Environmental sampling for the detection of foot-and-mouth disease virus and peste des petits ruminants virus in a live goat market, Nepal
Abstract
Livestock markets are considered vital parts of the agricultural economy, particularly in developing countries where livestock keeping contributes to both food security and economic stability. Animals from diverse sources are moved to markets, they mix while they are there and are subsequently redistributed over wide geographic areas. Consequently, markets provide an opportunity for targeted surveillance for circulating pathogens. This study investigated the use of environmental sampling at a live goat market in Nepal for the detection of foot-and-mouth disease virus (FMDV) and peste des petits ruminants virus (PPRV), both of which are endemic. Five visits to the market were carried out between November 2016 and April 2018, with FMDV RNA detected on four visits and PPRV RNA detected on all five visits. Overall, 4.1% of samples (nine out of 217) were positive for FMDV RNA and 60.8% (132 out of 217) were positive for PPRV RNA, though the proportion of positive samples varied amongst visits. These results demonstrate that non-invasive, environmental sampling methods have the potential to be used to detect circulation of high priority livestock diseases at a live animal market and, hence, to contribute to their surveillance and control.
1 INTRODUCTION
In developing countries, livestock contributes to both food security and economic stability for small holder farmers. In Nepal, around 65% of the population are involved in the agricultural sector, with livestock rearing remaining a significant part of income generation (Ministry of Agriculture and Livestock Development; Government of Nepal, 2016). In developing countries, such as Nepal, livestock markets play a key role in the sale and supply of animals. However, while movement of animals is an intrinsic part of the agricultural economy, it is also a significant factor in both the spread and control of disease (Fèvre et al., 2006). Markets which bring together significant numbers of animals from across large geographic regions can serve as hotspots for transmission and dispersal of multiple infectious diseases (Motta et al., 2019; Robinson & Christley, 2007; Vallée et al., 2013).
With a large proportion of the population relying on livestock as a source of income, endemic diseases such as foot-and-mouth disease (FMD) and peste des petits ruminants (PPR) can have a considerable economic impact (Diallo, 2006; Knight-Jones & Rushton, 2013). Both PPR and FMD are important transboundary animal diseases that are endemic in Nepal. PPR primarily affects small ruminants, with typical signs of infection including fever, nasal and ocular discharge, lesions in the mouth, diarrhoea and, in severe cases, pneumonia (Parida et al., 2015). FMD infects cloven hooved livestock species such as cattle, pigs, sheep and goats (Alexandersen et al., 2003). Clinical signs in goats are often mild and include fever, vesicles in the mouth or on the feet and lameness. PPRV exists as a single serotype with four genetically different lineages (Senthil Kumar et al., 2014), whereas FMD virus (FMDV) exists as seven immunologically distinct serotypes (Mahapatra & Parida, 2018).
Employing surveillance at live animal markets has the potential to provide information on both the movements of animals and the pathogens that they carry (Fournié & Pfeiffer, 2013). Sampling at live bird markets across Asia has been useful in investigating the prevalence of avian influenza (Chen et al., 2014), and use of environmental sampling has been an efficient way of monitoring for presence of these viruses (Vergne et al., 2019). Environmental sampling has the advantage of being low cost in terms of sample collection and requires little prior knowledge of diseases or handling of animals. The aim of this study was to investigate the use of non-invasive, environmental sampling methods (Brown et al., 2021; Colenutt et al., 2018) to detect relevant livestock viruses at a market that serves as a focal point for the movement, sale and distribution of goats within Nepal.
2 METHODS
2.1 Sampling site
Five visits to the Khasibazar goat market in Kathmandu, Nepal, were made between November 2016 and April 2018 (Table 1). Environmental sampling of pens holding live goats (Figure 1) was carried out with permission of traders, facilitated by a representative of the Ministry of Agriculture and Livestock Development, Nepal. The goats at the market were not systematically examined at each visit, but goats with nasal discharge were observed on several occasions. The market is managed by a union of traders who pay an annual rent for use of the area. Each individual trader pays a monthly rent for use of dedicated selling pens. Market days are usually Sunday and Thursday, although animals arrive at the market throughout the week, adding to the stock already in pens, and can remain in the market for up to 5 days. According to verbal reports from the traders, around 80% of the traded goats are from India, mainly from the states Uttar Pradesh and Bihar which both neighbour Nepal. The remaining 20% of goats are from regions within Nepal. Written records of origins and destinations of traded animals were not kept at the time of the study. Goats are sold for a mixture of local consumption and retention with few being dispersed longer distances in Nepal. Even those consumed locally may have contact with resident household goats before slaughter. There were no reported requirements for use of vaccinations and no veterinary inspections relating to the health of animals were carried out at the markets at the time of the study. No cleaning of pens, aside from removal of faeces, or biosecurity measures to prevent mixing of goats from different locations were reported by traders.
FMDV | PPRV | |||||||
---|---|---|---|---|---|---|---|---|
Visit | Pena | Number of samples | Number of positive | % positive | CT valuesb | Number of positive | % positive | CT valuesb |
November 2016 | 1 | 10 | 1 | 10.0 | 38.6 | 6 | 60.0 | 24.6–32.2 |
2 | 12 | 0 | 0.0 | 4 | 33.3 | |||
3 | 10 | 0 | 0.0 | 0 | 0.0 | |||
April 2017 | 1 | 8 | 0 | 0.0 | 34.9 | 4 | 50.0 | 30.0–34.8 |
2 | 8 | 0 | 0.0 | 7 | 87.5 | |||
3 | 8 | 0 | 0.0 | 7 | 87.5 | |||
4 | 8 | 1 | 12.5 | 8 | 100.0 | |||
5 | 8 | 0 | 0.0 | 7 | 87.5 | |||
6 | 3 | 0 | 0.0 | 3 | 100.0 | |||
November 2017 (A) | 1 | 15 | 0 | 0.0 | – | 9 | 60.0 | 26.9–35.0 |
2 | 15 | 0 | 0.0 | 10 | 66.7 | |||
3 | 13 | 0 | 0.0 | 10 | 76.9 | |||
4 | 15 | 0 | 0.0 | 2 | 13.3 | |||
November 2017 (B) | 1 | 12 | 0 | 0.0 | 35.1–35.4 | 9 | 75.0 | 29.4–34.9 |
2 | 14 | 2 | 14.3 | 11 | 78.6 | |||
3 | 14 | 0 | 0.0 | 10 | 71.4 | |||
4 | 14 | 0 | 0.0 | 8 | 57.1 | |||
April 2018 | 1 | 30 | 5 | 16.7 | 33.8–38.8 | 17 | 56.7 | 29.4–34.9 |
Total | 217 | 9 | 4.1 | 33.8–38.8 | 132 | 60.8 | 24.6–35.0 |
- a Pen numbers indicate different pens sampled during the same visit and not the same pens sampled across visits.
- b CT values are given as ranges where multiple samples were positive.

2.2 Sample collection
Electrostatic dust cloths were used to swab surfaces that were deemed most likely to have come into contact with excretions and secretions from livestock, for example, feed buckets, troughs and fencing (Figure 1). No sampling or handling of live animals was necessary for this study. After swabbing a surface, cloths were placed inside in a screw top tube. Either immediately or within 3 h, material was eluted by the addition of 5 ml of impinger fluid [Glasgow minimum essential medium (Gibco, UK) with antibiotics (penicillin–streptomycin and amphotericin B {Gibco, UK}), 5% bovine serum albumin (Sigma–Aldrich, UK) and 1 M HEPES (Gibco)] and the contents were shaken to fully saturate the cloth with the medium. A disposable wooden spatula was used to remove the cloth, and at the same time, it was pressed to extract as much medium as possible. An aliquot of the medium was then added directly to a guanidinium thiocyanate-based lysis buffer (catalogue number AM8500; Thermo Fisher Scientific, UK) at a ratio of 1:2.6. Samples were then stored at 4°C for up to 33 weeks before transport to The Pirbright Institute (UK) for testing by real-time reverse transcription-PCR (rRT-PCR).
2.3 Sample processing
Viral RNA was extracted from samples using the Kingfisher Flex automated extraction platform (Thermo Fisher Scientific) with the MagMAX viral RNA isolation kit (Thermo Fisher Scientific). FMDV RNA was detected by rRT-PCR on the ABI 7500 system (Applied Biosystems) using a PCR that targets the polymerase 3D region of the FMDV genome (Callahan et al., 2002). A serotype-specific rRT-PCR assay targeting the VP1 region of the FMDV genome (Knowles et al., 2016a) and Sanger dideoxy-sequencing methods for determining VP1 sequence (Knowles et al., 2016b) were used for further analysis of samples positive in the 3D assay. PPRV RNA was detected by rRT-PCR using an assay that targets the nucleocapsid (N) protein gene of the PPRV genome following the method as described previously (Batten et al., 2011), except 9 µl of RNA was added in place of 3 µl to maximize detection. All samples were tested at least in duplicate and the average of the two CT values were used in subsequent analysis. Five selected samples that were positive for PPRV in rRT-PCR were further analyzed in a gel-based RT-PCR to amplify the C-terminus of the nucleocapsid (N)-gene with the aim to generate nucleotide sequence for lineage determination as per methods described previously (Baazizi et al., 2017).
3 RESULTS
3.1 Goat market
Positive detections of both FMDV and PPRV RNA were made from environmental samples collected at the Khasibazar goat market using rRT-PCR (Table 1 and Data S1), with 4.1% of samples (nine out of 217) positive for FMDV RNA and 60.8% of samples (132 out of 217) positive for PPRV RNA. FMDV RNA was detected in four out of five visits and PPRV RNA was detected in every visit. Of the 18 pens tested, four were positive for FMDV RNA, while all the pens were positive for PPRV RNA except pen three in the November 2016 sample collection (Table 1). Due to collection and storage methods used with these samples, assays to measure infectious virus were not possible, so all results refer to detection of viral genome rather than infectious virus.
3.2 Sample analysis
Samples that were positive for FMDV RNA in rRT-PCR were also tested using a serotype-specific rRT-PCR assay, though no positive results were obtained. Amplification and sequencing of VP1 from the FMDV genome was also unsuccessful for samples collected at the goat market, although the method has been successfully used with an environmental sample collected at a farm with an active outbreak of FMDV from a companion study (Colenutt et al., 2018).
Five samples positive for PPRV RNA were selected for lineage identification using a diagnostic gel-based PCR for amplification of the C-terminus of the N-gene. Amplification of a correct size product (of around 350 bp) was observed in only one sample (Figure S1). Attempts to sequence the PCR product were not successful indicating a low viral load and/or poor quality of RNA obtained from the environmental samples.
4 DISCUSSION
In this study, we have demonstrated the ability to detect FMDV and PPRV, two high priority livestock diseases (Knight-Jones & Rushton et al., 2013; OIE & FAO, 2015), using non-invasive, environmental sampling methods at a live animal market. Use of surveillance at livestock markets has the potential to provide information on pathogens that are present within a particular animal population, supplementing information from traditional surveillance approaches (Bates et al., 2003; Ranabijuli et al., 2010). Improving surveillance for livestock diseases in endemic regions is crucial to support control plans, particularly where eradication is a long-term goal (FAO, 2011; OIE & FAO, 2015).
FMDV and PPRV are both shed in secretions and excretions from infected goats and shedding occurs despite only mild or inapparent disease in the case of FMDV (Alexandersen et al., 2003) or before the onset of clinical signs in the case of PPRV (Parida et al., 2020). PPRV is inactivated rapidly once shed by an infected animal (Baron et al., 2016), whereas FMDV can survive in the environment (Bartley et al., 2002). The survival of genomic RNA from these two viruses in the environment has not been compared to our knowledge, though in this study, the proportion of samples positive for PPRV RNA was much higher than that for FMDV (Table 1). This is consistent with the numbers of cases reported to the World Organisation for Animal Health (OIE) over the same time period, which were higher for PPR than for FMD (Table S1).
We were not able to sample or examine goats in the pens to confirm whether any were shedding FMDV or PPRV. However, in a companion study undertaken at the same time as the market sampling, we carried out environmental sampling at households with cattle in the Kathmandu valley (Colenutt et al., 2018). Here, clinical inspections of the animals and testing of oral swabs were used to confirm FMD, which gives us confidence that detection of FMDV in environmental samples is related to the shedding of FMDV by infected animals.
At livestock markets it is difficult to link detection of viral RNA in the environment to individual animals or even batches of goats. Movement of animals through the market in Kathmandu is relatively rapid and viral RNA from both viruses can persist in the environment after any shedding animals have moved on. In this particular market, little or no thorough cleaning activities take place, and each pen is likely to have a mix of goats that arrived in different batches, sourced from different areas, further complicating any potential tracing. Consequently, environmental samples lack the temporal information that clinical samples have (i.e., when and how many animals were shedding virus). The value of market surveillance could be enhanced by collecting information on animals entering and leaving the market, especially if market pen locations were also recorded. If connections between markets are stable over a sufficiently long period, timescales associated with positive detections are less important, but, if trading connections change frequently, traceability would present more of a challenge. Alternatively, sampling livestock transport vehicles when they arrive at markets could provide clearer temporal resolution, as well as making movement tracing easier. However, there may still be carry over of viruses between batches due to lack of cleaning.
Due to the methods used for sample collection and storage in the present study, only the presence of genetic material could be determined, rather than whether infectious virus was present. Detection of both PPRV and FMDV RNA was carried out through use of rRT-PCR assays which are designed to detect all lineages and serotypes of the respective viruses. These assays do not provide information on lineages or serotypes, though such data would be useful in informing vaccination selection or for tracing viral movements. Serotype-specific PCR assays are available for FMDV, but these need to be designed specifically for a geographic region (Knowles et al., 2016a; Saduakassova et al., 2017). In case of PPRV, the available gel-based RT-PCR assays can detect all four lineages, and usually viral genome sequencing is necessary for lineage determination. In general, environmental samples will have a lower viral content and genetic material will be of poorer quality than in clinical samples (Ganime et al., 2015; Colenutt et al., 2018), making it difficult to obtain more refined sequence data, as was the case in this study. With development of methods such as probe enrichment (Singanallur et al., 2019) or nested PCR assays to obtain appropriate PCR amplicons, recovering sequence data from samples with poorer quality and lower levels of viral RNA may become feasible in the future.
Although the frequency of movements and persistence of viral RNA in the environment complicates interpretation of the results, this study has demonstrated the potential use of environmental sampling at livestock markets for surveillance of FMDV and PPRV. The value of this approach could be increased in future as developments in diagnostic techniques and assays provide more scope for generating viral sequence information from the samples. The sampling methods could also be extended to use with many other pathogens, increasing surveillance capacity with a relatively easy sampling technique. Increasing the knowledge of livestock viruses present in a region by use of basic sampling techniques will aid with surveillance and control programmes.
ACKNOWLEDGMENTS
Fieldwork was carried out as part of the European Commission for the Control of Foot-and-Mouth Disease (EuFMD) Real Time Training program. This program is funded by the Department of Agriculture and Water Resources, Government of Australia, and is facilitated by the Department of Livestock Services, Government of Nepal. We are grateful for the help received from veterinary technicians in Nepal and to all traders who permitted sampling to take place. This work was funded by the Department for the Environment, Food and Rural Affairs (grant code: SE2816). Satya Parida and Simon Gubbins acknowledge funding from the Biotechnology and Biological Sciences Research Council (BBSRC) (grant codes: BBS/E/I/00007036 and BBS/E/I/00007037). Satya Parida and Mana Mahapatra also acknowledge funding from BBSRC (grant code: BB/T004096/1).
ETHICS STATEMENT
The authors confirm that the ethical policies of the journal, as noted on the journal's author guidelines, have been adhered to. No ethical approval was required as no animals were sampled in the present study.
CONFLICT OF INTEREST
The authors declare no conflict of interest.
Open Research
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available in the supplementary material of this paper.