Interferon lambda 4 (IFNL4) gene polymorphism is associated with spontaneous clearance of HCV in HIV-1 positive patients

Abstract Approximately one-third of the individuals infected with human immunodeficiency virus type 1 (HIV-1) are co-infected with hepatitis C virus (HCV). Co-infected patients have an increased risk for developing end-stage liver diseases. Variants upstream of the IFNL3 gene have been associated with spontaneous and treatment-induced clearance of HCV infection. Recently, a novel polymorphism was discovered, denoted IFNL4 ΔG > TT (rs368234815), which seems to be a better predictor of spontaneous clearance than the IFNL4 rs12979860 polymorphism. We aimed to determine the prevalence of the IFNL4 ΔG > TT variants and to evaluate the association with spontaneous clearance of HCV infection in Brazilian HIV-1 patients. The IFNL4 ΔG > TT genotypes were analyzed by polymerase chain reaction followed by restriction digestion in 138 HIV-1 positive patients who had an anti-HCV positive result. Spontaneous clearance of HCV was observed in 34 individuals (24.6%). IFNL4 genotype distribution was significantly different between individuals who had spontaneous clearance and chronic HCV patients (p=0.002). The probability of spontaneous clearance of HCV infection for patients with the IFNL4 TT/TT genotype was 3.6 times higher than for patients carrying the IFNL4 ΔG allele (OR=3.63, 95% CI:1.51-8.89, p=0.001). The IFNL4 ΔG > TT polymorphism seems to be better than IFNL4 rs12979860 to predict spontaneous clearance of the HCV in Brazilian HIV-1 positive patients.

Human immunodeficiency virus type 1 (HIV-1) infects approximately 38 million people worldwide. About one-third of these individuals are also co-infected with hepatitis C virus (HCV) because of the shared transmission routes (Soriano et al., 2007;World Health Organization, 2013). Co-infected patients have an increased risk for developing end-stage liver diseases, including cirrhosis and hepatocellular carcinoma .
Genome-wide association studies (GWAS) identified single nucleotide polymorphisms upstream of interferon-l3 gene (IFNL3; formerly known as IL28B), strongly associated with spontaneous and treatment-induced clearance of HCV infection (Ge et al., 2009;Suppiah et al., 2009;Tanaka et al., 2009;Rauch et al., 2010). The IFNL3 gene, located on chromosome 19, encodes interferon-l3 protein, a cytokine which has antiviral properties in vitro and in vivo (Hou et al., 2009) through activation of the JAK-STAT pathway and up-regulation of interferon-stimulated genes (ISGs) (Zhou et al., 2007). Several studies confirmed the significant association between the CC genotype of the rs12979860 polymorphism (located approximately 3Kb upstream of IFNL3) and spontaneous and treatment-induced clearance of HCV infection in both HCV mono-infected and HIV/HCV co-infected patients (Matsuura et al., 2014). However, it was not clear how the genotypes of the rs12979860 polymorphism affect the interferon signaling pathways (Balagopal et al., 2010;Prokunina-Olsson et al., 2013). Recently, the IFNL4 gene was discovered. It is controlled by a dinucleotide polymorphism, denoted IFNL4 DG > TT (rs368234815; c.65_66delAAinsC; p.Glu22AlafsTer25) and located in exon 1 of IFNL4. The IFNL4 DG allele creates a novel gene which encodes the interferon-l4 protein, moderately similar to interferon-l3. The alternative allele IFNL4 TT does not create this protein. IFNL4 DG > TT was identified in high linkage disequilibrium with rs12979860, now more properly called IFNL4 rs12979860 because of its location within intron 1 of IFNL4 (Prokunina-Olsson et al., 2013;O'Brien et al., 2014). IFNL4 DG > TT seems to be a better predictor of HCV clearance than IFNL4 rs12979860, mainly in individuals of African ancestry (Prokunina-Olsson et al., 2013;Bibert et al., 2013;O'Brien et al., 2014). The present study aimed to determine the prevalence of the IFNL4 DG > TT variants and to evaluate the association with spontaneous clearance of HCV infection in Brazilian HIV-1 positive patients.
The study was performed with HIV-1 positive patients from a cross-sectional study conducted previously (July 2008 to January 2009) in a reference outpatient treatment center for HIV testing and AIDS treatment in Canoas (Rio Grande do Sul, Brazil). A total of 580 adult patients (aged ³ 18 years) were consecutively enrolled, and 57 refused to participate in the study. Socio-demographic and clinical variables of the studied sample were described previously (Lunge et al., 2012;Simon et al., 2014). All study participants signed an informed consent form. The study was approved by the Research Ethics Committee of the Universidade Luterana do Brasil (process 139H/2007).
Blood samples were collected by venipuncture in 5 mL tubes, using ethylenediamitetraacetic acid (EDTA) as anticoagulant, and centrifuged for plasma and cell separation. All plasma samples were submitted to anti-HCV and HCV-RNA detection, as described previously (Lunge et al., 2012). The viral genotype was identified in all HCV-RNA positive samples. Spontaneous HCV clearance was defined as HCV seropositivity and negative HCV RNA results (< 50 UI/ml) without any hepatitis C specific treatment (either interferon or interferon and ribavirin).
IFNL4 genotyping was performed in all patients with a positive result for anti-HCV. Total DNA of each clinical sample was purified by a standard silica-based procedure (Boom et al., 1990). IFNL4 genotypes were determined by polymerase chain reaction (PCR) using primers (5'-GCCTGCTGCAGAAGCAGAGAT-3' and R 5'-GCTCCAGCGAGCGGTAGTG-3') described previously (Prokunina-Olsson et al., 2013). The amplification mixture consisted of ultrapure distilled water (DNase and RNase free), 10 mM Tris-HCl, pH 8.5, 50 mM KCl, 0.75 mM MgCl 2 , 0.0625 mM of deoxynucleoside triphosphates, 0.25 mM of each primer, 1 U of Taq DNA polymerase (Cenbiot Enzimas, Brazil) and 20-100 ng of DNA. All reactions were performed with the following cycling parameters: 1 cycle at 94 ºC for 3 min followed by 35 cycles at 94 ºC for 10 s, 61 ºC for 30 s and 72 ºC for 30 s, followed by a final extension step at 72 ºC for 5 min. The amplified DNA was digested for restriction fragment length polymorphism (RFLP) analysis using ApeK I restriction enzyme according the manufacturer instructions (New England Biolabs). Digested fragments were separated by electrophoresis on a 10% polyacrylamide gel stained with silver nitrate. The IFNL4 TT allele presented three fragments (of 92, 32 and 5 bp, respectively), while the IFNL4 DG allele presented four fragments (of 75, 32, 16 and 5 bp, respectively). PCR amplified products of the three IFNL4 genotypes were sequenced to validate the PCR-RFLP assay results.
Data were expressed as mean and standard deviation (± S.D.) or frequency percentage (%).Variables were compared between groups using Student's t-test or the nonparametric Mann-Whitney test for categorical variables, and the chi-square test for qualitative variables. Allele frequencies were obtained by direct counting of alleles. Allele and genotype frequencies were compared between groups using the chi-square test. The linkage disequilibrium between IFNL4 DG > TT and IFNL4 rs12979860 polymorphisms was calculated using the CubeX program (Gaunt et al., 2007). All tests were two-tailed and P < 0.05 values were considered statistically significant.
Allele and genotype frequencies of the IFNL4 DG > TT polymorphism in the HIV-1/HCV co-infected patients are presented in Table 2. The IFNL4 DG allele frequency was 37%. The genotypic frequencies are in Hardy-Weinberg equilibrium for the total sample. A significant difference was observed between HCV chronic infected patients and individuals with spontaneous clearance (p=0.002). The probability of spontaneous resolution of the HCV infection in patients with IFNL4 TT/TT genotype was 3.6 higher than for patients carrying DG allele (OR = 3.63, 95% CI: 1.51-8.89; p=0.001). The distribution of IFNL4 DG > TT genotypes in comparison to IFNL4 rs12979860 genotypes is shown in Table 3. The variants are in strong linkage disequilibrium (D' = 0.883; r 2 = 0.733).
Previous studies demonstrated that the IFNL4 rs12979860 polymorphism was significantly associated with spontaneous clearance and response to therapy in HCV infected patients (Ge et al., 2009;McCarthy et al., 2010). However, recent studies reported that the IFNL4 DG > TT polymorphism has a better prediction value than IFNL4 rs12979860. The IFNL4 DG > TT polymorphism has been proposed as a candidate to provide a causal link between variants nearby IFNL3 or within IFNL4 and spontaneous clearance of the HCV, thus solving some confusing results (Covolo et al., 2014). In the present study we observed that the genotype distribution was significantly different between HCV chronic infection patients and individuals who had spontaneous clearance. There was an association between the IFNL4 TT/TT genotype of the IFNL4 DG > TT polymorphism and spontaneous clearance. In our previous study, the HIV-1/ HCV co-infected group presenting the CC genotype of the IFNL4 rs12979860 polymorphism had a 2.8 times higher probability for spon-376 IFNL4 gene in HIV/HCV patients HIV: human immunodeficiency virus; HCV: hepatitis C virus; HAART: highly active antiretroviral therapy; NA: not applicable a Data are reported as mean and standard deviation (± S.D.) or frequency and percentage (%). b Totals do not coincide due to lack of data from participants in the study (n=15).  taneous clearance than the other genotypes (Lunge et al., 2012). The current results now showed that patients with the IFNL4 TT/TT genotype had a 3.6 times higher probability for spontaneous HCV clearance than patients carrying the DG allele. Our finding is in agreement with other studies that investigated the association of the IFNL4 DG > TT polymorphism with impaired spontaneous or treatment-induced clearance of HCV in HIV-1 infected patients. A cohort of 207 patients treated with interferon was investigated and the IFNL4 DG > TT polymorphism was a better predictor of treatment failure than IFNL4 rs12979860 (Franco et al., 2014). Similar results were found in an analysis of 890 HIV-1/HCV co-infected women. HCV clearance was three-fold higher in black women with the IFNL4 TT/TT genotype than in those with TT/DG or DG/DG genotypes (Aka et al., 2014). However, another study that analyzed 206 HIV/HCV co-infected patients and 162 HCV mono-infected patients found that the IFNL4 DG > TT polymorphism was strongly associated with the response to interferon/ribavirin therapy in mono-infected patients, but not in co-infected ones (Krämer et al., 2013).
Other groups have studied this polymorphism in HCV mono-infected patients only. In a Swiss cohort of 540 HCV patients, the IFNL4 DG > TT polymorphism was a better predictor of HCV clearance than IFNL4 rs12979860 (Bibert et al., 2013). Interestingly, these authors attributed the effect of the IFNL4 DG variant to reduced expression of IFNL3 and interferon-g-inducible protein-10 (IP-10). In Italian patients, the IFNL4 DG > TT polymorphism was significantly associated with a marker for ISG activation (IP-10), but with regard to spontaneous and treatmentinduced clearance, IFNL4 DG > TT had a predictive value similar to IFNL4 rs12979860 (Covolo et al., 2014). Two other studies also reported similar predictive values of IFNL4 DG > TT and IFNL4 rs12979860 polymorphisms (Keshvari et al., 2014;Stättermayer et al., 2014). Nonetheless if, as reported, one accepts the argument that IFNL4 is a functional polymorphism in the process of HCV clearance, then it would make sense to include IFNL4 DG > TT genotyping in clinical decisions (Keshvari et al., 2014).
Studies reported that IFNL4 DG > TT and IFNL4 rs12979860 polymorphisms are in linkage disequilibrium, but this association varies according to ethnic group. The IFNL4 DG allele is completely correlated with the unfavorable IFNL4 rs12979860 T allele in Asians (r 2 = 1.00) and highly so in Europeans (r 2 = 0.92). However, in Africans, this correlation is only moderate (r 2 = 0.71). The IFNL4 DG > TT polymorphism has been highly associated with HCV clearance in individuals of African ancestry, with the IFNL4 DG allele being better than the IFNL4 rs12979860 T allele in predicting impaired clearance and treatment failure in HCV infection (Prokunina-Olsson et al., 2013). The allele frequencies of both polymorphisms also differ accord-ing to ethnic groups. According to the HapMap project (Prokunina-Olsson et al., 2013;Franco et al., 2014), the frequencies for the IFNL4 DG allele are 7% in Asians, 32% in Europeans, and 78% in Africans. The Brazilian population is remarkably heterogeneous, which reflect its history of extensive admixture, mainly from Europeans, Africans and Amerindians (Cavalcante et al., 2011). In this regard, the present study comprised a sample of individuals of mixed ancestry, but the majority (59.4%) of the subjects self-reported skin color as white. The frequency of IFNL4 DG > TT polymorphism in the Brazilian populations is not known yet. Our study found a frequency of 37% for the IFNL4 DG allele, which is associated with impaired HCV clearance. The frequency of the IFNL4 DG allele in our study was similar to that reported in Swiss (38%) (Bibert et al., 2013) and European (32%) (Franco et al., 2014) cohorts.
Certain limitations must be considered in this study. First, the statistical power to detect differences was limited due to small sample size, which prevented us to perform a multivariate analysis. Second, it was not possible to identify the time of infection for the two viruses (HCV and HIV-1) and to clarify the precise relation between HIV-1 infection and HCV clearance.
To investigate factors that influence spontaneous clearance of HCV, as well as clearance induced by IFNbased treatment is important to understand the natural history of infection and establish new insights about treatment strategies. It has been reported that IFNL4 polymorphisms are associated with treatment outcomes based on directacting antivirals (DAAs) (Chu et al., 2012;Zeuzem et al., 2013;Meissner et al., 2014). However, IFN-free DAA treatments allow high cure rates, but come with matching high price tags (The Lancet, 2014). In this sense, IFNL4 genotyping could to be useful tool before and until global access to DAA can be achieved, helping to prioritize DAA treatment.
In conclusion, our findings suggest that the IFNL4 DG > TT polymorphism of the IFNL4 gene seems to be better than IFNL4 rs12979860 to predict spontaneous clearance of the HCV in HIV-1 positive patients.