ABSTRACT
Equid herpesviruses (EHVs) are pathogens of equid and nonequid hosts that can cause disease and fatalities in captivity and in the wild. EHVs establish latent infections but can reactivate, and most EHVs are shed via the nasal passage. Therefore, nasal swabs are generally used for EHV monitoring. However, invasive sampling of wild equids is difficult. While feces is a commonly used substrate for detecting other pathogens, to our knowledge, EHVs have never been detected in feces of naturally infected equids. We systematically tested zebra feces for EHV presence by (i) establishing nested PCR conditions for fecal DNA extracts, (ii) controlling for environmental EHV contamination, and (iii) large-scale testing on a free-ranging zebra population. A dilution minimizing inhibition while maximizing viral DNA concentrations was determined in captive Grévy’s zebra (Equus grevyi) fecal samples from individuals shedding EHV nasally. Sixteen of 42 fecal samples (38%) were EHV positive. To demonstrate that the EHV positivity was not a result of environmental contamination, rectal swabs of wild zebras were screened (n = 18 [Equus quagga and E. zebra]), and 50% were EHV positive, indicating that the source of EHV in feces is likely the intestinal mucosa and not postdefecation contamination. Out of 270 fecal samples of wild zebras, 26% were EHV positive. Quantitative PCRs showed that the amount of virus DNA in feces was not significantly smaller than that in other samples. In summary, fecal sampling facilitates large-scale screening and may be useful to noninvasively investigate phylogenetic EHV diversity in wild and domestic equids.
IMPORTANCE Equid herpesviruses (EHVs) establish latent infections, and many EHVs are shed and transmitted via nasal discharge primarily through droplet and aerosol infection. Obtaining nasal swabs and other invasive samples from wildlife is often not possible without capture and physical restraint of individuals, which are resource intensive and a health risk for the captured animals. Fecal EHV shedding has never been demonstrated for naturally infected equids. We established the conditions for fecal EHV screening, and our results suggest that testing fecal samples is an effective noninvasive approach for monitoring acute EHV shedding in equids.
INTRODUCTION
Equid herpesviruses (EHVs) are common pathogens with a prevalence of over 90% in all global equid populations (1–3) and are responsible for considerable economic losses in the equine industry (4, 5). The known EHVs belong to the family Herpesviridae and group into two of the three known Herpesviridae subfamilies, the Alphaherpesvirinae (e.g., EHV-1, EHV-4, and EHV-9) and the Gammaherpesvirinae (e.g., EHV-2, EHV-5, and EHV-7) (6). Although different viruses from both subfamilies may cause a wide variety of clinical signs during initial infection or reactivation, viruses belonging to the Alphaherpesvirinae subfamily are typically associated with more-severe outcomes (5, 7, 8). Acute EHV infection can cause clinical signs including pharyngitis, pneumonia, pyrexia, lymphadenopathy, abortion, and neuropathies due to acute myeloencephalopathy (3, 5, 9, 10). Furthermore, it has been shown that EHVs can occasionally infect a range of nonequid species in captivity, with potentially fatal outcomes (11–16).
Herpesviruses remain primarily latent in host neural and lymphoid tissues (17). Viral latency is characterized by the absence of lytic viral replication and minimal, if any, viral gene expression, despite the presence of the viral genome in the nucleus of the infected cell (17, 18). However, herpesviruses can be reactivated (17), leading to shedding of infectious virus into the environment. In equids, nasal shedding is predominant and can be accompanied by viremia; thus, testing nasal swabs and invasive samples such as blood is considered the most reliable method for monitoring EHV shedding in horses (3, 10). Sampling wild equids generally requires invasive procedures due to the necessity of physical restraint, which can cause severe stress or risk to animal health for large mammals (19, 20). Moreover, due to the considerable logistic and physical difficulties associated with the capture and immobilization of wild animals, noninvasive sampling would be a preferable method for pathogen screening. For example, for captive wild equids, the swabbing of surfaces such as feed troughs or behavioral enrichment toys has been shown to be an effective approach for surveying EHV shedding (21, 22). However, this approach may not be feasible for wild equids in a (semi)natural environment, due to the lack of surfaces to swab, the difficulty of assigning environmental samples to individuals, and the potentially low encounter rates or lack of interaction with enrichment objects in the natural environment.
Noninvasive pathogen monitoring from feces has been used successfully in wildlife (23–26). However, EHV monitoring has relied mostly on nasal or nasopharyngeal swabs given that nasal secretions are expected to contain herpesviruses during lytic reactivation. To our knowledge, EHVs have never been systematically examined from fecal samples of naturally infected equids, with limited evidence for fecal shedding from experimentally infected foals (27).
In the present study, we evaluated feces as a source for noninvasive EHV monitoring in wild equids by (i) establishing the optimal EHV PCR conditions for fecal DNA extracts, (ii) examining rectal swabs to confirm that fecal EHV DNA originated from the gastrointestinal tract, and (iii) testing the approach on free-ranging zebras. Fecal samples of captive Grévy’s zebras (Equus grevyi) were screened for EHV after opportunistic sampling subsequent to a stressful event which had caused nasal EHV shedding as determined from trough swabbing. EHV screening conditions were determined from these samples to apply to a free-ranging population. The results are described in the context of the effectiveness of fecal samples for noninvasive EHV monitoring.
RESULTS AND DISCUSSION
From fecal samples of captive Grévy’s zebras, PCR using 1.4 μl of undiluted DNA extract produced 3 positive samples, out of 42, and PCR with 0.8 μl of undiluted DNA extract produced 2 positive samples (Table 1). Using DNA standardized to a concentration of 25 ng/μl, 12 of the 42 samples were positive, 1 of which was detected previously, whereas 4 previously positive samples were negative. In total, 16 fecal samples (38%) from captive Grévy’s zebras tested EHV positive, but different amounts of input DNA produced substantial variation in EHV detectability among samples. This may be an effect of PCR-inhibiting substances common in feces (e.g., bile salts, plant secondary metabolites, and complex sugars) (28). Thus, EHV screening from fecal samples may need to be performed at different dilutions in order to obtain the optimal ratio of target DNA to PCR inhibitors, which likely differs among samples. Comparing fecal sampling with feed trough swabs (cumulative result of both concentrations), the same number of positive samples was found with either method (n = 16), although not all sample pairs matched with regard to the day of sampling. This incongruence is likely due to (i) false-negative results, which can be expected with both noninvasive sampling methods, and (ii) a fecal gut passage time of 1 to 2 days (29). For example, infectious particles in feces may originate from swallowed nasal secretions, appearing in feces only with a delay.
EHV-positive and -negative samplesa
Of 18 rectal swabs taken from free-ranging Namibian zebras, 9 (50%) were EHV positive. DNA extracts at 25 ng/μl produced the same number of or more positive results (n = 8) than the PCRs using 0.8 μl or 1.4 μl of a nonstandardized DNA extract (n = 6 and 8 positive results, respectively) (Table 2). Using other sample types (nasal swabs, blood, or tissue), a total of 14 animals were EHV positive (78%). In fecal samples from free-ranging zebras in the Serengeti National Park (n = 270), EHV DNA was detected in 69 samples (26%), using DNA extracts at a concentration of 25 ng/μl.
EHV-positive and -negative PCR results of samples from wild plains zebras and mountain zebras in Namibia from PCRs using different amounts of input DNA of rectal swab DNA extracts and EHVs detected in other substratesa
All recovered EHVs from rectal swabs of wild zebras and from all samples of captive Grévy’s zebras were gammaherpesviruses. For captive zebras, we exclusively detected EHVs most closely related to EHV-7 (92 to 99% identity). In Namibian zebras, the EHVs most closely matched the GenBank entries for EHV-2, asinine herpesvirus 5 (AsHV-5), EHV-5, EHV-7, wild ass herpesvirus (WAH), and Equus zebra herpesvirus (ZHV) (88 to 99% identity) (Table 2). In three cases, the viruses identified in fecal samples did not match those found in other sample types of the same individual (Table 2). Of the 69 PCR products amplified from Serengeti plains zebras, 1 alphaherpesvirus (EHV-1) and 68 gammaherpesviruses (most closely related to EHV-2 [n = 1], EHV-5 [n = 29], AsHV-5 [n = 31], ZHV [n = 6], and EHV-7 [n = 1]) were identified. That almost exclusively gammaherpesviruses were found is in agreement with the absence of clinical signs of disease observed in any of the zebras, as viruses of this subfamily are often reactivated subclinically in equids (3, 9). Pairwise distances between all gammaherpesviruses ranged from 0% to 20% and showed considerable among-virus diversity (see Fig. S1 in the supplemental material). However, due to the short sequence lengths and few phylogenetically informative sites, phylogenetic relationships of EHVs would need to be established by sequencing of fragments substantially larger than those produced here. Sequence data are available from the Mendeley data repository (https://doi.org/10.17632/jm673jrwn3.1).
EHV-1 was quantitated by quantitative PCR (qPCR) (Table 3). The EHV-2 qPCR generated positive results for 1 out of 2 EHV-2-positive samples, and the EHV-5 qPCR produced 8 positive results out of 29 samples previously identified as EHV-5 positive and none in samples identified as AsHV-5 positive (n = 11). The highest genome copy number was found in a nasal swab (51,836 copies/μl DNA extract), followed by rectal swabs and fecal samples (n = 2 each; averages of 4,364 and 2,273 viral genome copies, respectively). Blood and tissue samples (n = 4) produced the lowest copy numbers (on average 112 genome copies/μl each). No significant difference in genome copy numbers was found among sample types using a Kruskal-Wallis test (P = 0.41). The overall lower success rate of the qPCRs than the nested generic herpesvirus PCR is likely due to the nonnested approach and primer specificity for reference EHV sequences, which may fail to amplify when similarity decreases, e.g., as low as 88% (Table 2).
Results for samples previously screened for EHV by EHV-1, EHV-2, and EHV-5 qPCRs
Conventional glycoprotein B (gB) PCR produced fragments of the expected size in 8 out of 28 samples (2 nasal swabs, 3 rectal swabs, and 3 tissue samples). Fecal samples, however, produced only unspecific bands in this PCR; thus, a nested PCR approach would likely produce better results from feces. Furthermore, only viruses previously identified as EHV-2 or EHV-5 produced amplicons, whereas samples containing viruses more closely related to AsHV-5 or ZHV did not. This suggests considerable genetic divergence between EHV-5 and AsHV-5, which likely prevented the qPCR from amplifying AsHV-5 DNA.
EHVs are major pathogens of both domestic and wild equids which can cause severe diseases in equids and nonequids alike. Therefore, monitoring of viral shedding is crucial in a wide variety of settings ranging from domestic horse facilities to zoos. Noninvasive methods to assess pathogen loads have become an increasingly important tool in wildlife research, given the considerable advantages compared to invasive sampling, including the avoidance of the stress of capture and immobilization and the associated logistic efforts and health risks (30). This applies to wildlife both in situ and in captivity. In the captive environment, fecal sampling has the considerable advantage that individual sampling is possible even when animals are not housed separately. Moreover, for wildlife, in situ fecal sampling may be the method of choice for noninvasive monitoring of EHV shedding when invasive procedures are impractical or impossible. Our results demonstrate that fecal sampling is a promising approach for monitoring EHV shedding in wild equids.
The drawback of fecal EHV screening is a potentially higher rate of false-negative results than with invasively obtained samples (including nasal swabs) due to potentially smaller amounts of target DNA and higher concentrations of PCR-inhibiting substances (28). Furthermore, viruses shed in wildlife feces may be insufficiently well characterized to be detected using domestic horse-based assays. However, for wild equid species, obtaining large numbers of invasive samples may be prohibitive such that noninvasive methods may provide a tradeoff between false-negative results with much higher sample numbers and high costs and effort coupled with small sample numbers for invasive collections. Various different EHVs could be retrieved from fecal samples, including those which have been reported relatively rarely in previous studies (e.g., EHV-7, WAH, and ZHV) (31). The use of feces for EHV screening facilitates sampling on a larger scale, which may be useful for investigating the phylogenetic diversity of EHVs in wild and domestic equids. Furthermore, as more viral genomes become available from wildlife, the diagnostics will continue to improve.
Transmission of EHV is assumed to mainly occur directly among animals via nasal discharge (3, 10). However, it has been shown that EHVs can remain infectious in the environment for a considerable length of time (32). Thus, EHV transmission via feces may play a role in pathogen transmission, as equids frequently sniff the feces of conspecifics, and coprophagy may also occur, which could facilitate pathogen spread (33, 34). It remains to be investigated whether EHVs are shed into the intestinal tract or if virus material is swallowed with nasal mucus and passes through the digestive system, which may affect its infectivity after defecation. In the present study, the virus taxa differed between nasal and fecal sampling in three individuals, which may suggest that EHVs found in feces originate from within the intestines rather than from swallowing nasal secretions. Further studies are needed to investigate the diversity and shedding patterns of different viruses within individuals.
MATERIALS AND METHODS
Captive zebras.Fecal samples from captive Grévy’s zebras (n = 4) were collected on a daily basis for 14 consecutive days in July 2017 at Tierpark Berlin, Germany. One of the zebras had been moved to Tierpark Berlin, Germany, from Zoo Mulhouse, France. The move was shown to be associated with nasal EHV shedding in the transported animal and two of the resident zebras (22). Zebras were kept in an outdoor enclosure for most of the day but were separated into individually assigned stalls each day for several hours for feeding of pelleted grain. Individual fecal samples were collected from day 1 to day 14 after translocation. To control for an unequal distribution of EHV DNA in the samples, material from several boluses was pooled. In addition, swabs were collected (dipped in phosphate-buffered saline) from each zebra’s feed trough for screening of nasal EHV shedding (21). Feed troughs were cleaned after swabbing. Fecal samples and swabs were stored frozen until DNA extraction. No clinical signs of acute viral infection, such as excessive nasal or ocular discharge, lethargy, or conjunctivitis, were observed in any of the captive zebras during this study.
This study was approved by the Internal Committee for Ethics and Animal Welfare of the Leibniz Institute for Zoo and Wildlife Research (approval no. 2016-09-04). All protocols adhered to the laws and guidelines of Namibia and Germany, respectively. Permission to conduct research in Namibia was granted by the Ministry of Environment and Tourism (MET) (permit no. 2094/2016). Permission to export sample material from Namibia was granted by an MET export permit (no. 105336), and samples were transported to Germany in full compliance with the Convention on International Trade in Endangered Species (CITES) (permit no. 108448) and in compliance with the Nagoya Protocol on Access to Genetic Resources. Permission to conduct research in Tanzania was granted by the Tanzania Commission of Science and Technology and the Tanzania Wildlife Research Institute (TAWIRI) (permit no. 2094/2016). Permission to export sample material from Tanzania was granted by a TAWIRI export permit (reference no. TWRI/RS-85/VOL.IV/86/128).
Swabs and tissue samples of wild zebras.Fecal samples may potentially be contaminated by nasal exudate when animals that are actively shedding virus sniff defecation piles of their conspecifics. Thus, to exclude potential environmental contamination, rectal swabs were collected. Rectal and nasal swabs (Mini-UTM kit; Copan Diagnostics Inc., Murrieta, CA) and blood samples (12 ml in EDTA) of free-ranging mountain zebras (n = 11) and plains zebras (n = 7) in Etosha National Park and Khomas region, Namibia, were collected after immobilization (35). Furthermore, swabs, blood, and tissue samples from mountain zebras which had been hunted for meat production were collected (stored in RNAlater). All samples were stored frozen until DNA extraction. No clinical signs were observed in any of the zebras.
Fecal samples of wild zebras.In order to determine the occurrence of fecal shedding of EHV in an undisturbed population of zebras on a larger scale, we collected fecal samples from free-ranging plains zebras (n = 270) in the Serengeti National Park, Tanzania, during the months of January to March, May to July, and October 2016 (36). Samples were collected opportunistically immediately after defecation, with care being taken to avoid soil contamination. To control for an unequal distribution of virus DNA in the boluses, material was collected from several remote parts of each bolus and then pooled and thoroughly mixed. Samples were stored in RNAlater (Sigma-Aldrich, Taufkirchen, Germany) in the field and frozen for shipment and further storage until DNA extraction.
EHV screening.DNA was extracted by using commercially available kits (NucleoSpin tissue kit [Macherey-Nagel, Düren, Germany] for all swabs and tissue samples and NucleoSpin soil kit [Macherey-Nagel] for feces) according to the manufacturer’s instructions, using 400 to 500 mg feces. DNA extracts were eluted in a volume of 100 μl, and the total DNA concentration was measured on an Agilent 2200 TapeStation (Agilent Technologies, Waldbronn, Germany), using genomic screen tapes. Concentrations ranged from 26.4 to 85.7 ng/μl. We made every effort to avoid contamination, including separate UV-illuminated hoods for extraction and PCR setup.
A nested PCR for herpesviruses targeting the DNA polymerase gene was performed as described previously (21), with a total reaction mixture volume of 22 μl. To reduce PCR inhibition (by, e.g., fecal bile salts and complex sugars), 0.6 μl of bovine serum albumin (BSA) was added per reaction for all sample types and concentrations. DNA extracts of feces of captive zebras and of rectal swabs from wild zebras were PCR screened using three different amounts of input DNA: (i) 1.4 μl of undiluted DNA extracts, (ii) 0.8 μl of undiluted DNA extracts, and (iii) 1.4 μl of DNA extracts standardized to a concentration of 25 ng/μl. The nasal swabs, tissues, blood, and feed trough swabs were screened with 1.4 and 0.8 μl of undiluted DNA extracts. Fecal samples of Serengeti plains zebras were screened using 1.4 μl of DNA extracts standardized to a concentration of 25 ng/μl. PCR products were visualized on a 1.5% agarose gel. Bands of the expected product size (225 bp) were excised from the gel and purified using a commercial kit (NucleoSpin gel and PCR cleanup; Macherey-Nagel), according to the manufacturer’s instructions. Purified PCR products were Sanger sequenced by LGC Genomics GmbH, Berlin, Germany. Virus sequences were queried against GenBank (37) using BLAST.
Quantitative PCR for detection of EHV-1, EHV-2, and EHV-5.To determine the quantity of EHV present in DNA extracts of swabs, tissue, and fecal samples, qPCR was carried out. The qPCR was performed on a randomly selected subset of samples that had previously tested positive by conventional nested PCR screening (n = 12 fecal samples, n = 8 rectal swabs, n = 11 nasal swabs, and n = 12 blood and tissue samples). All qPCRs for EHV-1, -2, and -5 were TaqMan based, and the reaction mixtures contained their given primers, probes, 10 μl of the SensiFAST Probe Lo-ROX DNA polymerase (Bioline, Germany), and 5 μl of the DNA extract standardized to a concentration of 25 ng/μl. All qPCRs were carried out in 96-well microtiter plates using the 7500 Fast real-time PCR system (Applied Biosystems, CA, USA) under the following cycling conditions: 95°C for 2 min, followed by 40 cycles of 95°C for 3 to 10 s and 60°C for 30 s and a final hold at 60°C for 1 min. Each sample was run in triplicate.
For EHV-1 qPCR, we amplified a 106-bp sequence of the glycoprotein B (gB) gene (GenBank accession no. M36298), as previously described (38). The reaction mixture contained 100 nM the fluorogenic TaqMan probe FAM (6-carboxyfluorescein)-TGA GAC CGA AGA TCT CCT CCA CCG A-BHQ1 (black hole quencher 1) and 450 nM forward primer 5′-CAT ACG TCC CTG TCC GAC AGA T-3′ and reverse primer 5′-GGT ACT CGG CCT TTG ACG AA-3′. EHV-1 DNA was quantified by regression to the slope of the curve from serial dilutions of isolated DNA from EHV-1 strain Ab4 cloned as a bacterial artificial chromosome (BAC) (39). EHV-2 qPCR amplified a 174-bp sequence of the gB gene (GenBank accession no. NP_042604); the protocol was adapted from methods reported previously (40). The EHV-2 reaction mixture contained 100 nM the fluorogenic TaqMan probe FAM-TGA CAT ACC CAC CCT ACA CAC CAT AG-BHQ1 and 200 nM forward primer 5′-AGG ACT ACT ACT ATG TCA G-3′ and reverse primer 5′-ATG GTC TCG ATG TCA AAC AC-3′. The EHV-2 DNA was then quantitated by regression to the slope of a standard curve from serial dilutions of the 174-bp EHV-2 gBlock gene fragment generated from the gB gene of EHV-2. For EHV-5 qPCR, we amplified a 297-bp sequence of the gB gene (GenBank accession no. NC_026421.1); the protocol was adapted from methods described previously (41). For the EHV-5 reaction, 100 nM the fluorogenic TaqMan probe FAM-TCC ATC CAC GAT GGC AGG GA-BHQ1 and 200 nM forward primer 5′-ATG AAC CTG ACA GAT GTG CC-3′ and reverse primer 5′-CAC GTT CAC TAT CAC GTC GC-3′ were used. The EHV-5 DNA was then quantitated by regression to the slope of a standard curve from serial dilutions of the 297-bp EHV-2 gBlock gene fragment generated from the gB gene of EHV-5.
Glycoprotein B PCR.Several samples were characterized as AsHV-5/EHV-5 by sequencing but could not be detected by EHV-5 qPCR. In order to determine potential phylogenetic diversity between EHV-5 and these viruses, a gB fragment was amplified from a subset of samples. Primers were designed to amplify a 641-bp fragment of the gB gene of EHV-2 and EHV-5 and to include the region complementary to the qPCR probe (forward primer 5′-CAC CAG CGT CAT GAG CGC CA-3′ and reverse primer 5′-AAC ACC CCG CTG GCC ACG TT-3′) and were used under the following PCR conditions: an initial denaturation step at 95°C for 2 min, followed by 35 cycles of 95°C for 20 s, 63°C for 20 s, and 72°C for 40 s and a final elongation step at 72°C for 2 min. Purified PCR products were Sanger sequenced.
ACKNOWLEDGMENTS
This study was supported by a grant from the Leibniz Gemeinschaft (SAW-2015-IZW-1 440) and the Leibniz Institute for Zoo and Wildlife Research.
We thank the animal keepers of Tierpark Berlin for their assistance with sample collection. We thank the Ministry of Environment and Tourism (MET), Namibia; the Tanzania Commission of Science and Technology; and the Tanzania Wildlife Research Institute (TAWIRI) for their research permissions. For assistance with sample collection in Namibia, we thank Carl-Heinz Moeller and Cheri Morkel.
FOOTNOTES
- Received 13 September 2018.
- Accepted 9 November 2018.
- Accepted manuscript posted online 16 November 2018.
Supplemental material for this article may be found at https://doi.org/10.1128/AEM.02234-18.
- Copyright © 2019 American Society for Microbiology.
