By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Due to essential maintenance the subscribe/renew pages will be unavailable on Wednesday 26 October between 02:00- 08:00 BST/ 09:00 – 15:00 SGT/ 21:00- 03:00 EDT. Apologies for the inconvenience.
HCV sequences have been submitted to GenBank under accession number AY629502-AY629554.
Liver cirrhosis caused by chronic hepatitis C virus (HCV) infection is the main indication for liver transplantation (LT). There is little information on HCV genetic evolution following transplantation. The aim of this study was to carefully assess early evolution of HCV quasispecies in a cohort of 18 liver transplant recipients followed prospectively. Quasispecies analysis was performed by sequence analysis of the hypervariable region 1 (HVR1) before transplantation and at day 4 and week 4 following LT. A predominant variant was present in 12 (67%) of the 18 patients before transplantation and the same variant was propagated and remained predominant after LT in 6 (50%) of these patients. In the remaining individuals, there were major changes in the quasispecies composition, mostly occurring during the first days after LT. There was a progressive decrease in the nonsynonymous (dN)/synonymous (dS) ratios from baseline (1.2) to day 4 (.6) (P = .08) and to week 4 after LT (.3) (P = .015). Similarly, genetic distance (GD) declined from baseline (.1) to day 4 (.03) (P = .07) and to week 4 (.04) (P = .04). We did not find any differences in HCV genetic evolution between patients with mild (n = 10) or severe (n = 8) disease recurrence. In conclusion, during the first days following transplantation, HCV quasispecies becomes more homogenous, even after major changes in its composition. Importantly, these changes persist and even increase during the 1st month after transplantation. The “bottleneck” effect caused by the implantation of a new graft and the lack of selective pressure due to the strong immunosuppression most likely explain this particular pattern of genetic evolution. (Liver Transpl 2004;10:1131–1139.)
Liver cirrhosis caused by chronic hepatitis C virus (HCV) infection is the main indication for liver transplantation (LT). HCV infects the liver graft in almost all HCV-infected patients following transplantation and causes a persistent infection that leads to chronic hepatitis and cirrhosis in a significant proportion of patients.1–3
HCV replication is associated with a high rate of nucleotide mutations, which explains that in infected individuals, HCV circulates as a mixture of closely related but distinct genomes (quasispecies).4, 5 The quasispecies nature of HCV has been implicated in viral persistence, and, indeed, genetic evolution of the virus might permit HCV to escape the host immune surveillance.6 LT is a unique opportunity to study HCV genetic evolution, since a circulating viral quasispecies that is in dynamic equilibrium, requires a quick adaptation to a situation derived from the removal of the main source of virus production (the liver) and the implantation of an uninfected graft. To propagate, circulating virions need to attach and enter into hepatocytes, the first step of the HCV life cycle. These events will most likely result in the selection of certain viral variants and, therefore, in significant changes in the quasispecies composition. It is obvious that studying genetic evolution in this setting is complex, since infection occurs in an environment of already established immune responses against the virus but also under strong immunosuppression.
There is only 1 study focusing on changes in HCV populations immediately following transplantation. HCV quasispecies isolated from the graft only a few hours after graft reperfusion was significantly more homogeneous than pretransplantation circulating quasispecies, suggesting a specific interaction of viral envelope proteins with a receptor.7 The main aim of our study was to carefully describe HCV genetic evolution during the early phase of LT. Our hypothesis was that there must be a common pattern of changes in HCV populations when HCV infects a new graft in the setting of strong immunosuppression. As a secondary endpoint, we analyzed if HCV genetic evolution differed between patients with different clinical outcomes after transplantation.
A total of 18 patients undergoing LT for HCV-related cirrhosis were included in the study. Inclusion criteria were the presence of HCV-ribonucleic acid (RNA) before LT and a follow-up liver biopsy demonstrating either mild or severe liver fibrosis (see below). Exclusion criteria were coinfection with the hepatitis B virus, fibrosing cholestatic hepatitis, and graft/patient survival shorter than 1 year. Indication for LT was decompensated cirrhosis in 7 patients and hepatocellular carcinoma in the remaining 11 patients.
Induction immunosuppression regimen was cyclosporine A or tacrolimus plus prednisone, as previously described.8 Mycophenolate mofetil was added in patients who required cyclosporine or tacrolimus dose reduction or discontinuation. Prednisone tapering was initiated 1 month after LT and the drug was discontinued between 6 and 12 months thereafter.
All individuals were followed prospectively and underwent protocol liver biopsies at 3, 12, and 24 months following LT. Histological assessment was performed by a single pathologist using the Scheuer score.9 The presence of bridging fibrosis or cirrhosis (F3 or F4) was considered severe recurrence, whereas the absence of fibrosis or fibrosis restricted to the portal tract was considered mild disease recurrence.
HCV Genotyping and Viral Load Determination
HCV genotype was determined by restriction fragment length polymorphism analysis of the 5′ untranslated region, as previously described.10 Viral load was measured by a commercially available assay (Amplicor Monitor Roche Diagnostic Systems v.2, Branchburg, NJ) in serum samples collected before LT, daily during the first week, and weekly during the first month after LT. Samples with HCV-RNA concentration exceeding 500,000 IU/mL were diluted to 1/100 and retested.
Amplification and Cloning of the Hypervariable Region 1 (HVR1)
HCV quasispecies were studied in serum samples taken immediately before LT and at day 4 and week 4 following LT. In hemodynamically stable patients, the Swan-Ganz catheter (Baxter, Irvine, CA), which is usually left the first 48 hours after transplantation, was replaced by a catheter in the hepatic veins (HV). A serum sample taken at day 4 directly from the HV was used for HCV quasispecies analysis.3
Total RNA was extracted from 100 μL of serum using a sensitive method (Trizol Reagent; Invitrogen S.A., Scotland, UK), following the manufacturer's instructions. Amplification of the HVR1 was performed as previously described, using specific primers for each genotype.8
All PCR products were analyzed in agarose gels stained with ethidium bromide, gel purified (QIAquick PCR purification kit; Qiagen, Hilden, Germany), incubated with Taq DNA polymerase in the presence of dATP, and ligated into pGEM-T easy using T4 DNA ligase (Promega, Madison, WI). Competent E. coli JM109 cells were transformed and plated on Luria broth agar plates containing ampicillin (100 μg/mL), IPTG (.5 mM), and X-Gal (80 μg/mL), and incubated overnight at 37°C. A minimum of 10 clones per sample were randomly grown in Luria broth medium containing 100 μg/mL of ampicillin. Plasmid DNA from the cultures was extracted with the Wizard Plus SV Minipreps DNA Purification System (Promega) and then purified by ethanol precipitation. All purified plasmids were sequenced using an ABI Prism Big Dye Terminator Ready Reaction Kit, v 2.0 (Applied Biosystems, Foster City, CA) using pGEM-T specific primers. The sequencing reactions were performed following the manufacturer's instructions.
Sequences were analyzed by visual inspection of the electropherograms (Sequence Navigator v1.0.1, Perkin Elmer Software) and aligned with CLUSTAL W.11 Quasispecies analysis was performed using the MEGA and DAMBE programs.12, 13
The genetic diversity, defined as the frequency of mutations within the different isolates of HCV quasispecies, is expressed in genetic distance (GD) terms. The intrasample GD of the quasispecies was estimated by pairwise comparison of all amino acid sequences using the p-distance method.14
The average number of synonymous (silent) nucleotide substitutions per synonymous site (dS) and the number of nonsynonymous (amino acid replacement) nucleotide substitutions per nonsynonymous site (dN) were estimated with the Nei Gojobori Method using the Jukes Cantor correction.15
The genetic complexity, defined as the number of viral variants within HCV quasispecies, is expressed as normalized Shannon Entropy.16 Shannon Entropy was calculated at the nucleotide and amino acid level using the following equation:
(where pi is the relative frequency of each sequence in the viral quasispecies and N is the total number of sequences analyzed). Shannon Entropy theoretically varies from 0 (all isolates identical) to 1 (all isolates are different).
Quantitative variables are expressed as median (range). For categorical variables, differences between groups were calculated by the Fisher's exact test. For quantitative variables, differences between groups were analyzed using a non-parametric test (Mann-Whitney for unpaired groups and Wilcoxon for paired groups).
The median age of the patient cohort was 57 years (range 25–65 years); 11 patients were male. A total of 13 patients were infected with genotype 1b, 3 with genotype 1a, 1 with genotype 2, and 1 with genotype 4. Median viral load before transplantation was 5.5 log10 IU/mL (range 4.2–6.2). One patient received antiviral therapy before transplantation, but did not achieve virological response. HCV infection recurred in all 18 patients after LT. None of the 18 patients received antiviral therapy before week 4 following LT.
Median follow-up of the patient cohort was 33 months (range 24–36). In 8 of the 18 patients, follow-up liver biopsies established the presence of bridging fibrosis or cirrhosis (F3 or F4) and disease recurrence was considered severe. In fact, portal hypertension (hepatic venous portal gradient > 6 mm Hg) was demonstrated in all 5 patients in whom a hepatic hemodynamic study was available. In the remaining 10 individuals, follow-up liver biopsies showed no fibrosis or fibrosis restricted to the portal tract and disease recurrence was considered mild; portal pressure was normal in all 6 patients in whom a hemodynamic study was available.
Longitudinal Analysis of Quasispecies Parameters in the Entire Cohort
As stated above, HCV quasispecies were studied in serum samples taken immediately before LT, at day 4, and week 4 following LT. Only 1 sample tested negative for HCV-RNA (day 4 for patient 13) and could not be analyzed. In 7 patients, a serum sample obtained directly from the HV was available for quasispecies analysis. Overall, a total of 610 clones were sequenced and analyzed (median 12 clones per time point, range 10–20).
The main findings regarding quasispecies evolution in our patient cohort are summarized in Table 1. To exclude laboratory cross-contamination, a phylogenetic tree was constructed with the consensus HVR1 sequences from all patients at the 3 analyzed time points. As expected, posttransplantation HCV isolates were more closely related to their respective pretransplantation isolates than any other variant (data not shown).
Table 1. Genetic Evolution of HCV After Liver Transplantation (LT)*
At the amino acid level, a predominant variant (arbitrarily defined as the presence of a variant representing >40% of the quasispecies population) was present in 12 (67%) of the 18 patients before LT. The same variant was propagated and remained predominant after LT in 6 (50%) of these patients. In the remaining individuals, there were major changes in the quasispecies composition following LT. Importantly, the amino acid consensus sequence changed in 12 of 18 individuals following LT. Most amino acid substitutions occurred during the first week after transplantation (Fig. 1); thereafter, major changes in the consensus sequence were detected in only 4 individuals (patients 1, 6, 16, and 18).
Viral load decreased during the first hours following LT and reached its lowest value 12 hours after the procedure. By day 4 after LT, however, HCV-RNA concentrations were significantly higher than the lowest value in 12 (67%) of 18 patients (median increase 35%, range 11–106%). In the remaining 6 patients, viral load increased at a significantly slower rate. These differences in viral kinetics were not associated either with the presence of a predominant variant before LT or with its propagation after transplantation.
Changes in Quasispecies Parameters Following LT
Entropy values decreased in 11 (61%) of 18 patients following LT. The reduction in entropy over time, however, did not reach statistical significance (Table 1 and Fig. 2).
There was a progressive decrease in dN over time after LT that resulted in a significant reduction in the dN/dS ratio from baseline (1.2, range .35–9.7) to day 4 (.6, range .1–1.9) (P = .08) and to week 4 after LT (.3, range .01–5) (P = .015) (Table 1 and Fig. 2). In fact, dN/dS ratios decreased in 12 (67%) of 18 patients after LT. In those 6 patients in whom the ratio did not decrease after LT, dN/dS at baseline (before LT) was significantly lower (.48, range .35–1.47) than in the remaining 12 patients (1.4, range .39–9.8) (P = .03).
There was also a sharp decrease in GD following LT. GD declined from .1 (range .006–.3) before LT to .03 (range .00–.4) at day 4 (P = .07) and to .04 (range .006–.3) at week 4 after LT (P = .04) (Table 1 and Fig. 2). In fact, GD decreased in 12 (70%) of 17 patients at day 4 following transplantation and in 13 (72%) of 18 patients at week 4. As occurred with dN/dS ratio, GD before transplantation was significantly lower in those patients in whom GD did not decrease after LT (.06, range .006–.13) compared to the remaining 13 patients (.22, range .05–.28) (P = .01). Changes in GD using nucleotide distance methods were similar to those described above: GD decreased in 12 (67%) of 18 patients at day 4 and week 4 after LT, but the differences did not reach statistical significance (P = .08 and P = .2 for day 4 and week 4, respectively).
Changes in quasispecies parameters were nearly identical if analysis was restricted to individuals infected with genotype 1b (Table 1). In fact, there was a significant decrease in dN/dS ratio from baseline (1.03, range .35–9.7) to day 4 (.47, range .1–1.9) (P = .06) and to week 4 after LT (.38, range .01–1.4) (P = .005) (Table 1). Similarly, there was a sharp decrease in GD from baseline (.13, range .006–.3) to day 4 (.02, range .00–.4) (P = .03) and to week 4 after LT (.03, range .006–.3) (P = .03) (Table 1).
We investigated if changes in quasispecies parameters differed between patients with a rapid increase in viral load (n = 12) and those individuals in whom HCV-RNA concentrations increased at a slower rate (n = 6). Changes in dN/dS ratio and in GD, however, were similar between both groups.
Quasispecies Parameters and Severity of HCV Disease Recurrence
We analyzed if changes in quasispecies populations were associated with disease outcome. A careful comparison of all quasispecies parameters between patients with (n = 8) and without (n = 10) severe disease recurrence did not disclose significant differences among them (Table 2). The median dN/dS ratio before transplantation was higher than 1 in patients with mild recurrence and lower than 1 in those with severe recurrence, but the difference did not reach statistical significance (P = .2).
Table 2. Genetic Evolution of HCV After Liver Transplantation*
Quasispecies parameters were analyzed in the hypervariable region 1 (HVR1) in patients with mild and severe disease recurrence. Values are expressed as median (range).
HVR1 region, mild recurrence (n = 10)
HVR1 region, severe recurrence (n = 8)
We did not find a particular pattern of quasispecies evolution in 5 individuals treated with methylprednisolone boluses due to acute cellular rejection.
Analysis of HCV Quasispecies in Serum Obtained From the Systemic Circulation (SC) and From the Hepatic Vein (HV)
Serum samples from the SC and HV obtained at day 4 following LT were available in 7 patients. Analysis of HCV quasispecies revealed a very similar distribution of viral variants in both sites, with slight differences in the composition of minor quasispecies (Fig. 3). The latter can be explained by random sampling. Interestingly, in 1 patient (patient 5), the predominant variant in the SC was a minor variant in the HV and a variant not represented in the SC was predominant in the HV.
We recently showed that following LT there is a sharp decrease in HCV-RNA concentrations; however, viral load increases very rapidly following transplantation and in some patients it reaches levels similar or even higher than before transplantation in only a few days.3 Based on these data, we decided to study HCV genetic evolution by analyzing quasispecies composition before transplantation and at 2 early time points (day 4 and week 4) after LT.
Following transplantation, we observed a rapid and significant decrease in the quasispecies dN/dS ratio and in GD, resulting in a considerably more homogenous viral population. Even in patients with major changes in HCV quasispecies, the population that propagated after transplantation was significantly more homogeneous than before transplantation. A similar phenomenon has been described after an acute HCV infection, where initial quasispecies are more uniform than those present in the infectious contact at the time of transmission.17, 18 After implantation of the new graft, circulating virions need to infect hepatocytes and initiate their life cycle. Although the mechanisms of HCV attachment and entry into hepatocytes are still controversial,19, 20 its is likely that not all pre-existing viral variants within a quasispecies will have the same ability to infect the graft. Recently, Hughes et al.7 demonstrated that only 1.5 to 2.5 hours after graft reperfusion, the viral quasispecies found in the liver was significantly more homogenous than that found in the serum before transplantation. These very early changes in the quasispecies composition strongly suggest that attachment and entry of HCV into hepatocytes generate a “bottleneck” effect4, 5, 21 by selecting certain variants from the viral population. We can not exclude, however, that these early changes in quasispecies composition reflect different replication abilities among distinct HCV variants. In fact, data generated using the HCV replicon system suggest that the appearance of adaptive mutations may result in a significant increase in replication efficiency in cell culture.22
The progressive decrease in dN/dS ratios and in GD that occurred immediately after transplantation persisted or became even more significant 3 weeks later. After the initial “bottleneck” effect caused by implantation of the new graft, the increasing homogeneity of HCV quasispecies is most likely explained by the deficient immune responses against HCV. In fact, immunosuppression is highest during the first 4 weeks after LT. Similar changes in HCV genetic evolution have been documented in individuals with different forms of immunosuppression. It is well known that in patients with agammaglobulinemia, where antibody production is severely compromised, there is a decrease in the accumulation of mutations in the regions encoding the envelope proteins.23–25 Furthermore, when HCV infected patients acquire the human immunodeficiency virus, there is a decrease in dN/dS ratio that reveals lower immune pressures.26 On the contrary, the increase in genetic diversity observed in immunocompetent patients with acute hepatitis progressing to chronicity occurs after anti-HCV seroconversion, that is, after immune responses against the virus are elicited.27
Except for 1 patient, we did not find significant differences in the HCV quasispecies composition between samples obtained from the HV and samples obtained from the SC at day 4 after LT. This results, along with a rapid and significant increase in viral load in a significant proportion of patients, support that the liver is the major HCV production site. In 1 of the analyzed patients, however, the composition of HCV quasispecies differed significantly between both sites. In this individual, we cannot exclude the existence of a second compartment (such as peripheral blood mononuclear cells) with active HCV replication that contributed to the observed differences.28
A secondary aim of our study was to analyze evolution of HCV quasispecies in patients with different disease outcomes. Our data do not support a relationship between the severity of HCV recurrence after LT and a specific pattern of quasispecies evolution in the early posttransplant period.8 We are aware, however, that these results need to be interpreted with caution due to the small sample size and the fact that variables not inherently related to the infecting HCV variants may influence the outcome of the disease. Other authors have shown a different pattern of quasispecies evolution depending on the severity of HCV recurrence in the liver graft. Pessoa et al.29 found that in patients with severe HCV recurrence, evolution of HCV quasispecies was more pronounced and HVR1 divergence occurred earlier in the course of transplantation. The data published by Pessoa can not be compared with ours for various reasons. First, patients included in the above mentioned study had a particular severe form of HCV recurrence (severe cholestatic hepatitis leading to liver failure) and these patients were excluded from our analysis. Second, analysis of HCV quasispecies was performed considerably late after transplantation. Other studies found that genetic diversification increased after LT in patients with mild recurrence, whereas quasispecies populations became more homogenous in patients with severe recurrence.30, 31 Despite the similarities between the patients described in the above mentioned studies and ours, viral evolution was evaluated several months after transplantation. Immunosuppression is significantly weaker after the third month following transplantation, and it is entirely possible that restoration of the immune responses contributes to the observed patterns of HCV genetic evolution. Although we did not observe any significant differences in pretransplantation quasispecies composition and long-term outcome of HCV recurrence,32 there was a trend towards a higher baseline dN / dS ratio in patients with mild compared to those with severe HCV disease recurrence. This might suggest a stronger selective pressure (most likely immune pressure) in those individuals with a better outcome after transplantation.
Our study presents some caveats. First, when analyzing a limited number of clones per time point, the sequenced variants may not always be fully representative of the whole population of circulating quasispecies. Second, we have assumed that quasispecies composition at day 4 is representative of newly replicating virions produced in the liver, and not the result of residual circulating virions or virus released from extrahepatic sites. However, the significant increase in viral load by day 4 after transplantation in a significant proportion of patients, and a similar genetic evolution of HCV quasispecies in individuals with different HCV kinetics profiles, support early viral replication in the new graft. Moreover, the similar composition of HCV quasispecies in blood obtained from the peripheral circulation and from the HV at day 4 further supports this hypothesis.
In summary, during the first days and weeks following LT, HCV quasispecies becomes more homogenous despite the occurrence of major changes in its composition. The “bottleneck” effect caused by the implantation of a new graft and the strong immunosuppression most likely explain this particular pattern of HCV genetic evolution. HCV is able to adapt rapidly to the new environment generated after transplantation and replicate efficiently.
The authors thank Amaia Amador from Unitat de Genòmica, Serveis Cientifico-tècnics, Universitat de Barcelona, Parc científic for her excellent technical support in sequencing. The authors also thank Antonio Mas from Universitat Pompeu Fabra for his scientific input.