analysis. We cultivated CCA cell lines with high or low levels of pyruvate, and then analyzed the protein levels of HDAC3 and apoptotic markers via Western Blotting. We then explored the reasons of low levels of pyruvate by using seahorse analysis and 13C6 metabolites tracing analysis, and then confirmed the results using patient tissue protein samples through Western Blotting. Bioinformatics analysis and transfection assay were used to confirm the upstream target of the low levels of pyruvate status in CCA. The regulation of cMyc by HDAC3 was studied through immunoprecipitation and Western Blotting.
We confirmed downregulated pyruvate levels in CCA, and defined that high pyruvate levels correlated with reduced cell proliferation levels. Downregulated pyruvate levels decreased the inhibition to HDAC3 and consequently protected CCA cells from apoptosis. Synergistically upregulated LDHA, PKM2 levels resulted in low levels of pyruvate, as well as poor patient survival. We also found that low levels of pyruvate contributed to proliferation of CCA cells and confirmed that the upstream target is cMyc. Conversely, high activity of HDAC3 stabilized cMyc protein by preferential deacetylating cMyc at K323 site, which further contributed to the low pyruvate levels. Finally, this creates a positive feedback loop that maintained the low levels of pyruvate and promoted CCA proliferation.
Collectively, our findings identify a role for promoting the low pyruvate levels regulated by c-Myc, and its dynamic acetylation in cancer cell proliferation. These targets, as markers for predicting tumor proliferation in patients undergoing clinical treatments, could pave the way towards personalized therapies.
KeywordscMyc Pyruvate HDAC3 Apoptosis Cholangiocarcinoma
polyADP ribose polymerase
Muscle specific Pyruvate Kinase 2
We found cMyc decreases pyruvate levels by promoting LDHA and PKM2 levels, this can consequently decrease the inhibition to HDAC3 and protect cancer cells from apoptosis. Conversely, high activity of HDAC3 stabilizes the cMyc protein by preferentially deacetylating cMyc at K323 site, which further contributes to low pyruvate levels. This creates a positive feedback loop that promotes the Warburg effect and cell proliferation of the tumor.
Together, this suggest the low pyruvate levels, regulated by c-Myc and its dynamic acetylation, can serve as a marker for predicting tumor proliferation in patients undergoing clinical treatments. These potential targets could pave the way towards personalized therapies.
Cancer cells mainly rely on aerobic glycolysis to generate enough energy and intermediates for the malignant behaviors. This so-called Warburg Effect can convert the glycolysis-induced pyruvate into lactate, thus making low pyruvate status in cancer cells . The low pyruvate levels in cells are due to both reduced production and excessive consumption. The reduced pyruvate production comes from the expression of pyruvate kinase, the enzyme responsible for the generation of pyruvate in glycolysis. The splice variant PKM2 (musclespecific pyruvate kinase 2) is expressed specifically in cancer cells in the dimeric form with low catalytic activity, and is predictive of a poor prognosis in CCA patients . The dominant pyruvate consumption comes from the conversion of pyruvate into lactate, and this reaction is mediated by lactate dehydrogenase (LDH). There are different isoforms of tetrameric LDH: LDHA and LDHB, and LDHA is effective in the conversion of pyruvate into lactate. Since tumor cells robustly convert pyruvate into lactate, one would expect a high expression of LDHA in cancer cells as well as cholangiocarcinoma (CCA) .
As an oncogene, c-Myc has attracted extensive interest as its potential role for contributing to tumorigenesis. MYC, and c-Myc in particular, is one such oncogene. MYC was discovered in studies of fulminant chicken tumors caused by oncogenic retroviruses. Subsequently, genomic sequencing efforts identified c-Myc as one of the most highly amplified oncogenes in many different human cancers [4, 5]. There are various mechanism of MYC-induced tumorigenesis, including increased Warburg effect, and many studies have found that MYC increased metabolic proteins, such as LDH and PKM2 [6, 7]. Therefore, many studies focus on the therapeutic value of targeting Myc. So far, no small molecules can directly target c-Myc in vivo. Both suppressing c-Myc transcription by bromodomain inhibitors targeting BRD4 and destabilizing c-Myc protein level by SIRT2 inhibition significantly reduced cancer cell proliferation [5, 8]. As the stability of c-Myc contributed to tumorigenesis, additional studies have found that the stability of c-Myc protein is related to the low acetylation at K323 [9, 10]. The treatment of HDAC inhibitors (HDACi), but not SIRT inhibitors, induced c-Myc K323 acetylation as well as tumorigenesis inhibition, suggesting that at least one of HDACs is the deacetylase of c-Myc [11, 12]. Although cMyc have often been described as preferentially an oncoprotein that contributes to the Warburg effect and tumor proliferation, mechanisms of action still remain unclear.
Genetic or epigenetic alterations, which disrupt proliferation and cell death pathways, are the fundamental event for initiation and progression of cancer . Imbalanced epigenetic networks have been identified in all types of cancers and involve multiple metabolic changes. Unlike genetic mutations, epigenetic modifications are potentially reversible, thereby allowing drugs acting on specific enzymes involved in the epigenetic regulation of gene expression and raising the possibility of epigenetic therapies [12, 14]. Among the epigenetic alterations, acetylation has emerged as a key post-translational modification and acetylases have been identified as key metabolic enzymes in cellular regulation . Deacetylases are designated to four classes (I-IV), depending on their amino acid sequence structure . Class I HDACs (1, 2, 3 and 8) play an important role in tumorigenesis and may be candidate targets for many cancer treatments [16, 17]. Recently, we confirmed high levels of HDAC3 expressed in CCA tissues, and found that this was associated with poor survival in CCA patients. Mechanistically, HDAC3 induced proliferation and protected CCA cells from apoptosis . HDACs inhibitors have been noted for their ability to induce cell cycle arrest and apoptosis of a broad spectrum of cancer cells [13, 16, 17, 18, 19]. Some agents have shown signs of efficacy in clinical trials. SAHA and romidepsin are U.S. Food and Drug Administration (FDA) approved for the treatment of cutaneous T-cell lymphoma . Novel Class I HDACs inhibitors were found to be beneficial for tumorigenesis inhibition as well as the acetylation-induced cMyc degradation [18, 20, 21, 22].
Collectively, the mechanisms that underlie cMyc’s ability to promote tumorigenesis via the Warburg effect are poorly understood. Thus, the question of whether cMyc acetylation is beneficial for its induced-tumorigenesis still remains unanswered. Here we set out to identify whether low pyruvate levels regulated by c-Myc, and its dynamic acetylation, promote cancer cell proliferation. These targets, as markers of predicting tumor proliferation in patients undergoing clinical treatments, could pave the way towards therapeutic intervention in the treatment of CCA.
Materials and methods
Ethics, consent and permissions
All experiments utilizing animal and human samples were approved by the Ethical Committee of Medical Research, Nanjing Drum Tower Hospital, Affiliated Hospital of Nanjing University Medical School.
Cell culture and reagents
One human intrahepatic biliary epithelial cell line HIBEpiC and six human cholangiocarcinoma (CCA) cell lines HuCCT1, OZ, HuH28, Hccc9810, RBE, and QBC939 were used. HIBEpiC were obtained from the ScienCell (ScienCell, CA, Carlsbad, USA). HuCCT1, OZ, HuH28, and Hccc9810 cells were obtained from the Japanese Collection of Research Bioresources (JCRB) (Tokyo, Japan). RBE cells were obtained from the Institute of Biochemistry and Cell Biology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences (Shanghai, China). QBC939 cells were kindly provided by Professor Shuguang Wang from The Third Military Medical University (Chongqing, China). Cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) (Invitrogen, Carlsbad, CA, USA) containing 10% fetal bovine serum (Invitrogen), penicillin (Invitrogen) (100 U/ml) and streptomycin (Invitrogen) (100 U/ml). For apoptosis experiments, a moderate concentration of 1 mM ethyl pyruvate is present in order to maintain a proper basal apoptotic rate. Full-length HDAC1–3, PKM2, LDHA (wildtype) and cMYC (wildtype and K323R) plasmids were kindly provided by the Zhao lab of Fudan University (Shanghai, China). Cells were transfected with Lipofectamine 3000 (Invitrogen) according to the manufacturer’s protocol. The HDAC3 siRNA was commercially purchased from RiboBio (Guangzhou, China), siRNA-HDAC3–1: CCATGACAATGACAAGGAA, siRNA-HDAC3–2: GCATTGATGACCAGAGTTA, siRNA-HDAC3–3: GAATATGTCAAGAGCTTCA. HDAC3 shRNA (h) lentiviral particles were commercially purchased from Santa Cruz Biotechnology. RGFP966 (MCE, Monmouth Junction, NJ, USA) was commercially purchased.
Cell viability assay
Cell viability was determined using the CCK-8 colorimetric assay in 96-well plates (2 × 103 cells/well) (Dojindo, Minato-ku, Tokyo, Japan) and cultured at 37 °C with 5% CO2. After treatment at the indicated times, 10 ul of CCK-8 solutions was added to each well. Then, cells were incubated for one and a half hours. The absorbance of the samples at 450 nm was recorded using a scanning multi-well spectrophotometer. Relative cell viability (%) = (absorbance 450 nm of treated group - absorbance 450 nm of blank)/(absorbance 450 nm of control group − absorbance450 nm of blank) × 100.
Cellular pyruvate levels assay
Cellular pyruvate levels were detected by using Pyruvate Colorimetric/Fluorometric Assay Kit (K609–100, Biovision, Milpitas, California, USA) following the manufacturer’s instructions.
Cells were lysed with 0.5% NP40 lysis buffer and proteins were blotted following standard protocol; except for the detection of acetylation, which used 50 mM Tris (pH 7.5) with 10% (v/v) Tween 20 and 1% peptone (AMRESCO, Solon, OH, USA) as a blocking buffer. Primary and secondary antibodies were diluted in 50 mM Tris (pH 7.5) with 0.1% peptone. Signals were probed using the chemiluminescence ECL plus reagent (Thermo, Grand Island, NY, USA) and detected using a Typhoon FLA9500 scanner (GE, Fairfield, CT, USA). Primary antibodies were as follows: HDAC1 (Abcam, Cambridge, UK), HDAC2 (Abcam), HDAC3 (Abcam), cleaved caspase-3 (CST, Danvers, MA, USA), cleaved PARP (CST), PARP (CST), PKM2 (Abcam), LDHA (Abcam), cMYC (Abcam), β-actin (Sigma), FLAG (Abcam), and HA (provided by the Zhao lab of Fudan University).
Cells were lysed in NP-40 buffer containing 50 mM Tris-HCl (pH 7.5) (Sigma, St Louis, MO, USA), 150 mM NaCl (Sangon, Shanghai, China), 0.5% Nonidet P-40 (Sigma), 1 μg/ml aprotinin (Sigma), 1 μg/ml leupeptin (Sigma), 1 μg/ml pepstatin (Sigma), 1 mM Na3VO4 (Sigma) and 1 mM PMSF (Sigma). For immunoprecipitation, 500 μl of cell lysate was incubated with HA antibody (provided by the Zhao lab of Fudan University) for 3 h at 4 °C with rotation. Then, 30 μl of Protein A Agarose (Millipore, Billerica, MA, USA) was added for 12 h at 4 °C with rotation, and the beads were washed three times with lysis buffer before proteins were dissolved in loading buffer. Deacetylation assays were carried out in the presence of 5 μg enzyme and 0.3 μg peptide in 30 μl reaction buffer [30 mM HEPES (Sigma), 0.6 mM MgCl2 (Sangon), 1 mM DTT (Sigma), 1 mM NAD+ (Sigma), 10 mM PMSF (Sigma)]. The deacetylation reaction was incubated for 2 h at 37 °C before the mixture was desalted by passing it through a C18 ZipTip (Millipore). The desalted samples were analyzed using a MALDI-TOF/TOF mass spectrometer (Applied Biosystems, Grand Island, NY, USA). The acetylated peptide used in the assay was TRKDYPAAK (Ac) RVKLDSVR (Glssale, Shanghai, China).
Mitochondrial oxidative phosphorylation analysis
Oxygen consumption rate (OCR) was detected in real-time with the XF96 Extracellular Flux Analyzer from Seahorse Bioscience, Inc. (North Billerica, MA, USA) following the manufacturer’s instructions.
13C metabolism labeling experiments have previously been documented . Metabolite extracts were collected and 2ul of metabolite extract samples were injected for the GC-MS analysis using an Agilent 6980 GC coupled to an Agilent 5973 MS system. Relative metabolite abundances were determined by normalizing abundances of each metabolite to the internal standard and to cell number.
DEGs of paired-CCA from TCGA data
The CCA RNA-Seq data were downloaded from the TCGA database using The GDC Data Portal (https://portal.gdc.cancer.gov). The mRNA expression data included a total of 18 samples consisting of 9 normal sample and 9 paired-CCA samples. The sequencing data were all publicly available and no ethical issues were involved. The edgeR package in Bioconductor was used to screen the DEGs in CCA and normal liver tissue samples. The edgeR package is based on the negative binomial (NB) distribution, which can correct the overdispersion problem in RNA-seq data by using a Poisson model and a Bayes procedure. Data with expression values of zero were removed. The genes were deemed to be DEGs if |FoldChange| > 2, respectively, both with p-value < 0.01 and false discovery rate (FDR) < 0.05.
The Database for Annotation Visualization and Integrated Discovery (DAVID) online tool (https://david.ncifcrf.gov/) was used to conduct the functional and pathway enrichment analyses in our study. We performed GO and KEGG pathway enrichment analyses to detect the potential biological functions and pathways of the High and low expression genes in CCA.
Tumor specimens from human and mice were fixed in 4% formalin and embedded in paraffin. Two to five human CCA tumor specimens from one patient were used for the IHC study. Standard procedures were followed for IHC except for the detection of targets. The staining intensity for each tissue was calculated by multiplication and the range of this calculation .
Thirty-six hours following transfection, cells were lysed in 1% SDS buffer (Tris (Sigma), 0.5 mM EDTA (Sigma) and 1 mM DTT (Sigma), pH = 7.5), as well as boiled for 10 min. For immunoprecipitation, the lysates were diluted 10-fold in Tris-HCl buffer. Analyses of ubiquitination were performed using anti-HA blotting.
CCA xenograft model
Nude mice were purchased from the Department of Laboratory Animal Science, Nanjing Drum Tower Hospital. HuCCT1 cells (5 × 106) in FBS-free medium were subcutaneously injected into the abdomen of mice. The Animal Welfare Committee of Nanjing Drum Tower Hospital approved all procedures involving animals.
Data was expressed as means ± standard error of the mean (SE). The data was analyzed through one-way ANOVAs followed by post hoc Duncan tests (SPSS 17.0). P < 0.05 was considered significant.
High levels of intracellular pyruvate inhibit the proliferation of CCA cells
Pyruvate decreases the proliferation of CCA cells by inhibiting HDAC3
Changes in metabolic enzymes contribute to low levels of pyruvate
cMYC leads to a change in metabolic enzymes in CCA cells
cMYC-induced metabolic enzyme changes contribute to a poor prognosis in CCA patients
HDAC3 deacetylates cMYC at K323 and protects cMYC from ubiquitinated degradation
HDAC3 inhibition induces CCA cell apoptosis by decreasing cMYC
The Warburg effect represents a serious worldwide problem threatening the health of millions of cancer patients. However, how metabolites of the Warburg effect are beneficial for its induced-tumorigenesis, and what is the downstream target of these metabolites, still remains an unanswered question. Here, we set out to identify a promoting role for the low pyruvate levels as regulated by c-Myc and its dynamic acetylation in cancer cell proliferation. Low pyruvate levels contributed via downstream targets (PKM2 and LDHA) of c-Myc and attenuated the inhibition of HDAC3 as well as decreased HDAC3-regulated apoptosis. On the contrary, a high activity of HDAC3 stabilizes the cMyc protein by preferential deacetylation of cMyc at the K323 site, which further contributes to low pyruvate levels. This creates a positive feedback loop that promotes the Warburg effect and cell proliferation of the tumor.
It is well known that the dominant metabolite of the Warburg effect is lactate, and many previous studies focus on lactate and its role in tumor proliferation . Since lactate is converted from pyruvate and the end product of enhanced glycolysis is pyruvate in cancer cells, we inquired whether this decreased pyruvate benefitted tumor proliferation. The underlying notion here is that decreased generation of pyruvate in CCA is the direct result of increased glycolysis and protects the tumor from apoptosis. We demonstrated an inhibitory effect of high levels of pyruvate on CCA cells, and it is noteworthy that intracellular pyruvate levels were negatively correlated with cell viability. This may be due to the fact that the expression of SLC5A8 (the gene coding for the Na + −coupled pyruvate transporter that regulates the entry of blood-borne pyruvate into cancer cells) is variant in CCA cells . The glycolysis induced low pyruvate levels are due to both reduced production and excessive consumption. This reduced pyruvate production comes from the expression of pyruvate kinase, the enzyme responsible for the generation of pyruvate in glycolysis. We found the splice variant PKM2 (musclespecific pyruvate kinase 2) is expressed specifically in CCA with low catalytic activity. Moreover, we confirmed the high expression of LDHA in CCA, which is effective in the conversion of pyruvate into lactate and contributed to the predominant pyruvate consumption. Collectively, the high expression of PKM2 and LDHA maintained the low intracellular levels of pyruvate in cancer cells. Interestingly, inhibition of any one of these targets could neither induce tumor proliferation, nor reverse CCA metabolic type. Yet inhibition of these three targets increased pyruvate levels while promoting apoptosis.
Since pyruvate is an energy-rich nutrient necessary for growth in non-malignant cells, there must be an upstream regulator to maintain low levels of this metabolite in cancer cells. Our findings of the present study found that the upstream regulator of pyruvate is cMYC. As an oncogene, c-Myc has attracted extensive interest for its potential role in contributing to tumorigenesis. There are various mechanisms of cMYC-induced tumorigenesis, and an increased Warburg effect is one such mechanism. Many studies have also found that MYC increased metabolic proteins, such as LDHA and PKM2 [6, 7]. Thus, it has been considered a promising cancer target. Our studies demonstrate that cMYC is highly expressed in CCA and predicted a poor prognosis. Moreover, cMYC is positively correlated with PKM2 and LDHA, and contributed to the low pyruvate level. The stability of c-Myc protein is related to its acetylation at K323 [9, 10], and HDACi treatment, but not SIRTi treatment, induced c-Myc K323 acetylation as well as tumorigenesis inhibition [11, 12]. Employing immunoprecipitation, we further inquired as to which HDAC interacted with cMYC, and found that HDAC3 deacetylated cMYC at K323 and further protected cMYC from ubiquitinated degradation. Our work establishes cMYC inhibition as a strategy to accumulate pyruvate, which is effective in promoting apoptosis in CCA cells.
After confirming that the upstream regulator of pyruvate is cMYC, we moved to the other side and inquired as to what is the downstream target of pyruvate in CCA. Previous studies have concluded that the tumor suppressive function of pyruvate is related to its ability to inhibit HDACs; pyruvate is an HDAC inhibitor and a tumor suppressor . Here we confirmed HDAC3 is the downstream target of pyruvate. It is important to note that HDAC3, which is capable of being inhibited by pyruvate, is up-regulated in CCA and protects CCA cells form apoptosis . The elevation of HDAC3 activity is presumably necessary for the cancer cells to maintain their malignant phenotype . Therefore, CCA cells must maintain low intracellular levels of pyruvate, lest HDAC3 will be inhibited and cell growth prevented by enhanced apoptosis.
In the present study, we set out to identify a promoting role for low pyruvate levels regulated by c-Myc and its dynamic acetylation in cancer cell proliferation. Moreover, we reported a unique phenomenon of the positive feedback loop with cMyc and HDAC3. Therapeutic strategies targeting pyruvate may be applicable to cancer treatment in a wide variety of tissues, since high expression of c-Myc is a common phenomenon in cancer.
We thank the Zhao lab for offering their help.
This work was supported by grants from the National Natural Science Foundation of China (No. 81602076, 81672935 and 81472756), Outstanding Youth Project of Nanjing City (No. JQX17002), the Jiangsu Clinical Medical Center of Digestive Disease (BL2012001), the Natural Science Foundation from the Department of Science & Technology of Jiangsu Province (BK20160113), the Fund of Jiangsu Provincial Commission of Health and Family Planning (No. Q201611), the Fundamental Research Funds for the Central Universities (No. 021414380244) and Macau Science and Technology Development Fund (009/2017/A1).
Availability of data and materials
All data generated or analysed during this study are included in this published article [and its supplementary information files].
MZ, LW and XZ designed the study; QZ, DT, YL and YP did the cell experiments; YL, YY, YW and LX collected the tissue samples; QZ and YL performed the protein analysis; LZ, YW, MZ and YL drafted the manuscript and performed the immunohistochemistry experiment; BK, LW and JW performed the metabolite analysis; RGD and MZ wrote the manuscript, MZ, HF, SZ and XZ supported the study. All authors read and approved the final manuscript.
Ethics approval and consent to participate
All experiments utilizing animal and human samples were approved by the Ethical Committee of Medical Research, Nanjing Drum Tower Hospital, Affiliated Hospital of Nanjing University Medical School.
Consent for publication
The authors declare that they have no competing interests.
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.