Abstract
Streptococcus suis is an opportunistic pathogen or pathobionts in pigs and a zoonotic pathogen for humans. S. suis Serotype 2 (SS2) is the predominant serotype isolated from diseased pigs but rarely isolated from healthy pigs. This study aims to investigate whether the growth of SS2 can be inhibited by commensal S. suis non-serotype 2 (non-SS2) sharing tonsillar habitats. Agar co-culture (CI<1.0) and transwell assays confirmed that two non-SS2 strains inhibited the growth of patients, diseased and asymptomatic pig SS2 isolates by contact-dependent mechanism. Kinetic co-culture demonstrated that the consequence of this competition was the coexistence of SS2 at the low density. Moreover, non-SS2 also interfered SS2 adhesion on IPEC-J2 cells. Comparative whole genome analysis elucidated that contact-dependent growth inhibition involved with T7SS homologs and apparently non-lethal interaction. Complete semi-phosphorylative Entner-Doudoroff, CMP-Neu5Ac, and Catechol meta-cleavage pathways distinctively found in non-SS2 and their increased expression during co-culture might facilitate them to outcompete the growth of SS2. In conclusion, this study identified a novel contact-dependent growth competition phenotype in non-SS2 that might contribute to maintaining a low density of the SS2 target in natural habitats.
Subject terms: Ecology, Microbiology, Medical research
Introduction
Streptococcus suis is a zoonotic swine pathogen that can cause several severe diseases in young pigs, including meningitis, septicemia, endocarditis, arthritis, and pneumonia, causing economic losses in the global swine industry1,2. To date, S. suis is known as opportunist or pathobiont in pigs. The most predominant serotype of S. suis isolated from disease pigs and patients worldwide is serotype 23. However, S. suis serotype 2 (SS2) is commonly isolated at low frequency from healthy pigs compared to other serotypes in zoonotic endemic areas4–8. Therefore, it is possible that during the healthy (asymptomatic) stage of pigs, commensal non-SS2 strains may be able to inhibit the growth of SS2 and result in the rare isolation of SS2 from the healthy pig samples.
In commensal habitat, most of the bacteria interact and coexist in a high-density community and regulate the number of each strain to utilize the limited nutrients and occupy space. The primary phenotype of competition for population number control is the interference competition or direct competition, resulting in the growth inhibition of the target bacterial strain. Therefore, the low frequency isolation of SS2 from the healthy pig upper respiratory tract or tonsils may possibly be the consequence of growth competition by the major commensal component, non-SS2 strains12. The interaction patterns between SS2 and non-SS2 strains in pig commensal habitat and their impact on the growth of each other were currently partially known.
Recently, several systems have been demonstrated to facilitate niche competition of S. suis novel capsular locus 4 (NCL4) strain WUSS351. This strain harboured ABC transporter system SstFEG, conferring resistance to bacitracin produced by other microflora in vivo12. Moreover, S. suis NCL4 strain WUSS351 also secrete lantibiotic bacteriocin and lactococcin972-like bacteriocin Lcn351 that specifically target SS2 and might provide themselves the survival advantage in vivo12,13. Moreover, strain WUSS351 also carried putative type VII secretion system (T7SS) machinery and LXG domain-containing effector proteins homologous to those found in serotype 9 S. suis. To date, T7SS is the only known system that mediates contact-dependent growth competition in gram-positive cocci. However, the contribution of T7SS and effector protein homologues in S. suis growth competition has not yet been elucidated. Therefore, this study aims to characterize and investigate whether the growth of SS2 can be competitively inhibited by contact-dependent mediated mechanisms from non-SS2.
Results
Intra- and interspecies growth inhibition of pathogenic SS2
Only 14 out of 180 non-SS2 strains showed a complete range of growth inhibitory effects to SS2 isolated from patients, diseased pigs and asymptomatic carriers. A representative of clear inhibition zone on SS2 lawn and the details of SS2 growth inhibition by non- SS2 strains are shown in Fig. 1a and Table 1, respectively. Subsequently, these 14 strains were investigated further to differentiate whether a contact-dependent mechanism mediated the growth competition.
Fig. 1.
The ability of non-SS2 strains to inhibit the growth of SS2. (a) Representative growth inhibition of SS2 (lawn; SS2 strain 399) mediated by non-SS2 strains (spots; non-SS2 strains 121S, 121B, 389.2, 391, 82S, 82B,128, 475.2, 147, 29S, 29B, 37, 27, and 83) were demonstrated as the clear inhibition zones in direct challenge assay (white arrows). (b) Agar diffusion assay showed an unstained zone of growth inhibition on SS2 strain 281 lawn (black arrows) caused by the soluble substances produced by non-SS2 strain in the wells (non-SS2 strains 147, 34 and 150, respectively). These strains were then excluded for further experiments. SS2 growth around the wells was stained with methylene blue. (c) A proximity assay used to preliminary investigate whether the non-SS2 strains were mediated by a contact-dependent mechanism. At the closest proximity (0.4 cm) non-SS2 inhibited the growth of SS2. The edge of non-SS2 spots protruded as the bulge while that of SS2 became concaved. The results were obtained from the duplicate tests of three separate experiments.
Table 1.
Intra- and inter-species growth inhibition of SS2 strains by direct challenge assay, agar diffusion assay and proximity assay.
| Characteristic of non-SS2 inhibitor strains | Number of SS2 target strains that were growth-inhibited by non-SS2 (total 68 strains) |
||||
|---|---|---|---|---|---|
| No | Genus | Specie | Direct challenge | Soluble substance | Proximity |
| 95 | Streptococcus | Porcinus | 68 (100.0%) | 68 (100.0%) | |
| 463 | Streptococcus | Suis | 65 (95.6%) | 68 (100.0%) | |
| 29 | Streptococcus | Suis | 65 (95.6%) | 62 (91.1%) | |
| 150 | Streptococcus | Suis | 62 (91.1%) | 64 (94.1%) | |
| 34 | Streptococcus | Suis | 62 (91.1%) | 32 (42.7%) | 65 (95.5%) |
| 147 | Streptococcus | Suis | 61 (89.7%) | 16 (20.6%) | 61 (89.7%) |
| 18 | Streptococcus | anginosus (S. milleri) | 64 (94.1%) | 67 (98.5%) | |
| 26 | Streptococcus | Hyointestinalis | 66 (97.0%) | 67 (98.5%) | |
| 121 | Enterococcus | Cecorum | 65 (95.6%) | 67 (98.5%) | |
| 41 | Streptococcus | Porcinus | 66 (97.0%) | 46 (67.7%) | |
| 60 | Streptococcus | hyointestinalis | 57 (83.8%) | 30 (41.2%) | 51 (75.0%) |
| 196 | Enterococcus | Casseliflavus | 62 (91.1%) | 68 (100.0%) | |
| 391 | Enterococcus | sp. | 65 (95.6%) | 68 (100.0%) | |
| 83 | Enterococcus | Casseliflavus | 63 (92.7%) | 68 (100.0%) | |
Among these 14 non-SS2 inhibitor strains, 11 strains were excluded because they produced soluble substances that can diffuse through an agar medium to inhibit the growth of SS2 (Fig. 1b). Consequently, only three non-SS2 strains, including two S. suis (no. 34 and 147) and a Streptococcus hyointestinalis (no. 60) could not inhibit the growth of SS2 target strains when they were separately grown in the wells in agar diffusion assay. Then, contact-dependent growth inhibition mediated by these three non-SS2 was further investigated by proximity assay. It was found that these three inhibitors can inhibit SS2 targets only at closest proximity (0.4 cm.), where the edges of non-SS2 and SS2 cultured spots directly contact (Fig. 1c). Our data showed that the growth of 75 to 95.5% of SS2 target strains from various source could be inhibited by non-SS2 and probably mediated by contact-dependent mechanism (Table 1).
Confirmation for contact-dependent mediated growth inhibition by non-SS2
Only the co-culture pairs yielded CI higher than one indicated that the growth of SS2 and non-SS2 were unaffected by each other in transwell growth partitioning assay hence mediated by contact-dependent.
The non-SS2 strains 34 and 147 inhibitors inhibited the growth of more than 70% of asymptomatic pig SS2 isolates. However, strain 147 was superior to 434 in growth inhibition of patient SS2 isolates, whilst strain 34 had higher growth inhibition activities for disease pig SS2 isolates (Table 2 and Table S1).
Table 2.
Confirmation of contact-dependent growth inhibition of SS2 by non-SS2 using transwell partitioning assay.
| Non-SS2 inhibitor strain | Number of SS2 target isolates inhibited by contact-dependent mechanism of non-SS2 | |||
|---|---|---|---|---|
| No | Specie | Patient isolates (total 13) | Disease pig isolates (total 26) | Asymptomatic pig isolate (total 7) |
| 34 | Streptococcus suis | 7 (53.8%) | 25 (96.2%) | 5 (71.4%) |
| 147 | Streptococcus suis | 10 (76.9%) | 18 (69.2%) | 5 (71.4%) |
| 60 | Streptococcus hyointestinalis | 0 (0.0%) | 10 (38.5%) | 1 (14.3%) |
Competitive interaction phenotype of Non-SS2
Eight SS2 target strains were individually co-cultured with non-SS2 strains 34 and 147 on agar medium as described in Materials and Methods to identify their interaction phenotype and the consequence of growth inhibition. The successful SS2 growth inhibition represented by CI lesser than one. It was found that the growth of seven SS2 strains was inhibited during co-culture with non-SS2 no. 34 (Fig. 2a), while only four SS2 strains showed growth inhibition during co-culture with non-SS2 no. 147 (Fig. 2b).
Fig. 2.
Growth inhibition of SS2 on agar medium following co-culture with non-SS2 strains. (a) CI of non-SS2 strain 34 co-culture with SS2 and (b) CI of non-SS2 strain 147 co-culture with SS2. Growth inhibition was represented by a CI lesser than 1. Sequence types of SS2 strains were shown in the bracket. CI ± SD was obtained from the duplicate tests of three separate experiments.
To further investigate and identify interaction phenotype and consequence of growth competition, the kinetics of growth competitions of the 11 competitive pairs (Fig. 2; CI lesser than 1) were individually co-cultured with inhibitor strains in DMEM/F-12 tissue culture medium to mimic in vivo conditions. All competitive pairs were found to interact with competitive phenotype as a consequence of the growth rate of both target and inhibitor strains in co-culture being lower than that of the respective strains in sole culture (Fig. 3). The remaining four competitive pairs were classified as atypical competitive interaction phenotypes because the density of the non- SS2 inhibitors at the end of co-culture incubation was equivalent to their respective sole culture. These suggested that the growth non- SS2 inhibitors might gain the benefit in co-culture to restore their growth (Fig. 3b).
Fig. 3.
Kinetic growth inhibition of SS2 after co-culture with non-SS2 strains represented (a) a typical and (b) atypical competitive interaction phenotype. Grey line: growth of non-SS2 (no. 34 and 147); pink line: growth of SS2; solid line: growth after co-culture; dash line: respective sole culture (control). CFU/ml ± SD were obtained from the duplicate tests of three separate experiments. P-value : * < 0.01, ** < 0.001, *** < 0.0001.
At the end of the kinetic incubation, although, the density of SS2 targets were low but still higher than that of the initial inoculums suggesting that the consequence of growth competition were the coexistence of SS2 targets with non-SS2 inhibitors at low number in the shared niche (Fig. 3).
SS2 adhesion to IPEC-J2 was reduced when co-culture with Non-SS2 inhibitors
Reduction of adhesion of SS2 indicated by CI lesser than one. It was found that the adhesion of six SS2 targets to IPEC-J2 interfered during co-infection with non-SS2 strain no. 34 (Fig. 4a). Interestingly, inhibitor strain147 can interfere diminished the adhesion of all SS2 targets to IPEC-J2 (Fig. 4b).
Fig. 4.
Reduction of SS2 adhesion to IPEC-J2 during co-infection with non-SS2 strains. Coinfection of SS2 strains with inhibitors (a) 34 and (b) 147 to IPEC-J2 cells. Reduction of adhesion of SS2 to IPEC-J2 cells indicated by CI lesser than one. Sequence type of each SS2 strain was shown in the bracket. CI ± SD were obtained from the duplicate tests of three separate experiments.
Comparative genome analysis of SS2 target and non-SS2 inhibitor strains
Genome sequences of all two non-SS2 inhibitor and eight SS2 target strains were analyzed, assembled and annotated to predicted factor(s), that might attribute to contact-dependent growth inhibition mediated by non-SS2 strains.
T7SS was not involved in contact-dependent growth inhibition mediated by non-SS2
T7SS was previously reported to mediate contact-dependent growth inhibition in Gram-positive bacteria, such as S. aureus and S. intermedius. Moreover, T7SS components were also present in the ST243 serotype 9 S. suis strain GZ0565 and the NCL4 S. suis strain WUSS351. None of non-SS2 inhibitors and SS2 strains in our study carried complete T7SS locus. Annotated genes and open reading frames of non-SS2 inhibitors and SS2 strains were then aligned to compare with the sequence of the known 18 T7SS genes in gram-positive bacteria. It was found that both non-SS2 inhibitors and SS2 strains shared 31% and 49% DNA homology to essC from GZ0565 and WUSS351, respectively (Table 3 and Table S2).
Table 3.
Presence of 18 T7SS genes in SS2 and non-SS2 strains.
| Gene of T7SS locus | SS2 target strain (ST) | non-SS2 inhibitor strain |
Similar gene source | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| P1/7 (ST1) |
06 (ST1) | 05 (ST104) | 279 (ST104) | 281 (ST104) | 676 (ST104) | 879 (ST104) | 399 (ST1) | 34 SS29 (ST319) |
147 NCL1 (ST1086) |
||
| esxA | - | - | - | - | - | - | - | - | - | - | Staphylococcus aureus strain WU1 |
| esaA | - | - | + | + | + | + | + | - | - | - | |
| esaB | - | - | - | - | - | - | - | - | - | - | |
| essA | - | - | - | - | - | - | - | - | - | - | |
| essB | - | - | - | - | - | - | - | - | - | - | |
| essC | + | + | + | + | + | + | + | + | + | + | |
|
DUF5083/ DUF5084 |
- | - | - | - | - | - | - | - | - | - | |
| DUF4064 | - | - | - | - | - | - | - | - | - | - | |
|
DUF5079/ DUF5080 |
- | - | - | - | - | - | - | - | - | - | |
|
DUF3130/ DUF5344 |
- | - | - | - | - | - | - | - | - | - | Listeria monocytogenes strain GIMC2124:LmcUH35 |
| DUF5085 | - | - | - | - | - | - | - | - | - | - | |
|
DUF5082/ DUF3958 |
- | - | - | - | - | - | - | - | - | - | Streptococcus oralis strain SF100 |
| esaE | - | - | - | - | - | - | - | - | - | - | Staphylococcus aureus JICS127 |
| DUF4176 | - | - | - | - | - | - | - | - | - | - | Streptococcus agalactiae strain SS1 |
| DUF600 | - | - | - | - | - | - | - | - | - | - | Staphylococcus aureus subsp. aureus JH9 |
| telA | - | - | - | - | - | - | - | - | - | - | Streptococcus intermedius |
| telB | - | - | - | - | - | - | - | - | - | - | |
| telC | - | - | - | - | - | - | - | - | - | - | |
Non-SS2 inhibitors inhibited growth of SS2 with out killing
Since T7SS has been absent in non-SS2 strains, it was necessary to test whether they inhibit the growth of SS2 by contact-dependent killing. A co-culture assay was performed. Dead bacteria were stained with PI and visualized and enumerated using a confocal laser scanning microscope. It was found that the ratio of live/dead bacterial cells in co-culture was not significantly different from those sole culture of each strain(Fig. 5 and Table S3). These data indicated that non-SS2 inhibitor strains inhibited the growth of weaker competitor SS2 targets by a non-lethal mechanism.
Fig. 5.
Live/dead assay to quantify the death of bacteria during co-culture of SS2 with non-SS2 inhibitors. Live and PI-stained dead bacteria were visualized and enumerated under the confocal laser scanning microscope to calculate live/dead ratio.
Growth advantage of non-SS2 over SS2 supported by carbohydrate metabolism, xenobiotic degradation pathway, and capsular polysaccharide precursor biosynthesis
COG and KEGG analyses of whole genome sequences suggested the unique presence of three pathway modules for non-SS2 and SS2 strains (Table 4). The first and second KEGG modules were absent in all SS2 target strains but present in both non-SS2 strains. Within the semi-phosphorylative Entner-Doudoroff pathway (M00308) of central carbohydrate metabolism, gapN coding for glyceraldehyde-3-phosphate dehydrogenase (NADP +) designated as K00131 was absent in all SS2 target strains. This enzyme catalyzes the irreversible oxidation of glyceraldehyde 3-phosphate to 3-phosphoglycerate (glycerate-3P) while reducing NADP + to NADPH. Therefore, K00131 activities provides NADPH, which is substituted for those usually produced through the two oxidizing reactions in the pentose phosphate pathway (PPP), absent in S. suis and other streptococci. Hence, lack of this enzyme in SS2 may impede their energy production through glycolysis and essential cofactorszz for macromolecule biosynthesis (Table 4 and Fig. 8). Second module was catechol meta-cleavage (M00569), aromatic xenobiotic degradation pathway, K07104 encoded by catE was present in only non-SS2 inhibitor strains (Table 4). This pathway degrades toxic catechol and methylcatechol, naturally derived from aromatic polymers or hormones. This degradation provides pyruvate and acetyl-CoA, precursors for glycolysis and TCA cycle, respectively. Many microorganisms have developed similar catabolic pathways to utilize the available non-toxic and toxic compounds in their natural environments, such as forests, soil, water, and sediments, for gaining energy in a niche adaptation fashion.
Table 4.
Presence and absence of KEGG pathway genes in SS2 and non-SS2 strains.
| KEGG module | Sequence validation |
non-SS2 inhibitor strains | SS2 target strains | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 34 SS29 (ST319) |
147 NCL1 (ST1086) |
06 (ST1) |
05 (ST104) |
05 (ST104) |
399 (ST1) |
P1/7 (ST1) |
279 (ST104) |
281 (ST104) |
676 (ST104) |
879 (ST104) |
||
| Pathway module | ||||||||||||
| Carbohydrate metabolism | ||||||||||||
| Central carbohydrate metabolism | ||||||||||||
| M00308 Semi-phosphorylative Entner-Doudoroff pathway | ||||||||||||
|
(GapN) |
KEGG | + | + | - | - | - | - | - | - | - | - | - |
| BlastN | + | + | - | - | - | - | - | - | - | - | - | |
| K00134 | KEGG | + | + | + | + | + | + | + | + | + | + | + |
| BlastN | + | + | + | + | + | + | + | + | + | + | + | |
| K00874 | KEGG | + | + | + | + | + | + | + | + | + | + | + |
| BlastN | + | + | + | + | + | + | + | + | + | + | + | |
| K00927 | KEGG | + | + | + | + | + | + | + | + | + | + | + |
| BlastN | + | + | + | + | + | + | + | + | + | + | + | |
| K01625 | KEGG | + | + | + | + | + | + | + | + | + | + | + |
| BlastN | + | + | + | + | + | + | + | + | + | + | + | |
| Xenobiotics biodegradation | ||||||||||||
| Aromatics degradation | ||||||||||||
| M00569 Catechol meta-cleavage | ||||||||||||
| K01821 | KEGG | + | + | - | - | - | - | - | - | - | - | - |
| BlastN | + | + | + | + | + | + | + | + | + | + | + | |
|
K07104 (CatE) |
KEGG | + | + | - | - | - | - | - | - | - | - | - |
| BlastN | + | + | - | - | - | - | - | - | - | - | - | |
| Glycan metabolism | ||||||||||||
| Lipopolysaccharide metabolism/capsular polysaccharide precursor biosynthesis | ||||||||||||
| M00922 CMP-Neu5Ac biosynthesis | ||||||||||||
|
(NeuB) |
KEGG | - | - | + | + | + | + | + | + | + | + | + |
| BlastN | - | - | + | + | + | + | + | + | + | + | + | |
|
(NeuC) |
KEGG | - | - | + | + | + | + | + | + | + | + | + |
| BlastN | - | - | + | + | + | + | + | + | + | + | + | |
| K24257 | KEGG | - | + | + | + | + | + | + | + | + | + | + |
| BlastN | - | + | + | + | + | + | + | + | + | + | + | |
Fig. 8.
Proposed mechanism for growth inhibition mediated by non-SS2 inhibitor over target SS2 strains. Thick black arrow: possible mechanism/pathway according to the gene presented only in the non-SS2 genome; Thin black arrow: mechanism/pathway presented in both the inhibitor and target strains; Dark blue letter: mechanism/pathway presented in only the SS2 genome; Green blue letter: products of the pathway9–11.
The third pathway was capsular polysaccharide precursor, CMP-Neu5Ac biosynthesis (M00922), neuBC were absent in non-SS2 strains 34 and 147 (Table 4 and Fig. 8). On the other hand, SS2 strains possess the completed sialic acid biosynthesis pathway. CMP-Neu5Ac biosynthesis is essential for S. suis CPS biosynthesis. However, the Neu5Ac can be imported from the extracellular milleu for use as a carbon source and CPS biosynthesis precursor for those non-SS2 strains.
Increase expression of gapN and catE during co-infection of IPEC-J2
As mentioned previously , gapN and catE were absent in genomes of all SS2 targets. During co-culture, it was found that gapN and catE were significantly increased (Fig. 6), suggesting the involvement of the complete semi-phosphorylative Entner-Doudoroff pathway (M00308) and aromatic xenobiotic degradation, catechol meta-cleavage (M00569) in growth advantage of non-SS2 over SS2.
Fig. 6.
Relative expression of (a) catE and (b) gapN (fold change) while bacterial co-culture determined by qRT-PCR. Relative fold change ± SD were obtained from the duplicate tests of three separate experiments. P-value : * < 0.05.
CatE inhibitor mediated the loss of non-SS2 ability to interfere SS2 IPEC-J2- adhesion during co-infection
The 3-chlorocatechol inhibitor for CatE was used to confirm whether non-SS2 CatE function involved in the interference of SS2 adhesion. The presence of CatE inhibitor resuted in the inversion of competitive index to be higher than one (Fig. 7). These data indicated that CatE function were not only facilitating non-SS2 growth advantage but also contributed to their competition on eukaryotic adhesion and colonization.
Fig. 7.
SS2 adhesion to IPEC-J2 was not reduced during co-infection with non- SS2 strains along with 3-chlorocatechol inhibitor. Reduction of adhesion of SS2 to IPEC-J2 cells indicated by CI lesser than one. CI ± SD were obtained from the duplicate tests of three separate experiments.
Discussion
The ability of an organism to maintain a polymicrobial niche was partly the reflection of the differences in nutrients and mechanisms of interaction among the bacterial community in that particular microenvironment12–15. Several species live together, leading to competition and regulating the population number of each strain to use the limited nutrients and share the colonizing spaces, influencing the bacterial composition and maintaining homeostasis16. In this study, 14 non-SS2 inhibitor stains had a broad range of growth inhibitory effects to SS2 target strains. As reported previously, commensal bacteria produce molecular factors to outcompete the pathogenic bacteria by taking advantage of using limited nutrients or occupying colonizing spaces, contributing to the better survival and directly inhibiting the growth of pathogenic bacteria17,18.
Commensal bacteria in the respiratory tract, intestine, or soil often coexist in a community using positive and negative interactions. Diversity and density of these populations have been mediated by interference competition, whereby a stronger population hinders or prevents the weaker one from exploiting the particular limited resources in the shared niche (interference competition). This interference competition could be mediated by two mechanisms, contact-independent and contact-dependent mechanisms16,19. In the gastrointestinal tract, probiotic Clostridium difficile employed the intra-species contact-dependent mechanism for inhibiting the colonization of pathogenic C. difficile, resulting in the mitigation of colitis20. S. suis are the most predominant commensal population inhabiting the mucosal surface of the pig palatine tonsil. This study emphasized on the investigation of capacities of non-SS2 inhibitors to inhibit growth and eukaryotic cell adhesion of SS2 targets by the intra-species contact-dependent mechanisms21. Two non-SS2 inhibitors in our study were confirmed to mediate the growth inhibition of pathogenic SS2 by contact-dependent mechanisms.
Organism interactions were classified based on the determination of the ecological outcome for the individual organism in a pairwise interaction22. We found that two non-SS2 inhibitor strains used contact-dependent mechanisms to inhibit the growth of all tested SS2 strains with a competitive interaction phenotype (Fig. 4). At the end of the experimental interaction, the non-SS2 inhibitor strains dominated the SS2 target and resulted in the coexistence of the SS2 targets at a low level4–8. Similar intra-species interactions were also reported in Streptococcus pneumoniae23. Moreover, this competitive interaction mechanism were completely non-lethal (Fig. 5).
Abilities of the virulence SS2 to adhere and invade the host cells could be investigated in vitro by using porcine intestinal epithelial cells, IPEC-J2 cells24,25. We demonstrated that the contact-dependent competitive interaction of non-SS2 significantly reduced the IPEC-J2 adhesion of target SS2 during the in vitro co-infection, suggesting that this interaction phenotype also hindered the SS2 adhesion prerequisite step for SS2 pathogenesis.
In our study, whole genome sequence analyses elucidated that genes involved in contact-dependent mediated growth inhibition systems previously reported in other Gram-positive bacteria, such as outer membrane exchange (OME) and contact-dependent inhibition (CDI), were absent in the genome of non-SS2 inhibitor strains. Furthermore, complete T7SS gene loci previously reported in S. aureus and S. intermedius were also missing. In addition, although T7SS were previously identified from ST243 serotype 9 S. suis strain GZ0565 and NCL4 SS2 strain WUSS351, their function has not yet been elucidated. In our study, two T7SS gene homologs, essC and esaA, were harboured by non-SS2 inhibitor and SS2 target strains, therefore, it was likely that they were not responsible for contact-dependent growth inhibition.
The different presence of KEGG modules in the genome of non-SS2 inhibitors and SS2 target strains were likely to contribute to the growth and eukaryotic cell adherence advantages of non-SS2 inhibitors (Fig. 8). The semi-phosphorylative Entner-Doudoroff pathway and catechol meta-cleavage generated additional substrates for entering glycolysis, which is a critical step associated with the metabolism of cellular respiration. Gapn and Cate of the two pathways uniquely presented non-SS2 inhibitor genomes and overexpressed during co-infection of IPEC-J2, suggesting a higher ability of non-SS2 inhibitors to produce energy and degrade additional substrates for their metabolic activities. Previous studies revealed that the ability to eliminate catechols derived from the host metabolism in vivo was associated with the performance of piglets by improving weight gain, therefore, the ability of non-SS2 inhibitors to degrade catechol will probably benefit to their host and provide them the favorable commensal strain selected by the host in the perspective of evolutionary selection26. 2-Hydroxy-5-methyl-cis and cis-muconic-semialdehyde are formate precursors that are not only beneficial for bacterial growth and maintaining metabolic homeostasis, but also improve the growth performance of pigs (Fig. 8)27–30. The CMP-Neu5Ac biosynthesis pathway was completed in SS2 trains. CMP-Neu5Ac is the active form of Neu5Ac (N-acetylneuraminic acid), the primary type of sialic acid, catalyzed by CMP-Neu5Ac synthetase, which is important for capsular polysaccharides (CPSs) biosynthesis31. CPS masks bacteria from host immune systems and facilitates bacterial invasion31,32. Nevertheless, CMP-Neu5Ac biosynthesis was truncated in non-SS2 inhibitors, resulting in bias glucose substrate to enter glycolysis (Fig. 8). The absence and presence of those three pathways suggested that non–SS2 inhibitors utilize glucose substrates more efficient than SS2 targets, leading to outcompete the neighbour SS2 target cells and interrupt capsular polysaccharides (CPSs) biosynthesis (Fig. 8). As CPS is a significant virulence factor SS2, therefore, interference of CPS biosynthesis in SS2 might contribute to increasing its susceptiblity to elimination by phagocytosis in the upper repiratory tract.31,33,34.
In conclusion, this study identified a novel contact-dependent growth competition phenotype in non-SS2 that might influence the maintenance of a low density of the SS2 target in the commensal of healthy pigs. Semi-phosphorylative Entner-Doudoroff, CMP-Neu5Ac, and Catechol meta-cleavage pathways distinctively found in non-SS2 and their increased expression during co-culture facilitate them in part to outcompete the growth and adherence of SS2. Although the specific genetic components directly mediating contact-dependent growth inhibition mechanism by non-SS2 remain elusive, it is novel.
Methods
Study design
The research and all associated experimental protocols were carried out in accordance with relevant guidelines and regulations with approval from the Thammasat University Institutional Biosafety Committee, Thailand (062/2565).
Bacterial strain and growth condition
A total of 67 SS2 strains isolated from patients (n = 35), disease pigs (n = 25), and asymptomatic pigs (n = 7) were used in this study. The SS2 reference strain P1/7 was isolated from a United Kingdom disease pig belonging to sequence type 1 (ST1). One hundred and eighty non-SS2 isolates were obtained from asymptomatic pigs. Among these, 157 were confirmed as S. suis by recN-PCR7. MALDI-TOF MS identified four of Streptococcus sp. and Vagococcus fluvialis strain, three of Streptococcus porcinus strain, two of Streptococcus hyointestinalis, Streptococcus uberis, and Enterococcus casseliflavus strain. One of Streptococcus anginosus, Streptococcus mitis, Streptococcus oralis, Streptococcus parasanguinis, Enterococcus cercorum, Enterococcus sp. strain. All strains were streaked on the surface of Todd-Hewitt yeast extract (THY)-agar (THB supplemented with 2% yeast extract) and incubated at 37˚C under 5% CO2 for 18 to 22 h.
Direct challenge assay
To screen whether non-SS2 and streptococci strains could inhibit the growth of S. suis serotype 2, direct challenge assay was performed as previously described35. Briefly, overnight cultures of SS2 (approximately 5X108 CFU/ml) was individually spread on the surface of THY agar. Five microliters (5X108 CFU) of each non-SS2 were dropped on THY agar, which had previously been spread with SS2. The inhibition zone was observed after incubating overnight at 37˚C under 5% CO2. Fresh new THY broth will be used as a negative control.
Agar diffusion assay
To investigate whether intra- and inter-species inhibition employ contact-independent mechanisms, the inhibitor and target strains were grown in the separated compartment as previously described36. Briefly, approximately 5X108 CFU/ml SS2 were individually spread on THY agar, and a gel cutting tip was used to punch the wells. The well was then filled with 20 µl of non-SS2 suspension, 2 mg/ml ampicillin, which was used as positive soluble substance control, and isogenic SS2 and THY were used as negative growth inhibition control. Plates were incubated overnight at 37˚C under 5% CO2 and observed inhibition zone. The cultured plates were stained with 0.04% methylene blue in water for better visualization37.
Proximity assay
To investigate whether S. suis non-serotype 2 able to inhibit the growth of S. suis serotype 2 mediated by close contact as previously described38. Briefly, overnight cultures of SS2 and non-SS2 were resuspended in THY broth (approximately 5X108 CFU/ml). Then, 5 µl cell suspension of target and inhibitor strains were dropped at a distance of 0.4, 0.6, and 0.8 cm on the surface of THY agar. After overnight incubation, the bulge and concave shapes of culture spots were observed. Isogenic testing strains were used as controls.
Transwell assay
The non-SS2 that inhibit the growth of SS2 was further confirmed for the contact-dependent mechanism by transwell assay as previously described39. Briefly, overnight cultures of both strains were adjusted to approximately 1X108 CFU/ml. The suspension of non-SS2 inhibitors was added into the well of the 24-well plate (lower chamber). An equal volume of SS2 suspension was transferred into the transwell insert with 0.4 µm polycarbonate membrane (upper chamber). After overnight incubation, the target and inhibitor strains suspension were serially diluted and enumerated on THY agar with and without penicillin G. The number of each strain was expressed as CFU/ml. The total number of S. suis were from THY agar without antibiotic supplements. The number of non-SS2 was obtained from THY supplemented with penicillin G, and the number of SS2 was obtained from the subtraction of the number of non-serotype 2 S. suis from the total number of S. suis.
Agar coculture assay
To determine CI, type of competition and consequence of competition, an agar co-culture assay was performed as previously described40,41. Briefly, overnight cultures of both strains (approximately 5X108 CFU/ml) were mixed at a 1:1 ratio, a 10 µl of the mixture was dropped on THY-agar and incubated. Following the incubation, a mixture on the plate was scraped and resuspended in THY broth. The suspension was serially diluted and enumerated on THY agar supplemented with and without penicillin G. The number of each strain was calculated as described previously. The change in the inhibitor and target strains ratio from that of the initial inoculum was calculated and presented as the competitive index (CI). CI = [number of target(Tx) / number of inhibitor(Tx)] / [number of target(T0) / number of inhibitor(T0)]42.
DMEM/F-12 coculture assay
Dynamic of growth competition was investigated using DMEM/F-12 co-culture assay as previously described39,42. Briefly, overnight cultures of both strains (approximately 5X108 CFU/ml) were mixed at a 1:1 ratio, and a 10 µl of the mixture was seeded in each well, that containing 1 ml DMEM/F-12 supplemented with 1% pig serum and incubated. A co-culture mixture in each well was collected at one-hour intervals for eight hours, serially diluted, and enumerated on THY agar supplemented with and without penicillin G. The dynamic of growth inhibition was calculated and plotted.
IPEC-J2 co-colonization assay
IPEC-J2 cells were derived from the Institute of Clinical Physiology, Charité – Universitätsmedizin Berlin, Campus Benjamin Franklin, Berlin, Germany. The cultivation method was performed as described by Zakrzewski et al.43. IPEC-J2 cells were seeded at 4X 104 cells/well and incubated overnight. After incubation, cells were washed two times and fresh new complete media (DMEM/F-12 supplemented with 1% pig serum) was added. Overnight cultures of SS2 and non- SS2 were mixed at a 1:1 ratio, infected IPEC-J2 at MOI 100 and then incubated at 37˚C under 5% CO2 for three hours. After incubation, IPEC-J2 were washed and lysed. The suspension was serially diluted and enumerated on THY agar supplemented with and without penicillin G. The number of each strain was calculated. The data were presented as the competitive index (CI).
Confirmation of inhibitory effect by dead/live assay
The Inhibitory effect of the non-SS2 inhibitor on SS2 target was further confirmed using DMEM/F-12 co-culture assay. The overnight culture of both strains (approximately 55X108 CFU/ml) was mixed at a 1:1 ratio, and incubated at 37˚C under 5% CO2 for two hours. After incubation, co-culture mixture was collected and stained with propidium iodide at final concentration 1.875 µM (1.25 µg/ml) for 15 min, mixed with 1.2% low meling point THB agar and observed to confirm inhibitory effect under confocal laser scanning microscope.
DNA extraction, Whole genome sequencing, and data analysis
Genomic DNA of non-SS2 and SS2 strains were extracted using DNeasy Blood and Tissue Kit (Qiagen, Hilden, Germany) and sequenced by Illumina NextSeq2000 according to the manufacturer’s instructions (Illumina, San Diego, CA, USA). The sequencing data were checked for quality using FastQC v.1.0.0, and low-quality data were eliminated using DRAGEN FASTQ Toolkit v.1.1.0. SPAdes Genome Assembler v.3.9.0 were used to assemble the sequence. The genome sequence of each strain was annotated by Prokka Genome Annotation v.1.11.1. Kyoto Encyclopedia of Genes and Genomes (KEGG) was used to determine different presence and absence pathway modules in each strain.
Loss of function on IPEC-J2 co-colonization assay
IPEC-J2 cells were seeded at 4X104 cells/well and incubated overnight. After incubation, cells were washed two times and fresh new complete media (DMEM/F-12 supplemented with 1% pig serum) supplemented with and without 3-chlorocatechol inhibitor were added. Overnight cultures of SS2 and non-SS2 were mixed at a 1:1 ratio, infected IPEC-J2 at MOI 100 and then incubated at 37˚C under 5% CO2 for three hours. After incubation, IPEC-J2 were washed and lysed. The suspension was serially diluted and enumerated on THY agar supplemented with and without penicillin G. The number of each strain was calculated. The data were presented as the competitive index (CI).
Target genes expression during bacterial co-culture
RNA of non-SS2, SS2 and co-culture mixture were extracted using RNeasy Mini Kit (Qiagen, Hilden, Germany) and amplified by by one-step SYBR-based real-time RT-PCR (Agilent Technologies, Brilliant II SYBR Green QRT-PCR Master Mix Kit,USA) with CFX Opus 96 Dx Real-Time PCR Systems according to the manufacturer’s instructions (Biorad, Ho Chi Minh City,Vietnam). Cut-off in the RT-PCR was at Ct value 30.
Supplementary Information
Acknowledgements
This work was supported by the Thailand Science Research and InnovationFundamental Fund Fiscal year 2023 (Grant number 180178) and Thammasat University Research Unit in Molecular Pathogenesis and Immunology of Infectious Diseases. PS was supported by Innovet-AMR, IDRC Grant No. 110297-002.
Author contributions
P.P. and BA performed experiments. P.P. analyzed data and prepared the draft of manuscript. P.S. conceptualized, designed experiments, supevised, and reviewed and edited the manuscript. P.S. and J.T. performed funding aquisition . P.S., J.T., O.K. and P.T. reviewed the manuscript. All authors have read and agreed to the published version of the manuscript.
Data availability
The assembled genomic sequences in the current study were submitted under the BioProject PRJNA1129316 with accession number of non-SS2 including SAMN42150037 (strain 34) and SAMN42197248 (strain 147), and accession number of SS2 including SAMN42213117 (strain 06), SAMN42213056 (strain 05), SAMN42213018 (strain 279), SAMN42212905 (strain 281), SAMN42212581 (strain 676), SAMN42212570 (strain 879), SAMN42204707 (strain 399).
Declarations
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary Information
The online version contains supplementary material available at 10.1038/s41598-025-04541-z.
References
- 1.Wertheim, H. F., Nghia, H. D., Taylor, W. & Schultsz, C. Streptococcus suis: an emerging human pathogen. Clin. Infect. Dis.48(5), 617–625 (2009). [DOI] [PubMed] [Google Scholar]
- 2.Sanford, S. E. & Tilker, M. E. Streptococcus suis type II-associated diseases in swine: observations of a one-year study. J. Am. Vet. Med. Assoc.181(7), 673–676 (1982). [PubMed] [Google Scholar]
- 3.Goyette-Desjardins, G., Auger, J. P., Xu, J., Segura, M. & Gottschalk, M. Streptococcus suis, an important pig pathogen and emerging zoonotic agent-an update on the worldwide distribution based on serotyping and sequence typing. Emerg. Microbes Infect.3(6), e45 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Meekhanon, N. et al. Potentially hazardous Streptococcus suis strains latent in asymptomatic pigs in a major swine production area of Thailand. J. Med. Microbiol.66(5), 662–669 (2017). [DOI] [PubMed] [Google Scholar]
- 5.Zhang, C. et al. Capsular serotypes, antimicrobial susceptibility, and the presence of transferable oxazolidinone resistance genes in Streptococcus suis isolated from healthy pigs in China. Vet. Microbiol.247, 108750 (2020). [DOI] [PubMed] [Google Scholar]
- 6.Thongkamkoon, P., Kiatyingangsulee, T. & Gottschalk, M. Serotypes of Streptococcus suis isolated from healthy pigs in Phayao Province, Thailand. BMC Res Notes.10(1), 53 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Nutravong, T. et al. Identification of major Streptococcus suis serotypes 2, 7, 8 and 9 isolated from pigs and humans in upper northeastern Thailand. Southeast Asian J Trop Med Public Health.45(5), 1173–1181 (2014). [PubMed] [Google Scholar]
- 8.Sedano, S. A. et al. Epidemiology and genetic diversity of Streptococcus suis in smallhold swine farms in the Philippines. Sci. Rep.13(1), 21178 (2023). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Kanehisa, M. Toward understanding the origin and evolution of cellular organisms. Protein Sci.28(11), 1947–1951 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Kanehisa, M., Furumichi, M., Sato, Y., Matsuura, Y. & Ishiguro-Watanabe, M. KEGG: biological systems database as a model of the real world. Nucleic Acids Res.53(D1), D672–D677 (2024). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kanehisa, M. & Goto, S. KEGG: kyoto encyclopedia of genes and genomes. Nucleic Acids Res.28(1), 27–30 (2000). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Fierer, N. & Jackson, R. B. The diversity and biogeography of soil bacterial communities. Proc. Natl. Acad. Sci. U.S.A.103(3), 626 (2006). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Zorz, J. et al. Drivers of regional bacterial community structure and diversity in the northwest Atlantic ocean. Front. Microbiol.10.3389/fmicb.2019.00281 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Fatimawali, K. B. J., Gani, M. A. & Tallei, T. E. Comparison of bacterial community structure and diversity in traditional gold mining waste disposal site and rice field by using a metabarcoding approach. Int. J. Microbiol.2020, 1–8 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ali, B. et al. Culture independent diversity of bacterial communities indigenous to lower altitude at Laohugou glacial environment. Geomicrobiol J.38(1), 1–13 (2021). [Google Scholar]
- 16.Ghoul, M. & Mitri, S. The ecology and evolution of microbial competition. Trends Microbiol.24(10), 833–845 (2016). [DOI] [PubMed] [Google Scholar]
- 17.Khan, R., Petersen, F. C. & Shekhar, S. Commensal bacteria: an emerging player in defense against respiratory pathogens. Front. Immunol.10, 1203 (2019). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Abt, M. C. & Pamer, E. G. Commensal bacteria mediated defenses against pathogens. Curr. Opin. Immunol.29, 16–22 (2014). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Rendueles, O. & Ghigo, J.-M. Mechanisms of competition in biofilm communities. Microbiol. Spectrum10.1128/microbiolspec.mb-0009-2014 (2015). [DOI] [PubMed] [Google Scholar]
- 20.Leslie, J. L. et al. Protection from lethal clostridioides difficile infection via intraspecies competition for cogerminant. MBio10.1128/mBio.00522-21 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Liang, Z. et al. The antimicrobial systems of Streptococcus suis promote niche competition in pig tonsils. Virulence13(1), 781–793 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Weiland-Bräuer, N. Friends or foes—microbial interactions in nature. Biology.10(6), 496 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Valente, C., Cruz, A. R., Henriques, A. O. & Sá-Leão, R. Intra-species interactions in streptococcus pneumoniae biofilms. Front. Cell Infect. Microbiol.11, 803286 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ferrando, M. L. et al. Host-pathogen Interaction at the intestinal mucosa correlates with zoonotic potential of streptococcus suis. J. Infect Dis.212(1), 95–105 (2015). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Ferrando, M. L. et al. Streptococcal Adhesin P (SadP) contributes to Streptococcus suis adhesion to the human intestinal epithelium. PLoS ONE12(4), e0175639 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Bugenyi, A. W., Cho, H. S. & Heo, J. Association between oropharyngeal microbiome and weight gain in piglets during pre and post weaning life. J. Anim. Sci. Technol.62(2), 247–262 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Luise, D., Correa, F., Bosi, P. & Trevisi, P. A review of the effect of formic acid and its salts on the gastrointestinal microbiota and performance of pigs. Animals (Basel)10.3390/ani10050887 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Oh, J. K. et al. Multispecies probiotics alter fecal short-chain fatty acids and lactate levels in weaned pigs by modulating gut microbiota. J. Anim. Sci. Technol.63(5), 1142–1158 (2021). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Pietzke, M., Meiser, J. & Vazquez, A. Formate metabolism in health and disease. Mol. Metab.33, 23–37 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Quan, J. et al. A global comparison of the microbiome compositions of three gut locations in commercial pigs with extreme feed conversion ratios. Sci. Rep.8(1), 4536 (2018). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Feng, Y. et al. Attenuation of Streptococcus suis virulence by the alteration of bacterial surface architecture. Sci. Rep.2, 710 (2012). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Hussain, M. R. M. et al. The microbial pathology of Neu5Ac and gal epitopes. J. Carbohydr. Chem.32(3), 169–183 (2013). [Google Scholar]
- 33.Zhang, Y. et al. Effect of the glycosyltransferases on the capsular polysaccharide synthesis of Streptococcus suis serotype 2. Microbiol. Res.185, 45–54 (2016). [DOI] [PubMed] [Google Scholar]
- 34.Charland, N., Harel, J., Kobisch, M., Lacasse, S. & Gottschalk, M. Streptococcus suis serotype 2 mutants deficient in capsular expression. Microbiology144(2), 325–332 (1998). [DOI] [PubMed] [Google Scholar]
- 35.Ruenchit, P., Reamtong, O., Siripanichgon, K., Chaicumpa, W. & Diraphat, P. New facet of non-O1/non-O139 Vibrio cholerae hemolysin A: a competitive factor in the ecological niche. FEMS Microbiol. Ecol.10.1093/femsec/fix113 (2017). [DOI] [PubMed] [Google Scholar]
- 36.Liu, X., Wu, B., Nakamoto, S. T., Imamura, J. L. & Li, Y. Inhibition of Listeria monocytogenes and escherichia coli O157:H7 growth by Ohelo berry (Vaccinium calycinum) fractions: anthocyanins, non-anthocyanin phenolics, and organic acids. Microorganisms.10(11), 2231 (2022). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Bhattacharjee, M. K. Better visualization and photodocumentation of zone of inhibition by staining cells and background agar differently. J. Antibiot.68(10), 657–659 (2015). [DOI] [PubMed] [Google Scholar]
- 38.Menetrey, Q., Dupont, C., Chiron, R., Jumas-Bilak, E. & Marchandin, H. High occurrence of bacterial competition among clinically documented opportunistic pathogens including achromobacter xylosoxidans in cystic fibrosis. Front. Microbiol.10.3389/fmicb.2020.558160 (2020). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Aoki, S. K. et al. Contact-dependent inhibition of growth in Escherichia coli. Science309(5738), 1245–1248 (2005). [DOI] [PubMed] [Google Scholar]
- 40.Filek K. Contact-dependent growth inhibition in Escherichia coli EC93 2018. [DOI] [PMC free article] [PubMed]
- 41.Whitney, J. C. et al. A broadly distributed toxin family mediates contact-dependent antagonism between gram-positive bacteria. Elife6, e26938 (2017). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Koskiniemi, S. et al. Rhs proteins from diverse bacteria mediate intercellular competition. Proc. Nat. Acad. Sci.110(17), 7032–7037 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Zakrzewski, S. S. et al. Improved cell line IPEC-J2, characterized as a model for porcine jejunal epithelium. PLoS ONE8(11), e79643 (2013). [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The assembled genomic sequences in the current study were submitted under the BioProject PRJNA1129316 with accession number of non-SS2 including SAMN42150037 (strain 34) and SAMN42197248 (strain 147), and accession number of SS2 including SAMN42213117 (strain 06), SAMN42213056 (strain 05), SAMN42213018 (strain 279), SAMN42212905 (strain 281), SAMN42212581 (strain 676), SAMN42212570 (strain 879), SAMN42204707 (strain 399).