Skip to main content
PMC home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Jan 3.
Published in final edited form as: Virology. 2006 Sep 11;357(2):165–174. doi: 10.1016/j.virol.2006.08.005

Cell-specific adaptation of two flaviviruses following serial passage in mosquito cell culture

Alexander T Ciota 1,*, Amy O Lovelace 1, Kiet A Ngo 1, An N Le 1, Joseph G Maffei 1, Mary A Franke 1, Anne F Payne 1, Susan A Jones 1, Elizabeth B Kauffman 1, Laura D Kramer 1
PMCID: PMC3249649  NIHMSID: NIHMS344505  PMID: 16963095

Abstract

West Nile Virus (WNV) is a mosquito-borne flavivirus that was introduced into the U.S. in the New York City area in 1999. Despite its successful establishment and rapid spread in a naive environment, WNV has undergone limited evolution since its introduction. This evolutionary stability has been attributed to compromises made to permit alternating cycles of viral replication in vertebrate hosts and arthropod vectors. Outbreaks of a close relative of WNV, St. Louis encephalitis virus (SLEV), occur in the U.S. periodically and are also characterized by limited genetic change overtime. We measured both phenotypic and genotypic changes in WNVand SLEV serially passaged in mosquito cell culture in order to clarify the role of an individual host cell type in flavivirus adaptation and evolution. Genetic changes in passaged WNVand SLEV were minimal but led to increased relative fitness and replicative ability of the virus in the homologous cell line C6/36 mosquito cells. Similar increases were not measured in the heterologous cell line DF-1 avian cells. These phenotypic changes are consistent with the concept of cell-specific adaptation in flaviviruses.

Keywords: West Nile virus, St. Louis encephalitis virus, Flavivirus, Viral fitness, Virus evolution

Introduction

West Nile Virus (WNV) is a mosquito-borne flavivirus that was introduced into the U.S. in the New York City area in 1999. Since its introduction, WNV has spread rapidly across the U.S. and into Canada, Mexico, and Central and South America (Lanciotti et al., 1999; Dupuis et al., 2003; 2005; Austin et al., 2004). Outbreaks of its close relative, St. Louis encephalitis virus (SLEV), occur in the U.S. periodically (Chandler et al., 2001; Day and Stark, 2000). Understanding the adaptability and selective pressures that drive genetic and phenotypic changes in these viruses is crucial to predicting their ability to persist and reemerge.

Both WNV and SLEV are maintained in nature in enzootic cycles in which they are transmitted between ornithophilic mosquitoes and avian hosts. Non-avian vertebrates become infected as a result of feeding by infected vectors, but virus is generally not perpetuated in this manner, due to the low level of viremia in most non-avian, vertebrate hosts (Kramer and Bernard, 2001). Evolutionary pressures on the virus, therefore, are applied predominately by the mosquito and avian environments.

RNA viruses have the capacity for rapid evolution due to their high mutation rates, short replication times, and large population sizes (Drake and Holland, 1999). Despite this, rapid rates of evolution have not been observed with arboviruses (Cilnis et al., 1996; Jenkins et al., 2002; Weaver et al., 1992). It has been shown, for example, that WNV has undergone limited evolution in the years during which it has been circulating in the U.S. (Davis et al., 2005; Ebel et al., 2001, 2004). The case of WNV is particularly interesting because the virus’s genotypic stability has not compromised its ability to succeed in new environments. One hypothesis for this genetic conservation is that the alternate cycles of viral replication in vertebrate hosts and arthropod vectors constrain evolution (Scott et al., 1994; Weaver et al., 1992). This implies that compromises in replicative ability are made regularly by virus populations, in both the arthropod vectors and vertebrate hosts, due to differential selection in each. Specifically, mutations exclusively advantageous to either host are purged by purifying selection if they are detrimental to replication in the alternative host; positive selection then generally results from the infrequent mutations that result in coadaptation. Reduced positive selection and increased purifying selection in vector-borne RNA viruses have been reported previously (Holmes, 2003; Jerzak et al., 2005; Woelk and Holmes, 2002). Further phenotypic and genotypic evidence for replicative compromises has been provided by previous studies including those on the flavivirus dengue 2 (Chen et al., 2003) and the togaviruses Sindbis (SINV) (Greene et al., 2005) and eastern equine encephalitis (EEEV) (Cooper and Scott, 2001;Weaver et al., 1999). However, different results were obtained when studies were performed with the rhabdovirus vesicular stomatitis virus (VSV) (Novella et al., 999; Zarate and Novella, 2004).

We assessed individual relative fitness, replicative ability, and genetic alterations of both WNV and SLEV during serial passage in C6/36 cell culture in order to better characterize the evolutionary pressures and adaptive ability of flaviviruses replicating in mosquito cells. Specifically, we sought to clarify the extent to which these flaviviruses are capable of cell-specific adaptation and the degree to which this adaptation alters both fitness in other cell types and genetic sequences. We hypothesized that adaptation to a single cell type would lead to decreases in viral fitness in the bypassed cell. Understanding the spectrum of phenotypic changes and degree of genetic alteration in virus serially passaged in a single host cell type will help us to understand the evolutionary stability observed in virus cycling in nature.

Results

Changes in relative fitness of WNV and SLEV following serial passage in mosquito cells

Competition assays to measure the relative fitness of WNV and SLEV were based on co-infection of three different cell lines with a mixture of test (biologically cloned virus) and control (monoclonal antibody [MAb] resistant mutant; MARM) virus. Growth curves were analyzed to confirm that replication of WNV and SLEV and their respective MARM is similar in each cell line (Fig. 1). Time of virus harvest for competition assays in each cell line was selected based on the time at which the peak titers of the test and control viruses were similar. Consequently, WNV competition assays were carried out for 48, 72, and 96 h post inoculation (hpi) in Vero (mammalian), DF-1 (avian), and C6/36 (mosquito) cells, respectively, and SLEV competition assays were harvested at 96 hpi in C6/36 cells and 48 hpi in DF-1 and Vero cells (Fig. 1). Although titers of SLEV and its MARM began to diverge at 48 hpi in DF-1 cells, the assays were harvested at a time at which neither strain had a significant replicative advantage.

Fig. 1.

Fig. 1

Open in a new tab

Replication in different three cell lines of WNV (A) and SLEV (B) unpassaged biological clones (UP) and monoclonal antibody resistant mutants (MARM). Multiplicity of infection for all growth curves is 0.1 PFU/cell. Arrows indicate the time of harvest for competition assays.

The results of competition assays measuring the relative fitness values of WNV and SLEV in three different cell lines are summarized in Fig. 2. Changes in unpassaged virus: MARM ratios were minimal during competition. Ratios fluctuated slightly in all cell lines for both viruses but were never significantly different from input ratios in any round of competition (data not shown). Values were adjusted such that a relative fitness value of 0 was assigned to these controls. Therefore, relative fitness values accurately represent changes resulting from passaging such that a positive value represents a gain in relative fitness and a negative value represents a loss in relative fitness.

Fig. 2.

Fig. 2

Open in a new tab

Relative fitness values of WNV and SLEV after passage in C6/36 cells and competition in three cell lines. Three to four independent C6/36 cell passage lineages, designated A, B, C, and D, were tested in each cell line and vector plots were constructed. Relative fitness values represent slopes calculated after normalizing the ratio for each round to the ratio of the initial input and plotting these values against the competition round. Each competition assay was performed in triplicate, and the average fitness values are presented here. In order to measure changes resulting from passaging, all values were adjusted further by normalizing to relative fitness of unpassaged virus controls, and slopes were compared by ANCOVA (GraphPad Prism, Version 4.0) to determine if significant fitness changes had occurred (*P<0.05).

Competition assays performed in the mosquito cell line C6/36 resulted in relative fitness gains for all lineages of WNV and SLEV after serial passage in C6/36 cells. These gains were significant in two of three lineages after 10 passages for WNV (WNV CP10) (P<0.05) and were highly significant in all three lineages by passage 19 (WNV CP19) (P<0.001) (Fig. 2A). Fitness continued to increase to passage 39 (WNV CP39), with a relative fitness advantage of approximately 1 (equivalent to 1 log10 increase per round of competition). Overall, the rate and magnitude of fitness gains for SLEV were greater than those of WNV (Fig. 2B). After three passages (SLE CP3), all SLEV lineages increased in fitness significantly (P<0.01). Fitness gains substantially increased and appeared to be maximized by passage 20 (SLE CP20) in two of four lineages. The remaining two lineages continued to gain in fitness to passage 40 (SLE CP40). SLEV relative fitness advantages ranged from approximately 1.5 to 2 for SLE CP40.

Competition assays of C6/36-passaged virus populations were performed in avian cells (DF-1) to determine whether an increased competitive advantage in mosquito cells affected fitness in the natural alternative host cell line (Fig. 2). Measurements of relative fitness are equivalent to normalized slopes of vector plots resulting from serial rounds of competition (see Materials and methods). A representation of how all vector plots were constructed is shown in plaque assay portion of Fig. 3. With the exception of SLE CP3, increases in relative fitness were not observed in any lineage of either virus in avian cells (Fig. 2B). All three lineages of C6/36-passaged SLEV and WNV exhibited slight decreases in fitness in avian cells, but these decreases were statistically significant (P<0.05) for only one lineage of each virus. Furthermore, no decreases were comparable in magnitude to the increases measured in C6/36 cells.

Fig. 3.

Fig. 3

Open in a new tab

Comparison of the relative fitness of C6/36-passed SLEV, measured by plaque assay and RT-PCR. Four rounds of competition between test (unpassaged SLEV and passages 3, 20, and 40) and control (MARM) viruses were completed in triplicate in C6/36 cells. Each round consisted of 96 h of competitive replication. Changes in ratios of test:control virus were quantified by Vero cell plaque assay (duplicate) or real-time RT-PCR (triplicate). An MOI of 0.01 was maintained for each round of infection.

The results of competition assays performed in Vero cells were more variable. Some lineages of C6/36-passaged viruses demonstrated no fitness changes, while others incurred modest increases or decreases in relative fitness at various passage numbers. Some of these differences were statistically significant, but none was comparable to the increases seen in C6/36 cell competition assays.

The competition assay results (Fig. 2) were based on plaque titration on Vero cells in the presence or absence of MAb. The gains in relative fitness of infectious SLEV determined by plaque assay were compared, for representative samples, to trends measured by RT-PCR assays designed to distinguish the amount of test and control virus RNA in samples. Although some of the slopes in the fitness vector plots of two C6/36 cell passage lineages (A and C) are steeper (higher relative fitness) with plaque assay quantitation, similar trends were measured with RT-PCR analysis (Fig. 3). The differences in magnitude of relative fitness measurements for passaged SLEV can be attributed to an increase in cross reactivity of SLE Kern and MARM as ratios increase. Experiments with control mixtures determined that cross reactivity begins to affect accuracy of quantitation at approximately 100:1 and increases with further gains (data not shown). For this reason, fitness advantages for SLE CP20 and SLE CP40 are dampened somewhat in later rounds of competition when assessed by RT-PCR. Statistically, results from both plaque and RT-PCR assays for lineages A and C demonstrated that SLE CP20 and SLE CP40 had relative fitness values that were significantly greater than relative fitness values of unpassaged SLEV (SLE UP) (P<0.05). SLE CP3 results were more variable. For lineage A, the plaque assay demonstrated that SLE CP3 relative fitness was significantly greater than SLE UP relative fitness and less than SLE CP20 and CP40 relative fitness, thus showing a gradual gain in fitness during serial passaging. This difference was not seen for lineage C by plaque assay or for either lineage by RT-PCR.

Changes in viral growth patterns following serial passage in mosquito cells

Individual growth curves were analyzed in both mosquito and avian cells to determine whether fitness gains correlated with replicative advantages for passaged viruses. Both WNV and SLEV demonstrated higher titers in C6/36 cells compared to unpassaged virus following passage in C6/36 cells (Figs. 4A and B). For WNV, a difference in mean titer between passaged (WNV CP19 and WNV CP39) and unpassaged virus (WNV UP) was observed by 24 hpi, and the difference increased through 96 hpi when it peaked with a greater than 100-fold advantage of passaged virus (Fig. 4A). This peak corresponded to the harvest time for competition assays and reflected the increase in relative fitness measured by competition assay. A difference in titer between WNV CP19 and WNV CP39 was not observed at any time point (Fig. 4A), despite differences in relative fitness (Fig. 2A).

Fig. 4.

Fig. 4

Open in a new tab

Replication of C6/36 passaged (CP) and unpassaged viruses (UP) in C6/36 and DF-1 cell culture. Multiplicity of infection for all growth curves is 0.01 PFU/cell. Results are presented as the means± SD of duplicate or triplicate assays. (A) WNV replication in C6/36 cell culture. (B) SLE replication in C6/36 cell culture. (C) WNV replication in DF-1 cell culture. (D) SLE replication in DF-1 cell culture.

C6/36-passed SLEV also demonstrated an accelerated rate of replication in C6/36 cells, resulting in differences 2.5 and 3.5 log10 plaque forming units (PFU)/ml by 24 hpi for SLE CP20 and CP40, respectively (Fig. 4B). The differences in titer decreased to 1–2 log10 PFU/ml (SLE CP20) and 2–3 log10 PFU/ml (SLE CP40) at 48 hpi. Although these results are consistent with the accelerated fitness values measured by competition assay with SLEV (Fig. 2B), they differ slightly in that SLE CP40 had a replicative advantage over SLE CP20 despite no measurable difference in relative fitness.

Neither SLEV nor WNV showed similar gains when growth curves were analyzed following infection in avian cells, indicating a specific adaptation to mosquito cells following passage in these cells (Figs. 4C and D). There were no measurable differences in replicative ability between passaged and unpassaged WNV in avian cells (Fig. 4C), and passaged SLEV displayed only moderately decreased replicative ability in avian cells (Fig. 4D).

Genetic changes in WNV and SLEV following serial passages in mosquito cells

Genetic changes resulting from passage of WNV and SLEV in mosquito cells were identified in the nucleotide sequences of the full-length genomes of unpassaged and passaged virus from representative lineages (Table 1). Although relatively few mutations accumulated during passaging of either virus, more mutations were identified in SLEV than in WNV (8 vs. 3). Only three mutations were generated in the full-length genome of WNV CP39, two of which were nonsynonymous. Both of the resultant amino acid changes were conservative. One change occurred at nucleotide position 1712 of the envelope in WNV CP39, but this change was not identified in WNV CP19, by which time an increase in viral fitness had been demonstrated. One amino acid change also occurred in NS4A, a nonstructural region which has no known role in replication. This was the only mutation identified in the full-genome analysis of WNV CP19. All mutations in WNV were confirmed in a second distinct lineage. Three of the six nonsynonymous changes found in SLE CP40 occurred in the NS4A region, one of which was confirmed in SLE CP20. Two amino acid changes in the envelope region were generated in SLE CP40, while one was found in the prM region. Of these three changes, only the prM mutation was also found in SLE CP20. The prM change and envelope change at position 1428 were both nonconservative amino acid changes.

Table 1.

Nucleotide (NT) and amino acid (AA) changes in sequences of C6/36-passaged WNV and SLEV populations relative to unpassaged clonal populations following full-genome analyses

Virus NT position Region NT change AA change
WNV CP19 6687 NS4A G → T Q → H
WNV CP39 1712 ENV A → G K → R
3663 NS2A T → C None
6687 NS4A G → T Q → H
SLEV CP20a   679 prM A → G E → G
6720 NS4A T → A T → A
SLEV CP40   679 prM A → G E → G
1334 ENV G → T K → N
1428 ENV T → G S → A
1796 ENV T → C None
4572 NS2B C → T None
6720 NS4A T → A T → A
6968 NS4B T → A V → E
7533 NS4B A → C I → L
Open in a new tab
a

Sequence obtained only for regions where changes had been identified in SLEV CP40.

Discussion

We have developed an in vitro system to measure relative fitness of two flaviviruses, which was modified from a method developed for VSV (Holland et al., 1991). The system is based on the development of surrogate wild-type MARM viruses that replicate similarly to those of the test viruses and can be accurately distinguished during competition assays. As a validation of the plaque assay results, we demonstrated that patterns of fitness change could also be measured by RT-PCR. We also performed viral growth curve analyses and demonstrated that, although changes in relative fitness often are reflected in growth patterns, there are inconsistencies between the two complementary methods that reflect the importance of each. Using these methods, we sought to clarify the contributions of an individual host cell type to adaptation and viral evolution of WNV and SLEV.

Alternate replication in vertebrates and invertebrate vectors has been hypothesized to explain the evolutionary stability of WNV and other arboviruses (Scott et al., 1994; Weaver et al., 1992). This hypothesis implies a compromise between presumably nonintersecting fitness peaks, which in turn slows accumulation of mutations that are exclusively advantageous to either vertebrate or vector. The experiments presented here demonstrate that both WNV and SLEV, after serial passage in mosquito cell culture, showed large relative fitness gains in the homologous system (mosquito cells) that did not correspond to gains in an alternate system (avian cell culture). This cell-specific adaptation in fitness was also reflected in virus replicative ability. Although the two flaviviruses, WNV and SLEV, are closely related genetically and structurally, SLEV showed an ability to adapt more quickly and to attain larger gains in fitness than did WNV. Furthermore, more mutations (8 vs. 3) were generated in SLE CP40 than in WNV CP39. This result is somewhat surprising, given the fact that SLEV had consistently grown to an approximately 1 log10 lower titer than WNV at the time of harvest (Fig. 4), and it demonstrates that closely related RNA viruses may differ in their rates of adaptation and evolution. These differences could reflect the inherent robust and catholic nature of WNV as compared with the specificity of SLEV. WNV has disseminated rapidly and successfully throughout diverse ecological habitats where it has infected a broad range of mosquito and bird species worldwide (Hayes et al., 2005), while SLEV has a much narrower host range (Monath, 1980).

The specific adaptation of SLEV and WNV to the mosquito cells in which they were passed is consistent with findings of studies done with two alphaviruses, EEEV and SINV, which also exhibited cell-specific fitness gains (Greene et al., 2005; Weaver et al., 1999; Cooper and Scott, 2001). The use of cell culture passage producing virus strains which are attenuated in other in vitro and in vivo systems is well documented for the flaviviruses dengue and yellow fever (Barrett et al., 1990; Butrapet et al., 2000; Halstead and Marchette, 2003). Dengue studies, however, used cell culture systems that did not represent natural host cell types and, therefore, give us little insight into adaptations important in perpetuation of flaviviruses in nature. Although some yellow fever studies passaged in a relevant cell type, attenuation was measured only in vivo and, therefore, results are not directly applicable to comparison with the results presented here. The idea that alternating replication constrains evolution predicts that a gain in fitness in the invertebrate vector should result in a loss of fitness in the vertebrate host. In contrast to the work of Weaver and coworkers (1999), the gains in fitness for WNV and SLEV we measured in mosquito cells do not correspond to large fitness losses in a vertebrate (avian) cell line. Although fitness losses in DF-1 cells were seen in most lineages, they were modest and inconsistent. This difference could be attributed to the cell lines used to represent the ‘bypassed’ portion of the transmission cycle. For WNV, SLEV, EEEV and SINV, the natural reservoir host is avian. Hamster cells (BHK) were used as ‘bypassed’ cell line in the previous studies with EEE and SINV, whereas chicken (DF-1) cells were used here. The lack of change in avian cells could therefore be attributed to the inherent adaptation of these viruses to their natural host cell type. The failure to detect significant losses could also be attributed to the adaptability of the virus population as a result of its genetic diversity. It has been shown that WNV in nature exists as highly diverse populations (Jerzak et al., 2005), a characteristic typical of RNA viruses (Domingo et al., 1998; Eigen, 1993; Eigen and Biebricher, 1988). This diversity could provide genetic variants that allow rapid adaptability upon entry into a new environment. If this is the case, some mutations may in fact be detrimental to replication in vertebrate cells. However, this effect could not be accurately measured without a genetically homogenous population with these changes. The impact of WNV and SLEV quasispecies on both fitness and population adaptability is currently under investigation.

Studies with VSV suggest that arbovirus evolution is not constrained by alternating cellular environments and, rather, that virus populations are, in effect, generalists that find replication in disparate cellular environments indistinguishable (Novella et al., 1999). If this were the case for flaviviruses, adaptation to mosquito cells should result in fitness gains in avian cells. Although we did not measure substantial gains or declines in viral fitness in avian cells, our results clearly indicate that mosquito and avian cells do not constitute similar environments for flavivirus replication. Additional studies with VSV suggested that differences in replicative strategy between invertebrate vectors (persistent replication) and vertebrate hosts (acute infection) play a major role in shaping arbovirus evolution (Novella et al., 1995; Zarate and Novella, 2004). Since in our study all infections were acute, the results clearly indicate that cell type is sufficient for specific adaptation of flaviviruses. However, replicative strategy, together with the further constraints of limited cell types in vitro, is a factor to consider when assessing how the adaptation measured here is applicable to the intricacy of adaptation in natural viral populations. Further studies to investigate whether in vitro adaptations translate to in vivo systems are in progress.

The relatively small number of consensus mutations, taken together with the nature of the mutations (predominately conservative) following sequential mosquito cell culture passages, is not necessarily compatible with the idea that evolutionary stability can be attributed to replication in alternating environments. Nevertheless, the degree of genetic change identified here agrees with that seen in previous studies (Chen et al., 2003; Greene et al., 2005; Novella et al., 1999; Weaver et al., 1999) and, without a comparison to alternating environments, the constraints of cycling cannot be assessed accurately. Though low in number, the genetic changes measured here were coupled with profound phenotypic effects, indicating that adaptation to a specific environment may require very few changes in the consensus sequence. Such a phenomenon has been observed with the dissemination of WNV throughout North America, i.e., rapid success in a naive environment with few changes in consensus sequence (Davis et al., 2005). It also has been shown, for VSV, that fitness gains can occur with no changes in consensus sequence (Novella and Ebendick-Corp, 2004). For WNV, we confirmed the sequence changes in a second distinct lineage. This suggests positive selection for these changes and identifies them as potentially significant in generating fitness advantages.

One and two synonymous mutations were detected in the envelope region of WNV and SLEV, respectively. Our expectation was that the majority of important changes would accumulate here due to this region’s clearly defined role in binding and replication (Chambers et al., 1990; Scherret et al., 2001). Surprisingly, though, the amino acid changes common in both WNV CP19 and CP39, and SLE CP20 and CP40, were not found in the envelope region. In both WNV CP19 and SLE CP20, a single amino acid change was found in the NS4A gene, a region that has no known role in replication. Since this represents one of two changes confirmed in SLE CP20 and the only change identified in full-genome sequences of WNV CP19, it seems plausible that this region is playing a role in creating fitness advantages in mosquito cells. Furthermore, three of the six amino acid changes identified in SLE CP40 were within the NS4 region of the genome. Studies using reverse genetics are underway to investigate the roles of NS4 and other potentially important genes in flavivirus replication.

Materials and methods

Cells and media

African green monkey kidney cells (Vero, ATCC #CCL-81) were grown in minimal essential medium (MEM, Gibco, Invitrogen Corp, Carlsbad, CA) supplemented with 10% fetal bovine serum (FBS, Hyclone, Logan, UT), 2 mM l-glutamine, 1.5 g/l sodium bicarbonate, 100 U/ml of penicillin, and 100 µg/ml of streptomycin. Aedes albopictus mosquito cells (C6/36, ATCC #CRL-1660) were maintained in MEM supplemented with 10% FBS, 2 mM l-glutamine, 1.5 g/l sodium bicarbonate, 0.1 mM non-essential amino acids, 100 U/ml of penicillin, and 100 µg/ml of streptomycin. Chicken embryo fibroblast cells (DF-1, ATCC #CRL-12203, cell line derived from East Lansing ELL-0 chicken eggs) were grown in DMEM (ATCC #30–2002) supplemented with 10% FBS, 100 U/ml of penicillin and 100 µg/ml of streptomycin. Vero, C6/36, and DF-1 cells were grown and maintained at the optimum temperature for each cell line: 37 °C, 28 °C, and 39 °C, respectively. Confluent monolayers of each cell line were prepared for infection by seeding six-well plates (Costar, Corning, NY) with 6 × 105 cells/well in 3 ml of appropriate media and incubating at the proper temperature for 3 days. Confluent cell monolayers and virus-infected cells were maintained in the appropriate media for the cell line, supplemented with 2% FBS.

Virus strains and MAb

Biological clones of WNV (3356.1.1.1) and SLE Kern (217.3.1.1) were isolated by three rounds of plaque purification on Vero cells. WNV 3356.1.1.1 was derived from WNV NY003356, a primary isolate from the kidney tissue of an American crow that was collected in 2000 in Staten Island, NY and prepared by one round of amplification in Vero cells (Ebel et al., 2001). SLE Kern 217.3.1.1 was derived from the SLE Kern 217 strain that was originally isolated in 1989 from Culex tarsalis from Kern County, CA, and passaged twice in Vero cells (obtained from Dr. William Reisen, University of California at Davis; Kramer and Chandler, 2001). The biological clones were propagated for use in these fitness studies by two (WNV) or three (SLE) rounds of Vero cell amplification and quantified by plaque assay on Vero cells (Payne et al., 2006). Replication of biological clones was compared to replication of original isolates in many in vitro and in vivo systems, and no significant differences were identified (data not shown). WNV and SLE MARMs were developed by three rounds of plaque purification in Vero cells in the presence of highly neutralizing MAb. MARM WNV 34.1 and SLE MARM-8 were isolated in the presence of WNV MAb 5H10 (BioReliance Invitrogen Bioservices #81-003, Rockville, MD) and SLE MAb 6B5A-2 (obtained from John Roehrig, Centers for Disease Control and Prevention, Fort Collins, CO), respectively. Sequence analysis demonstrated amino acid substitutions T→I at position 652 in the envelope of WNV 34.1 and L→P at position 388 in the envelope of SLE-MARM 8. MARM isolates were propagated for use in fitness studies by two (WNV) or three (SLE) rounds of Vero cell amplification in the presence of the appropriate MAb. All titrations of MARMs were performed by plaque assay on Vero cells in the presence of the appropriate MAb in the agar overlay to prevent phenotypic mixing and masking (Holland et al., 1989). MARM titers were identical in the presence or absence of MAb. Multiple aliquots of all viral stocks were stored frozen at −80 °C.

Serial passage of virus

Three lineages each of WNV 3356.1.1.1 and SLE Kern 217.3.1.1 were obtained by 40 serial passages in C6/36 cells, using a multiplicity of infection (MOI) of 0.1 PFU/cell, based on Vero cell titers, for each passage. Virus was adsorbed to confluent cell monolayers in six-well plates for 1 h at 28 °C, with frequent rocking. Following adsorption, the inoculum was removed, cells were washed with MEM, 3 ml of maintenance media was added, and plates were incubated at 28 °C. Medium from each well was harvested at 72 hpi and stored, in aliquots, at −80 °C. Following each passage, the viral harvest was quantified by plaque assay on Vero cells and diluted to produce an MOI of 0.1 PFU/cell for each subsequent passage.

Viral growth curve analysis

Six-well plates containing confluent monolayers of Vero, C6/36, or DF-1 cells were infected with virus, in duplicate or triplicate, at an MOI of 0.1 or 0.01 PFU/cell, based on Vero cell titer. At the end of a 1 h absorption period at 37 °C (Vero), 28 °C (C6/36), or 39 °C (DF-1), the inoculum was removed, 3 ml of maintenance media was added to each well, and the plates were returned to the appropriate temperatures. Samples, consisting of 100 µl of media, were taken at 12, 24, 36, 48, 72, 96, and 120 hpi, diluted 1:10 in BA-1 containing 20% FBS, and stored at −80 °C. All samples were titered, in duplicate, by plaque assay on Vero cells, and growth curves were constructed, using the mean titer for each time point.

Determination of viral fitness by competition assay

Relative fitness of WNV and SLEV was evaluated by competition assays in C6/36, Vero, and DF-1 cells, based, with some modifications, on the protocol used by Holland et al. (1991) for VSV. Briefly, confluent cell monolayers in six-well plates were infected in triplicate with a mixture of control and test virus at an MOI of 0.01 PFU/cell, based on Vero cell titer. The controls were WNV MARM and SLEV MARM, each serving as surrogate wild-type virus. The test virus consisted of WNV or SLEV biological clones before and after passage in C6/36 cells. The ratio of test:control virus in the inoculum mixture used for the first competition round, determined by preliminary assays, was 1:1 for all competition assays, except for SLE in C6/36 cells, which was 1:4. After a 60-min absorption period at 28 °C (C6/36), 37 °C (Vero), or 39 °C (DF-1), the infected monolayers were washed three times, overlaid with 3 ml of maintenance medium, and incubated at the temperatures used for the infection. Medium from the infected cultures was harvested at 48, 72, or 96 hpi, diluted 1:10 in growth medium supplemented with 20% FBS, and frozen at −80 °C for later titration, in duplicate, on Vero cells. The time of harvest for each cell line was chosen from previous growth curves (Fig. 1). A total of three or four successive competition rounds were carried out, using Vero cell titers by plaque assay or fluorescent focus assay (Payne et al., 2006), to determine the dilution needed to maintain an MOI of 0.01 PFU/cell. The purpose of using multiple rounds of competition was to correct for random fluctuations in ratios which could be misinterpreted as fitness changes if a single passage of competition was used. Multiple rounds of competition allowed construction of vector plots which accurately measure changes in fitness. This is particularly important in validating modest changes. The competition assays were evaluated by measuring the amount of control and test virus present in the initial input mixture, and at the end of each competition round, by duplicate plaque assay on Vero cells, in the presence and absence of highly neutralizing MAb in the agar overlay. The quantity of test virus was obtained by subtracting the titer in the presence of MAb (MARM control titer) from the titer without MAb (total titer), and the ratio of test:control virus was determined for each of the three competition round replicates. Fitness vector plots were constructed by normalizing the ratio for each round of infection to the ratio of the initial input and plotting log10 transformed values against the competition round. Relative fitness values were obtained from the slope of each vector line (Duarte et al., 1992; Martinez et al., 1991). Post tests following regression analysis were performed to confirm that no data set was significantly nonlinear (GraphPad Prism, Version 4.0). This allowed us to confidently conclude that relative fitness changes quantified were a result of passaging and not subsequent rounds of competition. The slopes were then compared by ANCOVA to determine whether significant fitness changes had occurred (P<0.05).

Competition assays of C6/36-passed SLEV in C6/36 cell culture also were analyzed by real-time RT-PCR, using primer-probe sets designed to distinguish between the envelope regions of SLE MARM-8 and Kern 217.1.1.1. MARM-8 was detected with forward primer 5′-AGT-ATT-GTT-ACG-AAG-CAACTT-TGG-A, reverser primer 5′-TCT-CCT-GTT-GTA-GGGCAC-CTT, and TaqMan probe 6FAM-ACG-CCG-TCA-ACAGTG-NFQ. Kern 217.1.1.1 was detected with forward primer 5′-CGT-GAG-TAT-TGT-TAC-GAA-GCA-ACC-T, reverser primer 5′-TCT TCT-CCT-GTT-GTA-GGG-CAC-CTT, and TaqMan probe 6FAM-ACA-CGC-TGT-CAA-CAG-TG-NFQ. RT-PCR assays were performed on an ABI Prism 7700 sequence detector, using TaqMan one-step RT-PCR master mix (Applied Biosystems, Foster City, CA), as previously described (Shi et al., 2001).

Sequencing

RNA was extracted from WNV and SLEV using RNeasy spin columns (Qiagen, Valencia, CA) according to manufacturer’s protocol. Primers for WNV were designed from GenBank AF260967, and for SLEV, from M16614 and AF1160194. One-step RT-PCR (Qiagen) was conducted using primers (Table 2) to generate nine overlapping PCR products. Reverse transcription reactions were carried out at 50 °C for 30 min followed by inactivation of the transcriptase at 95 °C for 15 min. Amplification was then carried out for 40 cycles at 94 °C for 20 s, 5 °C for 30 s, 72 °C for 2 min, with final elongation at 72 °C for 10 min. PCR products were visualized on a 1.5% gel, and then bands were then allowed to run through 1% Nusieve GTG low-melting agarose (Cambrex BioScience, Rockland, ME). Sequencing was performed with ABI 3700 automated sequencers (Applied Biosystems) using overlapping primers to give a minimum of two-fold redundancy. Sequences were compiled and edited by using DNASTAR software package (Madison, WI). Novel full-length sequence generated for SLE 217.3.1.1 was deposited into GenBank under accession number DQ525916.

Table 2.

Primers used for full-genome sequencing of WNV and SLEV

WNV SLE
Primer name Sequence (5′–3′) Map position Primer name Sequence (5′–3′) Map position
WN1S AGTAGTTCGCCTGTGTGAGCTGAC 1–24 SLE1S ATGTTCGCGTTGGTGAGTGGAGAG 1–23
WN1R CAGACTGCTTCGTGGCGTGTG 1700–1720 SLE1R AACTGTTCCATACTCGCCCATGTTG 1496–1520
WN2S TCAGGCAGGGAGATTCAGC 1452–1470 SLE2S TTGGTGATCATGCTGATGCTGATT 924–947
WN2R GGTGGCGGTGAGGCGTTTAGGTG 2768–2790 SLE2R AACGTCGTTGTACACGAAGATGC 2518–2540
WN3S AAGGCTGTCCATCAAGTGTTC 2284–2305 SLE3S GGCACCACCCAGATTAACTACCAC 2127–2150
WN3R CCCACCAAGATGACATAGCGTAAC 3687–3710 SLE3R AGGCTGCTCCAATAACCATCAAGA 3835–3858
WN4S ATTGACTTCGATTACTGCCCAG 3292–3313 SLE4S GTGCGATACCTGGTCCTTGTTGG 3684–3706
WN4R ACCTCATCCTGCCCGTTCCACTTG 4872–4895 SLE4R AATCATTTGGACCTCCTCAGTTC 4867–4889
WN5S CAAGCAGGAGCGGGCGTGAT 4714–4733 SLE5S CAAAAGTGTATCCAAAGTGTGAGA 4624–4647
WN5R GATCCGAGTACACCCTGGCGTCAA 6401–6424 SLE5R ACGCAATCGAAACTCCCCATCC 6152–6173
WN6S CTACCAACCAGAGCGTGAGAA 6138–6158 SLE6S TGGACCCATGGCAATCACA 5927–5945
WN6R TACCCTCTGACTTCTTGGACTCTT 7965–7988 SLE6R CTGTTCTCCGCTGGGCTGCTCT 7308–7329
WN7S CAACAACTGCCATCGGACTC 7574–7593 SLE7S ATCTCAAGCTGGTGTACTGTTAG 7124–7147
WN7R GGCGCTCCTCCATTGATTC 8934–8952 SLE7R ACGTTCTGTATGGGTTGTTGTGGT 8560–8583
WN8S GTTCGCTGGTCAATGGAGTGGTCA 8639–8662 SLE8S ATGGGGAGGATGGACAAACAGA 8391–8412
WN8R TCCGATGATTGCTCTGACTTGGTT 10,297–10,320 SLE8R ACATAAGCAACCACATCTGAGCATAG 9956–9981
WN9S AGACAAAACCCCAGTGGAGAAATG 10,170–10,193 SLE9S CCCTTCTGCTCACACCACTTCAAT 9804–9827
WN9R AGATCCTGTGTTCTCGCACCACCA 11,006–11,029 SLE9R GGGTCTCCTCTAACCTCTAG 10,794–10,814
Open in a new tab

Acknowledgments

The authors thank the Wadsworth Center Molecular Genetics Core for providing sequencing data and the Wadsworth Center Media and Tissue Culture Facility for providing cells and media for this work. We also appreciate the efforts of the Wadsworth Center Arbovirus Laboratories in assisting with cell culture work. We thank G.D. Ebel for critical review of this article. We also thank William Reisen for funding support. This work was supported partially by federal funds from the National Institute of Allergy and Infectious Disease, National Institutes of Health contract number NO1-AI-25490 and National Institutes of Health grant numbers RO1-AI-47855 and RO1-AI-50758.

References

  1. Austin RJ, Whiting TL, Anderson RA, Drebot MA. An outbreak of West Nile virus-associated disease in domestic geese (Anser anser domesticus) upon initial introduction to a geographic region, with evidence of bird to bird transmission. Canadian Veterinary Journal. 2004;45:117–123. [PMC free article] [PubMed] [Google Scholar]
  2. Barrett AD, Monath TP, Cropp CB, Adkins JA, Ledger TN, Gould EA, Schlesinger JJ, Kinney RM, Trent DW. Attenuation of wild-type yellow fever virus by passage in HeLa cells. J. Gen. Virol. 1990;71(pt. 10):2301–2306. doi: 10.1099/0022-1317-71-10-2301. [DOI] [PubMed] [Google Scholar]
  3. Butrapet S, Huang CYH, Pierro DJ, Gwong-Jen JC, Bhamarapravati N, Gubler DJ, Kinney R. Attenuation markers of a candidate dengue type 2 vaccine virus, strain 16681 (PDK-53), Are defined by mutations in the 5′ noncoding region and nonstructural proteins 1 and 3. J. Virol. 2000;74:3011–3019. doi: 10.1128/jvi.74.7.3011-3019.2000. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Chambers TJ, Hahn CS, Galler R, Rice CM. Flavivirus genome organization, expression, and replication. Annual Review of Microbiology. 1990;44:649–688. doi: 10.1146/annurev.mi.44.100190.003245. [DOI] [PubMed] [Google Scholar]
  5. Chandler LJ, Parsons R, Randle Y. Multiple genotypes of St. Louis encephalitis virus (Flaviviridae: Flavivirus) circulate in Harris County, Texas. American Journal of Tropical Medicine and Hygiene. 2001;64:12–19. doi: 10.4269/ajtmh.2001.64.1.11425155. [DOI] [PubMed] [Google Scholar]
  6. Chen WJ, Wu HR, Chiou SS. E/NS1 modifications of dengue 2 virus after serial passages in mammalian and/or mosquito cells. Intervirology. 2003;46:289–295. doi: 10.1159/000073208. [DOI] [PubMed] [Google Scholar]
  7. Cilnis MJ, Kang W, Weaver SC. Genetic conservation of highlands J viruses. Virology. 1996;218:343–351. doi: 10.1006/viro.1996.0203. [DOI] [PubMed] [Google Scholar]
  8. Cooper LA, Scott TW. Differential evolution of eastern equine encephalitis virus populations in response to host cell type. Genetics. 2001;157:1403–1412. doi: 10.1093/genetics/157.4.1403. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Davis CT, Ebel GD, Lanciotti RS, Brault AC, Guzman H, Siirin M, Lambert A, Parsons RE, Beasley DW, Novak RJ, Elizondo-Quiroga D, Green EN, Young DS, Stark LM, Drebot MA, Artsob H, Tesh RB, Kramer LD, Barrett AD. Phylogenetic analysis of North American West Nile virus isolates, 2001–2004: evidence for the emergence of a dominant genotype. Virology. 2005;342:252–265. doi: 10.1016/j.virol.2005.07.022. [DOI] [PubMed] [Google Scholar]
  10. Day JF, Stark LM. Frequency of Saint Louis encephalitis virus in humans from Florida, USA: 1990–1999. Journal of Medical Entomology. 2000;37:626–633. doi: 10.1603/0022-2585-37.4.626. [DOI] [PubMed] [Google Scholar]
  11. Domingo E, Escarmis C, Sevilla N, Baranowski E. Population dynamics in the evolution of RNA viruses. Adv. Exp. Med. Biol. 1998;440:721–727. doi: 10.1007/978-1-4615-5331-1_93. [DOI] [PubMed] [Google Scholar]
  12. Drake JW, Holland JJ. Mutation rates among RNA viruses. Proc. Natl. Acad. Sci. U. S. A. 1999;96:13910–13913. doi: 10.1073/pnas.96.24.13910. [DOI] [PMC free article] [PubMed] [Google Scholar]
  13. Duarte E, Clarke D, Moya A, Domingo E, Holland J. Rapid fitness losses in mammalian RNA virus clones due to Mueller’s ratchet. Proc. Natl. Acad. Sci. U. S. A. 1992;89:6015–6019. doi: 10.1073/pnas.89.13.6015. [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Dupuis AP, Marra PP, Kramer LD. Serologic evidence of West Nile virus transmission, Jamaica, West Indies. Emerging Infect. Dis. 2003;9:860–863. doi: 10.3201/eid0907.030249. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Dupuis AP, Marra PP, Reitsma R, Jones MJ, Louie KL, Kramer LD. Serologic evidence for West Nile virus transmission in Puerto Rico and Cuba. American Journal of Tropical Medicine and Hygiene. 2005;73:474–476. [PubMed] [Google Scholar]
  16. Ebel GD, Dupuis AP, II, Ngo KA, Nicholas DC, Kauffman EB, Jones SA, Young DM, Maffei JG, Shi P-Y, Bernard KA, Kramer LD. Partial genetic characterization of West Nile virus strains, New York State. Emerging Infect. Dis. 2001;7:650–653. doi: 10.3201/eid0704.010408. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Ebel GD, Carricaburu J, Young D, Bernard KA, Kramer LD. Genetic and phenotypic variation of West Nile virus in New York, 2000–2003. American Journal of Tropical Medicine and Hygiene. 2004;71:493–500. [PubMed] [Google Scholar]
  18. Eigen M. Viral quasispecies. Sci. Am. 1993;269:42–49. doi: 10.1038/scientificamerican0793-42. [DOI] [PubMed] [Google Scholar]
  19. Eigen M, Biebricher DK. Sequence space and quasispecies distribution. In: Domingo E, Holland JJ, Ahlquist P, editors. RNA Genetics. Boca raton, Florida: CRC Press; 1988. pp. 211–245. [Google Scholar]
  20. Greene IP, Wang E, Deardorff ER, Milleron R, Domingo E, Weaver SC. Effect of alternating passage on adaptation of Sindbis virus to vertebrate and invertebrate cells. J. Virol. 2005;79:14253–14260. doi: 10.1128/JVI.79.22.14253-14260.2005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Halstead SB, Marchette MJ. Biological properties of Dengue virus following serial passage in primary dog kidney cells. American Journal of Tropical Medicine and Hygiene. 2003;69(6 Suppl.):5–11. doi: 10.4269/ajtmh.2003.69.6_suppl.0690005. [DOI] [PubMed] [Google Scholar]
  22. Hayes EB, Komar N, Nasci RS, Montgomery SP, O’Leary DR, Campbell GL. Epidemiology and transmission dynamics of West Nile Virus disease. Emerging Infect. Dis. 2005;11:1167–1173. doi: 10.3201/eid1108.050289a. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Holland JJ, DeLaTorre JC, Steinhauer DA, Clarke D, Duarte E, Domingo E. Virus mutation frequencies can be greatly under-estimated by monoclonal antibody neutralization of virions. J. Virol. 1989;63:5030–5036. doi: 10.1128/jvi.63.12.5030-5036.1989. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Holland JJ, DeLaTorre JC, Clarke DK, Duarte E. Quantitation of relative fitness and great adaptability of clonal populations of RNA viruses. J. Virol. 1991;65:2960–2967. doi: 10.1128/jvi.65.6.2960-2967.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Holmes EC. Patterns of intra- and interhost nonsynonymous variation reveal strong purifying selection in dengue virus. J. Virol. 2003;77:11296–11298. doi: 10.1128/JVI.77.20.11296-11298.2003. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Jenkins GM, Rambaut A, Pybus OG, Holmes EC. Rates of molecular evolution in RNA viruses: a quantitative phylogenetic analysis. Journal of Molecular Evolution. 2002;54:156–165. doi: 10.1007/s00239-001-0064-3. [DOI] [PubMed] [Google Scholar]
  27. Jerzak G, Bernard KA, Kramer LD, Ebel GD. Genetic variation in West Nile virus from naturally infected mosquitoes and birds suggests quasispecies structure and strong purifying selection. J. Gen. Virol. 2005;86:2175–2183. doi: 10.1099/vir.0.81015-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Kramer LD, Bernard KA. West Nile virus in the western hemisphere. Curr. Opin. Infect. Dis. 2001;14:519–525. doi: 10.1097/00001432-200110000-00004. [DOI] [PubMed] [Google Scholar]
  29. Kramer LD, Chandler LJ. Phylogenetic analysis of the envelope gene of St. Louis encephalitis virus. Arch. Virol. 2001;146:2341–2355. doi: 10.1007/s007050170007. [DOI] [PubMed] [Google Scholar]
  30. Lanciotti RS, Roehrig JT, Deubel V, Smith J, Parker M, Steele K, Crise B, Volpe KE, Crabtree MB, Scherret JH, Hall RA, MacKenzie JS, Cropp CB, Panigrahy B, Ostlund E, Schmitt B, Malkinson M, Banet C, Weissman J, Komar N, Savage HM, Stone W, McNamara T, Gubler DJ. Origin of the West Nile virus responsible for an outbreak of encephalitis in the northeastern United States. Science. 1999;286:2333–2337. doi: 10.1126/science.286.5448.2333. [DOI] [PubMed] [Google Scholar]
  31. Martinez MA, Carrillo C, Gonzalez-Candelas F, Moya A, Domingo E, Sobrino F. Fitness alteration of foot-and-mouth disease virus mutants: measurement of adaptability of viral quasispecies. J. Virol. 1991;65:3954–3957. doi: 10.1128/jvi.65.7.3954-3957.1991. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Monath TP. St. Louis encephalitis. Washington, D.C.: American Public Health Association; 1980. [Google Scholar]
  33. Novella IS. Molecular basis of fitness loss and fitness recovery in vesicular stomatitis virus. J. Mol. Biol. 2004;342:1423–1430. doi: 10.1016/j.jmb.2004.08.004. [DOI] [PubMed] [Google Scholar]
  34. Novella IS, Clarke DK, Quer J, Duarte EA, Lee CH, Weaver SC, Elena SF, Moya A, Domingo E, Holland JJ. Extreme fitness differences in mammalian and insect hosts after continuous replication of vesicular stomatitis virus in sandfly cells. J. Virol. 1995;69:6805–6809. doi: 10.1128/jvi.69.11.6805-6809.1995. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Novella IS, Hershey CL, Escarmis C, Domingo E, Holland JJ. Lack of evolutionary stasis during alternating replication of an arbovirus in insect and mammalian cells. J. Mol. Biol. 1999;287:459–465. doi: 10.1006/jmbi.1999.2635. [DOI] [PubMed] [Google Scholar]
  36. Payne AF, Binduga-Gajewska I, Kauffman EB, Kramer LD. Quantitation of flaviviruses by fluorescent focus assay. J. Virol. Methods. 2006;134:183–187. doi: 10.1016/j.jviromet.2006.01.003. [DOI] [PubMed] [Google Scholar]
  37. Scherret JH, MacKenzie JS, Khromykh AA, Hall RA. Biological significance of glycosylation of the envelope protein of Kunjin virus. Ann. N. Y. Acad. Sci. 1999;951:361–363. doi: 10.1111/j.1749-6632.2001.tb02719.x. [DOI] [PubMed] [Google Scholar]
  38. Scott TW, Weaver SC, Mallampalli VL. Evolution of mosquito-borne viruses. In: Morse SS, editor. The Evolutionary Biology of Viruses. New York: Raven Press, Ltd.; 1994. pp. 293–324. [Google Scholar]
  39. Shi P-Y, Kauffman EB, Ren P, Felton A, Tai JH, Dupuis AP, II, Jones SA, Ngo KA, Nicholas DC, Maffei JG, Ebel GD, Bernard KA, Kramer LD. High throughput detection of West Nile virus RNA. Journal of Clinical Microbiology. 2001;39:1264–1271. doi: 10.1128/JCM.39.4.1264-1271.2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Weaver SC, Rico-Hesse R, Scott TW. Genetic diversity and slow rates of evolution in New World alphaviruses. Curr. Top. Microbiol. Immunol. 1992;176:99–117. doi: 10.1007/978-3-642-77011-1_7. [DOI] [PubMed] [Google Scholar]
  41. Weaver SC, Brault AC, Kang W, Holland JJ. Genetic and fitness changes accompanying adaptation of an arbovirus to vertebrate and invertebrate cells. J. Virol. 1999;73:4316–4326. doi: 10.1128/jvi.73.5.4316-4326.1999. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Woelk CH, Holmes EC. Reduced positive selection in vector-borne RNA viruses. Mol. Biol. Evol. 2002;19:2333–2336. doi: 10.1093/oxfordjournals.molbev.a004059. [DOI] [PubMed] [Google Scholar]
  43. Zarate S, Novella IS. Vesicular stomatitis virus evolution during alternation between persistent infection in insect cells and acute infection in mammalian cells is dominated by the persistence phase. J. Virol. 2004;78:12236–12242. doi: 10.1128/JVI.78.22.12236-12242.2004. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES