Serology-based diagnostics for the control of bovine neosporosis
Stefano Guido1, Frank Katzer1, Ian Nanjiani3, Elspeth Milne2, Elisabeth A. Innes1
1Moredun Research Institute, Pentlands Science Park, Bush Loan, Penicuik, Midlothian, EH26 0PZ, UK
2The Royal (Dick) School of Veterinary Studies, University of Edinburgh, Easter Bush, Roslin, Midlothian, EH25 9RG, UK
3 Westpoint Veterinary Group, Dawes Farm, Bognor Road, Warnham, West Sussex, RH12 3SH, UK
Corresponding author: Guido, S.(; Tel: +44 131 445 5111; Fax: +44 131 445 5111).
Key words: Neospora; neosporosis; diagnostics; cattle; control.
Abstract
The protozoan Neospora caninum is a primary infectious cause of abortion in cattle that causes significant economic losses worldwide. As effective vaccines and licensed pharmacological treatments are currently unavailable, control measures rely on biosecurity and management practice. Serological diagnosis plays a crucial role in the identification of infected animals and a number of tests have been developed. However, due to the particular dynamics of the host-parasite interaction and to the characteristics of the currently used diagnostic tools, a proportion of infected cattle may not be reliably identified and can potentially undermine efforts towards the control of bovine neosporosis. Here, current diagnostic methods for N. caninum infection in cattle and the advancements required to support effective control strategies are discussed.
Neospora caninum in cattle: why be concerned?
Neospora caninum is a cyst-forming protozoan parasite of the phylum Apicomplexa.It is regarded as a major reproductive pathogen in cattle causing abortion and perinatal mortality [1]. Due to its worldwide distribution and efficient transmission, N. caninum impacts the cattle industry globally resulting in significant economic losses and production inefficiency [2].
The parasite is characterised by a heteroxenous life-cycle in which dogs and related canids are definitive hosts and cattle as well as a range of other species can act as intermediate hosts (Figure 1). Since N. caninum emerged as a major threat to dairy and to a lesser extent beef herds [2] nearly three decades ago, significant research efforts have been invested in the development of control measures. At present, vaccination appears the most desirable option [3] for controlling the disease in herds with high prevalence of infection [4]. However, despite different experimental vaccination strategies [reviewed by 3] and some promising results [5, 6] there are currently no commercially available vaccines to help prevent the disease [7]. Less attractive but also experimentally investigated is the chemotherapeutic option. Although several compounds showed inhibitory effects against N. caninum both in vitro and in vivo[8-12], presently there are no licensed therapies for bovine neosporosis. As a result, current measures aimed at reducing the impact of the disease are restricted to biosecurity and management practices [13].
Discrimination between infected and uninfected animals is the basis of disease management and a number of diagnostic tools have been developed for this purpose [reviewed by 14]. This review discusses current methods used in the diagnosis of N. caninum in live cattle and how they are currently applied to support disease prevention and control measures.
In vivo diagnosis of bovine neosporosis
In live cattle, N. caninum infection is primarily diagnosed by serology, namely the detection of specific antibodies in serum, plasma or milk [15]. Alternative techniques, such as the detection of parasite DNA by PCR in blood or semen [16, 17] and the assessment of pro-inflammatory cytokines as markers of exposure to the parasite[18], could also be applied, however being transitory indicators of infection their use is confined to the research field.
Numerous serological techniques have been developed for the determination of N. caninum serological status in cattle (Table 1).[19] Routinely, ELISAs (Enzyme-linked Immunosorbent Assay) represent the technique of choice for high throughput screening hence they are commonly used at the herd level to support the control of bovine neosporosis.Evaluating and comparing the performances of diagnostic tests for N. caninum infection is problematic since there is no true gold standard assay. In a recent comparative study, the performances of ten commercial ELISAs [15]were assessed and compared based on two definitions of gold standard: (1) majority of tests (i.e. samples that were classified as positive or negative by the majority of the tests evaluated were considered as reference positive or negative samples), (2) pre-test information (i.e. epidemiological, clinical and previous serological data using experimental ELISAs). Good test agreement and shared high performances in terms of specificity and sensitivity were observed, leading to the conclusion that serological diagnosis of N. caninum infection is accurate in measuring tachyzoite specific antibodies in cattle [15]. The lack of a perfect reference assay has been addressed by applying non gold-standard Bayesian modelling using field data. The performances of two commercial antibody ELISAs were evaluated within two different scenarios: diagnosis in aborting cows and testing of purchased animals. The tests showed comparable high accuracy and were classified as “fit for purpose” when applied to the designated purpose [20]. All currently commercially available ELISAs are based on tachyzoite antigens. However, would these tools also work to identify persistently infected animals when the parasite is relatively quiescent within tissue cysts? In the author’s opinion this point represent one the main criticalities of current commercial serological assays. Although the problem was addressed by complementing diagnostic ELISAs based on tachyzoite antigens with assays based on bradyzoite-specific antigens [21] (Table 1), none of these tests is currently marketed.
True or false seronegative cattle?
Reports of N. caninum serologically negative dams giving birth to seropositive calves [22-24] or aborting foetuses in which parasite DNA was detected [23] as well as post-mortem evidence of N. caninum infection found in tissues of seronegative non-aborting cows [25] expose a relevant issue that can be encountered when approaching the diagnosis of bovine neosporosis. The presence of these serologically elusive animals may be attributable to the variations in tachyzoite-specific antibody titres observed as a result of the changing dynamics of the host-parasite interaction and the characteristics of the diagnostic tests used.
Detectable specific antibody responses after N. caninum exposure are represented by the early onset of IgM antibodies which peak at two weeks before declining at four weeks post-infection followed by the appearance of IgG antibodies. Specific IgG concentrations rise for 3 to 6 months and are believed to persist for life in infected animals [14]. However, fluctuations occur depending on the physiological status of the dam [26-28] and the activity of the parasite [29]. Marked rises in antibody titres are observed during the second half of gestation and are associated with vertical transmission in experimental and field studies [26, 30, 31]. In persistently infected individuals this may reflect recrudescence of a persistent infection [32] in which bradyzoites, residing within tissue cysts, reconvert into the more active tachyzoites that spread throughout the body boosting the host immune responses. Conversely, in some cases tachyzoite-specific antibody titres may drop below the detection limits of serological assays [25, 33] so that previously serologically positive animals may become seronegative (Figure 1, Key Figure). This may occur in both aborting and non-aborting persistently infected cows at any stage during gestation [34]. A recent prospective serological study in dairy cattle found that more than one third of cows, with low antibody titres (1:200) during pregnancy, became serologically negative at the end of gestation [22]. In addition, two studies conducted in Argentina reported that 5% [35] and 3% [36] of dams that were seronegative at calving gave birth to pre-colostrally seropositive calves. Besides reporting fluctuations in the antibody titres, these findings highlight the limitations of some serological techniques.
Most if not all the work reporting fluctuations in specific N. caninum antibody concentrations in cattle refers to those humoral responses recognising the tachyzoite stage of the parasite since they are targeted by the vast majority of commercial and in house diagnostic tests. Concerns about the sensitivity of tests based only on tachyzoite antigens were raised following an interesting observation of a cow whose antibody response was able to recognise bradyzoite-specific and not tachyzoite-specific antigens [25]. This animal was kept under experimental conditions as a negative control for a N. caninum infection experiment and tested repeatedly negative with commercial and experimental tachyzoite-based ELISAs; however, parasite DNA was detected in several tissues post-mortem and further serological analysis carried out with a bradyzoite-specific (SAG4) antigen-based ELISA highlighted seropositivity to antigens related to the quiescent bradyzoite stage of N.caninum.
In experimentally infected animals, humoral responses to N. caninum bradyzoite-specific antigens show individual variability[21]. Indeed, antibody responses against bradyzoite antigens will depend on the intensity and duration of specific antigen exposure during the host-parasite interaction in cattle. In addition, rupture of tissue cysts may also be required to enhance detectable host immune responses against bradyzoite antigens. As suggested for some bradyzoite-specific antigens of the closely related apicomplexan Toxoplasma gondii, antigens specific to the quiescent stage may hardly be exposed [37]. There is no conclusive information on the extent these antigens are actually exposed to the host immune system as this may be difficult to assess and subject to individual variability in immunocompetence. The limited immunogenicity observed for the N. caninum bradyzoite-expressed BRS4 and SRS9 antigens was ascribed to a possible late up-regulation of expression during persistent infection with only a transient antigenic exposure to the host immune system [38]. What emerges is the need for more reliable diagnostic tests perhaps using a combination of antigens relevant for different stages of the disease that could enable the reliable identification of serologically elusive animals [34].
An alternative explanation for those animals that, despite harbouring N. caninum infection do not show any detectable immunological response, is the occurrence of acquired or innate immunotolerance i.e. unresponsiveness of the host immune system to certain pathogen-specific antigens [23, 39, 40]. This is supported by the fact that not all N.caninum infected calves born from seropositive dams are pre-colostrally seropositive indicating that congenital infection may occur without seroconversion [39]. This may be due either to infection of the foetus prior to reaching immunocompetence with the establishment of immunotolerance or to infections at a very late stage during pregnancy where the foetus has not had time to develop a serological response. Immunotolerance is well documented followingin utero infections with ruminant pestiviruses that may result in seronegative yet infected offspring [41]; however, its occurrence during N. caninum infection has not yet been clarified. With regards to diagnosis of bovine neosporosis for disease control purposes, immunotolerant animals would be a potential source of infection that would currently be difficult to identify.
Overall, the frequency, hence the epidemiological impact ofserologically elusive animals, is difficult to assess due to their intrinsic evasiveness to current diagnostic tests. Nevertheless, the possibility of false negative results should be considered when undertaking serological testing for bovine neosporosis.
Individual and herd testing
In cattle, the serological status with regards to N. caninum may be assessed in individual animals through serum, plasma or milk sampling or in groups of lactating cows by bulk milk testing[42].
Individual serology is a good indicator of the relative risk of abortion. In several studies, the risk of abortion observed in serologically positive cows was 2 to 26 times higher than in serologically negative[43, 44]. High antibody titres were also correlated to increased risk of abortion in parous cows [44] but not in heifers in which the lower antibody titres observed compared with older cows could not be associated with an increased risk of abortion [45]. These findings may be explained by the higher chance of repeated exposure to the parasite, either by secondary horizontal infection or reactivation over subsequent pregnancies, of older cows compared to heifers in which primary infections with a lower antigenic stimulus is more likely to occur [33].
The time of testing plays a role in the reliable identification of infected animals. In general, animals should be tested with serology when older than six months of age as there is evidence of colostral antibodies persisting for several months that may interfere with serological assays[46]. Most importantly, as antibody fluctuates sampling should be undertaken when there is the highest chance of detecting most of the infected animals which is likely to be during the second half of gestation when antibody titres are higher [34] or following an abortion. However, drawing conclusions about the effective serostatus of individual animals based on one testing period only may lead to the wrong conclusions given the potential occurrence of false negative results, as previously discussed [47]. Therefore, repeat sampling over subsequent pregnancies to confirm positive results is highly recommended.
Pre-colostral serology in newborn calves can be used to assess vertical transmission and indirectly assess infection status of the dam[46]. In ruminants, transplacental transfer of maternal immunoglobulins does not occur because of the syndesmochorial placenta.Passive immunity is transferred from dam to calf after birth through colostrum. Consequently, specific antibody responses in pre-colostral calves derive from the activity of the foetal immune system following exposure to a pathogen in utero. In the majority of calves born from N. caninum infected dams, the timing of in utero infection will determine the development of specific antibodies [48]. The absence of specific antibodies in stillborn or newborn pre-colostral calves would suggest that N. caninum infection is unlikely [14]. Effectively, a detectable pre-colostral antibody response may depend on the stage of pregnancy hence the maturity of the foetal immune system at the time of infection [29]. Obtaining and testing pre-colostral sera from calves is not a practical diagnostic option, although it is very informative for research purposes.
In dairy herds, the detection of N. caninum antibodies in bulk milk is a useful tool for measuring the within herd seroprevalence [49, 50], namely an estimate of the seroprevalence within the group of animals that contribute to the milk sample. Several studies showed good correlation of bulk milk results with herd seroprevalence as assessed through testing of individual serum or plasma samples [51-53]. However, specific considerations are required for the interpretation of results. In milk, antibodies appear later and at lower concentration, about 30 times less, than in serum [54]. In addition, besides the number of serologically positive animals in a herd, bulk milk antibody concentrations are dependent on the stage of lactation and the milk yield [55]. Although several ELISA tests have been adapted for use on bulk milk samples, sensitivity is reported to be limited. A minimum of 10 - 15% serologically positive animals appears to be required to produce positive bulk milk testing results [51, 53, 56]. Therefore, bulk milk testing may underestimate the proportion of infected herds, and misclassify those with low seroprevalence as negative. Nevertheless, ELISAs on bulk milk are considered cost-effective non-invasive indicators of the herd status that can be applied for control and surveillance purposes.
Application of serology-based diagnostics to the control of bovine neosporosis
At present, serological techniques represent a useful tool for approaching the control of N. caninum infection in cattle that may be applied at different phases and for different purposes during control programmes.
Initial assessment
In the preliminary phases of control programmes, serology is employed to assess whether N. caninum is directly related to the abortion cases, which is the predominant route of infection and what is the within-herd seroprevalence[13]. Collectively this information is required to shape strategies and take action according to the specific situation encountered that may vary substantially between farms.
In individual cases, positive post-abortion serology is highly suggestive of infection; however, it is not sufficient to prove that the parasite caused the abortion. Persistently infected cows may show detectable N. caninum antibody titres that are not necessarily related to the recent abortion event [14]. In order to justify costs and efforts of a control programme for bovine neosporosis, the infection-abortion relationship should be investigated. This can be approached by comparing the seropositivity rates in aborting and non-aborting cows at calving. If the former group have a significantly higher seropositivity rate than the latter, a relationship between parasite infection and occurrence of abortion can be confirmed [23, 57].
Serology may also help in the investigation of the predominant route of transmission (vertical or horizontal) by testing serum samples from dams and their offspring and, where possible, from pre-colostral calves [58]. In herds in which the transmission is predominantly vertical, the distribution of seropositive animals is uniform across age groups with both dams and their offspring having specific N. caninum antibodies. However, if the mode of transmission is mainly horizontal there is no association between the serological status of dams and offspring; serologically positive animals are in age clusters and may have either seronegative dams or seronegative offspring [59].
N. caninum abortions may follow an epidemic pattern characterised by abortion storms defined as the abortion of more than 10% of the cattle at risk (i.e. pregnant) within a period of 12 weeks or an endemic pattern in which the abortion problem persists for several months or years within a herd [60]. Epidemic abortions are believed to be due to a primary horizontal infection of a group of naive animals whereas endemic abortions occur as a result of recurrent transplacental transmission within family lines [61]. The abortion pattern can be investigated by estimating the odds ratio: a parameter that expresses the risk of abortion in the population at risk [62]. Endemic patterns of abortion are associated with odds ratios of around 2, whereas in cases of epidemic abortions higher odds ratios are found. Associating information about the abortion pattern with serology using avidity tests can provide further information on the predominant transmission route. It is recommended to determine avidity values on samples obtained immediately after abortion from a representative number (8 to 10) of seropositive aborting cows. High avidity observed in the presence of an endemic pattern of abortion would suggest that the vertical route is the predominant route of infection whereas low avidity antibodies associated with an epidemic pattern are indicative of recent exposure by the horizontal route [62-64].