Autophagy-Related 5 Gene rs510432 Polymorphism Is Associated with Hepatocellular Carcinoma in Patients with Chronic Hepatitis B Virus Infection.

ABSTRACT Background: Despite the identification of autophagy-related protein 5 (ATG5) as a molecule involved in the activated autophagy machinery during hepatitis B virus (HBV) infection and hepatocarcinogenesis, the consequences of ATG5 mutation carriage for patients with chronic HBV infection remain unclear. This study examined the association of ATG5 polymorphisms with HBV-related diseases including hepatocellular carcinoma (HCC). Patients and Methods: Two functionally relevant polymorphisms ATG5 rs573775 and rs510432 were genotyped by ligase detection reaction-polymerase chain reaction in 403 patients with chronic HBV infection (171 chronic hepatitis, 119 cirrhosis and 113 HCC) and 196 healthy controls. Univariate and multivariate logistic regression was performed to evaluate factors associated with HCC. Results: The rs573775 genotype and allele frequencies had no significant differences between patients with different clinical diseases. However, HCC patients had significantly higher frequency of rs510432 genotype AA (odds ratio [OR] 2.185, 95% confidence interval [CI] 1.042–4.581, P = 0.037, P value by Bonferroni correction [Pc] = 0.074) and allele A (OR 1.435, 95% CI 1.023–2.013, Pc = 0.036) than chronic hepatitis patients. In multivariate analyses, rs510432 allele A-containing genotypes (AA+GA) were independently associated with cirrhosis in comparison to chronic hepatitis (OR 1.927, 95%CI 1.011–3.017, P = 0.032). The rs510432 genotypes AA+GA were also independently associated with HCC in comparison to chronic hepatitis (OR 2.583, 95% CI 1.025–3.911, P = 0.006) or chronic HBV infection without HCC (OR 2.632, 95% CI 1.067–3.482, P = 0.032). Conclusion: These results indicate that rs510432 genotypes AA+GA are associated with disease progression and HCC risk in chronic HBV infection, providing novel evidence for a role of ATG5 in the pathogenesis of HBV-related HCC. Abbreviations: HBV: hepatitis B virus; HCC hepatocellular carcinoma; TNFSF10: tumor necrosis factor superfamily member 10; ATG5: autophagy-related protein 5; DNA: deoxyribonucleic acid; LDR-PCR: ligase detection reactions-polymerase chain reaction; PCR: polymerase chain reaction; SLE: systemic lupus erythematosus; BD: Behçet’s disease; IL-10: interlukin-10; LPS: lipopolysaccharide; PBMC: peripheral blood mononuclear cells; CWP: coal workers’ pneumoconiosis; TNF-α: tumor necrosis factor-α


Introduction
Chronic hepatitis B virus (HBV) infection is associated with chronic hepatitis, liver cirrhosis and hepatocellular carcinoma (HCC) and remains a severe public health issue worldwide (Schweitzer et al., 2015). The natural course of chronic HBV infection is a complex dynamic process of the interaction between the virus and the host immune responses (European Association for the Study of the Liver, 2017). The pathogenic mechanisms of chronic HBV infection in relation to the various clinical outcomes remain largely to be investigated.
Autophagy is regarded as a cellular process directed at recycling cellular proteins and eliminating intracellular microorganisms. Increasing evidence suggests that autophagy has diverse physiological functions and deficiency in autophagy has been implicated in a variety of diseases including HBV-associated liver diseases. Studies have shown that HBV can induce and enhance the autophagic process and this autophagic response can enhance HBV replication and is involved in the pathogenesis of HBV infection (Sir et al., 2010;Tian et al., 2011;Xie et al., 2018). Autophagy may influence the efficiency priming of virus-specific CD8+ T cells (Uhl et al., 2009) which have been demonstrated to play a pivotal role in both the virus control and liver damage of chronic HBV infection (Maini et al., 2000). Autophagy is also involved in the HBV evasion from tumor necrosis factor superfamily member 10 (TNFSF10)-mediated antiviral immunity (Shin et al., 2016). Dysregulation of autophagy has also been shown to be significantly correlated with HBV infection and HBV-associated HCC (Kunanopparat et al., 2016;Lan et al., 2014).
Autophagy-related protein 5 (ATG5), an essential component of the autophagic machinery encoded by Autophagy-related 5 gene (ATG5), is involved in the crucial phase of autophagy, autophagosome formation (Vij et al., 2016). In terms of HBV and HBV-related diseases, ATG5 is involved in the activated autophagy machinery during HBV infection to enhance HBV replication (Li et al., 2011) and plays an essential role for HBV capsid biogenesis (Döring and Prange, 2015;Döring et al., 2018). Dysregulation of immune response is involved in pathogenesis of cirrhosis and HCC (Elwan et al., 2018). ATG5 appears to block innate antiviral immune responses (Jounai et al., 2007) and its expression is involved in the autophagic influence on the efficiency of priming of virus-specific CD8+ T cells (Uhl et al., 2009). ATG5 mRNA expression was significantly increased in HBV-infected hepatic cells (HepG2.2.15 cells) and its protein levels were increased in tumor liver tissues mainly from HCC patients with HBV infection (Kunanopparat et al., 2016). ATG5 is involved in the decreased apoptosis of HCC cells and increased HCC tumor growth  and its expression shows a significant correlation with HBV infection and overall survival rate of HCC patients (Lan et al., 2014).
In view of the significant involvement of autophagy and ATG5 in HBV replication and HBV-related diseases including HCC, we hypothesize that ATG5 gene polymorphisms, especially the functionally relevant polymorphisms, may be associated with chronic HBV infection and the development of HBV-related liver diseases. A study in Thai population indicated that polymorphisms of the ATG5 gene might be involved in HBV-related HCC (Wisetsathorna et al., 2017). However, the possible associations of ATG5 polymorphisms with chronic HBV infection and HBV-related HCC have not been addressed in other ethnic populations so far. Therefore, the present study determined ATG5 polymorphisms in patients with chronic HBV infection of various liver diseases and analyzed the associations with clinical diseases in Chinese Han population.

Study population
Adults with chronic HBV infection were enrolled from the First Affiliated Hospital of Xi'an Jiaotong University from March 2009 to May 2013. Fasting blood was collected in the morning. Three milliliter coagulation blood was used for serum separation by centrifugation and the serum samples were frozen at −20°C until use and 2 mL of EDTA-anticoagulated whole blood was frozen at −30°C for the extraction of human genomic DNA. The diagnosis of the patients was in accordance with the diagnostic criteria of guidance (Terrault et al. 2018). Other liver diseases (hepatitis A, hepatitis C and hepatitis E, drug-induced liver injury, steatohepatitis, alcoholic hepatitis, autoimmune hepatitis, and Wilson's disease), diseases associated with metabolism disorders (diabetes, acquired immunodeficiency syndrome and hyperthyroidism), and complications with severe cardiovascular and respiratory diseases as well as renal impairment were excluded. Patients under 18 years of age were excluded. A history of antiviral therapy (interferon or nucleos(t)ides) and pregnancy were also excluded in the patients with chronic HBV infection. Individuals who had no history of hepatitis with normal liver function and had no other diseases were included as healthy controls. The study recruited 403 patients with chronic HBV infection (male/ female, 283/120; mean age, 40.37 ± 13.68 [18-78] years) and 196 healthy controls (male/female, 127/69; mean age, 38.58 ± 14.29 [18-76] years). All the subjects are of Chinese Han ethnicity, were permanent residents of Shaanxi Province and had no close kinship. The study was approved by the Ethics Committee of the First Affiliated Hospital of Xi'an Jiaotong University and performed in accordance with the Declaration of Helsinki. All subjects completed the informed consent and voluntarily participated in the study.
Genotyping of rs573775 and rs510432 polymorphisms ATG5 rs573775 (a C/T single-nucleotide variation on human chromosome 6) and rs510432 (an A/G single-nucleotide variation on human chromosome 6) polymorphisms were selected given their potential relevance of function demonstrated by previous studies (López et al., 2013;Martin et al., 2012;Shao et al., 2017;Zheng et al., 2015).
Human genomic DNA was extracted by commercial DNA Extraction Kit (Beijing Tiangen Biochemical Technology Co., Ltd., Beijing, China). Genotyping of rs573775 and rs510432 was performed using high-temperature ligase detection reaction-polymerase chain reaction (PCR).
The primers were designed using oligo6.0 and primer5.0 software and synthesized by Shanghai Biotechnology Co., Ltd (Shanghai, China). The polymorphic regions of the ATG5 gene were amplified by a multiplex PCR with specific primers (Supplementary Table 1) which produced a 285 and 314 bp product, respectively.
For ligation reaction, the primers for rs573775 and rs510432 were shown in Supplementary Table 1. The ligation reaction was carried out in a volume of 10 µL containing 1 μL of 10× ligation buffer (TaKaRa Bio, Dalian, China), 0.2 μL of high temperature ligase (New England Biolabs, Ipswich, MA, USA), 0.4 μL (1 µM) of 5ʹ ligation primer mixture (TaKaRa Bio, Dalian, China), 0.4 μL (2 µM) of 3ʹ ligation primer mixture (TaKaRa Bio, Dalian, China), 2 μL of purified multiplex PCR products and 6 μL of ddH 2 O. The ligation program was as follows: 38 cycles of 94°C for 1 min and 56°C for 4 min; and 1 cycle of 4°C forever. A volume of 0.5 μL of the diluted ligation product, 0.5 μL of Liz500 SIZE STANDARD (Applied Biosystems, Carlsbad, CA, USA) and 9 μL of Hi-Di (Applied Biosystems, Carlsbad, CA, USA) were thoroughly mixed, denatured at 95°C for 5 min and then analyzed by ABI 3730 XL sequencer (ABI, USA). The data were analyzed by GeneMapper 4.1 (Applied Biosystems, CA, USA).

Power analysis
At the 5% significance level and a minor allele frequencies (MAFs) of 0.05 and 0.15, there is ≥80% power for detecting a dominant effect with an odds ratio (OR) of 1.2 and 1.5, respectively, for chronic hepatitis, an OR of 1.6 and 1.1, respectively, for liver cirrhosis, and an OR of 1.9 and 1.2, respectively, for HCC (Supplementary Table 2). The Genetic Power Calculator (Purcell et al., 2003) was used for power calculations, setting 'prevalence' of chronic hepatitis to 42.4%, liver cirrhosis to 29.5%, and HCC to 28.0% with the available sample size of patients with chronic HBV infection, D-prime to 1, and type 1 error rate to 0.05.

Statistical analysis
Statistical analysis was performed using SPSS 16.0 software (SPSS Inc. Chicago). The t-test or χ 2 test was used to compare the general data of patients with chronic HBV infection, and healthy controls. Genotype frequencies were tested for Hardy-Weinberg equilibrium. The χ 2 test was used for the difference in genotype and allele frequencies between patients with chronic HBV infection, and healthy controls. Univariate and multivariate analyses with logistic regression were used to analyze independent risk factors associated with HBV-related cirrhosis or HCC. P < 0.05 was considered statistically significant. The Bonferroni procedure was applied to account for multiple testing where appropriate.

Results
Demographics and Hardy-Weinberg equilibrium of the genotypes of ATG5 rs573775 and rs510432 polymorphisms in the study population The study enrolled 403 patients with chronic HBV infection (male/female, 283/120; mean age, 40.37 ± 13.68 [18-78] years) and 196 healthy controls (male/female, 127/ 69; mean age, 38.58 ± 14.29 [18-76] years). There were no significant differences in age and gender between the two groups (both P > 0.05). The genotype distribution of the genotypes of ATG5 rs573775 and rs510432 in patients with chronic HBV infection and healthy controls were all in Hardy-Weinberg equilibrium (Supplementary Table 3).

Genotype and allele frequencies of ATG5 polymorphisms in chronic HBV infection and healthy controls
The genotype and allele frequencies of rs573775s and rs510432 between patients with chronic HBV infection and healthy controls had no significant differences (Supplementary Table 4).

Genotype and allele frequencies of ATG5 polymorphisms in chronic HBV-infected patients with different clinical diseases
The clinical diagnosis of the 403 patients with chronic HBV infection included 171 chronic hepatitis, 119 liver cirrhosis, and 113 HCC. Calculation of sample size and power showed that the minimum effect size in which we had ≥80% power at 5% significance level and a MAF of 0.05 and 0.15 for chronic hepatitis, cirrhosis, and HCC was 167, 103, and 100, respectively (Supplementary Table 2).

Associations of ATG5 polymorphisms with cirrhosis and HCC in chronic HBV infection by univariate and multivariate analyses
Univariate and multivariate analyses were performed to identify whether ATG5 polymorphisms were factors associated with cirrhosis or HCC in chronic HBV infection. In comparison with chronic hepatitis, univariate analysis showed that age and albumin level were associated with cirrhosis (Table 4). Multivariate analysis showed that rs510432 genotype and age were independently associated with cirrhosis, with cirrhosis patients having higher rs510432 genotypes allele A-containing genotypes (AA+GA) (OR 1.927, 95% CI 1.011-3.017, P = 0.032, Table 4).
In comparison with cirrhosis, univariate analysis showed that gender, age and albumin were associated with HCC (Table 6). In multivariate analysis, gender and age were independent factors associated with HCC while rs510432 genotype GG was not shown to be an independent factor associated with HCC (OR 0.841, 95% CI 0.376-1.275, P = 0.088, Table 6).
In comparison with all the patients with chronic HBV infection without HCC (chronic hepatitis and cirrhosis), univariate analysis showed that gender, age, HBV DNA, ALT, and albumin were associated with HCC (Table 7). ATG5 rs510432 genotype, together with gender and age, was independently associated with HCC, with HCC patients having higher rs510432 genotypes AA+GA carriers (OR 2.632, 95% CI 1.067-3.482, P = 0.032, Table 7). Data are presented as n (%). OR: Odds ratio; 95% CI: 95% confidence intervals. a P c (P value by Bonferronis' correction) = 0.024. b P c < 0.001. c P c < 0.001.

Discussion
This study investigated the rs573775 and the rs510432 polymorphisms in the ATG5 gene in patients with chronic hepatitis, liver cirrhosis and HCC associated with chronic HBV infection. The genotypes of rs573775 and rs510432 were not shown to have significant differences between patients with chronic HBV infection and healthy controls. However, in relation to clinical diseases, HCC patients had higher rs510432 AA genotype and significantly higher A allele frequencies than chronic hepatitis patients. Patients with chronic HBV infection had significantly lower rs573775-rs510432 haplotypes C-G and T-G but higher haplotype T-A than healthy controls. Chronic hepatitis patients had higher haplotype C-G than HCC patients. The rs510432 AA+GA genotypes were shown to be an independent factor associated with cirrhosis in comparison with chronic hepatitis. The rs510432 AA+GA genotypes were also shown to be an independent factor associated with HCC in comparison with chronic hepatitis or chronic HBV infection without HCC. The insignificance of rs510432 polymorphism in HCC in relation to cirrhosis may be related to the small number of patients with cirrhosis. These results suggested that rs510432 allele A and AA+GA were factors predisposing to the disease progression from chronic hepatitis to the development of cirrhosis and HCC in chronic HBV infection. ATG5 rs573775 polymosphism has been shown to influence systemic lupus erythematosus (SLE) susceptibility in Spanish (López et al., 2013) and Italian populations (Ciccacci et al., 2018). This polymorphism has also been shown to influence the susceptibility to Behçet's disease (BD) in Chinese Han population (Zheng et al., 2015). The allele T of rs573775 was shown to confer a higher risk of developing SLE in Spanish (López et al., 2013) and Italian populations (Ciccacci et al., 2018) but a lower risk of developing BD in Chinese Han population (Zheng et al., 2015). Functionally, the T allele is associated with high interleukin (IL)-10 producer genotype in SLE patients of Spanish population (López et al., 2013) and the TT genotype is associated with increased ATG5 expression by lipopolysaccharide (LPS) stimulated peripheral blood mononuclear cells in Chinese Han BD patients and the level of ATG5 mRNA in active BD patients was significantly increased (Zheng et al., 2015). However, the present study did not observe an association between rs573775 polymorphism and HBV-related liver diseases including HCC in Chinese Han population. This may be a reflection of the disease differences in relation to the susceptibility with genetic background of the host. ATG5 rs510432 polymorphism was found to be associated with the development of asthma in white/Caucasian (Martin et al., 2012), the risk of coal workers' pneumoconiosis in a Chinese population (Yuan et al., 2017a) and the progression and mortality of sepsis in Chinese patients (Shao et al., 2017). The rs510432 was also shown to be associated with response to therapy with anti-tumor necrosis factor (TNF)-α drug adalimumab in Crohn's disease patients (Deželak et al., 2016). Furthermore, rs510432 polymorphism is found to be related to cancers. For example, rs510432 was found to be associated with stage and nonbrisk tumor infiltrating lymphocytes in melanoma (White et al., 2016). Among epidermal growth factor receptormutant patients with advanced lung adenocarcinoma, rs510432 was found to contribute to disease prognosis and to be associated with primary or acquired resistance to gefitinib (Yuan et al., 2017b). A recent study in Thai population indicated that ATG5 rs510432 polymorphism might be involved in HBV-related HCC (Wisetsathorna et al., 2017). The present study showed that rs510432 was associated with the disease progression of chronic HBV infection, especially the development of HBV-related HCC. These findings provide novel information for the role of ATG5 rs510432 polymorphism in disease predisposition of viral infection and carcinogenesis.
The rs510432 is located in the 5ʹ untranslated region, 335 bp upstream of the transcription start site of human ATG5 gene. It has been shown to be functionally relevant and confers significant effect on promotor activity. The rs510432 G allele had higher promotor activity than the A allele and associated with the increased gene expression of ATG5 in asthmatics (Martin et al., 2012) while the rs510432 allele A appeared to be related to lower promoter activity in sepsis (Shao et al., 2017). The rs510432 polymorphism was also indicated to influence the expression levels of ATG5 which was decreased with the severity of sepsis (Shao et al., 2017). The mononuclear cell of rs510432 allele A carriers exhibited decreased levels of ATG5 expression which led to enhanced releases of TNF-α and IL-1β under LPS stimulation in vitro (Shao et al., 2017). ATG5 is involved in HBV-associated HCC (Kunanopparat et al., 2016) and autophagy suppresses tumorigenesis of HBV-associated HCC (Lan et al., 2014). The significant association of rs510432 allele A and AA+GA with HCC found in the present study is consistent with the functional relevance of the genotypes. The lower promoter activity related to rs510432 allele A may lead to the decreased expression of ATG5, compromising the inhibitory effect of ATG5 in autophagy on tumorigenesis in chronic HBV infection and predisposing the development of HCC.
This study was performed in Chinese Han patients with a relatively small sample size. There are other polymorphisms in ATG5 which have not been investigated in this study. Therefore, further studies in large sample sizes of patients and controls and different ethnic populations to examine more ATG5 polymorphisms are needed to confirm the findings in this study and to clarify the contribution of ATG5 polymorphisms to HBVassociated diseases including HCC. Notably, there are other host genetic polymorphisms such as HLA class II alleles which have also been found to be associated with HBV infection (Matei et al., 2018). Therefore, further studies are also needed to investigate the potential coeffects of various host genetic polymorphisms on chronic HBV infection and the development of HBV-related HCC.
In conclusion, this study showed that ATG5 rs573775 appeared to have no significant association with chronic HBV infection and HBV-related liver diseases. However, rs510432 was associated with the disease progression in chronic HBV infection, especially the development of HBV-related HCC, highlighting a likely important role of ATG5 rs510432 in chronic HBV infection and hepatocarcinogenesis. These associations may be critical to understanding the role of autophagy in chronic HBV infection and HBVrelated HCC. Further investigations in large patient populations are warranted to characterize the contribution of ATG5 rs510432 and other polymorphisms to the disease progression of chronic HBV infection and the development of HBV-related HCC as well as the responses to antiviral therapy or to anticancer therapy in HCC patients.