Genetic variants of programmed cell death 1 are associated with HBV infection and liver disease progression

The inhibitory effects of programmed cell death 1/programmed cell death ligand 1 (PD-1/PD-L1) modulates T-cell depletion. T-cell depletion is one of the key mechanisms of hepatitis B virus (HBV) persistence, in particular liver disease progression and the development of hepatocellular carcinoma (HCC). This case–control study aimed to understand the significance of PD-1 polymorphisms (PD-1.5 and PD-1.9) association with HBV infection risk and HBV-induced liver disease progression. Genotyping of PD-1.5 and PD-1.9 variants was performed by direct Sanger sequencing in 682 HBV-infected patients including chronic hepatitis (CHB, n = 193), liver cirrhosis (LC, n = 183), hepatocellular carcinoma (HCC, n = 306) and 283 healthy controls (HC). To analyze the association of PD-1 variants with liver disease progression, a binary logistic regression, adjusted for age and gender, was performed using different genetic models. The PD-1.9 T allele and PD-1.9 TT genotype are significantly associated with increased risk of LC, HCC, and LC + HCC. The frequencies of PD-1.5 TT genotype and PD-1.5 T allele are significantly higher in HCC compared to LC patients. The haplotype CT (PD-1.5 C and PD-1.9 T) was significantly associated with increased risk of LC, HCC, and LC + HCC. In addition, the TC (PD-1.5 T and PD-1.9 C) haplotype was associated with the risk of HCC compared to non-HCC. The PD-1.5 CC, PD-1.9 TT, genotype, and the CC (PD-1.5 C and PD-1.9) haplotype are associated with unfavorable laboratory parameters in chronic hepatitis B patients. PD-1.5 and PD1.9 are useful prognostic predictors for HBV infection risk and liver disease progression.

www.nature.com/scientificreports/ by the activation of the inhibitory checkpoint PD-1/PD-L1 (Programmed cell death 1/Programmed cell death ligand 1) signaling pathway [14][15][16][17][18][19][20] . PD-1 is a transmembrane protein and an immunoinhibitory receptor, that are largely expressed on the surface of T cells, B cells, natural killer T cells and activated monocytes [21][22][23] . The PD-1 immunomodulatory ligands (PD-L1 and PD-L2) maintain immune homeostasis, but can also attenuate T-cell proliferation and regulate cytokine responses 24 . The phenomenon of persistent overexpression of PD-1 and its ligands is a common feature in chronic infections and cancer [25][26][27] , and blocking inhibitory signal transduction has led to increased T-cell function, and thus improved clinical outcomes. Blocking the PD-1/PD-L1 signaling with anti PD-1 or anti-PD-L1 molecules in combination with antiviral therapies has led to better immunological and clinical responses in HBV infections 28,29 . Clinical studies have also used expression of PD-1 and CTLA-4 as prognostic indicators, in evaluating antiviral HBV therapy responses 30 .
The investigated PD-1 variants are located in the 5′ upstream regions of the PD-1 locus (Genbank ID: 5133). The variant PD-1.5 (rs2227981) is associated with cancer susceptibility 46 ; while the variant PD-1.9 (rs2227982) is associated with risk of HBV infection and clinical outcome 47 . In this context, we conducted a case-control study with the aim to determine the association of PD1 (PD-1.5 and PD-1.9) variants with HBV infection risk, liver disease progression and clinical outcome in the Vietnamese population.

Materials and methods
Ethics statement. This study was performed in accordance with the relevant national guidelines and regulations. Informed written consent was obtained from all participants after explanation of the study at the time of sampling. Informed consent was obtained from a parent and/or legal guardian for minors. The study was approved by the institutional review board of the 108 Institute of Clinical Medical and Pharmaceutical Sciences, Hanoi, Vietnam.
Patients and liver specimens. A total of 682 unrelated Vietnamese HBV-infected patients were randomly recruited in a case-control design at 108 Institute of Clinical Medical and Pharmaceutical Sciences, Hanoi, Vietnam, between 2012 and 2015. Patients were assigned to subgroups of disease based on clinical manifestations, liver function tests, imaging modalities and histological confirmation for all HCC cases. Subgroups were chronic hepatitis (CHB, n = 193), liver cirrhosis (LC, n = 183), hepatocellular carcinoma (HCC, n = 306). The diagnostic criteria for the CHB patients and the HBV-related LC were previously described 48 . The HBV-related HCC group was characterized as patients infected with HBV and was diagnosed based on the AASLD practice guideline 49 . None of these HBV-infected patients had evidence of chronic comorbidities/conditions such as: autoimmune diseases, alcoholic liver disease, type 2 diabetes, cancers, addiction to smoking and alcohol. All patients were confirmed negative for anti-HCV and anti-HIV by ELISA assays. Laboratory parameters including HBV-DNA loads and liver function tests including alanine transaminase (ALT), aspartate transaminase (AST), total bilirubin and direct bilirubin, albumin, prothrombin were obtained from the patients' medical records. In addition, 283 healthy individuals (HC) were collected and assigned as the control group. These individuals were healthy individuals who presented themselves to the hospital for a routine medical check-up. Hematological and biochemical parameters were normal and serological tests were negative for hepatitis C virus and human immune deficiency virus. Finally, plasma and blood cells were separated and frozen at − 80 °C until use. PD-1 genotyping. Genomic DNA was isolated from 200 µl of whole blood using a DNA isolation kit (Qiagen, Hilden, Germany), following manufacturer's instructions. The process of PD-1 genotyping was followed the study protocol as described 50 . The amplicon containing the variants PD-1.5 (rs2227981) and PD-1.9 (rs2227982) was amplified by PCR (PCR1) using the primer pairs PD-1.5/9_F: 5′-GCA AGA ATG CCA GGG ACA TTT CAG AG-3′ and PD-1.5/9_R: 5′-TGC CTG GTG CAG GTG CAG-3′. PCR amplification was performed out in 25 μl reaction volumes containing: 1 × PCR buffer, 0.2 mM dNTPs, 1 mM MgCl 2 , 0.15 mM of each primer, 1 unit of Taq polymerase and 50 ng of genomic DNA. Cycling conditions: denaturation at 95 °C for 5 min, followed by 35 cycles of three-step cycling with denaturation at 94 °C for 40 s, annealing at 66 °C (PCR1) and 60 °C (PCR2) for 40 s, and extension at 72 °C for 45 s and a final extension at 72 °C for 7 min. PCR products were purified using the Exo-SAP-IT PCR product cleanup reagent (Affymetrix Santa Clara, USA) 5 µl of purified PCR products were used as templates. Sequencing was performed using the BigDye terminator v.1.1 cycle sequencing kit (Applied Biosystems, Foster City, CA, USA) on an ABI 3130XL DNA sequencer according to the manufacturer's instructions.
Statistical analysis. All statistical analysis was performed using R version 3.1.2 (http:// www.r-proje ct. org). Statistical analysis plan was tailored as described previously 50 . In brief, genotype and allelic frequencies www.nature.com/scientificreports/ were determined by simple gene counting and the haplotype frequency was estimated using the expectationmaximum algorithm method implemented in the Arlequin v.3.5.2.2. The deviations from Hardy-Weinberg equilibrium were calculated for each group. We used a binary logistic regression adjusted for age and gender to analyze association of PD-1 variants with HBV-related liver diseases applying for different genetic models and adjusted odds ratios (aOR) with 95% confidence intervals (CI) were calculated. Chi-square tests were used to test for significant differences of categorical variables and Mann-Whitney-Wilcoxon and Kruskal-Wallis tests were applied to compare quantitative variables between groups. Significance was set at a value of p < 0.05.

Baseline characteristics of HBV patients. The baseline profiles of 682 HBV-infected patients and 283
healthy controls (HC) are described in Table 1 and Fig. 1. Most of the HBV and HC patients are male (88% and 63%, respectively) and the HCs were younger than HBV patients. The ALT, AST liver enzymes were significantly higher in CHB patients than other subgroups (p < 0.0001). Albumin and prothrombin levels, red blood cells and platelets were lower in LC patients than in other patient groups (p < 0.0001). HCC patients had higher AFP levels than CHB and LC patients (p < 0.001).

PD-1 haplotypes and HBV-related liver diseases.
A total of four haplotypes were reconstructed from both PD-1.5 and PD-1.9 variants and the frequencies were presented in the Table 4. No significant association in the distribution of haplotypes in HC and HBV and patient subgroups was observed. The analysis within HBV patient subgroups showed that the haplotype CT (PD-1.5 C and PD-1.9 T) was frequently observed in patients

PD-1 genetic variant and laboratory parameters.
The PD-1.5 and PD-1.9 genetic variants were associated with distinct biochemical and laboratory parameters (Figs. 2 and 3). The total and direct bilirubin levels were higher in patients with genotype PD-1.5 CC compared to those with genotypes CT/TT (p = 0.016 and p = 0.05, respectively). In addition, prothrombin levels were significantly decreased among those with PD-1.5 CC genotypes (p = 0.00036). Equally other blood cell counts such as platelets, white and red blood cells were also significantly decreased among those carrying PD-1.5 CC genotypes (p = 0.042, 0.018 0.056 respectively). In the PD-1.9 variant, patients with the genotype TT had lower platelet, red blood cell, prothrombin and albumin levels compared to the CC/CT genotype (p = 0.009, 0.0019, 0.00025 and 0.047, respectively). In contrast, total bilirubin levels were significantly higher in the genotype TT than in the genotype CC/CT. There was no significant correlation between PD-1.9 and other laboratory parameters such as liver enzymes, HBV DNA and AFP. We also analyzed the association between constructed haplotypes and laboratory parameters (Fig. 4). We observed that patients with the TC haplotype had higher prothrombin and platelet levels than other CC and CT haplotypes (p = 0.0041, 0.0398, respectively). The TC haplotype was significantly associated with lower levels of total bilirubin compared to CC haplotypes (p = 0.026).  www.nature.com/scientificreports/

Discussion
PD-1 is a member of the immunoglobulin superfamily and its cytoplasmic domain contain immunoreceptor tyrosine inhibitory motif associated with inhibitory signaling. The interaction between PD-1 and PD-L1 is known to contribute primarily to the depletion of T cells and thus involved in carcinogenesis. Several variants in the PD-1 gene have been identified to associate with various infectious and autoimmune diseases, and cancers. In this case-control study, we investigated on the PD-1.5 (+ 7785T/C) and PD-1.9 (+ 7625C/T) variants in Vietnamese HBV-infected patients and controls. Two variants PD-1.5 and PD-1.9 were observed to influence the clinical outcome of HBV, in particular progression of liver disease. PD-1.5 is a synonymous substitution, whereas PD-1.9 is a non-synonymous substitution (valine to alanine) that modulates structural and functional properties of PD-1. Several studies have investigated on these two variants with different human malignancies 31,32 , autoimmune disorders 51-53 and infectious diseases, including HBV infection 41,47,54 . PD-1.5 and PD-1.9 variants susceptibility to HBV infection were earlier studied 47,54 , and in particular, the PD-1.9 T allele is considered as a predisposing factor for HBV susceptibility 47 . In addition, the frequency of TT and CT genotypes in HBV patients was significantly higher than healthy controls. Based on the reported results, Huang et al. pointed out that the CC genotype of rs2227982 and the C allele are protective factors in HBV infection 47 . In our study, we also reported that when analyzed with the age-and sex-adjusted logistic regression model, the frequency of CT genotype was significantly higher in HBV patients than in controls [co-dominant (CT:CC) model: OR = 1.4 (1.01-1.98), P = 0.042]. However, another study found no association between this variant and HBV infection 55 . For the PD-1.5 polymorphism, we found no significant association with HBV infection. Our conclusions were in contrast to a previous study, which found an association between this polymorphism and susceptibility to chronic HBV infection in the Chinese population 54 . In addition, this polymorphism was also reported to be associated with HCV infection 56 .
Few studies suggest that PD-1 as a convincing genetic indicator for different cancers in different world populations, however results were contradictory and ambiguous [31][32][33][34][35]46 . To date, comprehensive meta-analyses have been conducted to investigate the association of PD-1 polymorphisms, notably PD-1.5 and PD-1.9, with risk of cancer 32,46,57,58 . Concerning the PD-1.5 polymorphism, Dong et al. showed that the allele T (compared to the C allele) is associated with a reduced risk of cancer in the general population 32 . In addition, two other meta-analyses have shown that the PD-1.5 TT genotype (TT compared to CT/CC) significantly reduces the overall risk of cancer   www.nature.com/scientificreports/  It should be noted that these above meta-analyses included several case-control studies, but no studies have examined the association between PD-1.5 and PD-1.9 polymorphisms and HCC risk. To our knowledge, this is the first study in which we have examined the link between PD-1.5 and PD-1.9 polymorphisms and the progression of liver disease including HCC. We have shown that the PD-1.5 and PD-1.9 polymorphisms are associated with the progression of HBV-related liver diseases. Particularly, the variant PD-1.9 shows a recessively homozygous advantage in disease progression and might be a risk factor for the advanced disease progression including LC and HCC. The results from our study indicate that PD-1.9 polymorphism may influence liver disease progression and HCC development in chronic HBV infection. There is growing evidence that PD-1 overexpression is associated with T-cell dysfunction and exhaustion in chronic HBV infection and HCC development 59-61 .
As a single SNP site could not well represent the influence of PD-1 genetic variants in HBV infection, we analyzed the association of haplotypes constructed from variants PD-1.5 and PD-1.9 with the liver disease progression and outcomes in HBV-infected patients. The study results could show a way by integrating different sites to completely elucidate the potent role of PD-1 genetic variants in the risk of HBV infection and HBV-related clinical presentation. Still more works are needed to be done to examine the clinical significance of PD-1 genetic variants as the host genetic perspective in HBV infection.
This study has two limitations. First, a longitudinal monitoring of HCC patients would be an added advantage in understanding the overall survival rates among patients with HBV-related HCC. Secondly, quantification of PD-1 gene expression in liver tissues of HCC patients may additionally warrant on the relevance these variants in clinical routine. Taken together, this study demonstrates on the role of PD-1.5 and PD-1.9 genetic variants in predicting HBV infection risk and subsequent liver disease progression.