Research Article
A common polymorphism in the NCAN gene is associated with hepatocellular carcinoma in alcoholic liver disease Hans Dieter Nischalke1, , Philipp Lutz1,7, , Benjamin Krämer1,7, Jennifer Söhne1, Tobias Müller2, Jonas Rosendahl2, Janett Fischer2, Thomas Berg2, Kanishka Hittatiya3, Hans-Peter Fischer3, Michael Soyka4,5, Nasser Semmo6, Jacob Nattermann1,7, Tilman Sauerbruch1, Christian P. Strassburg1,7, Felix Stickel6,à, Ulrich Spengler1,7,⇑,à 1
Department of Internal Medicine I, University of Bonn, Germany; 2Department of Gastroenterology, University Hospital Leipzig, Germany; 3 Department of Pathology, University Hospital Bonn, Germany; 4Psychiatric Hospital Meiringen, Switzerland; 5Psychiatric Hospital, University of Munich, Germany; 6Hepatology Unit, Klinik Beau-Site, Department of Visceral Surgery and Medicine Inselspital, University of Bern, Switzerland; 7German Center for Infection Research, Germany See Editorial, pages 971–972
Background & Aims: The genetic background of alcoholic liver diseases and their complications are increasingly recognized. A common polymorphism in the neurocan (NCAN) gene, which is known to be expressed in neuronal tissue, has been identified as a risk factor for non-alcoholic fatty liver disease (NAFLD). We investigated if this polymorphism may also be related to alcoholic liver disease (ALD) and hepatocellular carcinoma (HCC). Methods: We analysed the distribution of the NCAN rs2228603 genotypes in 356 patients with alcoholic liver cirrhosis, 126 patients with alcoholic HCC, 382 persons with alcohol abuse without liver damage, 362 healthy controls and in 171 patients with hepatitis C virus (HCV) associated HCC. Furthermore, a validation cohort of 229 patients with alcoholic cirrhosis (83 with HCC) was analysed. The genotypes were determined by LightSNiP assays. The expression of NCAN was studied by RT-PCR and immunofluorescence microscopy. Results: The frequency of the NCAN rs2228603 T allele was significantly increased in patients with HCC due to ALD (15.1%) compared to alcoholic cirrhosis without HCC (9.3%), alcoholic controls (7.2%), healthy controls (7.9%), and HCV associated HCC (9.1%). This finding was confirmed in the validation cohort (15.7% vs. 6.8%, OR = 2.53; 95%CI: 1.36–4.68; p = 0.0025) and by
Keywords: NCAN; HCC; Liver cirrhosis; Alcohol; Neurocan; rs2228603; Polymorphism; SNP. Received 21 October 2013; received in revised form 17 May 2014; accepted 2 June 2014; available online 16 June 2014 q DOI of original article: http://dx.doi.org/10.1016/j.jhep.2014.08.001. ⇑ Corresponding author. Address: Department of Internal Medicine I, University of Bonn, Sigmund-Freud-Strasse 25, D–53129 Bonn, Germany. Tel.: +49 228 287 16789; fax: +49 228 287 19822. E-mail address: [email protected] (U. Spengler). These authors contributed equally to this work. à These authors share senior authorship. Abbreviations: AFP, alpha fetoprotein; NCAN, neurocan; NAFLD, non-alcoholic fatty liver disease; ALD, alcoholic liver disease; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; OR, odds ratio; CI, confidence interval; ECM, extracellular matrix; SNP, single nucleotide polymorphism; HBV, hepatitis B virus.
multivariate analysis (OR = 1.840; 95%CI: 1.22–2.78; p = 0.004 for carriage of the rs2228603 T allele). In addition, we identified and localised NCAN expression in human liver. Conclusions: NCAN is not only expressed in neuronal tissue, but also in the liver. Its rs2228603 polymorphism is a risk factor for HCC in ALD, but not in HCV infection. Ó 2014 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.
Introduction The net burden caused by alcohol consumption in EU countries is 1 in 7 deaths in men and 1 in 13 deaths in women [1], particularly due to 39,000 cases of liver cirrhosis [2]. Once liver cirrhosis has established, the risk for developing hepatocellular carcinoma (HCC) is 8% within 5 years [3]. Alcoholic cirrhosis is a complex trait in which host and environmental factors interact, and increasingly, host genetics are recognized as a pivotal modifier of the risk to develop significant liver disease [4]. Until now, a single nucleotide polymorphism (SNP) in the gene coding for patatin-like phospholipase domain-containing 3 (PNPLA3; adiponutrin; rs738409 C/G, M148I) is the only robustly confirmed marker indicating a genetic risk for alcoholic cirrhosis and HCC [5–10] which, however, only refers to approximately 27% of the population-attributable risk. Neurocan (NCAN) is a chondroitin sulphate proteoglycan belonging to the extracellular matrix (ECM) [11]. It has been intensively studied in neural tissue, where it is involved in brain development and remodelling [12]. Surprisingly, the SNP rs2228603 in the NCAN gene, that leads to an amino acid exchange (proline to serine) at position 92, was found to be closely associated with plasma LDL and triglyceride levels [13]. In line with this association, a genome-wide association study revealed that this SNP is linked to hepatic steatosis
Journal of Hepatology 2014 vol. 61 j 1073–1079
Research Article [14]. Further studies confirmed this finding [15,16], but did not investigate NCAN expression in liver tissue. However, the molecular mechanism linking NCAN and lipid metabolism remains elusive. Ncan/ mice do not show any obvious defects [17]. Since steatosis is universally present in heavy drinkers, we hypothesized that the NCAN rs2228603 polymorphism could be involved in ALD. To test this hypothesis, we studied the genotypic and allelic distribution of NCAN rs2228603 in large cohorts of patients with alcoholic cirrhosis with and without HCC, in healthy controls, alcoholic subjects without overt liver damage and, as a further control, in patients with hepatitis C virus (HCV)-associated HCC.
using the RNeasy Mini Kit (Qiagen, Hilden, Germany). Elimination of genomic DNA and reverse transcription of 1 lg total RNA was carried out using the QuantiTect Reverse Transcription Kit (Qiagen, Hilden, Germany). For each PCR 1 ll of the obtained cDNA was used as template. PCR was carried out in a final volume of 10 ll on a LightCycler instrument using the Maxima SYBR Green qPCR Master Mix (Thermo Fisher Scientific Inc., Waltham, MA) and 0.5 lM of primers. Primers for NCAN were commercially available from Qiagen (QuantiTect Primer Assay) and primers of the house-keeping gene b-actin (sense: 50 -TGGCATCGTGATGGACTCC-30 , antisense: 50 -AATGTCACGCACGATTTCCC-30 ) were obtained from TibMolbiol, Berlin, Germany. An identical amplification protocol was used for both PCRs as provided by the manufacturer. The LightCycler software version 3.5 was used in all PCR experiments. The NCAN Ct (threshold cycle) values were normalized by calculating the DCt values (DCt = Ct NCAN – Ct b-actin) [20]. The relative expression of NCAN mRNA was expressed as 1/DCt. Immunofluorescence microscopy
Patients and methods Patients We collected blood for genotyping and clinical data of 482 consecutive Caucasian patients with liver cirrhosis due to alcohol abuse (126 with HCC) from the Departments of Hepatology/Gastroenterology of the Universities Bonn, Berlin (both Germany) and Bern (Switzerland). 382 Caucasian patients with heavy alcohol abuse (>10 years), but normal sonographic appearance of the liver and normal liver enzyme levels (except for elevation of gamma-glutamyltranspeptidase) served as controls as reported previously [9,18]. DNA and clinical data from a cohort of patients with HCV-associated HCC described formerly [19] were included as controls. Furthermore, 362 healthy Caucasian controls were analysed to determine the genotype distribution in the background population. As a replication cohort, we used DNA and clinical data from 229 Caucasian patients with alcoholic cirrhosis (83 with HCC) from the Department of Gastroenterology, University Leipzig (Germany). The presence of cirrhosis was determined by the results of the liver biopsy (n = 174) or unequivocal clinical and laboratory evidence of cirrhosis as reflected by standard blood tests (coagulation tests, serum albumin concentration, platelet count), occurrence of cirrhosis-related complications such as encephalopathy, ascites or esophageal varices and consistent radiological findings. Patients were considered to have alcoholic cirrhosis if their history indicated average alcohol consumption to exceed 300 g ethanol per week. The diagnosis of HCC was made by contrast enhanced imaging according to established diagnostic criteria (current EASL and AASLD guidelines). Patients with concomitant viral infection (anti HCV positive or anti HBc positive) were excluded. Patients were screened for the development of HCC as long as they were seen in their department. Ethical approval The study protocol complied with the ethical guidelines set forth in the Declaration of Helsinki and was approved by the local ethics committees from all participating centres. Written informed consent from patients was obtained prior to inclusion into this study or via their written general consent to be subject of a retrospective study.
Liver specimens (cirrhotic liver explants from alcoholic cirrhosis, HCV-induced cirrhosis or surrounding normal tissue from resection of metastases (n = 4 for each group) and from alcoholic cirrhosis containing HCC and adjacent tissue [n = 2]) were shock-frozen in liquid nitrogen and then cut into 7 lm slices on a Leica microtome (Microsystems GmbH, Wetzlar, Germany). After drying overnight, sections were fixed with 4% paraformaldehyde for 10 min and blocked with 10% donkey serum and 1% bovine serum albumin solved in PBS. The following anti-human antibodies were used: anti-NCAN (polyclonal rabbit IgG, Atlas antibodies Stockholm, Sweden), anti Hep Par 1 (Clone OCH1E5, Dako, Hamburg, Germany), and anti-alpha-smooth muscle actin (alpha-SMA) (1:200, Abcam, Cambridge, UK). Double-staining was performed using NorthernLights coupled secondary antibodies NL-488, NL-637, and NL-557 (R&D Systems). Nuclei were counterstained with DAPI (Life Technologies-Invitrogen, Darmstadt, Germany). After mounting with Aqueous Mounting Medium (R&D Systems), sections were scanned with a Zeiss Axiovert 200 M Apotome using the AxioVision Software (Zeiss, Jena, Germany). Statistical analysis Statistical analysis was done with IBM SPSS Statistics software version 21 (IBM, New York, USA). For the analysis of quantitative data, the Wilcoxon-Mann-Whitney-U test and the Student’s t test were used as appropriate. The v2 test was applied to qualitative data. Correspondence of genotype frequencies to the Hardy-Weinberg equilibrium was checked by a web-based software (http://ihg.gsf.de/cgi-bin/hw/ hwa1.pl). Correlation was analysed by Spearman’s rank correlation. Multivariate forward conditional regression analysis was performed on the putative HCC risk factors in order to identify independent risk factors by using p 0.1 for exclusion of parameters in the final model. Functional effects of the NCAN rs2228603 SNP on the protein structure were assessed by a PolyPhen-2 analysis with a web-based software (http://genetics.bwh.harvard.edu/pph2/) [21]. A p value
A common polymorphism in the NCAN gene is associated with hepatocellular carcinoma in alcoholic liver disease Hans Dieter Nischalke1, , Philipp Lutz1,7, , Benjamin Krämer1,7, Jennifer Söhne1, Tobias Müller2, Jonas Rosendahl2, Janett Fischer2, Thomas Berg2, Kanishka Hittatiya3, Hans-Peter Fischer3, Michael Soyka4,5, Nasser Semmo6, Jacob Nattermann1,7, Tilman Sauerbruch1, Christian P. Strassburg1,7, Felix Stickel6,à, Ulrich Spengler1,7,⇑,à 1
Department of Internal Medicine I, University of Bonn, Germany; 2Department of Gastroenterology, University Hospital Leipzig, Germany; 3 Department of Pathology, University Hospital Bonn, Germany; 4Psychiatric Hospital Meiringen, Switzerland; 5Psychiatric Hospital, University of Munich, Germany; 6Hepatology Unit, Klinik Beau-Site, Department of Visceral Surgery and Medicine Inselspital, University of Bern, Switzerland; 7German Center for Infection Research, Germany See Editorial, pages 971–972
Background & Aims: The genetic background of alcoholic liver diseases and their complications are increasingly recognized. A common polymorphism in the neurocan (NCAN) gene, which is known to be expressed in neuronal tissue, has been identified as a risk factor for non-alcoholic fatty liver disease (NAFLD). We investigated if this polymorphism may also be related to alcoholic liver disease (ALD) and hepatocellular carcinoma (HCC). Methods: We analysed the distribution of the NCAN rs2228603 genotypes in 356 patients with alcoholic liver cirrhosis, 126 patients with alcoholic HCC, 382 persons with alcohol abuse without liver damage, 362 healthy controls and in 171 patients with hepatitis C virus (HCV) associated HCC. Furthermore, a validation cohort of 229 patients with alcoholic cirrhosis (83 with HCC) was analysed. The genotypes were determined by LightSNiP assays. The expression of NCAN was studied by RT-PCR and immunofluorescence microscopy. Results: The frequency of the NCAN rs2228603 T allele was significantly increased in patients with HCC due to ALD (15.1%) compared to alcoholic cirrhosis without HCC (9.3%), alcoholic controls (7.2%), healthy controls (7.9%), and HCV associated HCC (9.1%). This finding was confirmed in the validation cohort (15.7% vs. 6.8%, OR = 2.53; 95%CI: 1.36–4.68; p = 0.0025) and by
Keywords: NCAN; HCC; Liver cirrhosis; Alcohol; Neurocan; rs2228603; Polymorphism; SNP. Received 21 October 2013; received in revised form 17 May 2014; accepted 2 June 2014; available online 16 June 2014 q DOI of original article: http://dx.doi.org/10.1016/j.jhep.2014.08.001. ⇑ Corresponding author. Address: Department of Internal Medicine I, University of Bonn, Sigmund-Freud-Strasse 25, D–53129 Bonn, Germany. Tel.: +49 228 287 16789; fax: +49 228 287 19822. E-mail address: [email protected] (U. Spengler). These authors contributed equally to this work. à These authors share senior authorship. Abbreviations: AFP, alpha fetoprotein; NCAN, neurocan; NAFLD, non-alcoholic fatty liver disease; ALD, alcoholic liver disease; HCC, hepatocellular carcinoma; HCV, hepatitis C virus; OR, odds ratio; CI, confidence interval; ECM, extracellular matrix; SNP, single nucleotide polymorphism; HBV, hepatitis B virus.
multivariate analysis (OR = 1.840; 95%CI: 1.22–2.78; p = 0.004 for carriage of the rs2228603 T allele). In addition, we identified and localised NCAN expression in human liver. Conclusions: NCAN is not only expressed in neuronal tissue, but also in the liver. Its rs2228603 polymorphism is a risk factor for HCC in ALD, but not in HCV infection. Ó 2014 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.
Introduction The net burden caused by alcohol consumption in EU countries is 1 in 7 deaths in men and 1 in 13 deaths in women [1], particularly due to 39,000 cases of liver cirrhosis [2]. Once liver cirrhosis has established, the risk for developing hepatocellular carcinoma (HCC) is 8% within 5 years [3]. Alcoholic cirrhosis is a complex trait in which host and environmental factors interact, and increasingly, host genetics are recognized as a pivotal modifier of the risk to develop significant liver disease [4]. Until now, a single nucleotide polymorphism (SNP) in the gene coding for patatin-like phospholipase domain-containing 3 (PNPLA3; adiponutrin; rs738409 C/G, M148I) is the only robustly confirmed marker indicating a genetic risk for alcoholic cirrhosis and HCC [5–10] which, however, only refers to approximately 27% of the population-attributable risk. Neurocan (NCAN) is a chondroitin sulphate proteoglycan belonging to the extracellular matrix (ECM) [11]. It has been intensively studied in neural tissue, where it is involved in brain development and remodelling [12]. Surprisingly, the SNP rs2228603 in the NCAN gene, that leads to an amino acid exchange (proline to serine) at position 92, was found to be closely associated with plasma LDL and triglyceride levels [13]. In line with this association, a genome-wide association study revealed that this SNP is linked to hepatic steatosis
Journal of Hepatology 2014 vol. 61 j 1073–1079
Research Article [14]. Further studies confirmed this finding [15,16], but did not investigate NCAN expression in liver tissue. However, the molecular mechanism linking NCAN and lipid metabolism remains elusive. Ncan/ mice do not show any obvious defects [17]. Since steatosis is universally present in heavy drinkers, we hypothesized that the NCAN rs2228603 polymorphism could be involved in ALD. To test this hypothesis, we studied the genotypic and allelic distribution of NCAN rs2228603 in large cohorts of patients with alcoholic cirrhosis with and without HCC, in healthy controls, alcoholic subjects without overt liver damage and, as a further control, in patients with hepatitis C virus (HCV)-associated HCC.
using the RNeasy Mini Kit (Qiagen, Hilden, Germany). Elimination of genomic DNA and reverse transcription of 1 lg total RNA was carried out using the QuantiTect Reverse Transcription Kit (Qiagen, Hilden, Germany). For each PCR 1 ll of the obtained cDNA was used as template. PCR was carried out in a final volume of 10 ll on a LightCycler instrument using the Maxima SYBR Green qPCR Master Mix (Thermo Fisher Scientific Inc., Waltham, MA) and 0.5 lM of primers. Primers for NCAN were commercially available from Qiagen (QuantiTect Primer Assay) and primers of the house-keeping gene b-actin (sense: 50 -TGGCATCGTGATGGACTCC-30 , antisense: 50 -AATGTCACGCACGATTTCCC-30 ) were obtained from TibMolbiol, Berlin, Germany. An identical amplification protocol was used for both PCRs as provided by the manufacturer. The LightCycler software version 3.5 was used in all PCR experiments. The NCAN Ct (threshold cycle) values were normalized by calculating the DCt values (DCt = Ct NCAN – Ct b-actin) [20]. The relative expression of NCAN mRNA was expressed as 1/DCt. Immunofluorescence microscopy
Patients and methods Patients We collected blood for genotyping and clinical data of 482 consecutive Caucasian patients with liver cirrhosis due to alcohol abuse (126 with HCC) from the Departments of Hepatology/Gastroenterology of the Universities Bonn, Berlin (both Germany) and Bern (Switzerland). 382 Caucasian patients with heavy alcohol abuse (>10 years), but normal sonographic appearance of the liver and normal liver enzyme levels (except for elevation of gamma-glutamyltranspeptidase) served as controls as reported previously [9,18]. DNA and clinical data from a cohort of patients with HCV-associated HCC described formerly [19] were included as controls. Furthermore, 362 healthy Caucasian controls were analysed to determine the genotype distribution in the background population. As a replication cohort, we used DNA and clinical data from 229 Caucasian patients with alcoholic cirrhosis (83 with HCC) from the Department of Gastroenterology, University Leipzig (Germany). The presence of cirrhosis was determined by the results of the liver biopsy (n = 174) or unequivocal clinical and laboratory evidence of cirrhosis as reflected by standard blood tests (coagulation tests, serum albumin concentration, platelet count), occurrence of cirrhosis-related complications such as encephalopathy, ascites or esophageal varices and consistent radiological findings. Patients were considered to have alcoholic cirrhosis if their history indicated average alcohol consumption to exceed 300 g ethanol per week. The diagnosis of HCC was made by contrast enhanced imaging according to established diagnostic criteria (current EASL and AASLD guidelines). Patients with concomitant viral infection (anti HCV positive or anti HBc positive) were excluded. Patients were screened for the development of HCC as long as they were seen in their department. Ethical approval The study protocol complied with the ethical guidelines set forth in the Declaration of Helsinki and was approved by the local ethics committees from all participating centres. Written informed consent from patients was obtained prior to inclusion into this study or via their written general consent to be subject of a retrospective study.
Liver specimens (cirrhotic liver explants from alcoholic cirrhosis, HCV-induced cirrhosis or surrounding normal tissue from resection of metastases (n = 4 for each group) and from alcoholic cirrhosis containing HCC and adjacent tissue [n = 2]) were shock-frozen in liquid nitrogen and then cut into 7 lm slices on a Leica microtome (Microsystems GmbH, Wetzlar, Germany). After drying overnight, sections were fixed with 4% paraformaldehyde for 10 min and blocked with 10% donkey serum and 1% bovine serum albumin solved in PBS. The following anti-human antibodies were used: anti-NCAN (polyclonal rabbit IgG, Atlas antibodies Stockholm, Sweden), anti Hep Par 1 (Clone OCH1E5, Dako, Hamburg, Germany), and anti-alpha-smooth muscle actin (alpha-SMA) (1:200, Abcam, Cambridge, UK). Double-staining was performed using NorthernLights coupled secondary antibodies NL-488, NL-637, and NL-557 (R&D Systems). Nuclei were counterstained with DAPI (Life Technologies-Invitrogen, Darmstadt, Germany). After mounting with Aqueous Mounting Medium (R&D Systems), sections were scanned with a Zeiss Axiovert 200 M Apotome using the AxioVision Software (Zeiss, Jena, Germany). Statistical analysis Statistical analysis was done with IBM SPSS Statistics software version 21 (IBM, New York, USA). For the analysis of quantitative data, the Wilcoxon-Mann-Whitney-U test and the Student’s t test were used as appropriate. The v2 test was applied to qualitative data. Correspondence of genotype frequencies to the Hardy-Weinberg equilibrium was checked by a web-based software (http://ihg.gsf.de/cgi-bin/hw/ hwa1.pl). Correlation was analysed by Spearman’s rank correlation. Multivariate forward conditional regression analysis was performed on the putative HCC risk factors in order to identify independent risk factors by using p 0.1 for exclusion of parameters in the final model. Functional effects of the NCAN rs2228603 SNP on the protein structure were assessed by a PolyPhen-2 analysis with a web-based software (http://genetics.bwh.harvard.edu/pph2/) [21]. A p value
