The authors have declared that no competing interests exist.
A study was conducted to investigate the serum virome of sows with and without stillbirths after farrowing. Sera from sows with at least one stillbirth or with normal litters were collected immediately after farrowing. Viral DNA was extracted from serum pools and submitted to high throughput sequencing. No differences in the proportion of virus-related reads were found in both groups (p > 0.05). A variety of viral DNA genomes were identified, mostly representative of three viral families:
In swine reproduction, the term “stillbirth” refers to piglets with healthy appearance born dead at term. The average incidence of stillbirths in commercial farms may range between 3% to 8%, causing significant losses to swine producers [
To date, most studies focusing on viruses with potential association with stillbirths have used traditional approaches, such as serological tests on samples from sows or stillborn piglets. More recently, the polymerase chain reaction (PCR) has been employed in this kind of studies [
This study was approved by the Ethics Committee in the Use of Animals from the Veterinary Research Institute
Just farrowed sows from six commercial piglet producing farms located in five municipalities in Southern Brazil were selected. The sows were derived from farms certified as free of classical swine fever, pseudorabies, brucellosis, tuberculosis and mange. All sows were vaccinated to porcine circovirus 2, porcine parvovirus (i.e. ungulate protoparvovirus),
After centrifugation (~ 5,000 x g/5 min), serum samples were pooled and filtered through 0.45 and 0.22 μm filters (Millex-GV, PVDF, Millex®). These were then centrifuged on a sucrose cushion (25%) at an average 150,000 x g for 4 h at 4 °C. The obtained pellets were resuspended in 400 μL of milli-Q water and stored at -80 °C until further processing. Subsequently, the resuspended pellets were treated with DNase (2 U/μL, Turbo DNase Kit, Ambion) and RNase A (20 mg/mL, Invitrogen) to reduce the concentration of non-encapsidated nucleic acids [
The quality of the obtained reads was verified using FastQC software (
The number of raw reads matching selected viral sequences (contigs and concatenates previously classified by BLAST) were measured using the Geneious assembly tool, with all raw data output reads in the “low-sensitivity/fastest” mode. Virus-specific reads from each pooled sample were then normalized by the total number of viral reads generated in each pool. Differences between the data obtained from pools of sows with and without stillbirths were compared using the Wilcoxon T test. A
Phylogenetic analyses were performed based on predicted amino acid sequences (for
The viral genomes described here were deposited in GenBank under accession numbers MH170056 to MH170073.
The twelve pools generated a total of 7,918,254 reads (
Reads related to two eukaryotic viral families with DNA genomes (
The number of reads corresponding to each family was normalized by the number of eukaryotic viral reads. Dark blue columns represent the average reads obtained in each of the groups. The numbers followed by a letter in parenthesis correspond to the identification of the pools analyzed in this study. Numbers correspond to each of the farms (1–6); letters S (stillbirth) or H (healthy) refer with the occurrence of stillbirths. Reads related of
The second most frequently detected reads were those of CRESS DNA viruses (
The third most frequently detected reads were representatives of the
When comparing the viral serum profile on both groups (with and without stillbirths), no statistically significant differences (p > 0.05) were observed (
The first two principal components (PC1 and PC2) were maintained since they explained more than 62% of the observed variability. Black dots represent the pools of sera from sows with stillbirths (S); white diamonds represent the pools of sera from sows without stillbirths (H).
Virus | Stillbirths | Healthy | p-value | ||
---|---|---|---|---|---|
Median | (Min—Max) | Median | (Min—Max) | ||
9,46E-01 | (2,50E-03–1,00E+00) | 9,40E-01 | (3,20E-03–9,98E-01) | 0.92 | |
TTSuV 1b | 0,00E+00 | (0,00E+00–2,24E-01) | 0,00E+00 | (0,00E+00–9,00E-04) | 0.65 |
TTSuV 1a | 8,70E-01 | (2,50E-03–1,00E+00) | 9,40E-01 | (3,20E-03–9,98E-01) | 0.75 |
0,00E+00 | (0,00E+00–1,39E-01) | 0,00E+00 | (0,00E+00–1,00E-04) | 0.18 | |
Porcine circovirus 3 | 0,00E+00 | (0,00E+00–1,39E-01) | 0,00E+00 | (0,00E+00–0,00E+00) | 0.18 |
Porcine circovirus 1 | 0,00E+00 | (0,00E+00–2,00E-05) | 0,00E+00 | (0,00E+00–1,00E-04) | 0.32 |
CRESS DNA | 7,00E-03 | (0,00E+00–9,96E-01) | 6,10E-02 | (1,80E-03–8,89E-01) | 0.60 |
Rat stool associated circular virus | 0,00E+00 | (0,00E+00–2,93E-02) | 4,00E-03 | (0,00E+00–4,87E-02) | 0.50 |
Porcine stool associated circular virus | 0,00E+00 | (0,00E+00–9,91E-02) | 0,00E+00 | (0,00E+00–0,00E+00) | 0.18 |
Porcine serum associated circular virus | 1,50E-03 | (0,00E+00–3,85E-02) | 6,00E-04 | (0,00E+00–2,30E-03) | 0.14 |
Po-circo-like virus | 5,00E-04 | (0,00E+00–2,88E-01) | 3,00E-02 | (0,00E+00–8,89E-01) | 0.17 |
Odonata associated circular virus | 0,00E+00 | (0,00E+00–3,60E-01) | 0,00E+00 | (0,00E+00–2,00E-03) | 0.65 |
Hudisavirus | 0,00E+00 | (0,00E+00–1,52E-02) | 0,00E+00 | (0,00E+00–0,00E+00) | 0.32 |
Duck faeces associated circular virus | 5,00E-04 | (0,00E+00–8,92E-02) | 2,00E-04 | (0,00E+00–5,80E-03) | 0.69 |
Dromedary stool associated circular virus | 0,00E+00 | (0,00E+00–1,80E-03) | 0,00E+00 | (0,00E+00–1,80E-03) | 0.32 |
Unclassified CRESS DNA viruses | 0,00E+00 | (0,00E+00–1,95E-01) | 0,00E+00 | (0,00E+00–0,00E+00) | 0.18 |
No statistically significant difference was detected between the groups (p>0.05); the number of viral reads were divided by the total number of reads for each pool.
The viruses which are often associated with stillbirths, i.e. UPV1, PCV2, PRRVS, EMCV, CSFV and Influenza virus, were not detected in this study. It must be pointed that the sows were vaccinated to UPV1 and PCV2; therefore, genomes corresponding to these agents were not expected to be detected. Viral RNA genomes were not obtained since they could not be recovered from the samples. This might have been caused either by the true absence of viral RNA genomes in samples, or by its presence in such small quantities that it could have been missed by the extraction procedure. However, the protocol employed here is well stablished in the laboratory and has demonstrated efficiency at viral RNA extraction [
It must should be considered that HTS does not provide an actual quantification of the detected viromes. Biases such as those introduced by the enrichment method (MDA) must be born in mind, although the impact of MDA on HTS has been lessened and regarded as not significant enough to lead to data misinterpretation [
DNA viruses which are often associated with stillbirths, as UPV1 and PCV2, were not detected in this study. It must be pointed that the sows were vaccinated to UPV1 and PCV2; therefore, genomes corresponding to these agents were not expected to be detected.
It is important to highlight that, in the present study, the samples were examined in pools, which may hamper the detection of genomes present in low concentrations. It is possible that sequencing individual samples might provide some additional information. In addition, only viral genomes matching the GenBank virus database were evaluated here. Most of the reads (~ 72%) could not be related to previously published viral sequences; these may represent new genomes of viruses, as yet unknown, infectious agents. Such sequences should be subject of further investigation.
The findings reported here reveal a number of viral genomes circulating in the serum of sows at farrowing. These strongly suggest that there is no association between the presence of such viral genomes in sow’s sera at farrowing and the occurrence of stillbirths. Nevertheless, it is still possible that an as yet undetected virus (or other infectious agents), could putatively be associated to stillbirths, though just not present in sow’s sera at farrowing. Viremia is expected to precede transplacental infection [
Altogether, a total of 20 complete viral genomes were recovered from sows either with or without stillbirths, including twelve TTSuV, six CRESS DNA viruses and two PCV3 genomes. PCV3 genomes were recovered from two different pools of sows with stillbirths (farms 1 and 3) and were described elsewhere [
Next, a brief report is provided on the full viral genomes detected in this study.
All twelve genomes representative of members of the
(A) Schematic representation of one out of the twelve TTSuV genome recovered in this study. (B) Phylogenetic analysis based on the entire ORF1 at nucleotide level. Neighbor-joining method with p-distance and a bootstrap of 1000 replicates. Drawings of pigs highlight the sequences recovered in this study.
Sequence name | Genus | Accession number | Length | ||
---|---|---|---|---|---|
TTSuV 1a RS/1 | MH170062 | 2,910 | 7444.3 | 1H | |
TTSuV 1b RS/2 | MH170063 | 2,930 | 7305.2 | 2H | |
TTSuV 1b RS/3 | MH170064 | 2,904 | 3761.8 | 3H | |
TTSuV 1a RS/4 | MH170065 | 2,803 | 2198.6 | 4H | |
TTSuV 1b RS/5 | MH170066 | 2,906 | 1388.2 | 5H | |
TTSuV 1a RS/6 | MH170067 | 2,882 | 3326.6 | 1S | |
TTSuV 1b RS/7 | MH170068 | 2,891 | 1399.0 | 2S | |
TTSuV 1b RS/10 | MH170069 | 2,887 | 1207.3 | 5S | |
TTSuV 1b RS/12/A | MH170070 | 2,868 | 6558.7 | 6S | |
TTSuV 1a RS/12/B | MH170071 | 2,910 | 6455.1 | 6S | |
TTSuV k2a RS/2 | MH170072 | 2,901 | 15.3 | 2H | |
TTSuV k2a RS/8 | MH170073 | 2,826 | 2270.6 | 3S |
aGenome coverage = average of nucleotides/site; obtained by mapping raw reads to reference.
bFarms sampled are numbered (1 to 6). Letters refer to groups with (S) or without (H) stillbirths; e.g.; 1S = farm 1, sows with stillbirths; 1H = farm 1, sows without stillbirths.
Based on the criteria for TTSuV classification from ICTV (2015), multiple alignment analyses of the entire ORF1 were performed. Phylogenetic analyses were performed based on the entire ORF1. The twelve TTSuV ORF1 sequences reported here were compared to 40 previously available TTSuV sequences. In such phylogenetic reconstruction, six of the ORF1 sequences identified here clustered along with TTSuV 1b; four clustered with TTSuV 1a, whereas two clustered along with TTSuV k2a (
The TTSuV recovered here share 36.6% to 84.7% nucleotide identity, revealing two distinct genera. The TTSuV 1a sequences share 64.1–82.5% nucleotide identity, whereas the TTSuV 1b sequences shared 65.1–84.7% nucleotide identity in relation to each other. The two TTSuV k2a sequences exhibited a nucleotide sequence identity of 71.5%.
Six complete CRESS DNA genomes were recovered from four different serum pools, collected in three of the farms (
Accession number | Acronym | ORF | Identity | E-value | Cover | |||
---|---|---|---|---|---|---|---|---|
MH170056 | 2H | 10.5 | Rep | Duck faeces associated circular DNA virus 2 (NC_030133) | 86% | 0.0 | 99% | |
Cap | Porcine serum-associated circular virus (KU203356) | 53% | 1E-160 | 100% | ||||
MH170057 | PoSCV-2 2B RS/BR | 2H | 47 | Rep | Odonata-associated circular virus -17 (KM598400) | 63% | 2E-108 | 92% |
Cap | Odonata-associated circular virus -17 (KM598400) | 34% | 2E-15 | 75% | ||||
MH170058 | PoSCV-3 7A RS/BR | 2S | 123.7 | Rep | Duck faeces associated circular DNA virus 2 (NC_030133) | 86% | 0.0 | 96% |
Cap | Duck faeces associated circular DNA virus 3 (NC_030134) | 67% | 1E-180 | 100% | ||||
MH170059 | PoSCV-4 7B RS/BR | 2S | 27.3 | Rep | Hudisavirus sp. (MF351519) | 89% | 0.0 | 98% |
Cap | Hudisavirus sp. (MG522858) | 78% | 1E-158 | 87% | ||||
MH170060 | PoSCV-5 8 RS/BR | 3S | 142.9 | Rep | Porcine serum-associated circular virus (KU203352) | 94% | 0.0 | 100% |
Cap | Porcine stool-associated circular virus 7 (KJ577814) | 57% | 2E-137 | 100% | ||||
MH170061 | Porcine associated porprismacovirus 3, 12/RS/BR | 6S | 37.3 | Rep | Porcine stool-associated circular virus 2 (KJ577818) | 85% | 0.0 | 100% |
Cap | Porcine stool-associated circular virus 3 (LC133374) | 93% | 0.0 | 100% |
aPoSCV = Porcine serum associated circular DNA virus.
bThe numbers refer to the farms; letters (“S” for stillbirths; “H” for healthy) refer to group; e.g.; 2S = farm 2, sows with stillbirths; 2H = farm 2, sows without stillbirths.
cGenome coverage = average of nucleotides/site; obtained by mapping raw reads to reference.
dBased on BLASTp search.
The CRESS DNA genomes recovered here ranged in length from 1,846 to 2,568 nt, with two major ORFs that encode the putative Rep and Cap proteins. Four types of genome architectures previously described [
Phylogenetic tree was inferred using PhyML with the VT + G + I + F substitution model. Sequences recovered in this study are highlighted by pig drawings. The number represents the farm and the letter the group; e.g. 2S = farm 2, sows with stillbirths; 2H = farm 2, sows without stillbirths.
PoSCV-2 2B/RS/BR and PoSCV-4 7B/RS/BR exhibited a conserved nonanucleotide motif (NANTATTAC), while PoSCV-1 2A/RS/BR, PoSCV-3 7A/RS/BR and PoSCV-5 8/RS/BR displayed less common nonamers (
Genome | Nanonucleotide motif | RCR motifs | SF3 helicase motifs | ||||
---|---|---|---|---|---|---|---|
I | II | III | Walker-A | Walker-B | Motif C | ||
PoSCV-1 2A/RS/BR | TAATATTAA | LLTYPQ | EHYHA | YVKK | GKTGTGKT | IFDDI | IFTS |
PoSCV-2 2B/RS/BR | TAGTATTAC | - | KHIHI | YIEK | GDSGKGKT | VIEEF | YTCN |
PoSCV-3 7A/RS/BR | AAATATTAA | LLTYPQ | EHYHA | YVKK | GATGTGKT | IFDDI | IFTS |
PoSCV-4 7B/RS/BR | CAGTATTAC | CFTINN | PHYQG | YCRK | APPGTGKS | VFEEF | ITSN |
PoSCV-5 8/RS/BR | AGCAAGCAA | - | RHYQF | YVYK | EKGNSGKT | IIDTP | ILCN |
Porcine associated porprismacovirus 3 12/RS/BR | TAGTATTAC | MATIPH | KHIQC | YEKK | SEGNSGKT | IIIDI | CMTN |
PoSCV-1 2A/RS/BR and PoSCV-3 7A/RS/BR were most closely to sequences of CRESS DNA detected in serum of Brazilian pigs (KU203355, KU203356) and in duck feces in New Zealand (NC030133, NC030134) (
The PoSCV-2 2B/RS/BR genome showed no significant nucleotide identity to any of the previously known CRESS DNA genomes available on GenBank database. Despite this, PoSCV-2 genome contains characteristics identified in others CRESS DNA viruses. This genome of 1,900 nt in length is bidirectionally organized, encoding a putative Cap on the virion-sense strand and a putative Rep on the complementary strand. The large intergenic region (181 nt) contains a predicted stem-loop located at the 5’ end of Rep, with a conserved nonanucleotide at the apex (TAGTATTAC) (
Another unclassified CRESS DNA genome, named PoSCV-4 7B/RS/BR shares ~78% nucleotide identity with hudisavirus (MF351515). These viruses were recently discovered in samples of human diarrheal feces in Peruvian patients [
The recently proposed family of CRESS DNA viruses,
This study provides a picture of DNA viruses present in sera of just farrowed sows with or without cases of stillbirths, in which genomes of anelloviruses, circoviruses and CRESS DNA viruses were the most abundantly detected. No association could be established between any of the viral genomes here identified and the occurrence of stillbirths. Nevertheless, a number of novel viral genomes were identified; these seem to be part of the normal virome in the serum of farrowing sows.
Farm and city of origin of sows, occurrence/non-occurrence of stillbirths, and numbers of sows sampled in each pool are shown.
(DOCX)
Viral eukaryotic reads and bacteriophage reads were normalized by the number of viral reads in each pool.
(DOCX)
The authors are grateful to Brazilian veterinary practitioners and farm owners who helped to collect the samples and farm specific information.