Streptococcus suis serotype 2 is a world-wide agent of diseases among pigs including meningitis, septicemia and arthritis. This microorganism is also recognized as an important zoonotic agent. The pathogenesis of the meningitis caused by S. suis is poorly understood. We have previously shown that S. suis is able to adhere to human brain microvascular endothelial cells (BMEC), but not to human umbilical vein endothelial cells (HUVEC). The objective of this work was to study the ability of S. suis serotype 2 to induce the release of the pro-inflammatory cytokines tumor necrosis factor alpha (TNF-α), interleukin-1 (IL-1); IL-6 and the chemokines IL-8 and monocyte chemotactic protein-1 (MCP-1) by human BMEC and HUVEC, using a sandwich enzyme-linked immunosorbent assay. S. suis was able to stimulate the production of IL-6, IL-8 and MCP-1 by BMEC but not HUVEC, in a time- and concentration-dependent manner. Bacterial cell wall components were largely responsible for such stimulation. The human and pig origin of strains does not seem to affect the intensity of the response; indeed, a very heterogeneous pattern of cytokine and chemokine production was observed for the different strains tested in this study. In situ production of cytokines and chemokines by BMEC may be the result of specific adhesion of S. suis to this cell type, with several consequences such as increased recruitment of leukocytes and an increase in the blood-brain barrier permeability.
Brain microvascular endothelial cells
Streptococcus suis causes many swine diseases including meningitis, septicemia, arthritis and pneumonia. Of the 35 official serotypes described to date, serotype 2 is the most virulent and the most commonly isolated from diseased pigs. This miroorganism is also recognized as an agent of zoonosis. In fact, over 200 cases of human infection by S. suis have been reported, especially among persons in close contact with pigs or pig products. S. suis causes mainly meningitis in humans, with hearing loss as the most frequent sequela .
The pathogenesis of meningitis caused by S. suis is poorly understood and is probably a multistep process. It is not known how bacteria are able to traverse the mucosal epithelial barrier to reach the bloodstream . Once there, bacteria can travel inside monocytes  or free in circulation, as demonstrated by several studies during the last decade [4,5,13,32]. In fact, the presence of the capsular polysaccharide (CPS) protects bacteria against phagocytosis [4,32]. The mechanisms by which S. suis traverses the blood-brain barrier (BBB) into the subarachnoid space to cause meningitis are unknown. Other meningeal pathogens including Streptococcus pneumoniae, Escherichia coli K1 and group B Streptococcus (GBS), are known to interact directly with the BBB as free bacteria . This barrier, responsible for maintaining biochemical homeostasis within the central nervous system (CNS), is characterized by intercellular tight junctions that regulate fluid, macromolecule and cell traffic across the layer . The BBB is composed of the arachnoid membrane, the brain microvascular endothelial cells (BMEC) and the choroid plexus. The primary site of breakdown of the BBB in most bacterial meningitis appears to be the BMEC .
It is generally accepted that bacterial interactions with BMEC are mainly characterized by specific attachment and consequent invasion, toxicity and increased permeability . S. suis serotype 2 has been shown to adhere to human BMEC, but unlike other meningeal pathogens, invasion does not occur. The adhesion appears to be related to the cell type, since S. suis does not adhere to human umbilical vein endothelial cells (HUVEC) . Adhesion of S. suis to BMEC may have different consequences which may lead to increased permeability of the BBB. For example, some strains produce a toxin (suilysin) [12,17], that was reported as being toxic for BMEC and other cells [6,21]. However, only European strains of S. suis produce this hemolysin . In fact, production of toxic factors by the majority of virulent North American strains, that might lead to cell damage and BBB increased permeability, has not been described so far. This suggests that the pathogenesis of meningitis produced by European and North American strains may differ. In fact, other virulence-related proteins are produced mainly by European strains [13,45].
Increased BBB permeability may also be induced by inflammatory mediators that might be produced following adhesion of bacteria to cells. Recent work in our laboratory show that S. suis is not only able to interact with monocytes/macrophages, but is also able to induce the release of several pro-inflammatory cytokines and chemokines, such as tumor necrosis factor alpha (TNF-α), interleukin (IL)-1, IL-6, IL-8 and monocyte chemotactic protein-1 (MCP-1) [27,28]. However, the source of pro-inflammatory cytokines in the cerebrospinal fluid during meningitis may be microglial and endothelial cells as well as migrating leukocytes [13,35]. Recent studies show that pro-inflammatory cytokines are produced by BMEC that are stimulated either by other cytokines (such as TNF-α or IL-1) or by a direct interaction with microbial pathogens . Cytokines and adhesion molecules expressed by endothelial cells are known to be key players in regulating the recruitment of leukocytes to the sites of inflammation . The objective of this work was to study the ability of S. suis serotype 2 to induce the release of the pro-inflammatory cytokines TNF-α, IL-1, IL-6 and the chemokines IL-8 and MCP-1, by human BMEC and HUVEC.
2 Materials and methods
2.1 Bacterial strains and growth conditions
S. suis capsular type 2 virulent strains 89–1591 and S735, from North America (Canada) and Europe (The Netherlands), respectively, were used throughout this study. The isogenic unencapsulated mutant 2A, derived from strain S735 and obtained by Tn916 transposition, was also used . Other porcine and human strains of S. suis also used in this study are listed in Table 1. Bacteria, maintained as stock cultures in 50% glycerol-Todd–Hewitt broth (THB; Difco Lab., Detroit, MI, USA) at −80°C, were grown overnight on bovine blood agar plates at 37°C and isolated colonies were used as inocula for THB, that were incubated for 18 h at 37°C. Working cultures for endothelial cell stimulation were produced by inoculating 10 ml of these cultures in 200 ml of THB at 37°C with agitation until they reached the mid-exponential phase (6 h incubation time; OD540 of 0.4–0.5). Bacteria were washed twice in phosphate-buffered saline (PBS) pH 7.4, and diluted to approximately 109 CFU ml−1 in PBS. A more accurate determination of the CFU ml−1 in the final suspension was made by plating on THB agar. Bacteria were then killed by heat treatment at 60°C for 45 min (minimal experimental condition required for killing of S. suis) . Subcultures of the heat-treated suspension on blood agar plates were incubated at 37°C for 48 h to confirm the absence of viable organisms. Killed bacterial preparations were stored at 4°C and re-suspended in cell culture media just before stimulation assays.
Human BMEC line, originating from a brain biopsy of an adult human female with epilepsy was kindly provided by Dr. K. Kim, Johns Hopkins University School of Medicine, Baltimore, MD, USA. Cells had been immortalized by transfection with simian virus 40 large T antigen and were shown to maintain their morphologic and functional characteristics . Cells were grown in RPMI 1640 medium (Gibco, Burlington, VT, USA) supplemented with 10% heat-inactivated fetal bovine serum (FBS; Gibco), 10% Nu-serum IV supplement (Becton Dickinson, Bedford, MA, USA), l-glutamin (ICN Biomedical Inc., Aurora, OH, USA) and penicillin—streptomycin (5000 U ml−1) (Gibco). Flasks (Falcon) and 24-well tissue culture plates (Becton Dickinson) were precoated with rat tail collagen to support the cells . HUVEC derived from human umbilical cord were purchased from the American Type Culture Collection (ATCC CRL-1730). Cells were grown in F-12K medium (Sigma-Aldrich, Oakville, ON, Canada) supplemented with 10% heat-inactivated FBS, endothelial cell growth supplement (30 µg ml−1) (Becton Dickinson) and penicillin—streptomycin. Flasks and 24-well tissue culture plates were pre-coated with 1% gelatin to support the cells. Both types of endothelial cells were incubated at 37°C, with 5% CO2 in a humid atmosphere. Cells were used before passage 35 for all experiments.
2.3 Stimulation of cells
For stimulation assays, 48-h cultures of BMEC or HUVEC cells in flasks were trypsinized and diluted in culture media at 105 cells ml−1, and 1 ml of this suspension was distributed in 24-well plates and incubated to confluence. At confluence, medium was removed and heat-killed S. suis strains (1 ml) was added at appropriate dilutions made in culture media. Comparative studies of cytokine induction by S. suis strain S735 and its unencapsulated mutant 2A were always done in parallel . In separate experiments, cells were stimulated with different concentrations of purified CPS, cell wall or suilysin, purified as previously described [17,27,30]. Endothelial cells stimulated with lipopolysaccharide (LPS) from E. coli 0127:B8 (Sigma-Aldrich) (10 µg ml−1 for BMEC and 1 µg ml−1 for HUVEC) served as a positive control. Cells incubated in medium alone served as controls for spontaneous cytokine release. Cytokine induction plates were incubated at 37°C, 5% CO2 in a humid atmosphere. At different time intervals (see below), culture supernatants were harvested from individual wells, and the supernatants aliquoted and frozen at −20°C until cytokine determinations were performed. Each test of BMEC or HUVEC stimulation was repeated at least three times. All solutions and bacterial preparations used in this study were tested for the presence of endotoxin by a Limulus amebocyte lysate (LAL) gel-clot test (Pyrotell STV, Cape Cod, Falmouth, MA, USA) with a sensitivity limit of 0.03 EU ml−1. In some experiments, endotoxin contamination during stimulation of endothelial cells was controlled by parallel assays with Polymixin B (PmB; 10 µg ml−1). Results from the LAL test and/or data from PmB treatment demonstrated no significant levels of endotoxin contamination from different bacterial preparations (data not shown). Cell culture medium contained less than 0.03 EU ml−1. Absence of cell toxicity at all bacterial concentrations as well as with purified bacterial components was confirmed by the lactate dehydrogenase cellular injury assay, as previously described .
2.4 Enzyme-linked immunosorbent assays (ELISA) for cytokines
IL-1, TNF-α, IL-6, IL-8 and MCP-1 were measured by sandwich ELISA, using pair-matched monoclonal antibodies from R&D Systems (Minneapolis, MN, USA), as previously described . Standard curves were included in each ELISA plate (Nunc, VWR, Ville Mont Royal, QC, Canada) as two-fold dilutions of recombinant (R&D Systems) human IL-6 (1500 to 3 pg ml−1), IL-8 (600 to 5 pg ml−1), MCP-1 (500 to 8 pg ml−1), IL-1 (300 to 5 pg ml−1) or TNF-α (3000 to 46 pg ml−1). Supernatant dilutions giving optical density readings in the linear portion of the appropriate standard curve were used to determine the level of each cytokine in the samples. Standard and sample dilutions were added in duplicate wells to each ELISA plate.
2.5 Statistical analysis
ELISA tests were performed at least four times for each individual endothelial cell stimulation assay. Results were derived from linear regression calculations and expressed in pg ml−1 of cytokine. Differences were analyzed for significance by using the Student's unpaired t test (two-tailed P value), with a P value <0.05 considered as significant. Differences between the human and porcine origin group of strains and differences among strains within the same group were analyzed for significance by using general linear models, followed by Tukey—Kramer post-hoc tests for differences between strains. The SAS software (SAS, Cary, NC, USA) was used for these analyses.
3 Results and discussion
3.1 BMEC but not HUVEC produce IL-6, IL-8 and MCP-1 after S. suis whole cell stimulation
Previous studies have shown that stimulated endothelial cells are able to produce IL-6 and different chemokines [3,25,26,34,44,48]. Unstimulated BMEC yielded low basal levels of IL-6, IL-8 and MCP-1 expression. These basal values were subtracted to correct data obtained after S. suis or LPS stimulation throughout this work. The production of cytokines and chemokines by BMEC stimulated with S. suis varied with incubation time. Both S. suis North American 89–1591 and European S735 strains were able to induce high levels of IL-6, low levels of MCP-1 and intermediate levels of IL-8 from stimulated BMEC and no significant difference in production was observed between the two strains (P>0.05). Maximal release of IL-6 was achieved between 12 and 48 h of incubation with bacteria. On the other hand, LPS stimulation lead to a gradual increase of IL-6 with time, reaching its highest level at 48 h of incubation (Fig. 1a). Similar kinetics for IL-8 and MCP-1 production were observed with both bacteria and LPS, with maximal release observed after 48 h of incubation (Fig. 1b,c). However, cytokine levels released by cells stimulated with LPS were significantly higher (P<0.001). Endothelial cells have been shown to release cytokines and chemokines with [18,31,33] or without  pre-activation with IL-1 and/or TNF-α. Since the BMEC used in this study did not produce either cytokine (see below), the observed levels of IL-6, IL-8 and MCP-1 could be considered TNF-α- and IL-1-independent; however, a possible amplifying role of IL-1 and TNF-α in vivo should not be ruled out. The BBB is a complex system that involves dynamic interplay of BMEC with perivascular cells such as astrocytes and macrophages. Our system focused only on the capacity of BMEC to produce pro-inflammatory cytokines and chemokines and did not attempt to capture the full complexity of these interactions.
Time course of IL-6 (a), IL-8 (b) and MCP-1 (c) production by BMEC after stimulation with heat-killed S. suis serotype 2 (strains 89–1591 and S735; 109CFU ml−1). Purified LPS (10 µg ml−1) was used as a positive control. Culture supernatants were harvested at different time intervals and assayed for cytokine production by ELISA. Data are expressed as means±S.D. (in pg ml−1). Values for basal cytokine expression (cell culture medium alone) were corrected from data obtained after S. suis or LPS stimulation.
As done for BMEC, basal production of the cytokine and chemokines by HUVEC was subtracted from experimental results. Although a strong response was observed with LPS, S. suis was not able to induce any upregulation of IL-6, IL-8 or MCP-1 production by HUVEC (Fig. 2a–c). HUVEC response appeared to be more sensitive to LPS than BMEC, as a much higher level of production was observed even though the concentration of LPS used for HUVEC induction was 10 times lower than that used for BMEC. A possible explanation for the sensitivity of BMEC to S. suis is that the induction of cytokines is the result of bacteria—cell adhesion. As indicated previously, S. suis is able to adhere to BMEC but not to HUVEC . It has already been shown for other bacterial species that adhesion is needed to stimulate cytokine production in cells . For example, Streptococcus bovis is able to induce IL-8 expression after adhesion to endothelial cells  and only adherent Neisseria meningitidis induces the expression of TNF-α by endothelial cells . Similarly, the cytokine-stimulatory activity of the CPS of Staphylococcus aureus on endothelial cells resulted from ligand—receptor interactions . Another possible explanation is that some receptors, such as the toll-like receptor 2, which has been shown to confer responsiveness to a wide variety of Gram-positive bacterial cell wall components , are present in the BMEC but absent in the HUVEC tested in this study. It has been recently reported that S. suis is able to induce the upregulation of pro-inflammatory cytokines from monocytes by CD14-dependent and -independent pathways . Receptors activated by S. suis, other than CD14, are presently under study in our laboratory. Since S. suis is not able to invade BMEC , cytokine activation seems to take place without cell invasion. Similarly, activation and induction of cytokine production in endothelial cells stimulated with Listeria monocytogenes occurs without cellular invasion . S. suis is also able to adhere to, but is not ingested, by monocytes, and this interaction induces the release of large amounts of pro-inflammatory cytokines [28,27].
Time course of IL-6 (a), IL-8 (b) and MCP-1 (c) production by HUVEC after stimulation with heat-killed S. suis serotype 2 (strains 89–1591 and S735; 109CFU ml−1). Purified LPS (1 µg ml−1) was used as a positive control. Culture supernatants were harvested at different time intervals and assayed for cytokine production by ELISA. Data are expressed as means±S.D. (in pg ml−1). Values for basal cytokine expression (cell culture medium alone) were corrected from data obtained after S. suis or LPS stimulation.
BMEC and HUVEC tested in this study were not able to produce IL-1 or TNF-α after stimulation with high doses of LPS or S. suis, even if pre-stimulated with interferon-gamma (data not shown). Furthermore, no mRNA signal corresponding to any of these cytokines could be detected by RT-PCR after LPS stimulation of BMEC (unpublished observations). It has been shown that oral viridans streptococci are able to induce IL-6 and IL-8, but not TNF-α or IL-1, from stimulated endothelial cells  and N. meningitidis-stimulated endothelial cells induce the production of TNF-α only in the presence of monocytes . Other reports indicate that IL-1 production (or gene expression) occurs in endothelial cells stimulated with bacteria or LPS [3,23,48]. Since the cells used in this study did not produce any of these cytokines with the positive control used, no conclusion on a possible IL-1 and TNF-α induction by S. suis could be made. Interestingly, both cytokines were shown to be significantly upregulated in human monocytes activated by S. suis by using the ELISA detection tests described in the present study .
The effect of bacterial concentration on cytokine production was determined. Cell-culture supernatants were harvested after 24 h of stimulation to evaluate IL-6 production, and after 48 h to measure IL-8 and MCP-1 induction by BMEC exposed to different concentrations of heat-killed S. suis strains 89–1591 or S735. Cytokine induction varied directly with bacterial concentration, and only a high concentration of bacteria was able to induce cytokine production. In fact, a concentration greater than 2.5×108 CFU ml−1 was needed to obtain cytokine release (data not shown). This is in agreement with results observed with human monocytes and murine macrophages [27,28]. Interestingly, the presence of clinical signs and symptoms in diseased animals correlates with those high levels of virulent bacteria in the bloodstream .
3.3 Lack of relationship between the origin or the virulence of strains and cytokine induction
Despite the fact that S. suis serotype 2 is usually associated with severe occupational disease in humans , studies using strains of human origin are limited. Since cells used in this and earlier studies [21,28] are of human origin, it was relevant to compare the ability of porcine strains to induce cytokine release with those recovered from serious cases of human disease. A very heterogeneous pattern of cytokine production was observed, with no tendency for human strains to induce higher cytokine levels (Fig. 3). Tukey—Kramer post-hoc tests revealed significant differences between strains within each group (P<0.01). In fact, no consistent effect on cytokine production could be attributed to the origin of the strains, specially for those of human origin. Similar observations have been reported for S. suis interaction with human monocytes, for oral viridans streptococci and for S. aureus[28,44,48]. The observed variability may be due to the degree of exposure and/or type of components of the bacterial surface, such as bacterial cell wall, which can stimulate cytokine release from endothelial cells. Despite the epidemiological fact that pigs may be the sole source of human infections [7,39], cases of S. suis infection in individuals not associated with the porcine industry have also been reported . Thus, the clinical relevance of potential species-specific differences in reactivity to bacterial strains still remains unclear.
Comparative study of cytokine production by different S. suis strains. BMEC were stimulated by heat-killed (109 CFU ml−1) S. suis serotype 2 strains from human or porcine origin (Table 1). IL-6 (a), after 24 h incubation, and IL-8 (b) and MCP-1 (c) after 48 h incubation were measured by ELISA titration of stimulated cell supernatants. Lines represent average cytokine production by each group of strains. Each point represents one strain, and is the average of at least three separate experiments.
The ability of virulent and non-virulent strains of S. suis to induce IL-6, IL-8 and MCP-1 was also compared. The concept of virulence for S. suis is currently debated in the literature [11,13]. In this study, we considered a strain as ‘virulent’ or ‘non-virulent’ depending on the presence or absence of clinical disease after experimental infections in piglets (Table 1). In this work, no association was observed between the cytokine response and the virulence of the strain. This is in agreement with results reported by Segura et al. . Similarly, there were no observed differences between virulent and non-virulent strains in their adhesion levels to different types of host cells, including BMEC (unpublished observations). Unlike other important streptococcal species, information on S. suis virulence factors as well as on surface-expressed molecules is limited . It has been shown in other bacterial species, such us Rodococcus equi, that virulence is not necessarily correlated with the level of cytokine production . In the case of S. suis, it has been suggested that only virulent strains are able to survive at high numbers in the bloodstream and induce disease . Indeed, recent research indicates that, unlike non-virulent strains, virulent S. suis strains are able to survive in circulation at high concentrations for more than 6 days .
3.4 Relative role of bacterial components in cytokine production
Different bacterial structures and products have been potentially implicated in the pathogenesis of the S. suis infection [13,36], but understanding of the effect that these proposed virulence factors have on cytokine release is limited. In the present study, S. suis purified cell wall material was able to induce IL-6 and IL-8 production by BMEC. At a concentration of 500 µg ml−1 of cell wall, the cytokine production level was roughly equivalent to half of that produced by whole bacteria. It has previously been shown that the cell wall of S. suis is the main component responsible for cytokine induction by murine macrophages . Similarly, antigens extracted from the cell wall of S. bovis and pneumocci induce pro-inflammatory cytokines from different type of cells [9,42]. The possible role of cell wall components of S. suis on the upregulation of IL-6 and IL-8 was confirmed by the use of an unencapsulated mutant, that was able to induce higher levels of IL-6 and IL-8, than the encapsulated parent S735 strain (P<0.001) (Fig. 4a,b). In the present study, the capacity of the unencapsulated mutant to induce higher levels of IL-6 and IL-8 was probably not the result of greater adhesion of bacteria to cells, since both encapsulated and unencapsulated strains adhere similarly to BMEC .
Comparative study of IL-6 (a), IL-8 (b) and MCP-1 (c) production by different proposed virulence factors for S. suis serotype 2. BMEC were stimulated with different concentrations of purified CPS, cell wall or hemolysin. In addition, the cytokine induction by S. suis strain S735 was compared to that obtained with its unencapsulated mutant 2A. Data are expressed as means±S.D. (in pg ml−1). Values for basal cytokine expression (cell culture medium alone) were corrected from data obtained after stimulation.
Results indicating that the capsule itself has no effect on the upregulation of these cytokines were confirmed by testing different concentrations of purified CPS. Concentrations as high as 100 µg ml−1 did not induce significant levels of cytokine release compared to either negative control or whole bacteria (Fig. 4a,b). This is in agreement with results previously observed with murine macrophages . Several in vitro and in vivo studies with purified CPS or with unencapsulated mutants failed to demonstrate a major role for CPS of pathogenic Gram-positive cocci in cytokine induction . In the case of S. suis, however, the capsule may indirectly contribute to cytokine induction. In fact, the polysaccharide capsule is probably responsible for the progression of the disease by allowing S. suis to evade host defense mechanisms such as phagocytosis . As shown in this and previous works [27,28], a high concentration of bacteria is needed to upregulate the production of pro-inflammatory cytokines. Thus, only well-encapsulated bacteria may be protected and survive at high concentrations in the bloodstream to reach the BBB and stimulate cells.
Since it has been shown that several toxins can stimulate or modulate the inflammatory mediator cascade , the cytokine induction by the extracellular hemolysin (suilysin), a possible virulence factor among European strains , was determined. A high IL-6 and IL-8 response was obtained with purified suilysin (Fig. 4a,b). Rose et al.  have recently demonstrated that listeriolysin, a hemolysin produced by L. monocytogenes, is largely responsible for endothelial cytokine upregulation. The production of suilysin by European strains may contribute to a higher local inflammatory response. The fact that virulent European suilysin-positive strains present a higher virulence potential than virulent North American suilysin-negative strains has already been proposed . Furthermore, it has been shown that a suilysin-negative mutant was not virulent for mice and less virulent for pigs than its hemolytic parent strain . In the present study, heat-killed washed bacterial suspensions (free of suilysin) were used as stimuli for BMEC. Thus, the inherent capacity of suilysin-positive strains to induce cytokines by BMEC may have been underestimated.
The kinetics of MCP-1 production following stimulation with purified components of S. suis was somehow different from that obtained for IL-6 and IL-8. BMEC did not produce MCP-1 after stimulation with purified CPS but were extremely responsive to low concentrations of cell wall (Fig. 4c). In fact, purified cell wall concentrations as low as 1 µg induced MCP-1 levels similar to those obtained with whole bacteria. However, no activation was obtained with the unencapsulated mutant. Interestingly, purified cell wall was produced from the same mutant . It may be hypothesized that MCP-1-stimulating components on the bacterial surface have been affected in the mutant and cell wall purification methods could make these components available for BMEC stimulation. As previously observed with S. suis stimulated macrophages, bacterial molecules responsible for stimulating the upregulation of different cytokines are probably different and present in the cell wall . Finally, no significant upregulation of MCP-1 could be observed after stimulation of cells with suilysin (Fig. 4c).
The fact that S. suis-activated vascular endothelium expresses several different cytokines supports the contention that these active molecules act as secondary immune response modulators. Cytokines released by the BBB may act to modulate their activity or that of nearby cells, such as astrocytes and glial cells. These pro-inflammatory cytokines may play an important role in initiating changes in permeability or adhesion properties of the same BMEC that allow the immune cells to infiltrate the CNS in cases of meningitis caused by S. suis.
We wish to thank Dr. M. Kobisch (Centre National d'Études Vétérinaires et Alimentaires, Ploufragan, France), Dr. U. Vecht (DLO-Institute for Animal Science and Health, Lelystad, The Netherlands), Dr. T. Alexander (University of Cambridge, England), Dr. L. Brasme (Centre Hospitalier Universitaire de Reims, France), Dr. G. Grise (Hôpital des Feugrois, Elbeuf, France), Dr. B. Cattier (Centre Hospitalier Universitaire Bretonneau, Tours, France), Dr. B. François (Dupuytren Hospital, Limoges Cedex, France), and Dr. P. Norton (Institute for Animal Health, Compton, UK) for providing some of the S. suis type 2 strains used in this study. This work was supported by the Natural Sciences and Engineering Research Council of Canada (NSERC) Grant # 0680154280, by the Fonds pour la Formation des Chercheurs et l'Aide à la Recherche du Québec (FCAR-équipe) Grant # 99-ER-0214, and by the Canadian Research Network on Bacterial Pathogens of Swine.
(1995) Infection of bovine brain microvessel endothelial cells with Cowdria ruminantium elicits IL-1 beta, -6, and -8 mRNA production and expression of an unusual MHC class II DQ alpha transcript. J. Immunol. 154, 4032–4038.
(1999) Identification and characterization of the cps locus of Streptococcus suis serotype 2: the capsule protects against phagocytosis and is an important virulence factor. Infect. Immun. 67, 1750–1756.
(1995) Capsular polysaccharide types 5 and 8 of Staphylococcus aureus bind specifically to human epithelial (KB) cells, endothelial cells, and monocytes and induce release of cytokines. Infect. Immun. 63, 1380–1386.
(1997) Bovine brain microvascular endothelial cells transfected with SV40-large T antigen: development of an immortalized cell line to study pathophysiology of CNS disease. In Vitro Cell. Dev. Biol. 33, 243–247.
(1995) Microbe-endothelium interactions in blood-brain barrier permeability during bacterial meningitis: Although bacterial pathogens can directly disrupt the barrier, a role for host factors is still under study. ASM News 61, 294–298.
(1996) Cytokine production by human epithelial and endothelial cells following exposure to oral viridans streptococci involves lectin interactions between bacteria and cell surface receptors. Infect. Immun. 64, 3016–3022.