BACKGROUND AND INTRODUCTION
Neospora parasites were initially reported by Bjerkas, Mohn and Presthus (Reference Bjerkas, Mohn and Presthus1984) as causing encephalitis in dogs. The parasites were later isolated into tissue culture (Dubey et al. Reference Dubey, Hattel, Lindsay and Topper1988), named as Neospora caninum and were used to develop diagnostic reagents. The parasite has been found in several other intermediate hosts including sheep, goats, deer and horses and specific antibodies have been recorded in water buffalos, foxes, coyotes, camels and members of the cat family (Dubey, Reference Dubey2003). There is no good evidence to date that the parasite can infect and cause disease in humans (Graham et al. Reference Graham, Calvert, Whyte and Marks1999; Petersen et al. Reference Peterson, Lebech, Jensen, Lind, Rask, Bagger, Bjorkman and Uggla1999). The main clinical manifestations of disease are seen in dogs and cattle. Clinical signs in congenitally infected dogs include progressive paralysis of the hind limbs, muscle wasting and difficulty in eating and swallowing (Barber and Trees, Reference Barber and Trees1996). Analysis of foetal tissues from cases of bovine abortion implicated N. caninum as an important pathogen causing reproductive disease in cattle (Dubey, Leathers and Lindsay, Reference Dubey, Leathers and Lindsay1989). Clinical signs of bovine neosporosis include abortion, birth of a sickly calf that may show some neurological problems and have difficulty suckling or standing due to hind limb hyperextension (Dubey, Buxton and Wouda, Reference Dubey, Buxton and Wouda2006). In addition, calves may also be born from N. caninum-infected dams that appear healthy at birth although they are congenitally infected with the parasite.
Since its discovery, N. caninum has emerged as a major cause of abortion in both dairy and beef cattle worldwide (Dubey and Lindsay, Reference Dubey and Lindsay1996). A recent large-scale supranational comparison of seroprevalence levels in European cattle herds showed national seroprevalences of 16–76% for dairy cattle and 41–61% for beef cattle (Bartels et al. Reference Bartels, Arnaiz-Seco, Ruiz-Santa-Quitera, Bjorkman, Frossling, von Blumroder, Conraths, Schares, van Maanen, Wouda and Ortega-Mora2006a). Cattle infected with N. caninum are 3–7 times more likely to abort compared to uninfected cattle (Thurmond and Hietala, Reference Thurmond and Hietala1997; Moen et al. Reference Moen, Wouda, Mul, Graat and van Werven1998; Wouda, Moen and Schukken, Reference Wouda, Moen and Schukken1998). Economic losses associated with the disease include increased calving interval, reduced milk production, reduced value of stock and increased likelihood of culling (Trees et al. Reference Trees, Davison, Innes and Wastling1999; Dubey, Reference Dubey2003; Bartels et al. Reference Bartels, van Schaik, Veldhuisen, van den Borne, Wouda and Dijkstra2006b). Most abortions due to Neospora infection are diagnosed between 3–8 months of gestation (Dubey et al. Reference Dubey, Buxton and Wouda2006) although abortions occurring earlier may be more difficult to diagnose due to lack of foetal material to examine. A case study examining the breeding history of cattle of a closed dairy herd found a significant increase in the number of artificial insemination (AI) services per successful pregnancy in N. caninum sero-positive cattle compared to sero-negative cattle implying that infection with N. caninum may cause early loss of the foetus (Latham, Reference Latham2003).
Dogs have been identified as definitive hosts for the parasite (McAllister et al. Reference McAllister, Dubey, Lindsay, Jolley, Wills and McGuire1998; Basso et al. Reference Basso, Venturini, Venturini, Hill, Kwok, Shen and Dubey2001) along with coyotes (Gondim et al. Reference Gondim, McAllister, Pitt and Zemlicka2004). The oocyst stage of the parasite is shed in the faeces of acutely infected dogs following consumption of N. caninum-infected tissue (Dijkstra et al. Reference Dijkstra, Eysker, Schares, Conraths, Wouda and Barkema2001; Gondim, Gao and McAllister, Reference Gondim, Gao and McAllister2002). Cattle may become infected following consumption of N. caninum oocysts where the parasites excyst in the gut and the sporozoites invade and multiply within host cells (Dubey et al. Reference Dubey, Buxton and Wouda2006). The parasite is an obligate intracellular organism and the tachyzoite stage invades host cells and then multiplies by endodyogeny forming numerous daughter organisms that go on to invade other cells (Dubey and Lindsay, Reference Dubey and Lindsay1996; Hemphill, Vonlaufen and Naguleswaran, Reference Hemphill, Vonlaufen and Naguleswaran2006) thus providing a means for the parasite to spread via the circulation to different sites within the host (Okeoma et al. Reference Okeoma, Williamson, Pomroy, Stowell and Gillespie2004). In the pregnant animal the parasites reach the placenta via the circulation and invade the maternal caruncular septa in the placentome before infecting the foetal trophoblast villi cells and then spreading to the foetus (Dubey et al. Reference Dubey, Buxton and Wouda2006). Neospora caninum tachyzoites are able to invade and multiply within bovine trophoblastic binucleated cells and it is postulated that these cells may be involved in transplacental transmission of the parasite (Machado et al. Reference Machado, Mineo, Landim, Carvalho, Gennari and Miglino2007). Vertical transmission is a very efficient means of the parasite spreading to a new host and the rates of vertical transmission in pregnant animals infected with N. caninum are very high, estimated to be between 78–95% (Pare, Thurmond and Hietala, Reference Pare, Thurmond and Hietala1996, Davison, Otter and Trees, Reference Davison, Otter and Trees1999).
The invasion of host cells and subsequent multiplication of N. caninum tachyzoites alerts and activates the host's immune system to contain the spread of the parasite (Innes et al. Reference Innes, Andrianarivo, Bjorkman, Williams and Conrad2002). It is thought that, in response to the mobilisation of the host's immune response, the parasite differentiates into a slower replicating bradyzoite stage which can be observed within host tissues usually in the central nervous system (Dubey and Lindsay, Reference Dubey and Lindsay1996). The slow replicating bradyzoite stage is thought to persist within the host and may act as a source of infection to the foetus in subsequent pregnancies. It is not clear how long Neospora parasites will persist in the bovine host or whether there is strain variation among different parasite isolates in this respect. A characteristic feature of bovine neosporosis is that vertical transmission rates are very high and can occur over several generations and in successive pregnancies (Bjorkman et al. Reference Bjorkman, Johansson, Stenlund, Holmdahl and Uggla1996). The source of infection may be from a recrudescence of an endogenous infection (Trees and Williams, Reference Trees and Williams2005) or from a newly acquired infection and this may have important implications for the priming and effectiveness of the host immune response (Innes et al. Reference Innes, Andrianarivo, Bjorkman, Williams and Conrad2002, Reference Innes, Wright, Bartley, Maley, Macaldowie, Esteban-Redondo and Buxton2005).
This paper discusses some aspects of the relationship of N. caninum and its bovine host during pregnancy which is a complex and fascinating dynamic. It is hoped that research in this area, as well as being of interest to biologists and parasitologists, may also provide important insights into disease pathogenesis and host defence mechanisms that will lead to novel disease control strategies to combat bovine neosporosis.
NEOSPORA, PREGNANCY AND THE IMMUNE RESPONSE
Following infection with N. caninum the host immune system becomes activated with a wide range of different consequences for both host and parasite. Some of the immune responses will be host protective, some will be parasite protective and some may cause pathology. Many of the aspects of the host-parasite relationship are not well understood, in particular: (1) The early induction of the innate immune response and its role in priming the adaptive immune response; (2) The anti-parasite immune defence mechanisms; (3) The changes to the host immune response in pregnancy and how this effects parasite development in the host; (4) The triggers for inducing parasite differentiation in the host leading to persistent infection; (5) The sites for persistent infection in the host; (6) The triggers for recrudescence of infection leading to parasitaemia/parasitosis and transmission to the foetus and (7) The role of the immune response in causing disease.
Clinical disease is mainly seen in pregnant cattle and affects the developing foetus. Therefore the stage of gestation when the parasite is active in the host is critical in determining clinical outcome.
Stage of pregnancy: disease and transmission risk
Neospora-associated abortions are most commonly diagnosed from 3–8 months of gestation (Dubey et al. Reference Dubey, Buxton and Wouda2006) although as discussed above, N. caninum infection may also be involved in foetal death in early pregnancy which is difficult to diagnose and may present as fertility problems. Controlled experimental infections in cattle have shown that infection occurring in the first trimester of pregnancy may have more severe consequences for the foetus than infection occurring later in gestation (Barr et al. Reference Barr, Rowe, Sverlow, Bon Durant, Ardans, Oliver and Conrad1994; Buxton et al. Reference Buxton, Maley, Wright, Thomson, Rae and Innes1998; Williams et al. Reference Williams, Guy, McGarry, Guy, Tasker, Smith, MacEachern, Cripps, Kelly and Trees2000; Maley et al. Reference Maley, Buxton, Rae, Wright, Schock, Bartley, Esteban-Redondo, Swales, Hamilton, Sales and Innes2003; Macaldowie et al. Reference Macaldowie, Maley, Wright, Bartley, Esteban-Redondo, Buxton and Innes2004).
However, the risk of transmission to the foetus increases relative to gestational age. In experiments where cattle were given an intravenous challenge with N. caninum tachyzoites at 10 weeks of gestation there was an 83% chance of transmission to the foetus compared with 100% transmission rate when cattle were given a similar challenge at 30 weeks of gestation (Williams et al. Reference Williams, Guy, McGarry, Guy, Tasker, Smith, MacEachern, Cripps, Kelly and Trees2000). In addition, cattle challenged subcutaneously with N. caninum tachyzoites at 10 weeks showed 50% transmission to the foetus (Macaldowie et al. Reference Macaldowie, Maley, Wright, Bartley, Esteban-Redondo, Buxton and Innes2004) compared to 100% transmission when cattle were given a similar challenge at 20 weeks of gestation (Innes et al. Reference Innes, Wright, Maley, Rae, Schock, Kirvar, Bartley, Hamilton, Carey and Buxton2001; Maley et al. Reference Maley, Buxton, Rae, Wright, Schock, Bartley, Esteban-Redondo, Swales, Hamilton, Sales and Innes2003). This correlates well to observations with human congenital toxoplasmosis where there is an increased risk of transmission to the foetus as gestation progresses (Buxton, Reference Buxton1990). An important factor in determining parasite spread and transmission in the host is the maternal immune response.
Maternal immune response
There is very little currently understood about the early induction of the innate and adaptive immune responses following parasite invasion and multiplication within host cells (Hemphill et al. Reference Hemphill, Vonlaufen and Naguleswaran2006). Recent data have shown a role for innate immune mechanisms involving Natural Killer (NK) cells that were able to lyse N. caninum-infected fibroblasts and produce interferon gamma (IFNγ) following direct stimulation by N. caninum parasites (Boysen et al. Reference Boysen, Klevar, Olsen and Storset2006). This early induction of IFNγ may be important in helping to control intracellular parasite multiplication and, in addition, may also create the appropriate cytokine environment for the induction of the adaptive immune response towards a Th1-type response, as has been shown for Toxoplasma gondii (T. gondii) (Gazzinelli et al. Reference Gazzinelli, Amichay, Sharton-Kersten, Grunwald, Farber, Sher and Gross1996). As N. caninum is an obligate intracellular parasite, T cells recognising modifications to the infected host cells are thought to be important in protective immunity (Innes et al. Reference Innes, Buxton, Eperon and Gottstein2000, Reference Innes, Andrianarivo, Bjorkman, Williams and Conrad2002). Antigen-specific cell proliferation and production of IFNγ has been demonstrated in cattle experimentally infected with N. caninum (Marks et al. Reference Marks, Lunden, Harkins and Innes1998; Lunden et al. Reference Lunden, Marks, Maley and Innes1998; DeMarez et al. Reference DeMarez, Liddell, Dubey, Jenkins and Gasbarre1999; Williams et al. Reference Williams, Guy, McGarry, Guy, Tasker, Smith, MacEachern, Cripps, Kelly and Trees2000; Andrianarivo et al. Reference Andrianarivo, Barr, Anderson, Rowe, Packham, Sverlow and Conrad2001; Trees et al. Reference Trees, McAllister, Guy, McGarry, Smith and Williams2002) and in vitro studies have shown that the cytokines IFNγ and tumour necrosis factor alpha (TNFα) would inhibit parasite multiplication in fibroblast and bovine brain cell cultures (Innes et al. Reference Innes, Panton, Marks, Trees, Holmdahl and Buxton1995; Yamane et al. Reference Yamane, Kitani, Kokuho, Shibahara, Haritani, Hamaoka, Shimizu, Koiwai, Shimura and Yokomizo2000). CD4+ T-cells were also shown to be cytotoxic for N. caninum-infected cells (Staska et al. Reference Staska, McGuire, Davies, Lewin and Baszler2003, Reference Staska, Davies, Brown, McGuire, Suarez, Park, Mathison, Abbott and Baszler2005). Antibody responses are also known to be important in protective immunity to N. caninum (Eperon et al. Reference Eperon, Bronnimann, Hemphill and Gottstein1999), a likely function being to help reduce parasite invasion of host cells (Innes et al. Reference Innes, Andrianarivo, Bjorkman, Williams and Conrad2002; Haldorson et al. Reference Haldorson, Stanton, Mathison, Suarez and Baszler2006).
Neospora-specific antibody levels are known to fluctuate throughout pregnancy (Pare et al. Reference Pare, Thurmond and Hietala1997; Stenlund et al. Reference Stenlund, Kindahl, Magnusson, Uggla and Bjorkman1999; Guy et al. Reference Guy, Williams, Kelly, McGarry, Guy, Bjorkman, Smith and Trees2001; Andrianarivo et al. Reference Andrianarivo, Anderson, Rowe, Gardner, Reynolds, Choromanski and Conrad2005) and this can be used as an indicator of parasite activity in the host. An increase in specific antibody levels indicates an increase in parasite activity in the host. Studies have examined whether this might be used as an indicator of disease and noted that an increase in specific antibody titre around mid-gestation is more likely to occur in animals that would abort their foetus, compared to an increase in antibody titre towards the end of pregnancy which occurred in cattle that would vertically transmit the parasite but did not abort the foetus (Pare et al. Reference Pare, Thurmond and Hietala1997; Stenlund et al. Reference Stenlund, Kindahl, Magnusson, Uggla and Bjorkman1999; Guy et al. Reference Guy, Williams, Kelly, McGarry, Guy, Bjorkman, Smith and Trees2001). What is not well understood in bovine neosporosis are the triggers that cause the endogenous bradyzoite stage of the parasite to become active again allowing the parasite to invade the placenta, via the circulation, and then the foetus. Clearly, as discussed above, the timing of this event is critical to foetal survival. The earlier in gestation a parasitaemia/parasitosis occurs the more dangerous it is for the foetus. With the closely related parasite T. gondii, recrudescence of the bradyzoite stages of the parasite to become actively multiplying tachyzoite stages occurs when cell-mediated immune responses involving T cells and IFNγ are impaired as occurs in immunosuppressed individuals (Luft et al. Reference Luft, Brooks, Conley, McCabe and Remington1984). T cell responses and IFNγ are known to be important in inhibiting multiplication of N. caninum parasites and therefore any changes in the cell-mediated immune responses of the dam as a result of pregnancy may be a trigger that allows the parasite to become active again in the host (Innes et al. Reference Innes, Wright, Maley, Rae, Schock, Kirvar, Bartley, Hamilton, Carey and Buxton2001).
To enable the dam to carry the semi-allogeneic foetus without immunological rejection taking place, the cytokine environment at the crucial maternal-foetal interface, favours the more regulatory cytokines such as interleukin 10 (IL-10), interleukin 4 (IL-4) and transforming growth factor beta (TGF-B) (Entrican, Reference Entrican2002). These regulatory cytokines counteract the more inflammatory cytokines such as IFNγ which may be dangerous for the pregnancy and result in foetal death (Tangri and Ragupathy, Reference Tangri and Raghupathy1993). Infection with N. caninum tends to provoke a cell mediated immune response, involving CD4+ T cells and IFNγ to help control multiplication of the parasite which may contribute to disease pathogenesis in pregnant animals depending on the tissue location where the inflammatory response is provoked. Recent studies examining placental tissue in cattle, infected with N. caninum at 10 weeks of gestation, demonstrated the presence of increased numbers of CD4+ and CD8+ T cells, NK cells and IFNγ in the maternal inflammatory infiltrate associated with lesions in the placentas of cattle where foetal death had occurred (Maley et al. Reference Maley, Buxton, Macaldowie, Anderson, Wright, Bartley, Esteban-Redondo, Hamilton, Storset and Innes2006). In contrast, N. caninum infected cattle that did not abort, did not have any evidence of inflammatory immune responses in the placenta (Maley et al. Reference Maley, Buxton, Macaldowie, Anderson, Wright, Bartley, Esteban-Redondo, Hamilton, Storset and Innes2006).
What is not well understood in bovine neosporosis is whether the natural immunomodulation that occurs as a result of pregnancy impairs the ability of the dam to mount an effective cell mediated immune response involving pro-inflammatory cytokines and thus may compromise her ability to control parasite multiplication. Studies examining another intracellular parasite, Leishmania major, have shown that during pregnancy there was a bias in the immune response towards the regulatory cytokines, such as IL-10 and IL-4 and away from the inflammatory cytokines such as IFNγ and as a result the pregnant mice were less able to control the infection (Krishnan et al. Reference Krishnan, Guilbert, Russell, Wegmann, Mosmann and Belosevic1996). Studies in cattle have suggested that there is not a clear Th1 and Th2 bias in T cell clones, where cells were shown to express both IFNγ and IL-4 (Brown, Rice-Ficht and Estes, Reference Brown, Rice-Ficht and Estes1998). CD4+ T cell-enriched fractions of peripheral blood lymphocyes were collected from N. caninum infected cattle at 10 weeks of gestation and 30 weeks of gestation and were shown to express both IL-4 and IFNγ following antigen-specific stimulation in vitro (Rosbottom et al. Reference Rosbottom, Guy, Gibney, smith, Valarcher, Taylor and Williams2007).
These studies taken together indicate that an inflammatory immune response at the maternal-foetal interface may be detrimental to the foetus and compromise the pregnancy, whereas an inflammatory response in the periphery or other local lymph nodes may help to protect the host and contain the infection (Innes et al. Reference Innes, Wright, Bartley, Maley, Macaldowie, Esteban-Redondo and Buxton2005). More research is required to determine how the pregnancy-related changes to the immune system at different stages of gestation may favour parasite multiplication and regulate the potentially destructive pro-inflammatory type immune responses within placental tissue.
Foetal immune responses and congenital infection
The ability of the bovine foetus to mount an immune response to an invading pathogen increases progressively throughout gestation (Osburn, MacLachlan and Terrell, Reference Osburn, MacLachlan and Terrell1982). Mitogenic cellular responses can be detected from 12 weeks of gestation (Osburn et al. Reference Osburn, MacLachlan and Terrell1982) and at 14–15 weeks of gestation in bovine foetal spleen and thymus cells (Innes et al. Reference Innes, Wright, Bartley, Maley, Macaldowie, Esteban-Redondo and Buxton2005). Therefore if infection occurs early in gestation, the foetus is unlikely to be able to defend itself against the invading pathogen and the parasite will rapidly spread throughout the foetal tissues (Dubey et al. Reference Dubey, Buxton and Wouda2006). Studies looking at foetal immune responses around 4–7 months of gestation in Neospora-infected cattle, showed evidence of specific cell mediated and antibody responses (Andrianarivo et al. Reference Andrianarivo, Barr, Anderson, Rowe, Packham, Sverlow and Conrad2001; Almeria et al. Reference Almeria, De Marez, Dawson, Araujo, Dubey and Gasbarre2003; Bartley et al. Reference Bartley, Kirvar, Wright, Swales, Esteban-Redondo, Buxton, Maley, Schock, Rae, Hamilton and Innes2004). The immunocompetence of the foetus at the time of challenge will influence disease severity and parasite load. A recent study examined distribution of parasites within different foetal tissues and compared parasite loads and lesions in aborted foetuses of different gestational ages (Collantes-Fernandez et al. Reference Collantes-Fernandez, Arnaiz-Seco, Burgos, Rodriguez-Bertos, Aduriz, Fernandez-Garcia and Ortega-Mora2006). The highest parasite burdens were observed in foetuses from the first trimester of pregnancy with a decrease in parasite burden over time of gestation. Parasites were detected in several tissues (brain, heart, kidney and lung) early in gestation, whereas they were largely only detected in the brain in the third trimester. Lesions were also found to be more severe in foetuses from the first and second trimesters of pregnancy compared to the third trimester illustrating the influence of foetal age on disease severity (Collantes-Fernandez et al. Reference Collantes-Fernandez, Arnaiz-Seco, Burgos, Rodriguez-Bertos, Aduriz, Fernandez-Garcia and Ortega-Mora2006).
Another interesting aspect of N. caninum infection of the foetus in utero is the immunological consequence of congenitally acquired infection and how this relates to gestational age at the time of initial infection. Foetuses infected before 120 days of gestation with the non-cytopathogenic biotype of bovine viral diarrhoea virus (BVDV) may be born persistently infected with the virus but seem to be immunologically tolerant as they do not produce specific antibody responses against the virus (Brownlie and Clarke, Reference Brownlie and Clarke1993). It is not known if a similar phenomenon may also occur infrequently with calves congenitally infected with N. caninum. Studies looking at experimental infections of pregnant cattle have shown that occasionally calves have been born to infected dams that were seronegative to N. caninum in pre-colostral blood samples but were found to have N. caninum parasites in brain tissue on post-mortem (Innes et al. Reference Innes, Wright, Maley, Rae, Schock, Kirvar, Bartley, Hamilton, Carey and Buxton2001; Innes et al. unpublished observations). A further study described a calf born to a Neospora sero-positive dam that had low levels of specific antibody in a pre-colostral blood sample but then had N. caninum tissue cysts in the CNS tissue at post-mortem (Kyaw et al. Reference Kyaw, Suwimonteerabutr, Virakul, Lohachit and Kalpravidh2005). Neospora parasites are known to invade and multiply within host cells followed by cell lysis and invasion of new cells. The triggers to induce parasite differentiation into the slow multiplying bradyzoite stage are not well understood, nor is the influence of different parasite genotypes in this respect. These observations are interesting as they may have important implications for our understanding of disease epidemiology and may help to explain why some animals persistently transmit the parasite very efficiently in successive pregnancies and do not seem to develop good immunity to prevent this vertical transmission of their endogenous infection.
CONCLUDING REMARKS
The relationship between Neospora caninum and its bovine host is a very complex and fascinating biological interaction. Recent data, largely derived from experimental infection of cattle, has provided some information on disease pathogenesis, parasite transmission and maternal and foetal immune responses at different stages of gestation. However, there are many aspects of the host-parasite relationship that are not well understood including the early induction of the innate immune response and how this may prime the adaptive immune response; the triggers for causing the parasite to differentiate into the persistent bradyzoite stage and also to recrudesce from a persistent stage to the actively multiplying tachyzoite; the tissue sites for persistence in the bovine host and the influence of different parasite genotypes in this interaction. The natural immunomodulation occurring in the dam as a result of pregnancy may provide a favourable environment for the parasite to spread to a new host, the developing foetus. The immunological consequences of congenital infection are not well understood and it is not known whether congenitally infected cattle may pose more of a disease and vertical transmission risk compared to those cattle that were initially infected horizontally. Control strategies against the parasite are still quite limited and involve biosecurity advice about minimising contact between dogs and infected bovine tissues to help prevent oocyst shedding and to cover feed, silage and bedding materials to prevent contamination with dog faeces. There are no effective chemotherapeutic agents to prevent vertical transmission of the parasite or to cure animals infected with N. caninum (Dubey, Reference Dubey2003). There is much interest in using vaccination to control the disease, as cattle which have experienced an abortion due to N. caninum develop immunity that will reduce the likelihood of them experiencing a subsequent abortion due to Neospora infection (McAllister et al. Reference McAllister, Bjorkman, Anderson-Sprecher and Rogers2000). A commercial vaccine is available in certain countries, Bovilis® Neoguard, Intervet, and efficacy data of the vaccine under field conditions showed it to protect 50% of cattle against abortions occurring at 5–6 months of age in cattle in Costa-Rica (Schetters et al. Reference Schetters, Dubey, Adrianarivo, Frankena, Romero, Pérez, Heuer, Nicholson, Russell and Weston2004). Research examining the host-parasite relationship in pregnant cattle will provide important information that will help target a vaccination strategy and determine the feasibility of preventing endogenous transplacental transmission of Neospora caninum.
ACKNOWLEDGEMENTS
Elisabeth A. Innes would like to acknowledge the support of the Scottish Executive Environment and Rural Affairs Department and the Creative Science Company.
