neospora caninum abortion in cattle; neospora caninum associated lack of weight gain in calves; neospora-associated abortion; neospora-induced congenital myelitis and polyradiculoneuritis in calves; neuromuscular disease in calves
Neosporosis affects livestock in many countries, particularly as an important cause of abortion in cattle, and was first seen as a neuromuscular disease of dogs in Norway (Bjerkås et al., 1984). Dubey et al. (1988) described the causal parasite Neospora caninum in a new genus, Neospora,of the family Sarcocystidae in the phylum Apicomplexa. Recently, Dubey et al. (2002) published a redescription of Neospora caninum. N. caninum is a coccidian parasite that is closely related phylogenetically to Toxoplasma and Hammondia species, but differs in its biology, epidemiology and its importance for different livestock. A second species of the genus, called Neospora hughesi, has been recognized in horses (Marsh et al., 1998).
The life-cycle of N. caninum is not fully understood. Several intermediate hosts have been identified parasitologically (cattle, goats, deer, horses and sheep) and serologically (camels and water buffaloes). The dog can act as both an intermediate and definitive host. Thilsted and Dubey (1989) were the first to report Neospora-like organisms in bovine foetal brains from herds with persistent abortions in New Mexico. By 1991, N. caninum was reported as a major cause of abortion in dairy herds in California (Anderson et al., 1991), but the infection has been found retrospectively in cattle from 1974 (Dubey et al., 1990a). In 1993, the parasite was isolated from bovine foetuses in California (Conrad et al., 1993)and diagnostic tests were developed. It is now known to be an important cause of infertility, particularly abortion, in dairy and beef cattle, and neuromuscular disease in neonatal calves.
N. caninum infection has been detected parasitologically in cattle, goats, horses, deer, dogs, and, on two occasions, in sheep. The dog has been shown experimentally to act as a definitive host. Serological evidence suggests a wider host range, including water buffaloes (Dubey et al., 1998; Huong et al., 1998), camels (Hilali et al., 1998), coyotes (Lindsay et al., 1996a) and red foxes (Buxton et al., 1997b). The presence of infection in different deer species is indicative of a sylvatic cycle of N. caninum (Woods et al., 1994; Dubey et al., 1999a). Sheep are probably rare natural hosts for N. caninum because infection has been found only in one lamb (Dubey et al., 1990b) and not in aborted ovine foetuses (Otter et al., 1997b) or in sheep co-grazing with infected cattle (McGarry et al., unpublished observations). A range of animals and birds, including sheep, rodents and pigeons, are susceptible to experimental N. caninum infection.
N. caninum-associated abortions can occur all year, although a seasonal effect was reported in The Netherlands (Bartels et al., 1999) and California (Thurmond et al., 1995). In endemic herds, commonly 5 to 10 % of cattle abort (Moen et al., 1998; Davison et al., 1999a), but more than 50% of cattle can abort (Wouda et al., 1999a). A greater risk of abortion has been associated with the presence of dogs on farms (Paré et al., 1998; Bartels et al., 1999). A higher seroprevalence was found in dogs kept on farms than dogs living in rural areas of Japan (Sawada et al., 1998) and in dogs on farms with a high seroprevalence in cattle (Wouda et al., 1999b). Calves were shown to become infected when orally dosed with oocysts from an experimentally-infected dog (Marez et al., 1999). Dogs may become infected by ingestion of infected bovine material; 51% of over 300 foxhounds, which had been fed raw bovine material, were seropositive (Trees and Williams, 2000). Some livestock management practices, for example selection of (infected) high genetic merit cattle or feeding pooled colostrum, might affect the prevalence in herds (French et al., 1999; Uggla et al., 1998). Transmission between cattle, either by consumption of infected colostrum/milk or placental tissues, does not appear to be important in the transmission of N. caninum from naturally-infected cattle (Davison et al., 2001).
blood and circulatory system diseases of large ruminants blood and circulatory system diseases of pigs blood and circulatory system diseases of small ruminants mammary gland diseases of large ruminants multisystemic diseases of large ruminants multisystemic diseases of pigs multisystemic diseases of small ruminants nervous system diseases of large ruminants nervous system diseases of pigs nervous system diseases of small ruminants reproductive diseases of large ruminants reproductive diseases of pigs reproductive diseases of small ruminants respiratory diseases of large ruminants respiratory diseases of small ruminants skin and ocular diseases of large ruminants skin and ocular diseases of pigs skin and ocular diseases of small ruminants
The distribution in this summary table is based on all the information available. When several references are cited, they may give conflicting information on the status. Further details may be available for individual references in the Distribution Table Details section which can be selected by going to Generate Report.
Gross lesions are not common in N. caninum-aborted foetuses (Dubey and Lindsay, 1996) and none are pathognomonic. Foetuses may be autolysed or mummified with focal areas of necrosis in the brain, heart and skeletal muscle. In congenitally infected calves, there may be malacia and narrowing or deviation of the spinal cord (Dubey and Lindsay, 1996). Hydrocephalus and cerebellar hypoplasia have been reported in an aborted goat’s foetus (Dubey et al., 1996).
In affected cattle, there is a characteristic non-suppurative encephalomyelitis with multifocal infiltration (perivascular cuffs and gliosis) and necrosis (Dubey and Lindsay, 1996). Encephalitis, myocarditis and periportal hepatitis are frequently observed in aborted foetuses as well as adrenalitis, myositis, nephritis and pneumonitis (Wouda et al., 1997). Tissue cysts may be observed in sections, but tachyzoites are rarely seen unless stained with immunoperoxidase-labelled conjugates (immunohistochemistry). A non-suppurative inflammation of placental cotyledons may be present (Otter et al., 1995). Histopathological lesions are not always observed in congenitally infected calves.
N. caninum infection cannot be diagnosed solely based on clinical signs, because most infected animals are asymptomatic and the signs that do occur are non-specific. Abortion is the principle manifestation in adult cattle (and goats) and other signs of infertility can be observed. Neuromuscular disease in neonatal calves is uncommon, but typical signs of paresis and ataxia are suggestive of neosporosis.
In cattle and other livestock, infectious and non-infectious causes of abortion must be included in the differential diagnosis. A definitive diagnosis of N. caninum as the causal agent of abortion in either individual cattle or a herd can be problematic because finding cattle infected with N. caninum might only reflect the underlying prevalence of congenital infection in cattle, which may abort for other reasons (Thurmond et al., 1999). Viral, bacterial, rickettsial, protozoal and fungal causes of abortion that are known to occur locally should be investigated, including Bovine viral diarrhoea virus (BVDV), infectious bovine rhinotracheitis (IBR), Brucella abortus, Campylobacter fetus, Chlamydia species, Leptospira species, Listeriamonocytogenes, Salmonella species, Trichomonas fetus, Anaplasma species and Babesia species. Nutritional and toxic causes of abortion should also be considered. In neonatal calves, other causes of neurological signs (for example: trauma, abscess in the central nervous system, inherited cerebellar hypoplasia, congenital BVDV, poisoning and listeriosis) should be included in the differential diagnosis.
In horses, a range of clinical signs associated with Neospora, including abortion, diarrhoea, weight loss and ataxia have been reported. Neospora species should be included with Sarcocystis neurona in the differential diagnosis of Equine Protozoal Myeloencephalitis in endemic areas.
Maternal and foetal serology, histopathology and immuno-histochemistry are used to detect N. caninum infection, but require good laboratory facilities (Wouda, 2000). Isolation of the parasite is difficult (Conrad et al., 1993; Davison et al., 1999b) and not practicable for routine diagnosis. Serological investigations are useful, particularly at herd level.
Aborted foetuses have characteristic histopathological lesions in the brain, heart, and liver (Otter et al., 1995; Wouda et al., 1997). The brain is the most useful organ to examine, even if partially autolysed, but if it is not available the heart should be examined. Lesions of placental cotyledons are usually non-specific. The severity of foetal lesions may indicate whether N. caninum is likely to be the primary cause of the abortion, however, aborted foetuses from experimentally-infected cows can have minimal brain pathology. Immuno-histochemistry is useful to differentiate Toxoplasma gondii and Sarcocystis species (Otter et al., 1997a; Wouda et al., 1997), but is costly, relatively insensitive and requires special reagents. Polymerase chain reaction (PCR) and hybridisation techniques are used in research, but not for diagnostic purposes.
IgG antibodies can be detected using either an enzyme-linked immunosorbent assay (ELISA) (Paré et al., 1995a; Björkman et al., 1997; Williams et al., 1997) or an immunofluorescent antibody test (IFAT) (Paré et al., 1995b; Barber et al., 1997). ‘In-house’ tests are commonly used and few validated commercial tests are available (Mastazyme ELISA, Mast Diagnostics, Liverpool, UK, Williams et al., 1999; Biovet Inc, Québec, Canada). A non-commercial ELISA has been developed to detect antibodies in milk, and bulk milk was found to be positive if > 10 % lactating cows were seropositive (Björkman et al., 1997). The high cost of ELISAs and IFATs in terms of the reagents, equipment and technical expertise required precludes their use in many regions. A direct agglutination test can detect serum antibodies in different host species (Romand et al., 1998), and potentially would be a cheaper and more appropriate test in developing countries. An important feature of all N. caninum serological tests is that there should be no cross-reaction with antibodies against related protozoa (T. gondii, Hammondia species and Sarcocystis species). Foetal fluid serology is less useful, because a negative result may be obtained if the foetus was infected before immunocompetence in the fifth month of gestation. A positive result confirms that the foetus was exposed to the infection, but not that it caused foetal death.
A positive serological test result indicates previous exposure of the host to N. caninum. Seropositive animals are likely to also be infected, and it is not possible to distinguish between congenital and post-natal infections. Interpretation of serological results must consider the choice of test and cut-off threshold, and the age and reproductive status of animals tested. Antibody responses fluctuate over time and during gestation in cattle. For example, congenitally infected neonatal calves have very high responses, but 1 to 2 year-old infected cattle can have low, sometimes undetectable, responses (Dannatt, 1998; Davison et al., 1999a; Stenlund et al., 1999). High responses throughout gestation and increasing responses at mid-gestation have been associated with an increased risk of abortion (Paré et al., 1997), and probably reflect recrudescence of infection and parasitaemia. Antibody responses peak at the time of abortion and then decline, therefore a single sample at abortion is more useful than paired serology. One strategy to aid identification of infected adult cattle involves testing newborn calves and their dams because most infected dams infect their offspring, which have high, more easily detected responses. Potential recipient cows for embryos could be screened in this way–a negative result in both the cow and calf provides very strong evidence that the cow is not infected. For dam-calf sampling, calves should be tested prior to consumption of colostrum or milk from cattle other than their dams. A second strategy is to use a lower cut-off threshold, for example to screen stock prior to purchase or immature replacement heifers. Herd-based serology is useful to differentiate between N. caninum and other abortifacients in abortion outbreaks, in these cases sera from both cattle that have aborted (cases) and cattle that have not aborted (controls) are tested and the results compared.
In dogs, various IFATs with different cut-off thresholds are used and clinical cases of neosporosis typically have titres >1:800 (Barber and Trees, 1996). The identification of seropositive dogs on farms does not confirm that dogs are shedding oocysts. Finding oocysts in the faeces of definitive hosts is extremely difficult; there has only been one report in naturally infected canids (Basso et al., 2002), possibly due to the low numbers shed. N. caninum oocysts cannot be distinguished morphologically from H. heydorni oocysts, which are also found in canid faeces (Lindsay et al., 1999b).
The host immune response to N. caninum involves humoral and cell-mediated responses. The risk of abortion appears to be lower in adult cows compared with heifers, suggesting that infected cattle may develop partial protective immunity. Also, in a beef herd, cattle with evidence of previous exposure to N. caninum were less likely to abort than those with no exposure (McAllister et al., 2000). Cell-mediated immune responses (e.g., increased interferon gamma) are important because N. caninum is an intracellular parasite, and is likely to influence the outcome of infection. Experimental and field observations indicate that the host-parasite relationship is complex, for example, infection of the foetus early in gestation is usually fatal, but foetuses infected later in gestation often survive to be born alive and infected (Williams et al., 2000). Immune changes, which normally facilitate foetal survival, during gestation may allow multiplication of N. caninum and parasitaemia in the host.
The majority of hosts infected with N. caninum show no clinical signs. Abortion is the most common sign in cattle and the pattern of abortion may be sporadic, endemic or epidemic within a herd (McAllister et al., 1996; Moen et al., 1998; Davison et al., 1999a). Abortions are usually from 3 months of gestation to full term, but early foetal resorption or mummification can occur. About 5 % of N. caninum-infected cattle abort in successive pregnancies, whereas repeated abortion is not a feature of Toxoplasma gondii infection in sheep. Most congenitally infected calves are born without clinical signs, but occasionally abnormalities, including low body weight, paresis, incoordination or paralysis of the forelimbs and/or hindlimbs, are observed in calves less than two months of age (Dubey, 1999). Limbs may be flexed or hyperextended, with reduced patellar reflexes, and there may be loss of conscious proprioception. Exophthalmia and an asymmetrical appearance of the eyes have been reported (Dubey and Lindsay, 1996). Clinical signs can develop in apparently healthy calves within a few weeks of birth.
N. caninum infection has been associated with abortion in goats (Barr et al., 1992; Dubey et al., 1996) and death in a 2 month-old California black-tailed deer (Odocoileus hemionus columbianus; Woods et al., 1994). However, the disease has not been fully described in either these or other hosts, such as camels and water buffaloes.
N. caninum and N. hughesi have been found in aborted foals, congenitally infected foals with neurological abnormalities and adult horses with a range of clinical signs including weight loss, diarrhoea, abnormal behaviour and ataxia. Neospora infections have been found with other diseases, for example in a 19-year old horse with Cushing’s disease and a 20-year old horse with a pituitary tumour showing severe ataxia (Daft et al., 1996; Hamir et al., 1998). In the USA, 23% of 296 horses tested prior to slaughter were seropositive for N. caninum (Dubey et al., 1999b).
Like other hosts, most infected dogs have no associated clinical signs. Clinical signs have been reported in congenitally infected puppies and older dogs, but abortion is not a common finding. Barber and Trees (1996) reviewed 27 clinical cases of neosporosis in dogs that were aged between 2 days and 7 years old. The most common signs observed were progressive hindlimb paresis or ataxia with muscle atrophy. Rigid hyperextension, forelimb ataxia, dyspnoea, dysphagia, head tremor, collapse and death can occur. An unusual presentation is ulcerative dermatitis associated with high numbers of tachyzoites (Perl et al., 1998).
A diagram of the life cycle of N. caninum can be seen in Dubey (1999).
The life-cycle of N. caninum is typical of a coccidian parasite. Different intermediate hosts (e.g., cattle, goats, deer, horses, dogs), which can be infected with the two asexual parasite stages, tachyzoites and bradyzoites, have been identified (Dubey and Lindsay, 1996; Dubey, 1999). Tachyzoites divide rapidly and invade a range of tissues whereas bradyzoites divide slowly, and are found within cysts in neural tissue. Experimental studies have shown that dogs can also act as a definitive host and shed oocysts following ingestion of infected tissues (McAllister et al., 1998). However, no data are available yet on the frequency of oocyst shedding by naturally-infected dogs, survival of oocysts in the environment or whether other canids are definitive hosts. Little is known about the epidemiology of N. caninum in livestock other than cattle.
In cattle, epidemiological studies have shown that vertical transmission of N. caninum is very efficient with 72% to 100% of infected dams transmitting the infection to their offspring (Paré et al., 1996; Thurmond and Hietala, 1997b; Davison et al., 1999d). Recrudescence of existing infections during pregnancy results in reactivation of cysts and release of tachyzoites that cross the placenta and infect the foetus. Foetal survival may partially depend on the timing of parasitaemia during gestation, and therefore on the degree of foetal immunocompetence. Most congenitally infected calves are born alive, and apparently healthy, but remain infected for life so that successive generations can be infected (Björkman et al., 1996).
Little is known about the true extent of horizontal transmission of N. caninum either between cattle or to cattle from other hosts. Oocysts shed by a definitive host (McAllister et al., 1998; Marez et al., 1999; Lindsay et al., 1999a) or tachyzoites in colostrum, milk, foetal tissues or uterine fluids from infected cows are potential sources of post-natal infection (Uggla et al., 1998). N. caninum-associated abortion epidemics suggest that herds can be exposed to a point source of infection, for example oocyst-contaminated mixed feeds (McAllister et al., 2000). However, longitudinal studies of endemic herds found very low rates of post-natal seroconversion (indicative of horizontal transmission), for example out of several hundred heifers monitored from birth, only three (Davison et al., 1999d) or four (Hietala and Thurmond, 1999) heifers seroconverted. Nevertheless, even low rates of horizontal transmission might be important to allow persistence of infection (French et al., 1999) or the introduction of new infections into herds.
The economic costs associated with N. caninum infections include both the direct costs of reproductive failure (e.g., abortion) and indirect costs (e.g., reduced value of breeding stock). No accurate data on these economic costs have been published. Several authors have attempted to estimate costs for cattle production, but not for other livestock. For example, annual losses of US$ 35 million to the dairy industry in California (Dubey, 1999) and losses to Australian industry amount to AUS$85 million for dairy and AUS$25 million for the beef industry (Ellis, 1997) have been suggested. Such estimates are based on the prevalence of N. caninum found in aborted foetuses and do not account for the presence of the parasite in foetuses that may have been aborted due to other abortifacients, and therefore may over-estimate the significance of N. caninum in abortions. By contrast, other effects of N. caninum were not included in the calculations.
Direct effects of N. caninum infection in cattle can include 1) abortion, 2) stillbirths and neonatal mortality, 3) early foetopathy, 4) increased culling, 5) reduced milk production and 6) reduced value of breeding stock (Trees et al., 1999). Case-control studies are necessary to correctly attribute N. caninum as the cause of these effects, and using this approach, the proportion of bovine abortions attributed to this infection is 12.5% in the UK (Davison et al., 1999c) and 15% to 20% in The Netherlands (Wouda et al., 1997). Early foetopathy can result in return to service, increased calving to conception and inter-calving intervals. The role of N. caninum in stillborn calves and neonatal mortality has not been clearly established. The majority of calves congenitally infected with N. caninum appear normal and have similar weight gain and survival rates as uninfected calves (Paré et al., 1996; Jensen et al., 1999). However, infected cows were found to be 1.6 times more likely to be culled, even taking account of culling for abortion (Thurmond and Hietala, 1996), and infected first lactation dairy cows had a 4% lower milk yield (Thurmond and Hietala, 1997a). Indirect costs to consider include veterinary charges, diagnostic tests, reduced value of breeding stock and costs of management changes for disease control (improved feed storage, embryo transfer etc.).
There is no conclusive evidence that humans have been infected with N. caninum. No serological evidence of infection was found in 199 blood donors, 48 agricultural workers (Graham et al., 1999) or in 76 women with a history of abortion (Petersen et al., 1999). However, Tranas et al. (1999) did find 6.7% of 1,029 blood donors to be seropositive using an IFAT with a cut-off threshold of 1:100. Pregnant Rhesus macaques are susceptible to experimental N. caninum infection and transmit the infection to their foetuses (Barr and Conrad, 1994). Humans may be exposed to N. caninum, for example 2% of muscle and brain abattoir samples from adult cows in Switzerland were PCR-positive (Wyss et al., 2000), and therefore the possibility that some (e.g. immunocompromised) individuals are susceptible and might be exposed to N. caninum cannot be ruled out. Research workers adhere to strict guidelines to minimize the risk of accidental inoculation with live N. caninum cultures or infected tissues.
No treatment or prophylaxis is available for N. caninum infections in cattle. Several drugs are effective against tachyzoites in vitro, but none have been effective in vivo; for example, daily monensin treatment did not prevent abortion in N. caninum-infected heifers (Thurmond and Hietala, 1997b). Clinical signs in dogs improve after treatment with clindamycin, potentiated sulphonamides and/or pyrimethamine, but treated dogs can remain infected (Barber and Trees, 1996). No drugs are effective against tissue cysts, and therefore clearance of the infection is problematic.
Molecular studies are being conducted to identify genes that could be used in genetically-engineered vaccines. Currently there are no vaccines available for the control of N. caninum infections, except a vaccine (Bayer) based on killed tachyzoites that is on conditional licence in the USA; no efficacy data for this vaccine has been published. Future vaccination strategies must aim to prevent transplacental transmission of tachyzoites, oral infection in cattle, or oocyst shedding in dogs.
Husbandry Methods and Good Practice
Farm-level control strategies can be implemented in herds with N. caninum even though its epidemiology is not fully understood (Wouda, 2000). Vertical transmission is the major route of transmission in many herds and its impact can be minimized by culling seropositive cattle (if low numbers of cattle are infected), selecting uninfected heifer replacements and excluding congenitally infected calves. Seropositivity for N. caninum should be included with other criteria for decisions on culling. Embryo transfer has been attempted to utilize infected high genetic merit cows by implanting embryos into uninfected recipients. Purchased cattle should be serologically tested, preferably prior to purchase or on arrival. Good hygiene at calving should be practised, including the disposal of foetal membranes and placentae from all cattle to prevent ingestion by other cattle or dogs, and is prudent for control of other infectious diseases. Dogs and wild canids (e.g. foxes and coyotes) should not have access to placental material, aborted foetuses, dead calves, cattle feed or water supplies. Exposure of the herd to potential immunosuppressive factors, including mouldy feed (mycotoxins) and Bovine viral diarrhoea virus (BVDV), which may stimulate recrudescence of N. caninum infections, should be minimized.
National and International Control Policy
There are currently no national or international guidelines for the control of N. caninum. However, there is increasing awareness of the economic costs associated with this infection in livestock (Trees et al., 1999), and in future, serological screening and certification of traded livestock could be implemented to reduce the risk of introducing infected animals.
Chartier C, Baudry C et al. , 2000. Neosporosis in goats: results of two serological surveys in Western France. Point Veterinaire, 31:65-69.
Cheadle MA, Lindsay DS et al. , 2000. Prevalence of antibodies to Neospora sp. in horses from Alabama and characterisation of an isolate recovered from a naturally infected horse. International Journal for Parasitology, 30(5):677-677.
Conraths FJ, Schares G et al. , 2000. Seroepidemiological evidence for bovine neosporosis and Neospora caninum-associated abortions in the Russian Federation. International Journal for Parasitology, 30:890-891.
Coskun SZ; Aydyn L; Bauer C, 2000. Seroprevalence of Neospora caninum infection in domestic dogs in Turkey. Veterinary Record, 146:649-649.
Daft BM; Barr BC; Collins N; Sverlow K, 1996. Neospora encephalomyelitis and polyradiculoneuritis in an aged mare with Cushing's disease. Equine Veterinary Journal, 29(3):240-243.
Dannatt L, 1998. Neospora caninum antibody levels in an endemically-infected dairy herd. Irish Veterinary Journal, 51(4):200-201.
Davison HC, Guy CS et al. , 2001. Experimental studies on the transmission of Neospora caninum between cattle. Research in Veterinary Science (in press).
Dyer RM, Jenkins MC et al. , 2000. Serologic survey of Neospora caninum infection in a closed dairy cattle herd in Maryland: risk of serologic reactivity by production groups. Veterinary Parasitology, 90(3):171-181.
Ellis JT, 1997. Neospora caninum: prospects for diagnosis and control using molecular methods. In: Shirley MW, Tomley F M, Freeman B M, eds. Control of Coccidiosis into the Next Millenium. VII International Coccidiosis Conference European Union COST 820 Workshop. Compton, Berkshire, UK: Institute for Animal Health, 80-81.
McAllister MM, Björkman C et al. , 2000. Evidence of a point-source exposure to Neospora canium and protective immunity in a herd of beef cows. Journal of the American Veterinary Medical Association, 217(6):881-887.
Morales SE; Ramírez LJ; Trigo TF; Ibarra VF; Puente CE; Santa Cruz M, 1997. Description of a case of abortion in Mexico, in a cow associated with Neospora infection. Veterinaria México, 28(4):353-357; 45 ref.
Ooi HK; Huang CC; Yang CH; Lee SH, 2000. Serological survey and first finding of Neospora caninum in Taiwan, and the detection of its antibodies in various body fluids of cattle. Veterinary Parasitology, 90(1-2):47-55.
Ellis JT, 1997. Neospora caninum: prospects for diagnosis and control using molecular methods. In: Control of Coccidiosis into the Next Millenium. VII International Coccidiosis Conference European Union COST 820 Workshop, [ed. by Shirley MW, Tomley FM, Freeman BM]. Compton Berkshire, UK: Institute for Animal Health. 80-81.