Impact of hepatitis C virus heterogeneity on interferon sensitivity: An overview

Abstract

Hepatitis C virus (HCV) is a major cause of liver disease worldwide. HCV is able to evade host defense mechanisms, including both innate and acquired immune responses, to establish persistent infection, which results in a broad spectrum of pathogenicity, such as lipid and glucose metabolism disorders and hepatocellular carcinoma development. The HCV genome is characterized by a high degree of genetic diversity, which can be associated with viral sensitivity or resistance (reflected by different virological responses) to interferon (IFN)-based therapy. In this regard, it is of importance to note that polymorphisms in certain HCV genomic regions have shown a close correlation with treatment outcome. In particular, among the HCV proteins, the core and nonstructural proteins (NS) 5A have been extensively studied for their correlation with responses to IFN-based treatment. This review aims to cover updated information on the impact of major HCV genetic factors, including HCV genotype, mutations in amino acids 70 and 91 of the core protein and sequence heterogeneity in the IFN sensitivity-determining region and IFN/ribavirin resistance-determining region of NS5A, on virological responses to IFN-based therapy.

Key Words: Hepatitis C virus; Interferon; Genotype; Core protein; Nonstructural proteins 5A; Interferon sensitivity-determining region; Interferon/ribavirin resistance-determining region

Core tip: This review aims to cover recent updates on the impact of major hepatitis C virus (HCV) genetic factors, including HCV genotype, mutations in amino acids 70 and 91 of the core protein and sequence heterogeneity in interferon (IFN) sensitivity-determining region and IFN/ribavirin resistance-determining region of Nonstructural proteins 5A, on virological responses to IFN-based therapy.

INTRODUCTION

Since its discovery in 1989[1,2], hepatitis C virus (HCV) has been the subject of intense research and clinical investigations as its major role in human disease has emerged. Globally, HCV is estimated to infect 180 million people, who represent about 3% of the world’s population. HCV is a major cause of chronic liver disease, such as chronic hepatitis, liver cirrhosis and hepatocellular carcinoma (HCC)[3-6]. HCC is the third most common cause of cancer-related mortality worldwide[7]. In particular, HCV infection accounts for 30%-90% of HCC cases in Western Europe, United States and Asia[8]. Although the treatment of HCV infection is available, it is costly and requires long-term medical support and follow-up. Moreover, current therapies are still impractical for a substantial proportion of HCV-infected patients. The development of a protective vaccine remains a distant prospect.

HCV is an enveloped virus with a positive-strand RNA molecule of approximately 9600 nucleotides. HCV lacks a DNA intermediate; thus, it is incapable of integrating into host chromosomal DNA. Despite this, unlike most RNA viruses, HCV is capable of establishing persistent infection. This ability is central to HCV pathogenesis because it allows chronic infection to occur in 60% to 90% of infected individuals, and virtually all clinically significant HCV-related liver damage takes place during the chronic phase of infection[9,10].

The HCV genome encodes a single open reading frame that encodes a large polyprotein of approximately 3000 amino acids (aa). The polyprotein is processed by host cell peptidases and viral proteases to generate three structural (core, E1, E2) and seven non-structural (p7, NS2 to NS5B) proteins[11]. Both ends of the HCV genome contain highly conserved untranslated regions (5’- and 3’-UTRs) that are critical for genome replication and viral protein translation[12,13]. The 5’-UTR contains the HCV internal ribosome entry site (IRES), an RNA structural element that mediates ribosome binding for translation in a cap-independent manner, while the 3’-UTR is required for HCV RNA replication[14-17].

HCV displays a high nucleotide mutation rate that is estimated to be 1.44 × 10^-3 nucleotide changes per site per year over the whole genome[18,19]. This mainly arises from the error-prone nature of its RNA-dependent RNA polymerase, which lacks 3’-to-5’ exonuclease proofreading activity. This has resulted in diversification of HCV into distinct genotypes and subtypes. HCV exists in the host as quasispecies, which are a dynamic distribution of non-identical but closely related genomes[20,21]. This genetic diversity plays a vital role in HCV’s ability to establish persistent infection and to evade the various selective pressures exerted by immune responses and antiviral therapy. Also, different HCV genotypes exhibit different treatment responses and different pathogenicity. Consequently, the impact of sequence heterogeneity within particular regions of the HCV genome, such as the core, E2, NS3 and NS5A, on treatment responses has been a subject of interest for many researchers. In this review, we will discuss the updated information about major HCV genetic factors, including viral genotype and sequence heterogeneity within certain regions of the HCV genome, in particular the core and NS5A regions, that influence the outcome of interferon (IFN)-based therapy.

HCV GENOTYPE AND SUSCEPTIBILITY TO IFN-BASED THERAPY

HCV exhibits genetic variability at several different levels. Most obvious is the genetic divergence of the main genotypes of HCV. Phylogenetic analysis of nucleotide sequences recovered from infected individuals in different geographical regions has classified HCV into seven major genotypes and series of subtypes[22]. HCV genotypes differ in 30%-35% of nucleotide sites over the whole genome, while the subtypes within a given genotype differ in 20%-25% of their nucleotide sites, with more sequence variability concentrated in such regions as the E1 and E2 glycoproteins, while more sequence conservation is found in the 5’- and 3’-UTR, the core gene and some of the nonstructural protein genes, such as NS3[23].

HCV genotypes differ in three major properties that highlight the importance of genetic diversity among the different HCV genotypes: (1) the prevalence of certain HCV genotype is frequently associated with certain geographical ranges; for example, HCV genotype 1 is prevalent in North America and Japan, genotype 3 is most common on the Indian subcontinent, genotype 4 is the most common genotype in Africa and the Middle East, genotype 5 can be found in South Africa and genotype 6 in Southeast Asia[24]; (2) the pathogenicity of HCV infection varies among the different genotypes; for example, HCV genotype 3 infection is associated with a higher degree of liver steatosis[25-27] and genotype 1 infection associated with a higher risk of HCC development[28,29]; and (3) the response rates to IFN-based therapy vary significantly between the different HCV genotypes[30-34].

To date, IFN represents the backbone of HCV therapeutic options. Pegylated-IFN and ribavirin (PEG-IFN/RBV) combination therapy is the standard of care for the treatment of chronic hepatitis C infection of different HCV genotypes. Recently, new direct-acting antivirals (DAAs) have been introduced to the treatment of HCV genotype 1 infection[35,36]. HCV genotype is an important determinant of both treatment strategy and outcome. HCV genotype 1 and genotype 4 infections need longer (48 wk) treatment period than genotypes 2 and 3 (24 wk); HCV genotypes 1 and 4 are less responsive to PEG-IFN/RBV treatment, with a sustained virological response (SVR) rate hovering around 50%[30,37-41], while the SVR rate in HCV genotype 2 and genotype 3 infections approaches 80%[30,37]. These differences in the SVR rates observed among different HCV genotypes (Figure 1) have suggested that viral genome variability could play a role in determining treatment outcome. However, it is still unclear which genetic element(s) within the HCV genome accounts for the difference in treatment response rates among different HCV genotypes. Interestingly, a series of detailed phylogenetic analyses have shown that there is a significant correlation between the relative evolutionary age of HCV genotypes and the response rates to IFN-based therapy[42]. In these analyses, HCV genotype 2 branched first, genotypes 1 and 4 branched last, and genotypes 3, 5 and 6 branched in between. Thus, it has been hypothesized that genotypes that emerged earlier exhibit better treatment outcomes, while newly-emerged genotypes have higher rates of resistance to IFN-based therapy than their ancestors. This correlation might be attributable to selective pressures generated by the host immune system, which the IFN-based therapy relies on to a large extent.

Figure 1 Average values of sustained virological response to pegylated-interferon and ribavirin combination therapy in patients infected with different hepatitis C virus genotypes. Average of sustained virological response (SVR) rates were calculated based on SVR rates achieved in clinical studies cited in references for genotype 1 (GT1)^{[30,31,37,152-154]}, GT2^[32,155-160], GT3^{[32,155-158,160]}, GT4^{[30,38-40,161-163]}, GT5^[164,165] and GT6^[166-172]. In those studies, standard 48-wk treatment regimens were used for GT1 and GT4, and 24-wk treatment regimens were used for GT2 and GT3. Since the treatment regimens for GT5 and GT6 have not been established yet, the treatment duration varied among different studies where it ranged from 24 to 52 wk. HCV: Hepatitis C virus.

MUTATIONS IN GENOMIC REGIONS OF HCV AND RESISTANCE TO IFN-BASED THERAPY

As stated above, the response rates to IFN-based therapy vary between different HCV genotypes. More importantly, the sensitivity to IFN also varies between different HCV isolates of a given genotype and subtype. This has further highlighted a possible role for certain viral genetic factors in determining IFN responsiveness. In this context, genetic variations within two genomic regions, the core and NS5A, have been widely discussed for their correlation with treatment outcome. We will cover updated information regarding the impact of sequence heterogeneity within the core and NS5A regions of different HCV genotypes on treatment outcome, as described below.

CORE PROTEIN

At both the RNA and protein levels, HCV core plays a critical role in the virus life cycle. At the RNA level, a limited region of the first 45 to 60 nucleotides of the core gene, together with the 5’-UTR, is required for IRES function that initiates translation of the HCV polyprotein[14-17,43]. At the protein level, the HCV core is an RNA-binding protein that forms the capsid shell to protect the HCV genome while the virus passes from one cell to another or from one person to another. Furthermore, the core protein plays an important role(s) in the pathogenicity of HCV by modulating host cellular signaling pathways through interaction with a variety of cellular proteins. The core protein has been implicated in IFN resistance, liver steatosis, insulin resistance and hepatocellular carcinoma[44-49].

The HCV core protein shares high homology among different HCV genotypes. There are, however, certain polymorphisms, which are closely associated with the clinical outcome of IFN-based therapy. Akuta et al[50] first reported that aa mutations at positions 70 and 91 of the HCV genotype 1b (HCV-1b) core protein were associated with virological responses to PEG-IFN/RBV treatment. They found a significant correlation between core mutations at aa 70 (Arg⁷⁰ to Gln⁷⁰/His⁷⁰) and/or aa 91 (Leu⁹¹ to Met⁹¹) and poor treatment response. Gln⁷⁰/His⁷⁰ and/or Met⁹¹ were found in 100% of ultimate resistance cases, who tested positive for HCV RNA at the end of 48-wk PEG-IFN/RBV treatment, but in only 42% of responders. Subsequently, several clinical studies were conducted by the same investigators and others, including our group, on Japanese patients infected with HCV-1b to follow up this observation. Most of these studies corroborate the initial observation that identified the aa mutation at position 70 of the core protein as a negative predictive marker for resistance to PEG-IFN/RBV treatment[51-60]. The polymorphism at aa 70 is also a useful determinant for the virologic response to extended 72-wk PEG-IFN/RBV treatment[61]. However, it is worth noting that although the significance of the point mutation at aa 70 (Gln⁷⁰) in prediction of poor treatment response is consistent among most clinical studies, the possible significance of the mutation at aa 91 is contradictory.

Since the discovery of their importance in IFN resistance, mutations at aa 70 and 91 have been the subject of intensive investigation for different aspects of chronic HCV infection including disease progression and pathogenesis. In HCV-1b infection, Gln⁷⁰/His⁷⁰ and/or Met⁹¹ were significantly associated with severe insulin resistance[62], the severity of liver disease, elevated gamma-glutamyl transpeptidase (γ-GTP) levels, low platelet count and low albumin levels[63]. More importantly, Gln⁷⁰/His⁷⁰ and/or Met⁹¹ (non-wild core) are significantly associated with an increased risk of HCC development[64-69]. Notably, Gln⁷⁰ is the only HCV point mutation associated with both an increased risk of HCC and IFN treatment failure in multiple studies. Because increased HCC risk and IFN treatment failure are associated with the same viral point mutation, it is possible that both adverse outcomes result from disruption of the same cellular pathway, specifically, the IFN signaling pathway that is involved in both anti-proliferative and antiviral functions.

It is important to note that the majority of the studies that demonstrated the importance of aa 70 and 91 mutations of the core protein were carried out on Japanese patients infected with HCV-1b. This raised two important questions: (1) is the significance of these viral mutations commonly observed among other HCV genotypes/subtypes; and (2) is the significance of these viral mutations commonly observed among other host ethnicities? To answer the first question, the correlation between the core protein mutations and IFN treatment outcome was investigated in non-HCV-1b infections, such as HCV-1a, -2a, -2b and -4a[70-76]. There was no significant correlation between these mutations and IFN treatment outcome in non-HCV-1b infections, where sequence patterns were quite conserved at these positions. The probability with different aa residues appear at positions 70 and 91 of the core protein of different HCV genotypes is shown in Figure 2. In contrast, point mutations at positions 4 and 110 of the core protein of HCV-2a were significantly associated with responses to PEG-IFN/RBV treatment[70,71]. As for the second question regarding host ethnicity, to the best of our knowledge, there are only three independent studies that were carried out on American[77], Swedish[76] and Saudi patients[75] infected with HCV-1b to investigate the possible correlation between mutations at aa 70 and 91 and IFN treatment outcome. In those studies, the point mutation at aa 70, but not at aa 91, was significantly associated with PEG-IFN/RBV treatment outcome. Overall, the results thus far obtained suggest that the mutation at codon 70 of the HCV core protein can be used as a predictive marker of IFN treatment outcome in only HCV-1b (and possibly HCV-5a) infection, regardless of host ethnicity.

Figure 2 Probability of different aa residues at positions 70 (A) and 91 (B) of the core protein of different hepatitis C virus subtypes. The probability was estimated based on the core protein sequences available in hepatitis C virus (HCV) database (http://hcv.lanl.gov/content/index). The length of an aa symbol-letter represents its probability for a given HCV subtype. R: Arginine; Q: Glutamine; H: Histidine; L: Leucine; C: Cysteine; M: Methionine.

As observed with the standard PEG-IFN/RBV combination therapy, Gln⁷⁰ of the core protein is also significantly correlated with poor response to the recently approved triple therapy of PEG-IFN/RBV/protease inhibitor (telaprevir) for HCV-1b infection[78-81]. Interestingly, Gln⁷⁰ showed significant linkage disequilibrium with minor genotypes (T/G and G/G) of the rs8099917 single nucleotide polymorphism (SNP) near the IL28B gene, which has recently been proposed as the strongest host genetic factor that is associated with poor response to PEG-IFN/RBV combination therapy in HCV genotype 1 infection[78,80,82].

Despite accumulated clinical evidence that strongly supports the correlation between HCV-1b core protein mutations and responses to IFN-based treatment, the molecular mechanism underlying this correlation is still obscure. Three experimental attempts aimed primarily to investigate this issue; however these studies produced conflicting results. In two of the studies, there was no significant difference in IFN resistance between the wild-type core protein and mutants at aa 70 and 91[83,84], while in the third study, there was a strong association between the core protein polymorphisms and IFN resistance via IL-6-induced and SOCS3-mediated suppression of IFN signaling[85]. These contradictions may, in part, be attributable to the different experimental models used by each study. Furthermore, Eng et al[86] identified a novel family of HCV core isotypes, referred to as minicores, which contain the C-terminal portion of the classical core protein, but lack the N-terminal portion. Interestingly, the N-termini of two major minicore proteins are at or near aa 70 and 91, and importantly, mutations in aa 70 and 91 regulate the expression levels of 70 and 91 minicores. Accordingly, it was hypothesized that these clinically important mutations at aa 70 and 91 of the core protein alter HCV function through altered expression levels of minicores, which might lead to IFN resistance. Further investigations using biologically relevant experimental models are needed to elucidate the molecular mechanism(s) underlying the role for the core protein mutations in HCV pathogenesis.

NS5A

The NS5A protein has generated a wide range of interest in HCV research because of its ability to modulate host cell functions, including responses to IFN. NS5A is multifunctional phosphoprotein that is found in a basally phosphorylated form of 56 kDa and a hyperphosphorylated form of 58 kDa[87,88]. NS5A modulates HCV replication through interaction with other viral proteins and certain host proteins to form the HCV replication complex. Moreover, NS5A has been implicated in various forms of viral pathogenesis through interactions with a wide variety of cellular proteins. Thus, NS5A clearly plays multiple roles in mediating viral replication, host-cell interactions, and viral pathogenesis[89]. NS5A is most extensively studied among all the HCV proteins for its relationship with IFN responsiveness. We review recent information regarding the clinical implication of NS5A sequence heterogeneity in predicting treatment outcome of IFN-based therapy.

IFN SENSITIVITY-DETERMINING REGION OF NS5A

Initially, in the era of IFN monotherapy, Enomoto et al[90,91] gained a key insight into the clinical influence of NS5A sequence heterogeneity on responses to IFN treatment. In their study, they amplified an NS5A fragment by RT-PCR from sera of HCV-infected patients, determined their sequences at both the nucleotide and amino acid levels, and compared them with a standard (reference) sequence of a given HCV subtype (Figure 3) to determine the number of amino acid substitutions in a sample of a given patient. By comparing the data obtained from patients who successfully responded to IFN therapy and those who did not, the authors identified a region called the IFN sensitivity-determining region (ISDR) spanning from aa 2209 to 2248 of NS5A of HCV-1b, whose sequence heterogeneity was closely associated with IFN treatment outcome. They found that Japanese patients infected with HCV-1b isolates having four or more mutations in the ISDR (ISDR ≥ 4) compared to that of the HCV-1b prototype (HCV-J) successfully responded to IFN therapy whereas patients infected with viral isolates having ISDR ≤ 3 were non-responders. Since this discovery, the ISDR has been the subject of intense clinical and experimental investigations. Although subsequent studies conducted in Japan were consistent with the initial ISDR report[92-95], results obtained from studies conducted in Europe and North America were quite controversial[96-99]. This might be explained in part by the fact that HCV isolates in western countries have lower overall degree of sequence heterogeneity, particularly in the ISDR, than HCV isolates circulating in Japan; the prevalence of HCV isolates with ISDR ≥ 4 is lower in western countries than in Japan. Host genetic differences between western and Japanese populations and the difference in treatment regimens may also contribute, at least partially, to the apparent discrepancy. Despite this, a meta-analysis conducted by Pascu et al[100] on 1230 ISDR sequences obtained from European and Japanese patients infected with HCV-1b clearly confirmed the importance of the ISDR in determining IFN treatment outcome. In this connection, it should be noted that, in the majority of studies dealing with the ISDR heterogeneity, nested RT-PCR followed by direct sequencing of the amplified fragments, without subcloning each sequence of quasispecies, was adopted to obtain the ISDR sequences, which, in general, identifies the most dominant population of quasispecies but may miss minor populations of quasispecies present in the patients.

Figure 3 Sequence alignment of interferon sensitivity-determining region of major hepatitis C virus subtypes. References of aligned sequences are: hepatitis C virus (HCV)-1b, Enomoto et al^[90]; HCV-1a, AF009606; HCV-2a and -2b, Murakami et al^[104]; HCV-3a, GU814263; HCV-4a, El-Shamy et al^[73]; HCV-5a, AF064490; HCV-6a, DQ480512. ISDR: Interferon sensitivity-determining region.

The predictive value of ISDR initially described in the era of IFN monotherapy continues to be significant also in the era of PEG-IFN/RBV therapy[58,59,101,102]. However, the original criterion of ISDR ≥ 4 to predict SVR was substituted by ISDR ≥ 2. This might result from the selective impact of IFN monotherapy, whereby the prevalence of sensitive isolates with ISDR ≥ 4 was decreased while that of HCV isolates of ISDR ≤ 3 was increased. Subsequently, those IFN monotherapy-resistant isolates with ISDR ≥ 2 were selected by PEG-IFN/RBV as sensitive isolates and those with ISDR ≤ 1 as resistant ones[52,58,102,103]. As for the different genotypes, the degree of sequence heterogeneity in ISDR was significantly correlated with SVR rate in Japanese patients infected with HCV-2a, the second most prevalent genotype in Japan[71,104,105]. In contrast, ISDR sequence heterogeneity did not associate with treatment outcome in patients infected with HCV-2b, -3a or -4a[73,104,106]. Whether the value of ISDR varies with different HCV genotypes and host ethnicity needs further investigation.

The molecular mechanism of ISDR-mediated IFN resistance is still unclear. Some studies have revealed that NS5A binds to and suppresses the function of the IFN-induced double-stranded RNA-activated protein kinase (PKR)[107-109]. PKR is known to inhibit viral replication by inhibiting viral protein synthesis through phosphorylation of eukaryotic initiation factor (eIF)-2. The PKR-binding domain (PKR-BD) of NS5A, spanning from aa 2209 to 2274, consists of the ISDR and its downstream region of 26 aa. The NS5A-PKR interaction was shown to be weakened by the ISDR mutations observed with IFN-sensitive HCV isolates, which would result in weaker suppression of PKR activity. In this context, a significant correlation between sequence variation in PKR-BD and IFN responsiveness was also reported[110-112]. On the other hand, we and other investigators have proposed that NS5A may also play a role(s) in IFN-resistance in an ISDR-independent manner. We have reported that an N-terminal portion of NS5A (aa 1-148) that lacks the ISDR and PKR-BD physically interacts with and inhibits the antiviral activity of 2’,5’-oligoadenylate synthetase in cultured cells[113]. It has also been demonstrated in vitro that NS5A induces the expression of IL-8 at both the mRNA and protein levels. IL-8 is known to inhibit IFN-α signaling pathway. In agreement with this experimental observation, clinical data were reported that IL-8 levels in pretreatment sera were significantly higher in non-responders than in SVR patients[114,115].

IFN/RBV RESISTANCE-DETERMINING REGION OF NS5A

Despite the controversy about the ISDR concept in the era of IFN monotherapy, correlation between NS5A sequence heterogeneity and IFN responsiveness continued to be the subject of interest for many researchers. In this context, by using the same methodology as that for the ISDR analysis, a correlation between sequence heterogeneity of variable region 3 (V3) in a C-terminal portion of NS5A (aa 2356-2379) and responses to IFN-based therapy was also reported[110,112,116-120]. Our group further expanded this observation and gained a key insight when we identified a new region near the C-terminus of NS5A of HCV-1b spanning from aa 2334 to 2379, which we referred to as the IFN/RBV resistance-determining region (IRRDR)[121,122]. The IRRDR consists of the V3 region and its flanking upstream region, pre-V3 (aa 2334-2354). The reference sequences of different HCV subtypes are shown in Figure 4. In our initial study carried out on 47 Japanese patients infected with HCV-1b, we found that a higher degree of IRRDR sequence heterogeneity was closely associated with an early virological response (EVR) at week 16 of the 48-wk PEG-IFN/RBV treatment course[116]. Most importantly, in a follow-up study, the degree of sequence heterogeneity of the IRRDR was significantly associated with SVR. In particular, 16 (76%) of 21 SVR, but only 2 (8%) of 24 Non-SVR, had HCV with 6 or more mutations in the IRRDR (IRRDR ≥ 6). Accordingly, IRRDR ≥ 6 could predict SVR with a positive predictive value of 89% (16/18), while IRRDR ≤ 5 could predict non-SVR with a negative predictive value of 81% (22/27). Thus, we proposed that the degree of sequence heterogeneity of the IRRDR would be a useful positive and negative predictive marker for PEG-IFN/RBV treatment outcomes in HCV-1b infection[121,122]. Following this report, several reports were published by other groups and ours that support the initial idea of the importance of IRRDR in the prediction of treatment outcome[52,53,123]. In a pilot study conducted on patients who underwent liver transplantation, the value of viral genetic factors, including sequence polymorphisms in the core protein, ISDR and IRRDR, in prediction of PEG-IFN/RBV treatment outcome after transplantation was investigated[124]. Interestingly, IRRDR ≥ 6 was the strongest viral genetic factor associated with SVR. Moreover, in a well executed viral genome wide associated study that scanned the whole HCV genome for polymorphisms at certain amino acids or in genomic regions that are significantly associated with PEG-IFN/RBV treatment outcome in HCV-1b infection, a high degree of IRRDR sequence heterogeneity (IRRDR ≥ 4) was selected in multivariate analysis as the strongest factor to predict SVR among various host and viral genetic factors and baseline demographic parameters[125].

Figure 4 Sequence alignment of interferon/ribavirin resistance-determining region and its vicinity of major hepatitis C virus subtypes. The residues in interferon/ribavirin resistance-determining region (IRRDR) are written in bold face letters. References of aligned sequences are: hepatitis C virus (HCV)-1b, El-Shamy et al^[122]; HCV-1a, AF009606; HCV-2a and -2b, Murakami et al^[104]; HCV-3a, GU814263; HCV-4a, El-Shamy et al^[73]; HCV-5a, AF064490; HCV-6a, DQ480512.

Also, it is important to point out that the cutoff number of mutations in IRRDR that is associated with treatment outcome might possibly vary with different geographical regions: In certain geographical regions where HCV isolates with a high degree of sequence heterogeneity are predominant, a higher cutoff number of IRRDR mutations (such as 6 mutations) is applicable[51,53] whereas a lower cutoff number of IRRDR mutations (such as 4 mutations) is better applicable in regions where HCV isolates with a low degree of sequence heterogeneity are predominant[52,125].

While the predictive value of the IRRDR was initially identified in Japanese patients infected with HCV-1b, its predictive value was also confirmed in HCV-2a infection, the second most prevalent HCV genotype in Japan[74]. Furthermore, we investigated for the first time the impact of viral genetics, including the core protein and NS5A polymorphisms, on PEG-IFN/RBV treatment outcome in Egyptian patients with HCV genotype 4 infection. The result clearly demonstrated that the degree of sequence heterogeneity in the IRRDR was the only viral factor that was significantly associated with PEG-IFN/RBV treatment outcome[73]. Therefore, we proposed that the IRRDR would be a useful positive and negative predictive marker for treatment outcome in Egyptian patients infected with HCV genotype 4. Collectively, a series of our studies and others have suggested the significance of IRRDR sequence heterogeneity predicting treatment outcome in different ethnic populations infected with different HCV genotypes.

The clinical correlation between IRRDR sequence heterogeneity and virological responses to IFN-based therapy in HCV infection can be linked to an experimental observation by Tsai et al[126] that an HCV subgenomic RNA replicon containing NS5A of HCV-1b exerted more profound inhibitory effects on IFN activities than the original HCV-2a replicon, and that domain swapping between NS5A sequences of HCV-1b and -2a in the V3 and/or a C-terminal region including the IRRDR resulted in a transfer of their anti-IFN activities. Consistent with this observation, Kumthip et al[127] found that the overexpression of either HCV genotype 1 or genotype 3 NS5A proteins significantly inhibited IFN-induced signaling of IFN-stimulated response element, STAT1 phosphorylation and IFN-stimulated gene expression compared to the respective controls. NS5A of HCV genotype 1 exhibited stronger inhibitory effects on IFN signaling than did that of genotype 3. Furthermore, NS5A of HCV genotype 1 bound to STAT1 with a higher affinity compared to genotype 3. Interestingly, domain mapping revealed that the C-terminal region of NS5A, including ISDR and IRRDR, conferred these inhibitory effects on IFN signaling. Whereas IRRDR is among the most variable sequences across the different genotypes and subtypes of HCV[87], its upstream and downstream sequences show a higher degree of sequence conservation (Figure 4). We speculate that whereas the upstream and downstream sequences are conserved to maintain the capacity of NS5A to participate in RNA replication and virion production across all the HCV genotypes, IRRDR sequences have a genotype-dependent or even a strain-dependent modulatory function(s)[128-130]. Therefore, the sequence heterogeneity of the IRRDR and its significant correlation with IFN responsiveness suggest the possibility that the IRRDR is involved, at least partly, in the viral strategy to evade IFN-mediated antiviral host defense mechanisms. The IRRDR sequence heterogeneity also suggests genetic flexibility of this region and, indeed, a short stretch of sequence in a C-terminal portion of NS5A was shown to tolerate sequence insertions and deletions. This means that the short stretch of sequence is not essential for virus replication in cultured cells. It does not exclude the possibility, however, that the same region might play an important role in modulating the interaction with various host systems, including IFN responsiveness. It is also possible that the genetic flexibility of this region, especially the IRRDR, is accompanied by compensatory changes elsewhere in the viral genome and that these compensatory changes affect overall viral fitness and responses to IFN therapy. We hypothesize that the IRRDR functions as evolution-adaptation machinery for HCV to cope with changes in the surrounding environment. For example, when sequence evolution of NS5A was investigated during IFN treatment, most of the evolutionary mutations were accumulated in the C-terminus, including the IRRDR[119,131]. Furthermore, in a recent retrospective study we investigated sequence evolution of the core protein, NS3 and NS5A (IRRDR and ISDR) during the follow-up period from chronic hepatitis to HCC development by comparing the sequences between pre- and post-HCC isolates[69]. The results showed that IRRDR sequences tended to be more polymorphic at the time of HCC occurrence. The frequency of HCV isolates with IRRDR ≥ 6 was significantly higher in patients with HCC than in those without HCC, and also higher in post-HCC isolates than in pre-HCC isolates. This might imply the possibility that HCV utilizes IRRDR evolution to accommodate certain selective pressures encountered during the course of HCC development. Further studies are needed to elucidate the issue.

E2 AND NS3

E2, one of the two envelope proteins, is the viral component that is required for direct contact with the cell-surface receptors[132]. The first 27 amino acids (aa) of E2 were identified as hypervariable region 1 (HVR1) because of its significant sequence variability and have been reported to be an immunodominant target for neutralization antibodies. Sequence variations in this region might contribute to immune evasion and thereby the persistence of viral infection[133-135]. Also, a region of 12 residues between aa 659-670 of E2, designated as PKR/eIF-2α phosphorylation homology domain (PePHD), has been reported to interact and inhibit the antiviral activity of PKR[136,137]. Accordingly, sequence variations within the PePHD were suspected to influence responses to IFN-based therapy. However, data obtained from clinical studies investigating this issue were controversial[103,138-142].

The NS3 region of the HCV genome is less variable compared to E2 and NS5A, but still displays significant sequence diversity[143]. In a previous study, we demonstrated that polymorphisms in the secondary structure of an N-terminal region of NS3 of HCV-1b were associated with different virological responses to PEG-IFN/RBV therapy and proposed that the viral grouping based on the NS3 polymorphism could be used to predict the outcome of the therapy[144]. In addition, we have recently found that a particular combination of point mutations in aa 1082 and 1112 of NS3 (NS3-Tyr¹⁰⁸²/Gln¹¹¹²) is closely associated with HCC development in HCV-1b infection. We have also noticed that a combination of NS3-Tyr¹⁰⁸²/Gln¹¹¹² and core-Gln⁷⁰ is more strongly associated with HCC development than is the mutations of NS3 alone or the core protein alone[69]. Therefore, we propose that NS3-Tyr¹⁰⁸²/Gln¹¹¹² and core-Gln⁷⁰ would be independent predictive markers for development of HCV-1b-associated HCC.

Apart from the IFN responsiveness, NS3 has been an intense focus of attention since the introduction of NS3 protease inhibitors as DAAs for treatment of HCV infection. Mutations in four positions in the NS3 protease domain are known to be associated with resistance or reduced sensitivity to telaprevir; R155K/T/S/M, A156T/V/S, V36A/M and T54A[145-148]. Three mutations, T54A, V170A and A156S, conferred low to moderate levels of resistance to boceprevir while variants with the A156T mutation are highly resistant[145-149]. Deep sequencing analysis using “next-generation” sequencers revealed that those DAA-resistant mutations were present even before the initiation of treatment in patients who did not achieve SVR. Thus, the prevalence of the DAA-resistant variants is determined by their replicative fitness and selective pressure of the DAAs[150,151]. In this connection, deep sequencing analysis is also useful to study the possible importance of a viral factor(s) in disease manifestations, including IFN resistance, especially when the target variant(s) constitutes a minor population in the sample and, therefore, undetectable by ordinary direct sequencing.

CONCLUSION

HCV is an interesting virus to study because of its ability to evade host defense mechanisms, including both innate and acquired immune responses, so as to establish persistent infection, which causes a wide spectrum of pathogenicity, such as lipid and glucose metabolism disorders and HCC development. The HCV genome is characterized by a high degree of genetic diversity that can be associated with the viral sensitivity or resistance (reflected by different virological responses) to IFN-based therapy. In addition to the IL28B SNP as the most important host factor that governs the IFN responsiveness of the patients, a point-mutation at position 70 of the core protein and sequence heterogeneity of the ISDR and IRRDR in NS5A of HCV have significant impact on the outcome of a standard regimen of PEG-IFN/RBV combination therapy. Currently, the HCV therapeutic field is heading towards IFN-free treatment where there are several ongoing clinical trials testing new specific DAAs against HCV. Whether these DAAs can overcome the HCV genetic diversity barrier without the emergence of resistant variants should be carefully monitored and properly assessed. New technologies, such as second and third generations of deep sequencing technologies that are currently available, will open up new doors to further understand the impact of HCV genetics on HCV pathogenesis and treatment responsiveness in more detail.

ACKNOWLEDGMENTS

We thank Adeeb Rahman for editing the manuscript.