Deregulated Balance of Omega-6 and Omega-3 Polyunsaturated Fatty Acids following Infection by the Zoonotic Pathogen Streptococcus suis

Abstract

Streptococcus suis is an important swine pathogen and an emergent zoonotic pathogen. Excessive inflammation caused by S. suis is responsible for early high mortality in septic shock-like syndrome cases. Polyunsaturated fatty acids (PUFAs) may contribute to regulating inflammatory processes. This study shows that mouse infection by S. suis is accompanied by an increase of arachidonic acid, a proinflammatory omega-6 (ω-6) PUFA, and by a decrease of docosahexaenoic acid, an anti-inflammatory ω-3 PUFA. Macrophages infected with S. suis showed activation of mitogen-activated protein kinase pathways and cyclooxygenase-2 upregulation. Fenretinide, a synthetic vitamin A analog, reduced in vitro expression of inflammatory mediators. Pretreatment of mice with fenretinide significantly improved their survival by reducing systemic proinflammatory cytokines during the acute phase of an S. suis infection. These findings indicate a beneficial effect of fenretinide in diminishing the expression of inflammation and improving survival during an acute infection by a virulent S. suis strain.

INTRODUCTION

Streptococcus suis is an important swine pathogen responsible for a wide range of clinical disease, such as sudden death, meningitis, endocarditis, and arthritis (1). S. suis is also an emergent zoonotic pathogen where humans become infected mostly by handling sick pigs and pork carcasses or by consumption of uncooked raw products, as seen in some countries in Asia (2, 3). Among the 35 described serotypes, type 2 is most frequently associated with disease in both swine and humans (4). In addition to meningitis, infection in humans may result in cases of septic shock with high fever, hypotension, endocarditis, pneumonia, arthritis, and multiple organ failure, leading rapidly to death (5). An excessive inflammatory immune response has been suggested to be responsible for most of these symptoms. In fact, infections by highly pathogenic S. suis results in an exacerbated release of proinflammatory cytokines and chemokines, which would be the main cause of death (6, 7).

Polyunsaturated fatty acids (PUFAs) may have an important role in the innate immune response. PUFAs appear to be important regulators of the onset and resolution of the inflammatory response by functioning like signaling molecules to regulate inflammation (8). Arachidonic acid (AA), an omega-6 (ω-6) PUFA, is released from membrane phospholipids by phospholipases, such as cytosolic phospholipase A₂ (cPLA₂), in response to several activation stimuli, including proinflammatory mediators and other stress signals (9). It is subsequently metabolized by cyclooxygenases (COXs) and lipoxygenases (LOXs) to generate prostaglandins, thromboxanes, and leukotrienes, which are potent proinflammatory mediators (10). However, docosahexaenoic acid (DHA), an ω-3 PUFA, possesses anti-inflammatory properties and plays an important role in the resolution of inflammation. DHA can compete with AA for the same enzymes in the lipid metabolic pathway and thus can reduce AA metabolism (11). Therefore, the ratio between AA and DHA may determine whether the host can mount an effective and appropriate immune response to combat infection. A higher-than-normal AA/DHA ratio seems to be closely linked with inadequate control of the inflammatory process and may result in often irreversible, pathological effects associated with inflammation (12–15).

Fenretinide, also called N-(4-hydroxyphenyl) retinamide, is a synthetic analog of all-trans retinoic acid (16). It has cytotoxic effects against cancer cells and has been used in different clinical trials (17). Recent studies showed that fenretinide is emerging as a promising anti-inflammatory agent when used at low doses. In fact, fenretinide is able to normalize aberrant ratio between AA and DHA, allowing normalization of the PUFA imbalance observed in various conditions associated with chronic inflammation or injury (12, 18–20).

The status of PUFAs following an acute infection by S. suis is currently unknown. In order to better understand the impact S. suis infection has on PUFAs and their downstream metabolites, we assessed levels of AA and DHA in S. suis acutely infected mice. The anti-inflammatory effect of fenretinide, due to its well-proven corrective effect on AA/DHA imbalance, has never been studied during an acute bacterial infection. Thus, we verified whether this compound could help diminishing the excessive proinflammatory response after S. suis infection in vitro and in vivo and thus leading to better survival. Our results showed, for the first time, that an S. suis infection results in a shift toward an induction of proinflammatory lipid mediators and that the treatment with fenretinide can alleviate the very first symptoms of septic shock following a S. suis infection by reducing the excessive proinflammatory response.

MATERIALS AND METHODS

Bacterial strain and growth conditions.

The well-characterized and previously described virulent S. suis serotype 2 strain 31533 (21), isolated from a diseased pig afflicted with meningitis, was used in the present study. The general growth conditions have been previously described (22).

Mice and experimental infections.

CD1 (n = 27) and C57BL/6 (B6; n = 26) 6- to 8-week-old mice (Jackson Laboratory) were acclimatized to standard laboratory conditions as described previously (23). Both models have been previously used with S. suis (7, 22). Experiments involving mice were performed in accordance with the guidelines and policies set forth by the University of Montreal Animal Welfare Committee (permit RECH-1570). One milliliter of the bacterial suspension (5 × 10⁷ CFU/ml) or the vehicle solution (sterile Todd-Hewitt broth; Difco Laboratories) was administered by intraperitoneal injection, as previously described (23). Mice were euthanized at 24 h postinfection (p.i.) and, the spleens, livers, and blood were harvested for the PUFA analysis described below. This p.i. time point was chosen based on preliminary results of infection kinetics (data not shown).

Polyunsaturated fatty acid analysis.

Blood samples were mixed with 30 μl of EDTA 200 mM and centrifuged at 350 × g for 10 min at 4°C. The plasma was then removed and mixed with chloroform-methanol (2:1 [vol/vol]) containing 1 mM butylated hydroxyanisole (BHA; Sigma-Aldrich) for lipid extraction, as previously described (24). A total of 200 mg of liver and spleen tissue was also collected from euthanized mice and mixed with the chloroform-methanol-BHA solution. The levels of DHA and AA in the spleens, livers, and plasma from infected and control mice were assessed by gas chromatography as previously described (18, 25).

Cell culture and bacterial infection.

The ANA-1 macrophage cell line was used, which was derived from the bone marrow of B6 mice (26). Cells were grown in Dulbecco modified Eagle medium (DMEM) supplemented with 10% heat-inactivated fetal bovine serum (Invitrogen) and incubated at 37°C with 5% CO₂. Cells were plated in 24-well plates at a concentration of 10⁵ cells/ml for 2 h. Medium was removed and replaced with fresh medium or fresh medium containing fenretinide where appropriate (graciously provided by Robert Smith, National Institutes of Health, Bethesda, MD) at a final concentration of 1.25 μM (fenretinide pretreatment). This concentration was chosen based on results obtained in preliminary tests with different fenretinide concentrations (data not shown). Cells were incubated for 24 h, and medium was then removed again and replaced with identical medium before the infection. Thereafter, cells were infected with S. suis strain 31533 for 16 h (multiplicity of infection [MOI] of 10:1). Specific samples received fenretinide and a similar dose of S. suis simultaneously. Cell supernatants were then harvested for measuring the cytotoxicity and TNF concentrations (see below). Cells were then lysed as previously described (27), and COX-2 expression was measured as described below.

In assays evaluating intracellular signaling pathways involving mitogen-activated protein kinases (MAPKs), macrophages were infected with 10⁷ CFU of S. suis strain 31533/ml for 30 min, as previously described (27). Purified Escherichia coli lipopolysaccharide (LPS; 100 ng/ml; Invivogen) was used as a positive control. The extracellular signal-regulated kinase 1/2 (ERK1/2) inhibitor UO126 (Calbiochem) was added at a concentration of 10 μM (28) to appropriate samples 1 h before the bacterial infections. All experiments included untreated and/or mock-infected negative controls.

Cell viability, ELISA, and Western blotting.

Cell toxicity induced by S. suis and/or fenretinide was measured using a Cytotox 96 nonradioactive cytotoxicity assay kit (Promega) according to the manufacturer's protocol. For quantifying TNF production from cell supernatants, a sandwich enzyme-linked immunosorbent assay (ELISA; R&D) was performed as previously described (29). Western blots for determining activities of specific intracellular proteins during S. suis infection of macrophages were done as described previously (27). Rabbit anti-mouse cyclooxygenase-2 (COX-2; 1:1,000), anti-mouse phospho-ERK1/2 (1:1,000), anti-mouse ERK1/2 (1:1,000), and anti-mouse phospho-p38 (1:1,000) antibodies (all from Cell Signaling Technology) were used to probe their substrates overnight at 4°C. Horseradish peroxidase-conjugated goat anti-rabbit secondary antibody (1:2,000; Cell Signaling Technology) solution was used to blot the target proteins.

Protective role of fenretinide in vivo.

Fenretinide powder was dissolved in 95% ethanol at a concentration of 20 mg/ml and diluted 1:20 with a liquid diet of Peptamen (Nestlé). The Peptamen solution containing fenretinide was kept at 4°C and protected from light before administration to mice. Since there were no differences between the two mouse strains tested concerning the deregulation of PUFAs by S. suis infection, and to reduce the use of animals, the protective role of fenretinide in vivo was carried out with the inbred B6 mouse strain to reduce variability. Based on preliminary data, mice (n = 15/group) were treated with fenretinide at 10 mg/kg 24 h prior to infection, 2 h prior to infection, and 24 h after i.p injection of S. suis strain 31533 (10⁷ CFU/mouse). This fenretinide treatment was shown to maintain blood concentration of fenretinide to ∼1 μM (unpublished results). The drug was delivered by gavage to ensure accurate dosing at specific time points. Control infected mice received an equivalent amount of Peptamen supplemented with an equivalent amount of vehicle (95% ethanol) by gavage. Uninfected control group received similar fenretinide treatment. Preliminary studies showed that this group of mice behaved like nontreated control mice, including total absence of clinical signs (data not shown). Mice had free access to water and rodent chow before and after the gavage. Clinical signs of sepsis were monitored during the survival study (up to 96 h p.i.). These signs were evaluated and scored appropriately according to clinical signs listed in Table 1. If mice reached a combined score of 7, they were sacrificed according to guidelines and policies set forth by University of Montreal Animal Welfare Committee. Blood bacteremia was monitored in surviving mice by collecting a 5-μl blood sample from the tail as described previously (7, 22). Additional mice (n = 7) were included in each group and were euthanized at 12 h p.i. At this time point, blood samples were used to evaluate the concentration of the following cytokines and chemokines (using a Luminex platform [Millipore]): interleukin-12p70 (IL-12p70), IL-6, tumor necrosis factor (TNF), CCL2, CCL3, CCL4, CCL5, CXCL1, CXCL2, and CXCL10.

TABLE 1.

Clinical sign scores of mice infected with S. suis strain 31533

Clinical sign	Score
Normal behavior	0
Ruffled fur	1
Eye swelling/closed	1
Macular edema	1
Prostration	1
Depression	1
Transient respiratory difficulty	1
Lethargy	7
Nervous signsa	7
Frantic hopping	7
Pedaling	7
Sudden excitation, rapidly followed by death	7
Wt loss >15% of initial wt	7
Death	10

^aHyperexcitation, episthotonus, opisthotonus, bending of the head toward one side, walking in circles, strong locomotive problems, including paresis of the forelimbs; unilateral or bilateral exophthalmia, with corneal opacity, probably in association with increased intracranial pressure.

Statistical analysis.

Data are presented as the means ± the standard errors of the mean (SEM). AN unpaired Student t test, a Mann-Whitney test, or one-way analysis of variance (ANOVA) where appropriate were performed to measure statistical differences between groups. The Kaplan-Meier method and log-rank Mantel-Cox tests were used to compare the survival rates of the studied groups. Differences were considered statistically significant at P ≤ 0.05.

RESULTS

S. suis infection disrupts PUFAs toward a proinflammatory profile.

It has been reported that cystic fibrosis, as well as chronic inflammation caused by Pseudomonas aeruginosa, leads to the elevation of AA levels, while the DHA levels become downregulated, resulting in a dysregulated proinflammatory profile (12, 18, 20). Compared to mock-infected mice, proinflammatory AA levels were significantly higher in the plasma of both B6 and CD1 mice infected with S. suis (Fig. 1A). Conversely, levels of DHA, an ω-3 anti-inflammatory PUFA, were significantly downregulated in S. suis-infected mice (Fig. 1A). Since both AA and DHA levels were altered in response to infection, the ratios of AA/DHA were significantly distorted during S. suis infection in both B6 and CD1 mice compared to mock-infected mice (Fig. 1A). This dramatic increase of the AA/DHA ratios indicates that the infection triggers a strong proinflammatory lipid mediator status in these animals, resulting from a simultaneous depletion of anti-inflammatory DHA and a significant increase in proinflammatory AA. The aberrant AA/DHA ratios could be observed as early as 9 h p.i. in both mouse strains (data not shown).

FIG 1.

Arachidonic (AA) and docosahexaenoic (DHA) polyunsaturated fatty acids (PUFAs) are modulated toward a proinflammatory state in B6 (n = 26) and CD1 (n = 27) mice infected with a virulent S. suis strain. Mice were infected by intraperitoneal injection with 5 × 10⁷ CFU of S. suis strain 31533 and euthanized at 24 h p.i. Corresponding mock-infected animals were used as controls. Plasma (A), spleens (B), and livers (C) were harvested, mixed into a chloroform-methanol solution containing butylated hydroxyanisole, and kept at −80°C until further processing for determination of PUFA levels (AA and DHA). The data represent mean levels of AA, DHA, or AA/DHA ratio values ± the SEM. *, P < 0.05, indicating that an S. suis-infected group is significantly different from corresponding noninfected counterparts, as determined by an unpaired Student t test.

We have previously shown that S. suis infection in mice results in a systemic infection and that bacteria are found in various organs such as the spleen and liver (7). Therefore, modifications in the AA and DHA profiles in these organs were evaluated. As shown in Fig. 1B and C, aberrant AA/DHA ratios were also observed in the spleens and livers of infected mice compared to mock-infected controls. Distortion of AA/DHA balance in the spleen of infected animals was particularly pronounced (AA/DHA ratios of 33.5 ± 2.4 in B6 mice and 35.6 ± 1.5 in CD mice) compared to uninfected controls with AA/DHA of 12.4 ± 0.4 in B6 mice and 15.4 ± 0.9 in CD1 mice. The AA/DHA ratio was also significantly increased in the liver (Fig. 1C). Moreover, the similar AA and DHA profile modifications in B6 and CD1 mice demonstrate that changes in the AA/DHA ratios induced by a S. suis infection are not dependent on the mouse genetic background.

Fenretinide modulates TNF in murine macrophages infected with S. suis.

S. suis is known to rapidly induce proinflammatory mediators in cells after in vitro infections and in mice after experimental infections (22, 27, 30). Treatment of uninfected cells with fenretinide had no detrimental effect on cell survival (Fig. 2A). Cell infection with S. suis resulted in high production of TNF (Fig. 2B). Of interest, TNF production was significantly reduced by pretreatment of macrophages with 1.25 μM fenretinide, although it did not completely return to basal levels. Similar results were obtained with LPS (data not shown). A simultaneous fenretinide treatment and S. suis infection of macrophages failed to reduce TNF production (data not shown).

FIG 2.

Pretreatment with fenretinide downregulates TNF release by macrophages infected with a virulent S. suis strain. Murine macrophages were seeded in a 24-well plate, and appropriate wells received 1.25 μM fenretinide for 24 h. Macrophages were then infected with S. suis strain 31533 for 16 h (MOI = 10). LPS (100 ng/ml) was used as positive control. Supernatants were harvested, cytotoxicity was determined by using a Cytotox 96 nonradioactive cytotoxicity assay kit (A), and TNF protein release was determined by ELISA (B). The results are presented as mean values ± the SEM of four different experiments. Differences were analyzed using an unpaired Student t test and were considered to be significant at P ≤ 0.05. *, P < 0.05.

Fenretinide attenuates ERK MAPK and COX-2 activation in murine macrophages infected with S. suis.

MAPKs play an important role in macrophage activation and release of proinflammatory mediators. We and others previously showed that ERK1/2 MAPK is activated following infection by S. suis (27, 31). We have also previously demonstrated that fenretinide is able to downregulate ERK1/2 phosphorylation in LPS-treated macrophages (32). In the present study, macrophages pretreated with fenretinide prior to S. suis infection showed a significant reduction in ERK1/2 activation compared to nontreated S. suis-infected macrophages (Fig. 3A). Fenretinide decreased the activation of ERK1/2 induced by S. suis to control level but did not completely abrogate activation of ERK1/2, as was the case with the specific inhibitor for ERK1/2. Activation of p38 was also monitored, but it was not significantly modified in macrophages infected with S. suis and/or treated with fenretinide (data not shown).

FIG 3.

Pretreatment with fenretinide downregulates ERK1/2 MAPK activation and COX-2 expression in macrophages infected with a virulent S. suis strain. Murine macrophages were seeded in a 24-well plate, and appropriate wells received 1.25 μM fenretinide for 24 h. Macrophages were then infected with either S. suis strain 31533 (MOI = 10) for 30 min for determination of ERK1/2 activation or 16 h for determination of COX-2 expression. LPS (100 ng/ml) was used as a positive control. ERK1/2 inhibitor UO126 (10 μM) was used as a negative control of ERK1/2 activation where appropriate. Cells were washed with phosphate-buffered saline and lysed with sodium dodecyl sulfate 2× lysis solution. Phospho-ERK1/2 (A) and COX-2 (B) were analyzed by Western blotting, and total protein levels (ERK1/2) were used to normalize differences in loading. The results are presented as mean values ± the SEM of four different experiments. Differences were analyzed using an unpaired Student t test and were considered to be significant at P ≤ 0.05. *, P < 0.05.

Activation of ERK1/2 is known to lead induction of COX-2 (33). Since S. suis infection was associated with a dramatic increase in AA levels and COX-2 plays a major role in the generation of proinflammatory prostaglandins/eicosanoids from AA (34), the levels of COX-2 in S. suis-infected macrophages, with or without fenretinide-treatment, were evaluated. As shown in Fig. 3B, COX-2 was markedly induced in macrophages infected with S. suis. In fact, levels were comparable to those observed in macrophages treated with 100 ng of LPS/ml (results not shown). S. suis-mediated COX-2 expression was markedly decreased by macrophage pretreatment with fenretinide, although it did not return to basal levels observed in noninfected macrophages. Similar results were obtained with LPS-treated cells (results not shown).

Fenretinide treatment improves survival and clinical score of B6 mice infected with a virulent S. suis strain.

Excessive inflammation occurring rapidly during the infection of mice with virulent S. suis strains is known to be responsible, at least in part, for the observed lethality (22). In order to improve mouse survival during the acute phase of the infection, mice were treated with fenretinide by gavage at 24 and 2 h prior to S. suis infection and again at 24 h after infection. Compared to vehicle-treated mice, fenretinide-treated mice showed significantly improved survival (P = 0.012) during the acute phase of the infection (up to 72 h; Fig. 4A). However, this difference could not be observed at 96 h p.i. Mice treated with fenretinide clearly presented lower clinical scores than vehicle-treated mice up to 72 h p.i. (P < 0.05) (Fig. 4B). It has been previously reported for cystic fibrosis that fenretinide reduces the expression of proinflammatory mediators without impairing the ability to control bacterial infection (20). In the present study, bacteremia was similar in both groups at 12 h (Fig. 4C) and up to 72 h (data not shown).

FIG 4.

Fenretinide reduces severity of clinical signs during infection by a virulent S. suis strain. B6 mice received by gavage two doses of fenretinide or vehicle-solution at −24 h and at −2 h prior to infection and a third dose at 24 h p.i. by intraperitoneal injection with 10⁷ CFU of S. suis strain 31533. (A) Survival was monitored for 4 days. *, Statistically significant differences (P < 0.05) between fenretinide-treated and vehicle-treated mouse groups, as determined by a log-rank (Mantel-Cox) test. (B) Clinical signs of infected mice were monitored and given a score for 72 h. The results are presented as mean values ± the SEM. Differences were analyzed by using a Mann-Whitney test and were considered significant at P ≤ 0.05. *, P < 0.05. Symbols (A and B): ●, S. suis; ◆, fenretinide/S. suis. (C) Blood bacteremia was determined by plating serial dilutions of blood samples onto agar plates. Colonies were counted, and data are expressed as the CFU/ml of blood. No statistically significant differences were observed between groups, as determined by an unpaired Student t test.

Fenretinide treatment downregulates excessive production of cytokines and chemokines in B6 mice infected with a virulent S. suis strain.

Since lower bacterial levels were not observed in fenretinide-treated S. suis-infected animals, which could have explained the observed improvement in early survival and clinical signs, the production of several proinflammatory mediators in plasma, known to be induced during S. suis infection, was measured. Fenretinide-treated S. suis-infected animals showed significantly lower levels of the cytokines IL-12p70, IL-6, and TNF compared to S. suis-infected untreated mice (Fig. 5). The chemokines CCL2, CCL3, CCL5, CXCL1, and CXCL10, responsible for the recruitment and activation of immune cells, also had their production dramatically lowered in fenretinide-treated mice. In the case of IL-12p70, fenretinide-treated animals infected with S. suis had levels as low as those observed in uninfected animals (Fig. 5). No differences could be observed for CCL4 and CXCL2 production between fenretinide-treated and untreated S. suis-infected animals (Fig. 5).

FIG 5.

Fenretinide reduces production of proinflammatory cytokines and chemokines in mice infected with a virulent S. suis strain. B6 mice received by gavage two doses of fenretinide or vehicle solution at −24 h and at 2 h prior to infection by intraperitoneal injection with 10⁷ CFU of S. suis strain 31533. An uninfected control group received similar fenretinide treatment. At 12 h p.i., mice were euthanized, and blood was harvested by cardiac puncture. Plasma concentration of IL-12p70, IL-6, TNF, CCL2, CCL3, CCL4, CCL5, CXCL1, CXCL2, and CXCL10 were determined by Luminex assay. The data represent mean values in pg/ml ± the SEM. Groups that are significantly different are indicated by different italic letters (a, b, and c), as determined by one-way ANOVA with P ≤ 0.05.

DISCUSSION

Inflammation is an essential process to protect the body from various injuries, including pathogens. Past in vitro and in vivo research on the activation of the innate immune system by pathogenic S. suis has shown that an important inflammatory response following infection can be detrimental to the host if left uncontrolled (7, 22, 27, 30, 35, 36). These studies concentrated more specifically on the cytokine and chemokine arms of the immune response. Two other studies investigated involvement of bioactive lipids known as eicosanoids, namely, prostaglandin E₂ (PGE₂), in the response of macrophages and endothelial cells to S. suis infection (37, 38). The link between modulation of AA and DHA levels during the course of S. suis infection and the expression of inflammatory mediators, including COX-2, known to modulate PGE₂ production, remained undefined prior to the present study.

Using two well-established animal models of S. suis infection, we observed a rapid increase of the plasma levels of proinflammatory ω-6 PUFA AA, while the ω-3 anti-inflammatory PUFA DHA was downregulated. Similar results were obtained in liver and spleen. PUFAs are important mediators in the orchestration of inflammatory traffic as they function as signaling molecules that regulate diverse inflammatory processes. AA is an important contributor to the onset of inflammation during bacterial infections (39, 40). AA, released from membranes by phospholipase A₂, can be metabolized by cyclooxygenases, such as COX-2, and lipoxygenases to generate prostaglandins, thromboxanes, and leukotrienes, which are potent proinflammatory mediators (10). Conversely, ω-3 PUFAs, such as DHA, are known to have anti-inflammatory properties and thus play an important role in the resolution of inflammation. DHA is metabolized into various mediators possessing anti-inflammatory properties such as resolvins and protectins (13–15). The results in the present study showed that S. suis-infected mice have a reduction in their DHA levels after the onset of infection, which could impair the resolution of inflammation.

A possible mechanism by which AA-derived metabolites exert their action is by activating the ERK1/2 MAPK pathway, which in turn upregulates COX-2 expression (41). Our in vitro results showed that ERK1/2 MAPK and COX-2 were both strongly activated in S. suis-infected macrophages. Although upregulation of MAPKs by S. suis has previously been reported (27, 31), it is the first time that expression of COX-2 is shown to be induced by S. suis. This would suggest that the phospholipase A2/COX-2 axis would be activated in order to generate various AA-derived proinflammatory lipid mediators, such as PGE₂, known to be induced by S. suis-infected macrophages and endothelial cells (37, 38). COX-2 and PGE₂ have also been implicated in the severity of infection caused by Streptococcus pyogenes (40). DHA is known to diminish NF-κβ activation by preventing nuclear p65 NF-κβ translocation, resulting in the downregulation of COX-2 gene expression (42). Furthermore, administration of DHA also diminishes the activation of ERK1/2 MAPK (32), which is known to be an inducer of COX-2 (41). Therefore, a reduction of DHA levels in S. suis-infected mice could lead, at least in part, to the strong proinflammatory response observed following infection by pathogenic S. suis.

Fenretinide, a synthetic retinoid, possesses well described antitumor, proapoptotic, and chemopreventative activities (43). More recent studies have shown that fenretinide can, at lower concentrations, modulate PUFAs in experimental models of cystic fibrosis and spinal cord contusion injury in mice (12, 18–20). However, no previous study had examined the effect of fenretinide during an acute bacterial infection. First, we demonstrated that a pretreatment with fenretinide was able to reduce, in vitro, TNF production by macrophages infected with S. suis. In addition, a significant reduction of S. suis-induced ERK1/2 MAPK and COX-2 could also be observed, suggesting a potential effect of fenretinide in reducing inflammation. Confirming in vitro results, fenretinide preadministration to S. suis-infected mice significantly reduced the early mortality rate and the severity of disease symptoms. Since the treatment with fenretinide, at least at the dosage used, had no effect on bacterial load, this partial protection could not be attributed to any antibacterial activity but rather to a reduction of systemic proinflammatory cytokines and chemokines by fenretinide administration. Interestingly, fenretinide did not totally abrogate the inflammatory response and mortality was delayed in infected animals but could not be totally prevented. In fact, if the animals were treated only after the S. suis infection or if they were infected with an overwhelming amount of bacteria (a five times larger amount of S. suis), no clear effect of fenretinide was observed (data not shown). Nevertheless, the results presented here showed a marked reduction of clinical signs related to the inflammatory response. The amplified inflammatory reaction induced by S. suis is very complex and yet required for adequate elimination of the pathogen. Thus, it is difficult to control the excessive inflammatory response of the host without impairing a protective immune response.

Our findings show that the excessive inflammation associated with aberrant fatty acid metabolism resulting from the infection with a virulent strain of S. suis plays an important role in the regulation of clinical outcome. Furthermore, successful pharmacological control of inflammation during the course of a S. suis infection might be as important as the proper selection of antibiotic regime to control this deadly infection. In toxic shock-like syndrome cases seen during S. suis human outbreaks, mortality reached >60% even if patients received antibiotic treatment, probably due to the proinflammatory cytokine storm (3). If effective in reducing the very first clinical signs, treatment with fenretinide might offer a window for a cotreatment with antibiotics to more appropriately control the acute infection caused by this emerging pathogen. The use of fenretinide in pigs to increase the success rate of antibiotic preventive programs to control S. suis infections might represent an interesting treatment option to be tested.