Bulk-Tank Milk Longitudinal Serosurvey Reveals Decreasing Schmallenberg Virus Circulation in a Confined Population of Portuguese Sheep

In 2011, a new disease was detected in dairy cattle herds in The Netherlands and Germany, which resulted in a steep decrease in milk yield, diarrhea, sudden onset of illness, and occasional fever (Rasmussen et al. 2012). This disease was termed Schmallenberg virus (SBV) and was a few months later found to be arthropod borne, transmitted by Culicoides spp. midges (Rasmussen et al. 2012). SBV was then characterized as part of the Simbu serogroup, genus Orthobunyavirus, and family Bunyaviridae, having a negative-stranded tripartite RNA genome organized by a small (S), a medium (M), and a large (L) segment (Hoffmann et al. 2012). Owing to its ability to cross the placenta, SBV was also found to cause fetal malformations at the cerebrospinal and musculoskeletal levels, often observed at birth, named arthrogryposis–hydranencephaly–syndrome (Rasmussen et al. 2012).

Since its emergence reports have shown that SBV has spread across Europe, having reached the farthest western and eastern countries of Europe (Yilmaz et al. 2014, Esteves et al. 2016, 2018). Recent studies are starting to show a decrease in new SBV infections in Europe, potentially associated to the replacement (sales or death) of previously infected animals and also to vaccination, a circumstance that raises alerts for the possible re-emergence of SBV, if ideal conditions for the development of Culicoides occur (Stavrou et al. 2017). More recently a study showed the re-emergence of SBV on sheep flocks in United Kingdom in 2016/2017, calculating that the impact of this re-emergence was largely comparable with the findings reported back from 2011/2012 outbreak, and highlighting the ongoing effect of SBV on sheep farms (Stokes et al. 2018).

In this study, a 2-year longitudinal serological investigation for SBV infection at the herd level was performed by using bulk-tank milk of a specific sheep breed from central Portugal.

The study focused on the geographic subregion of Estrela Mountain (“Serra da Estrela”), the highest point in the Portuguese mainland territory (maximum of 2000 m above sea level), located in the central region of the country. This region encompasses the National Association of “Serra da Estrela” sheep breed (ANCOSE; www.ancose.com), which is in charge of managing the “Serra da Estrela” autochthonous breed, known for its continuous confinement to farm premises and for the highly valued milk production exclusively reserved for making the internationally recognized “Serra da Estrela” cheese. “Serra da Estrela” sheep movement restriction is hence likely to provide a clearer picture of the midge vector effect on SBV infection by excluding the influence of pathogen introduction by animal trade from regions with different SBV epidemiological status.

ANCOSE-listed sheep flocks (N = 180) were asked to participate in this study and all asked to provide bulk-tank milk samples (2 mL) in two periods (first collection in January/February 2015; second in January/February 2016). From the 180 sheep flocks, 78 agreed to participate and were hence enrolled (43.3%). Out of these 78, 10 did not provide either milk samples from at least one of the 2 years under study. Hence, 68 flocks were considered, from 46 parishes of 5 municipalities (Celorico da Beira, Fornos de Algodres, Gouveia, Seia, and Tábua). After collection, samples were transported to the laboratory at 4°C and processed according to the procedures of a previous study, with slight modifications (Chaintoutis et al. 2014). In brief, bulk-tank milk samples were centrifuged (1000 × g/10 min/4°C) followed by manual fat fraction removal by using a sterile spatula. The remaining lactoserum fraction was preserved frozen (−20°C) until testing. The bulk-tank milk lactoserum samples were tested for IgG antibodies to SBV by using a commercial indirect ELISA (ID Screen^® Schmallenberg virus Milk Indirect; IDvet, Grabels, France), according to the manufacturer's instructions. Differences in bulk-tank milk SBV seroprevalences between both years (2015 and 2016) were evaluated through Fisher's exact test (GraphPad Prism 5.0; GraphPad Software, Inc., San Diego, CA).

Results show that in 2015, of the bulk-tank milk samples from the 68 sheep flocks included in the study, 63 tested positive for anti-SBV IgG, which corresponds to a seroprevalence of 92.6% (95% confidence interval [CI]: 83.9–96.8). In contrast, of the 2016 sample collection, 53 presented anti-SBV IgG antibodies, corresponding to a seroprevalence of 77.9% (95% CI: 66.7–86.1). Differences in the 2015/2016 seroprevalences showed to be statistically significant (p = 0.027). This significant decrease in anti-SBV IgG seroprevalence at the herd level between 2015 and 2016 (92.6–77.9%) seems to be in agreement with reported data from other European countries (Stavrou et al. 2017). In fact, a high seroconversion was observed in the initial 2011–2012 SBV epidemic that was further followed by low-level circulation of SBV, with an overall decline of herd immunity (Stavrou et al. 2017). A recent study from our team has focused on seric antibody response at the animal level and suggested that anti-SBV seroprevalences have recently dropped (Esteves et al. 2018). Results from this study that uses a bulk-tank milk matrix confirms this decrease and also raise alerts, since increasingly favorable conditions (higher number of susceptible animals) are now present, potentially favoring SBV epidemics if improved conditions for midge replication occur in the future.

Despite the reported anti-SBV IgG seroprevalence decrease, there was still a large number of seropositive flocks observed both in 2015 and 2016. Alerted by these results and since SBV infection is known to have a negative impact in milk yield, we have subsequently performed a retrospective evaluation of the normalized production (NP) at 150 days of each of the analyzed 68 flocks, both in the 2015 and the 2016 seasons. For this we have used the Fleischmann's method for calculating milk yield in the lambing period (Ruiz et al. 2000) and the observed production between the end of lambing and the 150th day of lactation. Results show that the NP (lambing production + observed production until 150th day) per animal in each flock has increased in 44 of the 68 flocks between 2015 and 2016. Although very preliminary and aiming at measuring the impact of the decreased SBV circulation in 2015/2016, the calculated NP points to a concurrent increase in production in these two years.

Although the initial SBV epidemics studies focused on clinical aspects, the subsequent studies have addressed the impact of SBV infection on milk production and showed substantial losses in the affected cow farms (Wüthrich et al. 2016). These preliminary observations are in line since they suggest increased milk yield when flocks show anti-SBV IgG seroreversion.

In this study, we have used a bulk-tank milk anti-SBV ELISA to study SBV infection at the herd level. Bulk-tank milk ELISA is considered to be a fast and noninvasive diagnostic approach that is cost-effective and broadly used to monitor exposure to infectious agents (Collins et al. 2017). Noteworthy, a recent study has demonstrated highly and significant correlations between herd-level seroprevalence of SBV infection and bulk-tank milk ELISA results and that bulk-tank milk antibody titers are highly predictive of within-herd seroprevalence (Collins et al. 2017).

In conclusion, this study is the first in Portugal to provide prospective data showing a decrease in the occurrence of SBV infection based on the presence of anti-SBV IgG in bulk-tank milk. There is a clear need to expand knowledge on SBV epidemiology and impact in nonbovine ruminants, to reduce the impact of potential SBV re-emergence.