In the experiment, 37 multiparous sows (line Topigs 40) of various parity from Czech farms were used. Nineteen sows (mean parity 3.5) were given 175 μg of cloprostenol, an analogue of prostaglandin F2α (PGF Veyx® forte, Veyx-Pharma GmbH, Germany) intramuscularly on day 114 of gestation (induced sows). Eighteen sows (mean of parity 3.1) were not treated (non-induced sows). All parturitions during which the samples of colostrum were taken occurred up to 24 hours after cloprostenol administration. Colostrum samples from non-induced sows were taken on day 114 or 115 of gestation.
5.2 Sample collection and cell isolation
Samples of colostrum (30 mL) were collected by manual milking at the time interval between the birth of the first and fourth piglet. Colostrum was collected as a mixed sample from different teats. First two front pairs of teats were excluded from sampling to allow suckling of piglets born so far. Also, the first rear pair of teats was excluded to avoid contamination of the sample by dirt from sow’s rear legs. The teats were cleaned with 30% ethanol (Sigma) before milking. Samples were stored at 4°C no longer than 2 hours till processed.
Samples were diluted 1:2 in PBS (Lonza) and centrifuged at 600 x g. Supernatants were stored at -80°C for cytokine, lysozyme and complement assays. Pellets were resuspended in PBS and centrifuged once more at 600 x g. Cells were counted (Coulter Counter M4, Coulter Cientifica) and transferred into U-bottomed 96-well microtiter plates (BD Biosciences) for flow cytometry immunostaining.
Details of mAbs and second-step reagents used in this study are summarized in Table 2. In each staining step the reagents were incubated for 20 min at 4°C. Staining of cell surface markers was performed in two steps: incubation with primary antibodies was followed by a second incubation with isotype-specific secondary antibodies. For the staining of lymphocyte subpopulations, two additional incubation steps for cell surface marker labelling were performed: free binding sites of the goat anti-mouse antibody were blocked by incubation with mouse serum (10 μl per sample). Thereafter, samples were incubated with Zenon labelled antibodies specific for CD3 (see Table 2 for details on antibodies and Zenon). Cell surface marker labelling was performed in a washing and staining buffer (WSB: PBS containing 0.2% gelatine from cold water fish skin, 1 g NaN3 and 1.84 g of Na4EDTA per litre, all reagents from Sigma–Aldrich). Following each incubation step, the cells were washed twice with 200 μl WSB. Prior to incubation and following every washing step, the cells were resuspended using a plate shaker. Moreover, single-stain samples for each fluorochrome present in various panels were prepared as compensation controls. Antigens with the highest expected expression density were selected as compensation controls. These were the following: CD3-PE, CD8-Alexa647, CD4-DyLight405, CD3-PE-TXRD and CD3-Alexa647.
5.4 FCM analysis
Data were acquired using a flow cytometer (BD FACS Aria Fusion; BD Biosciences, USA) and analysed with FACSDiva 6.0 software (BD Biosciences, USA). In each measurement 500,000 events were collected and a percentage of positive cells was analysed. Doublet events were discriminated on the basis of forward scatter integral versus forward scatter peak height (Fig. 1A). Dead cells were excluded from analysis using propidium iodide staining (Fig. 1B). Leucocytes were determined as CD45+ cells (Fig 1C). Granulocytes were determined as CD45+CD172α+swc8+ (Fig 1D), macrophages as CD45+CD172α+swc8- (Fig. 1D) and lymphocytes as CD45+CD172α-swc8-. Lymphocyte subpopulations were determined using a different staining panel. In this panel, light scatter characteristics were used for rough determination of the lymphocyte gate and exclusion of debris (Fig 2A). Total number (100%) of lymphocytes was determined as all CD3+ and/or CD2+ cells (Fig 2B). Afterwards, from the total number of lymphocytes, the percentage of γδTCR+ cells was determined (Fig 2C). According to Sinkora et al. (2005), γδTCR+ cells were subdivided into three subclasses based on CD2 and CD8 expression: CD2+CD8+, CD2+CD8-, CD2-CD8- (Fig 2D). Individual subpopulations of γδTCR- cells were defined as follows: Th cells: CD2+CD3+CD4+CD8-γδTCR-; DP cells: CD2+CD3+CD4+CD8+γδTCR- and Tc cells: CD2+CD3+CD4-CD8hiγδTCR- (Fig 2F). In addition to T cells, we also defined NK cells as CD2+CD3-CD4-CD8loγδTCR- (Fig 2E).
5.5 Lysozyme concentration determination by radial diffusion
The amount of lysozyme was assessed in vitro by radial diffusion in agarose gel mixed with Micrococcus luteus (CCM 169). Samples of 20 μl of acellular colostrum supernatant from individual sows were applied in duplicates into the wells cut in the agarose placed in glass plates and incubated at room temperature (20°C). After 24 h, diameter of diffusion zone was measured, and the amount of lysozyme in the sample was converted to milligram per milliliter of colostrum according to the calibration curve (for details see Poisot et al. (2009)).
5.6 Complement activity determination
The total bacteriolytic activity was determined using a modified bioluminescence-based method [42–44]. Transformed E. coli K12 with luxABCDE gene, originating from soil bacterium Photorhabdus, expressing bacterial luciferase (Lux)  was used. This Gram-negative bacterium is very sensitive to complement but not to lysozyme as checked before . Bacterial luciferase catalyzes the oxidation of a long-chain aldehyde and the reduced flavin mononucleotide (FMNH2) with emission maxima at 490 nm. Bacteria were exposed to acellular colostrum supernatant (diluted 1:2 in PBS) from individual sows. As a negative control, a randomly selected sample of acellular colostrum inactivated at 65°C for 10 minutes was used. Plasmid of bacteria contains genes for enzyme luciferase and its substrate — long-chain aldehyde. The light emission of the reaction is positively correlated with the viability of E. coli which was measured using an LM01-T luminometer (Immunotech, Czech Republic). The time (in hours) required for 50% viability of E. coli was evaluated (in triplicates) using kinetic curves corresponding to complement activity of each sample. There is a reciprocal proportion between time of E. coli viability and complement activity; the shorter time represents higher complement activity in colostrum of identical concentration. For better comprehension, the complement activity was expressed as inverted values (in h− 1).
Cytokine concentration was determined by the following ELISA kits: TGF beta 2 Pig ELISA kit, TNF alpha Pig ELISA Kit, Interleukin 6 Pig ELISA Kit, Interleukin 1β Pig ELISA Kit, Interleukin 4 Pig ELISA Kit. Interleukin 10 Pig ELISA Kit, TGF beta 2 Pig ELISA kit (all Abcam) and Granulocyte-Macrophage colony-stimulating factor (CSF2) ELISA Kit (CUSABIO). All assays were performed according to the manufacturers’ protocols. All the samples were run in duplicates.
5.8 Data analysis
Data are expressed as the mean ± SD. Statistical analysis was performed with the Mann-Whitney U-test (comparison of samples from induced and non-induced sows) using data analysis software (Statistica, StatSoft, Inc.). The obtained data were considered significant if P-values were P<0.05.