Adaptability costs in immune escape variants of vesicular stomatitis virus

Virus Research 107 (2005) 27–34

Adaptability costs in immune escape variants of vesicular stomatitis virus Isabel S. Novellaa,b,∗ , Dorothy L. Gilbertsonb , Belen Borregoc , Esteban Domingoc , John J. Hollandb a

Department of Microbiology and Immunology, Medical College of Ohio, 3055 Arlington Avenue Toledo, OH 43614, USA b Division of Biology and Institute for Molecular Genetics, 9500 Gilman Drive, University of California, San Diego, La Jolla, CA 92093-0116, USA c Centro de Biolog´ıa Molecular “Severo Ochoa” (Consejo Superior de Investigaciones Cient´ıficas, Universidad Aut´onoma de Madrid), Cantoblanco, Madrid 28049, Spain Received 4 April 2004; received in revised form 16 June 2004; accepted 16 June 2004 Available online 4 August 2004

Abstract We have used vesicular stomatitis virus (VSV) to determine the cost of antiserum resistance during escape from a polyclonal immune response. Replication of VSV in the presence of polyclonal antiserum resulted in the selection of antibody-escape mutants, as shown by increased fitness in the presence of antiserum and by increased resistance to neutralization. However, resistance came at a cost of overall fitness loss in the BHK-21 host cells. Sequencing of the surface G glycoprotein showed that two to four mutations were fixed in each population, most of which mapped in the A1 and A2 antigenic sites. Selected resistant populations were passaged as large populations in BHK-21 cells under constant conditions, which would normally lead to fitness increases. Nevertheless, many of the populations showed little or no sign of recovery, although the resistant phenotype was maintained. These results suggest that while antiserum resistance can develop, it may come at a cost in fitness and further limitations in the adaptability of the populations. © 2004 Elsevier B.V. All rights reserved. Keywords: Fitness; Antibodies; Trade-off; Adaptability; RNA virus

1. Introduction Error-prone RNA virus replication results in complex populations of closely related mutants known as quasispecies (Domingo et al., 1985; Holland et al., 1992). In addition, RNA viruses can replicate very fast and to very high titers, allowing rapid adaptation to changes in the environmental conditions (Domingo et al., 2001). This potential has been exploited to answer many basic questions of evolutionary processes using RNA viruses as models (Domingo et al., 2001; Elena and Lenski, 2003; ∗

Corresponding author. Tel.: +1 419 383 6442; fax: +1 419 383 3002. E-mail address: [email protected] (I.S. Novella).

0168-1702/$ – see front matter © 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.virusres.2004.06.007

Novella, 2003). One of the most significant forces that viruses infecting vertebrates have to face during replication is the immune response of the host. Viruses typically respond to this challenge by the accumulation of mutations that abrogate antibody recognition and allow escape. Escape variants become antigenic targets of new specificity and induce more antibodies, which constitute a new selective pressure and promote further mutation accumulation in the virus. Successive waves of host antibody response and virus escape represent an evolutionary race between host and virus. This race conforms to the Red Queen hypothesis (VanValen, 1973) in that “it takes all the running you can do to stay in the same place.” Thus, the inability of the host to mount an effective immune response or inability of the virus to develop resistance may result in

28

I.S. Novella et al. / Virus Research 107 (2005) 27–34

death or extinction. Among human pathogens, human immunodeficiency virus type 1 (HIV-1) and hepatitis C virus (HCV) are two well-known illustrations of this type of interplay between viruses and the immune system of individual hosts (Brander and Walker, 2003; Shimizu et al., 1994). At an inter-host level as in human populations antibody-driven evolution is best exemplified by influenza A virus (Webster, 1999). Understanding antibody escape is of fundamental importance for the rational design of antiviral strategies. How can viruses escape from a polyclonal immune response, presumably targeted against multiple antigenic sites? What is the cost of antibody escape? We addressed these questions using a well-developed and very successful model of RNA virus evolution, vesicular stomatitis virus (VSV). The prototype of the Rhabdovirus family, VSV is an enveloped virus with a negative-sense single-stranded RNA genome of 11,161 nucleotides with five genes. The nucleocapsid is composed of the RNA and three viral proteins, the nucleoprotein (N) and the polymerase (L and P proteins), and it is bound to an external membrane envelope by the matrix (M) protein. The only external protein is the G glycoprotein, which is the main target for antibodies (Kelley et al., 1972). VSV has been one of the best the models to dissect the processes controlling gene expression in mononegavirales (Barr et al., 2002; De et al., 1997; Rose and Whitt, 2001). It has also proved a unique system to test general principles of population genetics, as well as specific topics of viral evolution (Domingo et al., 2001; Novella, 2003). The antigenic structure of VSV Indiana serotype has been well characterized (Lefrancois and Lyles, 1982a, 1982b; Nagata et al., 1992; VandePol et al., 1986; Volk et al., 1982) and comprises four or five antigenic sites. Earlier work demonstrated the ability of VSV populations to escape polyclonal antibodies and described the molecular basis of escape (VandePol et al., 1986). However, the effect of escape selection on viral replicative fitness was not determined. To answer this question we allowed extensive replication of VSV populations in the presence of rabbit polyclonal antiserum and characterized the evolved progeny. Our results showed that resistance developed quickly, but at a fitness cost and at an additional cost in adaptability.

2. Materials and methods 2.1. Cells and viruses BHK-21 cells and wild type VSV Indiana serotype (Mudd–Summers strain) were used in this study. Fitness assays were done using monoclonal antibody resistant mutant (MARM) U as neutral reference virus. Methods for cell culture and virus replication have been previously described in detail (Duarte et al., 1994; Holland et al., 1991).

2.2. Immunization of rabbits and neutralization assays Two New Zealand rabbits (labeled 160 and 161) were immunized with 100 ␮g of VSV in complete Freud’s adjuvant. Animals were boosted three and 6 weeks after the first inoculation and bled after the second boost. Antisera obtained from bleeding were lyophilized, and resuspended in deionized water prior to use. Neutralization titers were determined by mixing one volume containing 2 × 104 infectious particles with one volume of antiserum at several dilutions and incubating the mixture at 37 ◦ C for 30 min. After incubation virus was titrated by plaque assay on BHK-21 monolayers. Antiserum neutralization titers are expressed as the concentration of serum causing 99% inhibition compared to a control incubated with no serum. 2.3. Virus passages and fitness assays A wild type VSV stock was diluted to 4 × 105 PFU/ml and mixed with an equal volume of antisera at a concentration causing 99% reduction of virus titer. The appropriate dilutions were 0.45 × 10−4 for antiserum 160 and 0.55 × 10−4 for antiserum 161, and this treatment resulted in a virus population size of 2 × 103 PFU at each passage. After incubation for 30 min at 37 ◦ C, mixtures were used to start six passage series (labeled 160A–160F or 161A–161F) for each antiserum by infection of six BHK-21 monolayers as described (Holland et al., 1991). A lower concentration of antiserum (0.5 × 10−5 ) was added to the overlay medium to keep the selective pressure during replication. After complete cytopathic effect (24–48 h), the progeny virus was diluted, mixed with antiviral serum, and used for a second passage, and this process was repeated until 30 passages were reached. As virus resistant mutants appeared, antiserum concentration was gradually increased to keep population size constant. Virus populations were labeled with the number of the antiserum used followed by the letter of the series. For example, 160A corresponds to series A of virus passaged 30 times in the presence of antiserum 160. Six control series were passaged in parallel in the absence of anti-VSV serum; in these infections the virus was further diluted to reach a similar population size per passage (i.e. 2000 PFU) than in the passages in the presence of anti-VSV serum. These were labeled A through F. Four of the populations resulting after 30 passages in the presence of each antiserum were further passsaged. Two populations selected in the presence of each antiserum were randomly picked and passaged ten times in the presence of a constant antiserum concentration (the same as was employed for the 30th passage) and in the absence of antiserum. These were populations 160B and 160C, which were further passaged in the presence of a constant concentration of antiserum 160 or in its absence, and populations 161C and 161E, which were passaged in the presence of a constant concentration of antiserum 161 or in its absence. In some cases, after antiserum treatment, the virus still had to be diluted to keep the population size constant. These populations were labled

I.S. Novella et al. / Virus Research 107 (2005) 27–34

29

by adding to the progenitor’s label “10A,” etc. for passages in the absence of antiserum, “+10 (160)A,” (for antiserum 160) “+10 (161)A,” etc. (for antiserum 161) in the case of passages in the presence of a constant antiserum concentration. Fitness assays were done as described elsewhere (Holland et al., 1991). Briefly, mixture of test virus (sensitive to I1) and MARM U (surrogate neutral internal control) were used to infect BHK monolayers in the presence or absence of antiserum. Ratios wt:MARMU were determined by plaque assay and changes in those ratios through time were used to obtain a regression, the slope of which is the fitness value.

Table 1 Resistance of VSV populations after replication in the presence of serum

2.4. Nucleotide sequencing

Populations replicating in the presence of serum 160 160B 2.8 × 10−4 160C 8.0 × 10−4 160D 6.0 × 10−4 160E 2.7 × 10−4 160F 3.2 × 10−4

VSV RNA was extracted from 25 ␮l of infected cultures with Tri-Reagent as recommended by the manufacturer, dissolved in 10 ␮l of DEPC water, and employed to obtain cDNA of the G glycoprotein by RT-PCR. Random hexamers and Superscript II were used for reverse transcription of 5 ␮l of viral RNA. After reverse transcription, 35 cycles of amplification were performed with the following scheme: 35 denaturing at 95 ◦ C, 45 annealing at 48 ◦ C and 1 45 at 72 ◦ C. Two cDNA fragments were amplified, the first one comprising nucleotides 3022–3939, and the second 3787–4740. Amplification of cDNA was done with primers M232F with G861R for the first fragment, and G38F with SR4757A for the second fragment. Internal primers (available upon request) of each cDNA fragment were used to obtain the complete sequence (Rodriguez et al., 2002). Sequencing was done using uncloned cDNA fragments and BigDye mixtures for the reactions. An automatic sequencer ABI Prism 310 Genetic Analyzer was employed. This method detects the dominant (i.e. average or consensus) nucleotide at each position. It does not, however, provide information regarding variability within each population. Sequences after selective regimes in the presence of antiserum were compared to control wild type prior to passage.

3. Results 3.1. Phenotypic changes of VSV populations replicating under immune pressure All the series that underwent 30 passages in BHK-21 cells in the absence of anti-VSV serum showed fitness increases, in agreement with previous results (Clarke et al., 1993; Novella et al., 1996, 1995; Turner and Elena, 2000). Neutralization assays of populations selected in antiserum showed increased resistance to antiserum in every case (Table 1). These results were confirmed by fitness assays in the presence of antiserum, where the resulting populations showed a great increase in replicative fitness when compared with wt parental virus (Table 2), except for strains 160D, 160E and 161F, which were 30–60% resistant to I1 and thus could not be assayed. All resistant VSV populations replicated to titers indistin-

Strain

Serum 160 10−4

wt 0.45 × Populations replicating in the absence of serum A 0.42 × 10−4 B 0.66 × 10−4 C 0.44 × 10−4 D 0.44 × 10−4 E 0.43 × 10−4 F 0.63 × 10−4 Average

Average

0.50 × 10−4

4.5 × 10−4

Populations replicating in the presence of serum 161 161A 0.8 × 10−4 161D 2.0 × 10−4 Average

1.4 × 10−4

Serum 161 0.55 × 10−4 0.50 × 10−4 0.50 × 10−4 0.61 × 10−4 0.57 × 10−4 0.51 × 10−4 0.57 × 10−4 0.53 × 10−4 3.0 × 10−4 2.2 × 10−4 6.8 × 10−4 2.0 × 10−4 2.0 × 10−4 3.2 × 10−4 1.25 × 10−4 2.0 × 10−4 1.6 × 10−4

Values represent the serum dilution that causes 99% inhibition after neutralization as described in Section 2.

guishable from wt. However, fitness determinations showed that many viral populations lost fitness to a significant degree; others displayed no significant changes or slight gains. Only one population (161D) showed fitness gains comparable to controls. The results were similar to those observed after the operation of Muller’s ratchet in that there was variability among replicas and the fate of a specific population cannot be predicted (Clarke et al., 1993), but on average overall fitness loss in BHK-21 cells was observed for populations that had replicated in the presence of anti-VSV serum (Table 2). 3.2. Genotypic changes of VSV populations replicating under immune pressure We determined the sequence of the external G glycoprotein from resistant populations to antiserum 161 (Table 3). In Indiana serotype there are four major antigenic sites (Lefrancois and Lyles, 1982a). The results showed 14 point mutations, and all of them except two were nonsynonymous. Excess of non-synonymous mutations is a typical trait in populations evolving under positive selection, although the number of mutations in the present data set is too small for statistical analysis. Here selection for antibody resistance is obviously the major evolutionary force affecting these viral populations. However, the Asp → Tyr mutation in amino acid 171 (present in populations 161A and 161B) is unlikely to be involved in antibody escape by itself, since it occurs in other laboratory strains of VSV-Indiana that show wt sen-

30

I.S. Novella et al. / Virus Research 107 (2005) 27–34

Table 2 Fitness values of VSV population during competition in the presence or absence of serum Strain

+Serum

Populations replicating in the absence of serum A 2.0 B 2.1 C 1.4 D 3.5 E 3.4 F ND Average

2.5∗∗

Populations replicating in the presence of serum 160 160A 490 160B 10 160C 53 160D (I1 resistant) 160E (I1 resistant) 160F 42 Average

149∗∗∗

Populations replicating in the presence of serum 161 161A 26 161B 37 161C 56 161D 148 161E 26 161F (I1 resistant) Average

59∗∗∗

−Serum 1.8 2.0 2.5 5.8 6.0 5.4 3.9∗∗

1.4 0.68 0.55 – – 0.8 0.86∗ 0.4 1.1 0.5 1.8 0.4 – 0.84∗

Fitness was determined by direct competition against the genetically-marked surrogate wt (see Section 2). Statistically sugnificant differences are indicated with asterisks. ND = not determined. I1 resistace was between 30% and 60%. ∗ P < 0.05. ∗∗ P < 0.01. ∗∗∗ P < 0.001 (paired t-test).

sitivity to neutralization by antibodies. Several mutations map in or close to epitope A1, recognized by Mab I8 (Lefrancois and Lyles, 1982a; VandePol et al., 1986). The Gly → Ser substitution at amino acid 54 is located in the A2 epitope, recognized by the I3 Mab (Lefrancois and Lyles, 1982a; VandePol et al., 1986), and corresponds to one of the three T-helper epitopes described in the G glycoprotein, spanning amino acids 52–71 (Burkhart et al., 1994). Positions 240 and 242 may represent the fifth antigenic site reported by Volk et al. (1982). To our knowledge, these mutations have not been previously reported, but their repeated appearance in three populations is suggestive of their involvement in antibody-escape. A proposed secondary structure of the G glycoprotein (Walker and Kongsuwan, 1999) places these residues in a loop that is maintained by disulfide bridges between Cys 234 and Cys 239. The Trp in position 242 mutated into a third Cys in the region, opening the possibility for alternative disulfide bridges, and therefore changes in structure that would allow antibody escape. Neither position 257 nor 259 were mutated in the antiserum-resistant strains, in agreement with the overall sensitive phenotype of the evolved populations to the I1 Mab used to distinguish evolved strains from wt during competitions. 3.3. Lack of fitness recovery in antibody resistant VSV populations upon continuous replication in a constant environment It should be stressed that the neutralization assays during the passages were done regularly increasing the concentration of antiserum as needed, so inhibition and population size were kept constant. In a normal infection the antibody response does not increase indefinitely. The question of adaptability was addressed by selecting some of the antibody-resistant populations (160B, 160C, 161C, 161E) and carrying out additional passages. We followed four series of each passage regime, which consisted of 10 passages in the presence of constant levels of the corresponding antiserum, or 10 passages in the absence of any serum. The replicative fitness of the resulting populations was tested in the presence and absence of antiserum as in previous cases (Table 4). These additional passages provided some expected results, such as the maintenance or improvement of the antibodyresistant phenotypes in populations that kept replicating in the presence of antiserum. Strikingly, however, a significant subset of the populations did not recover fitness in BHK-21 cells upon further passage under constant levels of antiserum or in the absence of antiserum (Table 4). The progeny of 160B were the only ones to show remarkable overall fitness recovery, with improvement of up to 10–20-fold. The only other group that regained fitness, although to a limited extent, was 161E. Thus, it appears that elimination of the selective pressure imposed by the antiserum was not enough to guarantee fitness recovery.

I.S. Novella et al. / Virus Research 107 (2005) 27–34 Table 4 Fitness of antibody-resistant populations upon 10 additional passages in the absence of serum or in the presence of constant serum concentration Fitness Strain

+Serum

−serum

160B (from Table 2) +10A +10B +10C +10D +10 average +10 (160)A +10 (160)B +10 (160)C +10 (160)D +10 (160) average

10 43 67 48 59 54∗ 283 581 1100 550 629∗∗∗

0.68 7 7.5 3.5 4.5 5.6∗∗ 1.5 1.3 0.95 0.78 1.13∗

160C (from Table 2) +10A +10B +10C +10D +10 average +10 (160)A +10 (160)B +10 (160)C +10 (160)D +10 (160) average

53 43 41 64 67 54 380 305 192 358 309∗∗

0.55 0.8 0.54 0.64 0.6 0.64 0.55 0.62 0.52 0.56 0.56

161E (from Table 2) +10A +10B +10C +10D +10 average +10 (161)A +10 (161)B +10 (161)C +10 (161)D +10 (161) average

26 214 51 27 30 80∗ 43 41 64 67∗ 54

0.4 0.6 0.5 2 0.56 0.92∗ 0.78 0.8 0.66 0.6 0.71

161C (from Table 1) +10A +10B +10C +10D +10 average

56 ND ND ND ND –

0.5 0.45 0.5 0.75 0.64 0.58

Bold font indicates the parental populations, which are the progeny of wt after 30 passages in the presence of antiserum 160 or 161. Statistically significant differences are indicated with asteriscs. ND = not determined. ∗ P < 0.05. ∗∗ P < 0.01. ∗∗∗ P < 0.001 (paired t-test).

4. Discussion Many RNA viruses can develop resistance to immune sera in spite of the presence of multiple antibodies. This can be explained in part by immunodominance, which limits the number of epitopes targeted during infection, decreasing the number of mutations needed for escape. In addition, maturation of the immune response results in amplification of selected lymphocyte clones further limiting the diversity of targeted sites. Thus, escape can be achieved with a limited number of substitutions, as shown in this report

31

and others (Lambkin et al., 1994; Siebelink et al., 1993, 1999; Skraban et al., 1999; Thali et al., 1994; VandePol et al., 1986; Watkins et al., 1996). Clearly, immune responses elicited by peptide or subunit vaccines will favor even more the selection of resistant variants (Carman et al., 1990; Ni et al., 1995; Taboga et al., 1997). Escape is also favored if a single mutation can produce changes in multiple antigenic sites (Watkins et al., 1996). Moreover, the same antiviral molecules will often select for the same mutations leading to parallel (if having a common progenitor) or convergent (if having distinct progenitors) evolution. Parallel evolution of G sequences was observed in our results (Table 3) that cannot be explained by cross-contamination, since other nucleotide substitutions allowed identification of each individual G sequence. Indeed, several of the mutations observed in these resistant variants were reported nearly two decades ago after selection of the strain used in this report, VSV Mudd–Summers, by mouse Mabs, or by selection of the VSV Glasgow ts G31 strain by rabbit antiserum (VandePol et al., 1986). The former example illustrates parallel evolution, and the latter is a case of convergent evolution. Strong selective pressures, such as antibodies or inhibitory drugs, often select for mutants with suboptimal replication ability in the absence of the antiviral as proposed by Coffin (Coffin, 1995). Fitness cost of single and multiple mutations implicated in HIV-1 drug resistance has been amply documented for many drugs (Borman et al., 1996; Collins et al., 2004; Nijhuis et al., 2001). VSV populations selected during replication in interferon-treated cells show lower fitness in the BHK-21 host cells than sensitive wt (Novella et al., 1996). While the generation and characterization of antibody-resistant animal viruses has received a lot of attention, little has been done regarding the potential cost of escape. Work carried out with foot-and-mouth disease virus (FMDV) showed that trade-off also applies during the development of resistance to polyclonal antiserum. Thus, resistant FMDV mutants could only replicate to titers as low as 2 × 102 PFU/ml (compared to 107 –108 in wild type) (Borrego et al., 1993). The studies with FMDV involved passage of several clonal populations in the presence or the absence of polyclonal antisera directed against a synthetic peptide that represented a major antigenic site of the virus (Borrego et al., 1993). Some populations decreased in fitness as a result of passage in the presence of antiviral antibodies. However, in all cases, passage of such low fitness populations in the absence of antiserum often resulted in complete fitness recovery (Borrego et al., 1993). This capacity to regain fitness offers an interesting contrast with the results reported here with VSV (Table 4). It could be considered that capsids of non-enveloped viruses may be less tolerant to amino acid substitutions than glycoproteins of enveloped viruses, because of more strict geometrical constraints. The comparison of the results of fitness recovery of VSV and FMDV antibody-selected mutants suggest that the nature of the amino acid substitutions and their immediate structural consequence may be more relevant to determine fitness costs than the general geometrical

32

I.S. Novella et al. / Virus Research 107 (2005) 27–34

Fig. 1. Model of potential immune-driven evolution for viruses.

context in which the escape substitutions occur. The fitness costs of antigenic variation do not seem to be predictable from the type of viral particle in which it occurs. Similar results have been reported for antibody-escape influenza, in that escape mutants selected in the presence of polyclonal serum or cocktails of Mabs displayed lower titers than sensitive wt virus (Lambkin et al., 1994). Fitness cost during escape is due to conflict between high levels of replication and survival. In the absence of antiserum mutations that allows better replication will be fixed. However, in the presence of antiserum there are two selective pressure, that provided by the host cell and that provided by the antiviral activity of antibodies, and mutations allowing improvement in both will rarely be the same. The lower adaptability displayed by resistant strains when compared to wt cannot be interpreted as a complete inability to gain fitness. Additional passages may lead to fitness gains, but other strains with initial relative fitness similar to that of these resistant strains (about half of that of wt) are able to reach neutrality in only five passages (Novella et al., 1995). Thus, 10 passages were a conservative approach to the question of whether these mutants would be able to regain fitness under constant conditions. Natural VSV evolution is not driven by the immune response (Nichol et al., 1989; Vernon et al., 1990), even though antibody titers are high in livestock of endemic areas. Sequence analysis of several arboviruses has shown little evidence of positive selection (Woelk and Holmes, 2002), although some amino acid substitutions are determinant for the evolution of an enzootic into an epizootic virus (Brault et al., 2002; Powers et al., 1997). Other viruses like RSV are also in a similar situation (Sullender, 2000; Sullender and Edwards, 1999). It is intriguing that in spite of the presence of selection, all these viral species show no response, in evolutionary terms. We propose a model of evolution driven by antiviral responses, according to which most or all RNA viruses can generate antibody escape variants, but the fate of escape mutants will depend on the cost in fitness and in adaptability (Fig. 1). According to this model, animal viruses for which there is no fitness or adaptability costs will show evolution driven

by host immune responses or drug intervention. Viruses for which there is fitness and adaptability costs will show little or no antibody-driven evolution (such as for VSV). If there is fitness cost, but no adaptability cost, evolution will depend on the mechanism of recovery. Recovery by reversion will hamper antibody-driven evolution, while recovery by compensation will favor antibody-driven evolution (such as in FMDV). Results of anti-HIV therapy are consistent with this idea. Thus, multiple drugs do select for multiple mutations, and most have a fitness cost (Borman et al., 1996; Nijhuis et al., 2001). In addition, some level of fitness recovery can take place often by reversion (Gandhi et al., 2003) or sometimes by compensation (Liang et al., 1977; Qui˜nones-Mateu et al., 2002). However, viral load can be controlled and patients do not develop AIDS. The virus appears to reach an evolutionary dead-end. While this is likely to be a simplification, and additional factors need to be considered, testing the validity of this model with other human pathogens would be useful to design more effective antiviral strategies.

Acknowledgements We thank Leo Lefrancois for kindly providing I1 hybridoma cells. This research was supported by the NIH grant AI45686 (ISN), grants DGICYT BMC 2001-1823-C02-01, FISS 98/0054-01 and Fundaci´on Ram´on Areces (ED), and by NIH grant AI14627 (JJH).

References Barr, J.N., Whelan, S.P.J., Wertz, G.W., 2002. Transcriptional control of the RNA-dependent RNA polymerase of vesicular stomatitis virus. Biochim. Biophys. Acta 1577, 323–353. Borman, A.M., Paulos, S., Clavel, F., 1996. Resistance of human immunodeficiency virus type 1 to protease inhibitors: selection of resistance mutations in the presence and absence of the drug. J. Gen. Virol. 77, 419–426. Borrego, B., Novella, I.S., Giralt, E., Andreu, D., Domingo, E., 1993. Distinct repertoire of antigenic variants of foot-and-mouth disease virus in the presence or absence of immune selection. J. Virol. 67, 6071–6079. Brander, C., Walker, B.D., 2003. Gradual adaptation of HIV to human host populations: good or bad news? Nat. Med. 9, 1359–1362. Brault, A.C., Powers, A.M., Holmes, E.C., Woelk, C.H., Weaver, S.C., 2002. Positively charged amino acid substitutions in E2 glycoprotein are associated with the emergence of Venezuelan equine encephalitis virus. J. Virol. 76, 1718–1730. Burkhart, C., Freer, G., Castro, R., Adorini, L., Wiesmuller, K.-H., Zinkernagel, R.M., Hengartner, H., 1994. Characterization of T-helper epitopes of the glycoprotein of vesicular stomatitis virus. J. Virol. 68, 1573–1580. Carman, W.F., Zanetti, A.R., Karayiannis, P., Waters, J., Manzillo, G., Tanzi, E., Zuckerman, A.J., Thomas, H.C., 1990. Vaccine-induced escape mutant of hepatitis B virus. Lancet 336, 325–329. Clarke, D.K., Duarte, E.A., Moya, A., Elena, S.F., Domingo, E., Holland, J., 1993. Genetic bottlenecks and population passages cause profound fitness differences in RNA viruses. J. Virol. 67, 222–228.

I.S. Novella et al. / Virus Research 107 (2005) 27–34 Coffin, J.M., 1995. HIV population dynamics in vivo-implications for genetic variation, pathogenesis and therapy. Science 267, 483–489. Collins, J.A., Thompson, M.G., Paintsil, E., Ricketts, M., Gedzior, J., Alexander, L., 2004. Competitive fitness of nevirapine-resistant human immunodeficiency virus type 1 mutants. J. Virol. 78, 603–611. De, B.P., Das, T., Banerjee, A.K., 1997. Role of cellular kinases in the gene expression of nonsegmented negative strand RNA viruses. Biol. Chem. 378, 489–493. Domingo, E., Biebricher, C., Eigen, M., Holland, J., 2001. Quasispecies and RNA Virus Evolution: Principles and Consequenes. Landes Bioscience, Georgetown, TX. Domingo, E., Martinez-Salas, E., Sobrino, F., delaTorre, J.C., Portela, A., Ortin, J., Lopez-Galindez, C., Perez-Brena, P., Villanueva, N., Najera, R., VandePol, S., Steinhauer, D., DePolo, N., Holland, J., 1985. The quasispecies (extremely heterogeneous) nature of viral RNA genome populations: biological relevance-a review. Gene 40, 1–8. Duarte, E.A., Novella, I.S., Ledesma, S., Clarke, D.K., Moya, A., Elena, S.F., Domingo, E., Holland, J.J., 1994. Subclonal components of consensus fitness in an RNA virus clone. J. Virol. 68, 4295–4301. Elena, S.F., Lenski, R.E., 2003. Evolution experiments with microorganims: the dynamics and genetic bases of adaptation. Nat. Rev. Genet. 4, 457–469. Gandhi, R.T., Wurcel, A., Rosenberg, E.S., Johnston, M.N., Hellmann, N., Bates, M., Hirsch, M.S., Walker, B.D., 2003. Progressive reversion of human immunodeficiency virus type 1 resistance: mutations in vivo after transmission of a multiply drug-resistant virus. Clin. Infect. Dis. 27, 1693–1698. Holland, J.J., de la Torre, J.C., Steinhauer, D.A., 1992. RNA virus populations as quasispecies. Curr. Top. Microbiol. Immunol. 176, 1–20. Holland, J.J., de la Torre, J.C., Clarke, D.C., Duarte, E., 1991. Quantitation of relative fitness and great adaptability of clonal populations of RNA viruses. J. Virol. 65, 2960–2967. Kelley, J.R., Emerson, S.U., Wagner, R.R., 1972. The glycoprotein of vesicular stomatitis virus is the antigen that gives rise and reacts with neutralizing antibodies. J. Virol. 10, 1231. Lambkin, R., MClain, L., Jones, S.E., Aldridge, S.L., Dimmock, N.J., 1994. Neutralization escape mutants of type-A influenza virus are readily selected by antisera from mice immunized with whole virus—a possible mechanism for antigenic drift. J. Gen. Virol. 75, 3493–3502. Lefrancois, L., Lyles, D.S., 1982a. The interaction of antibody with the major surface glycoprotein of vesicular stomatitis virus. 1. Analysis of neutralizing epitopes with monoclonal antibodies. Virology 121, 157–167. Lefrancois, L., Lyles, D.S., 1982b. The interaction of antibody with the major surface glycoprotein of vesicular stomatitis virus. 2. Monoclonal antibodies to non-neutralizing and cross-reactive epitopes of Indiana and New Jersey serotype. Virology 121, 168–174. Liang, C., Li, X., Rong, L., Inouye, P., Quan, Y., Kleiman, L., Wainberg, M.A., 1977. The importance of the A-h loop in human immunodeficiency virus type 1 reverse transcription and infectivity. J. Virol. 71, 5750–5757. Nagata, S., Okamoto, Y., Inoue, T., Ueno, Y., Kurata, T., Chiba, J., 1992. Identification of epitopes associated with different biological activities on the glycoprotein of vesicular stomatitis virus by use of monoclonal antibodies. Arch. Virol. 127, 153–168. Ni, F., Fang, D., Gan, R., Li, Z., Duan, S., Xu, Z., 1995. A new immune escape mutant of hepatitis B virus with and Asp to Ala substitution in amino acid 144 of the envelope major protein. Res. Virol. 146, 397–407. Nichol, S.T., Rowe, J.E., Fitch, W.M., 1989. Glycoprotein evolution of vesicular stomatitis virus New Jersey. Virology 168, 281–291. Nijhuis, M., Deeks, S., Boucher, C., 2001. Implications of antiretroviral resistance on viral fitness. Curr. Opin. Infect. Dis. 14, 23–28. Novella, I., 2003. Contributions of vesicular stomatitis virus to the understanding of RNA virus evolution. Curr. Opin. Microbiol. 6, 399–405. Novella, I.S., Cilnis, M., Elena, S.F., Kohn, J., Moya, A., Domingo, E., Holland, J.J., 1996. Large-population passages of vesicular stomatitis

33

virus in interferon-treated cells select variants of only limited resistance. J. Virol. 70, 6414–6417. Novella, I.S., Duarte, E.A., Elena, S.F., Moya, A., Domingo, E., Holland, J.J., 1995. Exponential increases of RNA virus fitness during large population transmissions. Proc. Natl. Acad. Sci. U.S.A. 92, 5841–5844. Powers, A.M., Oberste, M.S., Brault, A.C., Rico-Hesse, R., Schmura, S.M., Smith, J.F., Kang, W.L., Sweeney, W.P., Weaver, S.C., 1997. Repeated emergence of epidemic/epizootic Venezuelan equine encephalitis from a single genotype of enzootic subtype ID virus. J. Virol. 71, 6697–6705. Qui˜nones-Mateu, M.E., Tadele, M., Parera, M., Mas, A., Weber, J., Rangel, H.R., Chakraborty, B., Clotet, B., Domingo, E., MenendezArias, L., Martinez, M.A., 2002. Insertions in the reverse transcriptase increase both drug resistance and viral fitness in a human immunodeficiency virus type 1 isolate harboring the multi-nucleoside reverse transcriptase inhibitor resistance 69 insertion complex mutation. J. Virol. 76, 10546–10552. Rodriguez, L.L., Pauszek, S.J., Bunch, T.A., Schumann, K.R., 2002. Fulllength genome analysis of natural isolates of vesicular stomatitis virus (Indiana 1 serotype) from North, Central and South America. J. Gen. Virol. 83, 2475–2483. Rose, J.H., Whitt, M.A., et al., 2001. Rhabdoviridae and their replication, fourth ed. In: Knipe, D.M., Howley, P.M. (Eds.), Field’s Virology, 1. Lipicott Williams and Wilkins, pp. 1121–1244. Shimizu, Y.K., Hijikata, M., Iwamoto, A., Alter, H.J., Purcell, R.H., Yoshikura, H., 1994. Neutralizing antibodies against hepatitis C virus and the emergence of neutralization escape mutant viruses. J. Virol. 68, 1494–1500. Siebelink, K.H.J., Rimmelzwaan, G.F., Bosch, M.L., Meloen, R.H., Osterhaus, A.D.M.E., 1993. A single amino acid substitution in hypervariable region 5 of the envelope protein of feline immunodeficiency virus allows escape from virus neutralization. J. Virol. 67, 2202–2208. Skraban, R., Matthiasdottir, S., Torseinsdottir, S., Agnasdottir, G., Gudmundson, B., Georgsson, G., Meloen, R.H., Andresson, O.S., Staskus, K.A., Thormar, H., ANdresdottir, V., 1999. Naturally occurring mutations within the 39 amino acids in the envelope glycoprotein of maedivisna virus alter the neutralization epitope. J. Virol. 73, 8064–8072. Sullender, W.M., 2000. Respiratory syncytial virus genetic and antigenic diversity. Clin. Microbiol. Rev. 13, 1–15. Sullender, W.M., Edwards, K.G., 1999. Mutations of respiratory syncytial virus attachment glycoprotein G associated with resistance to neutralization by primate polyclonal antibodies. Virology 264, 230–236. Taboga, O., Tami, C., Carrillo, E., Nunez, J.I., Rodriguez, A., Sainz, J.C., Blanco, E., Valero, M.L., Roig, X., Camarero, J.A., Andreu, D., Mateu, M.G., Giralt, E., Domingo, E., Sobrino, F., Palma, E.L., 1997. A large-scale evaluation of peptide vaccines against foot-andmouth disease: lack of solid protection in cattle and isolation of escape mutants. J. Virol. 71, 2606–2614. Thali, M., Charles, M., Furman, C., Cavacini, L., Posner, M., Robinson, J., Sodroski, J., 1994. Resistantralization by broadly reactive antibodies to the human immunodeficiency virus type 1 gp120 glycoprotein conferred by gp41 amino acid change. J. Virol. 68, 674–680. Turner, P.E., Elena, S.F., 2000. Cost of host radiation in an RNA virus. Genetics 156, 1465–1470. VandePol, S.B., Lefrancois, L., Holland, J.J., 1986. Sequences of the major antibody binding epitopes of the Indiana serotype of vesicular stomatitis virus. Virology 148, 312–325. VanValen, L., 1973. A new evolutionary law. Evol. Theory 1, 1–30. Vernon, S.D., Rodriguez, L.L., Letchworth, G.J., 1990. Vesicular stomatitis virus New jersey glysoprotein sequence and neutralizing epitope stability in an enzootic focus. Virology 177, 209–215. Volk, W.A., Snyder, R.M., Benjamin, D.C., Wagner, R.W., 1982. Monoclonal antibodies to the glycoprotein of vesicular stomatitis virus: comparative neutralizing activity. J. Virol. 42, 220–227. Walker, P.J., Kongsuwan, K., 1999. Deduced structural model for animal rhabdovirus glycoproteins. J. Gen. Virol. 80, 1211–1220.

34

I.S. Novella et al. / Virus Research 107 (2005) 27–34

Watkins, B.A., Buge, S., Aldrich, K., Davis, A.E., Robinson, J., Reitz, M.S., Robert-Guroff, M., 1996. Resistance of human immunodeficiency virus type 1 to neutralization by natural antisera occurs through single amino acid substitutions that cause changes in antibody binding at multiple sites. J. Virol. 70, 8431–8437.

Webster, R.G., 1999. Antigenic variation in influenza viruses. In: Domingo, E., Webster, R.G., Holland, J.J. (Eds.), Origin and Evolution of Viruses. Academic Press, pp. 377–390. Woelk, C.H., Holmes, E.C., 2002. Reduced positive selection in vectorborne RNA viruses. Mol. Biol. Evol. 19, 2333–2336.