Introduction

Interferons (IFN) are cytokines playing a major role in the innate immune response to viral infections (Fig. 1). IFN-λ belongs to the type III IFN family and maps on the human chromosome 19.1 IFN-λ induces antiviral activity and suppresses hepatitis C virus (HCV) replication in vivo and in vitro activating the Janus kinase–Stat pathway that inducts several IFN-stimulated genes, which are in turn responsible for the antiviral effect (Fig. 1).2 Both spontaneous and treatment-induced clearance of HCV in adults have been associated with certain genetic polymorphisms in the IFNL genes, such as the rs12979860, the rs368234815 and the rs4803217.3

Fig. 1
figure 1

Classification of interferons and cascade of IFNL4 response following hepatitis C virus infection.

Full size image

A previous study by Ruiz-Extremera and collaborators has shown that the C allele and the C/C genotype of the rs12979860 single nucleotide polymorphism (SNP) were associated with spontaneous clearance of HCV infection in children.4 This association was confirmed by our group in subsequent multiple studies.5,6,7,8 No data are available on the impact of the SNPs rs368234815 and rs4803217 on the natural history of the infection in children.

The aim of the present study was to confirm the role of the SNP rs12979860, and to evaluate the role of the SNPs rs368234815 and rs4803217 in a large cohort of children with vertical acquired chronic HCV infection. Part of the children’s cohort was involved in previous studies by our group on the association between rs12979860 and spontaneous clearance of HCV.5,6,7,8 Allele and genotype frequencies in the cohort of patients described in the present study were compared with ethnically matched controls whose HCV status was unknown and obtained, using the data of the 1000 Genomes Project (1000GP)9 and the International HapMap Project (IHMP).10

Methods

Design of the study

This is a retrospective study. All the children evaluated at the Meyer Children’s University Hospital of Florence between January 2011 to December 2019 during routine follow-up visits for vertical HCV infection were recruited if (1) born to a mother infected with HCV; (2) aged >30 months; (3) were hepatitis B surface antigen and human immunodeficiency virus antibodies negative; (4) were untreated against HCV; and (5) written informed consent for the study was provided by parents or guardians. At the time of the enrolment the following data were retrospectively collected: ethnic origin, source of infection, outcome of the infection (i.e., spontaneous clearance and chronic infection) and genotyping for the SNPs rs12979860, rs4803217 and rs368234815. The study was approved by the local ethical committee.

Definitions

Vertical transmission of HCV: transmission from the HCV-infected mother to the foetus or to the child occurring during pregnancy or in the perinatal period.11 Vertical infection: HCV ribonucleic acid (HCV RNA) detected during the first year of life in at least two serum samples taken at least 3 months apart, and/or positive antibodies against HCV in children >18 months of age.11 Spontaneous clearance of HCV: polymerase chain reactions (PCR) for HCV RNA negative and HCV antibodies positive in at least three blood samples taken 6 months apart in children >30 months of age.12 Chronic HCV infection: PCR for HCV RNA and HCV antibodies positive in at least two serum samples taken at least 6 months apart in children >30 months of age.

Patients

One hundred and sixty-seven children were consecutively enroled of whom 32 (19.2%) spontaneously cleared the virus and 135 (80.8%) had chronic hepatitis C.

Comparison with subjects studied in the 1000GPand IHMP

The rs12979860/rs4803217 and or the rs368234815 SNPs allele and genotype frequencies were examined, and the concordance with the subjects studied in the 1000GP (http://www.1000genomes.org) and in the IHMP (https://www.genome.gov/10001688/international-hapmap-project), respectively, were evaluated.

Laboratory methods

PCR for HCV was performed as previously described.13 A third-generation enzyme-linked immunosorbent assay (Ortho Diagnostic System Inc., Raritan, NJ) was used to study HCV antibodies. Confirmation was obtained by western blot (Innogenetics, Zwijndrecht, Belgium). IFNL rs12979860, rs4803217 and rs368234815 SNPs genotyping was performed using Allelic Discrimination assays (Applied Biosystems), according to manufacturer’s instructions. IFNL rs12979860-T/C genotyping was performed, as previously described.6 IFNL rs4803217-T/G genotyping was performed using Taqman custom-designed primers and probes as follows: forward primer, CTGTGTGTCTGACCCTTCCG; reverse primer, TCCTGGAGGTGAGTTGGATTTAC; probe for allele T, CAATAAATTAAGACAAGTGGCTA (VIC); and probe for allele G, ATAAATTAAGCCAAGTGGCTA (FAM). rs368234815-TT/ΔG was genotyped with Taqman custom-designed primers and probes: forward primer, GCCTGCTGCAGAAGCAGAGAT; reverse primer, GCTCCAGCGAGCGGTAGTG; probe for allele TT, ATCGCAGAAGGCC (VIC); and probe for allele ΔG, ATCGCAGCGGCCC (FAM). Before the PCR reactions, DNA was added with Allelic Discrimination Assay Mix and TaqMan Universal PCR Master Mix to MicroAmp Optical 96-Well Reaction Plates (Applied Biosystems). For the real-time PCR reactions, the 7500 Fast Real-Time PCR System was used. After preheating for 10 min at 95 °C, 45 cycles of 15 s at 95 °C and 1 min at 60 °C followed. The ABI SDS software (Applied Biosystems) was used to analyse the data.

Statistical analysis

The SPSSX package (SPSS, Inc., Chicago, IL) was used to evaluate the data. Age was reported as median and interquartile range. Differences in frequencies were evaluated by chi2 or Fisher’s exact tests using two-tailed p values with a level of significance set at <0.05, and OR and 90% CI were calculated. Hardy–Weinberg equilibrium (HWE) was tested. A p value <0.05 defined a deviation from HWE.

Results

HCV-infected children

The characteristics of the 167 children enroled, according to the outcome of the infection, are described in Table 1. Thirty-two children (19.2%) presented spontaneous clearance of HCV and 135 (80.8%) developed a chronic infection.

Table 1 Characteristics of the 167 children enroled according to the outcome of the infection.
Full size table

Data extracted from the 1000GP and from the IHMP databases

rs12979860 and rs4803217 SNPs

The alleles and genotypes frequencies of 2504 subjects were extracted for the rs12979860 and the rs4803217 SNPs from the 1000GP database (Tables 2 and 3). In this cohort, 107 of 503 Europeans (21.3%) were Italians. The highest frequencies of the C allele and of the C/C genotype were found in East Asians (rs12979860: 92% and 85%, respectively, and rs4803217 92% and 84%, respectively). Italians had a lower frequency of the C allele and of the C/C genotype as compared with the remaining Europeans (rs12979860, C allele 60% and 71.5%, respectively, p = 0.002; C/C genotype: 35% and 51.8%, respectively, p = 0.002; rs4803217, C allele 61% and 71.5%, respectively, p = 0.003; and C/C genotype: 35% and 52%, respectively, p = 0.001).

Table 2 Rs12979860 alleles and genotypes frequencies in the 2504 individuals enroled in the 1000 Genomes Project database.
Full size table
Table 3 Rs4803217 alleles and genotypes frequencies in the 2,504 individuals enroled in the 1000 Genomes Project database.
Full size table

rs368234815

The alleles and genotypes frequencies of 270 subjects were extracted for the rs368234815 SNP from the IHMP database (Table 4). The Asiatic population had the highest frequencies of the TT allele and of the TT/TT genotype (93% and 87%, respectively) and significantly higher than that found in Caucasians (TT allele: 68%, p = 0.001 and TT/TT genotype: 48%, p = 0.0001).

Table 4 Rs3682234815 alleles and genotypes frequencies in the 207 individuals enroled in the International Hap Map Project.
Full size table

Alleles and genotypes frequencies in the cohort of HCV-infected children

The frequencies of the rs12979860, rs368234815 and rs4803217 alleles and genotypes in the HCV cohort according to the outcome of the infection are described in Table 5.

Table 5 Distribution of the rs12979860, rs368234815 and rs4803217 genotypes according to the outcome of the infection.
Full size table

Comparison of the allele frequencies

rs12979860

The frequency of the rs12979860 C allele was greater among children with spontaneous HCV clearance than in those chronically infected (75% versus 56% p = 0.005; OR: 2.3; 90% CI: 1.4–3.9) and to the Italians enroled in the 1000GP database (75% versus 60%, p = 0.03; OR: 1.9; 90% CI: 1.1–3.2), but was similar to that found in subjects of European ancestry (75% versus 69%, p = 0.3). Conversely, the T-allele frequency was greater in children with chronic hepatitis C than in the European controls (44% versus 31%, p = 0.001; OR: 1.7; 90% CI: 1.3–2.2).

rs368234815

The frequency of the rs368234815 TT allele was greater among children with spontaneous HCV clearance than in those chronically infected (75% versus 54%; p = 0.002; OR: 2.5; 90% CI: 1.5–4.3), but was similar to that found in subjects of Caucasian origin (75% versus 68%, p = 0.3). Conversely, the ΔG-allele frequency was greater in children with chronic hepatitis C as compared with Caucasians from the IHMP (46% versus 32%, p = 0.003; OR: 1.8; 90% CI: 1.2–2.5).

rs4803217

The frequency of the C allele of rs4803217 SNP was 48.4% among children with spontaneous clearance of HCV and 39.4% among those with chronic infection (p = 0.19). The frequency of the C allele was lower in children who spontaneously cleared the infection than that observed in the Europeans (48.4% versus 69%, p = 0.001; OR: 0.4; 90% CI: 0.2–0.6) and in Italians from the 1000GP database (48.4% versus 61%, p = 0.080; OR: 0.6; 90% CI: 0.3–0.9%). Conversely, the A-allele frequency was significantly greater in children with chronic hepatitis C as compared with Europeans (60.6% versus 31%, p = 0.001 OR: 3.4; 90% CI: 2.7–4.3%) and with Italians (60.6% versus 39%, p = 0.001 OR: 2.3; 90% CI: 1.7–3.2%).

Comparison of genotype frequencies

Genotypes were in HWE in the children with HCV infection (p = 0.63), for Rs12979860 (p = 0.6), rs368234815 (p = 0.59) and Rs4803217 (p = 0.58) genotypes, in Europeans and Italians from the 1000GP database and in Caucasians from the IHMP database (Tables 2–4).

Rs12979860

Children with the C/C genotype had higher clearance rate than children with the C/T and T/T genotype combined (p = 0.01; OR: 2.6; 90% CI: 1.3–5), and with the T/T genotype (p = 0.007; OR: 10; 90% CI: 1.81–60.8). The clearance rate of HCV was similar between the C/T and T/T, and between the C/C and the C/T genotypes. The frequency of the C/C genotype of the children with spontaneous clearance of HCV was similar to that of Europeans (53.1% and 48%, respectively, p = 0.5), but greater to that of Italians from the 1000GP database (53.1% versus 35%, p = 0.05; OR: 2.1; 90% CI: 1.1–4.1).

Rs368234815

The clearance rate of HCV was similar between the TT/TT and ΔG/TT genotypes. Children with the TT/TT genotype had higher clearance than those with ΔG/ΔG (p = 0.004; OR: 12.6; 90% CI: 2.2–72.3), and with the ΔG/ΔG and ΔG/TT genotypes taken together (p = 0.01; OR: 2.8; 90% CI: 1.4–5.5). The frequency of the TT/TT genotype was similar in children with spontaneous clearance of HCV and Caucasians from the IHMP (53.1% and 48%, respectively, p = 0.6).

Rs4803217

No C/C genotype was found in the HCV cohort. Children with the A/C genotype had higher clearance than those with A/A (p = 0.01; OR: 8.3; 90% CI: 1.5–45.9). The frequency of the A/C genotype was greater in children with spontaneous clearance of HCV (96.9%) as compared to Europeans (42%, p = 0.02; OR: 7.3; 90% CI: 1.3–40) and Italians from the 1000GP database (52%, p = 0.02; OR: 7.7; 90% CI: 1.3–44.2).

Discussion

This study demonstrates that the SNPs of IFNL3 and IFNL4 influence the spontaneous viral clearance of vertical acquired HCV infection. In particular, it confirms that the genotypes rs12979860 C/C and, for the first time, demonstrates that the rs368234815 TT/TT of the IFNL4 and the genotype rs4803217 A/C of the IFNL3 are associated with a significantly higher probability of spontaneous clearance of the infection compared to the subjects with alternative alleles and genotypes.

The SNP rs12979860 (C/T) is located in the first intron of the novel human IFNL4 gene.14 Previous genome-wide association studies in adults have shown that the rs12979860 CC genotype was associated with higher spontaneous and treatment induced clearance of the virus.15,16 The rs12979860 is located 367 base pairs from rs368234815 (TT/ΔG; previously known as ss469415590) a functional dinucleotide variant located in the first exon of IFNL4.17,18 The IFNL4 rs368234815-ΔG allele creates an open reading frame for the IFN-λ4 protein, whereas the alternative rs368234815-TT allele eliminates the IFN-λ4 protein.17 The rs368234815-ΔG and the -TT alleles are negatively and positively associated with spontaneous clearance of HCV, respectively.3 Finally, the rs4803217 SNP is located in the 3′ untranslated region of IFNL3. Even though data on rs4803217 SNP and HCV clearance are limited, it is reported that rs4803217-C allele is associated with a decreased degradation of IFN-λ3 mRNA and with a favourable outcome of the infection.19 The present study, confirming previous results in adults, supports the hypothesis that both IFNL3 and IFNL4 promote spontaneous clearance of HCV, regardless of patient’s age.

The value of the results of the present study was confirmed by the comparisons of alleles and genotypes frequencies among children with chronic infection, spontaneous clearance and, where available, with ethnically matched controls. In the context of genome-wide association studies, it is important to select the proper reference population that is usually the one geographically closer.20,21 If the genetic structure is not is not properly evaluated, the associations found can be spurious as demonstrated in the present article. In the present study, for example, according to the 1000GP database, for the rs12979860 SNP the C allele was more common in East-Asians when compared with the Europeans and, among these, with Italians. While for the rs4803217 A/C genotype, the correlation with spontaneous clearance was confirmed by the comparison of the frequency observed in the cohort of children with spontaneous clearance with the genotype frequencies of both Europeans and Italians extracted from the 1000GP database, for rs12979860 C/C genotype, the correlation was confirmed only when the frequency observed in the cohort of children with spontaneous clearance was compared with that of Italians selected as reference. For rs368234815 SNP, the data extracted from the IHMP database were limited to the Caucasian population with no specific data from southern Europe or Italy. This is the possible explanation why the highest prevalence of the TT/TT genotype in children who presented spontaneous clearance of the infection could not be confirmed by the comparison, with the cohort of matched controls extracted by the IHMP database. It is well known that within Europe there is a major axis of genetic variation from North to South21 that is confirmed by the results obtained for the rs12979860 SNP according to the 1000GP database.

Spontaneous clearance of HCV has been demonstrated to be dependent on both the viral genotype, as it is more common for genotype 322 and on the host’s immune system (Table A, provided as Supplementary file).23 Children presenting spontaneous clearance of HCV following vertical transmission of the infection have higher blood aminotransferase levels in their first year of life than those who develop chronic infection.24 Since the values of transaminases are an expression of immune-mediated hepatic damage, it is possible to hypothesize that the outcome of the infection transmitted vertically is strictly dependent on the action of the immune system. Previous studies have shown that the number of natural killer (NK) cells and their differentiation phenotype influence the outcome of HCV infection in children after vertical transmission. The cytolytic function of NK cells characterized phenotypically by the presence of CD56 (dim), in particular, contributes to the elimination of HCV in children with spontaneous clearance.23 The role of IFNL4 and of IFNL3 has been shown to be fundamental in the response to pharmacologic therapy with pegylated IFN and in spontaneous viral clearance in adults.25

In the present study, the IFNL4 rs12979860 C allele and the C/C genotype were more common in children with spontaneous HCV clearance than in those with chronic infection. The results confirm previous studies performed in adults25 and children.4,5 Subjects with the protective rs12979860 C/C genotype have higher serum type III IFN levels, which has potent antiviral properties.26 Furthermore, patients with CC genotype have higher levels of inhibitory receptors on the surface of their NK cells and fewer lymphocytes that expose the ligand for tumour-induced necrosis factor-α apoptosis.27 The association between SNPs of rs12979860 of the IFNL4 gene and spontaneous and treatment-induced clearance of HCV were already supported by paediatric data4,5, but the present study adds new insights evaluating the impact of the IFNL4 rs368234815 and of the IFNL3 rs4803217 polymorphisms on the natural history of the infection.

The IFNL4 rs368234815 TT/TT genotype was found to be associated with HCV clearance, while the ∆G allele was more frequent in children with chronic infection. Although IFN-λ4 can be generated only by individuals who carry the IFNL4 rs368234815 ΔG allele, the present study supports that this variant is associated with the failure to clear HCV infection, as previously demonstrated in adults.28,29 The paradoxical functions of IFN-λ4, which appears to induce antiviral activity yet impair effective clearance of HCV,30 has been correlated recently with the possible preactivation of the Janus kinase–Stat signalling that leads to impaired HCV clearance by both IFN-α and IFN-λ.31

The present study shows that the the IFNL3 rs4803217 A/C genotype is associated with spontaneous clearance of HCV infection in Italian children. It should be noted that surprisingly, and without any clear explanation, the C/C genotype was not identified in any of the patients enroled although it was described in 34% of the Italians enroled in the 1000GP database. It has been demonstrated that the Rs4803217 C allele promotes spontaneous clearance dictating an increasing transcript stability, when compared to the A allele.19 Furthermore, in vitro studies demonstrated the expression of IFNL3 associated with the C allele is higher.32 Although the C allele was not associated with spontaneous clearance in our data, a clear correlation between the presence of the A allele and of the AA genotype in children with chronic infection was demonstrated.

In conclusion, this study provides insights in the immunological mechanism of the natural history of HCV infection in children.