Impact of Highly Active Antiretroviral Therapy and Immunologic Status on Hepatitis C Virus Quasispecies Diversity in Human Immunodeficiency Virus/Hepatitis C Virus-Coinfected Patients (original) (raw)

Abstract

This study analyzes the effect of highly active antiretroviral therapy (HAART), and thus immunologic status, on hepatitis C virus (HCV) load and quasispecies diversity in patients coinfected with the human immunodeficiency virus (HIV) and HCV. Three cohorts of coinfected patients were analyzed retrospectively over a period of 7 to 10 months: group A was antiretroviral drug naïve at baseline and then on HAART for the remainder of the study, group B did not receive antiretroviral therapy at any point, and group C was on HAART for the entire study. HCV quasispecies diversity was analyzed by sequencing hypervariable region 1. In a longitudinal analysis, there was no significant change from baseline in any immunologic, virologic, or quasispecies parameter in any of the three groups. However, in comparison to groups A and B, group C had significantly higher CD4+- and CD8+-cell counts, a trend toward a higher HCV load, and significantly increased number of HCV clones, entropy, genetic distance, and ratio of nonsynonymous substitutions per nonsynonymous site to synonymous substitutions per synonymous site (Ka/Ks). In addition, CD4+-cell count was positively correlated with HCV load, genetic distance, and Ka. Interestingly, patients infected with HCV genotype 2 or 3 had a significantly higher CD4+-cell count, HCV load, genetic distance, and Ka/Ks than those infected with genotype 1. These results suggest that there is no immediate effect of HAART on HCV but that, with prolonged HAART, immune restoration results in an increase in HCV load and quasispecies diversity.

As a result of common modes of transmission, the overall prevalence of hepatitis C virus (HCV) infection among human immunodeficiency virus (HIV)-infected individuals is approximately 30 to 50%, with rates of coinfection as high as 90% in injection drug users and almost 100% in hemophiliacs (7, 8). Previously, many HIV-infected individuals died before the appearance of HCV-related symptoms, whose onset may be 20 to 30 years from the time of initial HCV infection (20). However, the advent of highly active antiretroviral therapy (HAART) for HIV infection has resulted in a decrease in morbidity and mortality for many HIV-infected individuals. As a result of the increased life expectancy of this population, HCV has emerged as a pathogen of great importance in the clinical management of HIV/HCV-coinfected patients.

The clinical implications of coinfection have been the focus of intense research. HIV coinfection has been shown to increase the severity of liver disease in patients chronically infected with HCV (10, 34, 47). In addition, many studies have documented that HIV/HCV-coinfected patients have higher HCV loads than do HCV-monoinfected controls (4, 12, 44). In contrast to these well-documented effects of HIV on the course of HCV disease, the effect of HCV on the course of HIV disease is less clear. In some studies, coinfection with HCV has been shown to confer an increased risk for progression to AIDS in HIV-infected individuals (5, 14); however, other studies have failed to demonstrate this increased risk (53).

One of the hallmarks of HCV is its marked genetic diversity. On a population level, genotype 1 infections account for approximately two-thirds of all HCV infections in the United States (25) and may account for up to 83% of infections in the HIV/HCV-coinfected population (45). HCV also exists within an individual as a population of quasispecies (28). The region of the HCV genome with the greatest diversity is hypervariable region 1 (HVR1), located at the N terminus of the E2 envelope gene (19, 24). This region has been implicated to play a role in immune escape by virtue of its high degree of sequence variation (51). The significance of quasispecies diversity, however, is still unclear. Increased diversity within HVR1 has been associated with increased severity of liver disease (16, 21), although some studies have not found such an association (31). Increased quasispecies diversity has also been shown, during acute infection, to predict progression to persistent viremia and chronic HCV infection (9, 36). In addition, many studies have found an association between a higher pretreatment number of quasispecies in HVR1 and a poor response to interferon (IFN) therapy (13, 21).

The investigation into the evolution of HCV quasispecies in the HIV/HCV-coinfected population has yielded conflicting results. When comparing HIV/HCV-coinfected patients with HCV-monoinfected controls, two studies have found that coinfected patients have more quasispecies diversity (6, 43). If HIV infection were a surrogate marker for immunosuppression, this would imply that quasispecies diversity increases with immunosuppression. In support of this theory, Dove et al. have shown in coinfected patients that those with lower CD4+-cell counts have greater quasispecies evolution than those with higher CD4+-cell counts (L. M. Dove, Y. Phung, J. Wrock, M. Kim, and T. L. Wright, Abstract, Hepatology **30:**456A, 1999). However, a number of other studies in HIV/HCV-coinfected individuals suggest that the opposite is true, i.e., that quasispecies diversity decreases with immunosuppression, manifested as lower quasispecies variation in those individuals with lower CD4+-cell counts (27, 38, 50).

There has been much focus recently on the administration of HAART in HIV/HCV-coinfected individuals. The effect of HAART on HCV load is still controversial. Most studies have shown no change in HCV RNA titers following HAART (11, 37, 49), although some studies have shown a transient (34, 39) or sustained (35) increase in HCV load; yet others have shown a decrease in HCV RNA levels and in some cases even HCV clearance (55).

To our knowledge, the effect of HAART on the quasispecies profile of HIV/HCV-coinfected individuals has not been described. Given the potential predictive value of quasispecies variation as a marker for HCV-related disease and IFN resistance, determining how HAART affects the genetic diversity of HCV will have important implications in both understanding the course of HCV disease in the setting of HIV infection, as well as in helping to better define the therapeutic management of coinfected individuals. The present data on quasispecies diversity and evolution in coinfected patients are conflicting. However, given that (i) HVR1 responds to immune pressure with increased variability (36, 51) and (ii) in other disease models it has been shown that immunosuppression (and thus decreased immune pressure) is associated with a decrease in HCV genetic diversity (2, 23, 26, 29, 33), we hypothesized that HAART, via immune restoration and increased immune pressure, might cause an increase in HCV quasispecies diversity. As such, we sought to determine the effect of HAART-associated immune restoration on the HCV quasispecies profile in HIV/HCV-coinfected individuals. In addition, we analyzed the effect of HAART on HCV load, as well as the differential effect of HCV genotype on quasispecies evolution during HAART.

MATERIALS AND METHODS

Patient selection.

Three cohorts of patients were chosen retrospectively from a database of patients followed in the Immune Clinic at the Veterans Affairs Palo Alto Health Care System (Palo Alto, Calif.). These patients were previously identified to be seropositive for both HIV and HCV. Samples were collected during the course of routine clinical follow-up, and the Stanford University Institutional Review Board approved the project. Patients who had received HCV treatment at any point or for whom HCV viremia could not be documented were excluded from the study. No patients were coinfected with the hepatitis B virus. The cohorts were defined as follows (Table 1): (i) the first cohort of patients (group A) was antiretroviral drug naïve at baseline but initiated and maintained HAART during the observation period. All patients had achieved an HIV load of <50 copies/ml by the final time point analyzed. (ii) The second cohort of patients (group B) was antiretroviral drug naïve and did not receive HIV treatment at any point during the observation period. (iii) The third cohort of patients (group C) was on HAART for the entire duration of the observation period. All patients had an HIV load of < 50 copies/ml for an average of 9.4 ± 7.2 months before the first time point chosen, thereby ensuring that virologic containment of HIV replication had been achieved prior to the onset of the study period. The mean length of the study period for each group is shown in Table 1. Serum samples were obtained from three time points throughout this period for all patients except one patient in group A, from whom samples from four time points were obtained.

TABLE 1.

Sociodemographic and HCV genotype characteristics of study groupsa

HIV, HCV, and T-cell subset analysis.

HCV genotyping was performed using the INNO-LiPA HCV II line probe assay (Bayer Diagnostics, Tarrytown, N.Y.). HIV and HCV RNA quantification was performed using the VERSANT HIV-1 RNA 3.0 bDNA Assay and VERSANT HCV RNA 3.0 Quantitative bDNA Assay, respectively (Bayer Diagnostics). CD4+- and CD8+-cell counts were measured by Unilab (San Jose, Calif.).

Oligonucleotide primers.

The E1/E2 region targeted for amplification included the entire 81-bp HVR1 at the 5′ end of the E2 gene as well as 170 bp of non-HVR1 sequence at the 3′ end of the E1 gene. Because of the diversity of the HCV genome, and in order to avoid selective PCR amplification of different quasispecies, the primers for reverse transcription and amplification of the E1/E2 region were chosen based on the analysis by Toyoda et al. (50), in which primers were designed to be 80% homologous to 128 different reported HCV sequences. Primer sequences were as follows: HVR1 (TGGGACACATGATGATGAACTGGT) was used as the sense primer in both rounds of seminested PCR. HVR4 (CGGTGCTGTTTATGTGCCAACTGCC) was used as the antisense primer for both reverse transcription and the first round of PCR amplification. HVR3 (GATGTGCCAGCTGCCATTGG) was used as the antisense primer in the second round of PCR amplification.

RNA extraction and reverse transcriptase PCR (RT-PCR).

Total RNA was extracted from 140 μl of patient sera by using the QIAamp Viral RNA Mini Kit (Qiagen, Valencia, Calif.) and resuspended in 60 μl of RNase-free water containing 0.04% sodium azide. Serial dilutions, when indicated, were made by 1:10, 1:100, and 1:1,000 dilution of this RNA sample. Two and a half microliters of the resulting RNA sample or dilution was then reverse transcribed using the GeneAmp RNA PCR Kit (Applied Biosystems, Foster City, Calif.) in a reaction volume of 20 μl containing 0.75 μM HVR4 primer, 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 25 mM MgCl2, a 1 mM concentration of each deoxynucleoside triphosphate (dNTP), 1 U of RNase inhibitor, and 1 U of murine leukemia virus RT. The E1/E2 region was then amplified from the resulting cDNA by seminested PCR; the first round was performed in a 50-μl reaction volume containing 50 mM KCl, 10 mM Tris-HCl (pH 8.3), 2 mM MgCl2, 0.2 μM HVR1 sense primer, 0.3 μM HVR4 antisense primer, and 1.5 U of AmpliTaq DNA Polymerase (Applied Biosystems). PCR amplification was performed under the following conditions: 95°C for 105 s; 35 cycles of 95°C for 15 s and 60°C for 30 s; and 72°C for 7 min. The second round of PCR was performed in a 100-μl reaction volume containing 1× PCR buffer (Qiagen), 2 mM MgCl2, a 200 μM concentration of each dNTP, 0.3 μM HVR1 sense primer, 0.3 μM HVR3 antisense primer, and 2.5 U of Taq DNA polymerase (Qiagen). Thermal cycling was then performed as follows: 94°C for 105 s; 30 cycles of 94°C for 1 min, 56°C for 1 min, and 72°C for 1 min; and 72°C for 10 min. The resulting 318-bp PCR product was isolated by gel purification using the Qiaex II Gel Extraction Kit (Qiagen).

Cloning and sequencing.

The amplified E1/E2 fragments were cloned into the pCR4-TOPO vector, and the resultant plasmids were used to transform chemically competent TOP10 cells using the TOPO TA Cloning Kit for Sequencing (Invitrogen, Carlsbad, Calif.). Colonies were screened for E1/E2-containing plasmids by PCR amplification of colonies in a 40-μl reaction volume containing 1× PCR buffer (Qiagen), 2 mM MgCl2, a 200 μM concentration of each dNTP, 0.07 μM M13 Forward primer (GTAAAACGACGGCCAG), 0.07 μM M13 Reverse primer (CAGGAAACAGCTATGAC), and 1 U of Taq DNA polymerase (Qiagen). Thermal cycling was performed as follows: 94°C for 10 min; 30 cycles of 94°C for 1 min, 55°C for 1 min, and 72°C for 1 min; and 72°C for 10 min. Ten positive clones per time point per patient were then sequenced using AP Biotech DYEnamic ET Terminator cycle sequencing (Amersham Biosciences, Piscataway, N.J.) and ABI PRISM (Applied Biosystems) sequencing technology.

Genetic analysis.

The 10 E1/E2 sequences per time point were aligned at the nucleotide level using EditSeq and MegAlign software (DNASTAR, Inc., Madison, Wis.). All sequence analysis was then performed with the 81-bp (27-amino-acid [aa]) HVR1 as well as with a 162-bp (54-aa) region of non-HVR1 sequence in the E1 gene. The program MEGA version 2.1 (22) was used to analyze the aligned sequences for the following parameters: (i) mean genetic distance, calculated as the mean of all pairwise comparisons of genetic distance using the p distance, defined as the number of amino acid differences divided by the total number of amino acid sites compared; and (ii) numbers of synonymous substitutions per synonymous site (Ks) and nonsynonymous substitutions per nonsynonymous site (Ka) using the Nei-Gojobori method with the Jukes-Cantor correction for multiple substitutions (32). The ratio of Ka/Ks was then calculated. The complexity of the quasispecies population per time point was analyzed at the amino acid level by two different methods: (i) determination of the total number of different clones within the 10-sequence population and (ii) calculation of the normalized Shannon entropy (Sn) to take into account the frequency of each different quasispecies in the population, calculated as follows: Sn = −Σ(pi ln pi)/ln N, where pi is the frequency of each sequence in the population and N is the total number of sequences analyzed (52). Entropy of the non-HVR1 E1 region was determined by taking the mean of the Sn values calculated for two separate 81-bp sequence blocks. Phylogenetic analysis was performed with the MEGA version 2.1 program (22) by using the neighbor-joining method and the Kimura two-parameter model.

Statistical analysis.

Means are expressed as mean plus or minus standard deviation. Comparisons between groups were determined using the Fisher exact probability test, analysis of variance, or the Student t test, where appropriate. Comparison analyses and correlations were performed using StatView 5.0 software (SAS, Cary, N.C.). All reported P values are two-tailed, and a P of less than 0.05 was considered significant.

Nucleotide sequence accession number.

Sequences reported herein have been assigned EMBL accession numbers AJ510768 through AJ511257.

RESULTS

Subject characteristics.

Three cohorts of HIV/HCV-coinfected patients were selected based on their exposure to HAART, as described in detail in Materials and Methods. Briefly, group A was antiretroviral naïve at baseline and then received HAART throughout the study period such that all had an undetectable HIV load by the final time point analyzed, group B did not receive antiretroviral therapy at any time point throughout the study period, and group C was on HAART for the entire duration of the study and had an undetectable HIV load at baseline. The sociodemographic characteristics and HCV genotype distribution of the three cohorts are presented in Table 1. All patients were male except for one patient in group C. There were no significant differences between the groups in terms of age, race, alcohol and intravenous drug use (IVDU) history, or length of study period.

Quasispecies parameters and comparison between HVR1 and the E1 region.

HCV quasispecies were analyzed by assessing the diversity in HVR1, since this area has been shown to have a high degree of variation and has been implicated as playing a role in immune escape (19, 24, 51). Quasispecies parameters analyzed were (i) complexity, measured as both the total number of amino acid clones and Sn; (ii) mean genetic distance at the amino acid level; and (iii) Ka/Ks. Nonsynonymous changes (Ka) in HCV are likely to play a role in immune escape in response to immune pressure, especially in HVR1 (51). On the other hand, synonymous changes (Ks) represent genetic drift, which is more a function of the high replicative capacity and high mutation rate of HCV (36). Thus, the ratio of Ka/Ks is a good measure of immune pressure, since it normalizes for the genetic drift associated with replication and since values of Ka/Ks that are > 1 represent immune pressure (42).

In order to confirm that in our patient population the quasispecies evolution in HVR1 was indeed higher than that of the surrounding nonhypervariable E1 region, we compared the quasispecies parameters between these two regions of the HCV genome (Table 2). When the data for all patients were examined together, HVR1 had significantly greater quasispecies variation than the E1 region for all parameters analyzed: in HVR1, the number of clones was 1.6-fold higher (P < 0.0001), the entropy was 3.4-fold higher (P < 0.0001), the genetic distance was 11.7-fold higher (P < 0.0001), and the Ka/Ks ratio was 5.0-fold higher (P = 0.003). The results were similar when patients were stratified by treatment group (Table 2). Thus, given the significantly higher quasispecies complexity, genetic distance, and Ka/Ks in HVR1 than in an adjacent region of the E1 gene, HVR1 seemed to be an accurate marker of quasispecies variation in our HIV/HCV-coinfected population.

TABLE 2.

Comparison of quasispecies parameters between HVR1 and the non-HVR1 (E1) regiona

Clonal sampling and phylogenetic analyses.

To assess for cross-contamination between samples during RT-PCR amplification, a phylogenetic tree was created from the E1/E2 sequences at the baseline time point for all patients (Fig. 1). Of the 160 sequences analyzed, 159 of these clustered according to their patient source; only one clone, C3-8, was grouped with an adjoining cluster. Thus, phylogenetic analysis showed that 99.4% of sequences clustered according to patient source, indicating that there was no cross-contamination between patient samples during RT-PCR amplification.

FIG. 1.

Phylogenetic analysis of E1/E2 nucleotide sequences from the baseline time point of all patients. A phylogenetic tree was constructed from a total of 160 E1/E2 sequences using the neighbor-joining method and Kimura two-parameter model. The horizontalbranch lengths are drawn to scale, and the scale bar represents 0.1 nucleotide substitutions per site. The clone number is indicated at the end of each horizontal branch tip, and the brackets at the far right indicate groupings based on patient source. Patients in group A are indicated as A1 through A7, patients in group B as B1 through B3, and patients in group C as C1 through C6. Clone numbers are indicated by patient source followed by sequence number (e.g., A1-1, A1-2, etc.).

In order to show that the HCV load of the sample did not bias PCR amplification or clonal analysis—i.e., that samples with high HCV load did not have a larger number of variants amplified (and thus sampled) than those with low HCV load simply because of inefficient RT-PCR amplification, serially diluted RNA samples were amplified by RT-PCR. Dilutions of RNA were made from three samples with high HCV load (> 40 × 106 copies/ml), intermediate HCV load (10 × 106 copies/ml), and low HCV load (0.7 × 106 copies/ml) to 1:10, 1:100, and 1:1,000 of their original concentrations; these samples were then amplified by our standard RT-PCR protocol (as described in Materials and Methods). All diluted samples yielded an E1/E2 PCR product detectable by gel electrophoresis (data not shown). Thus, we concluded that our RT-PCR protocol was efficient in amplifying the E1/E2 region from samples with low HCV load and that the RT-PCR amplification process was unlikely to introduce bias to the resulting quasispecies pool based on poor sampling of the initial RNA product.

To ensure that analysis of only 10 sequences per time point was an accurate measure of the quasispecies population at that time point, 10 E1/E2 sequences were analyzed from each of two independent RT-PCRs using the same RNA template; five samples were randomly chosen for this analysis. To compare the quasispecies population obtained from each reaction, a phylogenetic tree was constructed for each of the five samples using both sets of 10 clones obtained from each RT-PCR for a total of 20 sequences per tree (Fig. 2). As can be seen in the two representative phylogenetic trees shown in Fig. 2, sequences do not cluster according to RT-PCR, indicating that both sets of sequences were drawn from a similar quasispecies pool. In order to quantitatively compare the HVR1 quasispecies parameters obtained for each RT-PCR, the mean values for the five samples in each group were compared (Table 3), a type of comparison similar to the intergroup analysis used elsewhere in this paper (see below). As such, there were no significant differences between the number of HCV clones, entropy, genetic distance, and Ka/Ks between the independent RT-PCR experiments for the group of five samples examined (P > 0.80). Taken together, the phylogenetic and quantitative quasispecies analyses above indicate that the sampling of 10 clones per time point was sufficient to accurately determine the quasispecies profile of that sample.

FIG. 2.

Phylogenetic analysis of E1/E2 nucleotide sequences from two independent RT-PCRs. Phylogenetic trees were constructed from 10 E1/E2 sequences for each RT-PCR (for a total of 20 sequences per sample) using the neighbor-joining method and Kimura two-parameter model. Representative trees are shown for two samples, A and B. The horizontal branch lengths are drawn to scale, and the scale bar indicates nucleotide substitutions per site. The clone number is indicated at the end of each horizontal branch tip. For each sample, clones from RT-PCR 1 are indicated as 1-1 through 1-10, and clones from RT-PCR 2 are indicated as 2-1 through 2-10.

TABLE 3.

HCV quasispecies parameters determined by two independent RT-PCRsa

Longitudinal analysis of immunologic, virologic, and quasispecies parameters within each cohort.

We next sought to determine whether or not there was a change in immunologic or virologic parameters over time within each cohort (Table 4). In group A, the HIV load became undetectable after a mean duration of 7 months of HAART and, at this point, the CD4+-cell count had increased from baseline by 148 cells/μl (P = 0.13) and the HCV load had increased by 7.4 × 106 copies/ml (P = 0.36), although neither change reached statistical significance. There was no statistically significant difference in any of the HCV quasispecies parameters analyzed. In group B, during the 10-month observation period without HAART treatment, the HIV load increased by 35,337 copies/ml (P = 0.65), the CD4 count decreased by 212 cells/μl (P = 0.35) and the HCV load decreased by 8.9 × 106 copies/ml (P = 0.53); again, none of these changes were significant. Similarly, there were no statistically significant differences in any of the HCV quasispecies parameters analyzed over time in group B. In group C, during the 8-month observation period while on HAART with an undetectable HIV load, the CD4+-cell count increased by 120 cells/μl (P = 0.59) and the HCV load decreased by −0.8 × 106 copies/ml (P = 0.94); again, these changes were not significant. There were also no significant differences in any quasispecies parameter analyzed in group C.

TABLE 4.

Immunologic, virologic, and quasispecies characteristics of coinfected patients at baseline and ending time points

Analysis of multiple time points as a more accurate assessment of the immune status and quasispecies profile.

It has been shown previously that HCV quasispecies diversity changes in an oscillatory manner during the natural course of infection in both monoinfected and HIV-coinfected patients (6). Thus, we hypothesized that analyzing the quasispecies at a single time point—a “snapshot in time”—might not accurately reflect the overall quasispecies profile in a patient over time. Indeed, in our study, we also found that the quasispecies composition varied in an oscillatory manner (data not shown). Thus, in order to eliminate any bias introduced by snapshot analyses, we assessed the quasispecies and immunologic profiles of our study subjects by taking the mean of all time points collected during the study period. These results are shown in Table 5.

TABLE 5.

Mean immunologic, virologic, and quasispecies characteristics of coinfected patients throughout the study perioda

Overall, group C had higher mean values for CD4+-cell count, CD8+-cell count, HCV load, number of HCV clones, HCV entropy, HCV genetic distance, and Ka/Ks than did either group A or B. The CD4+-cell count for group C was 564 cells/μl, significantly higher than 317 cells/μl for group A (P = 0.007) and 238 cells/μl for group B (P = 0.007). Similarly, the CD8+-cell count for group C was 1,280 cells/μl, compared with 1,004 cells/μl in group A (P = 0.13) and 628 cells/μl in group B (P = 0.015), although the former did not reach statistical significance. There was also a trend toward a higher HCV load in group C, with an RNA titer of 18.1 × 106 copies/ml, compared with 11.3 × 106 copies/ml in group A (P = 0.12) and 6.6 × 106 copies/ml in group B (P = 0.06), although these differences did not reach statistical significance. The mean number of clones in group C was 5.83, significantly higher than 3.75 in group A (P = 0.003) and 2.71 in group B (P = 0.001). Similarly, the entropy of HVR1 in group C was 0.632, significantly higher than 0.449 in group A (P = 0.014) and 0.265 in group B (P = 0.0005). The genetic distance in the HVR1 was also higher in group C at 0.142 than 0.091 in group A (P = 0.12) and 0.041 in group B (P = 0.03), although the former did not reach statistical significance. The marker of immune pressure, Ka/Ks, was also significantly higher in group C at 1.695 than 0.576 in group A (P = 0.0002) and 0.743 in group B (P = 0.02). Interestingly, there were no significant differences in any immunologic, virologic, or quasispecies parameter between groups A and B. Thus, in comparison with the other study groups, the group that was on HAART for the duration of the study period (group C) had (i) significantly higher CD4+ and CD8+ counts, (ii) a trend toward a higher HCV load, (iii) significantly higher quasispecies complexity and genetic distance, and (iv) evidence of increased immune pressure manifested as a significantly higher Ka/Ks.

Correlation between immunologic, virologic, and quasispecies parameters.

In order to further investigate the association between increased CD4+- and CD8+-cell counts with higher HCV load and increased quasispecies diversity, we next assessed the relationship between these parameters by correlation analysis (Table 6). CD4+-cell count was positively correlated with HCV load (r = 0.545, P < 0.0001), genetic distance (r = 0.315, P = 0.02), and Ka (r = 0.316, P = 0.02). The CD4+-cell count was not significantly correlated with the number of HCV clones, entropy, Ks, or Ka/Ks. Similarly, CD8+-cell count was positively correlated with HCV load (r = 0.287, P = 0.04). There was also a trend toward positive correlation between CD8+-cell count and both genetic distance (r = 0.203, P = 0.16) and Ka (r = 0.248, P = 0.08), although neither reached statistical significance. There was no correlation between CD8+-cell count and number of HCV clones, entropy, Ks, or Ka/Ks.

TABLE 6.

Correlation between immunologic, virologic, and quasispecies parametersa

We then further assessed the relationship between HCV load and quasispecies parameters. HCV load was positively correlated with HCV entropy (r = 0.279, P = 0.03), genetic distance (r = 0.46, P = 0.0002), Ka (r = 0.44, P = 0.0004), and Ks (r = 0.32, P = 0.01). There was also a trend toward a positive correlation between HCV load and both the number of HCV clones (r = 0.25, P = 0.06) and Ka/Ks (r = 0.20, P = 0.15), although these did not reach statistical significance. In order to further investigate the association between immune pressure and quasispecies diversity, we analyzed the correlation between Ka/Ks and all parameters of quasispecies variation. Ka/Ks was positively correlated with the number of HCV clones (r = 0.578, P < 0.0001), entropy (r = 0.53, P < 0.0001), and genetic distance (r = 0.41, P < 0.0001). The entropy and genetic distance were also positively correlated (r = 0.561, P < 0.0001), as were the number of clones and genetic distance (r = 0.557, P < 0.0001). Lastly, the two measures used to assess quasispecies complexity, the number of clones and Sn, were also positively correlated (r = 0.977, P < 0.0001).

Analysis of immunologic, virologic, and quasispecies parameters by genotype.

We then sought to determine whether or not there were any differences in mean immunologic, virologic, or quasispecies parameters between patients infected with HCV genotype 1 and those infected with genotype 2 or 3 (Table 7). Intragroup comparisons were not made for group B because all three patients in this group were infected with genotype 1. Interestingly, patients infected with HCV genotype 2 or 3 had higher mean CD4+-cell counts than those infected with genotype 1. This was true for all patients taken together, as well as when patients were stratified into their respective treatment groups. The difference was most dramatic in group C, where patients infected with genotype 2 or 3 had a mean CD4+-cell count of 864, while those infected with genotype 1 had a mean CD4+-cell count of 264 (P < 0.0001). There was no significant difference in the CD8+-cell count of patients based on HCV genotype. In group C, but not group A, patients infected with genotype 2 or 3 had a higher HCV load (27.7 × 106 copies/ml versus 8.6 × 106 copies/ml; P = 0.005). Overall, patients with genotype 2 or 3 also had an increased genetic distance (0.146 versus 0.070 for all groups taken together; P = 0.01). When patients were analyzed by group, there was only a trend toward an increased genetic distance in patients infected with genotype 2 or 3: 0.12 versus 0.055 for group A (P = 0.14) and 0.174 versus 0.109 for group C (P = 0.20). In group A, but not group C, there was a higher Ka/Ks associated with genotype 2 or 3 infection (0.839 versus 0.399; P = 0.004). There was no significant difference in the number of HCV clones or entropy based on genotype. Taken together, patients infected with genotype 2 or 3 had significantly higher CD4+-cell counts and, in some cases, had a significantly higher HCV load, genetic distance, and Ka/Ks.

TABLE 7.

Analysis of mean immunologic, virologic, and quasispecies parameters by genotypea

DISCUSSION

Three groups of HIV/HCV-coinfected patients, differing in their exposure to HAART, were analyzed for any change in immunologic, virologic, or quasispecies parameters over a period of 7 to 10 months. In a longitudinal analysis, no significant change was observed from baseline for CD4+- or CD8+-cell counts, HCV load, or any quasispecies parameter in any of the three groups. Thus, there seems to be no immediate effect of HAART on HCV load or quasispecies diversity. By analyzing samples from all time points collected in the study, no significant difference was found in quasispecies diversity between those who had just initiated HAART (group A) and those who did not receive any treatment (group B), suggesting again that there is no short-term effect of HAART on quasispecies variation. However, patients who had been on HAART with undetectable viral loads for approximately 9 months (group C) did have (i) significantly higher CD4+- and CD8+-cell counts, (ii) a trend toward a higher HCV load, (iii) significantly increased quasispecies complexity and genetic distance, and (iv) significantly increased immune pressure manifested as an increased Ka/Ks with a value that was > 1. It is of note that all of the quasispecies parameters were increased in group C, giving more weight to the conclusion that quasispecies variation is higher in these patients. The association between quasispecies diversity and CD4+-cell counts, in particular, was supported by results of correlation analysis, which showed that CD4+-cell counts were positively correlated with greater genetic distance and Ka. In addition, Ka/Ks, a marker of increased immune pressure, was correlated with increased number of clones, entropy, and genetic distance, consistent with the idea that quasispecies diversity in the HVR1 responds to immune pressure, as has been shown by others (9, 51).

As a result of the above findings, we propose the following model to describe the relationship of HCV with immune status during HAART: initially, during the virologic containment of HIV, HAART does not have a major impact on quasispecies diversity (group A in our study). However, once virologic containment is stable and immune restoration has been maximized, there is increased immune pressure (increased Ka/Ks), which causes increased HCV quasispecies diversity as a means for immune escape (group C in our study). In addition, there may also be an increase in HCV load during this later stage, which may go hand in hand with increased quasispecies diversity (see below). In summary, this model describes the adaptive evolution of HCV in response to immune pressure: greater immune competence in the host results in increased quasispecies diversity by selecting for escape mutants from the preexisting pool of randomly generated quasispecies. In this manner, a less effective immune response does not exert such pressure for change on the quasispecies pool, resulting in a more homogenous population of viral quasispecies.

The increased immune pressure seen in patients on long-term HAART (group C) is likely a result of both quantitative (i.e., increased CD4+- and CD8+-cell counts) as well as qualitative (e.g., modification of cytokine expression) changes in immunologic status. Patients in group C had already achieved virologic containment of HIV for an average of 9 months prior to the onset of the study period. In comparison to patients who received a shorter duration of HAART (group A), the long-term virologic suppression in these patients may have allowed for better immune recovery (both quantitative and qualitative), consistent with the observation that HIV suppression for >1 year may be necessary to allow for complete and effective recovery of the immune system (30). In addition, it has been shown that HAART normalizes alterations in cytokine patterns induced by HIV infection (17, 30). These qualitative improvements in immune function may explain our observation that CD4+-cell counts, while correlated with genetic distance and Ka, were not correlated with the number of clones, entropy, or Ka/Ks. Absolute CD4+-cell counts may thus be only part of the immune driving force behind increased quasispecies diversity in patients on long-term HAART.

Interestingly, we found that CD4+- and CD8+-cell counts were positively correlated with HCV load. This is in contrast to most other studies, which have shown either an inverse relationship between HCV load and CD4+-cell count (4, 47) or no association at all (44, 45). The reason for these conflicting reports is unclear, although it may be due to differences in the immune function of study subjects or to different methods of studying the relationship between CD4+-cell count and HCV load. The biological significance of any change in HCV load with CD4+-cell count, when assessed for relevance to disease severity or response to IFN therapy, is unclear. In our study, we showed a threefold difference between the HCV load of group C (18.1 × 106 copies/ml) and that of group B (6.6 × 106 copies/ml). Daar et al. have shown that, for every 10-fold increase in HCV load, there is a increased relative risk of 1.66 for progression to AIDS and 1.54 for AIDS-related mortality (5). Therefore, a threefold increase in HCV load is unlikely to cause a substantial increased risk for a more rapid progression to AIDS. In addition, the impact of HCV load on HCV disease is unclear, as many studies have failed to find a correlation between HCV load and severity of liver disease (12, 41). On the other hand, it has been shown that an increase in pretreatment HCV load of only 1.2 × 106 copies/ml may be associated with a poor response to IFN therapy (48). Thus, the increased HCV load associated with long-term HAART in this study may not be predictive of an increased severity of HCV or HIV-related disease but may be important in predicting the response to subsequent IFN therapy.

In this study, we showed a significant positive correlation between HCV load and entropy, genetic distance, and Ka, and a trend toward a positive correlation between HCV load and both the number of clones and Ka/Ks. The data from previous studies on this relationship are not conclusive, with some studies reporting no correlation between quasispecies diversity and HCV load (9, 31, 36, 38) and others showing a positive correlation (13, 18). The reasons for this conflicting data are likely differences in study populations and methods used to evaluate quasispecies variation. The association between increased HCV load and greater quasispecies diversity demonstrated in this study can be explained by the adaptive evolution of HCV in the face of immune pressure: escape mutants generated in this way can replicate more freely, resulting in a higher HCV load. Similarly, a less robust immune response cannot select for escape mutants with such replicative fitness, and so patients with less HCV quasispecies diversity would also have a lower HCV load.

In this study, an association was found between lower CD4+-cell counts and decreased HCV quasispecies diversity in HIV/HCV-coinfected patients. These data are in accordance with results from other studies in coinfected patients showing that HCV quasispecies diversity decreases with the degree of HIV-related immunosuppression (27, 38, 50). Our data are also consistent with what has been shown in other disease models of immunosuppression: studies of HCV quasispecies in patients with agammaglobulinemia/hypogammaglobulinemia (2, 23) and in patients undergoing immunosuppressive therapy for liver transplantation (26, 29) or bone marrow transplantation (33) have all shown a decrease in HCV quasispecies diversity during states of immunosuppression. However, a few studies in HIV/HCV-coinfected patients have found the opposite result—that HCV quasispecies diversity increases with HIV-induced immunosuppression. One study showed that coinfected patients with lower CD4+-cell counts had a greater percentage of new clones over a 1-year study period than did those with higher CD4+-cell counts (Dove et al., Hepatology **30:**456A, 1999); however, this study involved only nine patients, and the authors did not report whether or not there was a change in genetic distance or Ka/Ks. Two other studies compared HIV/HCV-coinfected patients with HCV-monoinfected controls and found that coinfected patients had more quasispecies diversity (6, 43). One study compared only two patients (6), and neither study addressed the effect of CD4+-cell count differences within the coinfected population (6, 43). In addition, none of the aforementioned studies stratified patients based on exposure to HAART (6, 43; Dove et al., abstract).

Our results showing that HCV quasispecies diversity increases with immune pressure are also consistent with what has been described for HIV quasispecies evolution during the course of HIV monoinfection. Wolinsky et al. showed that in HIV-positive individuals over time, rapid CD4+-cell loss was associated with evolutionary stasis of HIV, while lower rates of CD4+-cell loss were associated with a greater accumulation of mutations and, in particular, nonsynonymous substitutions, indicating that selective pressure plays a role in the evolution of HIV quasispecies (52). Thus, HCV and HIV act similarly in response to immune pressure: greater immunologic competence is associated with an adaptive increase in the genetic diversity and evolution of both viruses.

The increased HCV quasispecies diversity observed in our group of patients on long-term HAART has one of two possible effects: (i) the quasispecies diversity is a marker of increased immune pressure by a more effective immune system, which leads to better control of HCV infection and thus less severe liver disease, or (ii) the increased quasispecies diversity results in the potential for more virulent or immunoresistant clones that cannot be adequately contained by the immune system (even though it might be functioning overall at a higher level), thus leading to more severe liver disease. In support of the first hypothesis, data from studies of HCV recurrence after liver transplantation have shown that increased quasispecies diversity and increased Ka/Ks have been associated with less severe HCV recurrence (26, 41). However, the majority of data in HCV-monoinfected patients supports the second hypothesis—that increased diversity within HVR1 will lead to increased severity of liver disease (16, 21). Further investigation into the significance of quasispecies diversity in these patients is warranted.

Many studies have found an association between higher pretreatment HCV quasispecies diversity in HVR1 and a poor response to IFN therapy (13, 21). This has important implications in the debate as to which infection—HIV or HCV—to treat first in a coinfected patient. Yokozaki and colleagues (54, 55) have argued that HIV treatment should be undertaken first because HAART would increase CD4+-cell counts and may potentially decrease HCV load, thus providing a better starting point for IFN therapy since it has been shown that a lower HCV load (48) and higher CD4+-cell count (46) both predict a better response to IFN. Others have argued that HCV treatment should be undertaken first, since HAART may in fact increase HCV load, cause hepatotoxicity, and interact with HCV medications in a manner that may limit compliance (3). The data in our study support, in part, both arguments. On the one hand, the increase in CD4+-cell count associated with HAART suggests that treatment of HIV should precede treatment of HCV, as discussed above. However, on the other hand, we have shown that long-term HAART is associated with an increased HCV load and greater quasispecies diversity, both of which predict a poor response to IFN—this suggests that treatment for HCV should precede HAART in order to decrease the probability of a poor response to IFN.

A surprising finding in our study was that patients infected with genotype 2 or 3 had significantly higher CD4+-cell counts and, in some cases, had significantly higher HCV load, genetic distance, and Ka/Ks. Our results are in conflict with other studies in coinfected patients, which have shown an association between genotype 1 infection and increased quasispecies diversity (38) and HCV load (1, 49). However, none of these studies stratified patients based on exposure to HAART. Interestingly, genotype 1 has also been associated with increased severity of liver disease in HCV-monoinfected (56) and -coinfected (10) patients, as well as with a more rapid progression to AIDS (40), although patients in this study were on single-drug antiretroviral therapy and the data may not necessarily extrapolate to patients on HAART. The association of genotype 1 with increased severity of liver disease and more rapid progression to AIDS may be explained by a lower quasispecies diversity (see above discussion) and/or lower CD4+-cell count in these patients, although these parameters were not analyzed in these studies (10, 40, 56). To our knowledge, ours is the first report showing a difference in CD4+-cell count in association with HCV genotype. It may be that HCV, and in particular genotype 1, has an immunomodulatory effect in coinfected patients, possibly by direct interaction of HCV with HIV or via the alteration of cytokine patterns. For example, HCV coinfection has been associated with decreased levels of interleukin 18 (IL-18) and IL-1β in coinfected patients (15); this in turn could lead to decreased CD4 cell proliferation and, in particular, Th1 cell proliferation and differentiation through the IL-18 pathway. In addition, some studies have found that HCV coinfection is associated with a blunted CD4+-cell response to HAART (14; C. Sabin, B. Dauer, A. N. Phillips, T. Lutz, V. Miller, A. C. Lepri, and S. Staszewski, Abstr. 9th Conf. Retrovir. Opportunistic Infect., abstr. 639-M, 2002), although these studies did not stratify patients by HCV genotype. It is provocative to envision a differential immunomodulatory effect of genotype 1, as is suggested by our results here. However, given the small number of patients in our study, we cannot speculate further on the implications of these results, as they demand confirmation in a larger study.

The potential limitations of our study include the relatively small number of patients in each group and the fact that our study population of veterans is mostly representative of the male HIV/HCV-coinfected population, whose main risk factor for infection is IVDU. In addition, 62% of our patients were infected with genotype 2 or 3; this is a higher percentage than that of the general coinfected population in the United States, in which 16% of HCV infections are with genotype 2 or 3 (45). We also focused our analysis solely on HVR1, and there may be other areas of the HCV genome that respond differently to immune pressure and/or to HAART, such as the IFN sensitivity-determining region. Another potential drawback of our study is that we analyzed only 10 clones per time point. We recognize that, the greater number of clones sequenced per time point, the better the assessment of the quasispecies population. However, given that the HCV quasispecies profile of a given group of samples was nearly identical for two independently obtained sets of 10 sequences (Fig. 2 and Table 3), we felt that the sampling of 10 clones per time point was adequate to achieve an accurate analysis of the HCV quasispecies population. In addition, while single-stranded conformational polymorphism is a better way to analyze the complexity of a larger numbers of clones, it does not permit analysis of genetic distance or Ka/Ks. Finally, the use of PCR amplification can at times lead to the introduction of mutations during cloning. However, in almost all cases, a mutation at a given position in HVR1 was present in more than one distinct clone (data not shown) and so was unlikely to represent a mutation introduced during PCR.

In summary, this study analyzes the effect of HAART and immunologic status on HCV load and quasispecies diversity in HIV/HCV-coinfected patients. To our knowledge, this is the first study analyzing the effect of HAART on HCV quasispecies variation. We have shown that there is no immediate effect of HAART on HCV quasispecies or load but that, after long-term HAART, patients had a higher HCV load and increased quasispecies diversity. We hypothesize that the higher immune pressure associated with maximal immune recovery in a given patient after HAART drives HCV to evolve more extensively in an attempt to create escape mutants. The potential implications for increased HCV quasispecies diversity in the coinfected population have been discussed and demand further investigation, especially given the importance of HCV disease in the long-term management of HIV/HCV-coinfected patients.

Acknowledgments

This work was supported in part by the Medical Scholars program at the Stanford University School of Medicine.

We thank Bayer Diagnostics for providing HCV load assay results. We also thank Sharon Lindsay for performing HCV genotype analysis and Suparna Dutt for her assistance in the preparation of HCV RNA.

REFERENCES

• 1.Berger, A., M. von Depka Prondzinski, H. W. Doerr, H. Rabenau, and B. Weber. 1996. Hepatitis C plasma viral load is associated with HCV genotype but not with HIV coinfection. J. Med. Virol. 48**:**339-343. [DOI] [PubMed] [Google Scholar]
• 2.Booth, J. C., U. Kumar, D. Webster, J. Montajardino, and H. C. Thomas. 1998. Comparison of the rate of sequence variation in the hypervariable region of E2/NS1 region of hepatitis C virus in normal and hypogammaglobulinemic patients. Hepatology 27**:**223-227. [DOI] [PubMed] [Google Scholar]
• 3.Bruno, R., P. Sacchi, and G. Filice. 2001. Hepatitis C virus dynamics during antiretroviral therapy. Blood 97**:**3318.. [DOI] [PubMed] [Google Scholar]
• 4.Daar, E. S., H. Lynn, S. Donfield, E. Gomperts, M. W. Hilgartner, W. K. Hoots, D. Chernoff, S. Arkin, W.-Y. Wong, C. A. Winkler, and the Hemophilia Growth and Development Study. 2001. Relation between HIV-1 and hepatitis C viral load in patients with hemophilia. J. Acquir. Immune Defic. Syndr. 26**:**466-472. [DOI] [PubMed] [Google Scholar]
• 5.Daar, E. S., H. Lynn, S. Donfield, E. Gomperts, S. J. O'Brien, M. W. Hilgartner, W. K. Hoots, D. Chernoff, S. Arkin, W.-Y. Wong, C. A. Winkler, and the Hemophilia Growth and Development Study. 2001. Hepatitis C virus load is associated with human immunodeficiency virus type 1 disease progression in hemophiliacs. J. Infect. Dis. 183**:**589-595. [DOI] [PubMed] [Google Scholar]
• 6.Devereux, H. L., D. Brown, G. M. Dusheiko, V. C. Emery, and C. A. Lee. 1997. Long-term evolution of the 5′UTR and a region of NS4 containing a CTL epitope of hepatitis C virus in two haemophilic patients. J. Gen. Virol. 78**:**583-590. [DOI] [PubMed] [Google Scholar]
• 7.Dieterich, D. T. 1999. Hepatitis C virus and human immunodeficiency virus: clinical issues in coinfection. Am. J. Med. 107**:**79S-84S. [DOI] [PubMed]
• 8.Dodig, M., and A. S. Tavill. 2001. Hepatitis C and human immunodeficiency virus coinfections. J. Clin. Gastroenterol. 33**:**367-374. [DOI] [PubMed] [Google Scholar]
• 9.Farci, P., A. Shimoda, A. Coiana, G. Diaz, G. Peddis, J. C. Merpolder, A. Strazzera, D. Y. Chien, S. J. Munoz, A. Balestrieri, R. H. Purcell, and H. J. Alter. 2000. The outcome of acute hepatitis C predicted by the evolution of the viral quasispecies. Science 288**:**339-344. [DOI] [PubMed] [Google Scholar]
• 10.Garcia-Samaniego, J., V. Soriano, J. Castilla, R. Bravo, A. Moreno, J. Carbo, A. Iniguez, J. Gonzalez, F. Munoz, et al. 1997. Influence of hepatitis C virus genotypes and HIV infection on histological severity of chronic hepatitis C. Am. J. Gastroenterol. 92**:**1130-1134. [PubMed] [Google Scholar]
• 11.Gavazzi, G., G. Richallet, P. Morand, O. Bouchard, A. Bosseray, P. Leclercq, and M. Micoud. 1998. Effects of double and triple antiretroviral agents on the HCV viral load in patients coinfected with HIV and HCV. Pathol. Biol. 46**:**412-415. [PubMed] [Google Scholar]
• 12.Goedert, J. J., A. Hatzakis, K. E. Sherman, and M. E. Eyster. 2001. Lack of association of hepatitis C virus load and genotype with risk of end-stage liver disease in patients with human immunodeficiency virus coinfection. J. Infect. Dis. 184**:**1202-1205. [DOI] [PubMed] [Google Scholar]
• 13.Gonzalez-Peralta, R. P., K. Qian, J. Y. She, G. L. Davis, T. Ohno, M. Mizokami, and J. Y. Lau. 1996. Clinical implications of viral quasispecies heterogeneity in chronic hepatitis C. J. Med. Virol. 49**:**242-247. [DOI] [PubMed] [Google Scholar]
• 14.Greub, G., B. Ledergerber, M. Battegay, P. Grob, L. Perrin, H. Furrer, P. Burgisser, P. Erb, K. Boggian, J. C. Piffaretti, B. Hirschel, P. Janin, P. Francioli, M. Flepp, and A. Telenti. 2000. Clinical progression, survival, and immune recovery during antiretroviral therapy in patients with HIV-1 and hepatitis C virus coinfection: the Swiss HIV Cohort Study. Lancet 356**:**1800-1805. [DOI] [PubMed] [Google Scholar]
• 15.He, L., H. Terunuma, H. Hanabusa, A. Iwamoto, S. Oka, F. Tanabe, N. Chiba, M. Kurimoto, M. Ikeda, H. Okamura, J. Dai, Y. Iwatani, T. Ishida, and M. Ito. 2000. Interleukin 18 and interleukin 1β production is decreased in HIV type 1-seropositive hemophiliacs but not in HIV type 1-seropositive nonhemophiliacs. AIDS Res. Hum. Retrovir. 16**:**345-353. [DOI] [PubMed] [Google Scholar]
• 16.Honda, M., S. Kaneko, A. Sakai, M. Unoura, S. Murakami, and K. Kobayashi. 1994. Degree of diversity of hepatitis C virus quasispecies and progression of liver disease. Hepatology 20**:**1144-1151. [PubMed] [Google Scholar]
• 17.Imami, N., C. Antonopoulos, G. A. Hardy, B. Gazzard, and F. M. Gotch. 1999. Assessment of type 1 and type 2 cytokines in HIV type 1-infected individuals: impact of highly active antiretroviral therapy. AIDS Res. Hum. Retrovir. 15**:**1499-1508. [DOI] [PubMed] [Google Scholar]
• 18.Kao, J. H., P. J. Chen, M. Y. Lai, T. H. Wang, and D. S. Chen. 1995. Quasispecies of hepatitis C virus and genetic drift of the hypervariable region in chronic type C hepatitis. J. Infect. Dis. 172**:**261-264. [DOI] [PubMed] [Google Scholar]
• 19.Kato, N., Y. Ootusuyama, T. Tanaka, M. Nakazawa, M. Hijikata, and K. Shimotohno. 1994. Genetic drift in hypervariable region 1 of the viral genome in persistent hepatitis C virus infection. J. Virol. 68**:**4776-4784. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 20.Kiyosawa, K., T. Sodeyama, E. Tanaka, Y. Gibo, K. Yoshizawa, Y. Nakano, S. Furuta, Y. Akahane, K. Nishioka, R. H. Purcell, et al. 1990. Interrelationship of blood transfusion, non-A, non-B hepatitis and hepatocellular carcinoma: analysis by detection of antibody to hepatitis C virus. Hepatology 12**:**671-675. [DOI] [PubMed] [Google Scholar]
• 21.Koizumi, K., N. Enomoto, N. Kurosaki, T. Murakami, N. Izumi, F. Marumo, and C. Sato. 1995. Diversity of quasispecies in various disease stages of chronic hepatitis C virus infection and its significance in interferon treatment. Hepatology 22**:**30-35. [PubMed] [Google Scholar]
• 22.Kumar, S., K. Tamura, I. B. Jakobsen, and M. Nei. 2001. MEGA2: molecular evolutionary genetics analysis software. Bioinformatics 17**:**1244-1245. [DOI] [PubMed] [Google Scholar]
• 23.Kumar, U., J. Monjardino, and H. C. Thomas. 1994. Hypervariable region of hepatitis C virus envelope glycoprotein (E2/NS1) in an agammaglobulinemic patient. Gastroenterology 106**:**1072-1075. [DOI] [PubMed] [Google Scholar]
• 24.Kurosaki, M., N. Enomoto, F. Marumo, and C. Sato. 1993. Rapid sequence variation in the hypervariable region of hepatitis C virus during the course of chronic infection. Hepatology 18**:**1293-1299. [PubMed] [Google Scholar]
• 25.Lau, J. Y., M. Mizokami, J. A. Kolberg, G. L. Davis, L. E. Prescott, T. Ohno, R. P. Perrillo, K. L. Lindsay, R. G. Gish, K. P. Qian, et al. 1995. Application of six hepatitis C virus genotyping systems to sera from chronic hepatitis C patients in the United States. J. Infect. Dis. 171**:**281-290. [DOI] [PubMed] [Google Scholar]
• 26.Lawal, Z., J. Petrik, V. S. Wong, G. J. Alexander, and J. P. Allain. 1997. Hepatitis C virus genome variability in untreated and immunosuppressed patients. Virology 228**:**107-111. [DOI] [PubMed] [Google Scholar]
• 27.Mao, Q., S. C. Ray, O. Laeyendecker, J. R. Ticehurst, S. A. Strathdee, D. Vlahov, and D. L. Thomas. 2001. Human immunodeficiency virus seroconversion and evolution of the hepatitis C virus quasispecies. J. Virol. 75**:**3259-3267. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 28.Martell, M., J. I. Esteban, J. Quer, J. Genesca, A. Weiner, R. Esteban, J. Guardia, and J. Gomez. 1992. Hepatitis C virus (HCV) circulates as a population of different but closely related genomes: quasispecies nature of HCV genome distribution. J. Virol. 66**:**3225-3229. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 29.Martell, M., J. I. Esteban, J. Quer, V. Vargas, R. Esteban, J. Guardia, and J. Gomez. 1994. Dynamic behavior of hepatitis C virus in quasispecies in patients undergoing orthotopic liver transplantation. J. Virol. 68**:**3425-3436. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 30.Martín, M., S. Echecarría, F. Leyva-Cobián, I. Pereda, and M. López-Hoyos. 2001. Limited immune reconstitution at intermediate stages of HIV-1 infection during one year of highly active antiretroviral therapy in antiretroviral-naive versus non-naive adults. Eur. J. Microbiol. Infect. Dis. 20**:**871-879. [DOI] [PubMed] [Google Scholar]
• 31.Naito, M., N. Hayashi, T. Moribe, H. Hagiwara, E. Mita, Y. Kanazawa, A. Kasahara, H. Fusamoto, and T. Kamada. 1995. Hepatitis C viral quasispecies in hepatitis C virus carriers with normal liver enzymes and patients with type C chronic liver disease. Hepatology 22**:**407-412. [PubMed] [Google Scholar]
• 32.Nei, M., and T. Gojobori. 1986. Simple methods for estimating the numbers of synonymous and nonsynonymous nucleotide substitutions. Mol. Biol. Evol. 3**:**418-426. [DOI] [PubMed] [Google Scholar]
• 33.Ni, Y.-H., M.-H. Chang, P.-J. Chen, H.-Y. Hsu, T.-W. Lu, K.-H. Lin, and D.-T. Lin. 1999. Decreased diversity of hepatitis C virus quasispecies during bone marrow transplantation. J. Med. Virol. 58**:**132-138. [DOI] [PubMed] [Google Scholar]
• 34.Puoti, M., F. Gargulio, E. Q. Roldan, A. Chiodera, L. Palvarini, A. Spinetti, S. Zaltron, V. Putzolo, B. Zanini, F. Favilli, A. Turano, and G. Carosi. 2000. Liver damage and kinetics of hepatitis C virus and human immunodeficiency virus replication during the early phases of combination antiretroviral treatment. J. Infect. Dis. 181**:**2033-2036. [DOI] [PubMed] [Google Scholar]
• 35.Ragni, M. V., and F. A. Bontempo. 1999. Increase in hepatitis C virus load in hemophiliacs during treatment with highly active antiretroviral therapy. J. Infect. Dis. 180**:**2027-2029. [DOI] [PubMed] [Google Scholar]
• 36.Ray, S. C., Y.-M. Wang, O. Laeyendecker, J. R. Ticehurst, S. A. Villano, and D. L. Thomas. 1999. Acute hepatitis C virus structural gene sequences as predictors of persistent viremia: hypervariable region 1 as a decoy. J. Virol. 73**:**2938-2946. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 37.Rockstroh, J. K., A. Theisen, R. Kaiser, T. Sauerbruch, and U. Spengler. 1998. Antiretroviral triple therapy decreases HIV viral load but does not alter hepatitis C virus (HCV) serum levels in HIV-HCV co-infected haemophiliacs. AIDS 12**:**829-830. [PubMed] [Google Scholar]
• 38.Roque-Afonso, A. M., M. Robain, D. Simoneau, P. Rodriguez-Mathieu, M. Gigou, L. Meyer, and E. Dussaix. 2002. Influence of CD4 cell counts on the genetic heterogeneity of hepatitis C virus in patients coinfected with human immunodeficiency virus. J. Infect. Dis. 185**:**728-733. [DOI] [PubMed] [Google Scholar]
• 39.Rutschman, O. T., F. Negro, B. Hirsschel, A. Hadengue, D. Anwar, and L. H. Perrin. 1998. Impact of treatment with human immunodeficiency virus (HIV) protease inhibitors on hepatitis C viremia in patients coinfected with HIV. J. Infect. Dis. 177**:**783-785. [DOI] [PubMed] [Google Scholar]
• 40.Sabin, C. A., P. Telfer, A. N. Phillips, S. Bhagani, and C. A. Lee. 1997. The association between hepatitis C virus genotype and human immunodeficiency virus disease progression in a cohort of hemophiliac men. J. Infect. Dis. 175**:**164-168. [DOI] [PubMed] [Google Scholar]
• 41.Sánchez-Fueyo, A., M. Giménez-Barcons, F. Puig-Basagoiti, A. Rimola, J. M. Sánchez-Tapias, J. C. Sáiz, and J. Rodés. 2001. Influence of the dynamics of the hypervariable region 1 of hepatitis C virus (HCV) on the histological severity of HCV recurrence after liver transplantation. J. Med. Virol. 65**:**266-275. [DOI] [PubMed] [Google Scholar]
• 42.Seibert, S. A., C. Y. Howell, M. K. Hughes, and A. L. Hughes. 1995. Natural selection on the gag, pol, and env genes of human immunodeficiency virus 1 (HIV-1). Mol. Biol. Evol. 12**:**803-813. [DOI] [PubMed] [Google Scholar]
• 43.Sherman, K. E., C. Andreatta, J. O'Brien, A. Gutierrez, and R. Harris. 1996. Hepatitis C in human immunodeficiency virus-coinfected patients: increased variability in the hypervariable envelope coding domain. Hepatology 23**:**688-694. [DOI] [PubMed] [Google Scholar]
• 44.Sherman, K. E., J. O'Brien, A. G. Gutierrez, S. Harrison, M. Urdea, P. Neuwald, and J. Wilber. 1993. Quantitative evaluation of hepatitis C virus RNA in patients with concurrent human immunodeficiency virus infections. J. Clin. Microbiol. 31**:**2679-2682. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 45.Sherman, K. E., S. D. Rouster, R. T. Chung, and N. Rajicic. 2002. Hepatitis C virus prevalence among patients infected with human immunodeficiency virus: a cross sectional analysis of the US adult AIDS clinical trials group. Clin. Infect. Dis. 34**:**831-837. [DOI] [PubMed] [Google Scholar]
• 46.Soriano, V., J. Garcia-Samaniego, R. Bravo, J. Gonzalez, A. Castro, J. Castilla, P. Martinex-Odriozola, M. Colmenero, E. Carballo, D. Suarez, F. J. Rodriguez-Pinero, A. Moreno, J. del Romero, J. Pedreira, J. Gonzalez-Lahoz, et al. 1996. Interferon alpha for the treatment of chronic hepatitis C in patients infected with human immunodeficiency virus. Clin. Infect. Dis. 23**:**585-591. [DOI] [PubMed] [Google Scholar]
• 47.Soto, B., A. Sanchez-Quijano, L. Rodrigo, J. A. del Olmo, M. Garcia-Bengoechea, J. Hernandez-Quero, C. Rey, M. A. Abad, M. Rodriguez, M. Sales Gilabert, F. Gonzalez, P. Miron, A. Caruz, F. Relimpio, R. Torronteras, M. Leal, and E. Lissen. 1997. Human immunodeficiency virus infection modifies the natural history of chronic parenterally-acquired hepatitis C with an unusually rapid progression to cirrhosis. J. Hepatol. 26**:**1-5. [DOI] [PubMed] [Google Scholar]
• 48.Tong, M. J., L. M. Blatt, J. G. McHutchison, R. L. Co, and A. Conrad. 1997. Prediction of response during interferon alfa 2b therapy in chronic hepatitis C patients using viral and biochemical characteristics: a comparison. Hepatology. 26**:**1640-1645. [DOI] [PubMed] [Google Scholar]
• 49.Torre, D., R. Tambini, F. Cadario, G. Barbarini, M. Moroni, and C. Basilico. 2001. Evolution of coinfection with human immunodeficiency virus and hepatitis C virus in patients treated with highly active antiretroviral therapy. Clin. Infect. Dis. 33**:**1579-1585. [DOI] [PubMed] [Google Scholar]
• 50.Toyoda, H., Y. Fukuda, Y. Koyama, J. Takamatsu, H. Saito, and T. Hayakawa. 1997. Effect of immunosuppression on composition of quasispecies population of hepatitis C virus in patients with chronic hepatitis C coinfected with human immunodeficiency virus. J. Hepatol. 26**:**975-982. [DOI] [PubMed] [Google Scholar]
• 51.Weiner, A. J., H. M. Geysen, C. Christopherson, J. E. Hall, T. J. Mason, G. Saracco, F. Bonino, K. Crawford, C. D. Marion, K. A. Crawford, M. Brunetto, P. J. Barr, T. Miyamura, J. McHutchinson, and M. Houghton. 1992. Evidence for immune selection of hepatitis C virus (HCV) putative envelope glycoprotein variants: potential role in chronic HCV infections. Proc. Natl. Acad. Sci. USA 89**:**3468-3472. [DOI] [PMC free article] [PubMed] [Google Scholar]
• 52.Wolinsky, S. M., B. T. Korber, A. U. Neumann, M. Daniels, K. J. Kuntsman, A. J. Whetsell, M. R. Furtado, Y. Cao, D. D. Ho, and J. T. Safrit. 1996. Adaptive evolution of human immunodeficiency virus-type 1 during the natural course of infection. Science 272**:**537-542. [DOI] [PubMed] [Google Scholar]
• 53.Wright, T. L., H. Hollander, X. Pu, M. J. Held, P. Lipson, S. Quan, A. Polito, M. M. Thaler, P. Bacchetti, and B. F. Scharschimdt. 1994. Hepatitis C in HIV-infected patients with and without AIDS: prevalence and relationship to patient survival. Hepatology 20**:**1152-1155. [PubMed] [Google Scholar]
• 54.Yokozaki, S., J. Takamatsu, I. Nakano, Y. Katano, H. Toyoda, K. Hayashi, T. Hayakawa, and Y. Fukuda. 2001. Highly active antiretroviral therapy and interferon therapy for HIV-HCV-coinfected patients: which should precede? Blood 97**:**3318-3319.11368064 [Google Scholar]
• 55.Yokozaki, S., J. Takamatsu, I. Nakano, Y. Katano, H. Toyoda, K. Hayashi, T. Hayakawa, and Y. Fukuda. 2000. Immunologic dynamics in hemophiliac patients infected with hepatitis C virus and human immunodeficiency virus: influence of antiretroviral therapy. Blood 96**:**4293-4299. [PubMed] [Google Scholar]
• 56.Zein, N. N., J. Rakela, E. Krawitt, K. R. Reddy, T. Tominaga, and D. Persing. 1996. Hepatitis C virus genotypes in the United States: epidemiology, pathogenicity, and response to interferon therapy. Ann. Intern. Med. 125**:**634-639. [DOI] [PubMed] [Google Scholar]