Introduction

As a member of the family Parvoviridae, human parvovirus B19 (B19V) was a global and common infectious pathogen in humans1. B19V was first identified by Cossart et al.2 during the evaluation of tests for hepatitis B virus surface antigen in 1975. Subsequently, B19V infection has been reported worldwide. The transmission of B19V infection usually occurs through the respiratory route, and infections can also be transmitted vertically from mother to fetus and through the transfusion of blood products and bone marrow transplants3,4,5. B19V infection is common in childhood, and it can also occur throughout adulthood albeit at a lower rate6. The findings showed a significant negative correlation between viremia rates and age, and the positive rate of B19V DNA decreased with age, from 2.24% in 19–30 years to 0.87% in 41–50 years7,8. However, the prevalence of IgG antibodies directed against B19V ranges from 2 to 15% in children 1 to 5 years old, 15 to 60% in children 6 to 19 years old, 30 to 60% in adults, and more than 85% in the geriatric population9,10,11,12. The manifestations of B19V infection depend on the age, immunity and hematologic status of the host1,13. Most B19V infections are asymptomatic. The common clinical manifestation of B19V infection is erythema infectiosum in children, while arthropathy is a more common manifestation of infection in adults, particularly in women14. In immunocompromised hosts, persistent B19V infection presents with pure red cell aplasia and chronic anemia13,15. B19V is now recognized as the only etiologic agent of erythema infectiosum16,17,18. Due to similar eruption symptoms, erythema infectiosum can be confused with rubella19,20,21,22.

B19V is a small nonenveloped single-stranded DNA virus with an approximately 5.5-kb long genome. This DNA encodes six viral proteins, among which the three major proteins are nonstructural protein 1 (NS1), viral protein 1 (VP1) and viral protein 2 (VP2). In addition, there are three small nonstructural proteins of 7.5 kDa, X, and 11 kDa. Based on the phylogenetic analysis of the NS1-VP1u region, B19V was classified into three genotypes: genotype 1, genotype 2, and genotype 323. With the increasing number of genome sequences obtained, B19V was further subdivided into subgenotypes, which were useful for molecular epidemiological studies. Genotype 1 was further divided into two subgenotypes, 1a and 1b, and genotype 3 was divided into two subgenotypes, 3a and 3b24,25.

To date, investigations of B19V have been carried out by many researchers in recent years, including in China. However, most of them focused on viral infection in blood products from healthy donors, organ transplantation and the status of B19V infection in pregnant women26,27,28,29,30. There have been few studies on the prevalence of B19V in patients with febrile rash illnesses (FRIs). From January 2009 to June 2021, active surveillance of FRIs was conducted in ten provinces of China, including Anhui, Bei**g, Hebei, Henan, Hunan, Shandong, Shanxi, Shaanxi, Shanghai and **-stone sampling (SS). The best fit models for the genome and six proteins of B19V were determined by Bayesian factor analysis. The Markov Chain Monte Carlo method was performed for 50 million generations and sampled so that 1,000 trees were generated. Finally, the convergence of the chains and the effective sample size (> 200) were determined by Tracer software (version 1.7.1), and the uncertainty of the parameter estimates was assessed by a 95% HPD interval.

Selection pressure was also analyzed in six proteins of B19V based on the genome dataset. DnaSP6 software (Version 6.0) was used to calculate the ω values (ω = dN/dS) of B19V, where dN represents the nonsynonymous substitution rate and dS represents the synonymous substitution rate. Gene-specific and site-specific selection pressures were measured as the value dN/dS at each codon site and estimated using four different codon-based maximum-likelihood methods (FEL, SLAC, FUBAR and MEME), and Tamura-Nei model (TrN) or Hasegawa-Kishino-Yano model (HKY85) were used as nucleotide substitution models50. All methods were obtained available at the DataMonkey online version of the Hyphy package (www.datamonkey.org) with significance levels set at p < 0.05, and the posterior probability of the FUBAR algorithm was > 0.95.

Accession number

Seven genome sequences and three NS1-VP1u sequences of B19V in this study were submitted to GenBank with accession numbers OR533486 ~ OR533495 (Supplementary Table S1).

Ethics approval and consent to participate

This study was approved by the ethical review committee of the National Institute for Viral Disease Control and Prevention, Chinese Centers for Disease Control and Prevention. The informed consents were signed by patients or their legal guardians.