Genomic sequencing revealed recombination event between clade 1 and clade 2 occurs in circulating varicella-zoster virus in China

Abstract

Varicella-zoster virus (VZV), a member of the Alphaherpesvirinae subfamily, causes varicella in primary infections and establishing a latent stage in sensory ganglia. Upon reactivation, VZV causes herpes zoster with severe neuralgia, especially in elderly patients. The mutation rate for VZV is comparatively lower than the other members of other alpha herpesviruses. Due to geographic isolation, different genotypes of VZV are circulating on separate continents. Here, we successfully isolated a VZV from the vesicular fluid of a youth zoster patient. Based on the single-nucleotide polymorphism profiles of different open reading frames that define the genotype, this newly isolated VZV primarily represents genotype clade 2 but also has characteristics of genotype clade 1. The next-generation sequencing provided a nearly full-length sequence, and further phylogenetic analysis revealed that this VZV isolate is distinct from clades 1 and 2. The Recombination Detection Program indicates that a possible recombinant event may occur between the VZV isolate and clade 1. In summary, we found that there is a circulating VZV isolate in China that may represent a recombinant between clade 1 and clade 2, providing new concerns that need to be considered in the future VZV vaccination program.

Maintext

Varicella-zoster virus (VZV, hereafter) or the Varicellovirus humanalpha3 (HuAHV3) following the newly released International Committee on Taxonomy of Viruses (ICTV) report is a human herpesvirus responsible for two diseases: chickenpox (varicella), for primary infection and establishing a lifelong latent state in the sensory nerve ganglia; and shingles (zoster), which is caused by virus reactivation upon declining immunity [1]. Although usually self-limited, severe complications can occur, including pneumonia, encephalitis, and vasculopathy. VZV is a member of the genus Varicellovirus of the Alphaherpesvirinae subfamily (family Orthoherpesviridae). Its genome is linear, double-stranded DNA, approximately 125-kb long, which encodes at least 71 open reading frames (ORFs). Though the VZV genomes are considered the most stable among all alpha herpesviruses, the recombination between two or more virus strains can occur when the same cell is infected by two or more viruses [2, 3]. In 2010, Breuer et al. proposed a novel nomenclature for VZV containing five clades (1–5) and two putative clades (VI, VII) based on 27 single-nucleotide polymorphisms (SNPs) in place of the previously proposed SNP clade clarification [4]. A previous study showed that clades 1–5 were geographically segregated [5, 6], as clade 2 was predominant in China and neighboring countries. Moreover, a live-attenuated strain (vOka) was licensed in 1998 in China, which derived from the pOka strain, and both vOka and pOka are characterized in clade 2. In recent years, the SNP clarification of VZV isolates from all over the country has been determined by different groups [7–11]. Along with the advancement of next-generation sequencing (NGS), more VZV whole genome data were obtained, and thus the possible recombination events between different VZV strains have been indicated [5, 6, 11]. However, the whole genome of VZV is comparably limited, especially in China [12].

In August 2018, an undergraduate within the university developed herpes zoster. We collected the vesicular fluid by piercing the vesicle blister and inoculated it onto the African green monkey kidney Vero E6 cells (ATCC, CRL-1586). Cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM, HyClone) with 2% fetal bovine serum (FBS, Gibco) after 1 h adsorption and cultured at 37 °C with 5% CO₂. The cytopathic effects (CPEs) were observed and recorded daily. The initial CPE appeared 7 days post-inoculation; cells were rounded and detached from the bottom (Fig. 1A, Lower left); however, the CPE-positive area was relatively limited. Until 14 days post-inoculation, the CPE progressed significantly (Fig. 1A, Lower right). To confirm that the CPE of Vero E6 cells was caused by the propagated VZV, the CPE-positive cell was trypsin digested and seeded onto the coverslip along with the normal cell and observed using a BX60 fluorescence microscope (Olympus, Tokyo, Japan) following a standard immunofluorescence assay (IFA) protocol [13]. The VZV gE-specific antibody (Santa Cruz Biotechnology) detected obvious gE expression in the cytoplasmic area of vesicular fluid-inoculated cells other than the control group (Fig. 1B). Further, inoculated cells and control cells were lysed with Radio Immunoprecipitation Assay (RIPA) lysis buffer (Beyotime) to identify VZV replication within cells. A 20-µg aliquot of each cell lysate was electrophoresed and transferred to a poly vinylidene fluoride (PVDF) membrane (Millipore). PVDF membranes were incubated with the primary antibodies against VZV gE and glyceraldehyde-3-phosphate dehydrogenase (GAPDH, Proteintech), followed by secondary antibodies labeled with infrared dye (LI-COR Biosciences) and visualized using the Odyssey Infrared Imaging System. A typical prominent and broad gE complex pattern was observed (Fig. 1C), indicating the VZV replication was established. Aside from these immunological methods, the existence of infectious virions was checked by electron microscopy (EM) observation. Since VZV is mainly transmitted through the direct cell-to-cell route, the number of virions secreted into the culture medium is relatively low. This may be due to the retrograde assortment of VZV by insulin-like growth factor 2 receptor (IGF2R)-mediated virus re-entry, and the virion was transported into late endosomes for degradation [14]. Thus, the VZV-propagated cells were trypsinized and collected through centrifugation and performed EM observation directly. After washing twice with DPBS, the cell pellet was resuspended and fixed with 2.5% glutaraldehyde solution overnight at 4 °C. Then, 1% osmium tetraoxide fixation was performed for 3 h. The samples were dehydrated with ethanol and then embedded in EPON resin. Thin grids were stained with lead citrate and uranyl acetate and examined by transmission EM as previously indicated [15]. As shown in Fig. 1D, capsid virions were present in the nucleus, and enveloped virions existed within the cytoplasm. A consecutive life stage of VZV propagation could be observed from versatile virions present in the different subcellular organelles. In short, we successfully isolated a VZV isolate from the vesicular fluid sample and named it AMU01.

Fig. 1.

Characterization of the isolated VZV. A Bright-field microscope observation of VZV or mock-inoculated Vero E6 cells. Upper left, mock-inoculated Vero E6 cells (100×). Lower left, mock-inoculated Vero E6 cells (200×). Upper right,VZV-inoculated Vero E6 cells, 7 days post infection (100×). Lower right,VZV-inoculated Vero E6 cells, 14 days post infection (200×). B Immunofluorescence detection of gE in VZV-infected Vero E6 cells. gE was stained with FITC-conjugated antibody (green), and nuclei were labeled with Hoechst 33258 (blue). C Western blot of gE protein expression in VZV-infected Vero E6 cells. D Electron microscope examination of VZV-infected Vero E6 cells. Numbers 1, 2, 3, and 4 represent different life stages of the VZV lifecycle

Though the genome of VZV is thought to be stable among different isolates across the world, the different nucleoid acid distribution at certain positions aroused the different genotypes for VZV characterization [16, 17]. In 2008, a common nomenclature for VZV viral clades was proposed [4]. This system characterized the VZV with five genotypes (and two putative types) from different areas according to SNPs of distinct ORFs. Most domestic studies were based on restriction fragment length polymorphism (RFLP) analysis [18] or a previously designated genotyping system along with a limited SNP profile [8, 10, 19]. Different geographical distribution patterns were also noted in these studies, and previous studies proposed that clade 2 is the most prevalent VZV in temperate regions in East Asia [10, 18]. However, other groups also detected different clades in China [8, 16]. To determine the genotype of the VZV we isolated, we first performed genotyping PCR. The VZV genomic DNA from the culture supernatant was extracted by TIANamp Virus DNA/RNA Kit (TIANGEN Biotech) according to the manufacturer’s instructions. The corresponding VZV ORF SNP primers were synthesized at TSINGKE Biological Technology. The VZV genes in ORF 1, 6, 12, 16, 17, 21, 22, 35, 37, 50, 54, 55, 56, 60, and 66 were amplified by polymerase chain reaction (PCR) with Q5 High-Fidelity DNA Polymerase (New England Biolabs) using corresponding oligonucleotide primers [7, 18]. Briefly, 25 µl of PCR product was loaded into agarose gels, electrophoresed, and the targeted DNA bands were cut and sent for sequencing. The sequences were aligned with the Dumas strain genome (accession number X04370) and pOka (AB097933), and the SNP site was recorded and assigned to the corresponding genotypes (Fig. 2). Our isolate AMU01 could be assigned to clade 2 (reference strain AB097933) by the SNPs in ORFs 31, 37, 60, 62, 67, and 68 (Fig. 2) and also fulfill the genotype J due to the SNPs of ORF22 (Fig. 2). In addition, we observed some inconsistencies between the pOka strain and AMU01, an 18 082 site-specific SNP (ORF12) representing the pOka strain but with base C at site 18 082 other than T. This phenomenon has also been observed in 5 of 23 VZV isolates collected in the suburban area of Shanghai, China, but only 1 case bears this mutation [7]. Though the data are limited, considering the reported cases annually, we think this is not an isolated case, and many more cases bearing this mutation remain to be detected. And this unique position mutation trait may represent a new branch of the clade 2 genotype.

Fig. 2.

Genomic variations of the VZV isolate. Sequence positions are based on the published genomic sequence of the Dumas strain (GenBank accession no. X04370). Clade 2 markers are in green, and unique single-nucleotide mutations markers are in blue

Although SNP profiling has advanced recently in China, the information regarding the whole genome sequence is relatively limited [12]. To further characterize the feature of AMU01, we performed HiSeq sequencing to obtain the whole genomic sequence of AMU01 (Shanghai BioGerm Medical Biotechnology); the sequencing reads were assembled with SPAdes 3.9, and the sequence was deposited in the GenBank (MK531557). Since AMU01 represents traits of both clades 1 and 2, to investigate whether this phenomenon is caused by recombination or convergent evolution, we first performed the phylogenetic network analysis using the program SplitsTree5 [20]. By comparing the complete genomes of the representative VZV strains, the phylogenetic incongruences between AMU01 and clade 1 or clade 2 were detected (Fig. 3A). Recombination Detection Program (RDP) 5.23 was adopted to investigate whether there is interclade recombination between clade 1 and clade 2 for AMU01 [21]. And a unique recombination event was suggested between AMU01 and clade 1 (Dumas) by distinct methods with statistical support (Fig. 3B). The location of the recombination breakpoint within the VZV genome is at site between 105 946 and 105 947 bp according to the RDP. The potential major parent is clade 2 VarilRix strain (DQ008354.1), and the potential minor parent is Dumas strain. Hierarchical clustering of regions derived from major parent and minor parent using UPGMA (unweighted pair group method with arithmetic mean) was more closely related to clade 2 and clade 1, respectively (Fig. 4A and B). Considering the breakpoint located within the transcriptional regulator ICP4 encoding ORF62, which has two copies in the genome (105 141 ~ 109 133 HHV3 gp63 and 120 764 ~ 124 756, HHV3 gp72). We selected the first ORF62 conducted phylogenetic analysis and found that ORF62 of AMU01 was located between clade 1 and clade 3 (Fig. 4C), suggesting that ORF62 was more versatile between different VZV strains. Previous reports indicated that ORF62 exhibited the highest frequency of mutations [22], which possibly provided a high chance for recombination. Currently, certified vaccine vOka strain and most of the circulating strains in China were characterized in clade 2. Thus, the origin of other VZV clades that participated in the recombination suggested here is intriguing, and similar recombination or clade 2 strains recombinant with other clades may be detected in the future. Although there is a possibility that the VZV isolate AMU01 adopted this kind of feature through propagation in Vero E6 cells, as suggested in a previous study, prolonged cell culture may confer VZV isolates the feature of vaccine-type mutations (vOka strain) [22]. However, the passages in our system are quite limited, and the original vesicular fluid sample is no longer remained, which hinders the primary VZV genome sequencing.

Fig. 3.

Phylogenetic and interclade recombination analysis of reference VZV strains and isolate AMU01. A Phylogenetic network (SplitsTree5) based on the consensus sequence of selected VZV strains. Clade 1 (Dumas, X04370.1; MSP, AY548170.1; SD, DQ479953; Kel, DQ479954; 32 passage 5, DQ479961; NH29_3, DQ674250) is marked in red, clade 2 (pOka, AB097933; vOka, AB097932; VarilRix, DQ008354; VariVax, DQ008355; YC02, KJ767492) is marked in light green, clade 3 (03–500, DQ479957; 11, DQ479955.; 22, DQ479956; HJ0, AJ871403) is marked in purple, clade 4 (8, DQ479960) is marked in yellow, and clade 5 (CA123, DQ457052) is marked in dark green. B Recombinant clades suggested by the RDP5.23 and statistical support yielded by each method for respective recombination event. C Proposed exchange of the 105 947 to 126 488 region of AMU01 between clade 1 strains with clade 2 sequences

Fig. 4.

Phylogenetic analysis of subgenomic regions. A Phylogenetic analysis of the 1 to 105 946 region of AMU01. UPGMA of region derived from major parent (clade 2). B Phylogenetic analysis of the 105 947 to 126 488 region of AMU01. UPGMA of region derived from minor parent (clade 1). C Overview of maximum-likelihood trees computed with MEGAX of the ORF62 region of AMU01 (105 141 to 109 133). The diagrams are not drawn to scale. Numbers at major nodes indicate bootstrap values obtained after 1000 replications

To conclude, the AMU01 represents the SNPs similar to clade 2 and shares some SNPs’ features of clade 1. Different methods suggested there is a recombination event between clades 1 and 2, indicating that the current SNPs feature may not be sufficient to characterize the circulating VZV strains in China and thus calls for more whole-genome data to estimate the VZV circulating status through molecular epidemiology analysis. This has and provided new concerns that should be considered in the future VZV vaccination program.

Author contribution

LJ performed sample collection. WD performed the sample inoculation and cell propagation. WD and LJ performed most of the experiments. QLB, YYW, DYC, WY, and YM performed the PCR and SNP analysis. YW performed the genome sequencing with the phylogenetic analysis. LYL, LYF, and ZFL provided administrative support. WY, CLF, and YW offered financial support. YW and CLF conceived of the study. YW wrote the manuscript. LYF, ZFL, CLF, and YW checked and finalized the manuscript.

Funding

The present study was supported in part by grants from the National Natural Science Foundation of China (no. 31600131), the Key Laboratory Open Project Fund (no. SKLPBS1834), and university supporting grants (no. 2018JSTS03, 2020SWAQ09, and 2021JSTS10). The funding body had no role in this work’s design, interpretation, or submission for publication.

Data availability

The original contributions presented in the study are included in the article; the VZV genome sequence was deposited at GenBank (MK531557). Further inquiries can be directed to the corresponding authors.

Declarations

Ethics approval and consent to participate

The study was approved by the ethics committee of First Affiliated Hospital, FMMU (approval no. KY20173167-1). The participant provided written consent for the anonymous use of his specimens and clinical information for research.

Conflict of interest

The authors declare no competing interests.