Identification of Specific Components of the Eicosanoid Biosynthetic and Signaling Pathway Involved in Pathological Inflammation during Intra-abdominal Infection with Candida albicans and Staphylococcus aureus

ABSTRACT

Polymicrobial intra-abdominal infections (IAIs) are a significant cause of morbidity and mortality, particularly when fungal pathogens are involved. Our experimental murine model of IAI involving intraperitoneal inoculation of Candida albicans and Staphylococcus aureus results in synergistic lethality (∼80%) due to exacerbated inflammation. Monomicrobial infection results in no mortality, despite a microbial burden and dissemination similar to those in a coinfection. In the coinfection model, the immunomodulatory eicosanoid prostaglandin E₂ (PGE₂) was determined to be necessary and sufficient to induce mortality, implicating PGE₂ as the central mediator of the amplified inflammatory response. The aim of this study was to identify key components of the PGE₂ biosynthetic and signaling pathway involved in the inflammatory response and explore whether these can be targeted to prevent or reduce mortality. Using selective pharmacological inhibitors of cyclooxygenases (COX) or PGE₂ receptor antagonists in the C. albicans-S. aureus IAI mouse model, we found that inhibition of COX and/or blocking of PGE₂ receptor 1 (EP1) or PGE₂ receptor 3 (EP3) signaling reduced proinflammatory cytokine production, promoted interleukin-10 production, reduced cellular damage in the peritoneal cavity, and, most importantly, significantly improved survival. The greatest effect on survival was obtained by the simultaneous inhibition of COX-1 activity and EP1 and EP3 receptor signaling. Importantly, early inhibition of PGE₂ pathways dramatically improved the survival of fluconazole-treated mice compared with that achieved with fluconazole treatment alone. These findings indicate that COX-1 and the EP1 and EP3 receptors mediate the downstream pathological effects of PGE₂ during polymicrobial IAI and may serve as effective therapeutic targets.

INTRODUCTION

Intra-abdominal infections (IAIs) are reported to be the second most common cause of sepsis and mortality in intensive care units (1, 2). IAIs are predominantly polymicrobial in nature, often involving both bacterial and fungal species, whereby diagnosis and treatment can be particularly challenging (3–5). The two major pathogens coisolated from IAIs are the polymorphic fungal pathogen Candida albicans and the Gram-positive bacterium Staphylococcus aureus (6). Fungal involvement also leads to more severe disease scores and increased rates of relapse and mortality (7–9). Further complicating treatment is the presence of drug-resistant strains (10).

A murine model of IAI has proven useful to study the interaction between C. albicans and S. aureus (11, 12). Accordingly, intraperitoneal inoculation of C. albicans and S. aureus results in 70 to 80% mortality, while a monoinfection with either organism alone results in no mortality (12). Mortality from the coinfection was associated with dramatically increased production of proinflammatory cytokines and the immunomodulatory eicosanoid prostaglandin E₂ (PGE₂) but no differences in the local microbial burden or dissemination from those seen in monoinfections (12, 13). Interestingly, pretreatment with the nonsteroidal anti-inflammatory drug (NSAID) indomethacin reduced PGE₂ production and inflammation and also prevented mortality (12). This protective effect of indomethacin was reversed by the administration of exogenous PGE₂ (12). These findings provided strong evidence that PGE₂ is the key mediator in the amplified proinflammatory response. However, the specific components of the eicosanoid pathway involved in PGE₂ synthesis and the targeted downstream signaling receptors on innate immune cells during infection are not known.

PGE₂ is a lipid-signaling eicosanoid synthesized from arachidonic acid by cyclooxygenases (COX), of which two isoforms exist in mammals, constitutive COX-1 and inducible COX-2 (14). The downstream signaling effects of PGE₂ are mediated through its activation of four specific cell surface receptors (PGE₂ receptor 1 [EP1] to EP4) on target cells (15, 16). In the current study, we set out to identify key components of the eicosanoid pathway involved in PGE₂ biosynthesis and signaling during C. albicans-S. aureus IAIs using selective pharmacological inhibitors (reviewed in references 17 and 18).

RESULTS

Inhibiting C. albicans-S. aureus-induced PGE₂ biosynthesis and signaling via the EP1 or EP3 receptor enhances survival.

Using selective COX-1 or COX-2 inhibitors, we first confirmed that blocking PGE₂ biosynthesis by either selective COX inhibitor was effective in our C. albicans-S. aureus IAI mouse model. Accordingly, inhibition of COX-1 or COX-2 activity significantly increased the survival rate from ∼25% to 50% by day 10 post-microbial inoculation, similar to the rate achieved with a nonselective inhibitor (Fig. 1A). To determine which EP receptor that PGE₂ interacts with to mediate the inflammatory response, we treated mice with selective EP receptor antagonists that bind covalently to one of the four receptors, effectively blocking signaling, and examined the effect on survival. Pharmacological inhibition of EP1 or EP3 receptor signaling significantly delayed mortality compared to the time of mortality of untreated mice. Conversely, treatment with EP2 or EP4 receptor antagonists resulted in no significant improvement in survival, with EP4 receptor antagonist treatment resulting in a slightly increased rate of mortality (Fig. 1B).

FIG 1.

Effect of inhibition of COX activity or PGE₂ receptors in vivo during polymicrobial IAI. (A) Effect of inhibition of COX-1 and/or COX-2 activity on survival. ⇓, the last dose of COX inhibitor was administered 8 h post-microbial inoculation. (B) Effect of inhibition of PGE₂ receptor signaling on survival. ↓, the last dose of EP receptor antagonist was administered at day 5 post-microbial inoculation. Mice were inoculated i.p. with C. albicans and S. aureus, and inhibitors were administered as indicated in Materials and Methods. Control mice received vehicle (PBS, DMSO) alone. Mortality was assessed using the Mantel-Cox log-rank test (*, P < 0.05 compared to the control groups). Shown are cumulative data from three independent experiments (n = 10 mice/group/experiment).

We next explored whether the combined inhibition of COX activity and EP1 or EP3 receptor signaling would have enhanced effects on survival. The combined administration of a COX-1 inhibitor and an EP1 or EP3 receptor antagonist failed to improve survival over that achieved by treatment with the COX-1 inhibitor alone (Fig. 2A). However, the combined inhibition of COX-1 activity and EP1 and EP3 receptor signaling provided significantly enhanced protection compared with that achieved with vehicle or COX-1 inhibitor treatment alone, with 100% survival being observed through day 7 (Fig. 2A). Survival slowly declined by day 10 following the last scheduled administration of EP receptor antagonists on day 5 (Fig. 2A). On the other hand, the combined inhibition of COX-2 activity and EP1 and/or EP3 receptor signaling showed no significant effect on survival over that achieved with COX-2 alone (Fig. 2B).

FIG 2.

Effect of early combined inhibition of COX-1 or COX-2 activity and EP1 and/or EP3 receptor signaling on long-term survival during polymicrobial IAI. (A) Effect of inhibition of COX-1 activity and EP1 and/or EP3 receptor signaling on survival. ⇓, the last dose of COX inhibitor was administered 8 h post-microbial inoculation; ↓, the last dose of EP receptor antagonist was administered at day 5 post-microbial inoculation. (B) Effect of COX-2 activity and EP1 and/or EP3 receptor inhibition on survival. Mice were inoculated i.p. with C. albicans and S. aureus, and inhibitors were administered as indicated in Materials and Methods. Control mice received vehicle (PBS, DMSO) alone. Mortality was assessed over time using the Mantel-Cox log-rank test (*, P < 0.05 compared to the control groups; **, P < 0.01 compared to the control groups). Shown are cumulative data from two independent experiments (n = 5 mice/group/experiment).

Effect of pharmacological inhibition of COX-1 activity and/or EP1 and/or EP3 receptor signaling on inflammatory responses and host damage.

Because we observed significantly improved survival with combined inhibition of COX-1 activity and EP1 and EP3 receptor signaling, we analyzed the effects on local and systemic markers of inflammation. First, we analyzed the PGE₂ levels in the peritoneal fluid of coinfected mice under treated conditions. As shown in Fig. 3, pharmacological inhibition of COX-1 activity significantly reduced PGE₂ levels beginning at 24 h and was sustained through day 5, indicating that the COX-1 isoform is a major source of PGE₂ even at early time points. Simultaneous inhibition of COX-1 activity and EP1 or EP3 receptor signaling showed levels of PGE₂ similar to those detected after inhibition of COX-1 activity alone (data not shown). However, no effect on PGE₂ levels was observed by inhibition of EP1 or EP3 receptor signaling (Fig. 3). While these results are somewhat expected, they argue against a positive-feedback loop to perpetuate PGE₂ production via EP1 or EP3 receptor signaling.

FIG 3.

Effect of early combined inhibition COX-1 activity and EP1 and/or EP3 receptor signaling on PGE₂ levels. Mice were inoculated i.p. with C. albicans and S. aureus, and inhibitors were administered as indicated in Materials and Methods. The levels of PGE₂ measured as PGE metabolites (PGEM) were assessed in peritoneal lavage fluid obtained from sacrificed mice at the indicated post-microbial inoculation time points. Values are plotted as the mean cytokine level (in picograms per milliliter) ± standard error of the mean (SEM). Data were analyzed using ANOVA and the unpaired Student's t test (ns, not significant; *, P < 0.05; **, P < 0.01). Shown are cumulative data from two independent experiments (n = 5 mice/group/experiment).

Next, we examined the effect of inhibition of COX-1 activity and/or EP1 and/or EP3 receptor signaling on the secretion of the hallmark proinflammatory cytokines interleukin-1β (IL-1β) and IL-6 in the peritoneum of coinfected mice. Pharmacological inhibition of COX-1 activity and/or EP1 and/or EP3 receptor signaling significantly reduced the local levels of production of IL-6 and IL-1β at 24 h post-microbial inoculation, and this was sustained through day 5 (Fig. 4A and B). At day 7 post-microbial inoculation, 2 days following the termination of EP1 and/or EP3 receptor antagonist treatment, the levels of both IL-6 and IL-1β were increased again toward control levels (data not shown).

FIG 4.

Effect of early combined inhibition of COX-1 activity and EP1 and and/or EP3 receptor signaling on local IL-6 (A) and IL-1β (B) levels. Mice were inoculated i.p. with C. albicans and S. aureus, and inhibitors were administered as indicated in Materials and Methods. The levels of cytokines were assessed in peritoneal lavage fluid obtained from sacrificed mice at the indicated post-microbial inoculation time points. Values are plotted as the mean cytokine level (in picograms per milliliter) ± standard error of the mean (SEM). Data were analyzed using ANOVA and the unpaired Student's t test (*, P < 0.05; **, P < 0.01; ***, P < 0.001). Shown are cumulative data from two independent experiments (n = 5 mice/group/experiment).

To understand further the mechanism behind the enhanced survival and reduced proinflammatory response in coinfected mice administered inhibitors of COX-1 activity and/or EP1 and/or EP3 receptor signaling, we examined the local anti-inflammatory response in the peritoneal cavity by measuring IL-10 levels. Significantly higher levels of IL-10 (at least 2-fold) were detected in the peritoneal fluid of coinfected mice treated with inhibitors of COX-1 activity and EP1 or EP3 receptor signaling than in the peritoneal fluid of control mice predominantly within 24 h post-microbial inoculation, with the group treated with an EP3 receptor inhibitor and the group treated with inhibitors of COX-1 activity and EP3 receptor signaling showing sustained levels through to 72 h. All groups showed a decline at day 5 (Fig. 5).

FIG 5.

Effect of early combined inhibition of COX-1 activity and EP1 and/or EP3 receptor signaling on local anti-inflammatory cytokine IL-10 levels. Mice were inoculated i.p. with C. albicans and S. aureus, and inhibitors were administered as indicated in Materials and Methods. The levels of the cytokine were assessed in peritoneal lavage fluid obtained from sacrificed mice at the indicated post-microbial inoculation time points. Values are plotted as the mean cytokine level (in picograms per milliliter) ± standard error of the mean (SEM). Data were analyzed using ANOVA and the unpaired Student's t test (ns, not significant; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001). Shown are cumulative data from two independent experiments (n = 5 mice/group/experiment).

We also evaluated host damage in the peritoneal cavity via a lactate dehydrogenase (LDH) assay under the respective inhibition conditions. As shown in Fig. 6, significantly higher levels of LDH were present in the peritoneal cavity of control mice than in the peritoneal cavity of mice in the groups that received inhibitors of COX-1 activity and/or EP1 and/or EP3 receptor signaling during coinfection through day 5.

FIG 6.

Effect of early combined inhibition of COX-1 activity and EP1 and/or EP3 receptor signaling on host damage during polymicrobial IAI. Mice were inoculated i.p. with C. albicans and S. aureus, and inhibitors were administered as indicated in Materials and Methods. LDH levels were assessed in peritoneal lavage fluid obtained from sacrificed mice at the indicated post-microbial inoculation time points. Values are expressed as the mean optical density at 450 nm (OD₄₅₀) and were compared using ANOVA and the unpaired Student's t test (**, P < 0.01; ***, P < 0.001; ****, P < 0.0001). The data shown are representative of those from two independent experiments (n = 5 mice/group/experiment).

Azole therapy extends survival following inhibition of PGE₂ biosynthesis and signaling.

Because mortality occurred following termination of treatment with inhibitors of COX-1 activity and EP1 and EP3 receptor signaling (after day 5 postinfection), this suggested that inhibition of PGE₂ biosynthesis and signaling must be maintained for efficacy. In addition, this also suggests that targeting of the inflammatory responses does not reduce or eliminate the microbial burden. Therefore, we examined whether treatment with COX-1 inhibitors and/or EP1 and/or EP3 receptor antagonists improves the efficacy of postinfection antifungal therapy. Oral fluconazole treatment was initiated at day 5 post-microbial inoculation and at the point of COX-1 and EP receptor inhibitor treatment termination and was administered daily through day 12. In control animals (which received the vehicle only), fluconazole treatment was initiated earlier (day 3), prior to significant mortality, and was continued through day 10. With all combinations of COX-1 inhibitor and EP receptor antagonist administration, fluconazole significantly improved the rate of survival during coinfection (Fig. 7). However, fluconazole treatment alone delayed mortality slightly, but the difference in the rate of mortality from that in untreated mice was otherwise insignificant, despite the initiation of antifungal treatment at an earlier time point. Noteworthy is the finding that the inhibition of COX-1 activity and EP3 receptor signaling or COX-1 activity and EP1 and EP3 receptor signaling prior to fluconazole treatment significantly improved the rate of survival compared with that achieved with antifungal therapy alone.

FIG 7.

Effect of combined early inhibition of COX-1 activity and EP1 and/or EP3 receptor signaling and fluconazole treatment on survival during polymicrobial IAI. Mice were inoculated i.p. with C. albicans and S. aureus, and inhibitors were administered as indicated in Materials and Methods. Control mice received vehicle (PBS, DMSO) alone and/or fluconazole daily from day 3 to day 10 post-microbial inoculation. COX-1 and EP receptor inhibitor-treated groups received fluconazole daily from day 5 to day 12 post-microbial inoculation, with EP receptor antagonist treatment being terminated at day 5. Mortality was assessed over time using the Mantel-Cox log-rank test (ns, not significant compared to the control groups; *, P < 0.05 compared to the control groups; **, P < 0.01 compared to the control groups). Shown are cumulative data from two independent experiments (n = 10 mice/group/experiment).

DISCUSSION

Polymicrobial IAIs involving fungal species result in rates of morbidity and mortality higher than those in infections involving only bacterial species (8–10). In our murine model of a polymicrobial IAI with C. albicans and S. aureus, the polymicrobial IAI also results in synergistic effects on lethality (70 to 80% mortality), while a monoinfection with either organism alone results in no mortality (12). This effect on mortality is not limited to C. albicans but was also observed with other non-albicans Candida species (NAC), including C. krusei and C. tropicalis, underscoring the pathogenic role of fungal species in polymicrobial IAIs (19). We have also tested several mouse strains, including the C57BL/6 and C3H/HeN strains, and both male and female mice and observed similar synergistic effects of coinfection with C. albicans and S. aureus on lethality (20). Therefore, this polymicrobial IAI model is highly reproducible, independent of the host genetic background, and useful in interrogating potential interventions/therapeutics.

IAI with C. albicans and S. aureus is characterized by an amplified proinflammatory response mediated by the immunomodulatory eicosanoid PGE₂ (12). In other studies designed to identify the host cell types responsible for this pathological inflammation, we determined that Gr-1-positive leukocytes, but not macrophages or T/B cells, are critical to the pathogenesis (20). The mechanisms underlying the PGE₂-dependent inflammatory response are currently unknown. We previously demonstrated that prophylaxis with the nonselective NSAID indomethacin prevents mortality in mice coinfected with C. albicans and S. aureus (12). However, global inhibition of both the COX-1 and COX-2 pathways could be detrimental clinically, particularly in at-risk intensive care unit patients on complex therapeutic regimens. There are several murine and human studies that have examined the effects of nonselective or COX-2-selective eicosanoid inhibitors during peritonitis or sepsis. Some studies supported a protective effect, while others, particularly studies in humans, were inconclusive (21–24). However, it is likely that COX-1 is responsible for the initial production of PGE₂ because resident peritoneal macrophages are preloaded with this enzyme, while COX-2 is transcribed/translated only after cells are stimulated (postinfection).

The results of the in vivo experiments presented here show that pharmacological inhibition of PGE₂ synthesis and/or PGE₂ signaling protects mice from tissue-damaging inflammation and mortality. Protection (∼50%) was afforded by treatment with an inhibitor of COX-1 or COX-2 activity or EP1 or EP3 receptor signaling alone but not an antagonist of EP2 or EP4 receptor signaling. The greatest protection (in which ∼90% survival was sustained for 5 days after the termination of all treatments) was observed with the simultaneous inhibition of COX-1 activity and PGE₂ signaling via EP1 and EP3 receptors, all of which correlated with significantly lower levels of inflammation, including significantly lower PGE₂ levels. These findings implicate a role for COX-1 activity and the EP1 and EP3 receptors in mediating PGE₂-induced pathological inflammation during C. albicans-S. aureus IAI. C. albicans is also known to produce PGE₂ from host arachidonic acid, which can be inhibited with various NSAIDS, including indomethacin (25, 26). Therefore, we cannot rule out the possibility of effects of COX inhibitors on fungal eicosanoid production, which would be an added benefit during infection. On the other hand, enhanced protection was not observed when inhibition of COX-2 activity was combined with inhibition of the EP1 and/or EP3 receptor. The lack of a beneficial effect could be explained by the fact that COX-2 is induced following infection and the fact that the pathological effects observed occurred within 24 h post-microbial inoculation, when COX-1 is active (12). Consistent with our results, COX-1 has also been implicated in macrophage-dependent inflammation in several other models of sepsis (induced with lipopolysaccharide [LPS] or bacterial infusion) (27). Therefore, selective targeting of COX-1 during sepsis may be an effective therapeutic strategy, while it would limit the potential negative effects of nonselective inhibition of COX-1/COX-2 activity.

PGE₂ signaling via specific EP receptors can lead to pro- and/or anti-inflammatory responses, depending on the inflammatory stimulus and target cell (28, 29). Each EP receptor subtype acts via distinct signaling pathways. While EP1 to EP4 are all G-protein-coupled receptors (GPCR), each is coupled to different G proteins and acts via different secondary messengers (29). Signaling via EP2 and EP4 receptors can lead to both pro- and anti-inflammatory responses in different models of inflammation, while EP1 and EP3 receptors are mainly proinflammatory (29). In addition, EP1 and EP3 are known to play a role in the febrile response in models of LPS-induced septic shock (30). In this study, the enhanced survival in mice treated with COX-1 and EP1 and/or EP3 receptor antagonists correlated well with the reduced levels of PGE₂ and proinflammatory cytokines (IL-6 and IL-1β), which is consistent with the findings of other studies that have reported that IL-6 and IL-1β are potent inflammatory markers of sepsis in mammals and that have shown that suppressing their production enhances survival (31–33). This is also consistent with the reported role of EP1 and EP3 receptors in mediating proinflammatory responses and our data demonstrating a protective role of administration of either EP1 or EP3 receptor antagonists during IAI.

The enhanced survival in mice treated with COX-1 and EP1 and/or EP3 receptor antagonists also correlated with increased anti-inflammatory cytokine IL-10 levels and less peritoneal damage (lower LDH levels). IL-10, also known as human cytokine synthesis inhibitory factor (CSIF), is reported to be a negative regulator of cytokine production, with a loss of IL-10 function resulting in an excessive inflammatory response to gut infection (34). Therefore, the increased IL-10 levels observed in the mice treated with PGE₂ pathway inhibitors may be directly responsible for restricting the excessive inflammatory response. Reduced inflammation also limited damage to the peritoneal cavity. There was enhanced survival following EP1 or EP3 inhibition, even though these mice had PGE₂ levels comparable to those of the vehicle-only-treated group; however, these mice were presumably protected by the inhibition of PGE₂ signaling via EP1 or EP3, which reduced the proinflammatory response, enhanced IL-10 production, and resulted in less peritoneal damage. Conversely, PGE₂ signaling via EP2 or EP4 is known to enhance IL-10 production (35). Although it was not tested directly, the enhanced mortality observed in EP4 receptor antagonist-treated mice may have been due to the loss of IL-10, which enhanced the effects of the proinflammatory cytokines.

Another interesting observation from this study was that when treatment with inhibitors of COX-1 activity and EP1 and EP3 receptor signaling was discontinued at day 5 postinoculation, the mice began to succumb to infection, illustrating the importance of PGE₂ signaling pathways in eliciting pathological inflammation and the need for continuous inhibition to sustain beneficial outcomes. These results are also in agreement with those of our previous studies demonstrating that administration of PGE₂ alone following NSAID treatment caused mortality (12). While it is impossible to measure inflammatory mediators postmortem in the mice that succumbed following the cessation of treatment, we assume that mortality coincided with infection-driven septic inflammation and that long-term survival would also depend on microbial clearance. In support of this idea, fluconazole treatment administered following administration inhibitors of COX-1 activity and EP1 and EP3 receptor signaling significantly enhanced survival. Importantly, early inhibition of PGE₂ pathways (inhibition of COX-1 activity and EP1 and EP3 receptor signaling) dramatically improved the survival of fluconazole-treated mice compared with that of mice receiving fluconazole treatment alone. Delayed diagnosis and delayed treatment are major problems during IAI and sepsis, leading to increased mortality (36–38). Therefore, prophylactic treatment of at-risk patients with PGE₂ pathway inhibitors, many of which are used clinically, may prolong survival before antimicrobial/antifungal treatment is initiated.

In summary, these studies provide insight into the pathological roles of PGE₂, COX-1, and EP1 and EP3 receptors during C. albicans and S. aureus IAI. We observed the protection of mice from lethal infection by pharmacological inhibition of these key components that correlated with reduced local proinflammation and increased anti-inflammation and reduced peritoneal damage.

MATERIALS AND METHODS

Strains and growth conditions.

The methicillin-resistant S. aureus strain used in all experiments, strain NRS383, was obtained from the Network on Antimicrobial Resistance in S. aureus (NARSA) data bank. NRS383 is positive for the toxic shock syndrome toxin (tst) and δ-toxin genes. Frozen stocks were obtained at −80°C and streaked onto Trypticase soy agar (TSA) plates at 37°C prior to use. A single colony was transferred to 10 ml of TSA broth and shaken at 37°C overnight. On the following day, the overnight culture was diluted 1:100 in fresh growth medium and shaken at 37°C for 3 h to obtain cells in mid-log growth phase.

The C. albicans strain used in these experiments was DAY185, a prototrophic control strain derived from strain BWP17, an auxotrophic derivative of strain SC5314, into which the HIS1, URA3, and ARG4 genes were reinserted (39). Frozen stocks were obtained at −80°C and streaked onto yeast peptone dextrose (YPD) agar prior to use. A single colony was transferred to 10 ml of YPD broth and shaken at 30°C for 18 h. Prior to infection, both C. albicans and S. aureus were rinsed 3 times by centrifugation in sterile phosphate-buffered saline (PBS; pH 7.4), counted on a hemocytometer, and diluted in sterile PBS to prepare standardized inocula.

Chemical inhibitors.

The following chemical inhibitors were tested in vivo at the indicated concentrations (Table 1): SC-560 (200 μg/ml; Cayman Chemicals), NS-398 (100 μg/ml; Cayman Chemicals), SC 51322 (100 μg/ml; Tocris); PF 04418948 (100 μg/ml; Tocris), L-798,106 (100 μg/ml; Tocris), and ONO AE3 208 (100 μg/ml; Tocris). Each chemical was prepared freshly as a concentrated stock in dimethyl sulfoxide (DMSO) and diluted to the respective working concentration in sterile PBS (final DMSO concentration, 0.28%). The vehicle control contained 0.28% DMSO in PBS. All inhibitors were confirmed to have no growth-inhibitory effect on C. albicans or S. aureus in vitro prior to use (40).

TABLE 1.

Selective and nonselective COX enzyme inhibitors and EP receptor antagonists

Chemical name	Inhibitory target	Concn (mg/kg)
Indomethacin	COX-1 and COX-2 enzymes	5
SC-560	COX-1 enzyme	20
NS-398	COX-2 enzyme	10
SC 51322	EP1 receptor	10
PF 04418948	EP2 receptor	10
L-798,106	EP3 receptor	10
ONO AE3 208	EP4 receptor	10

Mouse model of C. albicans-S. aureus intra-abdominal infection.

All animals were housed and handled according to institutionally recommended guidelines. All animal protocols were reviewed and approved by the Institutional Animal Care and Use Committee (IACUC) of the LSU Health Sciences Center, New Orleans, LA. Mice were given access to food and water ad libitum. In all experiments, 5- to 7-week-old female outbred Swiss mice, purchased from Charles River, Frederick, MD (NIH), were used.

Intra-abdominal infection was conducted as previously described (9). Briefly, mice were injected intraperitoneally (i.p.) with 7 × 10⁶ CFU of C. albicans and 8 × 10⁷ CFU of S. aureus (8.7 × 10⁷ CFU total) in 0.2 ml of PBS using a 27-gauge (1/2-in.) needle (n = 10 mice/group, unless otherwise indicated). After inoculation, the mice were observed over 10 days for signs of morbidity (hunched posture, inactivity, lethargy, and ruffled fur) and mortality. Mice that were significantly moribund were euthanized according to institutionally recommended guidelines.

Chemical inhibitor treatment.

Groups of mice were intraperitoneally administered 0.1 ml of inhibitor. The COX inhibitor or vehicle control was administered 4 h prior to and 4 and 8 h after microbial inoculation, while the EP receptor antagonists or vehicle control was administered daily starting 1 day prior to and continuing for 5 days after microbial inoculation. Each selective inhibitor/antagonist and its concentration used in this study have been previously well established in various in vivo infection models (41–47).

Fluconazole treatment.

Stock solutions of fluconazole (Sigma) were prepared in ethanol and stored at −20°C. The stock solution was further diluted daily immediately before use and administered by oral gavage once daily at 10 mg/kg of body weight. Fluconazole treatment was started at day 3 post-microbial inoculation for the vehicle-only-treated groups or day 5 post-microbial inoculation for the inhibitor-treated groups and continued for 7 days in each case.

PGE₂ analysis.

PGE₂ (in picograms per milliliter) was quantified with a prostaglandin E metabolite enzyme-linked immunosorbent assay (ELISA) kit (Cayman Chemicals) according to the manufacturer's instructions. All samples were measured in duplicate, and the results were averaged.

Cytokine analysis.

The concentrations of IL-6, IL-1β, and IL-10 (in picograms per milliliter) in peritoneal lavage fluid were determined by ELISA (eBioscience, San Diego, CA). All samples were measured in duplicate, and the results were averaged.

LDH assay.

The levels of lactate dehydrogenase (LDH) released in peritoneal lavage fluid were determined colorimetrically with an LDH assay kit according to the manufacturer's instructions (Abcam, Inc.). The absorbance at 450 nm was read on a Multiskan Ascent microplate reader. All samples were measured in duplicate, and the results were averaged.

Statistical analysis.

All experiments used groups of 5 to 10 mice and were repeated at least in duplicate, except where noted. Survival data were analyzed using the log-rank (Mantel-Cox) test, which also incorporated chi-square analyses as a measure of overall survival. Analysis of variance (ANOVA) and the unpaired Student's t test were used to analyze all ELISA and LDH assay data. In all tests, differences were considered significant at a P value of <0.05. All statistical analyses were performed with GraphPad Prism software.