Report

The Outcome of Acute Hepatitis C Predicted by the Evolution of the Viral Quasispecies

See allHide authors and affiliations

Science  14 Apr 2000:
Vol. 288, Issue 5464, pp. 339-344
DOI: 10.1126/science.288.5464.339

Abstract

The mechanisms by which hepatitis C virus (HCV) induces chronic infection in the vast majority of infected individuals are unknown. Sequences within the HCV E1 and E2 envelope genes were analyzed during the acute phase of hepatitis C in 12 patients with different clinical outcomes. Acute resolving hepatitis was associated with relative evolutionary stasis of the heterogeneous viral population (quasispecies), whereas progressing hepatitis correlated with genetic evolution of HCV. Consistent with the hypothesis of selective pressure by the host immune system, the sequence changes occurred almost exclusively within the hypervariable region 1 of the E2 gene and were temporally correlated with antibody seroconversion. These data indicate that the evolutionary dynamics of the HCV quasispecies during the acute phase of hepatitis C predict whether the infection will resolve or become chronic.

HCV infection is an important public health problem worldwide (1) because it is a major cause of chronic hepatitis, cirrhosis, and hepatocellular carcinoma (2). Very rarely, HCV causes fulminant hepatitis (FH), the most severe form of acute hepatitis. Although the infection resolves in 15% of cases, it becomes chronic in up to 85% of infected individuals (3). The clinical course of chronic hepatitis C is highly variable. In about 70% of the patients the disease is mild and stable over several decades, whereas in the remaining 30% it is more rapidly progressive. Prospective studies of hepatitis C have failed to identify any clinical, serologic, or virologic features that predict the outcome of the disease (4).

The mechanisms responsible for the high rate of viral persistence and for the variable clinical course of hepatitis C are unknown, but are thought to represent a complex interplay between viral diversity and host immunity (5). Although HCV infection induces strong cellular and humoral immune responses (6, 7), they are generally insufficient to eradicate the virus or to prevent reinfection (8, 9). Over the past few years, evidence has accumulated to suggest that the genetic variation of HCV within the same individual—specifically, the simultaneous presence of different but closely related viral variants that are commonly defined as quasispecies (10)—may allow the virus to circumvent the immune response, leading to chronic infection. There is very limited information on the early evolution of the viral quasispecies during the acute phase of hepatitis C in patients who have been followed for a sufficient time to determine their long-term clinical outcome (11). Access to a well-defined cohort of prospectively studied patients enrolled in a study of post-transfusion non-A, non-B hepatitis gave us an opportunity to investigate the relationship between the genetic evolution of HCV early in the course of infection and the outcome of the disease.

We studied the number of viral variants, the genetic distance between the different variants (genetic diversity), and the evolution of HCV quasispecies, in parallel with the level of viral replication and the humoral immune response, during the incubation period and the acute phase of hepatitis C. Serial serum samples were obtained from 12 patients with hepatitis C (12) who were selected on the basis of their clinical outcome determined during prospective evaluation ranging up to 20 years (Table 1). Three patients had FH; three had acute, self-limited hepatitis characterized by clearance of serum HCV RNA within 16 weeks after transfusion; and six had acute hepatitis that progressed to chronicity, associated with persistent viremia. Among the latter six patients, the disease was mild and stable for more than 20 years in three (slow progressors), whereas in the remaining three it was severe and rapidly progressive, leading to liver-related death within 5 years of the onset of infection (rapid progressors) (Table 1).

Table 1

Clinical and serologic evaluation of hepatitis C in prospectively followed patients. All numbers in parentheses indicate the week after transfusion. The level of viremia was measured by branched-chain DNA test (19). Antibodies to HCV and to E2 protein were measured as described (21). ND, not determined; NA, not applicable, because the risk factor was unknown and therefore the time of infection could not be ascertained.

View this table:

The number of viral variants and the genetic diversity of the HCV quasispecies were assessed by examining viral sequences spanning the envelope genes (E1 and E2) both within and outside the hypervariable region 1 (HVR1) (13). For all patients with resolving and progressing hepatitis, we studied the first available polymerase chain reaction (PCR)–positive sample (within 2 to 5 weeks of transfusion, mean 3.1 ± 1.2 weeks), one sample before antibody seroconversion [either before or at the time of the alanine aminotransferase (ALT) peak], and one or two samples after antibody seroconversion. For patients with FH, two samples were available for the analysis, one before and one after antibody seroconversion, with the exception of patient 3 (Table 1) (12). DNA amplified by PCR from the E1/E2 genes (14, 15) was cloned and a mean of 10.6 molecular clones from each sample were sequenced (16) for a total of 414 sequences, each 558 nucleotides in length. Genetic diversity was calculated by analysis of the amino acid sequences using the mean Hamming distance (17). To determine the course of HCV viremia, we detected serum HCV RNA by a nested PCR assay (18) and quantified it with the branched-chain DNA test (19); we determined the HCV genotype by sequence analysis of part of theE1 gene (20). All sera were tested for antibodies to structural and nonstructural proteins of HCV (21).

The genetic diversity and the number of viral variants within the HCV quasispecies, both in the first available PCR-positive sample and in the pre–antibody seroconversion sample, did not differ significantly between patients who resolved their infection and those who developed chronic disease (Fig. 1, A and B). In contrast, after antibody seroconversion, patients with resolving hepatitis showed a decrease in the genetic diversity of HVR1, whereas those with progressing hepatitis showed a marked increase in diversity (Fig. 1A). These different patterns are further illustrated by two representative cases shown in Fig. 2. When we compared the changes in genetic diversity between the last sample just before antibody appears and the first sample after antibody seroconversion, the difference in genetic diversity between resolving and progressing hepatitis was statistically significant (Table 2). Patients with FH, despite high serum levels of HCV RNA, showed the lowest degree of genetic diversity, both before and after antibody seroconversion (Fig. 1A). When we analyzed the genetic diversity in theE1/E2 region outside HVR1, based on the analysis of 155 predicted amino acids, the viral diversity was consistently lower than within HVR1 and did not change over time in all patients examined, although patients with FH showed a trend toward an increase after seroconversion (Fig. 1A). Therefore, the different patterns observed in patients with resolving and progressing hepatitis were essentially due to genetic variation within HVR1. These data strongly suggest that this region is under selective pressure by the host immune system.

Figure 1

Genetic distance among the variants (diversity) and number of viral strains of the HCV quasispecies within and outside HVR1 in fulminant, resolving, and progressing hepatitis. (A) Genetic diversity within the viral quasispecies, as measured by mean Hamming distance (17). (B) Number of different viral strains detected in the viral quasispecies. The values indicate the number of variants per 31 amino acids both within and outside HVR1. The data represent the mean (±SEM) of the results obtained from all the patients within each group at different time points. In fulminant hepatitis the analysis before antibody seroconversion was extended to only two patients, because only a single sample was available from patient 3 [Table 1 (12)].

Figure 2

Evolution of HCV quasispecies during the course of acute hepatitis C in two representative patients, one with resolving hepatitis and one with acute hepatitis that progressed to chronicity with a stable course. Patient numbers are the same as inTable 1. (A and D) Clinical course of hepatitis C. The gray areas indicate ALT levels. The red horizontal bars indicate positive assays for serum HCV RNA by PCR. The black lines indicate the titer of serum HCV RNA by the branched-chain DNA assay, on a logarithmic scale. The yellow horizontal bars indicate antibodies to HCV detected by second-generation assays. Tx denotes the time of blood transfusion. (B and E) Number of viral variants and diversity (genetic distance among variants) of the HCV quasispecies within the 31 amino acids of HVR1. The vertical bars indicate the number and the proportion of viral variants within each sample. Within the vertical bars, each variant is identified by a different color. The dominant viral variant found in each patient at the first time point is indicated in turquoise; other variants are indicated by additional colors. The same color indicates identity between viral variants detected at different time points within each patient, but not between different patients. The viral population diversity (black line) was calculated by mean Hamming distance from the predicted amino acid sequences obtained from each sample (17). (Cand F) Number of viral variants and genetic diversity of HCV within 155 amino acids from the E1/E2 region, spanning map positions 318 to 503, with the exclusion of the 31 amino acids of HVR1. [For a similar analysis of two other cases, representing fulminant hepatitis and chronic hepatitis with a rapidly progressive course, see Science Online (www.sciencemag.org/feature/data/1047689.shl).]

Table 2

Comparison of changes in genetic diversity and in number of viral strains in fulminant, resolving, and progressing hepatitis.

View this table:

In contrast to the genetic diversity, the number of viral variants as assessed by HVR1 sequences did not show any specific correlation with disease outcome, although a slight increase was detected in patients who developed chronic hepatitis (Fig. 1B). Similar to the genetic diversity, the number of variants outside HVR1 was consistently lower than that within HVR1 in all patient groups throughout the course of the disease.

To investigate whether the different patterns of viral genetic variation were due to positive selection, we measured the number of synonymous (silent) nucleotide substitutions per synonymous site and the number of nonsynonymous (amino acid replacement) nucleotide substitutions per nonsynonymous site (22), both within and outside HVR1; sequences obtained from each time point were compared with the consensus (reference) sequence of the first time point. This analysis revealed a difference in virus evolution according to the outcome of the disease. Comparison of genetic distances between the earliest and the latest sample from each patient revealed that the mean number of nonsynonymous substitutions per site per week within HVR1 was significantly higher in progressing hepatitis [mean ± SEM, 7.0 (±1.7) × 10−3] than in resolving [2.0 (±0.9) × 10−3; P = 0.037] or fulminant [0.4 (±0.4) × 10−3; P = 0.011] hepatitis, whereas outside HVR1 it was consistently lower in all groups and did not differ significantly among progressing [1.1 (±0.3) × 10−3], resolving [0.3 (±0.1) × 10−3], and fulminant [0.5 (±0.03) × 10−3] hepatitis. By contrast, the mean number of synonymous substitutions within HVR1 was comparable among the three groups, although it was slightly higher in fulminant [4.9 (±3.9) × 10−3] than in progressing [4.1 (±1.1) × 10−3] and resolving [3.6 (±2.6) × 10−3] hepatitis; similarly, outside HVR1, it was higher in fulminant [4.1 (±2.6) × 10−3] than in progressing [2.1 (±0.6) × 10–3] and resolving [1.5 (±1.0) × 10–3] hepatitis. Thus, in FH there was a striking amino acid conservation both within and outside HVR1, despite a high rate of viral replication and synonymous substitutions.

Phylogenetic analysis of all the HVR1 amino acid sequences obtained from each patient at different time points showed two topological patterns according to the outcome of the disease. In resolving hepatitis, there was generally a monophyletic population with intermingling of sequences derived from different time points (Fig. 3A). In progressing hepatitis, there was a tendency to form clusters over time (Fig. 3B) and the branch lengths were consistently longer than those seen in acute resolving hepatitis (23). By contrast, when HVR1 was excluded, the phylogenetic reconstructions of the rest of the E1/E2 region showed a monophyletic population in each patient, with very short branch lengths, and no difference between resolving and progressing hepatitis (24). The greater genetic distances seen in patients with progressing hepatitis correlated with a higher accumulation rate of nonsynonymous substitutions, consistent with a positive selection for change within HVR1.

Figure 3

Representative phylogenetic reconstructions showing the evolutionary relationships of all viral amino acid sequences of HVR1, within patients. The phylogenetic trees were constructed with the neighbor-joining method, using the NEIGHBOR program in the PHYLIP package (32). Genetic distances were calculated with PROTDIST from the PHYLIP package based on Kimura's distance. A bootstrap analysis using 100 bootstrap replicates was performed to assess the reliability at each of the internal nodes of the trees. The numbers at branch points indicate the percentage of the bootstrap resamplings; frequencies greater than 75% are shown. Substantially similar trees were generated using the Fitch-Margoliash distance method using the Fitch program available from the PHYLIP package. Clones are shown by a number indicating weeks from the time of primary infection, with each time point represented by a different color. The scale bar below each tree indicates the genetic distance based on Kimvra's formula. (A) The viral sequences were selected to represent the pattern seen in patients with acute resolving hepatitis. The phylogenetic reconstruction showed a monophyletic viral population with intermingling of sequences from different time points. (B) The viral sequences were selected to represent the pattern seen in patients who developed progressing hepatitis. The phylogenetic reconstruction showed sequential shifts in the viral population after 8 weeks from the time of infection. Some intermingling of sequences from sequential time points was also observed.

We next determined whether the pattern of evolution of the viral quasispecies in progressing hepatitis differed between slow and rapid progressors. Although there was a distinct difference in the rate and extent of viral evolution between patients who resolved their infection and those who progressed to chronicity, the distinction between slow and rapid progressors during the first 4 months of infection was equivocal. We also examined whether the pattern of genetic evolution observed in progressing hepatitis correlated with specific clinical, virological, or immunological parameters. Our study showed that the degree of genetic heterogeneity did not correlate with either the number of units of blood received or the number of donors. Similarly, the genetic diversity and the number of viral variants did not correlate with the level of viremia. Also, no significant association was found between the serum level of HCV RNA or the HCV genotype and the outcome of the disease.

Our study provides evidence that the outcome of acute hepatitis C may be determined in the early phase of primary infection. Acute hepatitis that progressed to chronicity correlated with genetic evolution of HCV within the first 4 months of infection. In agreement with the paradigm proposed for HIV infection (25,26), the presence of diverse viral forms may reflect shifts in the virus population possibly related to changes in the host environment, such as the appearance of neutralizing antibodies or intrahepatic cytotoxic T lymphocytes. Under these selective constraints, viral population disequilibrium may emerge with the selection of variant viruses with enhanced ability to persist in the host (27). The evidence in our patients of a temporal association between viral evolution and the first emergence of a specific immune response corroborates this hypothesis. Moreover, because the amino acid substitutions occurred almost exclusively within HVR1, a critical target of selective pressure by the immune system (5, 28), this domain may play a role in the pathogenesis of HCV infection and, in particular, in the progression to chronicity.

Although our findings on progressing hepatitis C suggest that there was no predominant viral form with enhanced pathogenic potential, FH was characterized by a homogeneous viral population. These data suggest that in fulminant hepatitis C there is a trend to preserve the unique fitness of a particular viral variant, which supports the hypothesis that the inherent virulence of a specific HCV strain may lead to massive hepatocellular necrosis.

Spontaneous viral clearance occurs in about 15% of patients after primary HCV infection (3). Although the immunological correlates of HCV clearance are still undefined, the earliest events in the virus-host interaction are likely to determine the outcome of infection. In this context, the observation that viral clearance correlates with a reduction in viral diversity over time may appear counterintuitive, as it might suggest the absence of effective immune responses exerting a selective pressure for change. The alternative and, we feel, more likely hypothesis is that the specific humoral or cellular immune responses against HCV are more effective in patients who resolve their infection, as suggested by recent studies (29), and that the genetic diversity of HCV quasispecies declines as a result of the progressive clearance of individual variants. Thus, when the immune response against HCV is optimal, viral variation is eventually contained and the strains become increasingly homogeneous until the final variant is effectively cleared. A similar trend, with a significant decrease in genetic diversity and in the number of HCV variants, has been documented in patients with a favorable response to interferon therapy during the first 2 weeks of treatment (30). Our findings have prognostic implications, because an increase in HVR1 diversity during acute infection predicts progression to chronic disease, whereas a decrease correlates with resolution of the disease. The direct implication of HVR1, the most variable region of HCV, in the development of chronic HCV infection poses a major challenge for devising preventive and therapeutic strategies against HCV.

  • * To whom correspondence should be addressed. E-mail: farcip{at}pacs.unica.it

REFERENCES AND NOTES

View Abstract

Navigate This Article