Conventional (CV) and germ-free (GF) piglets were used for the study. All animals were from herds with a good epidemiological situation. Large White sows were a source of CV piglets. The CV experiments were performed in two independent runs at the Veterinary Research Institute (Brno, Czech Republic). Piglets were nursed throughout the course of the experiment and were euthanized on days of life 0, 7, 14, and 28 by bleeding the brachial artery under deep anesthesia (tiletamine 2mg + zolazepam 2 mg + ketamine 2 mg + xylazine 2 mg/per kg of body weight). Samples from 6-month-old pigs were collected from animals exsanguinated after electrical stunning at the institutional experimental slaughterhouse. Animal treatment and use protocols were approved by the Ethics Committee of the Veterinary Research Institute in accordance with the guidelines of the Animal Protection Act and were subsequently approved by the Branch Commission for Animal Welfare of the Ministry of Agriculture of the Czech Republic.

A source of GF piglets were Minnesota miniature/Vietnamese–Asian–Malaysian crossbred piglets held at the Laboratory of Gnotobiology (Novy Hradek, Czech Republic). Piglets were recovered from gilts by sterile hysterectomy on day 112 of gestation as previously described (26–29). After recovery, GF piglets were kept in sterile isolator units under GF conditions and were hand-fed by γ-irradiated sterile 6%-fat cow milk reconstituted from evaporated 9%-fat concentrate by dilution with water (30). Piglets were euthanized by intracardiac puncture under general anesthesia on days 7 and 28 after the day of recovery. All these experiments were approved by the Ethics Committee of the Institute of Microbiology of the Czech Academy of Science according to the guidelines of the Animal Protection Act.

Blood was collected from the jugular vein and immediately heparinized using 25 IU/mL of sodium heparin (Zentiva). The spleen and mesenteric lymph nodes (MLN) were collected immediately after euthanasia. A single-cell suspensions were obtained from the tissues using a 100 µm cell strainer and then washed three times with phosphate buffered saline (PBS). Cells were used for stimulation at a concentration of 10⁶/ml in RPMI-1640 medium (Sigma–Aldrich) supplemented with antibiotics (100 IU/ml penicillin, 100 μg/ml streptomycin) and 10% fetal bovine serum (FBS). Blood was diluted 1:1 with RPMI-1640 medium supplemented with antibiotics. Samples were stimulated with 15 nM/ml phorbol-myristate-acetate (PMA; Sigma–Aldrich) and 1 μg/ml ionomycin (Sigma–Aldrich) for 5 hrs. For flow cytometry, samples were stimulated in the presence of a protein transport inhibitor, brefeldin A (10 μg/ml, Sigma–Aldrich). Unstimulated control samples were cultivated with brefeldin A only. After ammonium chloride-mediated lysis of erythrocytes and washing steps with PBS, samples were stained for flow cytometry or stored in Tri-Reagent RT (Molecular Research Center) for subsequent preparation for quantitative reverse transcription PCR (RT-qPCR).

Flow cytometry analysis was performed using a combination of IL-17 intracellular staining and surface markers. The primary mouse anti-pig antibodies were as follows: CD3ϵ (PerCP-Cy™5.5, BB23-8E6-8C8, BD Biosciences), CD2 (unconjugated, PG168A, IgG3, WSU), TCRγδ (unconjugated, PGBL22A, IgG1, WSU), CD4 (unconjugated, 10.2H2, IgG2b, WSU), CD8α (unconjugated, 76-2-11, IgG2a, WSU). As secondary antibodies, isotype-specific, fluorochrome-labelled goat anti-mouse antibodies were used: Alexa Fluor 488 (IgG2a, Invitrogen-Termo Fisher Scientific), Alexa Fluor 647 (IgG2b, Invitrogen-Termo Fisher Scientific), Dylight 405 (IgG1, BioLegend), Phycoerythrin (IgG1, Invitrogen-Termo Fisher Scientific). Samples were stained as previously described (13). Briefly, surface markers were stained with unconjugated primary and then fluorochrome-conjugated secondary antibodies. After a blocking step with 10% mouse serum, PerCP-Cy5.5-conjugated anti-CD3 antibody was added. Prior to the fixation step, the LIVE/DEAD™ Fixable Yellow Dead Cell Stain Kit viability probe (for 405 nm excitation) was added. Samples were then fixed and permeabilized using an Intrastain kit (Dako). During the permeabilization step, anti-human cross-reactive PE-conjugated anti-IL-17 antibody (PE, SCPL1362, BD Biosciences) or isotype control antibody (mouse IgG1 negative control PE-conjugated; BD Biosciences) was added. Flow cytometry was performed using an LSR Fortessa flow cytometer, operated by Diva software, version 6.0 (BD Biosciences). Doublets (defined by plotting the width against the area of forward scatter) and dead cells were excluded from the analysis.

Only samples stained with surface markers were used for fluorescence-activated cell sorting (FACS). Primary and secondary antibodies described above and mouse anti- pig CD27 (unconjugated, B30C7, IgG1, BioRad) were used. Dead cells and doublets were excluded from the sorting. Samples were sorted on the BD FACS Aria Fusion operated by Diva software, version 8.0 (BD Biosciences) under the following conditions: nozzle size 70µm, pressure 70 psi, frequency 87 kHz, four-way purity sort mask. The sorted cells were collected into 5 ml tubes containing RPMI-1640 medium supplemented with antibiotics and 10% FBS. After sorting, the tubes were centrifuged and the cells were resuspended in fresh medium for subsequent stimulation with PMA and ionomycin. After five hours of stimulation, cells were centrifuged and the pellet was lysed with Tri-reagent RT for further IL17mRNA analysis by qRT-PCR.

The isolation of RNA, reverse transcription, and RT-qPCR were performed according to previously described methods (31). Briefly, the RNA phase was isolated from Tri-Reagent RT homogenate mixed with bromoanisole with subsequent RNA purification using an RNeasy Mini Kit (Qiagen). RNA was then reverse transcribed with M-MLV reverse transcriptase (Invitrogen- Termo Fisher Scientific) using oligo-dT primers. Measurements were performed using the QuantiTect SYBR Green PCR Kit (Qiagen), LightCycler 480 II with 384-well plate block (Roche) and Innovadyne Nanodrop robot (IDEX Health & Science LLC). Hypoxanthine phosphoribosyltransferase 1 (HPRT1) was used as a housekeeping reference gene. The threshold cycle values (Ct) of the gene of interest (IL-17) was first normalized to the Ct value of HPRT reference mRNA (ΔCt) and then the normalized mRNA levels were calculated as [1/(2Ct target gene)]/[1/(2Ct HPRT)]. The normalized mRNA levels of IL-17 mRNA are expressed as “HPRT units”. Primers used in the study were adopted from Stepanova et al., 2012 (13).

Data were analysed using nonparametric ANOVA (Kruskal–Wallis test). Differences between CV and GF piglets were analysed using the nonparametric Mann-Whitney test. For both, p-values < 0.05 were considered to indicate significant differences. Statistical analysis was performed using GraphPad Prism software (GraphPad Software, version 5.04).

Changes in IL-17 levels during the early postnatal period were evaluated for blood cells and selected tissues immediately after birth (day 0) and 7, 14, and 28 days after birth. These data were compared with those from 6-month-old pigs. The ability of cells to produce IL-17 after polyclonal stimulation was measured at the mRNA level using RT-qPCR and by flow cytometry as the percentage of IL-17 positive cells. Differentiation was based on the five surface markers CD3/CD2/CD4/CD8/TCRγδ in combination with intracellular IL-17 staining after stimulation with PMA and ionomycin. The αβ T-helper (CD3⁺CD4⁺) and γδ T (CD3⁺TCRγδ⁺) cells were found to be the strongest IL-17 producers ( Figure 1A ). The total percentage of IL-17 positive cells was evaluated within all T cells (defined as CD3+) according to the gating strategy shown in supplementary data ( Figure S1 – gating strategy).

The significant increase in IL-17 was confirmed at both mRNA and protein levels during the early postnatal period for cells isolated from blood ( Figure 1B ), spleen ( Figure 1C ) and MLN ( Figure 1D ). The trend was similar in both blood and solid immune organs. The age dependent changes were evaluated by Kruskal-Wallis test (used to determine whether 5 different age groups are significantly different from each other on the variable “HPRT units of IL-17 mRNA” and “percentage of IL17+CD3+ cells”) with following results for HPRT units of IL-17 mRNA (blood: p < 0.0001, spleen: p = 0.0014 and MLN: p < 0.0001) and for percentage of IL17⁺CD3⁺ cells (blood: p < 0.0001, spleen: p = 0.0014 and MLN: p < 0.0001). An increasing trend apparently continued during aging with a substantial augmentation between 28-day-old and 6-month-old pigs. Relatively high individual variability was observed, especially in older animals.

The strongest IL-17 producers αβ T-helper (CD3⁺CD4⁺) and γδ T (CD3⁺TCRγδ⁺) cells were evaluated in the context of early postnatal ontogenesis. The percentage of IL-17 producing cells was measured by flow cytometry according to the gating strategy shown in Supplementary Data ( Figure S1 ). The data obtained demonstrated a significant increase in IL-17 producing αβ T-helper cells in blood, spleen, and MLN ( Figure 2A ). Statistical significance was found in all compartments analysed (Kruskal-Wallis, p < 0.0001). These IL-17 positive αβ T-helper cells were almost undetectable in newborn CV piglets. The first evidence of these cells occurred on day 7. The obtained data showed relatively low individual variability during the early postnatal period. The increase in IL-17 positive cells was gradual until day 28, followed by a substantial increase detectable in the group of 6-month-old pigs. In contrast to αβ T-helper cells, IL-17 positive γδ T cells were detectable in newborn piglets and exhibited different kinetics during early postnatal development ( Figure 2B ). A gradually increasing trend was found only in MLN (Kruskal-Wallis, p = 0.0129). The Kruskal-Wallis test also revealed significant differences in blood (p = 0.0033) and spleen (p = 0.0415). However, the percentage of IL-17 positive γδ T cells varied within age categories. Relatively high levels were found in spleen and blood at day 7, but low frequencies were detectable in these compartments at day 28, followed by an increase in 6 months-old pigs. The data on IL-17 positive γδ T cells were characterized by high individual variability. In addition, the phenotype of IL-17 producing γδ T cells was further analysed based on the expression of CD2, and only CD3⁺TCRγδ⁺CD2^- subpopulation was found as IL-17 positive ( Supplementary Data , Figure S2 ).

The differentiation/activation status of IL-17 producing αβ T-helper cells was also assessed by CD8α and/or CD27 expression. The combination of sorting followed by RT-qPCR was used because flow cytometry data did not provide entirely clear results about the IL-17 positivity of particular subpopulation. The following subpopulations of CD3⁺CD4⁺ cells were sorted according to the gating strategy shown in Figure 3A : CD27⁺CD8^-, CD27⁺CD8⁺, and CD27^-CD8⁺. Sorted subpopulations were then stimulated with PMA and ionomycin, and IL-17 transcripts were measured using RT-qPCR. The data obtained clearly showed that specifically the CD8 positive subpopulation was a source of IL-17 in all monitored compartments ( Figure 3B ). The IL-17 transcripts were almost undetectable in CD4⁺CD8^-CD27⁺ subpopulation. Populations with both CD4⁺CD8⁺CD27^- and CD4⁺CD8⁺CD27⁺ phenotypes were proved as IL-17 sources in blood and spleen. In MLN, CD4⁺CD8⁺CD27^- cells exhibited higher IL-17 mRNA levels compared with their CD4⁺CD8⁺CD27⁺ counterparts.

Data from CV pigs were compared with data from GF piglets to demonstrate the role of the microbiota in the induction of IL-17 producing T cells. One-week-old (day 7) and one-month-old (day 28) piglets were included in this part of the study. The GF conditions had a significant impact on the induction of IL-17. The data showed a clear impact of microbiota on the postnatal development of both IL-17 positive αβ T-helper (CD3+CD4+) and γδ T (CD3+TCRγδ+) cells ( Figure 4 ). In contrast to the data from CV piglets, IL-17 positive αβ T-helper cells were almost undetectable at both time points, day 7 and day 28 postpartum ( Figure 4A ). This trend was evident in all compartments examined. The similar impact was observed for γδ T cells ( Figure 4B ). Significant differences between CV and GF piglets evaluated by Mann-Whitney test were found in blood and spleen at day 7, and in spleen and MLN at day 28.