Vaccination with the immunoglobulin M-degrading enzyme of Streptococcus suis, Ide Ssuis, leads to protection against a highly virulent serotype 9 strain.

Vaccination of weaning piglets with the recombinant IgM degrading enzyme of Streptococcus suis (S. suis), rIde Ssuis, elicits protection against disease caused by serotype (cps) 2 infection. In Europe, S. suis cps9 is at least as important as cps2 in causing severe herd problems associated with meningitis, septicemia and arthritis. The objective of this study was to determine humoral and cellular immunogenicities of rIde Ssuis suckling piglet vaccination and to investigate protection against a virulent cps9 strain. Vaccination in the 2nd and 4th week of life with rIde Ssuis and an oil-in-water adjuvant induced seroconversion against Ide Ssuis in 13 of 20 vaccinated piglets. In the 5th week, survival of the S. suis cps9 strain was significantly reduced in the blood of prime-booster vaccinated piglets. After a 2nd booster vaccination Ide Ssuis -reactive T helper (Th) cells partially producing TNF-α, IL-17A or IFN-ɣ were detectable in rIde Ssuis -vaccinated but not in placebo-treated piglets and frequencies of Ide Ssuis -reactive Th cells correlated with α-Ide Ssuis-IgG levels. An intravenous challenge, conducted with a cps9 strain of sequence type (ST) 94, led to 89% mortality in placebo-treated piglets due to septicemia and meningitis. In contrast, all rIde Ssuis prime-booster-booster vaccinated littermates survived the challenge despite signs of disease such as fever and lameness. In conclusion, the described rIde Ssuis vaccination induces humoral and detectable Ide Ssuis -reactive Th cell responses and leads to protection against a highly virulent cps9 strain.

Introduction

S. suis cps9 has become the most important serotype in some European countries with a huge pig industry such as Spain and the Netherlands [1], [2]. In the field, prophylaxis against cps9 is very problematic, because vaccination with a bacterin protects against mortality, but not morbidity [3] and does not reduce colonization and transmission [4]. Furthermore, cps9 is a strong biofilm inducer and endocarditis is a main manifestation, which might also occur in vaccinated piglets [3], [5], [6].

A few S. suis proteins have been shown to provide protection against cps2 challenge in pigs, including a combination of muramidase-released protein (MRP) and extracellular factor (EF) [7], surface antigen one (SAO) [8], HP0197 [9], SsPepO [10] and Ide [11]. Interestingly, only SAO has been shown to protect against a further serotype, namely cps1 [12]. However, most cps1 strains are closely related to cps2. Strains of both serotypes often belong to clonal complex (CC) 1 [13]. Noteworthy, other independent studies have shown that piglets with high antibody titers against SAO are highly susceptible to S. suis challenge and that these antibodies are not opsonizing [14], [15]. Thus, it is currently unknown whether any of the identified protective antigens has the cross-protective potential needed for a universal S. suis vaccine.

Different S. suis serotypes express a highly specific immunoglobulin M-degrading enzyme, designated Ide [16]. The protein is homologous to the IgG protease IdeS of S. pyogenes [17], but cleaves solely class M antibodies of swine. Mutants expressing no or a point-mutated Ide variant deficient in IgM cleavage show enhanced deposition of C3b on the bacterial surface indicating that IgM cleavage by Ide is involved in complement evasion [18], [19]. Vaccination of weaning piglets with rIde elicits antibodies neutralizing the IgM protease activity and protects against morbidity and mortality induced by cps2 challenge [11].

Read out parameters in vaccination trials with piglets focused on humoral immunity. However, cellular immunity is known to be crucial to restrict colonization of the respiratory tract by the related pathogen S. pneumoniae [20], [21]. Mice lacking CD4 were found to be more susceptible to systemic S. suis infection at a low infection dose [22], whereas in pigs the T cell immune response is not well characterized.

Here, we investigated humoral and cellular immunogenicities of Ide vaccinated suckling piglets, being the age class of choice for vaccination in the field. Furthermore, we asked if immunization with Ide protects against S. suis cps9, the most important and troublesome pathotype in Europe.

Materials and methods

Bacterial strains, growth conditions and profiling of virulence-associated factors

S. suis strain A3286/94 is a mrp* sly+ cps9 strain belonging to ST99 of CC16 originally isolated from a pig with meningitis [3], [15], [23], [24]. Strains 15-3/3, V5404/2 and 16085/3b were originally isolated from inner organs of diseased piglets with meningitis and/or septicemia and identified as mrp in a described multiplex (MP) PCR [25]. S. suis strain 10 (mrp+ epf+ sly+) is a virulent serotype 2 strain [11], [15], [26]. The isogenic mutant 10Δide deficient in IgM cleavage was included in the bactericidal assay to reveal effects mediated by antigen-specific immunity [11], [16]. S. suis was grown on Columbia agar plates supplemented with 6% sheep blood or in BactoTM Todd Hewitt broth (THB). Escherichia coli (E. coli) strains were cultured in Luria-Bertani (LB) medium including 100 µg/ml ampicillin, if appropriate.

Expression and purification of recombinant (r) proteins

The expression and the purification of His-tagged rIde of serotype 2 strain 10 [16] and His-tagged fibronectin binding domain of streptococcal fibronectin-binding protein I of Streptococcus pyogenes (SfbI) [27] was performed as described previously [16]. SfbI was chosen as a control protein as Streptococcus pyogenes is not found in pigs.

Animal experiments

Piglets were infected experimentally and cared for in accordance with the principles outlined in the EU Directive 2010/63/EU (http://ec.europa.eu/environment/chemicals/lab_animals/legislation_en.htm). All animal experiments or samplings were conducted by veterinarians and in accordance with the principles outlined in the European Convention for the Protection of Vertebrate Animals Used for Experimental and other Scientific Purposes and the German Animal Protection Law (Tierschutzgesetz). The animal experiment of this study was approved by the Landesdirektion Sachsen (permit no. TVV28/16), which includes approval through the registered committee for animal experiments. The collection of blood samples was approved by the Landesdirektion Sachsen (permit no. N01/16 and N19/14).

Five weeks prior to farrowing four pregnant sows of a German Landrace pig herd considered to be free of cps9+ S. suis strains were transported to the Faculty of Veterinary Medicine, Leipzig University. The classification as cps9 free was based on the genotyping results including a described MP-PCR [25] of S. suis isolates from the tonsils of more than 400 animals over the last 14 years. After birth the suckling piglets were treated as follows: Piglets of one litter were equally distributed to the vaccination and the placebo group with a final number of 20 piglets per group. At an age of 2 weeks a vaccine containing rIde as antigen or a placebo was applied intramuscularly, both supplemented with 20% [vol/vol] Emulsigen as adjuvant. The piglets were boostered 14 days later. Nine piglets per group were boostered a second time another 14 days later. One dose of rIde vaccination contained 0.4 mg rIde. After the first booster vaccination at an age of 4 weeks, piglets were weaned and moved to the trial pen. All piglets (n = 18) used for experimental infection were placed in one pen. The person (CGB) who made the final decision of euthanasia did not know whether a specific piglet was vaccinated with rIde or placebo-treated (partially blinded experiment). Piglets were challenged intravenously at an age of 8 to 9 weeks (17 days after the second booster) with 2 × 108 CFU of S. suis strain 16085/3b grown in Bacto™ Tryptic Soy Broth (TSB) without dextrose. Post infection, the health status of the animals was monitored every 8 h, including measurement of the inner body temperature, assessment of movements and feed intake (piglets were only fed at these time points). Based on predefined criteria (Table S1) clinical signs were scored. Piglets were classified as morbid if a body temperature of ≥40.2 °C or/and severe clinical signs of an acute disease were observed. In case of high fever (≥40.5 °C), apathy and anorexia persisting over 32 h as well as in all cases of central nervous system dysfunction or clinical signs of acute polyarthritis, animals were euthanized for animal welfare reasons. All surviving piglets were sacrificed 14 days post infection (dpi). After euthanasia every animal went through the same procedure of necropsy to collect the following samples for histological (h) and semi-quantitative bacteriological (b) investigations as described previously [3], [15]: cerebrospinal fluid (b); brain (b, h); tarsal and carpal joints (b, h), peritoneal, pleural and pericardial swabs (b), peritoneum, pleura and pericardium (h); cranial lobe of the left lung (b, h); liver (b, h); spleen (b, h); mitral valve (b, h) and tonsil (b, h). The histological screenings were scored as described [28] and briefly mentioned in the footnotes of Table 2. Isolation of the challenge strain was confirmed by MP-PCR detecting mrp, epf, sly, arcA, gdh, cps1, cps2, cps7 and cps9 [25].

Table 2

Scoring of fibrinosuppurative lesions of piglets challenged with S. suis cps9 strain 16085/3b.

Immunization antigen	Piglets without lesionsa	Piglets with lesions in two or more locationsa	Brain			Serosae			Joint			Spleen and liver			Lung			Heart
			meningitis, chorioiditis			pleuritis or peritonitis or pericarditis			synovialitis			splenitisb or hepatitis			pneumonia			endocarditis
			5c	3d	1e	4c	2d	1e	4c	2d	1e	4c	2d	1e	4c	2d	1e	4c	2d	1e	ωf
placebo	0/9	6/9	3/9	1/9g	0/9	0/9	3/9	1/9	1/9	2/9	0/9	0/9	6/9	1/9	0/9	0/9	1/9	0/9	3/9	1/9	3.0 h
rIdeSsuis	3/9	2/9	0/9	0/9	2/9	0/9	1/9	0/9	1/9	3/9	0/9	0/9	0/9	6/9	0/9	0/9	0/9	1/9	2/9	2/9	2.0

Only fibrinosuppurative lesions are considered. Individual single perivascular neutrophils are not counted.

Neutrophilic accumulation of the splenic red pulp.

Scoring of 4 and 5 indicates moderate to severe diffuse or multifocal fibrinosuppurative inflammations.

Scoring of 2 and 3 indicates mild focal fibrinosuppurative inflammation.

Individual single perivascular neutrophils received a score of 1.

ω = Σscoremax/nanimals.

diffuse, mild plexus choroiditis.

Four of the nine placebo-treated piglets died within 8 h after experimental infection. These piglets reached comparable low histological scores of 2 (n = 3) and 3 (n = 1).

Bactericidal assay

Survival of S. suis in porcine blood ex vivo was determined as previously described [11]. Briefly, 500 µl of heparinized blood (16 I. U. heparin/ml) was mixed with 6 × 105 CFU of exponentially grown bacteria (OD600: 0.5–0.6). The samples were incubated for 2 h at 37°C on a rotator. Blood for bactericidal assays was drawn from all piglets of the vaccination trial 7 and 17 days after first and second booster vaccinations, respectively. The bactericidal assays were conducted within 4 h after blood collection. The specific bacterial contents in CFU/ml were determined by plating serial dilutions at t = 0 min and t = 120 min and the survival factor of S. suis for each sample was calculated by dividing the two values.

Detection of anti (α) -Ide IgG

The detection of α-Ide IgG was performed as previously described [11]. Sera of piglets immunized with rIde and a truncated derivative (rIde_homologue) in the previous study served as reference serum and positive control in the α-Ide ELISA, respectively.

Detection of Ide-reactive Th-cells

Peripheral blood mononuclear cells (PBMCs) were isolated from heparinized blood samples by density gradient centrifugation on Biocoll (Merck-Biochrom, Berlin, Germany) and cryo-conserved at −80 °C until usage for restimulation. PBMCs were cultivated in 96-well plates (1 × 106 cells/well) in complete Iscove's Modified Dulbecco's Medium (IMDM, Pan Biotech, Aidenbach, Germany) containing 10% FCS (Gibco) and penicillin/streptomycin (100 U/ml and 100 µg/ml, respectively; Merck-Biochrome, Berlin, Germany). Antigen-specific restimulation was conducted with 5 µg/ml rIde or 5 µg/ml His-tagged rSfbI [27] as a non-relevant control antigen. Both recombinant antigens were expressed in E. coli and purified via Ni-affinity chromatography. rSfbI was included as a control to estimate background reactivity. PMA/Ionomycin (10 ng/ml and 500 ng/ml, respectively, both purchased from Merck-SIGMA-Aldrich, Taufkirchen, Germany) were used as a positive control added for the last 4 h of incubation time and medium alone was used as a negative control. PBMCs were stimulated for 18 h in presence of 2 µg/ml Brefeldin A (Enzo Life Science, Lörrach, Germany) for the last 14 h at 37°C, 5% CO2 in a humidified atmosphere. To label cells for flow cytometry analysis, the stimulated cells were harvested and stained as described in the section flow cytometry and as previously shown [29]. Measurement of samples was conducted at LSR-Fortessa™ (Beckton Dickinson, Germany, Heidelberg) recording 3–5 × 105 viable CD3+CD4+ cells. The frequency of antigen-reactive Th cells (CD154+) was determined from the antigen-experienced Th cell population, described as CD3+CD4+CD8α+ cells. Frequency of Ide-induced Th cells was calculated as the difference of the percentage of CD154+ cells after Ide re-stimulation minus the percentage of CD154+ cells of medium cultivated cells from the same sample.

Flow cytometry staining

To label antigen-reactive Th cells for flow cytometry analysis 6x106 PBMCs per stimuli were used after antigen-specific restimulation. Cells were washed two times with PBS before addition of fixable viability dye eFlour™ 506 (1:500, Thermo Fisher Scientific, San Diego, CA). For extracellular staining anti-porcine CD3-PE ~ Cy7 (clone: BB23-8E6-8C8), anti-porcine CD4-PE (clone: 74–12-4) and anti-porcine CD8α-FITC (clone: 76-2-11), all purchased from Beckton Dickinson (BD, Heidelberg, Germany), were used. Following fixation (2% paraformaldehyde) the cells were permeabilized in PBS buffer containing 3% FCS, 0.1% NaN3 and 0.5% saponin for intracellular staining with anti-human CD154-VioBlue (clone: 5c8, Miltenyi, Colone, Germany), anti-porcine IFN-γ-PerCP ~ Cy5.5 (clone: P2G10, BD), anti-human IL-17A-AlexaFlour®647 (clone: SCPL1362, BD) and anti-human TNF-α-BV421 (clone: MAb11, BioLegend, San Diego, CA).

Statistical analysis

The evaluation of more than two groups was carried out using one- or two- way analysis of variance (ANOVA) or Kruskall-Wallis with a subsequent Tukey’s or Dunn’s multiple comparisons test, respectively. Differences between two groups were analyzed with the Mann-Whitney U test. Correlation was calculated with the Pearson test. The Wilcoxon test was used for comparison of different time point values within the same group in case of no more than two repeated measures. The data presented in the Kaplan-Meier-diagrams were analyzed with the log rank test. Means and standard deviations of the results are shown. All statistical tests were conducted with GraphPad Prism 7.01 software. Probabilities lower than 0.05 were considered significant (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Results

S. suis 16085/3b exhibits increased bacterial survival in porcine blood in comparison to other cps9 strains

In Europe, S. suis cps9 is a major porcine pathogen causing herd problems associated mainly with meningitis, endocarditis and arthritis [2], [25]. Recently, a herd experienced an escalating S. suis cps9 problem in growing piglets due to meningitis and septicemia associated with cyanosis and sudden death. As the extent and severity of disease on this farm appeared extreme, the S. suis cps9 strain 16085/3b isolated from the spleen of a pig from this farm was compared to other S. suis cps9 strains by profiling of virulence-associated factors, multi locus sequence typing (MLST) and analysis of bacterial survival in porcine blood. Interestingly, strain 16085/3b is a mrp+ sly+ cps9+ strain of ST94 and thus genetically distinct to other invasive cps9 strains. As shown in Fig. 1 S. suis cps9 strain 16085/3b shows significantly higher bacterial survival factors than the other three tested cps9 strains of ST16 (V5402/2 and 15-3/3) and ST99 (A3286/94) in blood from cps9 free piglets at an age of 6.5, 7.5 and 8.5 weeks. In addition, for the 16085/3b strain we observed a significant increase of bacterial survival in blood of 8.5 week-old (mean = 2.8, SD = 3.3, n = 5) piglets in comparison to 4.5 week-old (mean = 0.9, SD = 0.6, n = 5) and 5.5 week-old piglets (mean = 1.4, SD = 0.5, n = 5) (Fig. 1). Due to this phenotype, we chose strain 16085/3b as challenge strain for this vaccination study, since previous experimental infection studies with other cps9 strains led to rather low rates of mortality making it difficult or even impossible to draw conclusions on protection against mortality [3], [4].

Fig. 1

Strain 16085/3b shows increased survival in porcine blood compared to other cps9 strains. Bactericidal assays were conducted with S. suis cps9 strains A3286/94, 15-3/3, V5404/2 and 16085/3b belonging to the indicated sequence types (ST) with blood drawn from the same S. suis cps9 free piglets (n = 5) at the specified ages. Note, that data of strain 16085/3b was published recently [36]. The survival factor represents the ratio of CFU at 120 min to CFU at time zero. Bars and error bars represent mean values and standard deviations, respectively. Significant differences were determined using two-way ANOVA and a subsequent Tukey’s multiple comparisons test. Significances are indicated (*p < 0.05, **p < 0.01, ***p < 0.001).

Early prime-booster vaccination of piglets in the 2nd and 4th week with rIde elicits specific IgG antibody titers and bactericidal immunity against S. suis cps9 strain 16085/3b

Induction of early protective immunity against S. suis by suckling piglet vaccination would be advantageous over weaning piglet vaccination, as diseases might occur after weaning. Thus, we investigated humoral immunogenicities of early prime-booster rIde vaccination in the 2nd and 4th week of life, respectively. Prior to vaccination in the 2nd week of life, all suckling piglets had α-Ide IgG levels below 5 ELISA units (mean = 1.7; SD = 0.9; n = 20, Fig. 2). Seven days after the 1st booster vaccination, at an age of 5 weeks, a mean of 20.9 ELISA units (SD = 14.4, n = 20) α-Ide IgG was recorded in vaccinated piglets, which is significantly higher than the antibody level found in their placebo-treated littermates (mean = 0.5 ELISA units, SD = 0.2, n = 20; Fig. 2). However, 7 of 20 vaccinated piglets showed no or only a low increase in α-Ide IgGs with a difference of less than 10 ELISA units between the pre and post vaccination-serum.

Fig. 2

rIde vaccination induces specific IgG antibodies in serum. Time course of α-Ide IgG antibodies in rIde-vaccinated and placebo-treated piglets during prime-booster-booster vaccination and after S. suis cps9 infection. IgG levels were determined in the 2nd week of life (pre immunization) and 7 days after the first booster in rIde-vaccinated (○, ●) and placebo-treated (■, □) piglets (n = 20/group). In the 6th week of life, nine piglets per group were boostered (●) or placebo-treated (□), respectively, for a second time and challenged 17 days later. Blood samples taken from euthanized piglets are indicated as final bleeding and include the time period from 1 until 14 dpi. Mean values are indicated by horizontal lines, standard deviations by error bars. Statistical analyses were conducted with the Mann-Whitney U test (placebo-treated vs. rIde-vaccinated), the Wilcoxon test (comparison of pre immunization and 7d post 1st booster immune sera, n = 20/group) or the one-way ANOVA and subsequently the Dunn’s multiple comparisons test (pre and post immune sera of the 18 piglets challenged with S. suis cps9). Significant differences are indicated (**p < 0.01, ****p < 0.0001).

A bactericidal assay conducted 7 days after first booster vaccination revealed significantly lower survival factors of the cps9 strain 16085/3b in the blood of the vaccinated piglets (mean = 0.3, SD = 0.2, n = 20) compared to the blood of placebo-treated piglets (mean = 0.5, SD = 0.3, n = 20; Fig. 3). The cps2 wt strain 10 also had a lower survival factor in the blood of vaccinated piglets (mean = 0.2, SD = 0.3, n = 20) in comparison to placebo-treated piglets (mean = 0.3, SD = 0.5, n = 20), though differences were not significant (Fig. 3). In conclusion, an early prime-booster Ide vaccination (in the 2nd and 4th week of life) leads to an increase of α-Ide-IgG in serum at an age of 5 weeks which is associated with reduced survival of S. suis in blood.

Fig. 3

Survival of S. suis cps9 strain 16085/3b is significantly reduced in blood of rIde-vaccinated piglets (n = 20) in comparison to placebo-treated littermates 7 days after the first booster vaccination. Bacterial survival of cps2 strain 10 and its isogenic mutant 10Δide was also determined in the blood of the same piglets. Bars and error bars represent mean values and standard deviations, respectively. The Mann-Whitney-U test was used for comparison of rIde-vaccinated versus placebo-treated group. Two-way ANOVA and Tukey’s multiple comparisons post-hoc test were used for comparison of survival factors of the three S. suis strains per group. Significant differences are indicated (***p < 0.001, ****p < 0.0001).

Vaccination with rIde protects growing piglets against mortality caused by S. suis cps9 infection

In contrast to the results of a previous vaccination trial with weaning piglets [11], early prime-booster vaccination with rIde did not elicit a prominent systemic IgG response against rIde in all vaccinated piglets. Piglets with a prominent IgG response against rIde following prime-booster vaccination, which additionally showed neutralization of the IgM protease in Western blot analysis (as described previously by Seele et al., 2015), were chosen for another experiment not dealing with cps9 and thus not part of the study described here. A 2nd booster was applied to the remaining nine prime-booster vaccinated piglets which apart from two animals (>40 α-Ide IgG ELISA units) had responded poorly (<22 α-Ide IgG ELISA units) to the vaccine in terms of seroconversion against rIde (Table S2). The 2nd booster led to a significant increase of the α-Ide IgG level to a mean of 64.4 ELISA units (SD = 43.9, n = 9) 17 days after the 2nd booster in the 8th week of life compared to a mean of 18.2 ELISA units (SD = 15.3, n = 9) 7 days after the 1st booster in these nine piglets (Fig. 2). These prime-booster-booster vaccinated piglets (n = 9) were used for the challenge experiment. A bactericidal assay conducted after the 2nd booster immunization suggested killing of the cps9 strain with a mean survival factor of 0.6 (SD = 0.6) in the rIde-immunized group in contrast to substantially higher survival in placebo-treated littermates with a mean survival factor of 1.3 (SD = 1.15), respectively, although these differences were statistically not significant (Fig. S1). The intravenous challenge with the cps9 strain 16085/3b was conducted on the same day to test protection against this important serotype. A clinical scoring system was applied as shown in Table S1. Six of nine placebo-treated piglets died or were killed after reaching a clinical score of 25 within 32 h after experimental infection (Fig. 4). Four of them displayed a peracute course of disease with the inability to rise, pain vocalization, vomiting and apathy within less than 8 h after challenge. Further two showed signs of central nervous system dysfunction including convulsions, opisthotonus and tremor the day after challenge (Table 1). Another placebo-treated piglet showed signs of central nervous system dysfunction as late as 14 dpi. In contrast, none of the nine vaccinated piglets died or showed signs of acute septicemia, central nervous system dysfunction or the inability to rise in the observation period after the cps9 challenge. Only one vaccinated piglet received a clinical score above 10 after experimental infection. However, short time fever during the first three days following the challenge were recorded in seven of nine vaccinated piglets (Fig. S2) and lameness was observed in four of nine piglets. These piglets recovered within 24–32 h in all cases except for one. Thus, prime-booster-booster rIde vaccination protects piglets against mortality but not morbidity in this intravenous cps9 challenge model.

Fig. 4

rIde vaccination protects against mortality caused by S. suis cps9 challenge. The Kaplan-Meier diagram shows mortality of rIde prime-booster-booster (n = 9) and placebo-treated (n = 9) growing piglets after infection. Piglets were challenged through intravenous application of 2 × 108 CFU of S. suis strain 16085/3b (mrp* sly+cps9) 17 days after 2nd booster immunization. In case of high fever (≥40.5 °C), apathy and anorexia persisting over 32 h as well as in all cases of central nervous system dysfunction or clinical signs of acute polyarthritis, animals were euthanized for animal welfare reasons. All surviving piglets were sacrificed 14 days post infection (dpi). Statistical analysis was conducted with the log-rank test (p value is shown below the diagram).

Table 1

Assessment of the protection induced by the indicated treatments against morbidity and mortality after intravenous S. suis serotype 9 challenge.

Immunization antigen	Morbidity	Mortality	Mean clinical scoree (SD)	Clinical signs			Max. body temperature (°C)
				CNSa	Lameness	no feed intakee	<40	≥40 and ≤40.2	>40.2
placebo	8/9	8b/9	21 (8.1)	4/9	3c/9	4(6)d/9	1d/9	0d/9	6d/9
rIdeSsuis	8/9	0/9	6 (3.9)	0/9	4/9	1/9	1/9	1/9	7/9

Signs of central nervous system (CNS) dysfunction such as convulsions and opisthotonus.

One placebo-treated piglet showed central nervous system dysfunction and reached a score of 25 on the last day of the experiment.

Three additional piglets became recumbent.

Two piglets died a few hours after experimental infection prior to the first feeding and measurement of body temperature.

For the detailed clinical scoring system see Table S1.

The histological screening revealed that six of nine placebo-treated piglets had typical fibrinosuppurative lesions in at least two inner organs (Table 2). Accordingly, S. suis cps9 was detected in seven of nine placebo-treated piglets in two or more inner organs indicating severe bacteremia or infection of multiple organs after challenge (Table 3). The pathological score ω was substantially lower in rIde-vaccinated piglets (2.0 versus 3.0) as moderate and severe fibrinosuppurative lesions were not detected in the brain of these animals (Table 2). Furthermore, the challenge strain was isolated from an inner organ of only one rIde-vaccinated piglet (Table 3). In conclusion, recording of fibrinosuppurative inflammations and bacteriology indicated protection in prime-booster-booster rIde-vaccinated piglets against meningitis and bacterial dissemination.

Table 3

Reisolation of the challenge strain from piglets after intravenous challenge with S. suis cps9 strain 16085/3b.

Immunization antigen	Number of piglets positive for the isolation of the challenge strain in an inner organb or in serosa or in joint fluid	Number of piglets positive for the isolation of the challenge strain in ≥ 3 inner organsb	Number of piglets in which the S.suis challenge straina was isolated from
			Tonsils	Lungc	Serosad	Spleen	Liver	Brain, CSFe	Joint fluidf	Endocard
placebo	8/9	7/9	0/9	5/9	4/9	7/9	5/9	7/9	6/9	6/9
rIdeSsuis	1/9	0/9	0/9	0/9	0/9	0/9	1/9	0/9	0/9	1/9

The challenge strain was identified by PCR.

Inner organ refers to lung, spleen, liver, brain, CSF or endocard but not the tonsils.

One cranial lobe was investigated.

Pleural, peritoneal or pericardial cavity.

Cerebrospinal fluid.

Punctures of both tarsal and carpal joints were investigated in each animal. In case of lameness additional joint punctures of the respective limb were screened.

Prime-booster-booster rIde vaccination elicits a detectable antigen-reactive Th cell response

We recently described a method to detect antigen-reactive Th cell responses in pigs using the transiently expressed Th cell activation marker CD154 [29]. Therefore we analysed antigen-experienced Th cells (pre-gated for lymphocytes > single cells > viable CD3+ > CD4+ CD8α+ ), described as CD3+CD4+CD8α+ [30], [31], shown in Fig. 5A. In comparison to piglets of the placebo-group the re-stimulation of PBMCs of rIde-immunized piglets with rIde induced a significantly increased number of CD154+ Th cells (Fig. 5B, upper left) within the antigen-experienced Th cell population. In contrast, stimulation with a non-relevant control antigen (recombinant fibronectin binding domain of SfbI) did not induce CD154+ Th cells in both groups that confirm the antigen-specific reactivity of antigen-experienced Th cells in the assay. Considering the cytokine profile of the rIde-reactive CD154+ Th cells, we found a significantly increased number of TNF-α (Fig. 5B, upper right) and IL-17A producer (Fig. 5B, lower left) in the vaccination group in comparison to placebo-treated animals. The same trend was found for IFN-ɣ producing CD154+ Th cells, but not statistically significant (Fig. 5B, lower right).

Fig. 5

Vaccination-induced Ide-reactive Th cell frequencies correlate with Ide-specific IgG levels. (A) Gating strategy to pre-gate antigen experienced Th cells (CD3+CD4+CD8α+). (B) Within this population frequency of Ide-reactive Th cells was determined 17 d post 2nd booster immunization (prior cps9 challenge). Therefore PBMCs from rIde- vaccinated and placebo-treated pigs (n = 9 per group) were stimulated with 5 µg/ml rIde or the recombinant fibronectin binding domain of SfbI as a non-relevant control antigen for 18 h and in presence of Brefeldin A (2 µg/ml) for the last 14 h to detect intracellular CD154 and cytokine expression (frequency of viable CD3+ cells; mean 41.8 ± SD 5.0%). As controls PMA/ionomycin (10 ng/ml and 500 ng/ml, respectively, positive control) and medium (med, negative control) were used. For statistical analysis non-parametric Kruskal-Wallis test and Dunn’s multiple comparisons post-hoc test was used (*p < 0.05, **p < 0.01, ***p < 0.001). (C) Pearson correlation between frequencies of Ide-induced CD154+ Th cells (calculation see materials/methods) and Ide IgG levels.

To analyze, whether the detected Ide-reactive Th cell frequencies are linked to the B cell-mediated Ide-specific IgG response described above (Fig. 2), correlation of the frequencies of rIde-induced antigen-reactive CD154+ Th cell frequencies (representing the difference of Ide-stimulation minus medium control of the same sample) and Ide-specific IgG levels was performed using the Pearson test. In contrast to the placebo group (r2 = 0.10), within the rIde-vaccinated group (n = 9) the number of Ide-reactive CD154+ Th cells correlated with the Ide-specific IgG level (r2 = 0.48) in this manner that animals with higher frequencies of rIde-induced CD154+ Th cells showed higher levels of anti-Ide-IgGs (Fig. 5B). Taken together, rIde-prime-booster-booster vaccination of piglets induces a Th cell response that is linked to Ide-specific IgG level.

In addition, we analyzed the frequency of Ide-reactive Th cells after challenge to prove whether vaccination-induced Th cells are re-activated by the challenge with cps9 strain 16085/3b. Therefore the two placebo-treated animals that survived until day 14 and five rIde-immunized piglets, all with a maximal cumulative clinical score ≥ 4 were selected. We could not detect an increase of Ide-reactive Th cells after challenge, neither for the frequency of all rIde-induced CD154+ cells nor for a cytokine producing subtype of them (Fig. S3A). Rather, by trend we observed a reduced number of Ide-reactive Th. We found a similar result for the α-Ide IgG levels, which did not increase after the challenge, but decreased in tendency (Fig. S3B).

In summary we observed an induction of Ide-reactive Th cells by immunization, but no recall upon challenge infection.

Discussion

S. suis cps2 and 9 have an enormous impact on animal health in Europe. A vaccine protecting against both serotypes is needed in the field to ensure return of investments in the pig industry and improve animal health and welfare. Though cps9 is a major porcine pathogen in the field, no recombinant S. suis vaccine has been tested in a challenge experiment with a cps9 strain [32]. This study revealed protection through prime-booster-booster vaccination with rIde against mortality induced by intravenous application of a S. suis cps9 strain. rIde vaccination has been shown to protect piglets also against S. suis cps2 challenge [11]. Thus, Ide is a cross-protective antigen covering at least related cps2 and cps9 strains. In contrast to the previous study including an intranasal cps2 challenge, we did not observe protection against morbidity following intravenous cps9 challenge, as vaccinated piglets showed temporarily elevated body temperature, reduced feed intake and lameness. However, sudden death, recumbency, convulsions and polyarthritis were observed in placebo-treated piglets only. As these signs occurred a few hours after challenge in contrast to the intranasal cps2 challenge, the intravenous application of the cps9 strain 16085/3b with 2 × 108 CFU conducted in this study is regarded as a very hard challenge. Noteworthy, sudden death is not a common sign in experimental infections with S. suis. Based on the rapid progression of disease and the high rate of mortality after challenge as well as the increased survival of this cps9 strain in porcine blood in comparison to the other investigated cps9 strains, strain 16085/3b is considered very virulent. MLST revealed that this strain is a ST94 strain which in contrast to other cps9 strains such as A3286/94 (ST99), V5402/2 and 15-3/3 (both ST16) does not belong to CC16. We cannot rule out differences in protective efficacies of rIde vaccination against cps9 strains of different clonal complexes. However, experimental infections with cps9 strains of CC16 have resulted in low morbidity, making it very difficult if not impossible to draw conclusions on protection [4].

Pathohistological screenings revealed fibrinosuppurative meningitis only in placebo-treated piglets (4/9) and in accordance a substantially lower pathohistological score in rIde-vaccinated piglets (ω = 2.0 versus ω = 3.0, respectively). Of note, meningitis is the most important pathology of S. suis infection in weaning piglets causing high numbers of losses [33]. Though fibrinous endocarditis was observed in three placebo-treated and also in three rIde-vaccinated piglets, the challenge strain was detected on the respective mitral valve in six placebo-treated but only in one rIde-vaccinated piglet. Whether rIde vaccination might be less protective against endocarditis needs to be further clarified. As S. suis forms biofilms covered with fibrin on heart valves, killing of streptococci by humoral immunity is very limited once this vegetation is formed [6]. However, as intravenous application bypasses mucosal immunity, rIde vaccination might still protect against endocarditis in mucosal infections. Compared to our previous S. suis cps2 study [11], the pathohistological score in placebo-treated piglets was notably lower (ω = 3.6 versus ω = 3.0, respectively). We suggest that this is due to the sudden death of four piglets on the day of infection. The peracute course of disease did most likely not leave enough time for severe lesions to develop. The piglets only reached pathohistological scores of 3 at the maximum.

We investigated cellular immunogenicities to find out if antigen-reactive Th cells are elicited by our vaccination protocol. Since Th cell-mediated B cell activation supports the production of highly affine specific IgGs [34], we investigated the correlation of rIde-reactive Th cell frequencies with rIde-specific IgG levels. For this Th cell support especially the activity of TfH cells (follicular T-helper cells) in draining lymph nodes is of importance, as recently also demonstrated in pigs [35]. However, lymph node tissue from immunized animals in a vaccination study is usually not accessible. Therefore, it is more beneficial to use readout-parameters for T cell reactivity from blood samples. The results of this study showed that prime-booster-booster vaccination with a vaccine containing rIde and 20% Emulsigen results in induction of antigen-reactive Th cells detectable in porcine blood. The antigen-specific reactivity of this read out parameter is confirmed by using a S. pyogenes antigen (the fibronectin-binding domain of SfbI) as control that was also expressed as His-tagged protein in E. coli and purified via Ni-affinity chromatography [27]. Furthermore, the number of Ide-reactive CD154+ Th cells correlated with the Ide-specific IgG level within the immunization group. This is in accordance with our previous findings that Ide-reactive CD154+ Th cell frequencies correlated with Ide-specific IgG levels [29], although in contrast to our previous study, we did not observe a correlation with the frequency of IFN-γ producing CD154+ cells. In addition, we analyzed Ide-reactive Th cells in PBMCs after cps9 challenge (14 dpi.). Based on our data the adaptive immune response to Ide was not recalled by the S. suis cps9 challenge for both, Ide-specific-IgGs and Ide-reactive-Th cells. It is very unlikely that the time point of measurement (14 dpi) was too late after challenge as Ide-reative Th cells were detectable 17 days post 2nd booster and infection was not restricted to the day of challenge as indicated by the course of body temperature (Fig. S2) and the detection of the challenge strain in the mitral valve of piglet #1 (Table S2). In experiments designed to analyse antigen-specific recall of Th cells by using DO11.10/RagKO mice immunized and infected with Ova-expressing S. pneumoniae, Trzciński et al. demonstrated that vaccination-induced Th17 cells are only protective against S. pneumoniae by antigen-specific recall during challenge infection [21]. However, based on our results we suppose that Ide is not an immunodominant T cell antigen under invasive infection and therefore a direct involvement of Ide-reactive Th cells in protective mechanisms is improbable.

As we could previously demonstrate, the vaccination with rIde elicits antibodies neutralizing the IgM protease activity of Ide [11]. Recently we demonstrated that the activity of complement is important for clearance of S. suis in porcine blood, whereas the complement deposition is reduced by the protease activity of Ide [19]. Thus, neutralization of the IgM protease by α-Ide-IgGs should improve the complement-mediated killing of S. suis in porcine blood.

Additionally, in our previous study we found a significantly lower bacterial survival factor of S. suis wt in blood of vaccinated piglets in comparison to the respective survival factor of the Ide mutant 10Δide [11]. In agreement with the location of Ide on the bacterial surface [16], the most plausible explanation for this observation is that α-Ide antibodies mediate opsonophagocytosis. However, a significant difference was not observed in this study (Figs. 3 and S1).

As mentioned above, prime-booster-booster vaccination was necessary to elicit α-Ide-IgG levels comparable to those of our previous study. Still, in case of five piglets less than 43 ELISA units were measured at 17 d post 2nd booster (Table S2). α-Ide-IgG levels and the clinical scores obtained following experimental infection did not show a significant correlation (data not shown). Further studies are necessary to clarify if the data obtained in the described α-Ide-IgG ELISA allows setting up a threshold for protection against severe disease.

In summary, this study demonstrates protection against S. suis cps9 through prime-booster-booster vaccination of piglets with rIde. This vaccination was associated with induction of α-Ide-IgG and Ide-reactive T cells. We suppose that Ide -reactive Th cells are important for the B cell immune response to Ide, but recall of Ide -reactive Th cells was not detectable after challenge and is thus not considered to be crucial for protection.

Declaration of Competing Interest

The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: This study was financially supported by IDT Biologika GmbH. IDT Biologika GmbH applied for a European patent concerning an Ide-based S. suis vaccine (Nr. 14 170 637.4).