Bovine coronavirus in naturally and experimentally exposed calves; viral shedding and the potential for transmission

Bovine coronavirus (BCoV) is an important livestock pathogen with a high prevalence worldwide. The virus causes respiratory disease and diarrhea in calves and winter dysentery in adult cattle. These diseases result in substantial economic losses and reduced animal welfare [1]. One way of reducing the negative consequences of this virus is to prevent virus transmission between herds. Inter-herd transmission is possible either directly via transfer of live animals [2, 3], or indirectly via contaminated personnel or equipment [4]. Measures to prevent virus spread between herds must be based upon knowledge of viral shedding, the potential for transmission to susceptible animals and the role of protective immunity. Several observational studies have been published on BCoV shedding in feces of diarrheic calves and after transportation to feedlots [3, 5–10]. However, relatively few studies on BCoV pathogenesis with emphasis on transmission potential under controlled conditions have been published.

BCoV belongs to the genus Betacoronavirus within the family Coronaviridae, also including the closely related HCoV-OC43, which causes respiratory infections in humans, and the human pathogens SARS-CoV and MERS-CoV [11–13].

BCoV consists of one serotype with some antigenic variation between different strains [14, 15]. Acutely infected animals develop antibodies that persist for a long period, possibly for several years [16–18]. However, the protective immunity is shorter and incomplete. In two experimental studies, infected calves were not protected against reinfection with a different BCoV strain three weeks after the first challenge, but did not develop clinical signs [19, 20].

BCoV is transmitted via the fecal-oral or respiratory route [15]. It infects epithelial cells in the respiratory tract and the intestines; the nasal turbinates, trachea and lungs and the villi and crypts of the small and large intestine, respectively [21, 22]. Replication leads to shedding of virus in nasal secretions and in feces. Important factors for the pathogenesis are still not fully explored, such as how the virus infects enterocytes shortly after introduction to an animal. Viremia has been detected in one study by Park et al. [21]. Clinical signs range from none to severe, and include fever, respiratory signs and diarrhea with or without blood [1, 15]. As the time of infection is usually unknown and laboratory diagnostics are usually not performed, occurrence of clinical signs is the most relevant parameter to relate to viral shedding. The majority of experimental studies have used BCoV inoculation as challenge procedure, which may influence clinical signs and viral shedding, and thereby the transmission potential compared to natural infection. It has been hypothesized that BCoV can cause chronic subclinical infections which could be an important virus source [15]. Kapil et al. documented viral antigen in the small and large intestines of infected calves three weeks post inoculation [23]. Crouch et al. found that ten cows were shedding BCoV-immune complexes in the feces for 12 weeks [24]. It is, however, difficult to establish whether there is true persistence of virus, or reinfection of partially immune animals and whether these animals represent a risk to other animals. There is a lack of experimental studies investigating viral shedding pattern for longer periods than two weeks, with sensitive detection methods. Viral load and infectivity also needs to be determined. This is of high practical relevance, since the farmers need guidance on biosecurity in trade and transport of live animals.

The present study showed that calves infected with BCoV shed viral RNA for five weeks, and harbored viral RNA in intestinal tissues and lymph nodes even longer. Interestingly, contact with these calves three weeks after challenge, when the clinical condition had improved and the calves had seroconverted, did not lead to infection in sentinel calves and virus isolation was not possible from calves shedding viral RNA at this time point.

In concordance with other studies [18, 29], all EG calves became BCoV positive shortly after contact with infected calves and shed viral RNA continuously for two weeks. This supports that introduction of BCoV into a naïve population leads to a high basic reproduction number (R₀). R₀ depends on the duration of the infectious period, the number of exposed susceptible individuals and the probability of a susceptible individual to be infected. In herds and transportation systems where cattle from different herds are commingled, the risk of virus transmission is high.

The detection of BCoV RNA in nasal swabs from naïve calves in EG shortly after exposure might be due to passive inhalation of virus excreted by the FG, or to virus replication in the respiratory tract. Since the viral load in the nasal swabs from EG exceeded that of FG at D2, the study confirms that BCoV replicated massively in the airways of EG calves already at D2. Fecal shedding started later than nasal shedding which is in concurrence with other studies [30]. Saif and colleagues found that when inoculating calves intranasally, BCoV was first detected in nasal epithelial cells and secondly in feces. In contrast, in calves inoculated orally, fecal detection of BCoV preceded detection in nasal swab specimens. They concluded that the infection route could determine the sequence of infection of the respiratory and intestinal tract [22]. The present study supports that the respiratory route is the most common infection route when calves are naturally infected by direct contact. With indirect virus spread, the fecal-oral route could be more common.

Nasal swabs were more often positive for BCoV than fecal samples in this trial, most likely due to a higher limit of detection for BCoV in feces than in nasal swabs. For diagnostic purposes, nasal swab specimens therefore seem advantageous to fecal samples for virus detection in calves with suspected BCoV related disease.

Moving and commingling are associated with stress, which has been found to affect the intestinal immune system [31]. It is possible that stress increased the BCoV RNA shedding observed in the EG calves after introduction of the sentinel calves. Buying and selling of calves often involve extended transportation and commingling with susceptible cattle. The stress response, and a possible increased fecal shedding of virus, would probably be higher under field conditions.

In the acute stage of the infection, the agreement between positive PCR results and clinical score was relatively high. Three weeks after exposure to BCoV, the clinical signs and detection of viral RNA varied more independently. In an experiment with porcine deltacoronavirus, the severity of the clinical signs did not correlate with the shedding of virus in conventionally reared piglets, only in gnotobiotic piglets [32]. This indicates that secondary pathogens and changes in microbiota are important for disease development and clinical signs. The present study supports that after the acute stage of disease other factors than virus replication are important for clinical signs; for instance secondary bacterial infections.

Although the sentinel calves did not get infected with BCoV, they showed sporadic unspecific signs during the trial, but below the mildest category “mild disease” in the clinical scoring system. Since acclimatization was not possible, the calves changed environment including feeding routines when enrolled in the experiment, which could cause the signs observed. Other infectious agents could also have been present, and if so, most likely less virulent pathogens. Bovine virus diarrhea virus and bovine herpesvirus 1 are not present in Sweden [33], and the sentinel calves showed no serologic response to BRSV. Co-infection between BCoV and other agents is likewise possible in FG and EG, as is the case under field conditions.

Unlike most enteric viruses, BCoV is enveloped and therefore susceptible to environmental inactivation [1]. One might expect that the conditions in the forestomaches and abomasum would inactivate BCoV and one possibility is that BCoV is transported from the oronasal cavity to the small intestines through the bloodstream. However, viremia was not detected in the present study, and transport of the virus to the intestines appears to have been through the digestive tract. Park and colleagues [21] detected BCoV RNA in serum samples from calves infected with a winter dysentery strain between day three and eight post inoculation. They used nested PCR for detection, which is generally a more sensitive method than RT-qPCR, but also more vulnerable for contamination [34]. Short viremic period or intake of a lower virus dose in naturally infected calves could also explain the negative results in the present study. Inhibition of the RT-qPCR by plasma components was tested and ruled out. Despite the absence of detectable viremia in the present study, BCoV RNA was found in mesenteric lymph nodes at late stages of the infection. Viral RNA must have been transferred in low concentrations in blood or lymph to the draining lymph node, by antigen presenting cells or as free virus particles.

The finding of BCoV RNA in lymph nodes, ileum and colon six weeks after infection indicates coronavirus persistence in calves, however, the importance of this persistence for virus transmission is uncertain. Other coronaviruses are known to create persistent or chronic infections in mice and cats [35, 36]. MERS-CoV is shown to be excreted for more than a month in humans [37] and human coronavirus 229E creates persistent infections in vitro [38]. Although fecal shedding of BCoV RNA was detected five weeks post infection in the present study, the transmission potential at this stage is most likely negligible, as at three weeks post infection.

BCoV VRC were quantified by RT-qPCR, which does not give information on the number of infective particles. The ratio of total to infective particles (T/I) is challenging to establish for BCoV due to difficulties in cultivating virus from clinical samples. In the present study, virus titration showed a T/I ratio of approximately 5 log₁₀. With this high T/I ratio it is not surprising that virus isolation was unsuccessful after D13, when the VRC numbers are decreasing. It also agrees with the sentinel calves not getting infected D21. In contrast, roughly 8.8 log₁₀ VRC were detected per nasal swab and gram feces from the seronegative FG calves that infected the EG calves. With a T/I ratio of 5 log₁₀, each nasal swab and gram of feces contained more than 3.8 log₁₀ infective virus particles.

The high T/I ratio and the failure of virus isolation after D13 could be due to either few infective particles or low sensitivity of the isolation method. Low levels of infective particles could be caused either by high production of defective particles or by neutralizing effect of antibodies. Low sensitivity could be caused by suboptimal conditions in cell culture compared to in vivo (particularly for virus from clinical samples not adapted to cell culture growth), dilution of viral content in the swab, and freezing and thawing of the material. For feline enteric coronavirus, the T/I increased from 3–4 log₁₀ during the first week after infection, to up to 8 log₁₀ 28 days post infection [39], the increase possibly caused by the antibody response.

Few methods are available for studying transmission potential apart from live animal experiments, although ethically challenging and resource demanding. Existing literature is based on experimental studies examining BCoV shedding for 14 [20, 22, 40] to 21 [19, 41] days. To the authors’ knowledge, the present study is the first to study the shedding for as long as six weeks under experimental conditions. In addition, it is also the first to study the impact of this shedding using sentinel calves. Although a low number of calves were used, the results indicate that calves are not infectious three weeks after exposure to BCoV. This information is important and relevant in order to produce scientific based advices on how to avoid introduction of BCoV into herds. Further investigation of calves at different stages of disease is recommended to verify and corroborate these findings. The effect of stress related to transport on viral shedding and infectivity should also be considered.

In the present study, the virus that caused winter dysentery in adult cattle primarily gave respiratory disease in calves. Niskanen et al. also found that BCoV derived from an outbreak of winter dysentery caused mainly respiratory disease in weaned calves [29], supporting that BCoV is an important cause of respiratory disease in calves [42, 43] and winter dysentery in adults [17]. The economic and welfare consequences of BCoV therefore include the combined effects of neonatal enteritis, respiratory disease in young cattle and winter dysentery in adults. Also considering the high prevalence worldwide, BCoV is an important loss-inflicting factor in the cattle industry.

The current study shows that calves infected with BCoV are RT-qPCR positive in nasal and fecal specimens for a longer period than earlier recognized. However, contact with naïve calves three weeks after exposure did not lead to infection. A low level of infective particles could be due to either production of a high level of defective particles and/or production of neutralizing antibodies. The study provides highly relevant information when designing biosecurity advice regarding animal trade and coronaviral disease in cattle.