- Email: [email protected]
Hepatitis C virus RNA recombination in cell culture
Research Article
Hepatitis C virus RNA recombination in cell culture Jochen Reiter1, ,#, Gemma Pérez-Vilaró2, , Nicoletta Scheller1,2, , Leonardo Bruno Mina2, Juana Díez2, Andreas Meyerhans1,3,⇑ 1
Department of Virology, Saarland University, D-66421 Hamburg, Germany; 2Molecular Virology group; 3ICREA Infection Biology Group, Department of Experimental and Health Sciences, Universitat Pompeu Fabra, 08003 Barcelona, Spain
Background & Aims: The Hepatitis C virus (HCV) exhibits large genetic diversity, both on a global scale and at the level of the infected individual. A major underlying mechanism of the observed sequence differences is error-prone virus replication by the viral RNA polymerase NS5B. In addition, based on phylogenetic comparisons of patient-derived HCV sequences, there is evidence of HCV recombination. However, to date little is known about the frequency by which recombination events occur in HCV and under what conditions recombination may become important in HCV evolution. We, therefore, aimed to set up an experimental model system that would allow us to analyze and to characterize recombination events during HCV replication. Methods: A neomycin-selectable, HCV replicon-based recombination detection system was established. HCV replicons were mutated within either the neomycin-phosphotransferase gene or the NS5B polymerase. Upon co-transfection of hepatic cells lines, recombination between the mutated sites is necessary to restore the selectable phenotype. Results: Recombinants were readily detected with frequencies correlating to the distance between the mutations. The recombinant frequency normalized to a crossover range of one nucleotide was around 4 108. Conclusions: An experimental system to select for HCV recombinants in cell culture was successfully established. It allowed deriving first estimates of recombinant frequencies. Based on these, recombination in HCV seems rare. However, due to the rapid virus turnover and the large number of HCV-infected liver cells in vivo, it is expected that recombination will be of biological importance when strong selection pressures are operative. Ó 2011 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.
Keywords: Recombination; RNA; HCV; HCV replicon; Virus evolution; Positivestrand RNA virus. Received 30 June 2010; received in revised form 14 December 2010; accepted 20 December 2010; available online 18 February 2011 ⇑ Corresponding author. Address: ICREA Infection Biology Group, Department of Experimental and Health Sciences, Universitat Pompeu Fabra, Doctor Aiguader, 88, Edificio PRBB-3er piso, 08003 Barcelona, Spain. Tel.: +34 933 160 831; fax: +34 933 160 901. E-mail address: [email protected] (A. Meyerhans). These authors contributed equally to this work. # Present address: GENEART AG, D-93053 Regensburg, Germany.
Introduction The Hepatitis C virus (HCV) is an enveloped, positive strand RNA ((+) RNA) virus of the family of Flaviviridae. Infection occurs primarily through exposure to HCV-contaminated blood and remains persistent in the majority of cases. It is estimated that around 170 million individuals are chronically infected worldwide. As persistent HCV infection frequently causes chronic hepatitis that can progress to liver cirrhosis and liver cell carcinoma, it is a major threat for human health [1,2]. HCV shows a high degree of genetic diversity. At a global scale, six major genotypes or ‘‘clades’’ with a greater than 30% nucleotide divergence have been described [3]. These clades can be further divided into subtypes that differ between 20% and 25% from each other. At the level of the HCV-infected individual, the virus exists as a population of related but genetically distinct virus variants collectively named a viral quasispecies [4–8]. By comparing plasmaderived HCV RNA consensus sequences over time, the rate of fixation of mutations was estimated to be around 2 103 per site per year [9]. This is in the typical range for RNA viruses which have RNA-dependent RNA polymerases that lack proof-reading activity [10]. Clinically, the genetic diversity of HCV is of prime importance for antiviral treatment. For example, the rate of sustained response after the standard of care treatment regimen of pegylated interferon-a plus ribavirin differs between HCV genotypes, with genotypes 2 and 3 showing a better outcome than genotype 1 [11]. Furthermore, experimental treatments targeting viral enzymes rapidly select viral escape mutants when given as monotherapy, directly illustrating the consequences of the quasispecies nature of HCV in infected individuals [12]. The accumulation of HCV point mutations over time is caused by error-prone replication of the RNA genome by the viral RNA polymerase NS5B. In addition to this so-called genetic drift, HCV recombinants have been detected phylogenetically both as novel circulating recombinant forms and within infected patients [13–21], suggesting that recombination may also play a role in HCV evolution. Recombination is an integral part of the biology of many RNA viruses and indeed a powerful mechanism in virus evolution [22–26]. Within a single replicative step, complex structural changes can be generated that may result in novel viruses with new pathogenic properties and a modified host range [15,22,27,28]. Furthermore, recombination can help to eliminate deleterious segments generated by error prone replication or it may accelerate the appearance of drug resistant variants in individuals undergoing antiviral therapy [22]. Extensive studies
Journal of Hepatology 2011 vol. 55 j 777–783
Research Article with poliovirus and more recently with viral replication models in yeast suggest (i) that recombination between parental, (+) RNA viruses replicating in the same cell may occur highly frequently and (ii) that recombination occurs predominantly by a copy choice mechanism [16,22,29]. For HCV, such fundamental studies on recombination are lacking. Only the existence of inter- and intrasubtype HCV recombinants has been described and their respective cross-over regions defined. As the detection of recombinants was infrequent, the general belief has been that recombination in HCV is a rare event [30]. However, a recent, PCR-based study suggested that this might not always be the case [21]. In order to evaluate the importance of recombination for HCV evolution and to derive first estimates on the magnitude of this process, we made use of the antibiotic-selectable HCV replicon system [31]. It reproduces all the steps of HCV RNA replication in hepatic cell lines and allows the introduction and subsequent selection of replication-active mutants. The general strategy was to construct mutant replicons that needed a recombination event to regain their complete functionality with respect to conferring neomycin resistance to their host cells or being able to replicate. With such a selectable, HCV replicon-based recombination detection system we show here that recombination in HCV occurs, albeit at a low frequency. However, as the number of HCV-infected cells within seropositive individuals is very large, it is expected that recombination will be phenotypically relevant under conditions of strong selection pressures like antiviral treatment. Moreover, as the defined recombinant frequencies linearly increased with the selectable cross over length, recombination breakpoints are expected to occur along the entire virus genome without major recombination hotspots.
Cell culture and HCV in vitro recombination assay Huh7-Lunet cells, a subclone of the hepatoma cell line Huh7, were described previously [35] and kindly provided by Ralf Bartenschlager. The replicon RNA was electroporated using described methods [33]. Briefly, 4 106 Huh7-Lunet cells were electroporated in Cytomix with 10 lg carrier RNA plus 4 lg of (i) wt, (ii) mutant HCV-replicon RNA or (iii) mutant HCV-replicon RNA in combination with 4 lg of other mutant HCV-replicon RNA. Recombination was detected by a colony formation assay after co-transfection of different combinations of defective replicons (Fig. 2 and Supplementary Table 1). Electroporated cells were transferred to the medium, seeded into cell culture dishes, and cultured at 37 °C with 5% CO2. Twenty-four hours later, 500 lg/ml neomycin (G418) (Sigma–Aldrich, St. Louis, MO, USA) was added to select for replicon-containing, neomycin-resistant cell colonies. The medium was changed once per week. After 3 weeks in culture, cells were stained with 1% crystal violet and cell colonies were counted. Sequence analysis of HCV recombinants Neomycin-resistant colonies from the recombination assays were isolated using cloning cylinders (Sigma–Aldrich, St. Louis, MO, USA) and expanded in neomycin-containing media. RNA was extracted using TRIZOL (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions and reverse transcribed with SuperScript III RT (Invitrogen, Carlsbad, CA, USA). Subsequently, the region between the nucleotides 127 and 7801 was amplified using specific primers that generated five overlapping regions (Supplementary Table 3). The resulting PCR products were sequenced directly according to the BIGDye 3.1 protocol (Applied Biosystems, Foster City, CA, USA) using specific primers (Supplementary Table 4). Calculation of HCV recombinant frequency The HCV recombination selection system established here links a recombination event with cell survival. Thus, the recombinant frequency equals the ratio of the number of neomycin-resistant cell colonies with recombined HCV replicons divided by the respective number with wildtype replicons. The experiments that resulted in zero neomycin-resistant cell colonies were not used for the calculation of average values because the number of recombinants was below the detection limit. The definition of recombinant frequency implies that a single recombination event is sufficient to allow cell colony formation under selection. Consequently our frequency values are minimal estimates.
Materials and methods Generation of plasmids
Results
Starting from the HCV genotype 1b replicons pFKi389neoNS3-30 -ET and pFKi389neoNS3-30 -DGDD that both carry the neomycin-phosphotransferase gene (npt-gene) [31,32], three additional replicons were generated by mutagenesis (see Supplementary Table 1). Replicons 50 m and 30 m have mutations that disrupt functional npt gene expression while D5B is derived from the NS5B-defective replicon NAD5B (=pFKi389neoNS3-30 -DGDD) and contains three cell-culture adaptative mutations in NS3 and NS4B namely E1202G, T1280I and K1846T (see Supplementary Table 1; [33,34]). Plasmids were constructed via a PCR-mediated mutagenesis approach. First, two separate fragments were amplified from the template DNA with primers ‘‘a’’ plus ‘‘b’’ and ‘‘c’’ plus ‘‘d’’. Primers ‘‘b’’ and ‘‘c’’ contained the desired mutations (see Supplementary Table 2 for primer sequences). Subsequently, the two fragments were combined and the complete region amplified with the non-mutated outer primers. The final PCR products were then ligated with pCRÒII-TOPOÒ (Invitrogen, Carlsbad, CA, USA), cloned and verified by sequencing. The DNA-fragments containing the desired mutations were subcloned into pFKi389neoNS330 ET using NruI and RsrII restriction sites and again verified by sequencing. pFKi389neoNS3-30 DGDDadapt was generated by cloning a BglII-fragment from pFKi389neoNS3-30 DGDD into BglII restricted pFKi389neoNS3-30 ET. The correct orientation of the insert was verified by sequencing.
In order to observe and quantify recombination events during HCV replication, a selectable, replicon-based recombination detection system in hepatic cell lines was established. A fundamental feature of this system is that it links replicon recombination with survival of replicon-containing host cells. Starting from neomycin-selectable, subgenomic HCV replicons, several mutations were introduced by directed mutagenesis such that individual mutated replicons were either neomycin-sensitive or replication defective. The mutations were located either at the 50 or the 30 part of the neomycin phosphotransferase gene npt or at the active site of the viral polymerase NS5B (Fig. 1, Supplementary Table 1). Thus, transfection of Huh7-Lunet cells with individual replicons should not give rise to G418-selectable cell clones. However, these should appear after recombination of co-transfected replicons with modifications at different locations. Since the number of selectable recombination events should increase with the possible cross-over length, the further apart the mutated sites, the more cell clones should appear. Fig. 1 outlines the various replicon combinations used in this study and highlights the cross-over length. By including replication-defective HCV replicons with or without three adaptive mutations that strongly influence the replication capacity (see Supplementary Fig. 1 and [34]), the effective cross-over range could be varied further. For example, the combination of replicons 30 m with NAD5B
In vitro transcription In vitro transcripts of HCV replicons were generated by using the RNAMaxx High Yield Transcription Kit (Stratagene, Amsterdam, NL) or the MEGAScript Kit (Ambion, Austin, TX, USA) with T7 polymerase according to the manufacturers instructions. All transcripts were tested for integrity prior transfection using a formaldehyde (1% v/v) containing 1% (w/v) agarose gel.
778
Journal of Hepatology 2011 vol. 55 j 777–783
JOURNAL OF HEPATOLOGY Cross-over length 5’m replicon +
0.6 kb
3’m replicon
3’m replicon 2 kb
+ NAΔ5B replicon
5’m replicon 2.6 kb
+ NAΔ5B replicon
3’m replicon +
6 kb
Δ5B replicon
5’m replicon +
6.6 kb
Δ5B replicon
wt replicon
Fig. 1. Genetic structure of mutated HCV replicons and their combination for recombinant selection. The features of all subgenomic HCV replicons are given. They are related to the wt replicon originally described as pFKi389neoNS3-30 -ET [31]. Replicons 50 m and 30 m are replication competent but have an inactive neomycinphosphotransferase (npt) gene due to mutations of the start codons of the core and the npt gene or to mutations in the active site of the npt enzyme, respectively. The two D5B replicons are replication defective due to a deletion in the active site of the HCV NS5B polymerase. The mutations are indicated by an asterik. The replicon NAD5B lacks the three cell culture-adapted mutations E1202G, T1280I, and K1846T in NS3 and NS4B as indicated by dots. This significantly lowers the replicative fitness of potential recombinants that lack these adaptive mutations and thus results in the shorter, effective cross-over length in the recombination selection experiments as indicated. The cross-over length for all replicon combinations is given on the right.
Journal of Hepatology 2011 vol. 55 j 777–783
779
Recombinant frequency (x10-4)
Research Article 30 m plus D5B, and 50 m plus D5B resulted in cross-over ranges of 0.6, 2, 2.6, 6, and 6.6 kb, respectively (Fig. 1). As controls, cells were electroporated with the wt replicon, the individual mutated replicons or with carrier RNA only. 24 h after electroporation, cells were set under selective pressure by the addition of G418 and the resulting resistant HCV replicon containing cell colonies were quantified by a colony-formation assay. While the different co-transfections resulted in the appearance of neomycin-resistant Huh7-Lunet colonies, the single transfections of mutated replicon RNA (50 m, 30 m and D5B) did not. The details for all experiments are given in Table 1. The lack of colonies from single replicon transfections demonstrated that these could not be repaired during RNA generation or any complex intracellular rearrangements. Furthermore, as the HCV NS5B polymerase only functions in cis [36], trans-complementation of the two D5B replicons by 50 m or 30 m seemed unlikely and thus the appearance of neomycin-resistant cell colonies strongly suggested replicon recombination between the mutated sites. To exclude the possibility that accidental integration of replicon sequences into the Huh7-Lunet genome had provided neomycin-resistance, we isolated 6 colonies obtained from a 50 m plus 30 m co-transfection experiment and cultivated them in the presence of 100 IU/ml IFN-a. This procedure had been previously shown to cure cells from HCV replicons but left integrated DNA from replicons unchanged [37]. After 2 weeks of culture, 500 lg/ml G418 was added. All expanded cell colonies were sensitive to G418 treatment, indicating that no replicon sequence was integrated into the genome. To verify that the G418-selected cell colonies were indeed the result of an HCV replicon recombination event and not due to any complex sequence rearrangements generating a functional npt gene, four colonies from 30 m plus D5B and 50 m plus D5B co-transfection experiments were isolated and expanded in
4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0
1
2 3 4 5 Cross-over range (kb)
6
7
Fig. 2. Linear relationship between the HCV recombinant frequency and the distance between the inactivating replicon mutations. The recombinant frequency is given as a function of the cross-over length in kilobases (kb). The mean values ± standard error of the mean are the results of four to eight independent experiments (see Table 1). Function of linear regression: y = 0.42x 0.04; r2 = 0.83.
has an effective cross-over range of only around 2 kb since a cross-over between the active site of NS5B and the non-adaptive mutations would generate a replicon with a highly diminished replication capacity and thus a significantly reduced number of selectable cell clones (see Supplementary Fig. 1). Accordingly, the given replicon mutants allowed to test for recombination with cross-over ranges between 0.6 and 6.6 kb. Co-transfection of Huh7-Lunet cells with different HCV replicons readily gave rise to G418-selectable cell colonies while individually transfected replicons did not. Huh7-Lunet cells were electroporated with different combinations of the mutated replicons to vary the effective cross-over range for recombination. Combination of 50 m plus 30 m, 30 m plus NAD5B, 50 m plus NAD5B,
Table 1. Total number of G418-resistant cell colonies after co-transfection of mutated HCV replicon RNA and calculated recombinant frequencies.
Experiment
Total number of colonies/recombinant frequency 5’m + 3’m
3’m + NAΔ5B
5’m + NAΔ5B
3’m + Δ5B
5’m + Δ5B
wt*
1
0
1/0.5x10-4
2/1.1x10-4
-
-
1.9x104
2
4/0.6x10-4
10/1.4x10-4
14/2x10-4
-
-
7x104
3
1/0.2x10-4
0
5/1x10-4
-
-
4.9x104
4
-
-
-
23/2.2x10-4
-
-4
5 6 7 8 9 10
3/0.3x10 1/0.1x10
-4
1/0.1x10
-4
1/0.1x10 6/0.6x10
0 0.3x10
average Cross-over length (kb)
0.6 4.3x10
-4
-
0
Recombinant frequency
Normalized recombinant
-4
-8
-
-
4/0.5x10
-4
8/0.8x10
-4
7/1x10
0.7x10-4
1.1x10
-4
2
2.6
3.3x10
-8
4x10
780
⁄⁄
15/1.5x10
10x104
-
16/1.6x10
-4
-
3.3x10
5.5x10
normalized to a cross-over length of one nucleotide.
Journal of Hepatology 2011 vol. 55 j 777–783
8x104 9.8x104
33/3.3x10
14/2x10-4
frequency** , Not tested; ⁄wild type HCV replicon transfected alone;
4.4x104
-4
53/6.6x10
-4
6 -8
12/2.7x10
-4
-4
-4
-4
-4
17/3.9x10
2.1x10
-4
10x104 7x104
-4
6.6 -8
10.4x104
10/1x10-4
3.2x10-8
JOURNAL OF HEPATOLOGY G418-containing media. The RNA of these colonies was subsequently extracted and reverse transcribed. The cDNA covering the complete open reading frame (ORF) of the HCV replicons from position 127–7801 was then amplified in five overlapping parts with specific primers (Supplementary Table 3) and the amplification products directly sequenced. The sequencing primers are given in Supplementary Table 4. The entire HCV replicon ORF including the npt gene and the region of the NS5B polymerase were intact indicating that indeed recombination had occurred between the selectable markers. In addition, several point mutations were detectable demonstrating ongoing error prone HCV replication in cell culture (Supplementary Table 5). These mutations were located in the NS3 and NS5B ORF and showed a transition over transversion bias as previously described [38]. The observed HCV recombinant frequency, which is the ratio of the number of neomycin-resistant cell colonies with recombined HCV replicons divided by the respective number with wt replicons, was dependent on the cross-over length (Fig. 2). With greater cross-over length, the recombinant frequency increased accordingly suggesting that there was no major recombination hotspot between the mutated sites. Averaging over the whole region, the mean recombinant frequency was calculated to be 4 108 per nucleotide or 4 104 per HCV genotype 1b genome.
Discussion This manuscript describes the first quantitative estimates of HCV recombination frequencies in cell culture. For this, an experimental system was needed that allowed detection of rare recombination events. It had to (i) enable selection for such events, (ii) have variable and large cross-over regions to increase the probability of recombination, and (iii) be robust against trans-complementation. An HCV replicon with an antibiotic-selectable marker at the 50 region and a defective NS5B polymerase gene was the system of choice. HCV replicons were previously shown to reproduce the entire replication cycle of HCV and thus should enable a cross-over between parental viral strands. Furthermore as the distance between the selectable marker was several kb long, the likelihood of recombination events was increased. Finally, current data support that a functionally defective NS5B polymerase cannot be complemented in trans and thus, complementation effects between two replicons with defects in the antibiotic-resistant gene and NS5B, respectively, are not expected. The classical system to quantify recombination in (+) RNA viruses was established by Kirkegaard and Baltimore for poliovirus [39]. They infected permissive cells with infectious mutant polioviruses of similar fitness, harvested the supernatants after 4–6 h and screened for parental viruses and recombinants by plaque assays under different, highly efficient selective conditions. A comparable infection system does not exist for HCV. To overcome this drawback, we have developed a replicon-based HCV recombination detection system that uses co-transfection of viral genomes to enter susceptible host cells and then links recombination events with cell survival. After co-transfection, the system is left for 24 h without applying selective conditions. Only then is neomycin added to score recombinants. 24 h was selected as a time point because the input RNA is then largely degraded unless expanded by replication [34].
The use of this selectable, HCV replicon-based recombination detection system enabled us to estimate HCV recombinant frequencies in hepatocyte cell lines. These frequencies are orders of magnitude lower when compared with the notoriously recombinogenic human immunodeficiency virus (HIV) that exhibits 3–30 RNA strand cross-overs during a single reverse transcription step [40,41]. Even when compared with other (+) strand RNA viruses like poliovirus or Brome mosaic virus (BMV) replicating in a yeast model system with recombination frequencies per nucleotide of around 3–7 106 [29,42,43] and around 5 106, respectively, the HCV recombination frequency is still 100-fold lower. Nevertheless, given the estimated number of around 2 1010 HCV-infected hepatocytes in a chronic carrier [44], HCV recombination in vivo may become important when strong selection pressures are operative. Indeed, the outgrowth of an HCV 2 k/1b recombinant under standard peglygated interferon plus ribavirin therapy in a multi-HCV-infected individual has recently been observed [18]. For some of the co-transfections, pairs of replication competent and replication incompetent replicons harboring defective NS5B were used to determine the HCV recombinant frequencies. Available in vitro data demonstrated trans-dominant negative effects of a recombinant mutated NS5B only when in excess over the wild type [45]. However, such effects of mutated NS5B polymerases have not been demonstrated in cell culture. As in our system equal amounts of replicons expressing wild type and mutated NS5B polymerase were used, we did not take this possibility into consideration for the calculation of the recombinant frequencies. In the near future when novel protease and polymerase inhibitors reach clinics, selection of recombinants may become more common. Given the assumptions that the recombination between the HCV replicons described here mimics recombination events within infected hepatocytes in vivo and that these hepatocytes contain more than one viral genome, 2 1010 times 4 104 = 8 106 HCV recombinants might be expected to be generated within one or few days (=the available time for the recombination event in our experiments) within a patient. Taking into account the distance between protease and polymerase resistant mutants as previously determined in clinical trials and cell culture [46–48], this number comes down to about 3 105–3 106 recombinants for a distance of 360 nucleotides (Telaprevir-resistant mutants at amino acids 36 and 156 of NS3) and around 4000 nucleotides (the distance between Telaprevir- and PS1-6130resistance mutants), respectively. Clearly, a significant fitness advantage is necessary to enable such recombinants to compete with the estimated 1012 total number of virus particles produced per day [49]. The reason for the large differences in the frequency of homologous recombination in HCV and other members of the group of (+) strand RNA viruses is still unknown and may occur at various levels. In order to recombine, a minimal requirement is the coexistence of parental RNA stands within a single cell. Considering then a replication-dependent, template switch mechanism as shown for poliovirus or BMV [29,42], both parental strands have to be efficiently recruited to the same replication site and the viral polymerases have to switch between both strands during replication. For poliovirus, the dual recruitment of viral genomes to the same replication sites after co-infection of HeLa cells with poliovirus type 1 and 2 strains has been well documented [50]; however, data for HCV are lacking. With respect to the structure
Journal of Hepatology 2011 vol. 55 j 777–783
781
Research Article of the poliovirus and HCV polymerases, a number of common structural features have been noted, yet they differ significantly in their thumb domains with the consequence of a more restrictive access to the polymerase active site in HCV [51]. Whether this causes a reduction in template switching is an attractive hypothesis and needs to be addressed experimentally. In summary, a neomycin-selectable, HCV genotype 1b replicon-based recombination detection system is described that enabled the first quantitative estimate for the recombination of HCV in a hepatic cell line. The recombinant frequency was low and dependent on the available cross-over range. This system will be of great help to better define the mechanism of HCV recombination, to study the involvement of cellular host factors in recombination and to elucidate the implications of HCV recombination in patients undergoing treatment.
Conflict of interest The authors who have taken part in this study declared that they do not have anything to disclose regarding funding or conflict of interest with respect to this manuscript.
Financial support This work was supported by the Deutsche Forschungsgemeinschaft (Me 1061/4-1), a HOMFOR grant from the University of the Saarland and the Spanish Ministerio de Educación y Ciencia Grants BFU 2007-66933/BMC and SAF2010-21336. Acknowledgments We thank Ralf Bartenschlager (Institute of Virology, University of Heidelberg, Germany) for kindly providing subgenomic HCV replicons pFKi389neoNS3-30 -ET and pFKi389neoNS3-30 -DGDD, and the Huh7-Lunet cells. Axel Scheidig (Department of Structural Biology, University Kiel, Germany) kindly helped in defining the npt-inactivating mutations of the 30 m replicon.
Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jhep.2010.12.038. References [1] Lavanchy D. The global burden of hepatitis C. Liver Int 2009;29:74–81. [2] Shepard CW, Finelli L, Alter MJ. Global epidemiology of hepatitis C virus infection. Lancet Infect Dis 2005;5:558–567. [3] Simmonds P. Genetic diversity and evolution of hepatitis C virus–15 years on. J Gen Virol 2004;85:3173–3188. [4] Argentini C, Genovese D, Dettori S, Rapicetta M. HCV genetic variability: from quasispecies evolution to genotype classification. Future Microbiol 2009;4:359–373. [5] Domingo E, Escarmis C, Sevilla N, Moya A, Elena SF, Quer J, et al. Basic concepts in RNA virus evolution. Faseb J 1996;10:859–864. [6] Domingo E, Gomez J. Quasispecies and its impact on viral hepatitis. Virus Res 2007;127:131–150. [7] Farci P, Purcell RH. Clinical significance of hepatitis C virus genotypes and quasispecies. Semin Liver Dis 2000;20:103–126. [8] Gomez J, Martell M, Quer J, Cabot B, Esteban JI. Hepatitis C viral quasispecies. J Viral Hepat 1999;6:3–16.
782
[9] Ogata N, Alter HJ, Miller RH, Purcell RH. Nucleotide sequence and mutation rate of the H strain of hepatitis C virus. Proc Natl Acad Sci USA 1991;88:3392–3396. [10] Duffy S, Shackelton LA, Holmes EC. Rates of evolutionary change in viruses: patterns and determinants. Nat Rev Genet 2008;9:267–276. [11] Kronenberger B, Zeuzem S. Current and future treatment options for HCV. Ann Hepatol 2009;8:103–112. [12] Sarrazin C, Zeuzem S. Resistance to direct antiviral agents in patients with hepatitis C virus infection. Gastroenterology 2010;138:447–462. [13] Colina R, Casane D, Vasquez S, Garcia-Aguirre L, Chunga A, Romero H, et al. Evidence of intratypic recombination in natural populations of hepatitis C virus. J Gen Virol 2004;85:31–37. [14] Cristina J, Colina R. Evidence of structural genomic region recombination in hepatitis C virus. Virol J 2006;3:53. [15] Kageyama S, Agdamag DM, Alesna ET, Leano PS, Heredia AM, AbellanosaTac-An IP, et al. A natural inter-genotypic (2b/1b) recombinant of hepatitis C virus in the Philippines. J Med Virol 2006;78:1423–1428. [16] Kalinina O, Norder H, Mukomolov S, Magnius LO. A natural intergenotypic recombinant of hepatitis C virus identified in St. Petersburg J Virol 2002;76:4034–4043. [17] Legrand-Abravanel F, Claudinon J, Nicot F, Dubois M, Chapuy-Regaud S, Sandres-Saune K, et al. New natural intergenotypic (2/5) recombinant of hepatitis C virus. J Virol 2007;81:4357–4362. [18] Morel V, Descamps V, Francois C, Fournier C, Brochot E, Capron D, et al. Emergence of a genomic variant of the recombinant 2k/1b strain during a mixed Hepatitis C infection: a case report. J Clin Virol 2010;47:382–386. [19] Moreno MP, Casane D, Lopez L, Cristina J. Evidence of recombination in quasispecies populations of a Hepatitis C Virus patient undergoing anti-viral therapy. Virol J 2006;3:87. [20] Noppornpanth S, Lien TX, Poovorawan Y, Smits SL, Osterhaus AD, Haagmans BL. Identification of a naturally occurring recombinant genotype 2/6 hepatitis C virus. J Virol 2006;80:7569–7577. [21] Sentandreu V, Jimenez-Hernandez N, Torres-Puente M, Bracho MA, Valero A, Gosalbes MJ, et al. Evidence of recombination in intrapatient populations of hepatitis C virus. PLoS One 2008;3:e3239. [22] Lai MM. RNA recombination in animal and plant viruses. Microbiol Rev 1992;56:61–79. [23] Nagy PD, Simon AE. New insights into the mechanisms of RNA recombination. Virology 1997;235:1–9. [24] Simmonds P. Recombination and selection in the evolution of picornaviruses and other mammalian positive-stranded RNA viruses. J Virol 2006;80:11124–11140. [25] Twiddy SS, Holmes EC. The extent of homologous recombination in members of the genus Flavivirus. J Gen Virol 2003;84:429–440. [26] Worobey M, Holmes EC. Evolutionary aspects of recombination in RNA viruses. J Gen Virol 1999;80 (Pt 10):2535–2543. [27] Bailes E, Gao F, Bibollet-Ruche F, Courgnaud V, Peeters M, Marx PA, et al. Hybrid origin of SIV in chimpanzees. Science 2003;300:1713. [28] Gallei A, Orlich M, Thiel HJ, Becher P. Noncytopathogenic pestivirus strains generated by nonhomologous RNA recombination: alterations in the NS4A/ NS4B coding region. J Virol 2005;79:14261–14270. [29] Garcia-Ruiz H, Ahlquist P. Inducible yeast system for viral RNA recombination reveals requirement for an RNA replication signal on both parental RNAs. J Virol 2006;80:8316–8328. [30] Simmonds P, Smith DB, McOmish F, Yap PL, Kolberg J, Urdea MS, et al. Identification of genotypes of hepatitis C virus by sequence comparisons in the core, E1 and NS-5 regions. J Gen Virol 1994;75:1053–1061. [31] Lohmann V, Korner F, Koch J, Herian U, Theilmann L, Bartenschlager R. Replication of subgenomic hepatitis C virus RNAs in a hepatoma cell line. Science 1999;285:110–113. [32] Bartenschlager R, Sparacio S. Hepatitis C virus molecular clones and their replication capacity in vivo and in cell culture. Virus Res 2007;127:195–207. [33] Lohmann V, Korner F, Dobierzewska A, Bartenschlager R. Mutations in hepatitis C virus RNAs conferring cell culture adaptation. J Virol 2001;75:1437–1449. [34] Lohmann V, Hoffmann S, Herian U, Penin F, Bartenschlager R. Viral and cellular determinants of hepatitis C virus RNA replication in cell culture. J Virol 2003;77:3007–3019. [35] Quinkert D, Bartenschlager R, Lohmann V. Quantitative analysis of the hepatitis C virus replication complex. J Virol 2005;79:13594–13605. [36] Tong X, Malcolm BA. Trans-complementation of HCV replication by nonstructural protein 5A. Virus Res 2006;115:122–130. [37] Frese M, Pietschmann T, Moradpour D, Haller O, Bartenschlager R. Interferon-alpha inhibits hepatitis C virus subgenomic RNA replication by an MxA-independent pathway. J Gen Virol 2001;82:723–733.
Journal of Hepatology 2011 vol. 55 j 777–783
JOURNAL OF HEPATOLOGY [38] Kato N, Nakamura T, Dansako H, Namba K, Abe K, Nozaki A, et al. Genetic variation and dynamics of hepatitis C virus replicons in long-term cell culture. J Gen Virol 2005;86:645–656. [39] Kirkegaard K, Baltimore D. The mechanism of RNA recombination in poliovirus. Cell 1986;47:433–443. [40] Jetzt AE, Yu H, Klarmann GJ, Ron Y, Preston BD, Dougherty JP. High rate of recombination throughout the human immunodeficiency virus type 1 genome. J Virol 2000;74:1234–1240. [41] Levy DN, Aldrovandi GM, Kutsch O, Shaw GM. Dynamics of HIV-1 recombination in its natural target cells. Proc Natl Acad Sci USA 2004;101:4204–4209. [42] Jarvis TC, Kirkegaard K. Poliovirus RNA recombination: mechanistic studies in the absence of selection. Embo J 1992;11:3135–3145. [43] Wimmer E, Hellen CU, Cao X. Genetics of poliovirus. Annu Rev Genet 1993;27:353–436. [44] Bartenschlager R, Lohmann V. Replication of hepatitis C virus. J Gen Virol 2000;81:1631–1648. [45] Wang QM, Hockman MA, Staschke K, Johnson RB, Case KA, Lu J, et al. Oligomerization and cooperative RNA synthesis activity of hepatitis C virus RNA-dependent RNA polymerase. J Virol 2002;76:3865–3872.
[46] Ali S, Leveque V, Le Pogam S, Ma H, Philipp F, Inocencio N, et al. Selected replicon variants with low-level in vitro resistance to the hepatitis C virus NS5B polymerase inhibitor PSI-6130 lack cross-resistance with R1479. Antimicrob Agents Chemother 2008;52:4356–4369. [47] Sarrazin C, Kieffer TL, Bartels D, Hanzelka B, Muh U, Welker M, et al. Dynamic hepatitis C virus genotypic and phenotypic changes in patients treated with the protease inhibitor telaprevir. Gastroenterology 2007;132: 1767–1777. [48] Zhou Y, Bartels DJ, Hanzelka BL, Muh U, Wei Y, Chu HM, et al. Phenotypic characterization of resistant Val36 variants of hepatitis C virus NS3-4A serine protease. Antimicrob Agents Chemother 2008;52:110–120. [49] Neumann AU, Lam NP, Dahari H, Gretch DR, Wiley TE, Layden TJ, et al. Hepatitis C viral dynamics in vivo and the antiviral efficacy of interferonalpha therapy. Science 1998;282:103–107. [50] Egger D, Bienz K. Recombination of poliovirus RNA proceeds in mixed replication complexes originating from distinct replication start sites. J Virol 2002;76:10960–10971. [51] Ferrer-Orta C, Arias A, Escarmis C, Verdaguer N. A comparison of viral RNAdependent RNA polymerases. Curr Opin Struct Biol 2006;16:27–34.
Journal of Hepatology 2011 vol. 55 j 777–783
783
Hepatitis C virus RNA recombination in cell culture Jochen Reiter1, ,#, Gemma Pérez-Vilaró2, , Nicoletta Scheller1,2, , Leonardo Bruno Mina2, Juana Díez2, Andreas Meyerhans1,3,⇑ 1
Department of Virology, Saarland University, D-66421 Hamburg, Germany; 2Molecular Virology group; 3ICREA Infection Biology Group, Department of Experimental and Health Sciences, Universitat Pompeu Fabra, 08003 Barcelona, Spain
Background & Aims: The Hepatitis C virus (HCV) exhibits large genetic diversity, both on a global scale and at the level of the infected individual. A major underlying mechanism of the observed sequence differences is error-prone virus replication by the viral RNA polymerase NS5B. In addition, based on phylogenetic comparisons of patient-derived HCV sequences, there is evidence of HCV recombination. However, to date little is known about the frequency by which recombination events occur in HCV and under what conditions recombination may become important in HCV evolution. We, therefore, aimed to set up an experimental model system that would allow us to analyze and to characterize recombination events during HCV replication. Methods: A neomycin-selectable, HCV replicon-based recombination detection system was established. HCV replicons were mutated within either the neomycin-phosphotransferase gene or the NS5B polymerase. Upon co-transfection of hepatic cells lines, recombination between the mutated sites is necessary to restore the selectable phenotype. Results: Recombinants were readily detected with frequencies correlating to the distance between the mutations. The recombinant frequency normalized to a crossover range of one nucleotide was around 4 108. Conclusions: An experimental system to select for HCV recombinants in cell culture was successfully established. It allowed deriving first estimates of recombinant frequencies. Based on these, recombination in HCV seems rare. However, due to the rapid virus turnover and the large number of HCV-infected liver cells in vivo, it is expected that recombination will be of biological importance when strong selection pressures are operative. Ó 2011 European Association for the Study of the Liver. Published by Elsevier B.V. All rights reserved.
Keywords: Recombination; RNA; HCV; HCV replicon; Virus evolution; Positivestrand RNA virus. Received 30 June 2010; received in revised form 14 December 2010; accepted 20 December 2010; available online 18 February 2011 ⇑ Corresponding author. Address: ICREA Infection Biology Group, Department of Experimental and Health Sciences, Universitat Pompeu Fabra, Doctor Aiguader, 88, Edificio PRBB-3er piso, 08003 Barcelona, Spain. Tel.: +34 933 160 831; fax: +34 933 160 901. E-mail address: [email protected] (A. Meyerhans). These authors contributed equally to this work. # Present address: GENEART AG, D-93053 Regensburg, Germany.
Introduction The Hepatitis C virus (HCV) is an enveloped, positive strand RNA ((+) RNA) virus of the family of Flaviviridae. Infection occurs primarily through exposure to HCV-contaminated blood and remains persistent in the majority of cases. It is estimated that around 170 million individuals are chronically infected worldwide. As persistent HCV infection frequently causes chronic hepatitis that can progress to liver cirrhosis and liver cell carcinoma, it is a major threat for human health [1,2]. HCV shows a high degree of genetic diversity. At a global scale, six major genotypes or ‘‘clades’’ with a greater than 30% nucleotide divergence have been described [3]. These clades can be further divided into subtypes that differ between 20% and 25% from each other. At the level of the HCV-infected individual, the virus exists as a population of related but genetically distinct virus variants collectively named a viral quasispecies [4–8]. By comparing plasmaderived HCV RNA consensus sequences over time, the rate of fixation of mutations was estimated to be around 2 103 per site per year [9]. This is in the typical range for RNA viruses which have RNA-dependent RNA polymerases that lack proof-reading activity [10]. Clinically, the genetic diversity of HCV is of prime importance for antiviral treatment. For example, the rate of sustained response after the standard of care treatment regimen of pegylated interferon-a plus ribavirin differs between HCV genotypes, with genotypes 2 and 3 showing a better outcome than genotype 1 [11]. Furthermore, experimental treatments targeting viral enzymes rapidly select viral escape mutants when given as monotherapy, directly illustrating the consequences of the quasispecies nature of HCV in infected individuals [12]. The accumulation of HCV point mutations over time is caused by error-prone replication of the RNA genome by the viral RNA polymerase NS5B. In addition to this so-called genetic drift, HCV recombinants have been detected phylogenetically both as novel circulating recombinant forms and within infected patients [13–21], suggesting that recombination may also play a role in HCV evolution. Recombination is an integral part of the biology of many RNA viruses and indeed a powerful mechanism in virus evolution [22–26]. Within a single replicative step, complex structural changes can be generated that may result in novel viruses with new pathogenic properties and a modified host range [15,22,27,28]. Furthermore, recombination can help to eliminate deleterious segments generated by error prone replication or it may accelerate the appearance of drug resistant variants in individuals undergoing antiviral therapy [22]. Extensive studies
Journal of Hepatology 2011 vol. 55 j 777–783
Research Article with poliovirus and more recently with viral replication models in yeast suggest (i) that recombination between parental, (+) RNA viruses replicating in the same cell may occur highly frequently and (ii) that recombination occurs predominantly by a copy choice mechanism [16,22,29]. For HCV, such fundamental studies on recombination are lacking. Only the existence of inter- and intrasubtype HCV recombinants has been described and their respective cross-over regions defined. As the detection of recombinants was infrequent, the general belief has been that recombination in HCV is a rare event [30]. However, a recent, PCR-based study suggested that this might not always be the case [21]. In order to evaluate the importance of recombination for HCV evolution and to derive first estimates on the magnitude of this process, we made use of the antibiotic-selectable HCV replicon system [31]. It reproduces all the steps of HCV RNA replication in hepatic cell lines and allows the introduction and subsequent selection of replication-active mutants. The general strategy was to construct mutant replicons that needed a recombination event to regain their complete functionality with respect to conferring neomycin resistance to their host cells or being able to replicate. With such a selectable, HCV replicon-based recombination detection system we show here that recombination in HCV occurs, albeit at a low frequency. However, as the number of HCV-infected cells within seropositive individuals is very large, it is expected that recombination will be phenotypically relevant under conditions of strong selection pressures like antiviral treatment. Moreover, as the defined recombinant frequencies linearly increased with the selectable cross over length, recombination breakpoints are expected to occur along the entire virus genome without major recombination hotspots.
Cell culture and HCV in vitro recombination assay Huh7-Lunet cells, a subclone of the hepatoma cell line Huh7, were described previously [35] and kindly provided by Ralf Bartenschlager. The replicon RNA was electroporated using described methods [33]. Briefly, 4 106 Huh7-Lunet cells were electroporated in Cytomix with 10 lg carrier RNA plus 4 lg of (i) wt, (ii) mutant HCV-replicon RNA or (iii) mutant HCV-replicon RNA in combination with 4 lg of other mutant HCV-replicon RNA. Recombination was detected by a colony formation assay after co-transfection of different combinations of defective replicons (Fig. 2 and Supplementary Table 1). Electroporated cells were transferred to the medium, seeded into cell culture dishes, and cultured at 37 °C with 5% CO2. Twenty-four hours later, 500 lg/ml neomycin (G418) (Sigma–Aldrich, St. Louis, MO, USA) was added to select for replicon-containing, neomycin-resistant cell colonies. The medium was changed once per week. After 3 weeks in culture, cells were stained with 1% crystal violet and cell colonies were counted. Sequence analysis of HCV recombinants Neomycin-resistant colonies from the recombination assays were isolated using cloning cylinders (Sigma–Aldrich, St. Louis, MO, USA) and expanded in neomycin-containing media. RNA was extracted using TRIZOL (Invitrogen, Carlsbad, CA, USA) according to the manufacturer’s instructions and reverse transcribed with SuperScript III RT (Invitrogen, Carlsbad, CA, USA). Subsequently, the region between the nucleotides 127 and 7801 was amplified using specific primers that generated five overlapping regions (Supplementary Table 3). The resulting PCR products were sequenced directly according to the BIGDye 3.1 protocol (Applied Biosystems, Foster City, CA, USA) using specific primers (Supplementary Table 4). Calculation of HCV recombinant frequency The HCV recombination selection system established here links a recombination event with cell survival. Thus, the recombinant frequency equals the ratio of the number of neomycin-resistant cell colonies with recombined HCV replicons divided by the respective number with wildtype replicons. The experiments that resulted in zero neomycin-resistant cell colonies were not used for the calculation of average values because the number of recombinants was below the detection limit. The definition of recombinant frequency implies that a single recombination event is sufficient to allow cell colony formation under selection. Consequently our frequency values are minimal estimates.
Materials and methods Generation of plasmids
Results
Starting from the HCV genotype 1b replicons pFKi389neoNS3-30 -ET and pFKi389neoNS3-30 -DGDD that both carry the neomycin-phosphotransferase gene (npt-gene) [31,32], three additional replicons were generated by mutagenesis (see Supplementary Table 1). Replicons 50 m and 30 m have mutations that disrupt functional npt gene expression while D5B is derived from the NS5B-defective replicon NAD5B (=pFKi389neoNS3-30 -DGDD) and contains three cell-culture adaptative mutations in NS3 and NS4B namely E1202G, T1280I and K1846T (see Supplementary Table 1; [33,34]). Plasmids were constructed via a PCR-mediated mutagenesis approach. First, two separate fragments were amplified from the template DNA with primers ‘‘a’’ plus ‘‘b’’ and ‘‘c’’ plus ‘‘d’’. Primers ‘‘b’’ and ‘‘c’’ contained the desired mutations (see Supplementary Table 2 for primer sequences). Subsequently, the two fragments were combined and the complete region amplified with the non-mutated outer primers. The final PCR products were then ligated with pCRÒII-TOPOÒ (Invitrogen, Carlsbad, CA, USA), cloned and verified by sequencing. The DNA-fragments containing the desired mutations were subcloned into pFKi389neoNS330 ET using NruI and RsrII restriction sites and again verified by sequencing. pFKi389neoNS3-30 DGDDadapt was generated by cloning a BglII-fragment from pFKi389neoNS3-30 DGDD into BglII restricted pFKi389neoNS3-30 ET. The correct orientation of the insert was verified by sequencing.
In order to observe and quantify recombination events during HCV replication, a selectable, replicon-based recombination detection system in hepatic cell lines was established. A fundamental feature of this system is that it links replicon recombination with survival of replicon-containing host cells. Starting from neomycin-selectable, subgenomic HCV replicons, several mutations were introduced by directed mutagenesis such that individual mutated replicons were either neomycin-sensitive or replication defective. The mutations were located either at the 50 or the 30 part of the neomycin phosphotransferase gene npt or at the active site of the viral polymerase NS5B (Fig. 1, Supplementary Table 1). Thus, transfection of Huh7-Lunet cells with individual replicons should not give rise to G418-selectable cell clones. However, these should appear after recombination of co-transfected replicons with modifications at different locations. Since the number of selectable recombination events should increase with the possible cross-over length, the further apart the mutated sites, the more cell clones should appear. Fig. 1 outlines the various replicon combinations used in this study and highlights the cross-over length. By including replication-defective HCV replicons with or without three adaptive mutations that strongly influence the replication capacity (see Supplementary Fig. 1 and [34]), the effective cross-over range could be varied further. For example, the combination of replicons 30 m with NAD5B
In vitro transcription In vitro transcripts of HCV replicons were generated by using the RNAMaxx High Yield Transcription Kit (Stratagene, Amsterdam, NL) or the MEGAScript Kit (Ambion, Austin, TX, USA) with T7 polymerase according to the manufacturers instructions. All transcripts were tested for integrity prior transfection using a formaldehyde (1% v/v) containing 1% (w/v) agarose gel.
778
Journal of Hepatology 2011 vol. 55 j 777–783
JOURNAL OF HEPATOLOGY Cross-over length 5’m replicon +
0.6 kb
3’m replicon
3’m replicon 2 kb
+ NAΔ5B replicon
5’m replicon 2.6 kb
+ NAΔ5B replicon
3’m replicon +
6 kb
Δ5B replicon
5’m replicon +
6.6 kb
Δ5B replicon
wt replicon
Fig. 1. Genetic structure of mutated HCV replicons and their combination for recombinant selection. The features of all subgenomic HCV replicons are given. They are related to the wt replicon originally described as pFKi389neoNS3-30 -ET [31]. Replicons 50 m and 30 m are replication competent but have an inactive neomycinphosphotransferase (npt) gene due to mutations of the start codons of the core and the npt gene or to mutations in the active site of the npt enzyme, respectively. The two D5B replicons are replication defective due to a deletion in the active site of the HCV NS5B polymerase. The mutations are indicated by an asterik. The replicon NAD5B lacks the three cell culture-adapted mutations E1202G, T1280I, and K1846T in NS3 and NS4B as indicated by dots. This significantly lowers the replicative fitness of potential recombinants that lack these adaptive mutations and thus results in the shorter, effective cross-over length in the recombination selection experiments as indicated. The cross-over length for all replicon combinations is given on the right.
Journal of Hepatology 2011 vol. 55 j 777–783
779
Recombinant frequency (x10-4)
Research Article 30 m plus D5B, and 50 m plus D5B resulted in cross-over ranges of 0.6, 2, 2.6, 6, and 6.6 kb, respectively (Fig. 1). As controls, cells were electroporated with the wt replicon, the individual mutated replicons or with carrier RNA only. 24 h after electroporation, cells were set under selective pressure by the addition of G418 and the resulting resistant HCV replicon containing cell colonies were quantified by a colony-formation assay. While the different co-transfections resulted in the appearance of neomycin-resistant Huh7-Lunet colonies, the single transfections of mutated replicon RNA (50 m, 30 m and D5B) did not. The details for all experiments are given in Table 1. The lack of colonies from single replicon transfections demonstrated that these could not be repaired during RNA generation or any complex intracellular rearrangements. Furthermore, as the HCV NS5B polymerase only functions in cis [36], trans-complementation of the two D5B replicons by 50 m or 30 m seemed unlikely and thus the appearance of neomycin-resistant cell colonies strongly suggested replicon recombination between the mutated sites. To exclude the possibility that accidental integration of replicon sequences into the Huh7-Lunet genome had provided neomycin-resistance, we isolated 6 colonies obtained from a 50 m plus 30 m co-transfection experiment and cultivated them in the presence of 100 IU/ml IFN-a. This procedure had been previously shown to cure cells from HCV replicons but left integrated DNA from replicons unchanged [37]. After 2 weeks of culture, 500 lg/ml G418 was added. All expanded cell colonies were sensitive to G418 treatment, indicating that no replicon sequence was integrated into the genome. To verify that the G418-selected cell colonies were indeed the result of an HCV replicon recombination event and not due to any complex sequence rearrangements generating a functional npt gene, four colonies from 30 m plus D5B and 50 m plus D5B co-transfection experiments were isolated and expanded in
4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 0.0 0
1
2 3 4 5 Cross-over range (kb)
6
7
Fig. 2. Linear relationship between the HCV recombinant frequency and the distance between the inactivating replicon mutations. The recombinant frequency is given as a function of the cross-over length in kilobases (kb). The mean values ± standard error of the mean are the results of four to eight independent experiments (see Table 1). Function of linear regression: y = 0.42x 0.04; r2 = 0.83.
has an effective cross-over range of only around 2 kb since a cross-over between the active site of NS5B and the non-adaptive mutations would generate a replicon with a highly diminished replication capacity and thus a significantly reduced number of selectable cell clones (see Supplementary Fig. 1). Accordingly, the given replicon mutants allowed to test for recombination with cross-over ranges between 0.6 and 6.6 kb. Co-transfection of Huh7-Lunet cells with different HCV replicons readily gave rise to G418-selectable cell colonies while individually transfected replicons did not. Huh7-Lunet cells were electroporated with different combinations of the mutated replicons to vary the effective cross-over range for recombination. Combination of 50 m plus 30 m, 30 m plus NAD5B, 50 m plus NAD5B,
Table 1. Total number of G418-resistant cell colonies after co-transfection of mutated HCV replicon RNA and calculated recombinant frequencies.
Experiment
Total number of colonies/recombinant frequency 5’m + 3’m
3’m + NAΔ5B
5’m + NAΔ5B
3’m + Δ5B
5’m + Δ5B
wt*
1
0
1/0.5x10-4
2/1.1x10-4
-
-
1.9x104
2
4/0.6x10-4
10/1.4x10-4
14/2x10-4
-
-
7x104
3
1/0.2x10-4
0
5/1x10-4
-
-
4.9x104
4
-
-
-
23/2.2x10-4
-
-4
5 6 7 8 9 10
3/0.3x10 1/0.1x10
-4
1/0.1x10
-4
1/0.1x10 6/0.6x10
0 0.3x10
average Cross-over length (kb)
0.6 4.3x10
-4
-
0
Recombinant frequency
Normalized recombinant
-4
-8
-
-
4/0.5x10
-4
8/0.8x10
-4
7/1x10
0.7x10-4
1.1x10
-4
2
2.6
3.3x10
-8
4x10
780
⁄⁄
15/1.5x10
10x104
-
16/1.6x10
-4
-
3.3x10
5.5x10
normalized to a cross-over length of one nucleotide.
Journal of Hepatology 2011 vol. 55 j 777–783
8x104 9.8x104
33/3.3x10
14/2x10-4
frequency** , Not tested; ⁄wild type HCV replicon transfected alone;
4.4x104
-4
53/6.6x10
-4
6 -8
12/2.7x10
-4
-4
-4
-4
-4
17/3.9x10
2.1x10
-4
10x104 7x104
-4
6.6 -8
10.4x104
10/1x10-4
3.2x10-8
JOURNAL OF HEPATOLOGY G418-containing media. The RNA of these colonies was subsequently extracted and reverse transcribed. The cDNA covering the complete open reading frame (ORF) of the HCV replicons from position 127–7801 was then amplified in five overlapping parts with specific primers (Supplementary Table 3) and the amplification products directly sequenced. The sequencing primers are given in Supplementary Table 4. The entire HCV replicon ORF including the npt gene and the region of the NS5B polymerase were intact indicating that indeed recombination had occurred between the selectable markers. In addition, several point mutations were detectable demonstrating ongoing error prone HCV replication in cell culture (Supplementary Table 5). These mutations were located in the NS3 and NS5B ORF and showed a transition over transversion bias as previously described [38]. The observed HCV recombinant frequency, which is the ratio of the number of neomycin-resistant cell colonies with recombined HCV replicons divided by the respective number with wt replicons, was dependent on the cross-over length (Fig. 2). With greater cross-over length, the recombinant frequency increased accordingly suggesting that there was no major recombination hotspot between the mutated sites. Averaging over the whole region, the mean recombinant frequency was calculated to be 4 108 per nucleotide or 4 104 per HCV genotype 1b genome.
Discussion This manuscript describes the first quantitative estimates of HCV recombination frequencies in cell culture. For this, an experimental system was needed that allowed detection of rare recombination events. It had to (i) enable selection for such events, (ii) have variable and large cross-over regions to increase the probability of recombination, and (iii) be robust against trans-complementation. An HCV replicon with an antibiotic-selectable marker at the 50 region and a defective NS5B polymerase gene was the system of choice. HCV replicons were previously shown to reproduce the entire replication cycle of HCV and thus should enable a cross-over between parental viral strands. Furthermore as the distance between the selectable marker was several kb long, the likelihood of recombination events was increased. Finally, current data support that a functionally defective NS5B polymerase cannot be complemented in trans and thus, complementation effects between two replicons with defects in the antibiotic-resistant gene and NS5B, respectively, are not expected. The classical system to quantify recombination in (+) RNA viruses was established by Kirkegaard and Baltimore for poliovirus [39]. They infected permissive cells with infectious mutant polioviruses of similar fitness, harvested the supernatants after 4–6 h and screened for parental viruses and recombinants by plaque assays under different, highly efficient selective conditions. A comparable infection system does not exist for HCV. To overcome this drawback, we have developed a replicon-based HCV recombination detection system that uses co-transfection of viral genomes to enter susceptible host cells and then links recombination events with cell survival. After co-transfection, the system is left for 24 h without applying selective conditions. Only then is neomycin added to score recombinants. 24 h was selected as a time point because the input RNA is then largely degraded unless expanded by replication [34].
The use of this selectable, HCV replicon-based recombination detection system enabled us to estimate HCV recombinant frequencies in hepatocyte cell lines. These frequencies are orders of magnitude lower when compared with the notoriously recombinogenic human immunodeficiency virus (HIV) that exhibits 3–30 RNA strand cross-overs during a single reverse transcription step [40,41]. Even when compared with other (+) strand RNA viruses like poliovirus or Brome mosaic virus (BMV) replicating in a yeast model system with recombination frequencies per nucleotide of around 3–7 106 [29,42,43] and around 5 106, respectively, the HCV recombination frequency is still 100-fold lower. Nevertheless, given the estimated number of around 2 1010 HCV-infected hepatocytes in a chronic carrier [44], HCV recombination in vivo may become important when strong selection pressures are operative. Indeed, the outgrowth of an HCV 2 k/1b recombinant under standard peglygated interferon plus ribavirin therapy in a multi-HCV-infected individual has recently been observed [18]. For some of the co-transfections, pairs of replication competent and replication incompetent replicons harboring defective NS5B were used to determine the HCV recombinant frequencies. Available in vitro data demonstrated trans-dominant negative effects of a recombinant mutated NS5B only when in excess over the wild type [45]. However, such effects of mutated NS5B polymerases have not been demonstrated in cell culture. As in our system equal amounts of replicons expressing wild type and mutated NS5B polymerase were used, we did not take this possibility into consideration for the calculation of the recombinant frequencies. In the near future when novel protease and polymerase inhibitors reach clinics, selection of recombinants may become more common. Given the assumptions that the recombination between the HCV replicons described here mimics recombination events within infected hepatocytes in vivo and that these hepatocytes contain more than one viral genome, 2 1010 times 4 104 = 8 106 HCV recombinants might be expected to be generated within one or few days (=the available time for the recombination event in our experiments) within a patient. Taking into account the distance between protease and polymerase resistant mutants as previously determined in clinical trials and cell culture [46–48], this number comes down to about 3 105–3 106 recombinants for a distance of 360 nucleotides (Telaprevir-resistant mutants at amino acids 36 and 156 of NS3) and around 4000 nucleotides (the distance between Telaprevir- and PS1-6130resistance mutants), respectively. Clearly, a significant fitness advantage is necessary to enable such recombinants to compete with the estimated 1012 total number of virus particles produced per day [49]. The reason for the large differences in the frequency of homologous recombination in HCV and other members of the group of (+) strand RNA viruses is still unknown and may occur at various levels. In order to recombine, a minimal requirement is the coexistence of parental RNA stands within a single cell. Considering then a replication-dependent, template switch mechanism as shown for poliovirus or BMV [29,42], both parental strands have to be efficiently recruited to the same replication site and the viral polymerases have to switch between both strands during replication. For poliovirus, the dual recruitment of viral genomes to the same replication sites after co-infection of HeLa cells with poliovirus type 1 and 2 strains has been well documented [50]; however, data for HCV are lacking. With respect to the structure
Journal of Hepatology 2011 vol. 55 j 777–783
781
Research Article of the poliovirus and HCV polymerases, a number of common structural features have been noted, yet they differ significantly in their thumb domains with the consequence of a more restrictive access to the polymerase active site in HCV [51]. Whether this causes a reduction in template switching is an attractive hypothesis and needs to be addressed experimentally. In summary, a neomycin-selectable, HCV genotype 1b replicon-based recombination detection system is described that enabled the first quantitative estimate for the recombination of HCV in a hepatic cell line. The recombinant frequency was low and dependent on the available cross-over range. This system will be of great help to better define the mechanism of HCV recombination, to study the involvement of cellular host factors in recombination and to elucidate the implications of HCV recombination in patients undergoing treatment.
Conflict of interest The authors who have taken part in this study declared that they do not have anything to disclose regarding funding or conflict of interest with respect to this manuscript.
Financial support This work was supported by the Deutsche Forschungsgemeinschaft (Me 1061/4-1), a HOMFOR grant from the University of the Saarland and the Spanish Ministerio de Educación y Ciencia Grants BFU 2007-66933/BMC and SAF2010-21336. Acknowledgments We thank Ralf Bartenschlager (Institute of Virology, University of Heidelberg, Germany) for kindly providing subgenomic HCV replicons pFKi389neoNS3-30 -ET and pFKi389neoNS3-30 -DGDD, and the Huh7-Lunet cells. Axel Scheidig (Department of Structural Biology, University Kiel, Germany) kindly helped in defining the npt-inactivating mutations of the 30 m replicon.
Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.jhep.2010.12.038. References [1] Lavanchy D. The global burden of hepatitis C. Liver Int 2009;29:74–81. [2] Shepard CW, Finelli L, Alter MJ. Global epidemiology of hepatitis C virus infection. Lancet Infect Dis 2005;5:558–567. [3] Simmonds P. Genetic diversity and evolution of hepatitis C virus–15 years on. J Gen Virol 2004;85:3173–3188. [4] Argentini C, Genovese D, Dettori S, Rapicetta M. HCV genetic variability: from quasispecies evolution to genotype classification. Future Microbiol 2009;4:359–373. [5] Domingo E, Escarmis C, Sevilla N, Moya A, Elena SF, Quer J, et al. Basic concepts in RNA virus evolution. Faseb J 1996;10:859–864. [6] Domingo E, Gomez J. Quasispecies and its impact on viral hepatitis. Virus Res 2007;127:131–150. [7] Farci P, Purcell RH. Clinical significance of hepatitis C virus genotypes and quasispecies. Semin Liver Dis 2000;20:103–126. [8] Gomez J, Martell M, Quer J, Cabot B, Esteban JI. Hepatitis C viral quasispecies. J Viral Hepat 1999;6:3–16.
782
[9] Ogata N, Alter HJ, Miller RH, Purcell RH. Nucleotide sequence and mutation rate of the H strain of hepatitis C virus. Proc Natl Acad Sci USA 1991;88:3392–3396. [10] Duffy S, Shackelton LA, Holmes EC. Rates of evolutionary change in viruses: patterns and determinants. Nat Rev Genet 2008;9:267–276. [11] Kronenberger B, Zeuzem S. Current and future treatment options for HCV. Ann Hepatol 2009;8:103–112. [12] Sarrazin C, Zeuzem S. Resistance to direct antiviral agents in patients with hepatitis C virus infection. Gastroenterology 2010;138:447–462. [13] Colina R, Casane D, Vasquez S, Garcia-Aguirre L, Chunga A, Romero H, et al. Evidence of intratypic recombination in natural populations of hepatitis C virus. J Gen Virol 2004;85:31–37. [14] Cristina J, Colina R. Evidence of structural genomic region recombination in hepatitis C virus. Virol J 2006;3:53. [15] Kageyama S, Agdamag DM, Alesna ET, Leano PS, Heredia AM, AbellanosaTac-An IP, et al. A natural inter-genotypic (2b/1b) recombinant of hepatitis C virus in the Philippines. J Med Virol 2006;78:1423–1428. [16] Kalinina O, Norder H, Mukomolov S, Magnius LO. A natural intergenotypic recombinant of hepatitis C virus identified in St. Petersburg J Virol 2002;76:4034–4043. [17] Legrand-Abravanel F, Claudinon J, Nicot F, Dubois M, Chapuy-Regaud S, Sandres-Saune K, et al. New natural intergenotypic (2/5) recombinant of hepatitis C virus. J Virol 2007;81:4357–4362. [18] Morel V, Descamps V, Francois C, Fournier C, Brochot E, Capron D, et al. Emergence of a genomic variant of the recombinant 2k/1b strain during a mixed Hepatitis C infection: a case report. J Clin Virol 2010;47:382–386. [19] Moreno MP, Casane D, Lopez L, Cristina J. Evidence of recombination in quasispecies populations of a Hepatitis C Virus patient undergoing anti-viral therapy. Virol J 2006;3:87. [20] Noppornpanth S, Lien TX, Poovorawan Y, Smits SL, Osterhaus AD, Haagmans BL. Identification of a naturally occurring recombinant genotype 2/6 hepatitis C virus. J Virol 2006;80:7569–7577. [21] Sentandreu V, Jimenez-Hernandez N, Torres-Puente M, Bracho MA, Valero A, Gosalbes MJ, et al. Evidence of recombination in intrapatient populations of hepatitis C virus. PLoS One 2008;3:e3239. [22] Lai MM. RNA recombination in animal and plant viruses. Microbiol Rev 1992;56:61–79. [23] Nagy PD, Simon AE. New insights into the mechanisms of RNA recombination. Virology 1997;235:1–9. [24] Simmonds P. Recombination and selection in the evolution of picornaviruses and other mammalian positive-stranded RNA viruses. J Virol 2006;80:11124–11140. [25] Twiddy SS, Holmes EC. The extent of homologous recombination in members of the genus Flavivirus. J Gen Virol 2003;84:429–440. [26] Worobey M, Holmes EC. Evolutionary aspects of recombination in RNA viruses. J Gen Virol 1999;80 (Pt 10):2535–2543. [27] Bailes E, Gao F, Bibollet-Ruche F, Courgnaud V, Peeters M, Marx PA, et al. Hybrid origin of SIV in chimpanzees. Science 2003;300:1713. [28] Gallei A, Orlich M, Thiel HJ, Becher P. Noncytopathogenic pestivirus strains generated by nonhomologous RNA recombination: alterations in the NS4A/ NS4B coding region. J Virol 2005;79:14261–14270. [29] Garcia-Ruiz H, Ahlquist P. Inducible yeast system for viral RNA recombination reveals requirement for an RNA replication signal on both parental RNAs. J Virol 2006;80:8316–8328. [30] Simmonds P, Smith DB, McOmish F, Yap PL, Kolberg J, Urdea MS, et al. Identification of genotypes of hepatitis C virus by sequence comparisons in the core, E1 and NS-5 regions. J Gen Virol 1994;75:1053–1061. [31] Lohmann V, Korner F, Koch J, Herian U, Theilmann L, Bartenschlager R. Replication of subgenomic hepatitis C virus RNAs in a hepatoma cell line. Science 1999;285:110–113. [32] Bartenschlager R, Sparacio S. Hepatitis C virus molecular clones and their replication capacity in vivo and in cell culture. Virus Res 2007;127:195–207. [33] Lohmann V, Korner F, Dobierzewska A, Bartenschlager R. Mutations in hepatitis C virus RNAs conferring cell culture adaptation. J Virol 2001;75:1437–1449. [34] Lohmann V, Hoffmann S, Herian U, Penin F, Bartenschlager R. Viral and cellular determinants of hepatitis C virus RNA replication in cell culture. J Virol 2003;77:3007–3019. [35] Quinkert D, Bartenschlager R, Lohmann V. Quantitative analysis of the hepatitis C virus replication complex. J Virol 2005;79:13594–13605. [36] Tong X, Malcolm BA. Trans-complementation of HCV replication by nonstructural protein 5A. Virus Res 2006;115:122–130. [37] Frese M, Pietschmann T, Moradpour D, Haller O, Bartenschlager R. Interferon-alpha inhibits hepatitis C virus subgenomic RNA replication by an MxA-independent pathway. J Gen Virol 2001;82:723–733.
Journal of Hepatology 2011 vol. 55 j 777–783
JOURNAL OF HEPATOLOGY [38] Kato N, Nakamura T, Dansako H, Namba K, Abe K, Nozaki A, et al. Genetic variation and dynamics of hepatitis C virus replicons in long-term cell culture. J Gen Virol 2005;86:645–656. [39] Kirkegaard K, Baltimore D. The mechanism of RNA recombination in poliovirus. Cell 1986;47:433–443. [40] Jetzt AE, Yu H, Klarmann GJ, Ron Y, Preston BD, Dougherty JP. High rate of recombination throughout the human immunodeficiency virus type 1 genome. J Virol 2000;74:1234–1240. [41] Levy DN, Aldrovandi GM, Kutsch O, Shaw GM. Dynamics of HIV-1 recombination in its natural target cells. Proc Natl Acad Sci USA 2004;101:4204–4209. [42] Jarvis TC, Kirkegaard K. Poliovirus RNA recombination: mechanistic studies in the absence of selection. Embo J 1992;11:3135–3145. [43] Wimmer E, Hellen CU, Cao X. Genetics of poliovirus. Annu Rev Genet 1993;27:353–436. [44] Bartenschlager R, Lohmann V. Replication of hepatitis C virus. J Gen Virol 2000;81:1631–1648. [45] Wang QM, Hockman MA, Staschke K, Johnson RB, Case KA, Lu J, et al. Oligomerization and cooperative RNA synthesis activity of hepatitis C virus RNA-dependent RNA polymerase. J Virol 2002;76:3865–3872.
[46] Ali S, Leveque V, Le Pogam S, Ma H, Philipp F, Inocencio N, et al. Selected replicon variants with low-level in vitro resistance to the hepatitis C virus NS5B polymerase inhibitor PSI-6130 lack cross-resistance with R1479. Antimicrob Agents Chemother 2008;52:4356–4369. [47] Sarrazin C, Kieffer TL, Bartels D, Hanzelka B, Muh U, Welker M, et al. Dynamic hepatitis C virus genotypic and phenotypic changes in patients treated with the protease inhibitor telaprevir. Gastroenterology 2007;132: 1767–1777. [48] Zhou Y, Bartels DJ, Hanzelka BL, Muh U, Wei Y, Chu HM, et al. Phenotypic characterization of resistant Val36 variants of hepatitis C virus NS3-4A serine protease. Antimicrob Agents Chemother 2008;52:110–120. [49] Neumann AU, Lam NP, Dahari H, Gretch DR, Wiley TE, Layden TJ, et al. Hepatitis C viral dynamics in vivo and the antiviral efficacy of interferonalpha therapy. Science 1998;282:103–107. [50] Egger D, Bienz K. Recombination of poliovirus RNA proceeds in mixed replication complexes originating from distinct replication start sites. J Virol 2002;76:10960–10971. [51] Ferrer-Orta C, Arias A, Escarmis C, Verdaguer N. A comparison of viral RNAdependent RNA polymerases. Curr Opin Struct Biol 2006;16:27–34.
Journal of Hepatology 2011 vol. 55 j 777–783
783