S. suis serotype (cps) 2 strain 10 is a virulent strain of sequence type 1 that has previously been used for pathogenesis studies by different groups (25, 26), kindly provided by H. Smith (DLO-Lelystad, The Netherlands) (27). Bacteria were grown overnight at 37 °C in Bacto™ Todd-Hewitt-Broth (THB) (BD, Heidelberg, Germany) or on Columbia agar plates with 6% sheep blood (Thermo Oxoid, Schwerte, Germany). S. suis glycerol stocks were prepared at the exponential growth phase (optical densitiy (OD)_{600 nm} = 0.5) and stored at −80 °C in 15% glycerol as single-use aliquots.

The recombinant Ide _Ssuis _homologue (rIde _Ssuis _h) is a truncated variant of the enzyme including the conserved IgM protease domain. Functionally it is equivalent to wild-type full-length Ide _Ssuis (25). Expression was induced in Escherichia coli BL21 (DE3), carrying the pETideSsuis_homologue plasmid, by the addition of 1 mM isopropyl-β-D-thiogalactopyranoside (IPTG) during the exponential growth phase and the protein was subsequently purified using Ni-affinity chromatography as described previously (16, 25).

Blood collection from conventional healthy post-weaning pigs (boar: Hypor Libra × sow: PIC408) naturally colonized by different S. suis serotypes, including cps 2, was approved under the permit number 81-02.04.40.2023.VG034 by North Rhine-Westphalia State Agency for Nature, Environment and Consumer Protection (Landesamt für Natur, Umwelt und Verbraucherschutz Nordrhein-Westfalen). Whole blood and serum samples from the same pigs at four and eight weeks were collected via the Vena cava cranialis into heparinized tubes (Monovette^®, Sarstedt, Nümbrecht, Germany) or serum tubes (Monovette^®, Sarstedt), respectively. Whole blood samples were centrifuged at 400×g for 10 min, and the plasma was removed. For isolation of peripheral blood mononuclear cells (PBMC), blood was diluted with an equal volume of phosphate buffered saline (PBS), layered onto Pancoll^® Separating Solution (density 1.077 g/mL, PAN-Biotech, Aidenbach, Germany) and centrifuged (900×g, 35 min at room temperature (RT), minimal acceleration and deceleration). Mononuclear cells from the interphase were collected and washed with medium (IMDM with L-Glutamine, 25 mM HEPES and 3.024 g/L NaHCO₃; PAN-Biotech) supplemented with 10% heat-inactivated fetal bovine serum (FBS, PAN-Biotech) (500×g, 12 min, RT), followed by another washing step with PBS (400×g, 10 min, RT). If required, lysis of remaining erythrocytes was performed by incubating cells at RT in 1-2 mL of lysis buffer (150 mM NH₄Cl, 8 mM KHCO₃, 2 mM EDTA in PBS, pH 7.0) for 2-3 min. Medium containing 10% FBS was added to stop the reaction. Following two washing steps (400×g, 6 min, RT), the cell number was determined using a Neubauer chamber (Laboroptik Lancing, UK), excluding dead cells via trypan blue staining (Merck-Sigma, Darmstadt, Germany). PBMC were used immediately for phenotyping of B cell subpopulations by flow cytometric analysis or cryo-preserved in liquid nitrogen until B cell sorting [freezing medium: 50% FBS, 40% IMDM, 10% dimethyl sulfoxide (DMSO, Merck-Sigma)].

Detection of anti-S. suis IgM and IgG in serum samples was performed as described previously (17). Briefly, Nunc MaxiSorp™ flat bottom plates (ThermoFisher, Darmstadt, Germany) were coated overnight (O/N) at 4°C with 1×10⁷ colony-forming units (CFU) of inactivated (0.2% paraformaldehyde O/N) S. suis cps 2 strain 10 per well in PBS. All incubations described below were followed by washing steps with PBST (PBS, 0.05% v/v Tween20, Carl Roth). After blocking with a combination of 0.5% w/v BSA and 0.1% v/v gelatine (Carl Roth) in PBS for 1 hour at 37°C, four-step serial dilutions of serum samples were added in duplicates and incubated for 1.5 hours at 37°C. Anti-S. suis IgG was detected using peroxidase-conjugated goat anti-swine IgG (50 ng/mL, Bethyl Laboratories, Montgomery, Texas), while IgM was detected using peroxidase-conjugated goat anti-swine IgM (100 ng/mL, Bethyl Laboratories). 3,3’,5,5’-Tetramethylbenzidine substrate (TMB, KPL, medac, Wedel, Germany) was added for colorimetric development. After 15 min (assay-optimized), the reaction was stopped with 1 M phosphoric acid (Carl Roth). The OD at 450 nm was measured using a Spectra-Max 340 ELISA reader with SoftMax Pro software v5.0 (Molecular devices, Munich, Germany). Blank-reduced ODs were converted to ELISA units using a standard serum from a pig prime-booster-vaccinated with S. suis cps 2 strain 10 as internal reference defined to include 100 ELISA units. The standard serum as well as the colostrum-deprived serum serving as negative control have been used previously (28).

Anti-S. suis IgM was determined in supernatants of stimulated B cell subpopulations (see sections 2.7 and 2.8) as described above for serum samples. Supernatants were used undiluted. Blank-reduced ODs were reported and compared.

For determination of total IgM in serum and supernatants of medium-incubated/stimulated B cell subpopulations, ELISA plates were coated with 2.4 µg/mL of monoclonal anti-porcine IgM (clone: K521C3) Biorad, Feldkirchen, Germany) in PBS. Supernatants of cells incubated in medium without stimuli were diluted 1:2, while serial dilutions of sera (starting from 1:10,000) as well as of supernatants from stimulated cells (CD40L plus IL-21 starting from 1:100; Pam3Cys-SK4 starting from 1:10) were used. Detection was performed as described above. Swine IgM whole molecule serum (Rockland Immunochemicals, Limerick, Pottstown) was used as a standard and total IgM concentrations were calculated based on the standard curve.

Survival of S. suis cps 2 strain 10 in porcine blood ex vivo was determined following the protocol of Seele et al., 2013 (25). Briefly, 500 μL of heparinized blood was mixed with 1×10⁶ CFU of exponentially grown bacteria. CFU were determined by serial dilutions at time zero and after incubation of the samples in a rotator for 2 hours at 37°C. The survival factor was defined as the ratio of CFU at 120 min to CFU at time zero. Where indicated, 10 µg of rIdeSsuis_h was added to evaluate the impact of IgM on bacterial survival in blood of pigs at four and eight weeks of age.

The frequency of conventional and non-conventional B cell subpopulations in peripheral blood was determined in the same pigs at four and eight weeks of age by flow cytometry. A fixable viability dye (eFluor 506, Thermo Fisher Scientific, Carlsbad, USA) was used according to the manufacturer’s protocol to allow for discrimination between viable and dead cells. To prevent binding via Fc receptors, cells were incubated with heat-inactivated normal serum derived from pig (30% in PBS). All surface staining steps described below were performed for 20 min on ice in the dark and separated by washing steps with FACS buffer (PBS, 3% FBS, 0.1% sodium azide (Merck-Sigma); 500×g, 3 min, 4°C). Detailed information regarding the applied primary and secondary antibodies for flow cytometric staining is provided in Table 1 . Initially, cells were incubated with anti-porcine CD11R1, anti-porcine IgG, and anti-porcine IgM. Detection of CD11R1 was achieved using a fluorescently labeled anti-mouse IgG1 secondary antibody ( Table 1 ). In the final surface staining step, cells were incubated with a mixture containing the remaining primary antibodies (anti-CD21, anti-CD3) as well as secondary antibodies for detection of IgM and IgG ( Table 1 ). For intracellular staining of the pan-B-cell marker CD79a ( Table 1 ), cells were permeabilized using the FoxP3/Transcription Factor Staining Buffer Set (Thermo Fisher Scientific, Carlsbad, USA) according to the manufacturer’s protocol. After permeabilization, an additional blocking step with 30% heat-inactivated porcine normal serum was done and cells were incubated with the staining antibody for 30 min at RT. Following acquisition with a BD LSR Fortessa™ flow cytometer (Becton Dickinson (BD), Heidelberg, Germany) equipped with BD FACS Diva ™ software version 6.1.3 (BD), data were analyzed using the FlowJo™10.10 software (BD). After exclusion of doublets and dead cells, PBMC were gated on CD79a⁺IgM⁺ B lymphocytes and the frequency of non-conventional IgM⁺CD21^-CD11R1⁺ and IgM⁺CD21^-CD11R1^- B-1-like subpopulations as well as of conventional IgM⁺CD21⁺ B-2 cells was analyzed. The frequency of IgG⁺ B cells was analyzed for comparison. Adequate gating was performed according to Fluorescence Minus One (FMO) controls included in the experiments.

The different conventional and non-conventional B cell subpopulations were isolated from PBMC of the same individual pigs at four and eight weeks of age by fluorescence-activated cell sorting (FACS). Viability dye and surface staining ( Table 1 ) were performed as described above, except for that FACS buffer was replaced by PBS, 3% FCS (without sodium azide) in the protocol. Stained cells were passed through a 30 µm filter (Sysmex Germany GmbH, Norderstedt, Germany) and immediately sorted through a 85 μm nozzle using a BD FACSAria™ III Cell Sorter (BD, Heidelberg, Germany) equipped with BD FACS Diva ™ software version 6.1.3. Four B cell subpopulations were collected: IgG⁺, conventional IgM⁺CD21⁺ B-2 as well as non-conventional IgM⁺CD21^-CD11R1⁺ and IgM⁺CD21^-CD11R1^- B-1-like cells. Purity was determined by post-sort re-analysis using FlowJo™ 10.10 software (BD).

The four sorted B cell subpopulations were seeded at a density of 5×10⁴ per well in culture medium (IMDM with L-Glutamine, 25 mM HEPES and 3.024 g/L NaHCO₃; PAN-Biotech) supplemented with 10% heat-inactivated FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin (both purchased from PAA Laboratories, Cölbe, Germany) containing 50 ng/µL porcine IL-2 (BioLegend, San Diego, United States) into a round-bottom 96-well cell culture plate (TPP, Trasadingen, Switzerland). To evaluate the IgM response of sorted B cell subpopulations to Toll-like-receptor (TLR) stimulation and T cell-associated signals, cells were stimulated with 10 µg/mL Pam3Cys-SK4 (InvivoGen, Toulouse, France) or 1 µg/mL human CD40L plus 50 ng/mL human IL-21 (both: BioLegend), respectively. To promote cell sedimentation, plates were centrifuged at 400×g for 2 min before start of culture (29). B cell subpopulations were incubated for 3 days at 37°C and 5% CO₂. The viability of the different B cell subpopulations was determined using a viability dye by flow cytometry after culture. Importantly, the viability of the different B cell subpopulations was comparable for the individual stimulation conditions allowing for a comparison of IgM secretion.

Statistical analysis was performed using GraphPad Prism 10.2.0 (GraphPad Software Inc., San Diego, USA). The Shapiro-Wilk test was applied to test for normal distribution. Normally distributed data are presented with means. Nonparametric data are shown with medians. Depending on normal distribution, the paired t-test or the Wilcoxon matched-pairs signed-rank test (both: two-tailed) were used for statistical analysis. Significance of stimulation-induced effects was calculated for each subpopulation by direct comparison of medium versus the CD40L plus IL-21 or Pam3Cys-SK4-stimulated equivalent. A Pearson correlation analysis was conducted between S. suis-binding IgM and total IgM. The level of confidence for significance is shown in figure legends.

Since invasive infections of S. suis mainly affects pigs at the age of five to ten weeks, understanding changes in the immune status after the forth week of life is of great importance. Here, in healthy, weaned pigs naturally colonized with different serotypes of S. suis, including cps 2, we analyzed anti-S. suis cps 2 antibody levels during this critical period.

While S. suis cps 2-binding IgM was low in 4-weeks-old pigs, it significantly increased until eight weeks of age ( Figure 1A ). In contrast, anti-S. suis cps 2 IgG decreased over the same period ( Figure 1A ), indicating stable IgM production without indication of isotype switching to IgG between four and eight weeks. These findings are in line with previous reports showing a similar course of anti-S. suis IgM and IgG levels in porcine blood for other serotypes (cps 1, 7, 14/1) (30, 31). Using an established in vitro whole blood bacterial survival assay (30), comparable survival of S. suis cps 2 was observed in the blood of the same pigs at four and eight weeks of age ( Supplementary Figure 1A ). To elucidate the role of IgM in control of S. suis cps 2 in porcine blood in vitro, Ide _Ssuis , a highly specific protease cleaving porcine IgM, but not IgG (25), was added in the whole blood bacterial survival assay. As expected from the low serum IgM levels ( Figure 1A ), incubation with the IgM protease did not alter the survival of S. suis cps 2 in blood of 4-weeks-old pigs ( Supplementary Figure 1B ). In contrast, presence of Ide _Ssuis significantly increased the survival of cps 2 in blood of the same pigs at the age of eight weeks ( Figure 1B ). In blood samples of five out of six 8-week-old pigs, even an increase in the survival factors above values of one was observed after addition of the IgM protease, indicating proliferation of S. suis cps 2 upon IgM cleavage ( Figure 1B ). Comparable survival of S. suis cps 2 in the blood of the same pigs at four and eight weeks of age ( Supplementary Figure 1A ) likely can be attributed to higher maternal anti-S. suis IgG levels in the former and higher anti-S. suis IgM levels in the latter ( Figure 1A ). In serum from 8-weeks-old pigs, levels of anti-S. suis IgM and total IgM correlated ( Supplementary Figure 2 ).

Taken together, our data indicate that the increase of anti-S. suis cps 2 IgM in porcine serum after the fourth week of life contributes to the restriction of survival and proliferation of the bacterial pathogen.

Next we were interested in the cellular source of anti-S. suis IgM. In mice, non-conventional B-1 cells produce protective IgM antibodies against S. pneumoniae (23). Here, we applied a multicolor flow cytometric staining protocol (24) to identify conventional and non-conventional B cell subpopulations in the peripheral blood of the same pigs at four and eight weeks of age ( Figure 2A ). PBMC from the same pigs whose serum IgM and IgG data are shown in Figure 1 were used. After excluding doublets and dead cells in flow cytometric analysis, we identified B cells using CD79a as a marker and analyzed four B cell subpopulations: non-conventional IgM⁺CD21⁻CD11R1⁺ and IgM⁺CD21⁻CD11R1⁻ B-1-like cells as well as conventional IgM⁺CD21⁺, and IgG⁺ B-2 cells. Our analysis revealed an increase in the frequency of non-conventional IgM⁺CD21⁻CD11R1⁺ (mean 1.8-fold) and IgM⁺CD21⁻CD11R1⁻ (mean 3.0-fold) B-1-like cells in porcine blood between four and eight weeks of age, while a relative decrease of conventional IgM⁺CD21⁺ B-2 cells was observed ( Figure 2 ). The relative proportion of IgG⁺ B-2 cells also decreased in this period ( Figure 2 ). Thus, the significant increase of S. suis cps 2-binding IgM in peripheral blood of four to eight weeks old pigs ( Figure 1 ) is associated with an increase of the frequency of non-conventional IgM⁺CD21⁻ B-1-like cells, most pronounced for the CD11R1⁻ subpopulation ( Figure 2B ).

To assess the potential of IgM secretion by the individual IgM⁺ B cell subpopulations, conventional IgM⁺CD21⁺ B-2 cells as well as non-conventional IgM⁺CD21⁻CD11R1⁺ and IgM⁺CD21⁻CD11R1⁻ B-1-like cell subpopulations were isolated by fluorescence-activated cell sorting from PBMC of 8-weeks-old pigs with high purity ( Figure 3A ). IgG⁺ B cells were isolated for control ( Figure 3A ). The sorted B cell subpopulations were incubated in medium for three days or stimulated with either the Toll-like receptor (TLR) 1/2 ligand Pam3Cys-SK4 to analyze innate B cell activation or, to characterize adaptive T cell dependent B cell activation, with the T cell-associated factors CD40L and IL-21 (32). Responsiveness to TLR stimulation is a well-known feature of B-1 cells (24, 33, 34) and murine B-1 cells have also been shown to respond to T cell signals in vitro (35). Of note, analysis of total IgM in supernatants of purified B cell subpopulations by ELISA revealed spontaneous IgM secretion under medium condition only by non-conventional IgM⁺CD21⁻CD11R1⁻ B-1-like cells ( Figure 3B ). Moreover, IgM⁺CD21⁻CD11R1⁻ B-1-like cells responded to TLR stimulation with highest IgM production ( Figure 3B ). Conventional IgM⁺CD21⁺ B-2 and non-conventional IgM⁺CD21⁻CD11R⁺ B-1-like cells also secreted IgM following TLR stimulation, albeit to a slightly lesser extent ( Figure 3B ). Similar to conventional IgM⁺CD21⁺ B-2 cells and in contrast to the IgM⁺CD21⁻CD11R1⁺ B-1-like subpopulation, IgM⁺CD21⁻CD11R⁻ B-1-like cells secreted high amounts of IgM in response to the T cell-associated signals CD40L and IL-21 ( Figure 3B ). As expected IgM was not detectable in supernatants from sorted IgG⁺ B cells ( Figure 3B ). Thus, IgM⁺CD21⁻CD11R1⁻ B-1-like cells are not only a potent source of induced IgM under innate and adaptive B cell stimulation conditions, but also constitutively secrete IgM conceivable with pronounced innate features of this B cell subpopulation.

Next, we investigated whether the constitutively produced IgM detected in the supernatant of IgM⁺CD21⁻CD11R1⁻ B-1-like cells and the stimulation-induced IgM detected in supernatants of conventional and non-conventional B cell subpopulations bind to S. suis cps 2. For that purpose, ELISA plates were coated with whole inactivated bacteria and supernatants analyzed for total IgM ( Figure 3B ) were additionally analyzed for binding to S. suis cps 2. As shown in Figure 4 , constitutively produced IgM binding to S. suis cps 2 was not detectable ( Figure 4 ) likely due to limited amounts of total IgM as compared with the >700-fold higher stimulation-dependent levels (see Figure 3B for difference between constitutive and induced total IgM secretion). Noteworthy, in contrast to conventional IgM⁺CD21⁺ B-2 cells, both non-conventional IgM⁺CD21⁻ B-1-like cell subpopulations secreted anti-S. suis IgM following stimulation with the TLR 1/2 ligand Pam3Cys-SK4 ( Figure 4 ). Furthermore, upon stimulation with the T cell-associated factors CD40L and IL-21, the non-conventional IgM⁺CD21⁻CD11R1⁻ B-1-like cell subpopulation demonstrated a higher potential to produce IgM that binds to S. suis cps 2 compared to non-conventional IgM⁺CD21⁻CD11R1⁺ and to conventional IgM⁺CD21⁺ cells ( Figure 4 ). Taken together, conventional and non-conventional B cell subpopulations of eight-weeks-old pigs are able to secrete anti-S. suis IgM with highest levels produced by IgM⁺CD21⁻CD11R1⁻ B-1-like cells.