1. Life Cycle and Transmission
Similar to bovine neosporosis, the life cycle of
N. caninum in sheep begins with infection, which can occur horizontally via the ingestion of sporulated oocysts present in food or water
[1][2], vertically via the transplacental route from a previously-infected sheep, which has recently been shown to be the main source of infection in sheep
[3][4], as it is in cattle and goats
[5][6][7][8]. Vertical transmission can occur after a pregnant sheep ingests sporulated oocysts, causing exogenous transplacental transmission, or after reactivation during gestation of a chronic infection, causing endogenous transplacental transmission
[2][9]. Recently, vertical transmission has been shown to be a key feature in the persistence of ovine neosporosis for several generations within a flock due to chronically infected sheep
[10]. A very similar—if not identical—route of transmission has been described previously in bovine neosporosis, where it has been suggested that during reactivation of a latent infection in pregnant animals (i.e., recrudescence), encysted
N. caninum bradyzoites located in tissue cysts differentiate into tachyzoites, and subsequently disseminate throughout the host via the bloodstream (i.e., parasitemia), allowing invasion of the placenta and infection of the fetus
[11]. This theory is supported by parasite DNA having been found in blood samples from chronically infected sheep in epidemiological studies
[12][13], as it has in cattle
[14][15].
There is scant information on the influence of variables such as sex, age, or breed on the transmission of the parasite or the susceptibility to
N. caninum infection. Furthermore, there is a lack of consensus between the small number of studies investigating these variables, as some report a higher prevalence of infection in adult sheep
[16][17][18]—which would suggest that the transmission of the parasite is mainly due to ingestion of oocysts, as occurs in ovine toxoplasmosis
[19]—while others state that age is not a risk factor
[20][21], which is consistent with most abortifacient disease occurring after vertical transmission, as it does in bovine neosporosis
[22]. The same lack of consensus occurs with sex, as some studies have shown a higher prevalence of infection in males
[23], others in females, and some found no difference
[24]. With respect to breed, a higher risk of infection in Merino sheep compared to crossbreeds has been reported
[25], but other studies found no difference
[17][20]. Due to variabilities in the experimental design of these studies, meaningful comparisons are difficult, as are definitive conclusions.
2. Clinical Signs and Lesions
Once primary infection or recrudescence occurs, clinical presentations are characterized by abortions, stillbirths, or the delivery of congenitally infected but healthy lambs
[10][26][27]. Although the route of transmission (exogenous or endogenous) is unknown in most of the published studies of naturally occurring infections, stillbirths and abortions during late gestation are the most frequent outcomes. Gross lesions in aborted fetuses are infrequent and, when present, are not pathognomonic
[9][28]. Affected fetuses or placenta may show mummification, maceration, or just autolysis.
As described in both natural and experimental cases of ovine neosporosis, microscopic lesions in the placenta and aborted fetuses are similar to those caused by
T. gondii infection, in that they are typical “protozoal lesions”
[29][30][31]. Most studies that have described microscopic lesions due to natural
N. caninum infection in sheep are case reports comprising a low number of samples, or studies focused on the incidence of abortions due to
N. caninum infection. Typically, microscopic lesions associated with ovine neosporosis in the placenta are multiple foci of coagulative necrosis, with occasional mineralization and variable, non-suppurative inflammation in the cotyledons (
Figure 1A). Occasionally, the placenta may contain multiple focal aggregations of polymorphic nuclear neutrophils and lymphohistiocytic vasculitis, as well as cyst-like tissue structures
[3][27][29][32] (
Figure 1A). In the fetus, the most common lesions are multiple foci of non-suppurative inflammation of the brain
[33][34] (
Figure 1B,C), typical of similar protozoal infections, such as toxoplasmosis
[35]. The encephalitis is characterized by randomly distributed glial foci, vacuolization and death of neurons, congestion, mononuclear cell perivascular cuffing with microglia/macrophages, and occasional dystrophic mineralization
[10][27][33][36].
N. caninum tissue cysts in the neuropil (
Figure 1D) and the soma of neurons are also observed, with and without associated cellular inflammation, although the latter is more frequent—especially in stillborn, full-term, and neonatal lambs
[3][10][28][36]. Other than the fetal brain lesions, which are the most frequently encountered, several studies have described histological changes, such as multifocal non-purulent myositis affecting the tongue, heart, or skeletal muscles, and multiple foci of necrosis and/or infiltration of mononuclear inflammatory cells in the liver and lungs (
Figure 1E)
[10][29][32].
Figure 1. Microscopic lesions of ovine neosporosis. (A) Placenta. Focus of necrosis and scant infiltration of inflammatory cells, mainly lymphocytes and macrophages, within the maternal–fetal interface area of the placenta. Note mineralization at the center of the necrotic area (deep purple-pigment, arrowheads); HE. 4×. (B) Fetal brain; encephalitis characterized by multifocal glial foci randomly distributed within the neuropil; HE. 2×. (C) Fetal brain; higher magnification of the mononuclear glial foci within the grey matter of the cerebral cortex. There is mild mononuclear cell inflammation within the adjacent meninges; HE. 4×. (D) Fetal brain; parasite tissue cysts (arrowheads) within the neuropil. Note the scant inflammatory cell reaction in relation to the tissue cysts; HE. 40×. (E) Fetal liver; focus of caseous necrosis and infiltration of mononuclear cells—mainly macrophages—within the hepatic parenchyma. The aggregates of dark, mononuclear cells are foci of hematopoietic tissue—a normal finding in the fetal liver; HE. 10×.
3. Diagnosis
3.1. DNA Detection by PCR
Detection of
N. caninum DNA by PCR was initially developed for the diagnosis of bovine neosporosis, and the same technique is used for suspected ovine neosporosis, but despite the high specificity and sensitivity of PCR, this technique requires reasonable tissue sample preservation. However, in cases of abortion, the fetus and the placenta are usually autolytic, and this may compromise molecular diagnostic techniques
[6]. Although most diagnostic investigations have used fetal and placental tissues, other samples from adult sheep—such as blood, brain, and muscle—have been used in epidemiological studies. For example, three studies undertaken in Mexico and one in New Zealand analyzed whole blood to detect the parasite
[12][13][37][38]. Surprisingly, prevalence rates in two of the Mexican studies based on the detection of parasite DNA showed greater prevalence than shown by serology (25% vs. 5.5% and 27% vs. 13.5%, respectively). Conversely, in the New Zealand study, the serology showed a greater seroprevalence than detection of parasite DNA. Detection of
N. caninum DNA in blood from infected ruminants (mainly cattle) has rarely been reported, and blood is not considered to be the sample of choice, because of the inconsistency in detecting
N. caninum DNA in pregnant livestock
[13][15]. This is supported by studies by Arbabi et al.
[39], Amdouni et al.
[17], and Dessi et al.
[40], using brain, heart, or neck muscle tissues collected from sheep in a slaughterhouse, of which 3.9%, 10.6%, and 72.5% were positive for
N. caninum DNA, respectively. Although skeletal muscle has not been commonly used for the diagnosis of ovine neosporosis, it should be noted that parasite DNA has been found in the muscle tissue of seronegative animals
[41][42], so its use in epidemiological studies may complement—and, indeed, be more accurate than—the results from serological analyses. Similarly, the heart could be a suitable sample for diagnostic purposes in adult sheep, as parasite DNA was identified more frequently when compared with brain tissue from the same animals (6.7% and 0.7%, respectively)
[39], despite
N. caninum cysts being present in the brains of chronically infected sheep
[10].
Molecular diagnosis of ovine neosporosis is essentially achieved by one of two nested PCRs: one conventional endpoint PCR, and one real-time PCR, targeting different
N. caninum-specific genes (e.g., the
ITS1 and
Nc5 genes)
[31][43][44][45]. More recently, a quantitative PCR was developed for the detection and quantification of parasite burden in experimental samples
[30]. Most studies on suspected natural infections of ovine neosporosis use a nested PCR, confirming that this is the routine technique of choice for molecular diagnosis. However, it should be noted that the use of different PCRs could give rise to different results due to differences in sensitivity; for example, a semi-nested PCR is fourfold less sensitive than nested PCR for ovine samples
[46], and this has hampered direct comparison between studies.
3.2. Serology
In addition to examination for histological changes or parasite DNA, investigation of specific antibodies against
N. caninum is recommended, as aborted fetuses and, especially, stillborn lambs may not have histological lesions, or the parasite burden might be below the level detectable by PCR
[3][9]. There are several serological techniques available, including (i) a wide variety of enzyme-linked immunosorbent assays (ELISAs) (in-house and commercially available tests), (ii) indirect fluorescent antibody tests (IFATs), (iii) modified agglutination tests (MATs), and (iv) Western blotting (WB). The sample of choice is the serum, although milk has been also suggested for serological surveys of the prevalence of exposure of sheep to
N. caninum [47]. At present, ELISAs have mostly replaced IFATs, and several tests have been developed or adapted specifically for ovine samples, but their reliability needs to be improved by further validation studies
[48]. Although IFATs and ELISAs are the most commonly used serological tests for the diagnosis of ovine neosporosis, only a moderate agreement between them has been reported
[49][50]. This disparity could be explained by differences in secondary antibodies, antigen preparations, composition of the panel of sera, and the subjectivity inherent in interpretation of IFATs
[51].
Commercially available ELISAs used for the diagnosis of ovine neosporosis are usually “multi-species”, whereas in-house ELISA tests have usually been validated in each laboratory
[9][52][53]. There are proven differences in sensitivity and specificity between the commercially available ELISAs used for the diagnosis of bovine neosporosis
[54], due to their different components, and these have never been validated specifically for use in ovine neosporosis. For example, the use of protein G as a conjugate in commercial multispecies tests, due to its high binding affinity for ruminant IgG, might yield different results to those tests using a specific anti-ovine IgG as a conjugate. Nevertheless, the validation of diagnostic assays is a process involving constant development and readjustment of performance characteristics for each target population
[55]. This may be due to sera from experimental infections commonly being used during validation of serological tests
[48], even though field sera are required due to naturally infected animals having a significantly lower immune response
[56]. For this reason, the major challenge in definitively validating any serological test is collecting the sera required from the large number of naturally infected seropositive sheep from different countries.
One relevant confounding factor that must be considered when evaluating any serological test is antibody cross-reactivity. Conventional ELISAs using soluble
N. caninum antigens have been shown to have a high degree of cross-reaction with
T. gondii when compared with IFATs. However, cross-reaction between
T. gondii and
N. caninum when using IFATs can be ruled out by using cutoff values equal to or higher than 1:50
[57][58]. Cross-reactivity might be reduced by using more specific antigens or antibodies
[58], and several authors have highlighted the need to characterize new
N. caninum antigens linked to active infection in the host (e.g., immunogenic ones) for the development of more specific diagnostic tests
[59][60]. For this reason, novel ELISAs based on recombinant proteins (e.g., NcSAG1, NcSRS2), and already used in cattle, have been used for diagnosis in small ruminants with promising results, due to low antibody cross-reactivity
[52][61][62][63][64]. Finally, it remains unclear whether cross-reactivity between
Sarcocystis spp.—which are highly prevalent in sheep—and
N. caninum antigens affect serological diagnosis. A recent study reported very high levels (69.5%) of co-infection with
N. caninum and
S. tenella in 138 samples from sheep at a slaughterhouse
[40]. These two protozoal parasites are closely related, and have common antigens. However, there is a paucity of information about cross-reactions between
Sarcocystis species that affect sheep (e.g.,
S. tenella and
S. gigantea) and
N. caninum, although cross-reactions between
Sarcocystis spp. and
N. caninum in cattle seem to be negligible
[58]. Unfortunately, the high percentage of co-infection suggests the possible occurrence of cross-reactivity in serological diagnosis.
4. Prevalence
4.1. America
N. caninum infection has been reported in six countries within the American continent (Argentina, Brazil, Grenada, Costa Rica, Mexico, and Uruguay). The main finding from all of these studies was that there were variations in the prevalence rates both between countries, and even between regions within the same country. For example, animal seroprevalence rates in Brazil have been found to be as low as 1.8% and as high as 60.6% in the states of Rio do Norte (northeastern) and Rondonia (western), respectively
[65][66]. In general terms, serological studies carried out in the American continent were restricted to Latin and Central American countries—mainly Brazil (29/37), which had an overall seroprevalence of 19.9% (from a total of 15,461 serum samples tested)—and it has been suggested that deficiencies in environmental management and sanitation that could favor the dissemination of
N. caninum in the sheep flocks may be responsible for this high seroprevalence
[67].
4.2. Africa
In Africa, only five countries have conducted prevalence studies of N. caninum in sheep: Egypt, Gabon, Senegal, Tanzania, and Tunisia. However, these studies are not directly comparable, as they used different techniques: serology (ELISA) or DNA parasite detection (PCR). The prevalence rates were 8.6%, 42%, 41.9%, 1.5%, and 10.6%, respectively, but more studies using a consistent methodology are needed in order to draw any conclusions on the incidence and geographic distribution of ovine neosporosis in Africa.
4.3. Asia
In Asia, only seven countries (China, Iran, Iraq, Israel, Jordan, Pakistan, and Turkey) have reported specific antibodies against
N. caninum in sheep. Despite investigation, specific antibodies were not found in sheep in Malaysia
[68]. The countries that have carried out higher numbers of studies are Israel, China, Iran, and Turkey, with a total of 4804, 3865, 1783, and 1166 sheep sampled, respectively; the three latter showed overall seroprevalence of 7.9%, 3.6%, and 4%, respectively. However, the seroprevalence in Israel, based on a 10-year retrospective study, was conspicuously high (67.4%), especially when compared to the
T. gondii seroprevalence (46.7%)
[69]. This could mean that ovine neosporosis is an endemic disease in Israel or that there is a problem with cross-reactivity in the tests being used. The remaining Asian countries sampled less than 700 sheep.
4.4. Europe
Animal seroprevalence rates from Europe were reported in six countries (Czech Republic, Greece, Italy, Poland, Spain, and Switzerland) (10.9% average). The studies with the highest animal seroprevalence were from Greece (16.8%) and Italy (19.3% and 44.4%)
[25][70], whereas the lowest animal seroprevalence (0%) was reported in Spain
[71], despite most studies (
n = 5) having been carried out there, and an overall mean seroprevalence of 4.28% was calculated for the country. The remaining countries were mostly represented by one study each, except for Italy, which showed an overall mean seroprevalence of 21.9%. The high prevalence rate observed (72.5%) in brain samples by PCR in Italy was not used in the calculation of the overall seroprevalence
[40].
4.5. Oceania
In Oceania, ovine neosporosis has rarely been studied, and has only been reported in Australia and New Zealand, with those countries having the two lowest prevalence rates (0.7% and 0.8%). There was a high degree of disagreement between the ELISAs (1.3%) and IFATs (43.5%) used
[72], suggesting that at least two different serological tests are required to arrive at an accurate diagnosis. The seroprevalence rates determined by IFAT were not used in the calculation of the overall seroprevalence.
4.6. Experimental Design Variables and Risk Factors
The variability in the results from different studies may be due to differences in sample sizes, husbandry systems, times of investigation, serological tests used, and geographic factors
[73]. Sample size is a highly relevant variable when determining how representative a study is
[74], and those studies with a low sample size might not be representative of the true seroprevalence in the sheep population of the whole countries. Ideally, samples should be selected randomly to avoid any bias, although this is not always possible, and convenience sampling is frequently all that is available. The size of flocks and the rearing system have also been suggested to influence the prevalence of ovine neosporosis. For example, small- and medium-sized flocks, as well as semi-extensive and extensive production systems, have all been found to be associated with a higher prevalence of
N. caninum infection
[25][75][76]. However, this is contested by others who state that the production system does not influence the seroprevalence of this disease
[24][77][78], and some studies even state the contrary—that permanently housed sheep (i.e., intensive production systems) are more likely to ingest
N. caninum oocysts from contaminated batches of prepared food
[79][80]. This situation is highly complex, as it could also be related to flock hygiene and management practices, which are linked to flock size, as small family flocks typically have worse hygiene, which may facilitate infection due to more frequent opportunities for contamination of food and water by oocysts
[81]. Conversely, high-health-status flocks with better hygiene together with accurate veterinary supervision tend to be associated with larger, intensively managed flocks that have lower seroprevalence rates
[76][82]. Most published studies have reported a strong association between seropositivity for
N. caninum or the occurrence of abortions and the presence of dogs in the flock
[73][81][82]. However, it is important to appreciate that
N. caninum can maintain its life cycle for an indeterminate duration without the involvement of the definitive host, via endogenous transplacental (vertical) transmission.
Some studies have found a correlation between high seroprevalence rates for
N. caninum and a high-humidity climate, suggesting that the latter contributes to prolonged viability of oocysts in the environment
[66]. However, due to the paucity of studies on (i) horizontal transmission of ovine neosporosis and (ii) the frequency of oocyst shedding by dogs, the role of climate in the epidemiology of the disease requires further investigation. In addition, if one considers vertical transmission after recrudescence as the main route of transmission, then ecological factors would be less relevant
[6].