Liver cancer has been identified as a growing clinical problem, being the second most common cause of cancer‐related deaths globally, and estimated to be responsible for nearly 9.1% of deaths in 2012 (GLOBOCAN,
Two mechanistic components have been identified that foster HCV‐associated hepatocarcinogenesis: First, an inflammatory stage induced by chronic hepatocyte damage leading to the release of reactive oxygen species, apoptosis signals, nucleotides, and hedgehog ligands, ultimately leading to the activation of hepatic stellate cells and the progression of fibrosis toward cirrhosis. Second, a direct effect of HCV core as well as nonstructural proteins on profibrogenic genes such as TGF‐beta, as well as intracellular signaling pathways, including the mitogen‐activated protein kinase cascade. Several viral proteins have been identified in liver carcinogenesis, with HCV core protein being the most prevalent regarding cellular transformation and direct interaction with the tumor suppressor protein p53.
Despite our growing knowledge of HCV interactions, HCV‐mediated transformation and the progression to hepatocellular cancer remains incompletely understood.
We previously uncovered a so far unidentified connection between HCV core and promyelocytic leukemia‐nuclear bodies (PML‐NBs). PML is a tumor suppressor protein, initially described in its role during the pathogenesis of acute promyelocytic leukemia (APL). PML localizes to nuclear bodies (NBs), which are multi‐protein complexes. As of today, loss of PML has been described in numerous human cancers. In our in vivo approach, we could show that the deficiency of PML leads to increased HCC development in HCV‐transgenic mice when animals were treated with an established protocol of phenobarbital and diethylnitrosamine (DEN). Of note, HCV core protein targets PML‐NBs and inactivates the tumor suppressor function of PML through interference with the apoptosis‐induction of PML‐isoform IV in human hepatoma cells.
We have now investigated these mice for spontaneous development of dysplastic nodules as well as HCC. We could show that by the age of 12 months, 40% of all mice of the HCV‐transgenic, PML‐deficient group (HCVtgPML−/−) developed liver tumors, when compared to mice that were just PML−/− (but HCV negative), or HCVtg (but PML wild type), or wild type animals. Our in vivo data demonstrate a direct effect of HCV toward spontaneous liver tumor development under PML deficiency. This highlights the importance and direct correlation of PML as a tumor suppressor during HCV‐related carcinogenesis. Importantly, we have also reflected results of gene expression profiling from these mice with a human cohort of patients that were liver transplanted for HCC.
PML−/− mice (within a 129Sv genetic background) were generated by Pier Paolo Pandolfi (Beth Israel Diaconess Medical Center) as described previously. HCV transgenic FL‐N/35 mice, carrying the full‐length protein‐coding region of HCV genotype 1b, (within a C3H/C57BL6 genetic background) were generated by Herve Lerat (INSERM) and Stanley M. Lemon (UTMB). The two mouse stains were crossed, and the following four genotypes were used for this study: (a) WT; (b) PML−/−; (c) HCVtg; (d) PML−/−HCVtg. Genotyping of all used genotypes was performed by PCR and analyzed on 2% agarose gels. For a complete list of primer sequences see Table S1.
Transgenic mice (PML−/−, HCVtg, PML−/−HCVtg) as well as WT mice were left untreated under frequent observation. After 1 year 10 male mice of each group were sacrificed and analyzed. The general habitus as well as bodyweight were determined. After complete necropsy, the liver tissue of all animals was macrodissected and processed according to the analyses described below.
Liver tissue of patients undergoing liver transplantation was collected and further processed for RNA extraction (for indications see Table S2) as described elsewhere, and liver tissue of patients undergoing liver resection was collected for protein analysis (for indications see Table S3). Patients had given informed consent on scientific use of resected liver tissue, and all human tissue samples were collected in accordance with the Declaration of Helsinki. Samples for further analysis were collected from HCC tissue (TT) and tumor‐surrounding liver tissue (TST), as well as from livers without evidence of hepatocellular carcinoma formation (NTT) of each patient's explanted liver at time of transplantation. The underlying liver disease of these patients is listed in Tables S2 and S3.
Liver samples were formalin‐fixed, paraffin‐embedded, sectioned at 4 µm, and processed routinely for H&E staining. Immunohistochemical staining of glutamine synthetase (ab49873; 1:10 000; Abcam) and Ki67 (ab66155; 1:1000; Abcam) was performed on formalin‐fixed, paraffin‐embedded liver sections with the Histar Detection Kit Star3000a according to the manufacturer's instructions (Bio‐Rad Laboratories, Inc). Morphometry was used to quantify the stained tissue area using ImageJ software (National Institute of Health). Slides were counterstained with hematoxylin.
For protein isolation, 30 mg of human liver or tumor tissue were macrodissected and lysed in ice‐cold RIPA buffer (50 mmol/L Tris, 150 mmol/L NaCl, 1% NP‐40, 0.5% sodiumdeoxycholate, 1 mmol/L EDTA, pH 7.4) complemented with protease and phosphatase inhibitor (Halt Phosphatase Inhibitor Single‐Use Cocktail, Thermo Scientific) by mechanic homogenization using TissueRuptor (Qiagen Inc). Protein lysate was incubated on ice for 30 minutes, and cell debris was removed through 20 minutes of centrifugation at 12 000 × g and 4°C. The supernatant was collected and total protein concentration was determined using the Pierce™ BCA Protein Assay Kit (Thermo Fischer Scientific).
Cultured cells were washed in ice‐cold PBS and lysed by adding 100 µL RIPA buffer (50 mmol/L Tris, 150 mmol/L NaCl, 1% NP‐40, 0.5% Sodium Deoxycholate, 1 mmol/L EDTA, pH 7.4) per well of a 12‐well plate. Cell lysates were cleaned by 10 minutes of centrifugation at 5000 g and 4°C. Total protein concentration was determined using the Pierce™ BCA Protein Assay Kit (Thermo Fischer Scientific).
For Western blots, 25 µg of total protein lysate was separated by SDS gel electrophoresis and transferred to PVDF‐membranes. For protein detection, the following antibodies were used: rabbit anti‐PML (H‐238; 1:1000; Santa Cruz Biotechnology, Inc); rabbit anti‐NLRP12 (ab93113; 1:200; Abcam); rabbit anti‐RASSF6 (ab220111; 1:200; Abcam); rabbit anti‐GAPDH (14C10; 1:1000; Cell Signaling Technology); anti‐rabbit mouse Antibody (31458; 1:10.000; Thermo Fisher Scientific).
For RNA isolation, 30 mg of murine or human liver or tumor tissue was lysed in QIAzol Lysis Reagent (Qiagen Inc) by mechanic homogenization using TissueRuptor (Qiagen Inc). Total RNA and miRNA were isolated by phenol‐chloroform extraction followed by isopropanol precipitation. RNA purification was done using the miRNeasy kit (Qiagen Inc) according to the manufacturer's instructions.
Cultured cells were lysed in 350 µL RLT buffer (mRNeasy kit [Qiagen Inc]) and RNA isolation was performed according to the manufacturer's instructions.
First strand cDNA synthesis was performed with 1 µg total RNA using M‐MLV reverse transcriptase (Thermo Fisher Scientific Inc) and 6 µM Random Primer Mix (New England Biolabs) according to the manufacturer's instructions. Quantitative rt‐PCR was performed with the QuantiFast SYBR green PCR Kit (Qiagen Inc) on the C1000 Touch Thermal Cycler (Bio‐Rad Laboratories, Inc). Gene expression analysis was performed with Microsoft Excel (Microsoft Corp.) and GraphPad Prism6 (GraphPad Software, Inc) software after normalization to β‐Actin (ACTB) in murine and cell culture samples and glyceraldehyde‐3‐phosphate dehydrogenase (GAPDH) in human samples. The sequences of all primers are listed in Table S1.
For whole genome expression analysis on the GeneChip® HT MG‐430 PM Array Plate (Affymetrix Inc), total RNA was transcribed with 3′ IVT Express Kit (Affymetrix Inc) according to the manufacturer's instruction and further processed with GeneTitanTM Hybridization, Wash, and Stain Kit for 3' IVT Arrays (Affymetrix Inc) and the GeneTitan® Wash Buffers A and B Module (Affymetrix Inc). The experiments were performed on the GeneTitan® Instrument (Affymetrix Inc). The data discussed in this manuscript have been deposited in NCBI's Gene Expression Omnibus and are accessible through GEO Series accession number
Background signal was excluded by setting the general expression value in WT samples >200. Using two‐sided t Test between WT and NTT as well as NTT and TST, genes showing significantly different expression levels were identified. Hierarchical cluster analysis was performed with relative expression values normalized to WT samples using the Multiple Experiment Viewer (MeV 4.9.0). We performed average linkage clustering by genes and samples.
For gene set enrichment analysis (GSEA) 3.0 (JAVA version) provided by
Huh7 and HepG2 cells were obtained from the American Type Culture Collection (ATCC). Both cell lines were cultured in Dulbecco's modified essential medium (Gibco) supplemented with 10% heat‐inactivated fetal bovine serum (Gibco) and 100 U/mL penicillin and 100 µg/mL streptomycin sulfate (Gibco) in a humidified incubator with 5% CO2 at 37°C.
For protein analysis, 1 × 105 Huh7 or HepG2 cells were seeded per well of a 12‐well plate. For RNA analysis, 5 × 104 Huh7 or HepG2 cells were seeded per well of a 24‐well plate. For proliferation assays, 1 × 104 Huh7 or HepG2 cells were seeded per well of a 96‐well plate.
All cells were transfected with 20 nmol/L control or gene‐specific si‐RNA (siCtrl, SI03650318; siPml, SI00034664, siRassf6, SI03243275, and SI04323991; siNlrp12, SI04272758, and SI04146933; Qiagen Inc). Transfection was performed with the siLentFect™ Lipid Reagent according to the manufacturer's instructions (Bio‐Rad Laboratories, Inc). After 72 hours, cells were harvested according to the requirements of the subsequent analysis.
For cell proliferation analysis, cells were incubated with BrdU for 2 hours and analyzed by BrdU incorporation assay (Roche, Basel, Switzerland) according to the manufacturer's instruction and measured by luminescence detection. Fold change of the fluorescence signal of gene‐specific siRNA transfected cells was measured and normalized to cells transfected with siCtrl. Cell proliferation assay was performed in triplicates of at least three independent analyses.
Fold change of gene and protein expression as well as cell proliferation is provided as median and range of at least three independent experiments when not stated otherwise. The comparison of two groups was performed by student's t test. Three or more groups were compared by two‐way ANOVA. A P‐value < 0.05 was considered statistically significant.
In order to investigate the contribution of PML and HCV to liver tumor development, we selected mice that either carried the HCV transgene (HCVtg), were PML‐deficient (PML−/−) or had a combined genotype of PML knockout and HCV transgene (PML−/−HCVtg) and compared these groups to wild‐type mice (WT; Figure S1A,B). Mice were observed until they reached the age of 12 months, after which we assessed the presence of macroscopic liver tumors at the surface of each animal's liver in each experimental group. Within the PML−/−HCVtg group, 40% of all animals presented with at least one macroscopic tumor by the age of 12 months, whereas there were no lesions present in livers of mice of the three other groups (HCVtg, PML or WT; Figure A). H&E staining showed no overall fibrotic changes within each liver of each experimental group (Figure B). Immunohistochemistry confirmed high expression of glutamine synthetase within the liver cell tumors of the PML−/−HCVtg group (Figure D), as typically observed in HCC tissue.
Since mice with a genetic combination of PML‐deficiency and presence of HCV‐transgene exhibit increased expression of glutamine synthetase as well as macroscopic liver tumors, we assessed the proliferation of hepatocytes in each experimental group using Ki67 immunohistochemistry (Figure A,B). Mice of the WT, PML−/−, and HCVtg group exhibited <10% of Ki67‐positive nuclei per field. When the PML−/−HCVtg group was analyzed, mice that did not develop liver tumors also showed <10% of Ki67‐positive nuclei. However, when liver tissue of PML−/−HCVtg mice that did develop liver tumors was assessed, there was a significant increase of Ki67‐positive nuclei of up to 14% in the tumor‐surrounding tissue (TST) and up to 60% in the tumor tissue (TT) itself when compared to all other experimental groups (Figure B). This indicates that a subgroup (40%) of PML−/−HCVtg animals that did develop liver tumors harbor a permissive environment that facilitates tumor growth.
We next aimed at identifying mechanisms that may contribute to the increased proliferation of hepatocytes in a subgroup of PML−/−HCVtg mice that did develop tumors vs the subgroup of PML−/−HCVtg mice that never developed lesions within 12 months, as well as WT animals. We performed gene expression profiling of whole liver mRNA from WT animals, as well as from whole liver of PML−/−HCVtg mice that never developed a liver tumor (NTT), and compared these with gene expression profiling of whole liver mRNA from tumorous tissue (TT) and tumor surrounding tissue (TST) from PML−/−HCVtg mice that did develop liver masses (Figure A, Table S4). GSEA identified multiple signaling pathways altered between WT and NTT, NTT and TST as well as TST and TT tissue, in particular gene sets related to liver tumor development (Table S5, Figure S2).
We next analyzed the gene expression data regarding genes showing a continuous and stepwise alteration of expression from WT through NTT to TST and TT. Here we identified two genes, RAS‐Association Domain Family 6 (Rassf6) and NOD‐like receptor Nlrp12, which are known to be associated with PML and NFκB signaling pathway. Rassf6 and Nlrp12 were significantly downregulated in TT of PML−/−HCVtg mice vs WT animals (Figure B). Of note, a stepwise decrease in expression of these genes was observed when comparing WT animals with PML−/−HCVtg mice that did not develop liver tumors (NTT), as well as TST of PML−/−HCVtg mice that did develop liver tumors.
We subsequently analyzed human liver samples for the expression of RASSF6 and NLRP12 in association with the expression of PML. Here, we utilized whole‐liver mRNA from tumor tissue (TT) and tumor‐surrounding tissue (TST) from livers of patients undergoing liver resection for HCC, and compared these findings to liver tissues resected due to various underlying liver diseases, but without the presence of HCC (NTT). We compared mRNA expression of PML, RASSF6, and NLRP12 in livers that were resected without evidence of HCC (NTT) to livers that harbored HCC. Of the livers that contained HCC, we selected TST as well as TT and compared expression of the above‐mentioned genes (Figure A). For PML there was no significant difference in expression pattern, even though PML in TT liver tissue was slightly decreased. When we analyzed for RASSF6 and NLRP12 expression, we could find a highly significant decrease in both RASSF6 and NLRP12 within TST, and predominantly within TT.
These findings were confirmed on a posttranscriptional level (Figure B,C; Table S3) in livers of patients undergoing liver transplantation for HCC. Explanted livers were analyzed for protein expression in TT vs TST. Protein expression of PML and RASSF6 is decreased in TT vs TST of five livers explanted due to HCC. Here, NLRP12 expression was equal between TST and TT in the five selected livers (Figure B). Quantification of immunoblot of a total of 10 livers confirms stable decrease in PML expression within the TT sample as compared to TST sample (Figure C). A significant decrease of RASSF6 in TT samples was found, whereas there was no notable difference for NLRP12 expression on a posttranscriptional level (Figure C). Table S3 gives a complete list of underlying liver disease of each selected patient. These data suggest that PML and RASSF6 both play a major role in HCC development, whereas NLRP12 is differentially regulated in TT vs TST liver tissue at least on an RNA level.
Since we found a differential expression pattern of PML, RASSF6, and NLRP12, we next aimed to assess the influence of these genes to cell proliferation in vitro. HUH7 and HepG2 cells were transfected using either control si‐RNA, or si‐RNA to PML, NLRP12, or RASSF6. We first confirmed a successful knockdown of these genes as confirmed on RNA and protein level (Figure A,B).
To determine whether decreased expression of PML, NLRP12, or RASSF6 leads to increased cell proliferation, cells were transfected using either control si‐RNA, or si‐RNA to PML, NLRP12 or RASSF6, after which cells were incubated with BrdU for 2 hours. Cells transfected with siNLRP12 showed a marked increase in cell proliferation when compared to cells transfected with control si‐RNA, or si‐RNA to PML or RASSF6, as demonstrated via nuclear BrdU incorporation (Figure C). The experiment was repeated using alternative siRNAs to RASSF6 as well as NLRP12, reflecting our findings (Figure S3).
These findings indicate a strong correlation between loss of PML, RASSF6, and NLRP12 and increased cell proliferation, both in vitro and in vivo.
Hepatocellular carcinoma is known to be not just one of the most common types of cancer but also one of the most common causes of cancer‐related deaths worldwide. The growing body of evidence for HCV as a major risk factor for the development of liver fibrosis, cirrhosis, and HCC also suggests that the virus plays a direct role in the neoplastic transformation of hepatocytes. However, the molecular mechanism to this sequence of events is still poorly understood. We have recently conducted a thorough analysis on HCC development under a standardized induction protocol using DEN injections in PML‐deficient, HCV‐transgenic mice. Our data suggested that PML deficiency increases susceptibility toward carcinogenic stimuli, that HCV promotes carcinogenesis in the liver, and that the oncogenic potential of HCV is supported by an inactivation of PML. Here, our aim was to analyze the spontaneous development of liver tumors in these mice. Interestingly, 40% of the PML−/−HCVtg mice developed liver tumors by the age of 12 months compared to all other genotypes. Further gene expression profiling of these livers (when compared with the PML−/−HCVtg livers that did not develop tumors) provided numerous genes associated with increased cell proliferation. RASSF6 has been shown to be expressed in low amounts in HCC, and its overexpression correlates with decreased cell proliferation and invasion in vitro, as well as attenuated tumor growth in a rodent model. The inhibitory effects are described to be established through suppression of FAK phosphorylation, which in turn leads to decreased MMP2/9 expression. Our analysis confirms this finding, since we established the downregulation of RASSF6 in liver tissue extracted from liver tumors. Interestingly, not only the tumorous tissue but also the tumor‐surrounding tissue of PML−/−HCVtg livers, and PML−/−HCVtg livers without tumor development showed a stepwise decrease in RASSF6 expression. PML‐deficiency in combination with HCV therefore is associated with decreased expression of RASSF6, correlating with increased cell proliferation and tumor growth in vivo and in vitro.
NLRP12, which encodes a negative regulator of innate immunity, has been described to promote specific commensals that can reverse gut inflammation. However, to the best of our knowledge, its contribution to liver carcinogenesis has not been described so far. Our data suggest that within PML‐deficient and HCV‐transgenic mice, NLRP12 is downregulated in tumor tissue on an RNA level, but not on a posttranscriptional level. This was confirmed in our patient cohort where protein levels were unaltered in contrast to differential expression within tumor tissue vs nontumorous and tumor‐surrounding tissue on an RNA level. Further exploration of this mechanism is intriguing, but beyond the scope of this paper.
Our in vitro data confirm the increase in cell proliferation when either PML, RASSF6, or NLRP12 are downregulated. We chose to leave the cells HCV‐negative in order to assess the unaltered effects of each gene alone.
The correlation of tumor abundance as well as PML deficiency as well as downregulated RASSF6 and NLRP12 expression was confirmed in our patient cohort. As an important point to mention, we chose to investigate liver tissue from patients with various underlying liver diseases, and patients undergoing liver transplantation as well as liver resection. This selection might have blunted the overall effect of gene and protein expression as demonstrated in our analysis. The majority of our patients were HCV positive. However, the heterogeneous underlying diseases reflect the reality of clinical presentation. Future prospective studies will be performed in order to further distinguish between underlying liver disease and correlation with expression patterns of selected genes.
Taken together, this study provides further insight into the role of PML fostering a permissive milieu for the liver toward tumor development, as well as supporting HCV‐related HCC development.
None declared.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Copyright John Wiley & Sons, Inc. Jul 2019
Abstract
Persistent infection with hepatitis C virus (HCV) is a known risk factor for the development of hepatocellular carcinoma (HCC). The lack of the tumor suppressor promyelocytic leukemia protein (PML) in combination with HCV fosters hepatocarcinogenesis via induction of HCC using diethylnitrosamine (DEN) in a rodent model. However, the spontaneous development of malignant lesions in PML‐deficient mice with an HCV‐transgene (HCVtg) has not been investigated thus far. We crossed PML‐deficient mice with HCV transgene expressing mice and observed the animals for a period of 12 months. Livers were examined macroscopically and histologically. Gene expression analysis was performed on these samples, and compared with expression of selected genes in human samples of patients undergoing liver transplantation for HCC. In vitro studies were performed in order to analyze the selected pathways. Genetic depletion of PML in combination with HCVtg coincided with an increased hepatocyte proliferation, resulting in development of HCCs in 40% of the PML‐deficient livers. No tumor development was observed in mice with either the PML‐knockout (PML−/−) or HCVtg alone. Gene expression profiling uncovered pathways involved in cell proliferation, such as NLRP12 and RASFF6. These findings were verified in samples from human livers of patients undergoing liver transplantation for HCC. Further in vitro studies confirmed that lack of PML, NLRP12, and RASFF6 leads to increased cell proliferation. The lack of PML in combination with HCV is associated with increased cell proliferation, fostering tumor development in the liver. Our data demonstrate that PML acts as an important tumor suppressor in HCV‐dependent liver pathology.
You have requested "on-the-fly" machine translation of selected content from our databases. This functionality is provided solely for your convenience and is in no way intended to replace human translation. Show full disclaimer
Neither ProQuest nor its licensors make any representations or warranties with respect to the translations. The translations are automatically generated "AS IS" and "AS AVAILABLE" and are not retained in our systems. PROQUEST AND ITS LICENSORS SPECIFICALLY DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING WITHOUT LIMITATION, ANY WARRANTIES FOR AVAILABILITY, ACCURACY, TIMELINESS, COMPLETENESS, NON-INFRINGMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Your use of the translations is subject to all use restrictions contained in your Electronic Products License Agreement and by using the translation functionality you agree to forgo any and all claims against ProQuest or its licensors for your use of the translation functionality and any output derived there from. Hide full disclaimer
Details
1 Department of Gastroenterology and Hepatology, Faculty of Medicine, University Hospital Essen, University of Duisburg‐Essen, Essen, Germany
2 Department of General‐, Visceral‐ and Transplantation Surgery, Faculty of Medicine, University Hospital Essen, University of Duisburg‐Essen, Essen, Germany
3 Institute of Pathology, Faculty of Medicine, University Hospital Essen, University of Duisburg‐Essen, Essen, Germany