Seroprevalence and molecular characteristics of varicella-zoster virus infection in Chinese children

Between August 2017 and September 2017, 3014 kindergarten students aged 4–6 years in Suzhou Industrial Park (Jiangsu, China) were invited to participate in the present study when they received an annual physical examination. Written informed consents were obtained from the parents of children after declaring the aims of the present study. A blood sample (5 mL) was taken from 3003 children because 11 children did not agree to provide their blood samples. The samples were immediately spun down and the serum was stored at − 40 °C for later analysis. Varicella vaccination information including the date of administration and trademark of vaccine was retrieved from the Immunization Registry System or the detailed vaccination records. The study protocol was approved by the ethics committees of Suzhou Centers for Disease Control and Prevention in accordance with the ethical guidelines of the 1975 Declaration of Helsinki.

The serum samples obtained from all children were tested for anti-VZV IgG and IgM antibody by enzyme-linked immunosorbent assay (ELISA) with the commercial kits of “Diagnostic Kit for IgG Antibody to Varicella-Zooster Virus” and“Diagnostic Kit for IgM Antibody to Varicella-Zooster Virus” (Beier Biotechnology, Beijing, China), respectively. The cut-off value was calculated for each plate according to the manufacturer’s instructions. When an ambiguous result was found, the sample was retested and confirmed as positive only if one of the repeats (2/3 of the total tests) was positive. According to the manufacture’s introductions, both sensitivity and specificity of IgG method and IgM method were more than 98%, and the CV% of two methods were less than 15%.

VZV DNA was extracted by Viral RNA/DNA Extraction Kit Version 5.0 (TaKaRa Biotechnology, Dalian, China) from 300 μL of serum. According to the previous reports [20, 21], a 359-bp region of the ORF22 gene (nt37870–nt38228, reference strain Dumas, XO4370) and a 419-bp region of ORF62 gene (nt106983–nt107401, reference strain Dumas, XO4370) were amplified by polymerase chain reaction (PCR). The primers used for ORF22 were ORF 22F: 5′-TAGCATGTCTGGAGGCAATGG and 22R: 5′-GGCCTTGGAAACCACATGATCG; and the primers used for ORF62 were ORF 62F: 5′- GGCCTTGGAAACCACATGATCG and 62R: 5′- CGTCTCCCGTTCCGCATGTAG. The PCR products were purified and sequenced using the ABI 3730 (Applied Bio systems, Foster City, USA). The genotype of isolated VZV strain was identified using phylogenetic analysis by the neighbor-joining algorithm in MEGA version 4.1 software. The reliability of the phylogenetic analysis was tested by bootstrap analysis with 1,000 replicates. VZV reference wild-type sequences of genotype E strains included Dumas, BC (AY548171) and SD (DQ479953), genotypes E2 strains included HJO (AJ871403) and 11 (DQ479955), genotype M1 strain CA123 (DQ457052), genotype M2 strains included DR (DQ452050) and 8 (DQ479960), and genotype J strains included pOka (AB097933), pOka-derived vaccine strain (vOka; AB097932), VariVax (DQ008355) and Varilrix (DQ008354) were used to identify nucleotide variability.

Statistical analysis was performed using SAS version 9.2 software (SAS Institute Inc., Cary, USA). The seroprevalence of anti-VZV IgG antibody between different groups were compared using the χ² test or trends χ² test. All tests were two sided, and differences were considered statistically significant if P < 0 05.

Among 3014 children, 701 boys (43.9%) and 656 girls (46.3%) were vaccinated with Vari-L varicella vaccine (Changchun Institute of Biology Products, Changchun, China), respectively, and only 21 children received the second vaccine dose. The mean ages of the first and the second dose of vaccine were 5.20 ± 0.83 and 5.71 ± 0.46 years, respectively. A total of 33.7, 45.6 and 53.6% of children were vaccinated in the 4, 5 and 6-year groups, respectively (Trend χ² = 81.678, P = 0.000, Table 1). Among of them, 137 (10.1%), 699 (51.5%), 331 (24.4%), 118 (8.7%), 29 (2.1%), 26 (1.9%) and 17 (1.3%) children were administered with the first dose of vaccine at < 1, 1–2, 2–3, 3–4, 4–5, 5–6 and ≥ 6 years of age, respectively (Fig.1).

In the present study, ELISA was performed on the serum samples of the 3003 participants (1592 boys and 1411 girls). As the results shown, 837 boys (52.6%) and 710 girls (50.3%) were positive for anti-VZV IgG, respectively (χ² = 1.776, P = 0.411). The seropositivity of VZV IgG increased with age, which was detected in 47.6, 49.2 and 55.5% of the children in 4, 5 and 6-year groups, respectively (Trend χ² = 27.071, P = 0.000).

Among of 1357 children with varicella vaccination, the seroprevalence of anti-VZV IgG was 54.4%, which was significantly higher than those in unvaccinated children (49.2%) (χ² = 8.206, P = 0.004. Table 2). Additionally, among of the vaccinated children, the detection rates of VZV IgG antibody increased with age, with 49.4, 50.9 and 58.9% in 4, 5 and 6-year groups, respectively (Trend χ² = 17.202, P = 0.002). However, there was no difference in anti-VZV IgG detection rates among unvaccinated children, with 46.7, 49.4 and 51.7% in 4 years, 5 years and 6-year groups, respectively (Trend χ² = 8.681, P = 0.070) (Table 2).

The results from subgroup analysis showed that among vaccinated children, the seroprevalence of anti-VZV IgG in 5 years old boys was significantly higher than those in 5 years old girls (55.0% vs 48.1%, χ² = 6.823, P = 0.033). However, no difference in VZV IgG antibody seropositivity was detected between unvaccinated boys and girls with 5 years of age. Additionally, no significant difference in anti-VZV IgG detection rates was detected between boys and girls in 4- and 6-year groups regardless of vaccination status (Table 3).

Twenty-six children (13 boys and 13 girls) were positive for anti-VZV IgM. Among of them, eight childrenhave received varicella vaccination, which was similar to those in unvaccinated children (0.6% vs 1.1%, χ² = 2.153, P = 0.142). Additionally, the seropositivity of anti-VZV IgM was 0.5, 1.1 and 1.0% in 4, 5 and 6-year groups. There was no significant difference in the detection rates of VZV IgM antibody among three groups (Trend χ² = 1.625, P = 0.444).

Twenty-six anti-VZV-IgM-positive samples were further analyzed to determine the existence of VZV DNA by amplification of 359-bp ORF22 sequence and 419-bp ORF62 sequence, respectively. However, only one ORF22 sequence was obtained from an unvaccinated 5-year boy. Compared to the reference VZV sequences, the nucleotide homology was estimated to be 99.7% with genotype J, but only 98.6, 98.6, 98.9 and 99.4% similar to the genotype E, E2, M1 and M2 VZV strains, respectively. Therefore, this VZV isolate, named Soochow_VZV (GenBank no. MK047349) was classified as genotype J (Fig. 2), which was widely distributed in eastern China.