- Email: [email protected]
Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence
G Model PRP-51734; No. of Pages 6
ARTICLE IN PRESS Pathology – Research and Practice xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Pathology – Research and Practice journal homepage: www.elsevier.com/locate/prp
Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence Hongxian Yan a,∗,1 , Zhaohui Li b,1 , Quan Shen a , Qian Wang c , Jianguo Tian a , Qingfeng Jiang a , Linbo Gao d a
Department of Hepatobiliary Surgery, Henan Provincial People’s Hospital, Zhengzhou, Henan 650000, PR China Secondary Department of General Surgery, Luoyang Central Hospital Affiliated to Zhengzhou University, Luoyang, Henan 471003, PR China c Department of Hepatobiliary Surgery, Henan Cancer Hospital, Zhengzhou, Henan 650000, PR China d Laboratory of Molecular and Translational Medicine, West China Institute of Women and Children’s Health, Key Laboratory of Obstetric & Gynecologic and Pediatric Diseases and Birth Defects of Ministry of Education, West China Second University Hospital, Sichuan University, Chengdu, Sichuan 610041, PR China b
a r t i c l e
i n f o
Article history: Received 28 September 2015 Keywords: Hepatocellular carcinoma Differentially expressed genes Functional enrichment analysis Pathway enrichment analysis Protein-protein interaction
a b s t r a c t This study aims to deepen our understanding of the molecular mechanism underlying the occurrence of hepatocellular carcinoma (HCC). We first downloaded a gene expression profile dataset GSE29721 (10 HCC and 10 control samples) from Gene Expression Omnibus database (http://www.ncbi.nlm.nih.gov/ geo/). Differentially expressed genes (DEGs) were identified by the paired t-test using limma package. Pathway and functional enrichment analyses were performed with DAVID tools. Transcription factors were annotated with TRANSFAC database and tumor associated genes (TAGs) were annotated with TAG and TSGene databases. Protein-protein interaction (PPI) network was conducted using STRING online tool and function module was further identified with BioNet package. Totally, 527 up-regulated DEGs and 587 down-regulated DEGs were identified. GO functional and KEGG pathway enrichment analyses showed that the up-regulated DEGs were mainly related to cell division and cell cycle, while the down-regulated DEGs were largely related to material metabolism, especially secondary metabolism. Proteins encoded by DEGs CDK1, BUB1, CDC20, NCAPG, NDC80, CDCA8, MAD2L1, CCNB1, CCNA2 and BIRC5 were hub genes with high degrees in the PPI network; further module analysis detected a subnetwork consisting of 55 proteins, such as CYP2B6, ACAA1, BHMT and ALDH2. Taken together, aberrant expression of cell cycle related genes (e.g., CDK1, CCNA2, CCNB1, BUB1, MAD2L1 and CDC20) and material metabolism related genes (e.g., CYP2B6, ACAA1, BHMT and ALDH2) may contribute to HCC occurrence. © 2017 Elsevier GmbH. All rights reserved.
1. Introduction Hepatocellular carcinoma (HCC) is the most common form of liver cancer accounting for 90% of all primary liver tumors, with an survival of 6–20 months if not intervened [1]. It is the fifth most common cancer and the third most common cause of death from cancer worldwide [2]. Strikingly, most HCC cases (>80%) occur in sub-Saharan Africa and Eastern Asia, and most astonishingly, China has 50% of HCC cases globally [3]. HCC brings heavy financial burden to both individuals and the society and this burden is expected to
∗ Corresponding author at: Department of Hepatobiliary Surgery, Henan Provincial People’s Hospital, Weiwu Road, NO. 7, Zhengzhou City, Henan 650000, PR China. E-mail address: [email protected] (H. Yan). 1 Hongxian Yan and Zhaohui Li contributed equally to this work.
increase in the coming years [4]. Unfortunately, the management of patients with HCC has not dramatically changed [5]. Chronic hepatitis B virus (HBV) and chronic hepatitis C virus (HCV) infection are the primary causes of liver cancer, which account for approx. 75–80% of all HCC cases [6,7]. Environmental factors, such as alcoholic, smoking and aflatoxin exposure are important risk factors for HCC [8–10]. Recently, Milgrom et al. claimed that non-alcoholic steatohepatitis would become the first cause of HCC in America due to the increase of obesity-related liver disease [11]. In addition, genetic alterations have also been implicated in HCC pathogenesis, such as mutations in PIK3CA [12], ARID2 [13], TP53 and ARID1A [14], SNP polymorphism in ADAMTS5 [15], overexpression of SALL4 [16] and TFIIB [17]. However, the mechanisms underlying hepatocellular tumorigenesis and progression remain poorly understood.
http://dx.doi.org/10.1016/j.prp.2017.01.019 0344-0338/© 2017 Elsevier GmbH. All rights reserved.
Please cite this article in press as: H. Yan, et al., Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.01.019
G Model
ARTICLE IN PRESS
PRP-51734; No. of Pages 6
H. Yan et al. / Pathology – Research and Practice xxx (2017) xxx–xxx
2
In the present study, we reanalyzed a gene expression profile dataset GSE29721 deposited in GEO by Stefanska et al. [18] using bioinformatics tools, with the attempt to deepen our understanding of HCC pathogenesis. 2. Materials and methods As the paper did not involve any human or animal’s study, the ethical approval or patients gave informed consent were not required. 2.1. Source of gene expression profile data The gene expression profile dataset of GSE29721 was downloaded from the GEO (Gene Expression Omnibus) database (http://www.ncbi.nlm.nih.gov/geo/), which is annotated based on Affymetrix Human Genome U133 Plus 2.0 Array platform. It was deposited in GEO by Stefanska et al. in June, 2011 [18]. A total of 20 cancerous and normal adjacent normal tissue samples were obtained from patients with HCC, respectively, including 10 samples of normal adjacent tissues (control group) and 10 samples of cancerous tissues (tumor group) were used for this analysis.
page show search=on) [28] database and protein-protein interaction pairs with combined score larger than 0.9 were selected. Then, the PPI network of the DEGs was visualized using Cytoscape software [29]. Only the PPI pairs that have got experimental verification, co-expression analysis or record in database were chosen for the PPI network construction. Finally, the interaction network was analyzed, and the degree of each protein (namely, the number of other proteins that were connected to a certain protein), was calculated. Furthermore, functional modules were further identified from the PPI network using the BioNet package [30]. The proteins with high degree in the PPI network and its module were considered as the hub nodes. FDR (false discovery rate) <0.0005 was set as the threshold for this analysis. 2.6. Validation of hub proteins Finally, the gene expression data of liver hepatocellular carcinoma that were available to the public in The Cancer Genome Atlas (TCGA) database (http://cancergenome.nih.gov/) were also downloaded (upgraded to May 5th, 2016) to validate the expression of the hub genes identified in the PPI network and the module. These data had been preprocessed, which were collected from 374 tumor samples and 50 normal samples.
2.2. Identification of differentially expressed genes The raw data of GSE29721 dataset were preprocessed using AFFY package in R/Bioconductor software [19]. The detailed preprocessing protocols included background correction, quantile normalization and probe summarization using Robust Multi-array Average (RMA) package [20] with defaulted parameters. Then the gene expression matrix was obtained. DEGs between the tumor and control samples were identified by the paired t-test using limma package in R language with cutoffs of p < 0.05 and |log2 FC (fold change)|>1 [21]. Meanwhile, DEGs between the tumor and control samples in the TGCA data were also identified by the paired t-test using the same cutoffs. 2.3. Functional and pathway enrichment analyses of DEGs Both Gene Ontology (GO) [22] enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) [23] pathway enrichment analysis were performed to investigate the functions and potential roles of the identified DEGs in the pathogenesis of HCC using the online tool Database for Annotation, Visualization and Integrated Discovery (DAVID, v6.7, https://david.ncifcrf.gov/) [24]. Meanwhile, the enrichment analyses were performed independently in up-regulated and down-regulated DEGs. P value <0.05 was set as the cut-off for enrichment analysis. 2.4. Identification of tumor associated genes With reference to the database TRANSFAC (http://www.generegulation.com/index2), DEGs that were predicted to be transcription factors were identified [25], and the resulting DEGs were finally identified to be transcription factors only when they were also involved in the GO term (Transcription activity). In addition, tumor associated genes (TAG) including oncogenes and tumor suppressor genes were also identified based on the Tumor Suppressor Gene (TSGene) database (http://bioinfo.mc.vanderbilt.edu/ TSGene/) [26,27]. 2.5. Protein-protein interaction (PPI) network construction and module detection DEGs were mapped to the STRING (http://string-db.org/cgi/ input.pl?UserId=OKHJlLJy0nly&sessionId=pJlFY0Zvk3VZ&input
3. Results 3.1. DEGs analysis A total of 1114 DEGs were identified. Among the 1114 DEGs, 527 DEGs were up-regulated, such as CDK1, CCNA2, CCNB1, BUB1, MAD2L1 and CDC20, and 587 DEGs were down-regulated, such as CYP2B6, ACAA1, BHMT, ADH1A, ADH1B, ADH4 and ALDH2. 3.2. Functional and pathway enrichment analyses of DEGs According to the GO functional enrichment analysis, the upregulated DEGs and the downregulated were enriched in 293 and 416 GO terms respectively. The top five GO terms were mainly related to cell mitosis and cell cycle, and the down-regulated GO terms were mostly related to material metabolism (Table 1). According to the KEGG pathway enrichment analysis, the upregulated and downregulated DEGs were enriched in 38 and 11 pathways respectively. The top three pathways enriched by the upregulated genes were related to cell cycle, DNA replication and oocyte meiosis respectively, and the top five pathways enriched by the upregulated genes were all related to metabolism (Table 2). 3.3. Prediction of tumor associated genes Among the downregulated DEGs, 15 genes were transcriptional factors, and 33 genes were TAGs, including 4 oncogenes, 24 tumor suppressor genes and 5 genes with unidentified role in the development of tumor (Table 3). Noticeably, no upregulated genes were identified as TFs or TAGs here. 3.4. PPI network construction and module detection Based on the PPI pairs in database STRING, we constructed an interaction network of proteins encoded by the DEGs (Fig. 1). The top 10 proteins with connection degree ≥60 were CDK1, BUB1, CDC20, NCAPG, NDC80, CDCA8, MAD2L1, CCNB1, CCNA2 and BIRC5, respectively. Strikingly, all the 10 DEGs were up-regulated. A module consisting of 55 proteins were further identified, including those encoded by DEGs CYP2B6, ACAA1, BHMT, ADH1A, ADH1B, ADH4 and ALDH2 (Fig. 2).
Please cite this article in press as: H. Yan, et al., Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.01.019
G Model
ARTICLE IN PRESS
PRP-51734; No. of Pages 6
H. Yan et al. / Pathology – Research and Practice xxx (2017) xxx–xxx
3
Table 1 The top five enriched GO terms. GO ID
GO Term
DEG counts
P value
Enriched by up-regulated genes
GO:0000070 GO:0000226 GO:0000278 GO:0000280 GO:0000819
mitotic sister chromatid segregation microtubule cytoskeleton organization mitotic cell cycle nuclear division sister chromatid segregation
20 50 133 79 21
0 0 0 0 0
Enriched by down-regulated genes
GO:0006082 GO:0006520 GO:0006629 GO:0006631 GO:0006805
organic acid metabolic process cellular amino acid metabolic process lipid metabolic process fatty acid metabolic process xenobiotic metabolic process
130 62 103 46 36
0 0 0 0 0
GO: Gene Ontology; DEGs: differentially expressed genes; GO ID: Gene Ontology Identification..
Table 2 The top five enriched KEGG pathways. KEGG pathway
DEG counts
P value
Enriched by up-regulated DEGs
Cell cycle DNA replication Oocyte meiosis Pyrimidine metabolism Mismatch repair
124 36 112 99 23
4.44E-16 2.31E-12 6.07E-06 2.42E-04 2.67E-04
Enriched by down-regulated DEGs
Metabolic pathways Fatty acid metabolism Retinol metabolism Caffeine metabolism Tryptophan metabolism
128 18 19 7 14
0 1.54E-13 4.14E-11 4.72E-10 3.00E-09
KEGG: Kyoto Encyclopedia of Gene and Genomes; DEGs: differentially expressed genes.
Table 3 The functional statistics of DEGs between cancer and control samples. Transcription factors (15)
Tumor associated genes (33) Oncogenes
Tumor Suppressor Gene
other
Down
SMAD6, RORA, NR4A2, NR3C2, NR1I3, NR1I2, NKX3-1, NFIC, LHX2, ID1, FOSB, ETS2, EGR1, CUX2, AR
FYN, ETS2, CSF1R, CCND1
NR4A2, NKX3-1, MCC, LHX2, FNDC5
Up
none
none
ZFP36, ZBTB16, TGFBR3, TFPI2, STEAP3, SEC14L2, RND3,PTPRD, PER1, NGFR, MT1G, MST1, MFSD2A, MAD1L1, IGFBP4, IGFBP3, HPGD, GNMT, GJB2, GADD45G,FAT4, EGR1, DIRAS3, CBFA2T3 none
none
DEGs: differentially expressed genes; HCC: hepatocellular carcinoma.
3.5. Validation of hub proteins The differential expressions of the hub genes encoding the hub proteins identified above, namely CDK1, CCNA2, CCNB1, BUB1, MAD2L1, CDC20, CYP2B6, ACAA1, BHMT and ALDH2, were further validated by differential expression analysis using the HCC gene expression data from the TCGA database. And it was found that the differential expressions of these hub genes were consistent when using data downloaded from the two different databases (Table 4).
4. Discussion In the present study, we identified 1114 DEGs between the tumor samples and control samples. GO functional and KEGG pathway enrichment analysis showed that the up-regulated DEGs were mainly related to cell division, while the down-regulated DEGs were mainly to material metabolism. Afterwards, a PPI network was constructed and the hub proteins with higher connection degrees were identified, such as CDK1, BUB1, CDC20, NCAPG, NDC80, CDCA8, MAD2L1, CCNB1, CCNA2 and BIRC510, and those with higher degrees in the module were also identified such as CYP2B6, ACAA1, BHMT and ALDH2. The abnormal expression of
Table 4 Results of the differential expressions analyses of the hub genes using data downloaded from the Gene Expression Omnibus(GEO) database and The Cancer Genome Atlas (TCGA) database. Gene symbol
CDC20 BUB1 CDK1 CCNA2 CCNB1 MAD2L1 ALDH2 BHMT CYP2B6
Log2 FC
P value
GEO
TCGA
GEO
TCGA
2.384369 1.477166327 2.405377529 1.770570154 3.007841563 2.098264872 −1.363923507 −2.567023168 −1.857965386
4.430958 4.117578 3.550964 3.432285 3.190918 2.164226 −1.971130 −3.262380 −3.848740
5.17E-05 0.002845 8.20E-06 0.000113 2.28E-05 4.13E-05 0.000262 0.001374 6.31E-05
7.43E-48 5.05E-46 7.71E-46 1.14E-42 2.01E-45 9.65E-30 6.75E-29 1.87E-13 4.41E-24
genes encoding these hub genes were validated by using another batch of gene expression data downloaded from TCGA database. 4.1. Uncontrolled cell division and overexpression of cell cycle related genes may contribute to hepatocellular carcinomatosis Functional and pathway enrichment analysis revealed that the upregulated DEGs were mainly enriched in cell cycle and cell
Please cite this article in press as: H. Yan, et al., Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.01.019
G Model PRP-51734; No. of Pages 6 4
ARTICLE IN PRESS H. Yan et al. / Pathology – Research and Practice xxx (2017) xxx–xxx
Fig. 1. PPI network of DEGs between HCC and control samples. Red nodes represent up-regulated DEGs; Green nodes represent down-regulated DEGs. PPI: protein-protein interaction; DEGs: differentially expressed genes; HCC: hepatocellular carcinoma. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
division related GO terms, such as GO terms “mitotic sister chromatid segregation, mitotic cell cycle, nuclear division and sister chromatid segregation”, and KEGG pathways, such as “cell cycle, DNA replication and oocyte meiosis”. These suggest that abnormal cell proliferation may contribute to hepatocellular carcinomatosis, which was in accordance with a previous finding that disturbance in the regulation of cell cycle is one of the factors leading to cancer [31]. The top 10 proteins in the PPI network, which were all upregulated, are also encoded by cell-cycle related genes. CDK1 encodes cyclin-dependent kinase 1, which is critical for the control of G2 -M transition, G1 progression and G1 -S transition in eukaryotic cell [32]. CCNB1 and CCNA2 encode two cyclins cyclin B1 and cyclin A2 respectively. The former binds with cdk1 to form maturation-promoting factor (MPF), which is essential for the control of G2 /M (mitosis) transition. Cyclin A2 is involved in the G1 /S and G2 /M (mitosis) transitions via binging with protein kinases CDK2 and CDK1 respectively [33,34]. Overexpression of CDKs and
cyclins has been reported in various cancers, such as lung cancer, colon cancer and adenomatous tissue carcinomas [31,35], which agrees with the up-regulation of CDK1, CCNA2 and CCNB1 here. The abnormal expression of CDK1 may interfere with the normal G1 /S and G2 /M (mitosis) transitions by interacting with CCNA2 and CCNB1 respectively, so that cells can’t go through interphase successfully, then aberrant cell division occurred and even further developed into cancer cell combined with the induction of other carcinogenic factors. CDC20 encodes cell-division cycle protein 20, which can activate APC/C (anaphase promoting complex/cyclosome), a large 11–13 subunit complex that initiates the metaphase-to-anaphase transition. MAD2L1 encoding the mitotic spindle assembly checkpoint protein MAD2A, is a component of a spindle assembly checkpoint (SAC) complex, which can inhibit the activity of APC/C and the onset of anaphase by sequestering CDC20 until all chromosomes are properly aligned at the metaphase plate [36]. BUB1 encodes mitotic checkpoint serine/threonine-protein kinase, which, once activated, can phosphorylate CDC20 directly
Please cite this article in press as: H. Yan, et al., Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.01.019
G Model PRP-51734; No. of Pages 6
ARTICLE IN PRESS H. Yan et al. / Pathology – Research and Practice xxx (2017) xxx–xxx
5
Fig. 2. Module analysis of PPI network. The depth of color is proportional to |log2 FC| of DEGs. Red nodes represent up-regulated DEGs, and green nodes represent downregulated. Square nodes represent genes with low importance, and circular nodes represent genes with high importance. PPI: protein-protein interaction; DEGs: differentially expressed genes; FC: fold change. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
[37]. The interaction between BUB1, MAD2L1 and CDC20 is critical for the regulation of APC/C activity and guarantee of correct chromosome alignment and separation. Abnormal expression of these genes may lead to the out of control of APC/C, cell cycle progressing into anaphase although chromosomes haven’t been aligned and separated correctly, accordingly leading to genetic changes and cell carcinogenesis incurred. This hypothesis is in accordance with our aforementioned enrichment analysis as chromosome separation related terms or pathways were obtained, such as the terms mitotic sister chromatid segregation and nuclear division. 4.2. Dysfunctional material metabolism (especially secondary metabolism) and down-regulated material metabolism related genes were involved in hepatocellular carcinomatosis A great many of the down-regulated genes were enriched in GO terms and KEGG pathways related to metabolic processes, such as organic acid metabolism, amino acid metabolism, and etc., indicating that abnormal material metabolism is implicated in hepatocellular carcinomatosis. Some of these genes were also observed in the module detected from the PPI network, such as CYP2B6, ACAA1, BHMT, ADH1A, ADH1B, ADH4 and ALDH2. CYP2B6 encodes a member of cytochromes P450 family enzymes, and this enzyme is involved in the metabolization of a
variety of drugs in clinical use [38]. Previously, Hoshida et al. have found that CYP2B6 has a significant association with good prognosis in HCC patients [39]. However, its role in HCC occurrence has not been reported yet. ADH1A, ADH1B and ADH4 belong to the alcohol dehydrogenase (ADH) gene family. ADH1 is responsible for the conversion of ethanol to acetaldehyde [40]. ALDH2 encodes aldehyde dehydrogenase 2, which is involved in the conversion of acetaldehyde to non-toxic acetate [41]. In the PPI network here, CYP2B6 was directly connected to the three ADHs (ADH4, ADH1A and ADH1B), and the three ADHs were directly connected to ALDH2. Thus, it is presumably that CYP2B6 may interact with ALDH2 indirectly through the three ADHs in HCC pathogenesis. BHMT encoding betaine-homocysteine S-methyltransferase, is implicated in methionine synthesis [42], which has been reported to show a decreased expression in HCC tissues [43]. Hence, the downregulation of these genes implies dysfunction of these genes in patients with HCC, which may contribute to the pathogenesis of this disease. Taken together, abnormal expression of cell cycle and material metabolism related genes may contribute to HCC occurrence, such as CDK1, CCNA2, CCNB1, BUB1, MAD2L1, CDC20, CYP2B6, BHMT and ALDH2. The present study is a pilot study. Although our findings have deepened our insight into HCC pathogenesis and may have clinical application potential in the future, they need to be validated by further experimental work.
Please cite this article in press as: H. Yan, et al., Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.01.019
G Model PRP-51734; No. of Pages 6
ARTICLE IN PRESS H. Yan et al. / Pathology – Research and Practice xxx (2017) xxx–xxx
6
Conflict of interest The authors declare that they have no conflicts of interest to disclose. References [1] A. Buzzanco, A. Gomez, E. Rodriguez, B.A. French, B.A. Tillman, S. Chang, E. Ganapathy, S. Junrungsee, A. Zarrinpar, V.G. Agopian, B.V. Naini, S.W. French Jr., S.W. French, Digital quantitation of HCC-associated stem cell markers and protein quality control factors using tissue arrays of human liver sections, Exp. Mol. Pathol. (2014). [2] B.L. Updegraff, K.A. O’Donnell, Stressing the importance of CHOP in liver cancer, PLoS Genet. 9 (2013) e1004045. [3] H.B. El–Serag, K.L. Rudolph, Hepatocellular carcinoma: epidemiology and molecular carcinogenesis, Gastroenterology 132 (2007) 2557–2576. [4] A.P. Venook, C. Papandreou, J. Furuse, L.L. de Guevara, The incidence and epidemiology of hepatocellular carcinoma: a global and regional perspective, Oncologist 15 (2010) 5–13. [5] M.S. Han, T. Barrett, M.A. Brehm, R.J. Davis, Inflammation mediated by JNK in myeloid cells promotes the development of hepatitis and hepatocellular carcinoma, Cell Rep. 15 (2016) 19–26. [6] A. Arzumanyan, H.M. Reis, M.A. Feitelson, Pathogenic mechanisms in HBVand HCV-associated hepatocellular carcinoma, Nat. Rev. Cancer 13 (2013) 123–135. [7] H.R. Rosen, Clinical practice. Chronic hepatitis C infection, N. Engl. J. Med. 364 (2011) 2429–2438. [8] J.A. Marrero, R.J. Fontana, S. Fu, H.S. Conjeevaram, G.L. Su, A.S. Lok, Alcohol, tobacco and obesity are synergistic risk factors for hepatocellular carcinoma, J. Hepatol. 42 (2005) 218–224. [9] S.C. Chuang, C. La Vecchia, P. Boffetta, Liver cancer: descriptive epidemiology and risk factors other than HBV and HCV infection, Cancer Lett. 286 (2009) 9–14. [10] M.C. Yu, J.-M. Yuan, Environmental factors and risk for hepatocellular carcinoma, Gastroenterology 127 (2004) S72–S78. [11] D.P. Milgrom, M.A. Maluccio, L.G. Koniaris, Management of hepatocellular carcinoma (HCC), Curr. Surg. Rep. 4 (2016) 1–8. [12] J.W. Lee, Y.H. Soung, S.Y. Kim, H.W. Lee, W.S. Park, S.W. Nam, S.H. Kim, J.Y. Lee, N.J. Yoo, S.H. Lee, PIK3CA gene is frequently mutated in breast carcinomas and hepatocellular carcinomas, Oncogene 24 (2005) 1477–1480. [13] M. Li, H. Zhao, X. Zhang, L.D. Wood, R.A. Anders, M.A. Choti, T.M. Pawlik, H.D. Daniel, R. Kannangai, G.J. Offerhaus, V.E. Velculescu, L. Wang, S. Zhou, B. Vogelstein, R.H. Hruban, N. Papadopoulos, J. Cai, M.S. Torbenson, K.W. Kinzler, Inactivating mutations of the chromatin remodeling gene ARID2 in hepatocellular carcinoma, Nat. Genet. 43 (2011) 828–829. [14] J. Huang, Q. Deng, Q. Wang, K.Y. Li, J.H. Dai, N. Li, Z.D. Zhu, B. Zhou, X.Y. Liu, R.F. Liu, Q.L. Fei, H. Chen, B. Cai, H.S. Xiao, L.X. Qin, Z.G. Han, Exome sequencing of hepatitis B virus-associated hepatocellular carcinoma, Nat. Genet. 44 (2012) 1117–1121. [15] C. Li, Y. Xiong, Z. Zhong, S. Zhang, Y. Peng, N. Dai, M. Li, T. Ren, L. Gan, D. Wang, Association between a variant in ADAMTS5 and the susceptibility to hepatocellular carcinoma in a Chinese Han population, Cell Biochem. Biophys. 72 (2015) 221–225. [16] S.S. Zeng, T. Yamashita, M. Kondo, K. Nio, T. Hayashi, Y. Hara, Y. Nomura, M. Yoshida, N. Oishi, H. Ikeda, M. Honda, S. Kaneko, The transcription factor SALL4 regulates stemness of EpCAM-positive hepatocellular carcinoma, J. Hepatol. 60 (2014) 127–134. [17] L. Li, A. Zhang, X. Cao, J. Chen, Y. Xia, H. Zhao, A. Shen, General transcription factor IIb overexpression and a potential link to proliferation in human hepatocellular carcinoma, Pathol. Oncol. Res. 19 (2013) 195–203. [18] B. Stefanska, J. Huang, B. Bhattacharyya, M. Suderman, M. Hallett, Z.G. Han, M. Szyf, Definition of the landscape of promoter DNA hypomethylation in liver cancer, Cancer Res. 71 (2011) 5891–5903. [19] R.C. Gentleman, V.J. Carey, D.M. Bates, B. Bolstad, M. Dettling, S. Dudoit, B. Ellis, L. Gautier, Y. Ge, J. Gentry, K. Hornik, T. Hothorn, W. Huber, S. Iacus, R. Irizarry, F. Leisch, C. Li, M. Maechler, A.J. Rossini, G. Sawitzki, C. Smith, G. Smyth, L. Tierney, J.Y. Yang, J. Zhang, Bioconductor: open software development for computational biology and bioinformatics, Genome Biol. 5 (2004) R80. [20] R.A. Irizarry, B. Hobbs, F. Collin, Y.D. Beazer-Barclay, K.J. Antonellis, U. Scherf, T.P. Speed, Exploration, normalization, and summaries of high density oligonucleotide array probe level data, Biostatistics (Oxford, England) 4 (2003) 249–264.
[21] G.K. Smyth, Limma: linear models for microarray data, in: R. Gentleman, V. Carey, W. Huber, R. Irizarry, S. Dudoit (Eds.), Bioinformatics and Computational Biology Solutions Using R and Bioconductor, Springer, New York, 2005, pp. 397–420. [22] M. Ashburner, C.A. Ball, J.A. Blake, D. Botstein, H. Butler, J.M. Cherry, A.P. Davis, K. Dolinski, S.S. Dwight, J.T. Eppig, M.A. Harris, D.P. Hill, L. Issel-Tarver, A. Kasarskis, S. Lewis, J.C. Matese, J.E. Richardson, M. Ringwald, G.M. Rubin, G. Sherlock, Gene ontology: tool for the unification of biology. The gene ontology consortium, Nat. Genet. 25 (2000) 25–29. [23] J. Du, Z. Yuan, Z. Ma, J. Song, X. Xie, Y. Chen, KEGG-PATH: Kyoto encyclopedia of genes and genomes-based pathway analysis using a path analysis model, Mol. Biosyst. 10 (2014) 2441–2447. [24] G. Dennis Jr., B.T. Sherman, D.A. Hosack, J. Yang, W. Gao, H.C. Lane, R.A. Lempicki, DAVID: database for annotation, visualization, and integrated discovery, Genome Biol. 4 (2003), P3. [25] E. Wingender, The TRANSFAC project as an example of framework technology that supports the analysis of genomic regulation, Brief. Bioinform. 9 (2008) 326–332. [26] J.S. Chen, W.S. Hung, H.H. Chan, S.J. Tsai, H.S. Sun, In silico identification of oncogenic potential of fyn-related kinase in hepatocellular carcinoma, Bioinformatics, Bioinformatics (Oxford, England) 29 (2013) 420–427. [27] M. Zhao, J. Sun, Z. Zhao, TSGene: a web resource for tumor suppressor genes, Nucl. Acids Res. 41 (2013) D970–976. [28] A. Franceschini, D. Szklarczyk, S. Frankild, M. Kuhn, M. Simonovic, A. Roth, J. Lin, P. Minguez, P. Bork, C. von Mering, L.J. Jensen, STRING v9.1: protein-protein interaction networks with increased coverage and integration, Nucl. Acids Res. 41 (2013) D808–815. [29] P. Shannon, A. Markiel, O. Ozier, N.S. Baliga, J.T. Wang, D. Ramage, N. Amin, B. Schwikowski, T. Ideker, Cytoscape: a software environment for integrated models of biomolecular interaction networks, Genome Res. 13 (2003) 2498–2504. [30] D. Beisser, G.W. Klau, T. Dandekar, T. Muller, M.T. Dittrich, BioNet: an R-package for the functional analysis of biological networks, Bioinformatics (Oxford, England) 26 (2010) 1129–1130. [31] M.B. Kastan, J. Bartek, Cell-cycle checkpoints and cancer, Nature 432 (2004) 316–323. [32] M. Qiao, P. Shapiro, M. Fosbrink, H. Rus, R. Kumar, A. Passaniti, Cell cycle-dependent phosphorylation of the RUNX2 transcription factor by cdc2 regulates endothelial cell proliferation, J. Biol. Chem. 281 (2006) 7118–7128. [33] N.R. Brown, E.D. Lowe, E. Petri, V. Skamnaki, R. Antrobus, L.N. Johnson, Cyclin B and cyclin A confer different substrate recognition properties on CDK2, Cell Cycle 6 (2007) 1350–1359. [34] E.T. Petri, A. Errico, L. Escobedo, T. Hunt, R. Basavappa, The crystal structure of human cyclin B, Cell Cycle 6 (2007) 1342–1349. [35] G.I. Evan, K.H. Vousden, Proliferation, cell cycle and apoptosis in cancer, Nature 411 (2001) 342–348. [36] K. Wassmann, V. Liberal, R. Benezra, Mad2 phosphorylation regulates its association with Mad1 and the APC/C, EMBO J. 22 (2003) 797–806. [37] Z. Tang, H. Shu, D. Oncel, S. Chen, H. Yu, Phosphorylation of Cdc20 by Bub1 provides a catalytic mechanism for APC/C inhibition by the spindle checkpoint, Mol. Cell. 16 (2004) 387–397. [38] S. Rendic, F.J.D. Carlo, Human cytochrome P450 enzymes: a status report summarizing their reactions, substrates, inducers, and inhibitors, Drug Metab. Rev. 29 (1997) 413–580. [39] Y. Hoshida, A. Villanueva, M. Kobayashi, J. Peix, D.Y. Chiang, A. Camargo, S. Gupta, J. Moore, M.J. Wrobel, J. Lerner, Gene expression in fixed tissues and outcome in hepatocellular carcinoma, N. Engl. J. Med. 359 (2008) 1995–2004. ˜ [40] E.J. Duell, N. Sala, N. Travier, X. Munoz, M.C. Boutron-Ruault, F. Clavel-Chapelon, A. Barricarte, L. Arriola, C. Navarro, E. Sánchez-Cantalejo, Genetic variation in alcohol dehydrogenase (ADH1A, ADH1B, ADH1C, ADH7) and aldehyde dehydrogenase (ALDH2), alcohol consumption and gastric cancer risk in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort, Carcinogenesis (2011) bgr285. [41] V. Vasiliou, A. Pappa, T. Estey, Role of human aldehyde dehydrogenases in endobiotic and xenobiotic metabolism, Drug Metab. Rev. 36 (2004) 279–299. [42] S.S. Szegedi, C.C. Castro, M. Koutmos, T.A. Garrow, Betaine-homocysteine S-methyltransferase-2 is an S-methylmethionine-homocysteine methyltransferase, J. Biol. Chem. 283 (2008) 8939–8945. [43] W. Sun, B. Xing, Y. Sun, X. Du, M. Lu, C. Hao, Z. Lu, W. Mi, S. Wu, H. Wei, Proteome analysis of hepatocellular carcinoma by two-dimensional difference gel electrophoresis novel protein markers in hepatocellular carcinoma tissues, Mol. Cell. Proteom. 6 (2007) 1798–1808.
Please cite this article in press as: H. Yan, et al., Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.01.019
ARTICLE IN PRESS Pathology – Research and Practice xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Pathology – Research and Practice journal homepage: www.elsevier.com/locate/prp
Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence Hongxian Yan a,∗,1 , Zhaohui Li b,1 , Quan Shen a , Qian Wang c , Jianguo Tian a , Qingfeng Jiang a , Linbo Gao d a
Department of Hepatobiliary Surgery, Henan Provincial People’s Hospital, Zhengzhou, Henan 650000, PR China Secondary Department of General Surgery, Luoyang Central Hospital Affiliated to Zhengzhou University, Luoyang, Henan 471003, PR China c Department of Hepatobiliary Surgery, Henan Cancer Hospital, Zhengzhou, Henan 650000, PR China d Laboratory of Molecular and Translational Medicine, West China Institute of Women and Children’s Health, Key Laboratory of Obstetric & Gynecologic and Pediatric Diseases and Birth Defects of Ministry of Education, West China Second University Hospital, Sichuan University, Chengdu, Sichuan 610041, PR China b
a r t i c l e
i n f o
Article history: Received 28 September 2015 Keywords: Hepatocellular carcinoma Differentially expressed genes Functional enrichment analysis Pathway enrichment analysis Protein-protein interaction
a b s t r a c t This study aims to deepen our understanding of the molecular mechanism underlying the occurrence of hepatocellular carcinoma (HCC). We first downloaded a gene expression profile dataset GSE29721 (10 HCC and 10 control samples) from Gene Expression Omnibus database (http://www.ncbi.nlm.nih.gov/ geo/). Differentially expressed genes (DEGs) were identified by the paired t-test using limma package. Pathway and functional enrichment analyses were performed with DAVID tools. Transcription factors were annotated with TRANSFAC database and tumor associated genes (TAGs) were annotated with TAG and TSGene databases. Protein-protein interaction (PPI) network was conducted using STRING online tool and function module was further identified with BioNet package. Totally, 527 up-regulated DEGs and 587 down-regulated DEGs were identified. GO functional and KEGG pathway enrichment analyses showed that the up-regulated DEGs were mainly related to cell division and cell cycle, while the down-regulated DEGs were largely related to material metabolism, especially secondary metabolism. Proteins encoded by DEGs CDK1, BUB1, CDC20, NCAPG, NDC80, CDCA8, MAD2L1, CCNB1, CCNA2 and BIRC5 were hub genes with high degrees in the PPI network; further module analysis detected a subnetwork consisting of 55 proteins, such as CYP2B6, ACAA1, BHMT and ALDH2. Taken together, aberrant expression of cell cycle related genes (e.g., CDK1, CCNA2, CCNB1, BUB1, MAD2L1 and CDC20) and material metabolism related genes (e.g., CYP2B6, ACAA1, BHMT and ALDH2) may contribute to HCC occurrence. © 2017 Elsevier GmbH. All rights reserved.
1. Introduction Hepatocellular carcinoma (HCC) is the most common form of liver cancer accounting for 90% of all primary liver tumors, with an survival of 6–20 months if not intervened [1]. It is the fifth most common cancer and the third most common cause of death from cancer worldwide [2]. Strikingly, most HCC cases (>80%) occur in sub-Saharan Africa and Eastern Asia, and most astonishingly, China has 50% of HCC cases globally [3]. HCC brings heavy financial burden to both individuals and the society and this burden is expected to
∗ Corresponding author at: Department of Hepatobiliary Surgery, Henan Provincial People’s Hospital, Weiwu Road, NO. 7, Zhengzhou City, Henan 650000, PR China. E-mail address: [email protected] (H. Yan). 1 Hongxian Yan and Zhaohui Li contributed equally to this work.
increase in the coming years [4]. Unfortunately, the management of patients with HCC has not dramatically changed [5]. Chronic hepatitis B virus (HBV) and chronic hepatitis C virus (HCV) infection are the primary causes of liver cancer, which account for approx. 75–80% of all HCC cases [6,7]. Environmental factors, such as alcoholic, smoking and aflatoxin exposure are important risk factors for HCC [8–10]. Recently, Milgrom et al. claimed that non-alcoholic steatohepatitis would become the first cause of HCC in America due to the increase of obesity-related liver disease [11]. In addition, genetic alterations have also been implicated in HCC pathogenesis, such as mutations in PIK3CA [12], ARID2 [13], TP53 and ARID1A [14], SNP polymorphism in ADAMTS5 [15], overexpression of SALL4 [16] and TFIIB [17]. However, the mechanisms underlying hepatocellular tumorigenesis and progression remain poorly understood.
http://dx.doi.org/10.1016/j.prp.2017.01.019 0344-0338/© 2017 Elsevier GmbH. All rights reserved.
Please cite this article in press as: H. Yan, et al., Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.01.019
G Model
ARTICLE IN PRESS
PRP-51734; No. of Pages 6
H. Yan et al. / Pathology – Research and Practice xxx (2017) xxx–xxx
2
In the present study, we reanalyzed a gene expression profile dataset GSE29721 deposited in GEO by Stefanska et al. [18] using bioinformatics tools, with the attempt to deepen our understanding of HCC pathogenesis. 2. Materials and methods As the paper did not involve any human or animal’s study, the ethical approval or patients gave informed consent were not required. 2.1. Source of gene expression profile data The gene expression profile dataset of GSE29721 was downloaded from the GEO (Gene Expression Omnibus) database (http://www.ncbi.nlm.nih.gov/geo/), which is annotated based on Affymetrix Human Genome U133 Plus 2.0 Array platform. It was deposited in GEO by Stefanska et al. in June, 2011 [18]. A total of 20 cancerous and normal adjacent normal tissue samples were obtained from patients with HCC, respectively, including 10 samples of normal adjacent tissues (control group) and 10 samples of cancerous tissues (tumor group) were used for this analysis.
page show search=on) [28] database and protein-protein interaction pairs with combined score larger than 0.9 were selected. Then, the PPI network of the DEGs was visualized using Cytoscape software [29]. Only the PPI pairs that have got experimental verification, co-expression analysis or record in database were chosen for the PPI network construction. Finally, the interaction network was analyzed, and the degree of each protein (namely, the number of other proteins that were connected to a certain protein), was calculated. Furthermore, functional modules were further identified from the PPI network using the BioNet package [30]. The proteins with high degree in the PPI network and its module were considered as the hub nodes. FDR (false discovery rate) <0.0005 was set as the threshold for this analysis. 2.6. Validation of hub proteins Finally, the gene expression data of liver hepatocellular carcinoma that were available to the public in The Cancer Genome Atlas (TCGA) database (http://cancergenome.nih.gov/) were also downloaded (upgraded to May 5th, 2016) to validate the expression of the hub genes identified in the PPI network and the module. These data had been preprocessed, which were collected from 374 tumor samples and 50 normal samples.
2.2. Identification of differentially expressed genes The raw data of GSE29721 dataset were preprocessed using AFFY package in R/Bioconductor software [19]. The detailed preprocessing protocols included background correction, quantile normalization and probe summarization using Robust Multi-array Average (RMA) package [20] with defaulted parameters. Then the gene expression matrix was obtained. DEGs between the tumor and control samples were identified by the paired t-test using limma package in R language with cutoffs of p < 0.05 and |log2 FC (fold change)|>1 [21]. Meanwhile, DEGs between the tumor and control samples in the TGCA data were also identified by the paired t-test using the same cutoffs. 2.3. Functional and pathway enrichment analyses of DEGs Both Gene Ontology (GO) [22] enrichment analysis and Kyoto Encyclopedia of Genes and Genomes (KEGG) [23] pathway enrichment analysis were performed to investigate the functions and potential roles of the identified DEGs in the pathogenesis of HCC using the online tool Database for Annotation, Visualization and Integrated Discovery (DAVID, v6.7, https://david.ncifcrf.gov/) [24]. Meanwhile, the enrichment analyses were performed independently in up-regulated and down-regulated DEGs. P value <0.05 was set as the cut-off for enrichment analysis. 2.4. Identification of tumor associated genes With reference to the database TRANSFAC (http://www.generegulation.com/index2), DEGs that were predicted to be transcription factors were identified [25], and the resulting DEGs were finally identified to be transcription factors only when they were also involved in the GO term (Transcription activity). In addition, tumor associated genes (TAG) including oncogenes and tumor suppressor genes were also identified based on the Tumor Suppressor Gene (TSGene) database (http://bioinfo.mc.vanderbilt.edu/ TSGene/) [26,27]. 2.5. Protein-protein interaction (PPI) network construction and module detection DEGs were mapped to the STRING (http://string-db.org/cgi/ input.pl?UserId=OKHJlLJy0nly&sessionId=pJlFY0Zvk3VZ&input
3. Results 3.1. DEGs analysis A total of 1114 DEGs were identified. Among the 1114 DEGs, 527 DEGs were up-regulated, such as CDK1, CCNA2, CCNB1, BUB1, MAD2L1 and CDC20, and 587 DEGs were down-regulated, such as CYP2B6, ACAA1, BHMT, ADH1A, ADH1B, ADH4 and ALDH2. 3.2. Functional and pathway enrichment analyses of DEGs According to the GO functional enrichment analysis, the upregulated DEGs and the downregulated were enriched in 293 and 416 GO terms respectively. The top five GO terms were mainly related to cell mitosis and cell cycle, and the down-regulated GO terms were mostly related to material metabolism (Table 1). According to the KEGG pathway enrichment analysis, the upregulated and downregulated DEGs were enriched in 38 and 11 pathways respectively. The top three pathways enriched by the upregulated genes were related to cell cycle, DNA replication and oocyte meiosis respectively, and the top five pathways enriched by the upregulated genes were all related to metabolism (Table 2). 3.3. Prediction of tumor associated genes Among the downregulated DEGs, 15 genes were transcriptional factors, and 33 genes were TAGs, including 4 oncogenes, 24 tumor suppressor genes and 5 genes with unidentified role in the development of tumor (Table 3). Noticeably, no upregulated genes were identified as TFs or TAGs here. 3.4. PPI network construction and module detection Based on the PPI pairs in database STRING, we constructed an interaction network of proteins encoded by the DEGs (Fig. 1). The top 10 proteins with connection degree ≥60 were CDK1, BUB1, CDC20, NCAPG, NDC80, CDCA8, MAD2L1, CCNB1, CCNA2 and BIRC5, respectively. Strikingly, all the 10 DEGs were up-regulated. A module consisting of 55 proteins were further identified, including those encoded by DEGs CYP2B6, ACAA1, BHMT, ADH1A, ADH1B, ADH4 and ALDH2 (Fig. 2).
Please cite this article in press as: H. Yan, et al., Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.01.019
G Model
ARTICLE IN PRESS
PRP-51734; No. of Pages 6
H. Yan et al. / Pathology – Research and Practice xxx (2017) xxx–xxx
3
Table 1 The top five enriched GO terms. GO ID
GO Term
DEG counts
P value
Enriched by up-regulated genes
GO:0000070 GO:0000226 GO:0000278 GO:0000280 GO:0000819
mitotic sister chromatid segregation microtubule cytoskeleton organization mitotic cell cycle nuclear division sister chromatid segregation
20 50 133 79 21
0 0 0 0 0
Enriched by down-regulated genes
GO:0006082 GO:0006520 GO:0006629 GO:0006631 GO:0006805
organic acid metabolic process cellular amino acid metabolic process lipid metabolic process fatty acid metabolic process xenobiotic metabolic process
130 62 103 46 36
0 0 0 0 0
GO: Gene Ontology; DEGs: differentially expressed genes; GO ID: Gene Ontology Identification..
Table 2 The top five enriched KEGG pathways. KEGG pathway
DEG counts
P value
Enriched by up-regulated DEGs
Cell cycle DNA replication Oocyte meiosis Pyrimidine metabolism Mismatch repair
124 36 112 99 23
4.44E-16 2.31E-12 6.07E-06 2.42E-04 2.67E-04
Enriched by down-regulated DEGs
Metabolic pathways Fatty acid metabolism Retinol metabolism Caffeine metabolism Tryptophan metabolism
128 18 19 7 14
0 1.54E-13 4.14E-11 4.72E-10 3.00E-09
KEGG: Kyoto Encyclopedia of Gene and Genomes; DEGs: differentially expressed genes.
Table 3 The functional statistics of DEGs between cancer and control samples. Transcription factors (15)
Tumor associated genes (33) Oncogenes
Tumor Suppressor Gene
other
Down
SMAD6, RORA, NR4A2, NR3C2, NR1I3, NR1I2, NKX3-1, NFIC, LHX2, ID1, FOSB, ETS2, EGR1, CUX2, AR
FYN, ETS2, CSF1R, CCND1
NR4A2, NKX3-1, MCC, LHX2, FNDC5
Up
none
none
ZFP36, ZBTB16, TGFBR3, TFPI2, STEAP3, SEC14L2, RND3,PTPRD, PER1, NGFR, MT1G, MST1, MFSD2A, MAD1L1, IGFBP4, IGFBP3, HPGD, GNMT, GJB2, GADD45G,FAT4, EGR1, DIRAS3, CBFA2T3 none
none
DEGs: differentially expressed genes; HCC: hepatocellular carcinoma.
3.5. Validation of hub proteins The differential expressions of the hub genes encoding the hub proteins identified above, namely CDK1, CCNA2, CCNB1, BUB1, MAD2L1, CDC20, CYP2B6, ACAA1, BHMT and ALDH2, were further validated by differential expression analysis using the HCC gene expression data from the TCGA database. And it was found that the differential expressions of these hub genes were consistent when using data downloaded from the two different databases (Table 4).
4. Discussion In the present study, we identified 1114 DEGs between the tumor samples and control samples. GO functional and KEGG pathway enrichment analysis showed that the up-regulated DEGs were mainly related to cell division, while the down-regulated DEGs were mainly to material metabolism. Afterwards, a PPI network was constructed and the hub proteins with higher connection degrees were identified, such as CDK1, BUB1, CDC20, NCAPG, NDC80, CDCA8, MAD2L1, CCNB1, CCNA2 and BIRC510, and those with higher degrees in the module were also identified such as CYP2B6, ACAA1, BHMT and ALDH2. The abnormal expression of
Table 4 Results of the differential expressions analyses of the hub genes using data downloaded from the Gene Expression Omnibus(GEO) database and The Cancer Genome Atlas (TCGA) database. Gene symbol
CDC20 BUB1 CDK1 CCNA2 CCNB1 MAD2L1 ALDH2 BHMT CYP2B6
Log2 FC
P value
GEO
TCGA
GEO
TCGA
2.384369 1.477166327 2.405377529 1.770570154 3.007841563 2.098264872 −1.363923507 −2.567023168 −1.857965386
4.430958 4.117578 3.550964 3.432285 3.190918 2.164226 −1.971130 −3.262380 −3.848740
5.17E-05 0.002845 8.20E-06 0.000113 2.28E-05 4.13E-05 0.000262 0.001374 6.31E-05
7.43E-48 5.05E-46 7.71E-46 1.14E-42 2.01E-45 9.65E-30 6.75E-29 1.87E-13 4.41E-24
genes encoding these hub genes were validated by using another batch of gene expression data downloaded from TCGA database. 4.1. Uncontrolled cell division and overexpression of cell cycle related genes may contribute to hepatocellular carcinomatosis Functional and pathway enrichment analysis revealed that the upregulated DEGs were mainly enriched in cell cycle and cell
Please cite this article in press as: H. Yan, et al., Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.01.019
G Model PRP-51734; No. of Pages 6 4
ARTICLE IN PRESS H. Yan et al. / Pathology – Research and Practice xxx (2017) xxx–xxx
Fig. 1. PPI network of DEGs between HCC and control samples. Red nodes represent up-regulated DEGs; Green nodes represent down-regulated DEGs. PPI: protein-protein interaction; DEGs: differentially expressed genes; HCC: hepatocellular carcinoma. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
division related GO terms, such as GO terms “mitotic sister chromatid segregation, mitotic cell cycle, nuclear division and sister chromatid segregation”, and KEGG pathways, such as “cell cycle, DNA replication and oocyte meiosis”. These suggest that abnormal cell proliferation may contribute to hepatocellular carcinomatosis, which was in accordance with a previous finding that disturbance in the regulation of cell cycle is one of the factors leading to cancer [31]. The top 10 proteins in the PPI network, which were all upregulated, are also encoded by cell-cycle related genes. CDK1 encodes cyclin-dependent kinase 1, which is critical for the control of G2 -M transition, G1 progression and G1 -S transition in eukaryotic cell [32]. CCNB1 and CCNA2 encode two cyclins cyclin B1 and cyclin A2 respectively. The former binds with cdk1 to form maturation-promoting factor (MPF), which is essential for the control of G2 /M (mitosis) transition. Cyclin A2 is involved in the G1 /S and G2 /M (mitosis) transitions via binging with protein kinases CDK2 and CDK1 respectively [33,34]. Overexpression of CDKs and
cyclins has been reported in various cancers, such as lung cancer, colon cancer and adenomatous tissue carcinomas [31,35], which agrees with the up-regulation of CDK1, CCNA2 and CCNB1 here. The abnormal expression of CDK1 may interfere with the normal G1 /S and G2 /M (mitosis) transitions by interacting with CCNA2 and CCNB1 respectively, so that cells can’t go through interphase successfully, then aberrant cell division occurred and even further developed into cancer cell combined with the induction of other carcinogenic factors. CDC20 encodes cell-division cycle protein 20, which can activate APC/C (anaphase promoting complex/cyclosome), a large 11–13 subunit complex that initiates the metaphase-to-anaphase transition. MAD2L1 encoding the mitotic spindle assembly checkpoint protein MAD2A, is a component of a spindle assembly checkpoint (SAC) complex, which can inhibit the activity of APC/C and the onset of anaphase by sequestering CDC20 until all chromosomes are properly aligned at the metaphase plate [36]. BUB1 encodes mitotic checkpoint serine/threonine-protein kinase, which, once activated, can phosphorylate CDC20 directly
Please cite this article in press as: H. Yan, et al., Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.01.019
G Model PRP-51734; No. of Pages 6
ARTICLE IN PRESS H. Yan et al. / Pathology – Research and Practice xxx (2017) xxx–xxx
5
Fig. 2. Module analysis of PPI network. The depth of color is proportional to |log2 FC| of DEGs. Red nodes represent up-regulated DEGs, and green nodes represent downregulated. Square nodes represent genes with low importance, and circular nodes represent genes with high importance. PPI: protein-protein interaction; DEGs: differentially expressed genes; FC: fold change. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
[37]. The interaction between BUB1, MAD2L1 and CDC20 is critical for the regulation of APC/C activity and guarantee of correct chromosome alignment and separation. Abnormal expression of these genes may lead to the out of control of APC/C, cell cycle progressing into anaphase although chromosomes haven’t been aligned and separated correctly, accordingly leading to genetic changes and cell carcinogenesis incurred. This hypothesis is in accordance with our aforementioned enrichment analysis as chromosome separation related terms or pathways were obtained, such as the terms mitotic sister chromatid segregation and nuclear division. 4.2. Dysfunctional material metabolism (especially secondary metabolism) and down-regulated material metabolism related genes were involved in hepatocellular carcinomatosis A great many of the down-regulated genes were enriched in GO terms and KEGG pathways related to metabolic processes, such as organic acid metabolism, amino acid metabolism, and etc., indicating that abnormal material metabolism is implicated in hepatocellular carcinomatosis. Some of these genes were also observed in the module detected from the PPI network, such as CYP2B6, ACAA1, BHMT, ADH1A, ADH1B, ADH4 and ALDH2. CYP2B6 encodes a member of cytochromes P450 family enzymes, and this enzyme is involved in the metabolization of a
variety of drugs in clinical use [38]. Previously, Hoshida et al. have found that CYP2B6 has a significant association with good prognosis in HCC patients [39]. However, its role in HCC occurrence has not been reported yet. ADH1A, ADH1B and ADH4 belong to the alcohol dehydrogenase (ADH) gene family. ADH1 is responsible for the conversion of ethanol to acetaldehyde [40]. ALDH2 encodes aldehyde dehydrogenase 2, which is involved in the conversion of acetaldehyde to non-toxic acetate [41]. In the PPI network here, CYP2B6 was directly connected to the three ADHs (ADH4, ADH1A and ADH1B), and the three ADHs were directly connected to ALDH2. Thus, it is presumably that CYP2B6 may interact with ALDH2 indirectly through the three ADHs in HCC pathogenesis. BHMT encoding betaine-homocysteine S-methyltransferase, is implicated in methionine synthesis [42], which has been reported to show a decreased expression in HCC tissues [43]. Hence, the downregulation of these genes implies dysfunction of these genes in patients with HCC, which may contribute to the pathogenesis of this disease. Taken together, abnormal expression of cell cycle and material metabolism related genes may contribute to HCC occurrence, such as CDK1, CCNA2, CCNB1, BUB1, MAD2L1, CDC20, CYP2B6, BHMT and ALDH2. The present study is a pilot study. Although our findings have deepened our insight into HCC pathogenesis and may have clinical application potential in the future, they need to be validated by further experimental work.
Please cite this article in press as: H. Yan, et al., Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.01.019
G Model PRP-51734; No. of Pages 6
ARTICLE IN PRESS H. Yan et al. / Pathology – Research and Practice xxx (2017) xxx–xxx
6
Conflict of interest The authors declare that they have no conflicts of interest to disclose. References [1] A. Buzzanco, A. Gomez, E. Rodriguez, B.A. French, B.A. Tillman, S. Chang, E. Ganapathy, S. Junrungsee, A. Zarrinpar, V.G. Agopian, B.V. Naini, S.W. French Jr., S.W. French, Digital quantitation of HCC-associated stem cell markers and protein quality control factors using tissue arrays of human liver sections, Exp. Mol. Pathol. (2014). [2] B.L. Updegraff, K.A. O’Donnell, Stressing the importance of CHOP in liver cancer, PLoS Genet. 9 (2013) e1004045. [3] H.B. El–Serag, K.L. Rudolph, Hepatocellular carcinoma: epidemiology and molecular carcinogenesis, Gastroenterology 132 (2007) 2557–2576. [4] A.P. Venook, C. Papandreou, J. Furuse, L.L. de Guevara, The incidence and epidemiology of hepatocellular carcinoma: a global and regional perspective, Oncologist 15 (2010) 5–13. [5] M.S. Han, T. Barrett, M.A. Brehm, R.J. Davis, Inflammation mediated by JNK in myeloid cells promotes the development of hepatitis and hepatocellular carcinoma, Cell Rep. 15 (2016) 19–26. [6] A. Arzumanyan, H.M. Reis, M.A. Feitelson, Pathogenic mechanisms in HBVand HCV-associated hepatocellular carcinoma, Nat. Rev. Cancer 13 (2013) 123–135. [7] H.R. Rosen, Clinical practice. Chronic hepatitis C infection, N. Engl. J. Med. 364 (2011) 2429–2438. [8] J.A. Marrero, R.J. Fontana, S. Fu, H.S. Conjeevaram, G.L. Su, A.S. Lok, Alcohol, tobacco and obesity are synergistic risk factors for hepatocellular carcinoma, J. Hepatol. 42 (2005) 218–224. [9] S.C. Chuang, C. La Vecchia, P. Boffetta, Liver cancer: descriptive epidemiology and risk factors other than HBV and HCV infection, Cancer Lett. 286 (2009) 9–14. [10] M.C. Yu, J.-M. Yuan, Environmental factors and risk for hepatocellular carcinoma, Gastroenterology 127 (2004) S72–S78. [11] D.P. Milgrom, M.A. Maluccio, L.G. Koniaris, Management of hepatocellular carcinoma (HCC), Curr. Surg. Rep. 4 (2016) 1–8. [12] J.W. Lee, Y.H. Soung, S.Y. Kim, H.W. Lee, W.S. Park, S.W. Nam, S.H. Kim, J.Y. Lee, N.J. Yoo, S.H. Lee, PIK3CA gene is frequently mutated in breast carcinomas and hepatocellular carcinomas, Oncogene 24 (2005) 1477–1480. [13] M. Li, H. Zhao, X. Zhang, L.D. Wood, R.A. Anders, M.A. Choti, T.M. Pawlik, H.D. Daniel, R. Kannangai, G.J. Offerhaus, V.E. Velculescu, L. Wang, S. Zhou, B. Vogelstein, R.H. Hruban, N. Papadopoulos, J. Cai, M.S. Torbenson, K.W. Kinzler, Inactivating mutations of the chromatin remodeling gene ARID2 in hepatocellular carcinoma, Nat. Genet. 43 (2011) 828–829. [14] J. Huang, Q. Deng, Q. Wang, K.Y. Li, J.H. Dai, N. Li, Z.D. Zhu, B. Zhou, X.Y. Liu, R.F. Liu, Q.L. Fei, H. Chen, B. Cai, H.S. Xiao, L.X. Qin, Z.G. Han, Exome sequencing of hepatitis B virus-associated hepatocellular carcinoma, Nat. Genet. 44 (2012) 1117–1121. [15] C. Li, Y. Xiong, Z. Zhong, S. Zhang, Y. Peng, N. Dai, M. Li, T. Ren, L. Gan, D. Wang, Association between a variant in ADAMTS5 and the susceptibility to hepatocellular carcinoma in a Chinese Han population, Cell Biochem. Biophys. 72 (2015) 221–225. [16] S.S. Zeng, T. Yamashita, M. Kondo, K. Nio, T. Hayashi, Y. Hara, Y. Nomura, M. Yoshida, N. Oishi, H. Ikeda, M. Honda, S. Kaneko, The transcription factor SALL4 regulates stemness of EpCAM-positive hepatocellular carcinoma, J. Hepatol. 60 (2014) 127–134. [17] L. Li, A. Zhang, X. Cao, J. Chen, Y. Xia, H. Zhao, A. Shen, General transcription factor IIb overexpression and a potential link to proliferation in human hepatocellular carcinoma, Pathol. Oncol. Res. 19 (2013) 195–203. [18] B. Stefanska, J. Huang, B. Bhattacharyya, M. Suderman, M. Hallett, Z.G. Han, M. Szyf, Definition of the landscape of promoter DNA hypomethylation in liver cancer, Cancer Res. 71 (2011) 5891–5903. [19] R.C. Gentleman, V.J. Carey, D.M. Bates, B. Bolstad, M. Dettling, S. Dudoit, B. Ellis, L. Gautier, Y. Ge, J. Gentry, K. Hornik, T. Hothorn, W. Huber, S. Iacus, R. Irizarry, F. Leisch, C. Li, M. Maechler, A.J. Rossini, G. Sawitzki, C. Smith, G. Smyth, L. Tierney, J.Y. Yang, J. Zhang, Bioconductor: open software development for computational biology and bioinformatics, Genome Biol. 5 (2004) R80. [20] R.A. Irizarry, B. Hobbs, F. Collin, Y.D. Beazer-Barclay, K.J. Antonellis, U. Scherf, T.P. Speed, Exploration, normalization, and summaries of high density oligonucleotide array probe level data, Biostatistics (Oxford, England) 4 (2003) 249–264.
[21] G.K. Smyth, Limma: linear models for microarray data, in: R. Gentleman, V. Carey, W. Huber, R. Irizarry, S. Dudoit (Eds.), Bioinformatics and Computational Biology Solutions Using R and Bioconductor, Springer, New York, 2005, pp. 397–420. [22] M. Ashburner, C.A. Ball, J.A. Blake, D. Botstein, H. Butler, J.M. Cherry, A.P. Davis, K. Dolinski, S.S. Dwight, J.T. Eppig, M.A. Harris, D.P. Hill, L. Issel-Tarver, A. Kasarskis, S. Lewis, J.C. Matese, J.E. Richardson, M. Ringwald, G.M. Rubin, G. Sherlock, Gene ontology: tool for the unification of biology. The gene ontology consortium, Nat. Genet. 25 (2000) 25–29. [23] J. Du, Z. Yuan, Z. Ma, J. Song, X. Xie, Y. Chen, KEGG-PATH: Kyoto encyclopedia of genes and genomes-based pathway analysis using a path analysis model, Mol. Biosyst. 10 (2014) 2441–2447. [24] G. Dennis Jr., B.T. Sherman, D.A. Hosack, J. Yang, W. Gao, H.C. Lane, R.A. Lempicki, DAVID: database for annotation, visualization, and integrated discovery, Genome Biol. 4 (2003), P3. [25] E. Wingender, The TRANSFAC project as an example of framework technology that supports the analysis of genomic regulation, Brief. Bioinform. 9 (2008) 326–332. [26] J.S. Chen, W.S. Hung, H.H. Chan, S.J. Tsai, H.S. Sun, In silico identification of oncogenic potential of fyn-related kinase in hepatocellular carcinoma, Bioinformatics, Bioinformatics (Oxford, England) 29 (2013) 420–427. [27] M. Zhao, J. Sun, Z. Zhao, TSGene: a web resource for tumor suppressor genes, Nucl. Acids Res. 41 (2013) D970–976. [28] A. Franceschini, D. Szklarczyk, S. Frankild, M. Kuhn, M. Simonovic, A. Roth, J. Lin, P. Minguez, P. Bork, C. von Mering, L.J. Jensen, STRING v9.1: protein-protein interaction networks with increased coverage and integration, Nucl. Acids Res. 41 (2013) D808–815. [29] P. Shannon, A. Markiel, O. Ozier, N.S. Baliga, J.T. Wang, D. Ramage, N. Amin, B. Schwikowski, T. Ideker, Cytoscape: a software environment for integrated models of biomolecular interaction networks, Genome Res. 13 (2003) 2498–2504. [30] D. Beisser, G.W. Klau, T. Dandekar, T. Muller, M.T. Dittrich, BioNet: an R-package for the functional analysis of biological networks, Bioinformatics (Oxford, England) 26 (2010) 1129–1130. [31] M.B. Kastan, J. Bartek, Cell-cycle checkpoints and cancer, Nature 432 (2004) 316–323. [32] M. Qiao, P. Shapiro, M. Fosbrink, H. Rus, R. Kumar, A. Passaniti, Cell cycle-dependent phosphorylation of the RUNX2 transcription factor by cdc2 regulates endothelial cell proliferation, J. Biol. Chem. 281 (2006) 7118–7128. [33] N.R. Brown, E.D. Lowe, E. Petri, V. Skamnaki, R. Antrobus, L.N. Johnson, Cyclin B and cyclin A confer different substrate recognition properties on CDK2, Cell Cycle 6 (2007) 1350–1359. [34] E.T. Petri, A. Errico, L. Escobedo, T. Hunt, R. Basavappa, The crystal structure of human cyclin B, Cell Cycle 6 (2007) 1342–1349. [35] G.I. Evan, K.H. Vousden, Proliferation, cell cycle and apoptosis in cancer, Nature 411 (2001) 342–348. [36] K. Wassmann, V. Liberal, R. Benezra, Mad2 phosphorylation regulates its association with Mad1 and the APC/C, EMBO J. 22 (2003) 797–806. [37] Z. Tang, H. Shu, D. Oncel, S. Chen, H. Yu, Phosphorylation of Cdc20 by Bub1 provides a catalytic mechanism for APC/C inhibition by the spindle checkpoint, Mol. Cell. 16 (2004) 387–397. [38] S. Rendic, F.J.D. Carlo, Human cytochrome P450 enzymes: a status report summarizing their reactions, substrates, inducers, and inhibitors, Drug Metab. Rev. 29 (1997) 413–580. [39] Y. Hoshida, A. Villanueva, M. Kobayashi, J. Peix, D.Y. Chiang, A. Camargo, S. Gupta, J. Moore, M.J. Wrobel, J. Lerner, Gene expression in fixed tissues and outcome in hepatocellular carcinoma, N. Engl. J. Med. 359 (2008) 1995–2004. ˜ [40] E.J. Duell, N. Sala, N. Travier, X. Munoz, M.C. Boutron-Ruault, F. Clavel-Chapelon, A. Barricarte, L. Arriola, C. Navarro, E. Sánchez-Cantalejo, Genetic variation in alcohol dehydrogenase (ADH1A, ADH1B, ADH1C, ADH7) and aldehyde dehydrogenase (ALDH2), alcohol consumption and gastric cancer risk in the European Prospective Investigation into Cancer and Nutrition (EPIC) cohort, Carcinogenesis (2011) bgr285. [41] V. Vasiliou, A. Pappa, T. Estey, Role of human aldehyde dehydrogenases in endobiotic and xenobiotic metabolism, Drug Metab. Rev. 36 (2004) 279–299. [42] S.S. Szegedi, C.C. Castro, M. Koutmos, T.A. Garrow, Betaine-homocysteine S-methyltransferase-2 is an S-methylmethionine-homocysteine methyltransferase, J. Biol. Chem. 283 (2008) 8939–8945. [43] W. Sun, B. Xing, Y. Sun, X. Du, M. Lu, C. Hao, Z. Lu, W. Mi, S. Wu, H. Wei, Proteome analysis of hepatocellular carcinoma by two-dimensional difference gel electrophoresis novel protein markers in hepatocellular carcinoma tissues, Mol. Cell. Proteom. 6 (2007) 1798–1808.
Please cite this article in press as: H. Yan, et al., Aberrant expression of cell cycle and material metabolism related genes contributes to hepatocellular carcinoma occurrence, Pathol. – Res. Pract (2017), http://dx.doi.org/10.1016/j.prp.2017.01.019