Identification of human parvovirus 4 genotypes 1 and 2 in Chinese source plasma pools

Abstract Human parvovirus B19 (B19V) and human parvovirus 4 (PARV4) are known to infect humans and transmit through contaminated blood and blood products. Globally, three genotypes of B19V, as well as PARV4, have been identified, respectively. The existence of different B19V genotypes in Chinese plasma donors has been investigated, however, the data regarding PARV4 were not available. The main objective of this study is to identify the genotypes of PARV4 circulating in Chinese plasma donors. By using a duplex quantitative polymerase chain reaction assay adapted for all genotypes of B19V and PARV4, 78 source plasma pools for fractionation were screened and quantified. Results showed that positive rates of B19V and PARV4 DNA in plasma pool samples were 25.64% and 14.10%, respectively. PARV4 sequences in two positive samples were next genotyped, and these two sequences belonged to PARV4 genotypes 1 and 2, respectively. In conclusion, the data present demonstrate the existence of PARV4 genotypes 1 and 2 in Chinese plasma donors for the first time and also show the relatively lower prevalence and level of PARV4 DNA in Chinese plasma donors in comparison with that of B19V DNA.

plasma of acutely infected but asymptomatic donors, thereby resulting in the highly contaminated source plasma pools for fractionation, (2) it is highly resistant to virus inactivation or removal methods used in the manufacture of blood products, and (3) sometimes it can cause severe diseases in at-risk recipients. 2 To mitigate the risk of B19V transmission, most manufacturers have been performing nucleic acid testing (NAT) for B19V DNA as an inprocess test for source plasma pools used for manufacturing certain or all kinds of blood products to limit the virus load, according to the guidance or standard from European Pharmacopoeia, the Plasma Protein Therapeutics Association, and U.S. Food and Drug Administration. [3][4][5] Similar to B19V, PARV4 is also a frequent contaminant of source plasma pools for the production of blood products and the final products. 6 Baylis et al. 7 demonstrated PARV4 was even more resistant than B19V to virus inactivation strategies used during the manufacture of blood products, such as pasteurization and low-pH treatment. The transmission of PARV4 by virally inactivated clotting factor concentrates raised concerns among the patients with hemophilia or other recipients of such products. 6 14 recently reported a strong association of PARV4 with severe respiratory illness. Same as B19V, PARV4 has been classified into three genotypes. 15 All these three genotypes have been detected in human blood or blood products although they seem to have different epidemiology. Genotypes 1 and 2 are prevalent in North America, Europe, and some countries in Asia while genotype 3 seems to be endemic in Ghana. 16 However, data on the existence of different PARV4 genotypes circulating in Chinese plasma donors are limited.
In our previous article, the prevalence of PARV4 in Chinese plasma pools has been reported, but the genotypes of PARV4 were not identified. 17 The main objective of this study is to identify the genotypes of PARV4 circulating in Chinese plasma donors.

| B19V and PARV4 quantitative polymerase chain reaction assay
The DNA samples were initially tested for presence and quantities of B19V and PARV4 DNA using a duplex quantitative polymerase chain reaction (qPCR) assay, which has been proved to be able to simultaneously detect and quantify all the known genotypes of B19V and PARV4, with an equal limit of quantification of 5 copies/ml. 19 Serial log 10 dilutions of the B19V and PARV4 standard plasmids containing the qPCR target sequences and the samples were all analyzed in triplicate. On the basis of the standard curve generated by the standard plasmids and the quantification cycle value of each sample, the concentration of each virus DNA in copies/ml was calculated.

| PARV4 nested PCR and sequencing
A 161-bp region related to the NS1 gene of PARV4 (positions 1564-1724 in AY622943) was amplified by nested PCR with the primers described previously. 20 PCR amplification products derived from each sample were purified and cloned into the pMD18-T vector (TaKaRa Bio) and subsequently sequenced on an ABI 3730XL DNA Analyzer.

| Phylogenetic analysis of PARV4 sequences
The PARV4 genotypes were determined by phylogenetic tree ana-

| Quantity of B19V and PARV4 DNA in source plasma pools
The levels of B19V and PARV4 DNA in source plasma pools were shown in Figure 1. The quantity of B19V DNA varied from 2.56 × 10 2 to 2.30 × 10 9 copies/ml plasma. Levels of B19V DNA was as high as 2.30 × 10 9 copies/ml plasma, although 70% of the positive samples were at low levels (10 2 -10 4 copies/ml plasma). For PARV4, the level of virus DNA was lower than that of B19V: viral loads ranged from 2.42 × 10 2 to 2.28 × 10 5 copies/ml plasma, and most samples contained 10 2 -10 3 copies/ml plasma. In the one coexistence sample, levels of B19V and PARV4 were low, equal to 1.31 × 10 3 copies/ml plasma and 2.42 × 10 2 copies/ml plasma, respectively.

| Genotypes of PARV4 sequences in source plasma pools
The genotypes of PARV4 sequences in source plasma pool samples were identified. Of the 11 samples found to be positive for PARV4 sequences by qPCR (Table 1), only 2 were positive for PARV4 nested PCR and available for further analysis. The PARV4 DNA quantity of these two samples was 5.28 × 10 2 and 9.75 × 10 3 copies/ml plasma, respectively. While no targeted PCR products were amplified from five samples containing greater than 5.28 × 10 2 copies/ml plasma of PARV4 DNA. These unexpected results might attribute to the incomplete sequence of PARV4 genome in these samples, in view of the different target regions between PARV4 qPCR and nested PCR.
Phylogenetic analysis of 2 sequences obtained in this study, together with 24 sequences retrieved from GenBank, revealed that one sequence (pool-7) clustering together with genotype-1 reference sequences belonged to PARV4 genotype 1, and the other one sequence (pool-14) clustering together with genotype-2 reference sequences belonged to PARV4 genotype 2 ( Figure 2).

| DISCUSSION
Since discovered in 2005, PARV4 has received much attention. 22 Unlike B19V, which took nearly 30 years after its initial discovery in 1975 to identify its three genotypes, PARV4 genotypes 2 and 3 were identified within 3 years. 15 21 In this study, the PARV4 sequences in two PARV4-DNA positive source plasma pool samples were genotyped and the results demonstrated that at least two PARV4 genotypes, 1 and 2, were currently present in China. No genotype 3 was detected, which was not unexpected since this genotype was reported to be endemic in Ghana. 16 Besides, the possibility that genotype 3 might be present at an extremely low level in the sample, and, therefore, escaped identification can not be completely ruled out. Furthermore, it should be noted that this study was restricted to one region of China, and, therefore, could not reflect the whole national circulating status of different PARV4 genotypes. B19V genotype 3 is originally endemic to Ghana as well, but as research continues it shows a wider distribution and has been found in Brazil, France, North India, the United States, and China. 21,28 Whether PARV4 genotype 3 will show an increasing spread also outside of Ghana, just like B19V, needs further research with large sample size and a wide geographical area.
These results also showed that the positive rates and levels of B19V DNA in Chinese plasma pool samples were relatively higher than that of PARV4 DNA, indicating that the prevalence of B19V DNA in the Chinese population might be higher than that of PARV4, consistent with results detected in the general population from a previous report. 29 On the other hand, regarding the frequent coinfection with hepatitis C virus and HIV, some PARV4 positive plasma samples might be excluded by infectious agents screening tests before pooling. 6 The prevalence of B19V DNA in source plasma pool samples tested in this study was lower than that reported in 2015 (104/141, 73.76%), whose F I G U R E 2 Phylogenetic analysis of human parvovirus 4 (PARV4) nucleotide sequences. The phylogenetic tree was constructed based on the 161-nt NS1 region of PARV4 and the neighbor-joining algorithm using the Kimura two-parameter model. Two PARV4 sequences from this study (labeled with black circles) and a set of PARV4 sequences downloaded from GenBank (labeled with their GenBank accession number and isolate or strain name) used as references for the different genotypes were analyzed. Bootstrap replication frequencies are indicated above each node. Branch lengths are drawn to scale samples derived from the same manufacture were collected between 2008 and 2013. 30  Given that associated disease or specific symptoms have not yet been confirmed, there is no need to exclude PARV4 from source plasma pools, at least for the time being. 6 It should be noted that although the use of source plasma pools as materials in this study increases the potential for the detection of PARV4, the data regarding PARV4 containing plasma pools were still relatively limited. In further research, screen and sequence more source plasma pool samples by extracting larger volumes of plasma or concentrating virions by immune adsorption could address this limitation.

| CONCLUSION
In conclusion, the data present demonstrate the existence of PARV4 genotype 1 and 2 in Chinese plasma donors for the first time and also show the relatively lower prevalence and level of PARV4 DNA in Chinese plasma donors in comparison with that of B19V DNA.