Analysis of HCV quasispecies dynamic under selective pressure of combined therapy

Abstract

Background: The quasispecies composition of Hepatitis C virus (HCV) could have important implications with regard to viral persistence and response to interferon-based therapy. The complete NS5A was analyzed to evaluate whether the composition of NS5A quasispecies of HCV 1a/1b is related to responsiveness to combined interferon pegylated (PEG-IFN) and ribavirin therapy.

Methods: Viral RNA was isolated from serum samples collected before, during and after treatment from virological sustained responder (SVR), non-responder (NR) and the end-of-treatment responder patients (ETR). NS5A region was amplified, cloned and sequenced. Six hundred and ninety full-length NS5A sequences were analyzed.

Results: This study provides evidence that lower nucleotide diversity of the NS5A region pre-therapy is associated with viral clearance. Analysis of samples of NRs and the ETRs time points showed that genetic diversity of populations tend to decrease over time. Post-therapy population of ETRs presented higher genetic distance from baseline probably due to the bottleneck phenomenon observed for those patients in the end of treatment. The viral effective population of those patients also showed a strong decrease after therapy. Otherwise, NRs demonstrated a continuous variation or stability of effective populations and genetic diversity over time that did not seem to be related to therapy. Phylogenetic relationships concerning complete NS5A sequences obtained from patients did not demonstrate clustering associated with specific response patterns. However, distinctive clustering of pre/post-therapy sequences was observed. In addition, the evolution of quasispecies over time was subjected to purifying or relaxed purifying selection. Codons 157 (P03), 182 and 440 (P42), 62 and 404 (P44) were found to be under positive selective pressure but it failed to be related to the therapy.

Conclusion: These results confirm the hypothesis that a relationship exists between NS5A heterogeneity and response to therapy in patients infected with chronic hepatitis C.

Figure 1

Alignment of sequence representation demonstrating the nonsense mutations observed in this study. A) Nonsense mutation at amino-terminal of the NS5A genomic region, positions 9, 47 and 84. B) One nonsense mutation located at the middle of the NS5A genomic region position 233, and one nonsense mutation located at the carboxy-terminal of the NS5A genomic region, position 399. Pink boxes indicate nonsense mutations. The next methionines are highlighted in gray. References sequences for genotypes 1a (H77_NS5A1a) and 1b (HCVJ_NS5A1b) are indicated.

Figure 2

Genetic variability of HCV quasispecies in samples collected over time. A) Genetic complexity calculated using normalized entropy (Shannon entropy). B) Diversity of HCV quasispecies calculated using genetic distance (p-distance). Patterns of response to therapy are represented by colours: sustained virological responders in red, non-responders in blue and end-of-treatment responders in green. C) Genetic diversity by the mean of genetic distance of each time-point collected. On the left: the end-of-treatment-responders. On the right: non-responders. D) Genetic diversity by the mean of genetic distance between every sample and baseline. Comparisons between non-responders and end-of-treatment responders. P=value calculated using Fisher’s test or paired t-test.

Figure 3

Quasispecies diversity within the NS5A protein based on the number of strains. A) The end-of-treatment responders. B) The non-responders. C) The sustained virological responders. Vertical bars represent viral variants. Inside the bars are the numbers of identical clones and, consequently, the number of identical quasispecies. The predominant quasispecies identified from the samples of each patient are identified by specific colors. The percentages of different quasispecies in ETR, NR and SVR samples are presented above the bars. The time of sample collection are indicated below the bars.

Figure 4

Unrooted phylogenetic trees reconstructed from the complete NS5A region. Phylogenetic tree reconstructed using the maximum-likelihood (ML) method, a heuristic search with Nearest Neighbor Interchange (NNI) branch-swapping algorithm. A) HCV genotype 1b tree was based on the TVMef+G substitution model. B) HCV genotype 1a tree was based on the GTR+G substitution model. References are utilized in both trees. Patterns of response are represented by colors: sustained virological responders in red, non-responders in blue and end-of-treatment responders in green.

Figure 5

Effective quasispecies populations among time-point analyzed. Skyride plots showing the relative genetic diversity of populations from patients ETR (A) and NR (B). The Black lines represent the median posterior distribution while blue shaded areas are the 95% Bayesian credible intervals. Red lines represent the start of the treatment.