Animals, inoculum and experimental design
Twenty four pregnant Holstein-Friesian cattle aged 1.3 to 4 years and seronegative for N. caninum, Toxoplasma gondii, bovine viral diarrhoea virus, infectious bovine rhinotracheitis and Leptospira hardjo were assigned into three groups. Pregnancy and foetal viability was confirmed in all experimental animals by ultrasound scanning 36 days after insemination. On day 70 of gestation, group 1 dams (n = 8) received an intravenous (iv) inoculation in the right jugular vein of 5×108 live N. caninum (NC1 isolate) tachyzoites. Group 2 dams (n = 8) received a subcutaneous (sc) inoculation of 5×108 live N. caninum (NC1 isolate) tachyzoites over the left pre-femoral lymph node. Group 3 (n = 8), the control animals each received an iv inoculation of 5×106 Vero cells. This dose of Vero cells was used, as it was the equivalent number of cells present in the parasite inocula. Blood was collected by weekly jugular venipuncture throughout the experiment for immunological analysis. Two animals from each group were sacrificed at days 14, 28, 42 and 56 post inoculation (pi). At post mortem samples of left pre-femoral lymph node (LPF), right pre-femoral lymph node (RPF), left uterine lymph node (LUL), right uterine lymph node (RUL), mesenteric lymph node (MLN), retropharyngeal lymph node (RLN), spleen and peripheral blood mononuclear cells (PBMC) were collected from each dam; When possible spleen, thymus and PBMC samples were collected from the foetuses.
Preparation of cells for immunological assays
Single cell suspensions of PBMC were prepared as previously described . Samples from lymph nodes collected at post mortem were prepared as previously described . In brief, excess fat was trimmed from the tissues, which were then cut into small pieces and placed in 10 ml wash buffer (Hank’s balanced salt solution (HBSS) supplemented with 2% foetal bovine serum (FBS) (Labtech International, Ringmer, UK) 100 IU/ml penicillin and 50μg/ml streptomycin) (Northumbria Biologicals, Cramlington, UK), placed in a stomacher bag (Seward Medical, Northampton, UK) and homogenised for 10 seconds. The resultant cell suspension was decanted into a sterile universal through sterile lens tissue to remove clumps of cells, washed twice by repeated centrifugation at 260 × g, counted using a Neubauer haemocytometer and resuspended at a final concentration of 2×106 cells/ml in cell culture media (CCM) (Iscoves modified Dulbecco’s media (IMDM) (Gibco, Paisley, UK) supplemented with 10% FBS, 100 IU/ml penicillin and 50μg/ml streptomycin).
Cell proliferation assays
Single cell suspensions of both PBMC and lymph node tissues were treated as previously described . In brief, equal volumes (100 μl) of cells (2×106/ml) and antigen were added in quadruplicate to 96-well round bottom plates (Nunc, Roskilde, Denmark). Water-soluble N. caninum tachyzoite antigen (NCA)  was used at a final protein concentration of 1μg/ml, the T-cell mitogen concanavalin A (Con A) was used as a positive control at a final concentration of 5μg/ml, CCM alone was used as a negative control to determine the background level of proliferation. A Vero cell lysate antigen at 1 μg/ml was used as a control antigen. The cultures were incubated at 37°C in a humidified 5% CO2 atmosphere for 5 days. The cultures were pulsed with 18.5 kBq 3 H Thymidine/well (Amersham Biosciences, Little Chalfont, UK) for the final 18 hours, before being harvested onto glass-fibre filters (Canberra Packard, Meriden, CT, USA) and the cell-associated radioactivity was quantified using a MATRIX 96TM gas proportional counter.
Duplicate cell proliferation assays were prepared to those described above. Cell free supernatants were collected after 4 days incubation, to measure the levels of secreted cytokines (interferon-gamma (IFN-γ), interleukin-4 (IL-4), IL-10 and IL-12). The supernatant samples were stored at −20°C prior to analysis.
Levels of IFN-γ production were quantified using a commercially available enzyme linked immunosorbent assay (ELISA) kit (CSL Veterinary, Parkville, Australia). A standard curve was generated using doubling dilutions of known quantities (ng/ml) of recombinant bovine IFN-γ (rBoIFN-γ) (Pfizer Animal Health, Parkville, Australia). Mean optical density (OD) values were plotted against ng/ml rBoIFN-γ and a standard regression curve was fitted to the data. Experimental samples were extrapolated against the standard regression curve to determine the levels of IFN-γ in the test samples.
Levels of bovine IL-10 were quantified using an ELISA method as previously described . In brief, 96-well ELISA plates (Greiner, Stonehouse, UK) were coated with 50μl (4μg/ml) per well with a primary anti-bovine IL-10 capture antibody and incubated at room temperature overnight. The plates were washed x 5 using phosphate buffered saline (PBS) supplemented with 0.05% Tween 20 (PBS-T) between each step, with the exception of the final TMB – H2SO4 stage. The plates were blocked at room temperature for 1 hour with PBS-T supplemented with 3% bovine serum albumin (BSA). Samples and standards (50μl each) were added and incubated at room temperature for 1 hour. Plates were then coated with (1μg/ml) secondary biotinylated anti IL-10 antibody (Diluted in PBS-T 1%BSA) (50μl per well) and incubated at room temperature for 1 hour. Streptavidin-horseradish peroxidase (HRP) (Dako Cytomation, Glostrup, Denmark) diluted 1:500 in PBS-T 1%BSA (50μl/well) was added and incubated at room temperature for 45 minutes. Colour was developed by the addition of TMB (3,3′,5,5′-tetramethylbenzidine) substrate (Insight Biotech. Ltd., Wembley, UK) (100μl/well) and incubated in the dark for 10–15 minutes. Reactions were stopped by adding 50μl/well 1 M H2SO4. The plates were read at 450/650 nm using a MRX II plate reader (Dynex, East Grinstead, UK). Doubling dilutions of known quantities of recombinant bovine IL-10 (rBoIL-10) were used to generate a standard regression curve against which the test sample concentrations were extrapolated.
Interleukin-12 (IL-12) was quantified using the same method described above for the detection of IL-10, with the following changes. A primary anti IL-12 capture antibody (4μg/ml) was used along with a secondary biotinylated anti IL-12 antibody (8μg/ml). Known quantities of recombinant ovine IL-12 (rOvIL-12) were used as standards and standard regression curve was fitted to the data .
Interleukin-4 (IL-4) was quantified using the same method described above for the detection of IL-10 and IL-12, with the following changes. A primary anti IL-4 capture antibody (6μg/ml) was used along with a secondary biotinylated anti IL-4 antibody (2μg/ml). Known quantities of recombinant bovine IL-4 (rBoIL-4) were used as standards and a standard regression curve was fitted to the data .
All primary and secondary antibodies used for the capture and detection of IL-4, IL-10 and IL-12 were purchased from AbD Serotec, (Oxford, UK). All rBoIL-4, rBoIL-10 and rOvIL-12 cytokines (Moredun Research Institute, Edinburgh, UK)
At post mortem examination blood was drawn from the foetuses (when available) into non-heparinised evacuated tubes (Vacutainer, Becton Dickinson, Oxford,UK) and allowed to clot before centrifugation at 2000 x g for 15 minutes, the serum was removed and stored at −20°C prior to being tested for IgM and IgG to N. caninum by an indirect fluorescent antibody test (IFAT), as previously described ; an IFAT titre of ≥1:64 was considered positive.
The maternal cell proliferation data (PBMC, lymph nodes and spleen) and IFN-γ ELISA data were analysed with a linear mixed model, using a first-order autoregressive model to specify the temporal covariance structure. Both the proliferation and IFN-γ ELISA data were normalised by logarithmic transformation (base 10) prior to the analysis. The linear mixed model included the animal effect as a random effect and the treatment, day and the interaction effect of treatment and day as fixed effects. Parameters of the linear mixed models were estimated using the REML method and p-values were estimated using the modified F-statistic. If the F-statistic was statistically significant (p ≤ 0.05), two-sided probabilities for each treatment comparison were obtained; these probabilities were then adjusted using a False Discovery Rate approach . This value, denoted in this paper as p
, therefore summarises the strength of evidence for there being a real difference in a way analogous to a standard p-value.
The foetal proliferation and IFN-γ data (PBMC and spleen) were analysed using non-parametric Kruskal-Wallis one way analysis of variance (ANOVA) using treatment as a grouping factor. No data was available for group 1 for the foetal thymus, (proliferation and IFN-γ ELISA) hence a two sample non-parametric Mann–Whitney test was conducted. All statistical analyses were carried out using GenStat 13th Edition software (VSN International, Hemel Hempstead, UK)