In vitro evolution of herpes simplex virus 1 (HSV-1) reveals selection for syncytia and other minor variants in cell culture

Abstract

The large dsDNA virus herpes simplex virus 1 (HSV-1) is considered to be genetically stable, yet it can rapidly evolve in response to strong selective pressures such as antiviral treatment. Deep sequencing has revealed that clinical and laboratory isolates of this virus exist as populations that contain a mixture of minor alleles or variants, similar to many RNA viruses. The classic virology approach of plaque purifying virus creates a genetically homogenous population, but it is not clear how closely this represents the mixed virus populations found in nature. We sought to study the evolution of mixed versus highly purified HSV-1 populations in controlled cell culture conditions, to examine the impact of this genetic diversity on evolution. We found that a mixed population of HSV-1 acquired more genetic diversity and underwent a more dramatic phenotypic shift than a plaque-purified population, producing a viral population that was almost entirely syncytial after just ten passages. At the genomic level, adaptation and genetic diversification occurred at the level of minor alleles or variants in the viral population. Certain genetic variants in the mixed viral population appeared to be positively selected in cell culture, and this shift was also observed in clinical samples during their first passages in vitro. In contrast, the plaque-purified viral population did not appear to change substantially in phenotype or overall quantity of minor allele diversity. These data indicate that HSV-1 is capable of evolving rapidly in a given environment, and that this evolution is facilitated by diversity in the viral population.

Keywords: HSV-1; evolution; minor variant; syncytia; viral fitness.

Figure 1.

Schematic of the in vitro evolution experiment approach and observed changes in plaque morphology: two populations of HSV-1 strain F were used to infect Vero cells at a multiplicity of infection (MOI) of 0.01. After infection was allowed to proceed for 72 h, progeny viral populations were harvested. This infectious cycle is referred to as a passage, and each viral population was carried through ten passages. Following each passage, each viral population was titered, visually examined for plaque morphology (see ‘Materials and Methods’ section for details), and prepared for sequencing. The entire series of 10 passages was performed in triplicate for the Mixed population and once for the Purified population. All replicates were titered and quantified for plaque morphology. Deep sequencing was performed on one lineage for each starting viral population. Plaque images were obtained by plating virus at limiting dilution on monolayers of Vero cells, then fixing and staining with methylene blue at 72 hpi. Tiled images (6 × 6) were exported from Nikon NIS-Elements software, and contrast inverted using Adobe Photoshop to show plaques more clearly. Scale bars indicate 2 mm.

Figure 2.

Gross changes in plaque morphology occur over sequential passages, while consensus-level genome changes are limited. (A) Plaque morphology was quantified for each passage of the Mixed and Purified viral populations. The Mixed population was passaged for three independent lineages (A, B, and C). (B) The viral titer was measured at each passage of the Mixed and Purified lineages. (C) A diagram of the full-length HSV-1 genome, depicts the structural repeats and major sites of tandem repeats (black boxes). An alignment of the consensus genome for each of ten passages of the Mixed and Purified viral populations are shown here. Each genome is represented by a gray bar and passages are displayed in order (P0–P10). Identical sequences are noted by gray in the identity plot above the alignment, with yellow and red indicating lesser levels of identity across the alignment. The terminal repeats (TRL/TRS) are trimmed from the consensus genomes as to not over-represent diversity in these duplicated regions. (UL, Unique Long; US, Unique Short; TRL, Terminal Repeat Long; IRL, Internal Repeat Long; IRS, Internal Repeat Short; TRS—Terminal Repeat Short, a/a’, repeat-containing site involved in genome cleavage and packaging).

Figure 3.

Minor variants are widespread in the mixed, but not the purified, viral populations. Each passage of the (A) Mixed and (B) Purified viral populations were sequenced and analyzed for minor variants. A subset of passages are shown here for space considerations. Each minor variant present in 2% or more of the viral popiulation is shown as a vertical bar, and multiple variants within one gene are connected with a horizontal line. A plot of sequence read depth across the HSV-1 genome is shown for P0 of each virus population (results are representative of all passages). Across all passages, the average coverage depth across the genome was 563 reads/position for the Mixed viral populations and 517 reads/position for the Purified viral populations. A diagram of the HSV-1 genome is shown in each panel for spatial orientation.

Figure 4.

Minor variants in the mixed viral populations occur at more diverse locations and at higher frequencies than in the purified viral populations. Minor variants were plotted according to their location in the genome (x-axis), as well as by their observed frequency in each passage of these viral populations (y-axis). Variants that occurred in the same genomic location are seen as vertical columns of dots. Lighter colors indicate variants observed in earlier passages, whereas darker colors indicate later passages. Overall, a larger number and higher frequency of minor variants are observed in (A) the Mixed viral population than in (B) the Purified viral population. Highly repetitive regions of the HSV-1 genome are denoted by black bars (OriLyt, UL36 PQ repeats, and the internal repeat region) and gray vertical columns.

Figure 5.

Minor variant dynamics shift over sequential passages in vitro. (A) A subset of the observed high-confidence minor variants in the Mixed population were plotted by their frequency in the viral population over passage (see Table 1 for full list of minor variants). Each variant would cause a change in the translated protein, whether through premature stop (UL13 S118[F.S.], W30*stop) or a missense variant (all others). Variants and their encoded proteins are listed in the legend according to their frequency at passage 10 (P10). (B) A diagram of the UL13 protein (518 AA in length) depicts the location of the six previously described catalytic domains of this kinase (Smith & Smith 1989), with an asterisk (*) denoting the catalytic lysine where a single point mutation can disrupt kinase activity (Cano-Monreal et al. 2008; Kawaguchi et al. 2003). The minor variants observed in this study, and consensus-level mutations observed in other virus strains in prior studies, are indicated by arrows and labels. GenBank accessions: strain F, GU734771 (Szpara et al. 2010); E25, HM585506; E06, HM585496; E11, HM585500 [all three from (Szpara et al. 2014)]; OD4, JN420342 (Lee et al. 2015). (C) The minor variants observed in the UL13 gene were separated for specific analysis. Each variant is predicted to produce a catalytically-inactive UL13 kinase. Additive Effect refers to the arithmetic addition of the frequency of each of these two individual variants. (D) New clinical isolates of HSV-1 were passaged in Vero cells, genome sequenced, and examined for minor variants in the UL13 gene at each passage. The frequency of genetic variants that would produce an inactive UL13 kinase are plotted over each passage.

Figure 6.

The combination of individual variants in gB (UL27) explained the syncytial phenotype observed in the Mixed viral population. The UL27 gene encoding gB includes two syncytia-inducing minor variants that encode an Arginine to Histidine change at amino acid 858 or a Leucine to Proline change at amino acid 817. Sequencing reads that span the nucleotides encoding both sites were identified for each passage. (A) Shown here is a random subset of the passage 7 (P7) sequencing reads surrounding this region, of which a total of 366 reads span both minor variants. The consensus genome is displayed at the top, with individual sequencing reads aligned below (yellow indicates forward-aligning sequencing reads, while purple indicates the reverse). Nucleotides matching the consensus genome are not shown, such that minor alleles that do not match the consensus sequence are the only ones highlighted. The relevant stretches of these sequencing reads are magnified, with any nucleotide that varies from the consensus genome noted (C–T for Arg858His or A–G for Leu817Pro). Note that the UL27 gene is on the reverse strand of the HSV-1 genome, so that the depicted nucleotides are in the reverse complement of what would be transcribed to produce the UL27 mRNA. (B) Across all ten passages, the percent of sequencing reads spanning this region that contained one, both, or neither variant is summarized here. Data for ‘both’ are included but are not visible on the graph, since only six sequence reads ever included both variants (three at P5, and three at P9). The yellow background panel in (B) connects the subset of P7 sequence read data shown in (A) to the relevant overall P7 percentage data graphed in (B). (C) The frequency of each variant in gB (UL27) was plotted alone, or as the sum of their frequency (Additive Variation (Var.)).

Figure 7.

Defined mixtures of purified syncytial and non-syncytial HSV-1 populations reveal a universal advantage of the syncytial phenotype during sequential passage in Vero cells. (A) Purified populations of HSV-1 that displayed a uniformly syncytial or non-syncytial phenotype were mixed in defined ratios (as shown in legend) and allowed to compete during sequential passages of infection on Vero cells. At each passage, the plaque morphology of the resulting viral population was counted and plotted as the percentage of total plaques that were syncytial. MOI (0.01) and time to harvest (72 hpi) of each passage are the same as the initial in vitro evolution experiment described in Figure 1. (B) The tail region of gB (encoded by UL27) that contains the L817P and R858H variants described in Figure 6 was Sanger sequenced from each purified starting population, the Mixed P10 population, and the 1:5 and 1:50 passage 10 (P10) populations. Representative traces are shown with the key nucleic acids highlighted, for the non-syncytial stocks (e.g., Purified), for the mixed population (e.g., Mixed), and for the syncytial stocks (shown here is 1:50 P10; similar data for purified syncytial and 1:5 P10 are not shown). Note that these reads are the reverse-complement of those shown in Figure 6.

Figure 8.

Passaged virus populations replicate more than input viral populations. Viral populations from the indicated passages of the Mixed or Purified stocks were used to infect Vero cells in (A) a single step growth curve (MOI = 10), or in (B) a multiple step growth curve (MOI = 0.01). Each assay was done in triplicate, with viral harvest and quantification by titering at the indicated timepoints. Data are plotted as titer compared to the first time-point for each experiment (2 h for A, 12 h for B). Error bars indicate standard deviation. In both growth curves, comparisons of passaged viral populations versus the input viral populations were significantly different at the final two time-points (two-way ANOVA, P < 0.05).