Journal of Virological Methods 195 (2014) 106–111
Contents lists available at ScienceDirect
Journal of Virological Methods journal homepage: www.elsevier.com/locate/jviromet
A two-step real-time PCR assay for quantitation and genotyping of human parvovirus 4 E. Väisänen a,∗ , A. Lahtinen a , A.M. Eis-Hübinger b , M. Lappalainen c , K. Hedman a,c , M. Söderlund-Venermo a a
Haartman Institute, Department of Virology, University of Helsinki, Helsinki, Finland University of Bonn Medical Centre, Bonn, Germany c Helsinki University Hospital Laboratory Division, Helsinki, Finland b
a b s t r a c t Article history: Received 5 June 2013 Received in revised form 26 September 2013 Accepted 1 October 2013 Available online 14 October 2013 Keywords: Real-time PCR Parvoviridae PARV4 Genotyping Quantification
Human parvovirus 4 (PARV4) of the family Parvoviridae was discovered in a plasma sample of a patient with an undiagnosed acute infection in 2005. Currently, three PARV4 genotypes have been identified, however, with an unknown clinical significance. Interestingly, these genotypes seem to differ in epidemiology. In Northern Europe, USA and Asia, genotypes 1 and 2 have been found to occur mainly in persons with a history of injecting drug use or other parenteral exposure. In contrast, genotype 3 appears to be endemic in sub-Saharan Africa, where it infects children and adults without such risk behaviour. In this study, a novel straightforward and cost-efficient molecular assay for both quantitation and genotyping of PARV4 DNA was developed. The two-step method first applies a single-probe pan-PARV4 qPCR for screening and quantitation of this relatively rare virus, and subsequently, only the positive samples undergo a real-time PCR-based multi-probe genotyping. The new qPCR-GT method is highly sensitive and specific regardless of the genotype, and thus being suitable for studying the clinical impact and occurrence of the different PARV4 genotypes. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Human parvovirus 4 (PARV4) is a small single-stranded DNA virus in the Parvoviridae family. It is one of the four parvoviruses known to infect humans alongside the human parvovirus B19, adeno-associated viruses and human bocaviruses. The first PARV4 sequences were identified in 2005 in a plasma sample from an intravenous drug user presenting with various symptoms of acute viral infection (Jones et al., 2005). A year later, a related virus, initially called PARV5, was found in plasma pools for manufacture of plasma-derived medical products and characterised as a PARV4 variant, genotype 2 (Fryer et al., 2006, 2007a). A third genotype was isolated in 2008 from two adult AIDS patients from Africa (Simmonds et al., 2008). Furthermore, PARV4-like viruses have been discovered in various animals, for example, in chimpanzees and other primates, bats, cows, sheep, pigs and wild boars (Lau et al., 2008; Adlhoch et al., 2010; Sharp et al., 2010a; Canuti et al., 2011; Tse et al., 2011). PARV4 infection is associated commonly with intravenous drug use and with seropositivity of human immunodeficiency virus (HIV), hepatitis B virus (HBV) or hepatitis C virus (HCV) in
∗ Corresponding author. Tel.: +358 9 191 26676. E-mail address: elina.vaisanen@helsinki.fi (E. Väisänen). 0166-0934/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jviromet.2013.10.011
Northern Europe, the United States and Asia (Manning et al., 2007; Simmonds et al., 2007; Lurcharchaiwong et al., 2008; Sharp et al., 2009; Lahtinen et al., 2011; Yang et al., 2011; Simmons et al., 2012; Yu et al., 2012; Maple et al., 2013). The virus has been detected in blood products (Fryer et al., 2006, 2007a,b; Lurcharchaiwong et al., 2008; Touinssi et al., 2010; Ma et al., 2012) and in persons undergoing repetitive blood and blood-product transfusions; i.e. haemophilia patients are at greater risk of infection (Simmonds et al., 2007; Sharp et al., 2009, 2012). Therefore, the parenteral route has been suggested to be the main route of infection for genotypes 1 and 2. However, for genotype 3 in sub-Saharan Africa higher prevalences of both PARV4 IgG and DNA have been detected, the latter in serum, stool, and nasal swabs, also in constitutionally healthy young children and adults without parenteral exposure (Panning et al., 2010; Drexler et al., 2012; Lavoie et al., 2012; May et al., 2012). This indicates that PARV4 genotype 3, apparently endemic in sub-Saharan Africa, may differ from genotypes 1 and 2 in epidemiology and transmission (Sharp et al., 2010b). The clinical significance of PARV4 is currently unknown. In a follow-up study of haemophilia patients receiving plasma-derived clotting factors, mild flu-like symptoms, rash and exacerbation of hepatitis were the most common symptoms prior to seroconversion (Sharp et al., 2012). PARV4 infection has been linked to severe disease in two case reports; in India, two children with unexplained encephalitis were reported to be PARV4 PCR-positive
E. Väisänen et al. / Journal of Virological Methods 195 (2014) 106–111
in their cerebrospinal fluid (Benjamin et al., 2011), and in Taiwan mothers and foetuses with hydrops fetalis showed PCR and IgG positivity in plasma (Chen et al., 2011). In the literature, PARV4 DNA has been detected with qualitative PCR or quantitative PCR (qPCR) conducted on conserved regions of the virus. No method has been able to genotype the virus without sequencing or amplification in parallel reactions, one for each genotype. This study describes a two-phase real-time quantitative PCR method that identifies, quantifies and distinguishes all three PARV4 genotypes currently known in a practical and economical manner, with an initial one-probe pan-PARV4 qPCR followed for the positive samples by quick single-tube real-time genotyping. This new assay, the qPCR-GT, provides a sensitive, cost-efficient and straightforward method for screening and genotyping large sample cohorts for the presence of these viruses. 2. Materials and methods 2.1. Plasmids Plasmids containing PARV4 sequences of genotypes (GT) 1, 2 or 3 were used for testing and optimisation of the PARV4 real-time quantitative PCR and genotyping (qPCR-GT) assay. Of note, for GT3, two separate plasmids were necessary due to variable nucleotides in the genotyping probe area (referred to as 3 and 3.2). The GT1 plasmid was a nearly full-length clone (Lahtinen et al., 2011), whereas plasmids for GTs 2, 3 and 3.2 were commercially synthesised, corresponding to nucleotides 1484–2593 of DQ873390, 1625–2735 of EU874248 and 1–577 of GU951556, respectively, and cloned into pUC57 by Genscript (Piscataway, NJ, USA). 2.2. Primers and hydrolysis probes For designing primers and hydrolysis probes for the twostep PARV4 qPCR-GT, a region combining both similarity and genotype-specific differences was selected based on a comprehensive PARV4 sequence alignment containing 48 sequences in GenBank (accessed on March 10th, 2012). The primers and probes were designed within ORF1, encoding the non-structural proteins, between nucleotides 1699 and 1857 (AY622943). The real-time qPCR pan-P4 probe was labelled with the fluorescent dye FAM, whereas genotyping probes were labelled with Texas Red, Cy5 and HEX dyes to minimise any possible carry-over signal from the firststep qPCR probe to the genotyping reaction. All primer and probe sequences and locations in the PARV4 genome used in this study are presented in Table 1. 2.3. Real-time quantitative PCR The primers selected based on the sequence analysis were tested initially with 1× Maxima SYBR Green qPCR Master Mix (Fermentas, Vilnius, Lithuania) to confirm the lack of primerdimers and amplification of human DNA. To determine the optimal concentrations of the primers and the probe for the actual hydrolysis-format qPCR, the following were tested: for primers, a matrix of 300 nM, 500 nM and 900 nM of forward and reverse primers, and for the probe 150 nM, 200 nM and 300 nM. All reactions included 1× Maxima Probe qPCR Master Mix (Fermentas) with 30 nM of ROX passive reference dye, and the amplifications were done with Stratagene Mx3005P (Agilent Technologies, CA, USA) with the following PCR programme: initial denaturation for 10 min at 95 ◦ C, then amplification for 45 cycles of 15 s at 95 ◦ C and 1 min at 60 ◦ C. Two temperatures, 58 ◦ C and 60 ◦ C, were tested for the combined annealing and elongation step due to the degenerative nature of the probe. The concentrations and
107
temperatures resulting in the earliest quantitation cycle (Cq ) and the best curve shape were selected for the final assay. The parameters giving the best results constituted the final qPCR reaction: 1 × Maxima Probe qPCR Master Mix with 30 nM of ROX passive reference dye, 900 nM of each primer PARV4 fwd and PARV4 rev (Table 1), 200 nM of the pan-P4 probe (Table 1), 5 l of template and H2 O up to a final volume of 25 l. The final amplification programme consisted of an initial denaturation for 10 min at 95 ◦ C, then amplification for 45 cycles of 15 s at 95 ◦ C and 1 min at 58 ◦ C. Fluorescent reporter signals from the FAM channel with an 8× gain setting were measured against the internal reference dye signal (ROX, 1× gain setting) to normalise for background fluorescence fluctuations between samples. A ten-fold dilution series of a plasmid containing the near full-length PARV4 genotype-1 genome was included in each run for the standard curve. One dilution of each GT plasmid, containing the sequences of PARV4 genotypes 2, 3 or 3.2, were amplified in each run, and water served as the negative control. The results were analysed with MxPro software (version 4.1., Agilent Technologies). The cut-off for Cq was set automatically at 10 times the standard deviation of the fluorescence values of cycles 5–9. Strict precautions were applied during sample and plasmid handling to avoid contaminations. Sample and master mix preparations, as well as amplification, were all done in separate rooms with single-use disposable materials and filtered tips.
2.4. Real-time genotyping assay In the genotyping reaction, the same primer pair (PARV4 fwd and PARV4 rev) was used with genotype-specific probes labelled with different reporter dyes: GT1 with Texas Red-BHQ2, GT2 with Cy5-BHQ3 and GT3 and GT3.2 with HEX-BHQ1 (Table 1). Of note, for genotype 3, two separate probes were necessary to cover all GT3 sequences, although the two probes were labelled with the same reporter dye. All genotyping reactions were optimised first in singleplex and then combined to multiplex. The reactions were tested by using as template the amplified products of high-copy number plasmids of every genotype to confirm that no cross-reactions existed. Different filter gain settings for each dye, from 1× up to 8×, were explored during the multiplex qPCR optimisation to achieve optimal curve shapes for interpretation of results. The temperature of the annealing and elongation step of the genotyping assay was set to 62 ◦ C to assure the annealing specificity of the genotype-specific probes. For the final real-time genotyping assay, the following parameters were chosen: 1× Maxima Probe qPCR Master Mix (Fermentas), 1 M of each primer, 0.3 M of P4 GT1, P4 GT2, P4 GT3 and P4 GT3.2 probes each (Table 1), 1.5 l of the PARV4 qPCR product as a template and H2 O up to a final volume of 25 l. The amplification was done by the same Stratagene Mx3005P qPCR machine with the following programme: initial denaturation for 10 min at 95 ◦ C, then amplification for 25 cycles of 15 s at 95 ◦ C and 1 min at 62 ◦ C. The fluorescent reporter signals from the ROX (for Texas Red), Cy5 and HEX channels were measured with the following gain settings: 1×, 2× and 2×, respectively. Water and plasmids containing each PARV4 genotype, amplified by the PARV4 qPCR, served as negative and positive controls in the genotyping reaction. The results were analysed with the R (multicomponent view) fluorescence mode of the MxPro software, and an unequivocal rise in the fluorescence curve directly from cycle 0 was interpreted as a positive result for the corresponding genotype (Fig. 2A–C). Of note, in most reactions the genotype can be identified within the first 5 cycles, i.e. all 25 cycles are not needed invariably, as evident in Fig. 2.
108
E. Väisänen et al. / Journal of Virological Methods 195 (2014) 106–111
Table 1 Sequences of primers and probes used in this study. Oligo name Primers PARV4 fwd PARV4 rev Probes PARV4 qPCR Pan-P4 probe Genotyping probes P4 GT1 P4 GT2 P4 GT3 P4 GT3.2
Sequence 5 –3
Nucleotidesa
GGTGAACAGGAACTGAATATGAAAG CCAATCATCTTCWGAATCCCA
1699–1723 1857–1837
FAM-AGTGCCAAGAAACTTTGACAARCARCCA-BHQ1
1746–1773
TexasRed-TCTTCCATTGGTGCACACAAGGCTYT-BHQ2 Cy5-TTCATCTATTGGTGCACAAAGAGCTT-BHQ3 HEX-CTCGTCTATTGGTGCACACAGAGC-BHQ1 HEX-CTCATCCATTGGTGCACATAGAGC-BHQ1
1823–1798 1824–1799 1824–1801 1824–1801
Mismatches towards other GTs GT2:5/GT3:5/GT3.2:4 GT1:6/GT3:4/GT3.2:4 GT1:5/GT2:3/GT3.2:3 GT1:5/GT2:3/GT3:3
All primers and probes were synthesised by Sigma–Aldrich, Haverhill, United Kingdom. W = A or T. R = G or A. Y = C or T. BHQ = Black Hole Quencher. a Corresponding to AY622943.
2.5. Analytical sensitivity and specificity A 10-fold plasmid dilution series with 1 × 100 to 1 × 106 plasmids/l of template was used to determine the sensitivity, specificity and inter- and intra-assay reproducibility of the qPCR, with or without the presence of 500 ng of human DNA per reaction. The inter-assay variability was assessed by analysing 10-fold dilution series of each genotype plasmid as triplicates in three different runs on different days and the intra-assay variability by testing the 10-fold dilution series of each genotype plasmid in triplicate in one run. The corresponding coefficients of variation (CV) were calculated from the inter- and intra-assay variability data (Table 2). Cross-reactivity towards other human parvoviruses was tested with 1 × 107 copies/l of parvovirus B19 (B19V) and human bocavirus 1 (HBoV1) plasmids. The products from positive and negative qPCR reactions were used further in specificity testing of the real-time genotyping assay.
2.6. Clinical samples To assess the suitability of the assay for analysing human samples, eight previously determined PARV4 DNA-positive (1 GT1, 4 GT2, 3 GT3) and 14 negative human serum/plasma samples were analysed with the qPCR-GT assay. The GT1-positive sample was from the first PARV4 patient described (Jones et al., 2005) and one of the GT2-positive samples was from an HIV-negative intravenous drug user collected in 2004, both kind gifts of Eric Delwart. The other three GT2-positive samples (two of which were from the same patient) and all of the negative samples were from an HCV-positive patient cohort (Lahtinen et al., 2011), whereas the three GT3-positive samples were from a cohort of constitutionally healthy children from Ghana (Panning et al., 2010). The DNA in all samples had been extracted in the original studies and stored thereafter at −20 ◦ C.
3. Results
Fig. 1. Real-time quantitative PCR amplification curves for genotype 1. Black solid lines: 10-fold dilution series of plasmids (5 × 106 to 5 × 101 copies per reaction) or H2 O as template; red lines marked with diamond: 5 × 105 or 5 × 102 copies of plasmids or H2 O with 500 ng of human DNA per reaction as template. Of note, the amplification curves for genotypes 2 and 3 were identical to those of genotype 1. (For interpretation of the references to color in the artwork, the reader is referred to the web version of the article.)
positives (Fig. 1). Furthermore, the assay did not amplify other human parvoviruses, B19V or HBoV1, during the 45-cycle protocol. 3.2. Inter- and intra-assay variability The inter- and intra-assay variability analyses showed low standard deviations and good reproducibility of the assay regardless of the genotype, with CVs ranging from 0.6% to 2.9% within three runs (inter-assay) and from 0.4% to 3.8% within one run (intraassay) (Tables 2A and 2B). The highest CV variations were observed with the lowest load, 5 × 101 copies per reaction. Similar reaction efficiencies were detected with each genotype, indicating robustness of the quantitation (Table 2A).
3.1. Analytical sensitivity of qPCR 3.3. Real-time genotyping assay The analytical sensitivity of the real-time qPCR assay was assessed with 10-fold plasmid dilutions of each genotype. The limit of detection was 5 copies per reaction for GT1 and 20 copies per reaction for GT2 and GT3. The qPCR showed linearity over a range of 50–5 × 106 copies per reaction for all three genotypes (Fig. 1). Human DNA did not interfere with the amplification or cause false
The genotype-specific probes were shown to be extremely specific: no inter-genotype cross-reactivity was observed, even with high-copy number templates (Fig. 2A–C). Human DNA did not affect the genotyping results, nor did multiplexing of the probes in the same reaction tube. Of note, genotyping reactions done without
E. Väisänen et al. / Journal of Virological Methods 195 (2014) 106–111
109
Table 2A Reproducibility of the first-step quantitative PCR for the three genotypes (GT). Inter-assay variation (CV) and mean quantitation cycles (Cq ) of three replicates performed on three different days. Copies/l
GT1
GT2
Cq ± SD 1
10 102 103 104 105 106
33.88 30.44 27.14 23.71 20.35 16.90
± ± ± ± ± ±
0.66 0.30 0.24 0.14 0.17 0.29
Efficiencya Slopea RSqa a
GT3
CV (%)
Cq ± SD
1.96 0.98 0.88 0.59 0.82 1.70
34.88 31.44 28.07 24.60 21.25 18.06
± ± ± ± ± ±
0.85 0.22 0.25 0.23 0.23 0.41
97.40% −3.387 0.997
CV (%)
Cq ± SD
2.43 0.69 0.89 0.92 1.07 2.25
35.43 31.41 27.95 24.55 21.23 17.89
± ± ± ± ± ±
CV (%) 1.04 0.37 0.22 0.29 0.22 0.16
97.97% −3.374 0.996
2.94 1.17 0.80 1.16 1.04 0.90 95.40% −3.439 0.995
Mean from the three runs.
Table 2B Reproducibility of the first-step quantitative PCR for the three genotypes (GT). Intra-assay variation (CV) and mean quantitation cycles (Cq ) of three replicates within a single run. Copies/l
GT1
GT2
Cq ± SD 101 102 103 104 105 106
33.65 30.58 27.31 23.74 20.25 16.99
± ± ± ± ± ±
0.30 0.29 0.32 0.13 0.08 0.18
GT3
CV (%)
Cq ± SD
0.88 0.94 1.18 0.54 0.40 1.03
34.58 31.35 28.06 24.60 21.47 18.39
± ± ± ± ± ±
0.70 0.20 0.18 0.17 0.11 0.40
CV (%)
Cq ± SD
2.02 0.64 0.63 0.67 0.50 2.16
36.14 31.32 27.95 24.75 21.13 17.83
± ± ± ± ± ±
CV (%) 1.38 0.37 0.18 0.18 0.19 0.10
3.83 1.18 0.64 0.71 0.90 0.57
Table 3 Characteristics of PARV4 DNA-positive human serum/plasma samples, analysed with the qPCR-GT assay. Genotype
Quantity (copies per ml)
Cohort
Country of origin
Reference
1 2 2 2 2 3 3 3
5.7 × 107 a 2.2 × 103 1.6 × 104 2.8 × 104 3.6 × 104 1.6 × 104 2.0 × 104 2.2 × 104
IDU, HIV-negative IDU, HIV-negative HCV-positive patients HCV-positive patients HCV-positive patients Healthy children Healthy children Healthy children
USA USA Finland Finland Finland Ghana Ghana Ghana
E. Delwart, Jones et al. (2005) E. Delwart Lahtinen et al. (2011) Lahtinen et al. (2011) Lahtinen et al. (2011) Panning et al. (2010) Panning et al. (2010) Panning et al. (2010)
IDU = intravenous drug user. a Per reaction.
any genotyping probes showed no curve rise in any fluorescence channel (data not shown). 3.4. Clinical samples The PARV4 DNA quantity was relatively low in 7 of 8 of the PARV4-positive human samples, ranging from 2.2 × 103 to 3.6 × 104 copies per ml of serum/plasma (Table 3). All of the 7 low positives represented genotypes 2 or 3 (Table 3 and Fig. 2B, C). The only genotype 1-positive sample was of high quantity, 5.7 × 107 copies per reaction (Table 3 and Fig. 2A). None of the 14 known negative samples showed any fluorescence signal during the 45 cycles of qPCR or in genotyping. In the real-time genotyping reaction, each of the PARV4-positive samples showed the correct result (Fig. 2A–C). No false signals were detected, neither from any positive sample towards wrong genotypes nor from the negative samples towards any probe, thus demonstrating that the genotyping assay had a high specificity also with clinical sample material. 4. Discussion The three genotypes of human parvovirus 4 share 92–93% identity at the nucleotide level, similar to that of parvovirus B19 genotypes (B19V) (Hokynar et al., 2002). However, unlike with
B19V, the PARV4 genotypes seem to have divergent transmission routes: types 1 and 2 appear to be transmitted parenterally in Northern Europe, Asia and Northern America, whereas genotype 3 is apparently endemic in sub-Saharan Africa, where it is detected in young children and adults without evidence of parenteral exposure (Sharp et al., 2009; Panning et al., 2010; Sharp et al., 2010b). Whether or to what extent the clinical picture varies among infections by the three genotypes is unknown. The three genotypes of PARV4 are identified currently only by labour-intensive sequencing. This paper describes a new assay that both quantifies and distinguishes the PARV4 types in a convenient and cost-efficient way. The method first employs a single hydrolysis-probe qPCR for screening and quantitation of the PARV4 DNA, followed by a multi-probe real-time PCR assay to genotype the virus in the positive samples. The two-step approach was adopted, since for analysing large sample materials a single-probe qPCR is more practical and economical than a multi-probe qPCR for a relatively rare virus. Furthermore, with primers and probe located in the conserved region of the virus, the single-probe qPCR is as sensitive for every genotype and also able to detect novel variants. Subsequent real-time genotyping of the positive samples also saves the valuable DNA extract as its template comes from the qPCR reaction. In addition, the genotyping assay based on a single reaction is fast to perform; no subsequent sequencing is needed. In all, the new qPCR-GT assay provides a sensitive, economical
110
E. Väisänen et al. / Journal of Virological Methods 195 (2014) 106–111
Fig. 2. Real-time PARV4 genotyping reactions after qPCR amplification, using the qPCR amplification products as templates in all reactions. The results in the three figures are from a single run with fluorescence detected with three different dye filters: (A) Texas red for GT1 probe, (B) Cy5 for GT2 probe, and (C) HEX for GT3 and GT3.2 probes. Red lines marked with O: a high-copy number plasmid sample corresponding to the shown genotyping probe(s); blue dashed lines marked with diamonds: a high-copy number plasmid sample corresponding to the shown genotyping probe(s) with 500 ng of human DNA; green lines marked with X: high- and low-load PARV4-positive human samples corresponding to the shown genotyping probe(s); black lines: high-copy number plasmid and PARV4-positive human samples of the other two genotypes not corresponding to the shown genotype, with and without 500 ng of human DNA, and PARV4-negative human samples and no-template controls. (For interpretation of the references to color in the artwork, the reader is referred to the web version of the article.)
and straightforward approach for screening and genotyping large sample cohorts for the presence of known genotypes of PARV4. Conventional PCR is time-consuming and often requires confirmation by other methods. Furthermore, it does not reveal the quantity of the virus, which can be an important factor in virus diagnostics, as exemplified by the human bocavirus 1 persisting at low loads in the respiratory tract, and parvovirus B19 in serum (Brown, 2004; Enders et al., 2006; Kantola et al., 2008). This PARV4 qPCR was shown to quantify all genotypes in a comparable fashion, even in human serum samples as low copy numbers. The single sample with a high copy number showed that PARV4 does cause high-titre viraemia, as demonstrated also by others (Tuke et al., 2010; Benjamin et al., 2011; Ma et al., 2012). This confirms that the assay detects, quantifies and genotypes similarly both high and low DNA quantities in human serum/plasma. The presence of human cellular DNA did not lower the sensitivity or the specificity. However, the functionality of this assay with other sample types needs to be confirmed by further studies. The genotyping reaction was demonstrated to be highly specific using both plasmids and human samples. This held true also for the distinction of the closest genotypes; a mismatch of only three
nucleotides was sufficient to discriminate between the genotypes. The two-step approach could therefore be useful also in genotyping other pathogens, especially when rapid results are needed or when sequencing is unavailable or is difficult to perform. In conclusion, a sensitive and specific assay for PARV4 quantitation and genotyping was developed, of which the first-step pan-PARV4 qPCR can also be used separately. The real-time genotyping assay is specific and fast to perform. The qPCR-GT method provides a straightforward way to analyse PARV4 and makes it feasible to study the impact and spread of the different genotypes. The approach is also readily applicable to the identification of other pathogens.
Acknowledgements We thank Eric Delwart for the two PARV4-positive human samples. This work was supported by grants from the University of Helsinki Research Fund, the Sigrid Jusélius Foundation, the Finnish Medical Foundation (FLS), the Instrumentarium Foundation and the Academy of Finland (grant # 1257964).
E. Väisänen et al. / Journal of Virological Methods 195 (2014) 106–111
References Adlhoch, C., Kaiser, M., Ellerbrok, H., Pauli, G., 2010. High prevalence of porcine Hokovirus in German wild boar populations. Virol. J. 7, 171-422X-7-171. Benjamin, L.A., Lewthwaite, P., Vasanthapuram, R., Zhao, G., Sharp, C., Simmonds, P., Wang, D., Solomon, T., 2011. Human parvovirus 4 as potential cause of encephalitis in children, India. Emerg. Infect. Dis. 17, 1484–1487. Brown, K.E., 2004. Detection and quantitation of parvovirus B19. J. Clin. Virol. 31, 1–4. Canuti, M., Eis-Huebinger, A.M., Deijs, M., de Vries, M., Drexler, J.F., Oppong, S.K., Muller, M.A., Klose, S.M., Wellinghausen, N., Cottontail, V.M., Kalko, E.K., Drosten, C., van der Hoek, L., 2011. Two novel parvoviruses in frugivorous new and old world bats. PLoS ONE 6, e29140. Chen, M.Y., Yang, S.J., Hung, C.C., 2011. Placental transmission of human parvovirus 4 in newborns with hydrops, Taiwan. Emerg. Infect. Dis. 17, 1954–1956. Drexler, J.F., Reber, U., Muth, D., Herzog, P., Annan, A., Ebach, F., Sarpong, N., Acquah, S., Adlkofer, J., Adu-Sarkodie, Y., Panning, M., Tannich, E., May, J., Drosten, C., EisHubinger, A.M., 2012. Human parvovirus 4 in nasal and fecal specimens from children, Ghana. Emerg. Infect. Dis. 18, 1650–1653. Enders, M., Schalasta, G., Baisch, C., Weidner, A., Pukkila, L., Kaikkonen, L., Lankinen, H., Hedman, L., Soderlund-Venermo, M., Hedman, K., 2006. Human parvovirus B19 infection during pregnancy – value of modern molecular and serological diagnostics. J. Clin. Virol. 35, 400–406. Fryer, J.F., Kapoor, A., Minor, P.D., Delwart, E., Baylis, S.A., 2006. Novel parvovirus and related variant in human plasma. Emerg. Infect. Dis. 12, 151–154. Fryer, J.F., Delwart, E., Bernardin, F., Tuke, P.W., Lukashov, V.V., Baylis, S.A., 2007a. Analysis of two human parvovirus PARV4 genotypes identified in human plasma for fractionation. J. Gen. Virol. 88, 2162–2167. Fryer, J.F., Delwart, E., Hecht, F.M., Bernardin, F., Jones, M.S., Shah, N., Baylis, S.A., 2007b. Frequent detection of the parvoviruses, PARV4 and PARV5, in plasma from blood donors and symptomatic individuals. Transfusion 47, 1054–1061. Hokynar, K., Soderlund-Venermo, M., Pesonen, M., Ranki, A., Kiviluoto, O., Partio, E.K., Hedman, K., 2002. A new parvovirus genotype persistent in human skin. Virology 302, 224–228. Jones, M.S., Kapoor, A., Lukashov, V.V., Simmonds, P., Hecht, F., Delwart, E., 2005. New DNA viruses identified in patients with acute viral infection syndrome. J. Virol. 79, 8230–8236. Kantola, K., Hedman, L., Allander, T., Jartti, T., Lehtinen, P., Ruuskanen, O., Hedman, K., Soderlund-Venermo, M., 2008. Serodiagnosis of human bocavirus infection. Clin. Infect. Dis. 46, 540–546. Lahtinen, A., Kivela, P., Hedman, L., Kumar, A., Kantele, A., Lappalainen, M., Liitsola, K., Ristola, M., Delwart, E., Sharp, C., Simmonds, P., Soderlund-Venermo, M., Hedman, K., 2011. Serodiagnosis of primary infections with human parvovirus 4, Finland. Emerg. Infect. Dis. 17, 79–82. Lau, S.K., Woo, P.C., Tse, H., Fu, C.T., Au, W.K., Chen, X.C., Tsoi, H.W., Tsang, T.H., Chan, J.S., Tsang, D.N., Li, K.S., Tse, C.W., Ng, T.K., Tsang, O.T., Zheng, B.J., Tam, S., Chan, K.H., Zhou, B., Yuen, K.Y., 2008. Identification of novel porcine and bovine parvoviruses closely related to human parvovirus 4. J. Gen. Virol. 89, 1840–1848. Lavoie, M., Sharp, C.P., Pepin, J., Pennington, C., Foupouapouognigni, Y., Pybus, O.G., Njouom, R., Simmonds, P., 2012. Human parvovirus 4 infection, Cameroon. Emerg. Infect. Dis. 18, 680–683. Lurcharchaiwong, W., Chieochansin, T., Payungporn, S., Theamboonlers, A., Poovorawan, Y., 2008. Parvovirus 4 (PARV4) in serum of intravenous drug users and blood donors. Infection 36, 488–491. Ma, Y.Y., Guo, Y., Zhao, X., Wang, Z., Lv, M.M., Yan, Q.P., Zhang, J.G., 2012. Human parvovirus PARV4 in plasma pools of Chinese origin. Vox Sang. 103, 183–185.
111
Manning, A., Willey, S.J., Bell, J.E., Simmonds, P., 2007. Comparison of tissue distribution, persistence, and molecular epidemiology of parvovirus B19 and novel human parvoviruses PARV4 and human bocavirus. J. Infect. Dis. 195, 1345–1352. Maple, P.A., Beard, S., Parry, R.P., Brown, K.E., 2013. Testing UK blood donors for exposure to human parvovirus 4 using a time-resolved fluorescence immunoassay to screen sera and Western blot to confirm reactive samples. Transfusion 53, 2575–2584. May, J., Drexler, J.F., Reber, U., Sarpong, N., Adjei, O., Panning, M., Drosten, C., EisHubinger, A.M., 2012. Human parvovirus 4 viremia in young children, Ghana. Emerg. Infect. Dis. 18, 1690–1692. Panning, M., Kobbe, R., Vollbach, S., Drexler, J.F., Adjei, S., Adjei, O., Drosten, C., May, J., Eis-Hubinger, A.M., 2010. Novel human parvovirus 4 genotype 3 in infants, Ghana. Emerg. Infect. Dis. 16, 1143–1146. Sharp, C.P., Lail, A., Donfield, S., Simmons, R., Leen, C., Klenerman, P., Delwart, E., Gomperts, E.D., Simmonds, P., 2009. High frequencies of exposure to the novel human parvovirus PARV4 in hemophiliacs and injection drug users, as detected by a serological assay for PARV4 antibodies. J. Infect. Dis. 200, 1119–1125. Sharp, C.P., LeBreton, M., Kantola, K., Nana, A., Diffo Jle, D., Djoko, C.F., Tamoufe, U., Kiyang, J.A., Babila, T.G., Ngole, E.M., Pybus, O.G., Delwart, E., Delaporte, E., Peeters, M., Soderlund-Venermo, M., Hedman, K., Wolfe, N.D., Simmonds, P., 2010a. Widespread infection with homologues of human parvoviruses B19, PARV4, and human bocavirus of chimpanzees and gorillas in the wild. J. Virol. 84, 10289–10296. Sharp, C.P., Vermeulen, M., Nebie, Y., Djoko, C.F., LeBreton, M., Tamoufe, U., Rimoin, A.W., Kayembe, P.K., Carr, J.K., Servant-Delmas, A., Laperche, S., Harrison, G.L., Pybus, O.G., Delwart, E., Wolfe, N.D., Saville, A., Lefrere, J.J., Simmonds, P., 2010b. Changing epidemiology of human parvovirus 4 infection in sub-Saharan Africa. Emerg. Infect. Dis. 16, 1605–1607. Sharp, C.P., Lail, A., Donfield, S., Gomperts, E.D., Simmonds, P., 2012. Virologic and clinical features of primary infection with human parvovirus 4 in subjects with hemophilia: frequent transmission by virally inactivated clotting factor concentrates. Transfusion 52, 1482–1489. Simmonds, P., Manning, A., Kenneil, R., Carnie, F.W., Bell, J.E., 2007. Parenteral transmission of the novel human parvovirus PARV4. Emerg. Infect. Dis. 13, 1386–1388. Simmonds, P., Douglas, J., Bestetti, G., Longhi, E., Antinori, S., Parravicini, C., Corbellino, M., 2008. A third genotype of the human parvovirus PARV4 in subSaharan Africa. J. Gen. Virol. 89, 2299–2302. Simmons, R., Sharp, C., McClure, C.P., Rohrbach, J., Kovari, H., Frangou, E., Simmonds, P., Irving, W., Rauch, A., Bowness, P., Klenerman, P., Swiss HIV Cohort Study, 2012. Parvovirus 4 infection and clinical outcome in high-risk populations. J. Infect. Dis. 205, 1816–1820. Touinssi, M., Brisbarre, N., Picard, C., Frassati, C., Dussol, B., Uch, R., Gallian, P., Cantaloube, J.F., Micco, P., Biagini, P., 2010. Parvovirus 4 in blood donors, France. Emerg. Infect. Dis. 16, 165–166. Tse, H., Tsoi, H.W., Teng, J.L., Chen, X.C., Liu, H., Zhou, B., Zheng, B.J., Woo, P.C., Lau, S.K., Yuen, K.Y., 2011. Discovery and genomic characterization of a novel ovine partetravirus and a new genotype of bovine partetravirus. PLoS ONE 6, e25619. Tuke, P.W., Parry, R.P., Appleton, H., 2010. Parvovirus PARV4 visualization and detection. J. Gen. Virol. 91, 541–544. Yang, S.J., Hung, C.C., Chang, S.Y., Lee, K.L., Chen, M.Y., 2011. Immunoglobulin G and M antibodies to human parvovirus 4 (PARV4) are frequently detected in patients with HIV-1 infection. J. Clin. Virol. 51, 64–67. Yu, X., Zhang, J., Hong, L., Wang, J., Yuan, Z., Zhang, X., Ghildyal, R., 2012. High prevalence of human parvovirus 4 infection in HBV and HCV infected individuals in shanghai. PLoS ONE 7, e29474.
Contents lists available at ScienceDirect
Journal of Virological Methods journal homepage: www.elsevier.com/locate/jviromet
A two-step real-time PCR assay for quantitation and genotyping of human parvovirus 4 E. Väisänen a,∗ , A. Lahtinen a , A.M. Eis-Hübinger b , M. Lappalainen c , K. Hedman a,c , M. Söderlund-Venermo a a
Haartman Institute, Department of Virology, University of Helsinki, Helsinki, Finland University of Bonn Medical Centre, Bonn, Germany c Helsinki University Hospital Laboratory Division, Helsinki, Finland b
a b s t r a c t Article history: Received 5 June 2013 Received in revised form 26 September 2013 Accepted 1 October 2013 Available online 14 October 2013 Keywords: Real-time PCR Parvoviridae PARV4 Genotyping Quantification
Human parvovirus 4 (PARV4) of the family Parvoviridae was discovered in a plasma sample of a patient with an undiagnosed acute infection in 2005. Currently, three PARV4 genotypes have been identified, however, with an unknown clinical significance. Interestingly, these genotypes seem to differ in epidemiology. In Northern Europe, USA and Asia, genotypes 1 and 2 have been found to occur mainly in persons with a history of injecting drug use or other parenteral exposure. In contrast, genotype 3 appears to be endemic in sub-Saharan Africa, where it infects children and adults without such risk behaviour. In this study, a novel straightforward and cost-efficient molecular assay for both quantitation and genotyping of PARV4 DNA was developed. The two-step method first applies a single-probe pan-PARV4 qPCR for screening and quantitation of this relatively rare virus, and subsequently, only the positive samples undergo a real-time PCR-based multi-probe genotyping. The new qPCR-GT method is highly sensitive and specific regardless of the genotype, and thus being suitable for studying the clinical impact and occurrence of the different PARV4 genotypes. © 2013 Elsevier B.V. All rights reserved.
1. Introduction Human parvovirus 4 (PARV4) is a small single-stranded DNA virus in the Parvoviridae family. It is one of the four parvoviruses known to infect humans alongside the human parvovirus B19, adeno-associated viruses and human bocaviruses. The first PARV4 sequences were identified in 2005 in a plasma sample from an intravenous drug user presenting with various symptoms of acute viral infection (Jones et al., 2005). A year later, a related virus, initially called PARV5, was found in plasma pools for manufacture of plasma-derived medical products and characterised as a PARV4 variant, genotype 2 (Fryer et al., 2006, 2007a). A third genotype was isolated in 2008 from two adult AIDS patients from Africa (Simmonds et al., 2008). Furthermore, PARV4-like viruses have been discovered in various animals, for example, in chimpanzees and other primates, bats, cows, sheep, pigs and wild boars (Lau et al., 2008; Adlhoch et al., 2010; Sharp et al., 2010a; Canuti et al., 2011; Tse et al., 2011). PARV4 infection is associated commonly with intravenous drug use and with seropositivity of human immunodeficiency virus (HIV), hepatitis B virus (HBV) or hepatitis C virus (HCV) in
∗ Corresponding author. Tel.: +358 9 191 26676. E-mail address: elina.vaisanen@helsinki.fi (E. Väisänen). 0166-0934/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.jviromet.2013.10.011
Northern Europe, the United States and Asia (Manning et al., 2007; Simmonds et al., 2007; Lurcharchaiwong et al., 2008; Sharp et al., 2009; Lahtinen et al., 2011; Yang et al., 2011; Simmons et al., 2012; Yu et al., 2012; Maple et al., 2013). The virus has been detected in blood products (Fryer et al., 2006, 2007a,b; Lurcharchaiwong et al., 2008; Touinssi et al., 2010; Ma et al., 2012) and in persons undergoing repetitive blood and blood-product transfusions; i.e. haemophilia patients are at greater risk of infection (Simmonds et al., 2007; Sharp et al., 2009, 2012). Therefore, the parenteral route has been suggested to be the main route of infection for genotypes 1 and 2. However, for genotype 3 in sub-Saharan Africa higher prevalences of both PARV4 IgG and DNA have been detected, the latter in serum, stool, and nasal swabs, also in constitutionally healthy young children and adults without parenteral exposure (Panning et al., 2010; Drexler et al., 2012; Lavoie et al., 2012; May et al., 2012). This indicates that PARV4 genotype 3, apparently endemic in sub-Saharan Africa, may differ from genotypes 1 and 2 in epidemiology and transmission (Sharp et al., 2010b). The clinical significance of PARV4 is currently unknown. In a follow-up study of haemophilia patients receiving plasma-derived clotting factors, mild flu-like symptoms, rash and exacerbation of hepatitis were the most common symptoms prior to seroconversion (Sharp et al., 2012). PARV4 infection has been linked to severe disease in two case reports; in India, two children with unexplained encephalitis were reported to be PARV4 PCR-positive
E. Väisänen et al. / Journal of Virological Methods 195 (2014) 106–111
in their cerebrospinal fluid (Benjamin et al., 2011), and in Taiwan mothers and foetuses with hydrops fetalis showed PCR and IgG positivity in plasma (Chen et al., 2011). In the literature, PARV4 DNA has been detected with qualitative PCR or quantitative PCR (qPCR) conducted on conserved regions of the virus. No method has been able to genotype the virus without sequencing or amplification in parallel reactions, one for each genotype. This study describes a two-phase real-time quantitative PCR method that identifies, quantifies and distinguishes all three PARV4 genotypes currently known in a practical and economical manner, with an initial one-probe pan-PARV4 qPCR followed for the positive samples by quick single-tube real-time genotyping. This new assay, the qPCR-GT, provides a sensitive, cost-efficient and straightforward method for screening and genotyping large sample cohorts for the presence of these viruses. 2. Materials and methods 2.1. Plasmids Plasmids containing PARV4 sequences of genotypes (GT) 1, 2 or 3 were used for testing and optimisation of the PARV4 real-time quantitative PCR and genotyping (qPCR-GT) assay. Of note, for GT3, two separate plasmids were necessary due to variable nucleotides in the genotyping probe area (referred to as 3 and 3.2). The GT1 plasmid was a nearly full-length clone (Lahtinen et al., 2011), whereas plasmids for GTs 2, 3 and 3.2 were commercially synthesised, corresponding to nucleotides 1484–2593 of DQ873390, 1625–2735 of EU874248 and 1–577 of GU951556, respectively, and cloned into pUC57 by Genscript (Piscataway, NJ, USA). 2.2. Primers and hydrolysis probes For designing primers and hydrolysis probes for the twostep PARV4 qPCR-GT, a region combining both similarity and genotype-specific differences was selected based on a comprehensive PARV4 sequence alignment containing 48 sequences in GenBank (accessed on March 10th, 2012). The primers and probes were designed within ORF1, encoding the non-structural proteins, between nucleotides 1699 and 1857 (AY622943). The real-time qPCR pan-P4 probe was labelled with the fluorescent dye FAM, whereas genotyping probes were labelled with Texas Red, Cy5 and HEX dyes to minimise any possible carry-over signal from the firststep qPCR probe to the genotyping reaction. All primer and probe sequences and locations in the PARV4 genome used in this study are presented in Table 1. 2.3. Real-time quantitative PCR The primers selected based on the sequence analysis were tested initially with 1× Maxima SYBR Green qPCR Master Mix (Fermentas, Vilnius, Lithuania) to confirm the lack of primerdimers and amplification of human DNA. To determine the optimal concentrations of the primers and the probe for the actual hydrolysis-format qPCR, the following were tested: for primers, a matrix of 300 nM, 500 nM and 900 nM of forward and reverse primers, and for the probe 150 nM, 200 nM and 300 nM. All reactions included 1× Maxima Probe qPCR Master Mix (Fermentas) with 30 nM of ROX passive reference dye, and the amplifications were done with Stratagene Mx3005P (Agilent Technologies, CA, USA) with the following PCR programme: initial denaturation for 10 min at 95 ◦ C, then amplification for 45 cycles of 15 s at 95 ◦ C and 1 min at 60 ◦ C. Two temperatures, 58 ◦ C and 60 ◦ C, were tested for the combined annealing and elongation step due to the degenerative nature of the probe. The concentrations and
107
temperatures resulting in the earliest quantitation cycle (Cq ) and the best curve shape were selected for the final assay. The parameters giving the best results constituted the final qPCR reaction: 1 × Maxima Probe qPCR Master Mix with 30 nM of ROX passive reference dye, 900 nM of each primer PARV4 fwd and PARV4 rev (Table 1), 200 nM of the pan-P4 probe (Table 1), 5 l of template and H2 O up to a final volume of 25 l. The final amplification programme consisted of an initial denaturation for 10 min at 95 ◦ C, then amplification for 45 cycles of 15 s at 95 ◦ C and 1 min at 58 ◦ C. Fluorescent reporter signals from the FAM channel with an 8× gain setting were measured against the internal reference dye signal (ROX, 1× gain setting) to normalise for background fluorescence fluctuations between samples. A ten-fold dilution series of a plasmid containing the near full-length PARV4 genotype-1 genome was included in each run for the standard curve. One dilution of each GT plasmid, containing the sequences of PARV4 genotypes 2, 3 or 3.2, were amplified in each run, and water served as the negative control. The results were analysed with MxPro software (version 4.1., Agilent Technologies). The cut-off for Cq was set automatically at 10 times the standard deviation of the fluorescence values of cycles 5–9. Strict precautions were applied during sample and plasmid handling to avoid contaminations. Sample and master mix preparations, as well as amplification, were all done in separate rooms with single-use disposable materials and filtered tips.
2.4. Real-time genotyping assay In the genotyping reaction, the same primer pair (PARV4 fwd and PARV4 rev) was used with genotype-specific probes labelled with different reporter dyes: GT1 with Texas Red-BHQ2, GT2 with Cy5-BHQ3 and GT3 and GT3.2 with HEX-BHQ1 (Table 1). Of note, for genotype 3, two separate probes were necessary to cover all GT3 sequences, although the two probes were labelled with the same reporter dye. All genotyping reactions were optimised first in singleplex and then combined to multiplex. The reactions were tested by using as template the amplified products of high-copy number plasmids of every genotype to confirm that no cross-reactions existed. Different filter gain settings for each dye, from 1× up to 8×, were explored during the multiplex qPCR optimisation to achieve optimal curve shapes for interpretation of results. The temperature of the annealing and elongation step of the genotyping assay was set to 62 ◦ C to assure the annealing specificity of the genotype-specific probes. For the final real-time genotyping assay, the following parameters were chosen: 1× Maxima Probe qPCR Master Mix (Fermentas), 1 M of each primer, 0.3 M of P4 GT1, P4 GT2, P4 GT3 and P4 GT3.2 probes each (Table 1), 1.5 l of the PARV4 qPCR product as a template and H2 O up to a final volume of 25 l. The amplification was done by the same Stratagene Mx3005P qPCR machine with the following programme: initial denaturation for 10 min at 95 ◦ C, then amplification for 25 cycles of 15 s at 95 ◦ C and 1 min at 62 ◦ C. The fluorescent reporter signals from the ROX (for Texas Red), Cy5 and HEX channels were measured with the following gain settings: 1×, 2× and 2×, respectively. Water and plasmids containing each PARV4 genotype, amplified by the PARV4 qPCR, served as negative and positive controls in the genotyping reaction. The results were analysed with the R (multicomponent view) fluorescence mode of the MxPro software, and an unequivocal rise in the fluorescence curve directly from cycle 0 was interpreted as a positive result for the corresponding genotype (Fig. 2A–C). Of note, in most reactions the genotype can be identified within the first 5 cycles, i.e. all 25 cycles are not needed invariably, as evident in Fig. 2.
108
E. Väisänen et al. / Journal of Virological Methods 195 (2014) 106–111
Table 1 Sequences of primers and probes used in this study. Oligo name Primers PARV4 fwd PARV4 rev Probes PARV4 qPCR Pan-P4 probe Genotyping probes P4 GT1 P4 GT2 P4 GT3 P4 GT3.2
Sequence 5 –3
Nucleotidesa
GGTGAACAGGAACTGAATATGAAAG CCAATCATCTTCWGAATCCCA
1699–1723 1857–1837
FAM-AGTGCCAAGAAACTTTGACAARCARCCA-BHQ1
1746–1773
TexasRed-TCTTCCATTGGTGCACACAAGGCTYT-BHQ2 Cy5-TTCATCTATTGGTGCACAAAGAGCTT-BHQ3 HEX-CTCGTCTATTGGTGCACACAGAGC-BHQ1 HEX-CTCATCCATTGGTGCACATAGAGC-BHQ1
1823–1798 1824–1799 1824–1801 1824–1801
Mismatches towards other GTs GT2:5/GT3:5/GT3.2:4 GT1:6/GT3:4/GT3.2:4 GT1:5/GT2:3/GT3.2:3 GT1:5/GT2:3/GT3:3
All primers and probes were synthesised by Sigma–Aldrich, Haverhill, United Kingdom. W = A or T. R = G or A. Y = C or T. BHQ = Black Hole Quencher. a Corresponding to AY622943.
2.5. Analytical sensitivity and specificity A 10-fold plasmid dilution series with 1 × 100 to 1 × 106 plasmids/l of template was used to determine the sensitivity, specificity and inter- and intra-assay reproducibility of the qPCR, with or without the presence of 500 ng of human DNA per reaction. The inter-assay variability was assessed by analysing 10-fold dilution series of each genotype plasmid as triplicates in three different runs on different days and the intra-assay variability by testing the 10-fold dilution series of each genotype plasmid in triplicate in one run. The corresponding coefficients of variation (CV) were calculated from the inter- and intra-assay variability data (Table 2). Cross-reactivity towards other human parvoviruses was tested with 1 × 107 copies/l of parvovirus B19 (B19V) and human bocavirus 1 (HBoV1) plasmids. The products from positive and negative qPCR reactions were used further in specificity testing of the real-time genotyping assay.
2.6. Clinical samples To assess the suitability of the assay for analysing human samples, eight previously determined PARV4 DNA-positive (1 GT1, 4 GT2, 3 GT3) and 14 negative human serum/plasma samples were analysed with the qPCR-GT assay. The GT1-positive sample was from the first PARV4 patient described (Jones et al., 2005) and one of the GT2-positive samples was from an HIV-negative intravenous drug user collected in 2004, both kind gifts of Eric Delwart. The other three GT2-positive samples (two of which were from the same patient) and all of the negative samples were from an HCV-positive patient cohort (Lahtinen et al., 2011), whereas the three GT3-positive samples were from a cohort of constitutionally healthy children from Ghana (Panning et al., 2010). The DNA in all samples had been extracted in the original studies and stored thereafter at −20 ◦ C.
3. Results
Fig. 1. Real-time quantitative PCR amplification curves for genotype 1. Black solid lines: 10-fold dilution series of plasmids (5 × 106 to 5 × 101 copies per reaction) or H2 O as template; red lines marked with diamond: 5 × 105 or 5 × 102 copies of plasmids or H2 O with 500 ng of human DNA per reaction as template. Of note, the amplification curves for genotypes 2 and 3 were identical to those of genotype 1. (For interpretation of the references to color in the artwork, the reader is referred to the web version of the article.)
positives (Fig. 1). Furthermore, the assay did not amplify other human parvoviruses, B19V or HBoV1, during the 45-cycle protocol. 3.2. Inter- and intra-assay variability The inter- and intra-assay variability analyses showed low standard deviations and good reproducibility of the assay regardless of the genotype, with CVs ranging from 0.6% to 2.9% within three runs (inter-assay) and from 0.4% to 3.8% within one run (intraassay) (Tables 2A and 2B). The highest CV variations were observed with the lowest load, 5 × 101 copies per reaction. Similar reaction efficiencies were detected with each genotype, indicating robustness of the quantitation (Table 2A).
3.1. Analytical sensitivity of qPCR 3.3. Real-time genotyping assay The analytical sensitivity of the real-time qPCR assay was assessed with 10-fold plasmid dilutions of each genotype. The limit of detection was 5 copies per reaction for GT1 and 20 copies per reaction for GT2 and GT3. The qPCR showed linearity over a range of 50–5 × 106 copies per reaction for all three genotypes (Fig. 1). Human DNA did not interfere with the amplification or cause false
The genotype-specific probes were shown to be extremely specific: no inter-genotype cross-reactivity was observed, even with high-copy number templates (Fig. 2A–C). Human DNA did not affect the genotyping results, nor did multiplexing of the probes in the same reaction tube. Of note, genotyping reactions done without
E. Väisänen et al. / Journal of Virological Methods 195 (2014) 106–111
109
Table 2A Reproducibility of the first-step quantitative PCR for the three genotypes (GT). Inter-assay variation (CV) and mean quantitation cycles (Cq ) of three replicates performed on three different days. Copies/l
GT1
GT2
Cq ± SD 1
10 102 103 104 105 106
33.88 30.44 27.14 23.71 20.35 16.90
± ± ± ± ± ±
0.66 0.30 0.24 0.14 0.17 0.29
Efficiencya Slopea RSqa a
GT3
CV (%)
Cq ± SD
1.96 0.98 0.88 0.59 0.82 1.70
34.88 31.44 28.07 24.60 21.25 18.06
± ± ± ± ± ±
0.85 0.22 0.25 0.23 0.23 0.41
97.40% −3.387 0.997
CV (%)
Cq ± SD
2.43 0.69 0.89 0.92 1.07 2.25
35.43 31.41 27.95 24.55 21.23 17.89
± ± ± ± ± ±
CV (%) 1.04 0.37 0.22 0.29 0.22 0.16
97.97% −3.374 0.996
2.94 1.17 0.80 1.16 1.04 0.90 95.40% −3.439 0.995
Mean from the three runs.
Table 2B Reproducibility of the first-step quantitative PCR for the three genotypes (GT). Intra-assay variation (CV) and mean quantitation cycles (Cq ) of three replicates within a single run. Copies/l
GT1
GT2
Cq ± SD 101 102 103 104 105 106
33.65 30.58 27.31 23.74 20.25 16.99
± ± ± ± ± ±
0.30 0.29 0.32 0.13 0.08 0.18
GT3
CV (%)
Cq ± SD
0.88 0.94 1.18 0.54 0.40 1.03
34.58 31.35 28.06 24.60 21.47 18.39
± ± ± ± ± ±
0.70 0.20 0.18 0.17 0.11 0.40
CV (%)
Cq ± SD
2.02 0.64 0.63 0.67 0.50 2.16
36.14 31.32 27.95 24.75 21.13 17.83
± ± ± ± ± ±
CV (%) 1.38 0.37 0.18 0.18 0.19 0.10
3.83 1.18 0.64 0.71 0.90 0.57
Table 3 Characteristics of PARV4 DNA-positive human serum/plasma samples, analysed with the qPCR-GT assay. Genotype
Quantity (copies per ml)
Cohort
Country of origin
Reference
1 2 2 2 2 3 3 3
5.7 × 107 a 2.2 × 103 1.6 × 104 2.8 × 104 3.6 × 104 1.6 × 104 2.0 × 104 2.2 × 104
IDU, HIV-negative IDU, HIV-negative HCV-positive patients HCV-positive patients HCV-positive patients Healthy children Healthy children Healthy children
USA USA Finland Finland Finland Ghana Ghana Ghana
E. Delwart, Jones et al. (2005) E. Delwart Lahtinen et al. (2011) Lahtinen et al. (2011) Lahtinen et al. (2011) Panning et al. (2010) Panning et al. (2010) Panning et al. (2010)
IDU = intravenous drug user. a Per reaction.
any genotyping probes showed no curve rise in any fluorescence channel (data not shown). 3.4. Clinical samples The PARV4 DNA quantity was relatively low in 7 of 8 of the PARV4-positive human samples, ranging from 2.2 × 103 to 3.6 × 104 copies per ml of serum/plasma (Table 3). All of the 7 low positives represented genotypes 2 or 3 (Table 3 and Fig. 2B, C). The only genotype 1-positive sample was of high quantity, 5.7 × 107 copies per reaction (Table 3 and Fig. 2A). None of the 14 known negative samples showed any fluorescence signal during the 45 cycles of qPCR or in genotyping. In the real-time genotyping reaction, each of the PARV4-positive samples showed the correct result (Fig. 2A–C). No false signals were detected, neither from any positive sample towards wrong genotypes nor from the negative samples towards any probe, thus demonstrating that the genotyping assay had a high specificity also with clinical sample material. 4. Discussion The three genotypes of human parvovirus 4 share 92–93% identity at the nucleotide level, similar to that of parvovirus B19 genotypes (B19V) (Hokynar et al., 2002). However, unlike with
B19V, the PARV4 genotypes seem to have divergent transmission routes: types 1 and 2 appear to be transmitted parenterally in Northern Europe, Asia and Northern America, whereas genotype 3 is apparently endemic in sub-Saharan Africa, where it is detected in young children and adults without evidence of parenteral exposure (Sharp et al., 2009; Panning et al., 2010; Sharp et al., 2010b). Whether or to what extent the clinical picture varies among infections by the three genotypes is unknown. The three genotypes of PARV4 are identified currently only by labour-intensive sequencing. This paper describes a new assay that both quantifies and distinguishes the PARV4 types in a convenient and cost-efficient way. The method first employs a single hydrolysis-probe qPCR for screening and quantitation of the PARV4 DNA, followed by a multi-probe real-time PCR assay to genotype the virus in the positive samples. The two-step approach was adopted, since for analysing large sample materials a single-probe qPCR is more practical and economical than a multi-probe qPCR for a relatively rare virus. Furthermore, with primers and probe located in the conserved region of the virus, the single-probe qPCR is as sensitive for every genotype and also able to detect novel variants. Subsequent real-time genotyping of the positive samples also saves the valuable DNA extract as its template comes from the qPCR reaction. In addition, the genotyping assay based on a single reaction is fast to perform; no subsequent sequencing is needed. In all, the new qPCR-GT assay provides a sensitive, economical
110
E. Väisänen et al. / Journal of Virological Methods 195 (2014) 106–111
Fig. 2. Real-time PARV4 genotyping reactions after qPCR amplification, using the qPCR amplification products as templates in all reactions. The results in the three figures are from a single run with fluorescence detected with three different dye filters: (A) Texas red for GT1 probe, (B) Cy5 for GT2 probe, and (C) HEX for GT3 and GT3.2 probes. Red lines marked with O: a high-copy number plasmid sample corresponding to the shown genotyping probe(s); blue dashed lines marked with diamonds: a high-copy number plasmid sample corresponding to the shown genotyping probe(s) with 500 ng of human DNA; green lines marked with X: high- and low-load PARV4-positive human samples corresponding to the shown genotyping probe(s); black lines: high-copy number plasmid and PARV4-positive human samples of the other two genotypes not corresponding to the shown genotype, with and without 500 ng of human DNA, and PARV4-negative human samples and no-template controls. (For interpretation of the references to color in the artwork, the reader is referred to the web version of the article.)
and straightforward approach for screening and genotyping large sample cohorts for the presence of known genotypes of PARV4. Conventional PCR is time-consuming and often requires confirmation by other methods. Furthermore, it does not reveal the quantity of the virus, which can be an important factor in virus diagnostics, as exemplified by the human bocavirus 1 persisting at low loads in the respiratory tract, and parvovirus B19 in serum (Brown, 2004; Enders et al., 2006; Kantola et al., 2008). This PARV4 qPCR was shown to quantify all genotypes in a comparable fashion, even in human serum samples as low copy numbers. The single sample with a high copy number showed that PARV4 does cause high-titre viraemia, as demonstrated also by others (Tuke et al., 2010; Benjamin et al., 2011; Ma et al., 2012). This confirms that the assay detects, quantifies and genotypes similarly both high and low DNA quantities in human serum/plasma. The presence of human cellular DNA did not lower the sensitivity or the specificity. However, the functionality of this assay with other sample types needs to be confirmed by further studies. The genotyping reaction was demonstrated to be highly specific using both plasmids and human samples. This held true also for the distinction of the closest genotypes; a mismatch of only three
nucleotides was sufficient to discriminate between the genotypes. The two-step approach could therefore be useful also in genotyping other pathogens, especially when rapid results are needed or when sequencing is unavailable or is difficult to perform. In conclusion, a sensitive and specific assay for PARV4 quantitation and genotyping was developed, of which the first-step pan-PARV4 qPCR can also be used separately. The real-time genotyping assay is specific and fast to perform. The qPCR-GT method provides a straightforward way to analyse PARV4 and makes it feasible to study the impact and spread of the different genotypes. The approach is also readily applicable to the identification of other pathogens.
Acknowledgements We thank Eric Delwart for the two PARV4-positive human samples. This work was supported by grants from the University of Helsinki Research Fund, the Sigrid Jusélius Foundation, the Finnish Medical Foundation (FLS), the Instrumentarium Foundation and the Academy of Finland (grant # 1257964).
E. Väisänen et al. / Journal of Virological Methods 195 (2014) 106–111
References Adlhoch, C., Kaiser, M., Ellerbrok, H., Pauli, G., 2010. High prevalence of porcine Hokovirus in German wild boar populations. Virol. J. 7, 171-422X-7-171. Benjamin, L.A., Lewthwaite, P., Vasanthapuram, R., Zhao, G., Sharp, C., Simmonds, P., Wang, D., Solomon, T., 2011. Human parvovirus 4 as potential cause of encephalitis in children, India. Emerg. Infect. Dis. 17, 1484–1487. Brown, K.E., 2004. Detection and quantitation of parvovirus B19. J. Clin. Virol. 31, 1–4. Canuti, M., Eis-Huebinger, A.M., Deijs, M., de Vries, M., Drexler, J.F., Oppong, S.K., Muller, M.A., Klose, S.M., Wellinghausen, N., Cottontail, V.M., Kalko, E.K., Drosten, C., van der Hoek, L., 2011. Two novel parvoviruses in frugivorous new and old world bats. PLoS ONE 6, e29140. Chen, M.Y., Yang, S.J., Hung, C.C., 2011. Placental transmission of human parvovirus 4 in newborns with hydrops, Taiwan. Emerg. Infect. Dis. 17, 1954–1956. Drexler, J.F., Reber, U., Muth, D., Herzog, P., Annan, A., Ebach, F., Sarpong, N., Acquah, S., Adlkofer, J., Adu-Sarkodie, Y., Panning, M., Tannich, E., May, J., Drosten, C., EisHubinger, A.M., 2012. Human parvovirus 4 in nasal and fecal specimens from children, Ghana. Emerg. Infect. Dis. 18, 1650–1653. Enders, M., Schalasta, G., Baisch, C., Weidner, A., Pukkila, L., Kaikkonen, L., Lankinen, H., Hedman, L., Soderlund-Venermo, M., Hedman, K., 2006. Human parvovirus B19 infection during pregnancy – value of modern molecular and serological diagnostics. J. Clin. Virol. 35, 400–406. Fryer, J.F., Kapoor, A., Minor, P.D., Delwart, E., Baylis, S.A., 2006. Novel parvovirus and related variant in human plasma. Emerg. Infect. Dis. 12, 151–154. Fryer, J.F., Delwart, E., Bernardin, F., Tuke, P.W., Lukashov, V.V., Baylis, S.A., 2007a. Analysis of two human parvovirus PARV4 genotypes identified in human plasma for fractionation. J. Gen. Virol. 88, 2162–2167. Fryer, J.F., Delwart, E., Hecht, F.M., Bernardin, F., Jones, M.S., Shah, N., Baylis, S.A., 2007b. Frequent detection of the parvoviruses, PARV4 and PARV5, in plasma from blood donors and symptomatic individuals. Transfusion 47, 1054–1061. Hokynar, K., Soderlund-Venermo, M., Pesonen, M., Ranki, A., Kiviluoto, O., Partio, E.K., Hedman, K., 2002. A new parvovirus genotype persistent in human skin. Virology 302, 224–228. Jones, M.S., Kapoor, A., Lukashov, V.V., Simmonds, P., Hecht, F., Delwart, E., 2005. New DNA viruses identified in patients with acute viral infection syndrome. J. Virol. 79, 8230–8236. Kantola, K., Hedman, L., Allander, T., Jartti, T., Lehtinen, P., Ruuskanen, O., Hedman, K., Soderlund-Venermo, M., 2008. Serodiagnosis of human bocavirus infection. Clin. Infect. Dis. 46, 540–546. Lahtinen, A., Kivela, P., Hedman, L., Kumar, A., Kantele, A., Lappalainen, M., Liitsola, K., Ristola, M., Delwart, E., Sharp, C., Simmonds, P., Soderlund-Venermo, M., Hedman, K., 2011. Serodiagnosis of primary infections with human parvovirus 4, Finland. Emerg. Infect. Dis. 17, 79–82. Lau, S.K., Woo, P.C., Tse, H., Fu, C.T., Au, W.K., Chen, X.C., Tsoi, H.W., Tsang, T.H., Chan, J.S., Tsang, D.N., Li, K.S., Tse, C.W., Ng, T.K., Tsang, O.T., Zheng, B.J., Tam, S., Chan, K.H., Zhou, B., Yuen, K.Y., 2008. Identification of novel porcine and bovine parvoviruses closely related to human parvovirus 4. J. Gen. Virol. 89, 1840–1848. Lavoie, M., Sharp, C.P., Pepin, J., Pennington, C., Foupouapouognigni, Y., Pybus, O.G., Njouom, R., Simmonds, P., 2012. Human parvovirus 4 infection, Cameroon. Emerg. Infect. Dis. 18, 680–683. Lurcharchaiwong, W., Chieochansin, T., Payungporn, S., Theamboonlers, A., Poovorawan, Y., 2008. Parvovirus 4 (PARV4) in serum of intravenous drug users and blood donors. Infection 36, 488–491. Ma, Y.Y., Guo, Y., Zhao, X., Wang, Z., Lv, M.M., Yan, Q.P., Zhang, J.G., 2012. Human parvovirus PARV4 in plasma pools of Chinese origin. Vox Sang. 103, 183–185.
111
Manning, A., Willey, S.J., Bell, J.E., Simmonds, P., 2007. Comparison of tissue distribution, persistence, and molecular epidemiology of parvovirus B19 and novel human parvoviruses PARV4 and human bocavirus. J. Infect. Dis. 195, 1345–1352. Maple, P.A., Beard, S., Parry, R.P., Brown, K.E., 2013. Testing UK blood donors for exposure to human parvovirus 4 using a time-resolved fluorescence immunoassay to screen sera and Western blot to confirm reactive samples. Transfusion 53, 2575–2584. May, J., Drexler, J.F., Reber, U., Sarpong, N., Adjei, O., Panning, M., Drosten, C., EisHubinger, A.M., 2012. Human parvovirus 4 viremia in young children, Ghana. Emerg. Infect. Dis. 18, 1690–1692. Panning, M., Kobbe, R., Vollbach, S., Drexler, J.F., Adjei, S., Adjei, O., Drosten, C., May, J., Eis-Hubinger, A.M., 2010. Novel human parvovirus 4 genotype 3 in infants, Ghana. Emerg. Infect. Dis. 16, 1143–1146. Sharp, C.P., Lail, A., Donfield, S., Simmons, R., Leen, C., Klenerman, P., Delwart, E., Gomperts, E.D., Simmonds, P., 2009. High frequencies of exposure to the novel human parvovirus PARV4 in hemophiliacs and injection drug users, as detected by a serological assay for PARV4 antibodies. J. Infect. Dis. 200, 1119–1125. Sharp, C.P., LeBreton, M., Kantola, K., Nana, A., Diffo Jle, D., Djoko, C.F., Tamoufe, U., Kiyang, J.A., Babila, T.G., Ngole, E.M., Pybus, O.G., Delwart, E., Delaporte, E., Peeters, M., Soderlund-Venermo, M., Hedman, K., Wolfe, N.D., Simmonds, P., 2010a. Widespread infection with homologues of human parvoviruses B19, PARV4, and human bocavirus of chimpanzees and gorillas in the wild. J. Virol. 84, 10289–10296. Sharp, C.P., Vermeulen, M., Nebie, Y., Djoko, C.F., LeBreton, M., Tamoufe, U., Rimoin, A.W., Kayembe, P.K., Carr, J.K., Servant-Delmas, A., Laperche, S., Harrison, G.L., Pybus, O.G., Delwart, E., Wolfe, N.D., Saville, A., Lefrere, J.J., Simmonds, P., 2010b. Changing epidemiology of human parvovirus 4 infection in sub-Saharan Africa. Emerg. Infect. Dis. 16, 1605–1607. Sharp, C.P., Lail, A., Donfield, S., Gomperts, E.D., Simmonds, P., 2012. Virologic and clinical features of primary infection with human parvovirus 4 in subjects with hemophilia: frequent transmission by virally inactivated clotting factor concentrates. Transfusion 52, 1482–1489. Simmonds, P., Manning, A., Kenneil, R., Carnie, F.W., Bell, J.E., 2007. Parenteral transmission of the novel human parvovirus PARV4. Emerg. Infect. Dis. 13, 1386–1388. Simmonds, P., Douglas, J., Bestetti, G., Longhi, E., Antinori, S., Parravicini, C., Corbellino, M., 2008. A third genotype of the human parvovirus PARV4 in subSaharan Africa. J. Gen. Virol. 89, 2299–2302. Simmons, R., Sharp, C., McClure, C.P., Rohrbach, J., Kovari, H., Frangou, E., Simmonds, P., Irving, W., Rauch, A., Bowness, P., Klenerman, P., Swiss HIV Cohort Study, 2012. Parvovirus 4 infection and clinical outcome in high-risk populations. J. Infect. Dis. 205, 1816–1820. Touinssi, M., Brisbarre, N., Picard, C., Frassati, C., Dussol, B., Uch, R., Gallian, P., Cantaloube, J.F., Micco, P., Biagini, P., 2010. Parvovirus 4 in blood donors, France. Emerg. Infect. Dis. 16, 165–166. Tse, H., Tsoi, H.W., Teng, J.L., Chen, X.C., Liu, H., Zhou, B., Zheng, B.J., Woo, P.C., Lau, S.K., Yuen, K.Y., 2011. Discovery and genomic characterization of a novel ovine partetravirus and a new genotype of bovine partetravirus. PLoS ONE 6, e25619. Tuke, P.W., Parry, R.P., Appleton, H., 2010. Parvovirus PARV4 visualization and detection. J. Gen. Virol. 91, 541–544. Yang, S.J., Hung, C.C., Chang, S.Y., Lee, K.L., Chen, M.Y., 2011. Immunoglobulin G and M antibodies to human parvovirus 4 (PARV4) are frequently detected in patients with HIV-1 infection. J. Clin. Virol. 51, 64–67. Yu, X., Zhang, J., Hong, L., Wang, J., Yuan, Z., Zhang, X., Ghildyal, R., 2012. High prevalence of human parvovirus 4 infection in HBV and HCV infected individuals in shanghai. PLoS ONE 7, e29474.
