ABSTRACT
Streptococcus suis is an important swine and human pathogen. Assessment of susceptibility to S. suis using animal models has been limited to monitoring mortality rates. We recently developed a hematogenous model of S. suis infection in adult CD1 outbred mice to study the in vivo development of an early septic shock-like syndrome that leads to death and a late phase that clearly induces central nervous system damage, including meningitis. In the present study, we compared the severities of septic shock-like syndrome caused by S. suis between adult C57BL/6J (B6) and A/J inbred mice. Clinical parameters, proinflammatory mediators, and bacterial clearance were measured to dissect potential immune factors associated with genetic susceptibility to S. suis infection. Results showed that A/J mice were significantly more susceptible than B6 mice to S. suis infection, especially during the acute septic phase of infection (100% of A/J and 16% of B6 mice died before 24 h postinfection). The greater susceptibility of A/J mice was associated with an exaggerated inflammatory response, as indicated by their higher production of tumor necrosis factor alpha, interleukin-12p40/p70 (IL-12p40/p70), gamma interferon, and IL-1β, but not with different bacterial loads in the blood. In addition, IL-10 was shown to be responsible, at least in part, for the higher survival in B6 mice. Our findings demonstrate that A/J mice are very susceptible to S. suis infection and provide evidence that the balance between pro- and anti-inflammatory mediators is crucial for host survival during the septic phase.
Streptococcus suis is one of the most important pathogens affecting the swine industry. This pathogen not only causes septicemia but also affects the central nervous system (CNS) and other tissues, leading to meningitis, endocarditis, pneumonia, and arthritis. Although 35 serotypes have been described so far, serotype 2 is still the most frequently isolated from diseased animals (20). The ability of S. suis to cause disease is not limited to pigs. In humans, the organism is responsible for cases of septicemia with septic shock, meningitis, endocarditis, and other pathologies (2). Human S. suis infections are reported mainly for people with occupational exposure to infected pigs or raw pork products, and the latest serious outbreaks in Asia have led to critical recognition of S. suis as an emerging agent of zoonosis. The most recent outbreak in China resulted in more than 200 human cases which were characterized by an unusual clinical presentation of toxic shock-like syndrome with an increased death rate as high as 20% (50). In Vietnam, S. suis is by far the most common cause of adult meningitis (31).
The pathogenesis of infection is poorly understood. To induce clinical disease in swine, it is believed that S. suis enters through the respiratory route to colonize the tonsils. In humans, however, the entrance is mainly through skin abrasions, although other routes of infection have also been proposed (3, 16). Thereafter, by mechanisms presently not well understood, it reaches the bloodstream, where it disseminates freely or as cell-bound bacteria attached to phagocytes (15) and causes bacteremia that can, in some cases, lead to fatal septicemia. If the host survives this critical phase of infection, S. suis may colonize the CNS and cause meningitis. Our knowledge of the bacterial components involved in the pathogenesis of S. suis infection is also limited. These components include the antiphagocytic polysaccharide capsule, a hemolysin known as suilysin, and different cell-wall associated and extracellular proteins, including enzymes (15, 20, 38). The wide distribution of these factors among strains isolated either from asymptomatic or clinically ill animals suggests that the pathogenesis of S. suis is multifactorial.
Clinical septicemia with septic shock and meningitis may be related to an exacerbated or uncontrolled inflammatory response that is also, in the case of meningitis, accompanied by an increase in the permeability of the blood-brain barrier (15). For example, S. suis is able to upregulate different adhesion molecules expressed by monocytes, thereby increasing leukocyte recruitment to sites of infection and providing a boost to the inflammatory response (1). In addition, it induces the production of different proinflammatory cytokines and chemokines by human, mouse, and swine leukocytes and by human and swine brain microvascular endothelial cells (39-41, 46). It was reported only recently that human and murine monocytes/macrophages recognize whole S. suis or its purified cell wall components mainly through a Toll-like receptor 2 (TLR2)-dependent pathway, with the possible participation of CD14. This recognition was associated with a triggering of the inflammatory response via a MyD88-dependent downstream signaling pathway and the subsequent production of proinflammatory cytokines and chemokines (17, 40).
Assessment of susceptibility to S. suis infection by use of experimental animal models has been limited to the evaluation of overall mortality (4, 23, 49). Recently, our laboratory developed a hematogenous model of S. suis infection in adult CD1 outbred mice to study the in vivo development of an early septic shock-like syndrome that causes high mortality and a late phase that clearly induces CNS damage, including meningitis (11). We observed that during the septic phase of infection, S. suis induces a Th1 response that is characterized by an upregulation of key proinflammatory cytokines such as tumor necrosis factor alpha (TNF-α), interleukin-1β (IL-1β), IL-12p40/p70, IL-6, and gamma interferon (IFN-γ), as well as chemokines including CCL2 (monocyte chemoattractant protein-1 [MCP-1]), CXCL1 (KC), and CCL5 (RANTES). We postulate that these inflammatory mediators are responsible for exacerbating the inflammation and for the high host mortality occurring during the septic phase of infection. In mice that develop clinical meningitis, S. suis is also able to trigger the expression of proinflammatory genes in the brain, including the TLR2, TLR3, CD14, IκBα (an index of NF-κB expression), IL-1β, TNF-α, and MCP-1 genes (11).
Development of a suitable animal model enabling researches not only to assess mortality caused by S. suis infection but also to gain new insights into host immune responses to infection with gram-positive bacteria might allow a delineation of factors modulating host susceptibility to disease. The aim of this study was to compare the severities of septic shock-like syndrome in adult C57BL/6J (B6) and A/J inbred mice. We measured sepsis-related clinical parameters, pro- and anti-inflammatory mediators, and bacterial clearance to dissect potential immune factors associated with genetic susceptibility to S. suis infection.
MATERIALS AND METHODS
Bacterial strain and growth conditions. S. suis serotype 2 strain 31533 was used for all experimental infections. This encapsulated, suilysin-positive, virulent strain has been used largely in cell stimulation studies (39-41) as well as for experimental infections of mice and swine (11, 26). Bacteria were grown overnight on sheep blood agar plates at 37°C and isolated colonies were inoculated into 5 ml of Todd-Hewitt broth (THB) (Difco Laboratories, Detroit, MI), which was incubated for 8 h at 37°C with agitation. Working cultures were prepared by transferring 10 μl of 1/1,000 dilutions of 8-h cultures into 30 ml of THB, which was incubated for 16 h at 37°C with agitation. Stationary-phase bacteria were washed twice in phosphate-buffered saline (PBS) (pH 7.3). The bacterial pellet was then resuspended and adjusted to a concentration of 5 × 108 CFU/ml. The inoculum for experimental infection was diluted in THB to obtain a final concentration of 1 × 107 CFU/ml. This final suspension was plated onto blood agar to accurately determine the CFU/ml.
Mice and experimental infections.Female 6- to 8-week-old B6 or A/J mice (Jackson Laboratory, Bar Harbor, ME) were acclimatized to standard laboratory conditions of a 12-h-light/12-h-dark cycle with free access to rodent chow and water. All experiments involving mice were conducted in accordance with the guidelines and policies of the Canadian Council on Animal Care and the principles set forth in the Guide to the Care and Use of Experimental Animals (http://www.ccac.ca/en/CCAC_programs/Guidelines_Policies/PDFs/ExperimentalAnimals_GDL.pdf ) and were approved by the Animal Welfare Committee of the Université de Montréal. On the day of the experiment, a 1-ml volume of either the bacterial suspension (1 × 107 CFU/ml) or the vehicle solution (sterile THB) was administrated by intraperitoneal (i.p.) injection. Two independent preliminary trials were performed to establish the optimal bacterial dose and time points (data not shown).
Study of mortality and clinical signs.A total of 19 mice per mouse strain were closely monitored daily to record mortality and clinical signs of disease, such as depression, rough appearance of hair coat, and swollen eyes (11). Weight loss associated with infection was also recorded daily. Mice exhibiting extreme lethargy were considered moribund and were humanely euthanized.
Histopathology studies.For histopathological examination, samples of the brain, heart, liver, and spleen were fixed in 10% buffered formalin. After paraffin embedding, 4-μm-wide tissue sections were stained with hematoxylin-eosin according to standard protocol and examined by light microscopy.
Study of the systemic inflammatory response.A total of 38 B6 and 36 A/J mice were assessed for bacteriology and systemic levels of inflammatory mediators. Samples were taken at 3, 6, 12, 24, 36, 48, and 72 h postinfection (p.i.) and analyzed as described below.
Determination of viable bacteria in organs.At each time point, at least three infected and two noninfected mice per mouse strain were euthanized with CO2 and sampled. Blood was collected by cardiac puncture, and the brains, livers, and spleens were obtained aseptically. The organs (0.05 g/organ) were trimmed, placed in 500 μl of PBS (pH 7.3), and homogenized with a vortex. Then, 50 μl of serial dilutions of the homogenate in PBS was plated onto blood agar plates. Blood samples (50 μl) were also processed. All samples were plated using an Autoplate 4000 automated spiral plater (Spiral Biotech, Inc., Norwood, MA). Blood agar plates were incubated overnight at 37°C. Colonies were counted and expressed as CFU/0.05 g for organ samples or as CFU/ml for blood samples.
Plasma collection and measurement of cytokines and chemokines.Blood from CO2-anesthetized mice was collected by cardiac puncture into heparinized tubes and centrifuged at 10,000 × g for 10 min to obtain plasma. Samples were preserved at −80°C until analysis. Levels of IL-1β, IL-6, IL-10, IL-12p40/p70, TNF-α, IFN-γ, CCL2 (MCP-1), CXCL1 (KC), and CCL5 (RANTES) in plasma were determined using a liquid multiarray system (Luminex Co., Austin, TX). Commercial multiplex-coated beads, biotinylated antibodies, and 96-well filter plates were obtained from Upstate Group, Inc. (Lake Placid, NY). Each multiplex assay was performed in duplicate following the manufacturer's specifications. Data were collected using the Luminex-100 system, version IS 2.2, and analyzed by MasterPlex quantitation software (MiraiBio, Inc., Alamada, CA). Standard curves for each cytokine and chemokine were obtained using the reference cytokine and chemokine concentrations supplied by the manufacturer.
Treatment of mice with antibodies or recombinant proteins.The importance of IL-10 for mouse survival of S. suis infection was determined using two separate approaches. In the first approach, neutralization of IL-10 in B6 mice was performed via a single i.p. injection of 1 mg/ml of rat anti-murine IL-10 receptor (IL-10R) monoclonal antibody (MAb) 1B1.3a. The anti-IL-10R MAb was administrated into 10 B6 mice 1 h before infection with S. suis. The control group, which consisted of six B6 mice, received 1 ml of pyrogen-free saline containing an equivalent quantity of normal rat serum immunoglobulin G (Jackson ImmunoResearch, West Grove, PA). In the second approach, the role of IL-10 in modulating the survival of A/J mice against S. suis infection was tested by i.p. administration of recombinant mouse IL-10 (rmIL-10) (R&D Systems, Minneapolis, MN) to 10 A/J mice. At different times relative to the infection, each mouse received a cumulative dose of 2.6 μg rmIL-10 diluted in 300 μl of pyrogen-free saline. The control group of 10 A/J mice received vehicle alone. Injection doses of rmIL-10 or MAb 1B1.3a were selected based on previous publications (5, 7, 8, 21, 25).
Statistical analyses.Overall survivals for the different treatment groups were described using Kaplan-Meier plots. Survival curves were compared using the log-rank test. Values for bacterial loads in different tissues as well as serum cytokine levels between A/J and B6 mice were compared using the Mann-Whitney test. A P value of <0.05 was considered statistically significant. All analyses were performed using the Sigma plot system (v.9; Systat Software, San Jose, CA).
RESULTS
B6 and A/J mouse strains show important differences in their clinical susceptibilities to S. suis infection.The inbred mouse strains A/J and B6 showed significant differences in clinical behavior and mortality after i.p. infection with S. suis. In A/J mice, clinical signs related to infection were observed at 2 h p.i., which quickly worsened and resulted in a mortality rate of 84% at 12 h p.i. and then 100% at 20 h p.i. (Fig. 1). Based on the clinical profile and progression, all deaths were considered related to septicemia. This was confirmed by bacteriological isolation of S. suis from different tissues (see below). S. suis infection in B6 mice had a slower and different pattern of progression, with subtle clinical signs of acute disease appearing at 8 h p.i. B6 mice showed high survival compared with A/J mice; at 24 h p.i., only 16% of B6 mice died due to septicemia (Fig. 1). Interestingly, from day 3 p.i. onward, 44% of surviving B6 mice developed sudden neurological signs, such as recumbence, difficult mobility, and rear limb weakness. Furthermore, 57% of B6 mice which showed neurological signs also presented unexpected clinical signs suggestive of cardiac distress. S. suis was isolated from different organs of all mice that died from the infection. Weight loss was another clinical characteristic observed for B6 mice infected with S. suis. At day 3 p.i., 69% of surviving B6 mice had suffered important weight loss of between 21 and 40% of their initial body weight. However, at day 15 p.i., the remainder of the surviving mice had recovered their initial body weight.
Survival curves of A/J and B6 mice after S. suis infection. Mice were injected i.p. with S. suis serotype 2 (1 × 107 CFU/ml) and mortality was recorded daily for 15 days. S. suis-infected A/J mice showed significantly high mortality during the first 20 h p.i. compared with B6 mice (P < 0.001).
No histopathological lesions were observed in the different tissues isolated from A/J and B6 mice that died during the first 48 h p.i. However, in B6 mice which died at day 3 p.i. or later, histopathological studies confirmed that the neurological signs were consistent with encephalitis and meningitis, while clinical signs of cardiac distress suggested endocarditis. The main histopathological findings from the brains of mice with clinical signs of neurological disease were the presence of foci of malacia, hemorrhages, and gliosis (Fig. 2A). S. suis was observed as bacterial emboli contained inside blood vessels with thickened walls due to neutrophil infiltration (Fig. 2B). A severe infiltration of the meninges, mainly with macrophages and lymphocytes and also with fibrin, was also observed (Fig. 2C). Endocarditis was diagnosed on the basis of partial occlusion of the atrioventricular right valves by thrombi formed by fibrin, a mix of inflammatory cells, cellular debris, and bacteria (Fig. 2D).
Relevant histopathological findings for B6 mice presenting clinical signs of neurological disease or sudden death (day 9 p.i.) after i.p. infection with S. suis. Hematoxylin-eosin staining of brain and heart tissue samples was performed. (A) Micrograph of the brain parenchyma showing an area of malacia and hemorrhages (black arrows). Magnification, ×100. (B) Micrograph of the brain cortex, in which is evident the presence of a bacterial embolus (*) surrounded by neutrophils. Magnification, ×400. (C) Micrograph of the meninges, which are severely and diffusely infiltrated by macrophages and other nonsuppurative inflammatory cells (black arrows). Magnification, ×100. (D) Micrograph of the heart at the atrioventricular valves with a thrombus occluding most of the valvular lumen (black arrows). The thrombus comprises fibrin, inflammatory cells, and bacteria. Magnification, ×200.
S. suis is present in different tissues.i.p. infection with S. suis in both A/J and B6 induced sustained bacteremia, as demonstrated by the recovery of bacteria from infected mice at different p.i. time points. A/J mice had high and sustained bacterial loads in the blood, livers, and spleens, with more than 1 × 108 CFU per ml or organ at 3 to 12 h p.i. (Fig. 3A to C). Interestingly, the number of S. suis bacteria recovered from the brain was low compared with those from other organs, although it increased with time from 5 × 104 CFU/organ at 3 h p.i. to 8 × 105 CFU/organ at 12 h p.i. (Fig. 3D). No data could be obtained at later times for this mouse strain because 100% of A/J mice died within 20 h p.i. Similar to A/J mice, B6 mice showed bacterial loads in the blood of above 1 × 108 CFU/ml and these loads remained constant up to 48 h p.i. Bacterial loads remained elevated at 72 h p.i. (∼1 × 105 CFU/ml), demonstrating the ability of virulent S. suis to survive successfully in the bloodstream (Fig. 3A). Likewise, bacterial loads were high in the livers and spleens of infected B6 mice, with more than 1 × 107 CFU/organ during the first 36 h p.i. (Fig. 3B and C). Similar to what was seen for A/J mice, S. suis was found in lower numbers in the brain; however, in these mice, the presence of the pathogen was sustained throughout the experiment, with 1 × 104 to ∼1 × 105 CFU/organ at 72 h p.i. (Fig. 3D). There were no statistical differences in site-specific bacterial loads between the A/J and B6 mouse strains at any of the time points tested throughout the experiment (P > 0.05).
Bacterial loads in different organs from A/J and B6 mice infected i.p. with S. suis. Bacterial loads are expressed as CFU/ml for blood (A) and as CFU/0.05 g of tissue for the livers (B), spleens (C), and brains (D). Results are expressed as mean (± standard error of the mean [SEM]) results for at least three infected mice per p.i. time point. No significant differences were found between the two mouse strains throughout the experiment (P > 0.05).
Profiles of the inflammatory response induced by S. suis differ remarkably between A/J and B6 mice.There were important differences in the kinetics of cytokine and chemokine production between the two mouse strains in response to infection with S. suis. The inflammatory response of A/J mice could be determined only during the first 12 h p.i. given that, as discussed earlier, all A/J mice died within 20 h after S. suis infection. Indeed, compared with what was seen for B6 mice, S. suis infection of A/J mice induced a rapid systemic production of diverse proinflammatory cytokines. There were notable differences in the kinetics of cytokine production, namely, those for TNF-α, IL-12p40/p70, IFN-γ, and IL-1β, between the two mouse strains. A significant difference in TNF-α secretion between the mouse strains was found at 6 h p.i. (P = 0.018). In A/J mice, TNF-α, one of the most important mediators of septic shock, increased rapidly, peaked at 6 h, and was still elevated at 12 h p.i. (Fig. 4A). In B/6 mice, TNF-α increased more slowly, peaking at 12 h p.i. and then declining to basal levels at 24 h p.i. (Fig. 4A). This rapid and important upregulation of TNF-α in plasma of S. suis-infected A/J mice may be related to the early development of clinical signs resembling septic shock in these mice and the associated death. A/J mice also presented a rapid and high increase in IL-12p40 production compared to the modest induction of IL-12p40 observed for B6 mice (P value of 0.024 at 6 h p.i.) (Fig. 4B). Levels of bioactive IL-12p70, a key cytokine bridging innate and adaptive immunities, and IFN-γ, a critical marker of sepsis, were significantly higher in A/J mice than in B6 mice (Fig. 4C and D). Production of IL-1β was lower than that of other cytokines, although a significant and sustained expression was observed for A/J compared with B6 mice at 6 h and 12 h p.i. (P values of 0.002 and 0.023, respectively) (Fig. 4E). Interestingly, the expression of IL-6, another classical indicator of sepsis, was elevated and sustained in both mouse strains, without any significant difference (P > 0.05) (Fig. 4F).
Kinetics of the expression of inflammatory mediators in A/J and B6 mice infected i.p. with S. suis. (A) TNF-α. (B) IL-12p40. (C) IL-12p70. (D) IFN-γ. (E) IL-1β. (F) IL-6. Cytokine concentrations in sera were assayed by a liquid multiarray system (Luminex), as explained in Materials and Methods. Data are expressed as mean ± SEM (pg/ml). Values for uninfected controls did not show statistically significant changes from 3 h to 72 h. As such, time zero h represents the mean (± SEM) results for noninfected mice throughout the experiment. Asterisks represent significant differences between mouse strains (P < 0.05).
Concerning the chemokines, the expression of KC, an important neutrophil chemoattractant, was rapidly induced similarly in both mouse strains, appearing as early as 3 h p.i. (P > 0.05) (Fig. 5A). Very little or no KC expression was detected for A/J mice after 12 h p.i, whereas in B6 mice S. suis-induced KC production started to decrease only at 36 h p.i. (Fig. 5A). High levels of MCP-1, a molecule with monocyte chemoattractant properties, were observed in both infected mouse strains, and no significant differences were observed (P > 0.05) (Fig. 5B). Similar to KC and MCP-1 expression, the kinetics of RANTES expression were comparable between the two mouse strains, although the levels detected were relatively modest (P > 0.05).
Kinetics of serum chemokine expression in A/J and B6 mice infected i.p. with S. suis. (A) KC. (B) MCP-1. (C) RANTES. Chemokine concentration was assayed by a liquid multiarray system (Luminex), as explained in Materials and Methods. Data are expressed as mean ± SEM pg/ml. Time zero h represents the mean (± SEM) results for noninfected mice throughout the experiment.
IL-10 plays an important role in controlling the severity of S. suis-induced septic infection.As described above, A/J mice showed an exacerbated inflammatory systemic response to S. suis infection and, to some extent, were also unable to mount an appropriate anti-inflammatory response. As shown in Fig. 6A, A/J mice secreted relatively low amounts of IL-10, with a highest mean level of 620 pg/ml detected at 12 h p.i. In contrast, B6 mice were able to secrete significantly higher and more sustained levels of IL-10 from 12 h to 36 h p.i., with a mean production peak of 2,610 pg/ml at 12 h p.i. (P value of 0.014 versus that for A/J mice) (Fig. 6A). We used two different approaches to elucidate the potential role of IL-10 in mediating host protection against lethal S. suis infection. First, systemic IL-10R was blocked in B6 mice to assess whether interference with the IL-10 pathway would increase susceptibility to infection and exacerbate mortality. During the first 12 h p.i., B6 mice treated with an anti-IL-10R MAb and those treated with a vehicle control both showed similar clinical signs, such as rough hair coat, depression, and recumbence. However, the severity of disease increased with time only in the group that received the anti-IL-10R MAb. Indeed, during the course of the septic phase of disease, B6 mice treated with the anti-IL-10 MAb had significantly higher mortality than that seen for control mice (P = 0.012). Of the B6 mice treated with anti-IL-10 MAb, 50% died at 20 h p.i. and 90% at 24 h p.i. (Fig. 6B). In contrast, only 17% of control mice died at 30 h p.i. and 50% at 72 h p.i. (Fig. 6B). Overall, the survival curves during the first 24 h p.i. for A/J mice (Fig. 1) and B6 mice treated with anti-IL-10R MAb were comparable (Fig. 6B). Confirming the protective role of IL-10, Fig. 7 shows that anti-IL-10R MAb-treated B6 mice produced significantly high levels of TNF-α 24 h p.i. compared to nontreated B6 mice. Interestingly, levels of TNF-α in treated B6 mice were similar to those observed for A/J mice (Fig. 4A).
Role of IL-10 in the survival of A/J and B6 mice after S. suis infection. (A) Kinetics of serum IL-10 production in A/J and B6 mice infected i.p. with S. suis. IL-10 concentration was assayed by a liquid multiarray system (Luminex), as explained in Materials and Methods. Data are expressed as mean ± SEM (pg/ml). Time zero h represents the mean (± SEM) results for noninfected mice throughout the experiment. The asterisk represents a significant difference between mouse strains (P < 0.05). (B) Survival of B/6 mice treated with anti-IL-10R MAb (1 mg/ml) prior to challenge with S. suis. Blockage of IL-10R impedes the protective role of IL-10 and thus renders animals significantly more susceptible to S. suis infection. This is evidenced by a course of mortality that is more rapid in treated B6 mice than that in nontreated controls (P = 0.012). (C) Survival of A/J mice treated with rmIL-10 prior to challenge with S. suis. A/J mice were treated with rmIL-10 at different final concentrations and times relative to the infection. It is notable that rmIL-10 treatment has a protective effect, significantly delaying mortality compared with what was observed for the nontreated group (P < 0.001).
Effect of blockage of IL-10R on cytokine production. B6 mice were treated with anti-IL-10R MAb (1 mg/ml) before i.p. infection with S. suis serotype 2 (1 × 107 CFU/ml). Serum samples were taken at 24 h p.i., and TNF-α levels measured by an enzyme-linked immunosorbent assay kit (R & D Systems). The asterisk indicates a significant difference between treated and nontreated groups (P < 0.001). Symbols represent values from individual mice, while the horizontal lines indicate the median for each group.
In the second approach, we investigated whether the administration of rmIL-10 to A/J mice would enhance their survival after S. suis infection. To this end, A/J mice received 1 μg of rmIL-10 1 h before and 3 h after i.p. infection with S. suis and a final dose of 0.6 μg at 24 h p.i. We observed striking differences in mortality between A/J mice treated with rmIL-10 and nontreated controls; at 20 h p.i., only 10% of the treated mice died compared with 80% of the nontreated controls (P < 0.001). At 36 h p.i., 100% of the nontreated controls succumbed to their infection (Fig. 6C), as reported above (Fig. 1), compared with 40% of the treated mice. Although the administration of rmIL-10 significantly prolonged the survival of infected A/J mice, it did not confer complete protection against S. suis, as 70% of rmIL-10-treated A/J mice died at 48 h p.i.
DISCUSSION
S. suis is one of the most important causative agents of septicemia, septic shock, and meningitis in pigs as well as in humans (20, 33). Most past research on the activation of the innate immune response by S. suis has been conducted using in vitro models of infection. However, animal models are essential to achieve a better understanding of the activation and development of host immune responses to this important pathogen. Although mouse models for some diseases do not reflect the disease outcome in the natural host, we have recently shown for the first time that an adult mouse model for S. suis administered using the i.p. route of infection in CD1 mice clearly reproduces clinical signs and lesions observed for swine (11). Kataoka et al. also showed that clinical signs, disease outcomes, and lesions in mouse models were similar to those seen for experimentally infected pigs when the same virulent and avirulent bacterial strains were used. In that study, authors observed differences in susceptibility among several mouse strains and suggested that differences in genetic background or immune response might explain these discrepancies. However, no further analyses were carried out (23). In this study, we compared the susceptibilities of two genetically different inbred mouse strains, A/J and B6, to S. suis infection. These mouse strains have been used frequently to investigate the genetic control of host immunity to infectious pathogens. Previous studies using these mouse strains have focused mainly on sepsis induced by endotoxin (lipopolysaccharide [LPS]) challenge, cecal ligation and puncture, and systemic candidiasis (28, 30, 34, 42, 44, 45). Study of the genetic factors underlying susceptibility to systemic streptococcal infection has been limited mainly to Streptococcus pneumoniae and group A streptococci (GAS; Streptococcus pyogenes) (13, 14, 22, 24, 28), and most of these studies evaluated only mortality and bacterial loads in different tissues. To our knowledge, no studies describing host susceptibility and inflammatory response to S. suis infection have been published to date.
In a previous study from our laboratory, S. suis showed intermediate virulence in outbred CD1 mice (11). Based on clinical behavior and mortality, the present study clearly demonstrated that A/J mice are significantly more susceptible to S. suis infection than are B6 mice, especially during the acute septic phase of infection. Indeed, the susceptibility of B6 mice to S. suis infection shown in this study is comparable to that previously observed for CD1 mice. These results are in agreement with studies of Candida albicans or GAS infection (14, 30), to which A/J mice are highly susceptible while B6 mice show intermediate susceptibility. Nevertheless, overall mortalities of B6 and A/J mice to GAS infection are similar, although mortality is delayed for 48 h in B6 mice (28). Interestingly, these results differ from studies of LPS and cecal ligation and single puncture, in which A/J mice were considerably more tolerant than B6 mice (42). B6 mice show either high or intermediate susceptibility to infection with S pneumoniae, depending on the serotype used (29). Therefore, the genetic susceptibility of the mouse strain may vary depending on the pathogen encountered or injury received.
Consistent with previous studies, mortality in both A/J and B6 mice during the first 48 h after S. suis infection can be attributed to a septic shock-like condition in which there is uncontrolled bacteremia and/or an exacerbated production of proinflammatory cytokines and chemokines (30, 45). In the present study, the numbers of CFU in the blood, liver, spleen, and brain were not significantly different between the mouse strains, arguing against the possibility that the higher mortality in S. suis-infected A/J mice was due to uncontrolled bacteremia. Compared with B6 mice, A/J mice are known to carry a loss-of-function mutation in the structural gene of the C5 component of the complement pathway (32, 35). In a model of C. albicans infection, there was no conclusive evidence that this deficiency is responsible for the higher fungal replication in the kidneys of A/J versus B6 mice (30, 45), and direct fungicidal activity by complement could not be demonstrated (30). Similarly, complement alone, in the absence of specific antibodies, does not contribute to phagocytosis and killing of S. suis by monocytes and neutrophils (6). If the complement were essential for bactericidal activity against S. suis, the bacterial load would have been expected to be higher in A/J than in B6 mice, whereas this study demonstrated no significant differences in the bacterial loads between the two strains.
Although a deficiency of complement might not modulate bacterial clearance, some studies suggest that it may play a role in the dysregulated cytokine response observed for A/J mice (30). Accordingly, the high mortality of A/J mice after S. suis infection might be caused by an uncontrolled release of proinflammatory cytokines. This hypothesis is supported by the lack of specific histopathological findings for A/J (as well as B6) mice which had succumbed to infection within the first 48 h p.i. Different cytokines released early after infection are thought to play opposing roles. Proinflammatory cytokines are necessary to control infection, although excessive or uncontrolled responses may disturb homeostasis and lead to increased disease severity and mortality (47). In the present study, we observed that A/J mice had significantly high production of TNF-α, IL-12p40, IL-12p70, IFN-γ, and IL-1β compared with B6 mice, particularly during the first 6 to 12 h p.i. Production levels of TNF-α, one of the most important acute-phase cytokines and a reliable indicator of septic shock in both gram-positive and -negative bacterial infections (9, 43), were comparable between B6 (this study) and CD1 (11) mice. On the other hand, A/J mice produced twice as much TNF-α as B6 or CD1 mice did. Likewise, A/J mice produced levels of IL-12p40, IL-12p70, and IFN-γ that were 3- to 19-fold higher than those seen for B6 (this study) or CD1 (11) mice. As previously observed for CD1 mice, the release of IL-12p40 preceded that of IL-12p70 following S. suis infection (11). In A/J mice, increased IL-12 expression may have been associated with the induction of high levels of IFN-γ, a cytokine known for its potent ability to activate macrophages and enhance TNF-α synthesis, thereby exacerbating mortality as previously reported (18, 19). A/J mice had relatively low levels of IL-1β, as previously reported for CD1 mice (11), but these levels were consistently higher than those for B6 mice, and they might be sufficient to exert a biological effect. Both A/J and B6 mice produced high levels of IL-6, but these levels were lower than those produced by CD1 mice (11). Interestingly, there were no significant differences in any of the three chemokines measured, namely, KC, MCP-1, and RANTES, between both mouse strains at all time points. The levels of chemokines observed here were nearly identical to those reported for CD1 mice (11). Since both A/J and B6 mice infected with S. suis had secreted similar levels of these chemokines, the difference in mortality observed between the two strains makes these chemokines to be unlikely candidates for contributing to the difference in the development of septic shock-like syndrome after S. suis infection between the two mouse strains.
In previous studies of sepsis in mice, a massive secretion of inflammatory mediators was postulated as a likely cause of severe disease and mortality (37). Data available for S. suis also suggest that this may be the case in both swine and humans. We have demonstrated that S. suis stimulated swine cells in a whole-cell blood system to induce high levels of proinflammatory cytokines (41). In addition, we have recently shown that pigs experimentally infected with virulent S. suis show high levels of IL-6 and IL-8 (G. Vanier, M.-P. Lecours, M. Segura, D. Grenier, and M. Gottschalk, presented at the First Symposium of the Centre de recherché en infectiologie porcine, Saint-Hyacinthe, QC, Canada, 28 and 29 May 2007). In humans, S. suis is able to induce the release of proinflammatory cytokines from brain microvascular endothelial cells (46), and cytokine levels in humans with streptococcal toxic shock-like syndrome due to S. suis infection were extremely high and shown to be identical to those observed for B6 mice experimentally infected with the strain that caused the Chinese outbreak (C. Ye, H. Zheng, J. Zhang, H. Jing, L. Wang, Y. Xiong, W. Wang, Q. Sun, X. Luo, H. Du, M. Gottschalk, and J. Xu, submitted for publication). Thus, levels of cytokines are a prognosis of human disease severity too.
The differential survival of genetically different mouse strains after infection with invading pathogens has been linked to the ability to control this exacerbated proinflammatory cytokine production. As such, Råberg et al. defined “resistance” by the manner in which a host can limit pathogen burden, whereas “tolerance” is defined by the manner in which the host can limit the damage done by a given pathogen burden (36). The data presented here indicate that the two mice strains have similar loads throughout the first 12 h, after which all A/J mice succumb to infection. This indicates that at least over the first 12 h, while both mice strains experience the same pathogen load, B6 mice are able to limit the induced damage and thus seem more tolerant. Indeed, the higher survival of B6 mice during the septic phase might be related to their ability to mount a competent anti-inflammatory response, as demonstrated by the high and sustained production of IL-10 from 12 h to 36 p.i. Similar results for CD1 mice were previously reported (11). Other studies have also demonstrated the importance of IL-10 for controlling cytokine production (10, 12) and for mouse survival of septic shock (48). Indeed, high mortality in IL-10 gene knockout mice is associated with elevated circulating TNF-α levels, and more importantly, early treatment with rIL-10 delays the onset of irreversible shock (27). In the present study, we demonstrated that early systemic blocking of IL-10R in B6 mice is detrimental for host survival to the septic phase of S. suis infection, as anti-IL-10R MAb treatment before S. suis infection resulted in 90% mortality in B6 mice as early as 24 h p.i. with a concomitant and dramatic increase in TNF-α levels. S. suis-infected A/J mice, on the other hand, showed low and nearly undetectable IL-10 production. We hypothesized that the low IL-10 production might be partially responsible for the elevated levels of proinflammatory cytokines and the septic shock observed for these mice. In support of this concept, A/J mice treated with rmIL-10 showed 90% survival compared with 20% survival in nontreated mice.
Together, these findings suggest that, in S. suis-infected mice, IL-10 plays an important role in the early control of sepsis-related inflammatory responses and consequently can improve host survival. Similar protection conferred by rIL-10 treatment has been reported for mice after challenge with CPL, LPS, or GBS infection (8, 21, 25). Interestingly, in agreement with these studies, rmIL-10 treatment delayed but did not abolish mortality due to S. suis in A/J mice, indicating that other immunoregulatory factors are also involved in controlling the inflammatory response and influencing long-term survival.
Although B6 mice were more tolerant than A/J mice to the acute phase of infection, surviving B6 mice remained susceptible to developing S. suis-induced meningitis and endocarditis. These pathologies are similar to those described for S. suis-infected swine, humans, and CD1 mice (2, 11, 20). The clinical neurological signs in S. suis-infected B6 mice were not as marked as those observed in CD1 mice (11). Nevertheless, our histopathological analysis confirmed neurological disease in the infected B6 mice. These findings raise the possibility that B6 mice are a potentially useful model for future studies of the pathogenesis of S. suis infection.
Our findings show a clear susceptibility of A/J mice to S. suis infection and provide evidence that the balance between pro- and anti-inflammatory mediators is crucial for survival during the septic phase. The different responses elicited in A/J and B6 mice following systemic infection with S. suis warrant future studies using a genome-wide scan with F2 populations derived from both mouse strains.
ACKNOWLEDGMENTS
IL-10R (1B1.3a) MAb was kindly provided by M. Stevenson, Centre for the Study of Host Resistance, Research Institute of the McGill University Heath Centre, Montreal, QC, Canada. We are grateful to Claude Lachance (Research Institute of the McGill University Health Centre, Montréal, QC, Canada) and Sonia Lacouture (Faculté de médecine vétérinaire, Université de Montréal, St-Hyacinthe, QC, Canada) for technical support.
This work was supported by Natural Sciences and Engineering Research Council of Canada (NSERC) grant no. 0680154280 and Centre de recherche en infectiologie porcine (Fonds québécois de la recherche sur la nature et les technologies). M.C.D.-P. is the recipient of an NSERC postgraduate scholarship.
FOOTNOTES
- Received 17 March 2008.
- Returned for modification 13 May 2008.
- Accepted 13 June 2008.
- Copyright © 2008 American Society for Microbiology
