Enterovirus E infects bovine peripheral blood mononuclear cells. Implications for pathogenesis?

Enterovirus E (EV-E) is a member of the large Enterovirus genus in the Picornaviridae family along with 11 other species of enteroviruses (A, B, C, D, F, G, H, I, J, K and L). Enteroviruses are small, nonenveloped, icosahedral, single-stranded positive-sense RNA viruses that can infect humans and animals. Bovine enteroviruses (BEV) belong to two species: enterovirus E (formerly bovine enterovirus type 1, BEV-1) and enterovirus F (formerly bovine enterovirus type 2, BEV-2) (16, 33).

Enteroviruses are common viral pathogens which usually cause mild and self-limiting diseases, a notable exception being human poliovirus. However, enteroviral infections other than poliovirus can occasionally result in significant morbidity. Host-to-host transmission of enteroviruses occurs via the faecal–oral route and is followed by virus replication in the mucosa of the respiratory or gastrointestinal tract. The local lymphatic tissues are also targets, by colonising which the virus can spread to the lymph nodes and bloodstream. Viraemia enables a virus to reach secondary replication sites (34, 35), and while it lasts, leukocytes can act as an additional reservoir of the virus, thereby playing a significant role in the pathogenesis of enterovirus infections, particularly chronic ones (35). The literature data confirm the ability of human enteroviruses to productively infect various types of human immune cells, both T and B lymphocytes, and dendritic cells, monocytes or granulocytes (8, 9, 12, 17, 18, 26, 34, 39). Infection of immune cells with enteroviruses usually results in increased synthesis of numerous cytokines and chemokines by cells (6, 9, 12, 37, 39); from the clinical point of view, the most important of these are the pro-inflammatory cytokines interleukin 1β (IL-1β), interleukin 6 (IL-6) and tumour necrosis factor α (TNF-α), which sustain the inflammatory process and can contribute to the damage of tissues or even the development of immune-mediated chronic diseases (34). Elevated levels of these cytokines in the blood and cerebrospinal fluid have been observed in patients with a severe and complicated course of enteroviral infection (6, 19).

Bovine enterovirus was first isolated in late 1950s from the faeces of a clinically healthy cow, and was originally classified as enteric cytopathogenic bovine orphan virus (14). Because this virus, widespread in the cattle population worldwide, usually causes subclinical or mild infections, its pathogenicity has not been the subject of intensive research so far. Nevertheless, there are reports in the literature about respiratory, gastrointestinal and fertility disorders caused by BEV, some of which are fatal (3, 36, 42, 43). Bovine enterovirus is also listed as one of the viral pathogens associated with the bovine respiratory disease complex (25). The literature data concerning interactions of BEV with immune cells are only fragmentary. In in vitro studies, the ability of bovine enterovirus to infect human cell lines of the monocyte, B cell and T cell lineage has been demonstrated (32, 35). In an in vivo study, an experimental infection of mice with BEV resulted in liver lesions, and viral antigens were located around Kupffer cells and neutrophils (21). In turn, following the experimental infection of calves with BEV, the virus was localised in macrophages and lymphocytes of the gastrointestinal lymphatic tissue (3). However, the character of the interactions of this virus with bovine immune cells and the role of such interactions in the pathogenesis of infections remain unknown.

The aim of this study was to determine the ability of enterovirus E to infect bovine peripheral blood leukocytes and learn its influence on the viability and activity of these cells. Its findings have revealed the effect of EV-E on the mitogenic response of leukocytes; synthesis of IL-1β, IL-6 and TNF-α; and ability of phagocytes to carry out oxidative burst.

Bovine enterovirus is highly prevalent in cattle populations worldwide. Isolated from both healthy and diseased individuals, the virus induces production of antibodies, which are detected in most of the tested animals (2). In the first stage of our study, we confirmed the presence of antibodies (titres from 1 : 20 to 1 : 160) against BEV-E in nine out of ten cows the blood of which was used as our research material. This finding is fully consistent with results reported by other authors (2, 11). For example, in their study conducted in Turkey, Birdane and Gür (2) determined the presence of specific antibodies against EV-E in 153 out of 155 clinically healthy dams (98.7%), and the titres ranged from 1 : 5 to 1 : 160. In another study conducted in Brazil, neutralising antibodies against BEV were detected in 411 out of 414 tested cattle of both sexes (99.2%), and the titres varied from 1 : 5 to >1 : 640 (11).

Our experimental results suggested that previous contact of animals with bovine enterovirus, as was confirmed by the presence of antibodies in their blood serum, did not affect the permissiveness of their PBMCs to EV-E. The infection of cells was productive; we confirmed not only the presence of intracellular viral RNA but also the presence of viral progeny in the supernatant from the cell cultures. The virus achieved relatively low CCID₅₀ (1.75–3.875 log/mL) titres; in comparison, the titre of the same strain of virus in the MDBK cell line in our earlier studies typically oscillated around 7 logs (40). The only literature data available concerning the replication of enteroviruses in immune cells concern human pathogens. The productive infection of human cells has been confirmed in the case of enterovirus 71 (EV71), coxsackievirus B, enterovirus D and echoviruses (8, 9, 12, 17, 18, 26, 34, 39). Enterovirus 71 in supernatant from human PBMCs and monocyte-derived macrophages was measured at titres of 10⁴–10⁵ PFU/mL and 4.5 log TCID₅₀/mL (9, 39), respectively, whereas echoviruses in dendritic cells were not seen to exceed titres of 3 logs (18). The factors influencing the levels of viral titres are undoubtedly viral tropism and the percentage of virus-permissive cells. The available data suggest that different enteroviruses demonstrate tropism to different types of immune cells. Enterovirus 71 and coxsackievirus B infected T and B lymphocytes as well as monocytes/macrophages (8, 9, 12, 17, 26), enterovirus D was able to replicate not only in lymphocytes and monocytes but also in granulocytes (34), while echoviruses infected dendritic cells (18). Typically, the percentage of immune cells infected with enteroviruses was low. In a study by Wongsa et al. (39), monocytes were more sensitive than lymphocytes to EV71 infection, both carrying rather low percentages of infected cells at around 10% and 3%, respectively. Also, the productive infection of B lymphocytes by coxsackievirus involved only 1 to 10% of these cells (26).

Regrettably, we were unable in our study to determine which populations of bovine PBMCs were sensitive to EV-E, or what percentage of cells was infected with the virus. The rather low titres of the virus may indicate the limited susceptibility of the cows’ peripheral blood cells to EV-E infection. We also determined that most of the analysed cells contained a small number of copies of the viral RNA. This finding can be attributed to the late time of determination (24 h after infection). Plekhova et al. (30) demonstrated that the release of newly synthesised virions of enteroviruses from macrophages takes place as early as 4 h after infection, and the number of cells containing the virus decreases substantially with time. In the study by these researchers, the number of cells detected to contain the virus decreased by 60% after 24 h, and then after 72 h post infection the virus was undetectable in the cells.

In the subsequent stage of our research, we verified that the infection of bovine cells with the highly infectious dose of enterovirus E considerably decreased the cells’ viability. However, we did not observe statistically significant differences in the lymphocyte distribution relative to the uninfected cells, which may indicate similar sensitivity of different populations of these cells to EV-E infection. The literature data on the impact of enteroviruses on the viability of immune cells studied in vitro are divergent and what they indicate depends on the tested virus or even its serotype. Coxsackievirus B3 did not decrease the viability of human monocytes considerably, although it is capable of productively infecting them (12). In turn, Smura et al. (34) found that the EV-94 serotype of enterovirus D significantly decreased the viability of monocytes, granulocytes and B and T cells, while the EV-68 serotype did not affect the viability of cells despite generating viral progeny. It was also confirmed that enterovirus 71 in a high dose stimulated the apoptosis of T cells, which may represent a mechanism of evasion (7). On the other hand, human dendritic cells infected with echovirus showed signs of both necrosis and apoptosis (18). The elucidation of the mechanism through which high infectious doses of EV-E affect the viability of bovine PBMCs requires further research.

Infection of bovine PBMCs with enterovirus E had an effect on their blastogenic response to mitogens. A considerable decrease in the proliferation of T lymphocytes infected with the highest dose of the virus correlated with and was probably due to the decrease in the viability of these cells. Our cytometric analysis confirmed a decrease in the number of ConA-stimulated cells. However, the only significant change in the percentages of main lymphocyte populations caused by the highest virus dose was a substantial rise in the percentage of the relatively small population of double positive CD4+ CD8+ T cells. In humans, DP T cells are mature effector memory cells engaged in acquired immune response to viral antigens (27). This scenario cannot be excluded in cattle, especially because the cells analysed in this research originated from animals which had had previous contact with EV-E. This previous contact was implicated by the presence of antibodies in their serum, as detected in the first stage of the study. On the other hand, Onah et al. (28), who observed induction of DP T cells in sheep infected with Trypanosoma evansi after vaccination, attributed this finding to the generalised immunosuppression caused by trypanosomes in hosts since this phenomenon was followed by a substantial decline in all lymphocyte subset sizes. According to these authors, the appearance of DP T cells suggested the induction of the programmed cell death of lymphocytes. In our study, apart from increasing the percentage of DP T cells, the high dose of EV-E also caused a considerable decline in the viability of bovine PBMCs, and therefore lymphocyte apoptosis is also likely in this case. Nevertheless, the clarification of the mechanism of action of a high EV-E dose calls for further, more detailed investigations.

Another interesting finding made in the course of our research was a contrary response of LPS-stimulated cells to the virus which depended on their initial responsiveness. Based on the proliferative response of PBMCs to LPS, cows can be divided into high and low responders. The cells of low responders either do not divide after LPS stimulation or their response is negligible. It has been confirmed that such cows are more susceptible to infections after calving (developing metritis, mastitis or interdigital dermatitis) than high responders (1, 5). Based on our research results, we divided our material into two analogous categories. The low-responding cells did not proliferate after 72 h incubation with LPS, and none of the infectious doses of the virus had a significant effect on their mitogenic response. On the other hand, all the doses of the virus significantly decreased the LPS-induced proliferation of high-responding cells. Immunophenotyping confirmed a decline in the number of lymphoblasts after the stimulation of infected cells with LPS and a considerable decrease in the percentage of B cells, correlated with a compensatory rise in the percentages of the main T-cell subsets. There are no literature data that could elucidate this phenomenon. However, some publications suggest that the infection of other types of immune cells with enteroviruses can lead to a decline in their responsiveness to bacterial LPS. Henke et al. (13) reported from their experiment that the infection of human monocytes with coxsackievirus B3 resulted in the activation of cells, but also made them unresponsive to further activating stimuli, thereby considerably inhibiting the production of cytokines and prostaglandin E2 in response to LPS. In the case of human monocyte-derived dendritic cells, enterovirus B not only failed to activate cells but also caused their loss of responsiveness to LPS, thus reducing their maturation and decreasing the production of TNF-α and IL-12p70 in comparison with uninfected cells stimulated with LPS (18). The poor responsiveness of B lymphocytes infected with EV-E to LPS confirmed in our study is a worrying finding. The literature data prove that B lymphocytes play an important role in enteroviral infections. Peripheral B-cell deficiency in humans predisposed patients to adverse outcomes of enteroviral infections (10); antibodies playing a critical role in the control of human infections with coxsackieviruses and EV71 (20, 26), an insufficiency of the B cells to stimulate antibody secretio hampers the immune system’s response to these pathogens. B lymphocytes are a site of early replication of coxsackieviruses, and participate in their dissemination, which can be of key importance in an infection’s course and whether it becomes chronic (15, 17, 26). In turn, infection of mice with the human enterovirus 71 resulted in a reduced number of B cells in the animals’ peripheral blood and spleen (20, 41). The amount of EV71 in tissues was particularly high in the case of B-cell–deficient mice, and the survival rate of these animals was considerably lower than that of wild-type mice (20). Likewise, in piglets infected orally with the porcine enterovirus T80, antibodies played a more important role in counteracting the infection, IgA antibodies being particularly prominent, while the specific cellular response was weak, local in character and not connected with the considerable antiviral activity (4).

In the present study, apart from its effect on the viability and blastogenic response of bovine PBMCs, enterovirus E intensified the production of pro-inflammatory cytokines (IL-1β, IL-6 and TNF-α) by unstimulated cells isolated from high-responder animals. These cytokines are particularly important in the pathogenesis of enteroviral infections in humans, and sometimes it is even suggested that they may be more important than the virulence of the virus (19). Although they initiate the inflammatory and immune responses, which is beneficial for overcoming the infection, their excessive or persistent release can lead to immunopathological processes (12, 37). Patients with severe courses of EV71 infection were observed to produce considerably more IL-1β, IL-6 and TNF-α (6, 19). In in vitro studies, EV71 also stimulated the production of TNF-α in human PBMCs (8), and IL-1β, IL-6 and TNF-α in human monocyte-derived macrophages (9). According to the authors of that research paper, the stimulation of TNF-α may be important during the early phases of enteroviral infection, and monocytes/macrophages, the main producers of this cytokines, can be both the target and the effector cells in an EV71 infection (9). Coxsackievirus B3 also stimulated the production of IL-1β, IL-6 and TNF-α in human monocytes, the production of the last of these being the most sensitive indicator of the activation of bovine cells, as it also was in the case of EV71 (12).

It is well known that primary viral infections predispose animals to secondary bacterial infections. Among the bovine viruses having had their immunosuppressive effects well described are nearly all those which participate in development of the bovine respiratory disease complex in common with bovine enterovirus, i.e. bovine herpesvirus type 1, bovine parainfluenza virus type 3, bovine respiratory syncytial virus and bovine viral diarrhoea virus (29). Bovine enterovirus has always been perceived as a minor viral pathogen of cattle because of its low virulence. However, it is now known that BEV is not as mild as was claimed. As research techniques have advanced over many years, the opinions of researchers have changed about “commensal viruses” which have been detected in healthy subjects and until recently were considered completely harmless. It has been demonstrated that their presence in an organism can be associated with the development of immunosuppression, immunological anergy or tolerance. Their constant presence in the organism can also influence the course of infections caused by bacterial pathogens (trans-kingdom interactions) (38). In light of the results obtained in this study, the potential effect of the bovine enterovirus on the immune system may prove to be far more significant from the clinical viewpoint than its virulence. Nonetheless, confirmation of this hypothesis needs further, more extensive studies.

Our research shows that bovine PBMCs are permissive to enterovirus E and the infection is productive. Despite the relatively small amount of the viral RNA in cells and the low titres of the viral progeny in the supernatant, EV-E affected the viability and functions of bovine immune cells. High infectious doses of the virus decreased the viability of cells and inhibited the proliferation of T lymphocytes. However, the virus was found to have the strongest impact on B cells high responding to LPS, as their proliferation was significantly inhibited by the virus, regardless of the size of its infectious dose. All doses of the virus also stimulated the production of early pro-inflammatory cytokines (IL-1β, IL-6 and TNF-α). In our opinion, the clinical importance of EV-E, a virus widespread in cattle populations, may be underestimated. There is a risk that this virus can increase the susceptibility of infected cattle to secondary infections, particularly bacterial ones, which to a large extent are otherwise controlled through the antibody response. The interactions between the virus and the immune cells observed in our experiment may also implicate a potential evasion mechanism of the virus. However, continued and more broadly based studies are needed to verify these hypotheses.