Human bocavirus is detected in human placenta and aborted tissues

Background To date, four human bocaviruses (HBoV) have been described. The most closely related viruses (bovine and canine parvoviruses) are associated with miscarriage in their hosts. The objective of this retrospective study was to determine the frequency of HBoV DNA in miscarriage. Study Design Tissue samples from 172 patients, in which miscarriage occurred, were included and tested with a published qPCR protocol. Positive PCRs were mutually confirmed by sequencing. Results 43 patients (25%) were positive for HBoV DNA. Of those, the majority of HBoV‐positive samples were tissues from miscarriage (placenta: 6; aborted tissue products of conception: 37 specimens). The samples were not paired; either placental or aborted tissue was available. Conclusions The results show that, as long as no animal model is available, the role of HBoV in the occurrence of miscarriage requires additional prospective studies in order to investigate its significance and causal involvements of this pathogen.


| OBJEC TIVE S
The objective of this study was to empirically analyze the percentage of HBoV DNA in placenta and aborted tissue products of conception of miscarriages with unknown etiology.

| Study design
A total number of 172 patients were included. Of those, aborted tissues samples were obtained from 148 patients, placenta samples were obtained from 24 patients. Due to the retrospective design of the study, in none of the cases paired samples were available, thus solely one sample per patient was available. When the miscarriage was clinically reported, the tissues were sent to the pathology lab to be investigated by histology. The sample selection was described in our earlier study in which we screened samples from miscarriage and spontaneous abortion of unknown origin for the presence of Zikavirus RNA. 15 Remaining samples of this earlier study were included in the present investigation.
As control group, we included 19 samples from elective interruptions of pregnancy, of those 10 placenta tissues from sibling births (ie, Caesarean section, healthy newborns) and 9 embryonic tissues which had malformations diagnosed previously by ultrasound examination. The number of included controls was limited due to the fact that the cases rarely present to our hospital.
All samples were available as FFPE specimens. DNA was extracted using the Maxwell FFPE DNA kit (Promega, Darmstadt, Germany). From all samples, qPCRs that detect and differentiate the HBoV strains 1-4 were performed as previously described by Kantola and coworkers 3 with the exception that no quantification was performed because no reliable ratio between copy number and amount of FFPE tissue could be determined. Control plasmids/stan-

| RE SULTS
The results are summarized in Table 1. Of the 172 tissues from 172 patients, 43 (25%) were positive for at least one HBoV strain.
Of those 43 samples, 6 samples were from placenta tissue, specimens from 37 patients were aborted tissue products of conception, of those 10 with histologically confirmed embryonic tissue portions. The gestational age of the aborted tissues ranged from 6 to 15 weeks (mean: 9.78 weeks, median 10 weeks). In none of the samples, any specific histological signs of inflammation were observed, and most samples were categorized as missed abortion.
In none of the cases, fetal tissues were observed by histological investigation, while embryonic tissues were identified by histological analyses in 11 cases. In those cases, in which the placenta tissue was HBoV-positive, placental abruption was observed in 4 cases 5 to 1 weeks before the predicted date of delivery. One case resulted in a stillbirth at gestation week 38 with histologically confirmed signs of fibrosis of the villi, the last case was accompanied by a gestational diabetes.
HBoV-1 was detected in 12 cases, HBoV-3 in 5 cases, and the residual cases were HBoV types 2 and 4 positive. Mixed infections were observed in 3 cases, of those 2 with HBoV 1 and 3, and one with HBoV-1 and 2/4. Positive DNA detections were mutually performed by cloning and sequencing and confirmed the PCR results.
In the control group consisting of placenta samples from otherwise healthy sibling Caesarian sections and embryonic tissues with malformations of non-infectious origin, no samples were tested positive for HBoV.

| D ISCUSS I ON
Since its discovery in 2005, it became more and more evident that the human bocaviruses are serious pathogens 4,16-19 being involved in acute and chronic disease development. The testing of some HBoV research-related hypothesis, however, is still hindered by the lack of an animal model and the restriction of cell culture models to some selected laboratories due to the need of highly specialized cell lines that are not freely available. 20,21 Moreover, it was shown independently by several labs that HBoV is able to persist in the infected host, [9][10][11] which in turn makes it difficult to discriminate between an active contribution to a given clinical situation or a simple coexistence or co-detection due to the viral persistence. Thus, there is still an urgent need to retrospectively address the question in which clinical entities HBoV occurs in order to set up future prospective studies that may allow even these discriminations.
Under these prerequisites, our study shows that HBoV may play a role in the development of abortions, not at least as its two closest relatives, the canine parvovirus and the bovine parvovirus, as well as the human parvovirus B19 are responsible for even this clinical entity in their respective hosts. The fact that despite a positive PCR result no histological signs of the HBoV infection such as cytopathic effects were observed in histology fits to the fact that the virus induces only minor or moderate effects in infected cell culture.
The stroma fibrosis observed in one case also is in agreement with earlier observations that HBoV is able to induce fibrosis in infected organs. 22 However, it has to be stated that the present study cannot demonstrate if the miscarriage was caused by HBoV or was just TA B L E 1 Summary of the major results and cohort characteristics Aborted tissue products of conception control, elective abortions 9 9 0 0% accompanied by the virus, although the presented data increase the likelihood of a causal involvement. Instead, our observations contradict the findings by Enders and colleagues 13 and Riipinen and coworkers. 12 Enders et al tested amniotic liquid for HBoV which may have been less positive as HBoV may not be shed into this liquid but appears to be more tissue bound. Viral shedding is dependent on the three-dimensional structure of the infected tissue 23  Thus, unless an animal model will be available, more prospective in-depth studies are required to elucidate this pathogen's role in miscarriages.

ACK N OWLED G EM ENT
The authors would like to thank Prof. Maria Söderlund-Venermo and Man Xu from the University of Helsinki for providing the detailed PCR protocol and plasmid controls according to their previous publication. 3

CO N FLI C T O F I NTE R E S T S TATE M E NT
None of the authors declares and conflict of interest related to this study.

E TH I C A L S TATEM ENT
All procedures were performed in agreement with the decla-