Genetic Diversity within Alphaherpesviruses: Characterization of a Novel Variant of Herpes Simplex Virus 2

Abstract

Very low levels of variability have been reported for the herpes simplex virus 2 (HSV-2) genome. We recently described a new genetic variant of HSV-2 (HSV-2v) characterized by a much higher degree of variability for the UL30 gene (DNA polymerase) than observed for the HG52 reference strain. Retrospective screening of 505 clinical isolates of HSV-2 by a specific real-time PCR assay targeting the UL30 gene led to the identification of 13 additional HSV-2v isolates, resulting in an overall prevalence of 2.8%. Phylogenetic analyses on the basis of microsatellite markers and gene sequences showed clear differences between HSV-2v and classical HSV-2. Thirteen of the 14 patients infected with HSV-2v originated from West or Central Africa, and 9 of these patients were coinfected with HIV. These results raise questions about the origin of this new virus. Preliminary results suggest that HSV-2v may have acquired genomic segments from chimpanzee alphaherpesvirus (ChHV) by recombination.

Importance: This article deals with the highly topical question of the origin of this new HSV-2 variant identified in patients with HIV coinfection originating mostly from West or Central Africa. HSV-2v clearly differed from classical HSV-2 isolates in phylogenetic analyses and may be linked to simian ChHV. This new HSV-2 variant highlights the possible occurrence of recombination between human and simian herpesviruses under natural conditions, potentially presenting greater challenges for the future.

FIG 1

Variant-specific duplex real-time PCR assay targeting the HSV-2v molecular signature within the UL30 (DNA polymerase) gene. The conserved regions and functional domains of the DNA polymerase (3,723 bp) gene are indicated by gray boxes. They are named as follows: Exo I (nucleotides 1092 to 1122), domain IV (nucleotides 1314 to 1440), delta-C region (nucleotides 1596 to 1884), domain II (nucleotides 2097 to 2223), domain VI (nucleotides 2331 to 2388), domain III (nucleotides 2430 to 2550), domain I (nucleotides 2658 to 2703), domain VII (nucleotides 2829 to 2853), and domain V (nucleotides 2877 to 2907) (9). HSV-2c (top), embodied by reference strain HG52, and HSV-2v (bottom), embodied by HSV-2v isolate 1 (HSV-2v [1]), are represented separately. Nucleotide mutations identified in HSV-2v isolate 1 relative to the nucleotide sequence of the HG52 reference strain (GenBank accession number Z86099) are indicated by vertical bars. The duplex real-time PCR assay was designed to use primers generating a 328-bp amplicon (forward primer H2_UL30-F and reverse primer H2_UL30-R) and sequence-specific probes for discrimination between the two types of virus on the basis of the molecular signature located between nucleotides 2880 and 3424 (H2V-P and H2C-P, 5′-labeled hydrolysis probes with FAM and HEX for the detection of HSV-2v and HSV-2c, respectively).

FIG 2

Linear map of the genomic distribution of the genes and microsatellites studied in the HG52 reference strain. The HSV-2 genome (HG52 reference strain [GenBank accession number Z86099]) is a linear double-stranded DNA molecule of 154,746 bp consisting of two covalently linked segments known as the unique long (UL) and unique short (US) sequences. Each segment is flanked by the terminal and internal inverted repeat sequences TRL-IRL (terminal internal repeat long) and IRS-TRS (terminal internal repeat short) (gray boxes). The hatched boxes represent the genes analyzed by conventional sequencing: DNA polymerase (POL) (UL30 gene), processivity factor of DNA polymerase (PF) (UL42 gene), thymidine kinase (TK) (UL23 gene), glycoprotein L (gL) (UL1 gene), gH (UL22 gene), gB (UL27 gene), gG (US4 gene), and gD (US6 gene). The labeled vertical bars represent microsatellite markers (M followed by position numbers, as previously identified [19]) analyzed by length polymorphism analysis: M20 (TRL), M28 (TRL), M37-38 (inter-UL10-UL11), M43 (inter-UL15-UL18), M50-51 (inter-UL25-UL26), M55 (inter-UL26-UL27), M63-64 (inter-UL37-UL38), M72 (inter-UL50-UL51), and M125 (inter-US4-US5).

FIG 3

Phylogenetic tree based on microsatellite markers, depicting the genetic relationships between variant and classical HSV-2 clinical isolates. The neighbor-joining unrooted tree was constructed from the distance matrix derived from microsatellite allele frequencies reported by Nei et al.(20). The HSV-2 isolates are identified beside the corresponding branches: HSV-2v isolates, marked with an asterisk (isolates 1 to 14), and classical HSV-2c isolates (isolates 15 to 22). The solid bar beside the phylogenetic tree indicates length based on similarity. The haplotype of the gHSV-2 laboratory strain is included.

FIG 4

Genetic analysis of the concatenated sequence alignments of the UL1, UL22, UL23, UL27, UL30, UL42, US4, and US6 genes from human and NHP simplex viruses. Phylogenetic trees were constructed by the maximum likelihood method, implemented in MEGA6 software (22). The bar indicates the number of substitutions per site. Sequences of HSV-2v isolates 1 to 14 correspond to identified HSV-2 variant isolates, marked with an asterisk, and sequences of HSV-2 isolates 15 to 18 and 19 to 22 correspond to HSV-2c isolates recovered from HIV-infected and HIV-negative patients, respectively. The corresponding concatenated nucleotide sequences from HSV-2 reference strains HG52 and SD90e (GenBank accession numbers Z86099 and KF781518, respectively); HSV-1 reference strains 17, F, H129, and KOS (GenBank accession numbers X14112, GU734771, GU734772, and JQ780693, respectively); Old World monkeys (cercopithecine herpesvirus 1, B virus [macaques], cercopithecine herpesvirus 16, HVP2 [baboons], cercopithecine herpesvirus 2, SA-8 [vervets], and chimpanzee alpha-1 herpesvirus [ChHV] [chimpanzee] [GenBank accession numbers AF533768, AY714813, DQ149153, and JQ360576, respectively]); and South American monkeys (saimiriine herpesvirus 1 [HVS-1] [GenBank accession number HM625781]) were also included. Panel B focuses on the HSV-2 and ChHV sequences.

FIG 5

Nucleotide sequence differences for UL30 (DNA polymerase [POL]) (A) and US4 (gG) (B) between classical HSV-2 and the HSV-2 variant in comparison with ChHV UL30 DNA polymerase (3,723 bp) and US4 glycoprotein G (2,100 bp). Nucleotide consensus sequences for HSV-2c (isolates 15 to 22) and HSV-2v (isolates 1 to 14) are compared with the respective gene sequences from ChHV indexed in the GenBank database under accession number JQ360576. For each gene, nucleotide sequence differences observed in HSV-2c (top) or HSV-2v (bottom) and ChHV with respect to the nucleotide sequence of the HG52 reference strain (GenBank accession number Z86099) are indicated by vertical bars. For each panel, a representative set of nucleotide (NT) and amino acid (AA) alignments for HSV-2c and HSV-2v (embodied by HSV-2 isolate 15 and HSV-2v isolate 1, respectively) and sequences from across the region with high similarity to ChHV were included.

FIG 6

Boot-scanning (A) and phylogenetic (B) analyses of evidence for recombination in UL30 gene sequences. (A) The query sequence from HSV-2v (embodied by HSV-2v isolate 1) is shown at the top. The sequences (classical HSV-2, embodied by HSV-2 isolate 15, the HG52 reference strain, and ChHV) compared with the query sequence are indicated on the right. Sequence similarity was calculated by the Kimura two-parameter method with a transition/transversion ratio of 2.0. SimPlot software generates a graph of the percentage of trees (on the y axis) permuted with respect to the nucleotide position in the UL30 sequence (x axis), obtained with a sliding window covering 200 nucleotides, shifted 20 nucleotides at a time. The recombination cutoff value of 70% is indicated by the dashed line. The positions of the putative recombination breakpoints are indicated in red at the bottom. The black line below the boot-scanning plot indicates the region of alignment used to substantiate the recombination signal in the phylogenetic analysis shown in panel B. (B) Phylogenetic tree (500 bootstrap replicates) based on the UL30 segment shown in panel A, encompassing the putative recombination breakpoints (nucleotide positions 1940 to 3370). This tree was constructed to validate the recombination signals obtained by the maximum-likelihood method, implemented in MEGA6 software (22). The bar indicates the number of substitutions per site. Sequences of HSV-2v isolates 1 to 14 correspond to the identified HSV-2 variant isolates, marked with an asterisk, and sequences of HSV-2 isolates 15 to 18 and 19 to 22 correspond to HSV-2c isolates recovered from HIV-infected and HIV-negative patients, respectively. The corresponding nucleotide sequences for HSV-2 reference strains HG52 and SD90e (GenBank accession numbers Z86099 and KF781518, respectively); HSV-1 reference strains 17, F, H129, and KOS (GenBank accession numbers X14112, GU734771, GU734772, and JQ780693, respectively); Old World monkeys (cercopithecine herpesvirus 1, B virus [macaques], cercopithecine herpesvirus 16, HVP2 [baboons], cercopithecine herpesvirus 2, SA-8 [vervets], and ChHV [chimpanzees] [GenBank accession numbers AF533768, AY714813, DQ149153, and JQ360576, respectively]); and South American monkeys (HVS-1 [GenBank accession number HM625781]) are included.