Equine serum hepatitis (i.e., Theiler’s disease (TD)) is a serious and potentially life-threatening disease and one of the most common causes of acute hepatitis and liver failure in horses [1
]. Specific treatment options are still lacking. TD was first reported in 1918 by Sir Arnold Theiler, after he observed signs of liver disease in animals vaccinated against African horse sickness with a combination of live virus and equine antiserum [2
]. Similar to historical outbreaks of human posttransfusion hepatitis, multiple outbreaks of Theiler´s disease have been observed following parenterally administered equine serum products [3
]. An incidence between 1.4%–18% of fulminant hepatitis among horses receiving an equine biological product has been reported [2
]. Given the association between prior treatment with equine serum/plasma and the appearance of Theiler’s disease an etiologic role for a contaminating toxin or infectious agent has been suggested [8
]. However, the exact pathogenic agent remained unknown for nearly a century.
comprise a large family of non-enveloped DNA-viruses, which is currently subdivided into eight genera collectively known as parvoviruses. Members of this family have been described to infect a wide array of hosts, including humans, domestic, and wild animals [9
]. Parvovirus infections have been associated with various severe and fatal diseases affecting the respiratory, gastrointestinal, and haematological systems and further potentially causing abortions [6
]. Most recently, a novel equine Parvovirus (equine parvovirus-hepatitis virus (EqPV-H)) was isolated from serum and liver tissue of a horse that died of TD following administration of tetanus antitoxin (TAT) [15
]. Administration of TAT contaminated with EqPV-H further resulted in seroconversion and acute hepatitis in experimentally infected horses, indicating that EqPV-H might be the causative agent of TD [15
]. A recent study further reported a high prevalence of EqPV-H among commercial equine serum pools, indicating the necessity of careful risk assessment for medical and research applications [17
]. However, despite its association with equine diseases, EqPV-H has not gained much attention of equine veterinarians and its worldwide prevalence and epidemiology remain poorly investigated.
Here, we examined the prevalence of EqPV-H among Thoroughbred breeding horses in northern and western Germany to identify potential risk factors for EqPV-H infections. A total of 392 serum samples from Thoroughbred broodmares and stallions were analysed for the presence of anti-EqPV-H antibodies, DNA, and viral sequences, respectively. Furthermore, an analysis of risk factors potentially affecting the prevalence of EqPV-H infections was performed to investigate natural routes of virus transmission.
2. Materials and Methods
2.1. Serum Sample Collection
A total of 392 serum samples from Thoroughbreds stabled on stud farms in northern and western Germany (Lower Saxony, North Rhine-Westphalia, Hamburg, Schleswig-Holstein) representing more than 25% of all registered breeding horses at The German Thoroughbred Studbook Authority (Cologne) were collected and processed between 2012 and 2015 [18
]. All samples were then stored at −80 °C until further analysis regarding the presence of EqPV-H.
2.2. Detection of EqPV-H DNA
Viral DNA was extracted with a viral DNA Kit from Qiagen (Cat. No. 1048147, Hilden, Germany) according to the manufacturer’s recommendations. DNA samples were stored at −20 °C until further analysis. A probe-based quantitative polymerase chain reaction (qPCR) was used with primers and probe designed and provided by Dr. Amit Kapoor as described previously [15
]. A serial dilution of a plasmid containing the EqPV-H VP1 sequence was generated as standard row for the quantification of EqPV-H within the samples tested. qPCR measurements were performed using the LightCycler 480 real-time PCR system (Roche, Mannheim, Germany).
2.3. Detection of Anti-EqPV-H Antibodies
Samples were analysed for the presence of anti-EqPV-H-VP1 antibodies using the previously described luciferase immunoprecipitation system (LIPS) [19
]. The EqPV-H-LIPS antigen VP1 was produced as described by Divers et al. [15
]. Following the LIPS assay, relative light units (RLU) were determined using a plate luminometer (LB 960 XS3; Berthold, Bad Wildbad, Germany). To calculate sensitivity the mean RLU plus three standard deviations (SD) of an EqPV-H negative horse serum was defined as a cut-off limit. A potential cross-reactivity between the LIPS and other related parvoviruses could not be excluded.
2.4. Data Collection and Study Design
Three different groups regarding the state of EqPV-H infection were distinguished: Seropositive and EqPV-H DNA positive (DNA+/AB+), seronegative and EqPV-H DNA negative (DNA−/AB−), and seropositive and EqPV-H DNA negative (DNA−/AB+). Information regarding gender, age, state of reproduction, breeding, and international transportation history of the study population were received earlier from the Association for breeding and racing of Thoroughbreds (Cologne, Germany). Furthermore, the established criteria were subdivided into different groups. Age groups of similar size were created: 3–6-year-old horses (n = 78), 7–10-year-old horses (n = 113), 11–15-year-old horses (n = 129), and 16–29-year-old horses (n = 72). Similarly, horses were assigned to groups regarding the breeding history (0–4 breeding years) and the stock size of the stud farms (1–9 horses, 10–39 horses and >40 horses). Additional information about the transportation history was sorted according to the target country (France, Great Britain, Ireland, Germany, etc.). The data were determined by the time of sample collection between 2012–2015.
2.5. Sequencing and Phylogeny
For sequence analysis, a PCR (I) was designed within the NS1 of EqPV-H [17
]. PCR was performed using the expand high fidelity PCR system (Roche Diagnostics, Basel, Switzerland) as described before [17
]. PCR products were visualised on a 2% agarose gel, excised, and purified using a Monarch®
DNA gel extraction kit (New England Biolabs, Ipswich, Massachusetts, United States). Purified products were then sent for Sanger sequencing using the applicable PCR primers. Highlighter plot analysis [22
] displaying nucleotide exchanges in EqPV NS1 in the screened cohort compared to a previously published strain from Europe (MK792434). Labelling of nucleotides according to IUPAC code. Length of bar scale of neighbour-joining tree indicating number of nucleotide exchanges. The input multiple sequence alignment was created with Mega X.
In this cross-sectional study, we examined the prevalence of EqPV-H among almost 400 Thoroughbreds in northern and western Germany representing the first analysed European country. Based on the information from the German Association for breeding and racing of Thoroughbreds (Cologne), a total of 1450 brood mares and about 80 stallions were registered as breeding horses in 2014. Thus, we investigated about a quarter of the actively breeding Thoroughbreds in Germany and the results can be considered representative for the region. We observed a frequent occurrence of EqPV-H infections with 34.69% seropositive and 7.14% viraemic horses. The phylogenetic analysis demonstrated a high level of conservation in NS1 sequences in comparison with the world-wide circulating strains implying low levels of genetic variability. A detection of EqPV-H DNA with concurrent absence of anti-EqPV-H-VP1 antibodies was not observed for any sample, indicating no acute infection at the time of sampling. Of note, a cross-reaction in the LIPS assay with other potential parvoviruses in the horse samples cannot fully be excluded. Previously, in the first report describing the identification of EqPV-H a PCR prevalence in serum of 13% (13/100) was reported in the USA [15
], and in China the average prevalence was 11.9% (17/143) in racehorses [11
]. Further prevalence studies with larger cohorts from various continents are required for a more detailed epidemiology of EqPV-H.
We also performed a descriptive comparison between potential risk factors affecting EqPV-H prevalence. Iatrogenic transmission of EqPV-H has been demonstrated by Divers et al. through the administration of equine origin blood products [15
], which should be confirmed ideally with a clonal viral EqPV-H inoculum in future studies. Of note, the virus can also be transmitted to horses that did not have such treatments [8
]. The relative number of viraemic and seropositive horses in our analysis was increased with advanced age, especially between 11 and 15 years of age, but also in the group of 16–29-year-old horses compared to DNA−
horses. Possible reasons can be persistent humoral immunity or high risk of re-infection with longer living times. Due to the study design, horses younger than three years were not included. Therefore, it is impossible to draw a conclusion whether young foals in which the immune system is still maturing are at an increased risk of infection. Stage of reproduction was additionally accompanied by slightly higher fraction of EqPV-H infections. However, given that an extended breeding history mostly involves an advanced age of the examined horse this observation might be due to an indirect correlation. Neither stock size of the stud farm, transportation history to a foreign country, or an EqHV infection could be identified as a predisposing factor.
Of note, our previous investigation of the same study cohort revealed that younger Thoroughbreds transported abroad were at a higher risk to get infected with EqHV than horses at the same age staying in Germany. It is described that transportation of horses [23
] and a changing environment as well as disruption of established social groups [26
] can lead to a stress-induced immunosuppression, which could favour viral transmission. This risk factor seems therefore not to be highly involved in the susceptibility of horses to EqPV-H infections. Furthermore, equine hepacivirus co-infection was not a determinant for increased EqPV-H infections. Recently, a high rate of co-infections of these two equine viruses was also described in the prevalence report by Lu et al. from China [11
]. As an additional way of viral spreading, we also investigated the potential of vertical transmission of EqPV-H in a single case. Postnatal serum samples were taken from a foal after delivery from an EqPV-H DNA positive and anti-VP1-positive mare (Figure S2
). Transfer of EqPV-H-specific antibodies to the foal via colostrum could be observed, but no viremia was detected. The specific antibody transfer was expected, since the colostrum represents the important route of antibody transfer in neonatal foals, whereas the placental barrier hampers intrauterine transfer of macromolecules such as immunoglobulins from the mare to the foetus [27
In conclusion, these data revealed that Thoroughbreds from central Europe have a frequent occurrence of anti-VP1 antibodies and EqPV-H DNA suggesting circulating EqPV-H infections in the horse population, endemic herds, or persistent shedding. In addition, age and multiple breeding were identified as potential risk factors for EqPV-H transmission.