A natural immune modulator attenuates stress hormone and catecholamine concentrations in polymicrobial peritonitis.

BACKGROUND: Activated hexose correlated compound (AHCC), derived from shiitake mushrooms, increases resistance to infection in immunocompromised hosts with positive effects on dendritic cells, natural killer cell function and interleukin 12 production. It may also be attenuating the systemic inflammatory response by regulating the secretion of cortisol and norepinephrine (NE). METHODS: Female Swiss-Weber mice were pretreated with AHCC (Amino Up Chemical Co., Sapporo, Japan) or water by gavage for 10 days before undergoing cecal ligation and puncture (CLP). Peritoneal exudate cells and blood samples were harvested at 4 hours and 24 hours following CLP. Plasma and peritoneal concentrations of cortisol and NE were obtained using enzyme-linked immunosorbent assay. Peritoneal bacteria were quantified by colony counts after 4 hours and 24 hours. Significance was denoted by a p < 0.05. RESULTS: Plasma and peritoneal cortisol concentrations were increased 4 hours after CLP compared with normal controls, with no difference between the pretreated groups. Concentrations of cortisol decreased from 4 hours to 24 hours after CLP with AHCC (plasma, p = 0.009; peritoneal, p < 0.001), and peritoneal cortisol at 24 hours was lower with AHCC as compared with water (p = 0.028). There was no change in plasma or peritoneal NE concentrations at 4 hours. At 24 hours, higher concentrations of NE were detected in both plasma and peritoneal fluid, with lower plasma concentrations in those gavaged with AHCC (p = 0.015). There was no significant difference in peritoneal bacteria counts. CONCLUSION: Enhanced immune function observed with AHCC could be caused by attenuated concentrations of stress hormones and catecholamines.

Liver injury and abscess formation in secondary murine peritonitis.

OBJECTIVE: To investigate liver damage and abscess formation in murine, secondary peritonitis. SUBJECTS: Male C57BL/6 mice. TREATMENT: Intraperitoneal injection with 10(3) CFU Klebsiella pneumoniae and treatment with gentamicin 5 mg/kg/day (BID), subcutaneously. METHODS: Animals were killed at 12, 24, 48 and 72 h after infection. Bacterial burden was determined in the blood and the liver. Liver abscess formation was assessed macroscopically and microscopically. Plasma levels of alkaline phosphatase (ALP) and alanine transaminase (ALT) were measured. Polymorphonuclear leukocyte (PMN) accumulation was assessed via tissue myeloperoxidase (MPO) concentrations. Liver interleukin-10 (IL-10) levels were determined by ELISA. RESULTS: K. pneumoniae was detectable in the blood and the liver at 12 h after infection. Liver abscess formation was visible earliest at 24 h after infection and most pronounced within the right liver lobes. ALP and ALT levels peaked at 12 and 24 h after infection, respectively. MPO was elevated in the right and left liver lobes at 12 h but only in the right lobes at 48 h after infection, compared to tissue levels in naïve mice. Liver IL-10 concentrations were not significantly increased. CONCLUSION: Peritonitis led to liver injury and abscess formation but did not significantly affect tissue concentrations of anti-inflammatory IL-10.

Endogenous IL-10 leads to impaired bacterial clearance and reduced survival in a murine model of chronic peritonitis.

We previously observed insufficient neutrophil accumulation and a lack of TNF-alpha response at the site of infection until bacteria numbers >10(5) colony forming units in our model of chronic murine peritonitis, suggesting a defective host response after bacterial challenge with Klebsiella pneumoniae (Klebsiella). The aim of this study was to determine a potentially immunosuppressive effect of IL-10 in this model of chronic peritonitis. Balb/c animals were injected with 10(3) colony forming units Klebsiella intraperitoneally. Gentamicin (5 mg/kg/day BID) was given subcutaneously (s.c.) for two days and then withdrawn. Animals were treated with anti-IL-10 antibody or IgG isotype control (s.c.) before or after Klebsiella administration. Survival was determined over 14 days. Similarly treated animals were harvested after 48 h to obtain liver tissue, peritoneal fluid and blood. Bacteria and neutrophil counts were determined. TNF-alpha and IL-10 were measured by ELISA. Anti-IL-10 antibody significantly increased survival and bacterial clearance in the observed compartments. Anti-IL-10 administration did not lead to an increase in TNF-alpha concentrations or neutrophil accumulation at the site of infection at lower levels of Klebsiella. We conclude that endogenous IL-10 is detrimental for survival and bacterial clearance in this model of chronic peritonitis.

Postresuscitation tissue neutrophil infiltration is time-dependent and organ-specific.

BACKGROUND: Hemorrhagic shock with conventional resuscitation (CR) primes circulating neutrophils and activates vascular endothelium for increased systemic inflammation, superoxide release, and end-organ damage. Adjunctive direct peritoneal resuscitation (DPR) with intraperitoneal instillation of a clinical peritoneal dialysis solution decreases systemic inflammation and edema formation by enhancing tissue perfusion. The aim of this study is to determine the effect of adjunctive DPR on neutrophil and fluid sequestration. METHODS: Anesthetized rats were hemorrhaged to 40% mean arterial pressure for 60 min. Animals were randomized for CR with the return of the shed blood plus two volumes of saline, or CR plus adjunctive DPR with 30 mL of intraperitoneal injection of a clinical peritoneal dialysis solution. Tissue myeloperoxidase (MPO) level, a marker of neutrophil sequestration, and total water content were assessed in the gut, lung, and liver in sham animals and at time-points 1, 2, 4, and 24 h postresuscitation. RESULTS: Resuscitation from hemorrhagic shock increases MPO level in all tissues in a near-linear fashion during the first 4 h following resuscitation. This occurs irrespective of the resuscitation regimen used. Tissue MPO level returned to baseline at 24 h following resuscitation except in the liver where CR and not adjunctive DPR caused a significant rebound increase. Adjunctive DPR prevented the CR-mediated obligatory fluid sequestration in the gut and lung and maintained a relative normal tissue water in these organs compared with CR alone (n = 7, F = 10.1, P < 0.01). CONCLUSION: Hemorrhagic shock and resuscitation produces time-dependent organ-specific trends of neutrophil sequestration as measured with tissue levels of myeloperoxidase, a marker of neutrophil infiltration. Modulation of the splanchnic blood flow by direct peritoneal resuscitation did not alter the time-dependent neutrophil infiltration in end-organs, suggesting a subordinate role of blood rheology in the hemorrhage-induced neutrophil sequestration. Vulnerable window for neutrophil-mediated tissue damage exists during the first 4 h following resuscitation from hemorrhagic shock in rats. Direct peritoneal resuscitation prevents the early obligatory fluid sequestration and promotes early fluid mobilization.

Decreased neutrophil response in a model of chronic intraperitoneal Klebsiella infection.

BACKGROUND: Previously, we developed an infection model by intraperitoneal (IP) injection of Klebsiella pneumoniae. The high early mortality rate prompted a study of the effect of gentamicin on the disease course. METHODS: Infection was induced in Balb/c mice by IP inoculation of 10(3) colony-forming units (cfu) of K. pneumoniae serotype 2. Mice then received either saline or gentamicin twice daily at 5 mg/kg/day. Survival and weight loss were determined daily over 14 days. Leukocyte counts (WBC) were performed on peritoneal lavage fluid (PL) and peripheral blood. Bacterial counts in blood, lung homogenate, and PL were determined by culture. RESULTS: A significant survival benefit was seen in the gentamicin group by 48 h (p < 0.05), which persisted at 14 days. Weight loss of 2 g (p < 0.05) occurred in all mice by day three, with recovery seen only in the gentamicin-treated mice. Mice that did not regain their weight by day seven or lost more than 4 g (20%) died. All animals had peritoneal bacteria recovered at all time points. The gentamicin group showed fewer bacteria at one day in lung, blood, and PL, the difference being significant for the PL. Gentamicin did not clear peritoneal bacteria, and counts remained at inoculum concentration (10(3) cfu), but elicited no significant neutrophil influx. Whereas there was a progressive increase in total lavage WBC over time, the gentamicin group showed no significant neutrophil influx in PL or lung until bacterial counts exceeded 10(5) cfu. CONCLUSIONS: Despite persistent intra-abdominal infection, gentamicin treatment significantly prolonged survival in this uniformly lethal model. Although antibiotic treatment did not suppress bacteremia, it did diminish bacterial spread from the blood to the lung. This procedure produced a clinically relevant, novel model of antibiotic-treated chronic intraperitoneal infection, which will allow future study of specific host defense mechanisms.

Antibiotic modulation in a clinically relevant model of chronic intraabdominal infection.

Continuous and twice-daily cefoxitin dosing was used in a highly lethal model of acute peritonitis in mice using intraperitoneal (IP) Klebsiella pneumoniae (Kpn). The purpose was to use antibiotics to create a model of chronic infection. Male Balb/c mice (averaging 20 g body weight) were inoculated IP with 10(3) colony-forming units (CFU) Kpn serotype 2. Controls received subcutaneous saline either twice daily or continuously. Antibiotic groups received 300 mg/kg per day of cefoxitin either twice daily or continuously. Survival and daily weight losses were determined. Another group was inoculated with 10(3) Kpn given twice daily saline or cefoxitin and harvested at 24 hours. Leukocyte counts were performed on peritoneal exudate cells (PEC) and peripheral blood. Cultures determined Kpn counts in blood, lung, and PEC. By 24 hours, saline-treated animals had lost more weight than cefoxitin mice (1 g vs. 2 g, P < 0.05). Continuous cefoxitin showed significant advantage with 50 per cent mortality at 5 days. Kpn levels were not significantly altered by cefoxitin. Cefoxitin treatment extended chronicity by preventing weight loss and increasing survival in a highly lethal, monomicrobial peritonitis model. This model will allow future study of specific host defense mechanisms over a prolonged time period.

Genetic background influences natural killer cell activation during bacterial peritonitis in mice, and is interleukin 12 and interleukin 18 independent.

Some mouse strains produce strong pro-inflammatory, T-helper (Th)1 responses (e.g. C57BL/6), or strong anti-inflammatory, Th2 responses (e.g. BALB/c). The exact mechanisms for development of distinct immune responses to infection are not completely understood, although cytokines such as interleukin (IL)-12, IL-18 and IL-4 are known to play roles. Natural killer T (NKT)/natural killer (NK) cells are important regulators of immune responses in infection and non-infection models, and NKT/NK activation is also regulated by IL-12 and IL-18 in many models. We investigated the role of IL-12/IL-18 in NKT/NK activation in murine bacterial peritonitis, as well as differential NKT and NK cell activation in C57BL/6 and BALB/c mice. No differences in NKT or NK cell activation or intracellular interferon (IFN)-gamma were determined between mice given control, anti-IL-12 or anti-IL-18 antibodies or in NKT/NK cell activation in STAT4-/- mice (deficient in IL-12 signaling) or wild type controls. However, there were significant differences in the activation of NKT and NK cells between C57BL/6 mice and BALB/c mice, with NKT/NK cytokine production following Th1 or Th2 lines dependent on strain. This suggests a role for NKT and NK cell activation in the development of Th1 and Th2 responses during bacterial infection independently of IL-12 or IL-18.

Leukotriene B4 receptor (BLT-1) modulates neutrophil influx into the peritoneum but not the lung and liver during surgically induced bacterial peritonitis in mice.

Leukotriene B4 (LTB4) is a rapidly synthesized, early neutrophil chemoattractant that signals via its cell surface receptor, BLT-1, to attract and activate neutrophils during peritonitis. BLT-1-deficient (BLT-1(-/-)) mice were used to determine the effects of LTB4 on neutrophil migration and activation, bacterial levels, and survival after cecal ligation and puncture (CLP). Male BLT-1(-/-) or wild-type (WT) BALB/c mice underwent CLP. Tissues were harvested for determination of levels of bacteria, myeloperoxidase (MPO), LTB4, macrophage inflammatory protein 2 (MIP-2), and neutrophil (polymorphonuclear leukocyte [PMN]) numbers at 4 and 18 h after CLP. PMN activation was determined by an assessment of phagocytosis ability and CD11b expression. Survival was also determined. BLT-1(-/-) mice had decreased numbers of PMNs in the peritoneum at both 4 and 18 h after CLP but increased numbers of PMNs in the blood at 18 h compared with WT mice. Liver and lung MPO levels were significantly higher in BLT-1(-/-) mice at both 4 and 18 h after CLP, with increased bacterial levels in the blood, the liver, and peritoneal fluid at 4 h. Bacterial levels remained higher in peritoneal fluid at 18 h, but blood and liver bacterial levels at 18 h were not different from levels at 4 h. PMN phagocytosis and CD11b levels were decreased in BLT-1(-/-) mice. LTB4 levels were similar between the groups before and after CLP, but MIP-2 levels were decreased both locally and systemically in BLT-1(-/-) mice. Survival was significantly improved in BLT-1(-/-) mice (71%) compared with WT mice (14%) at 48 h post-CLP. Thus, LTB4 modulates neutrophil migration into the mouse peritoneum, but not the lung or liver, after CLP. Despite higher bacterial and PMN levels at remote sites, there was increased survival in BLT-1(-/-) mice compared to WT mice. Decreased PMN activation may result in less remote organ dysfunction and improved survival.

Mortality in murine peritonitis correlates with increased Escherichia coli adherence to the intestinal mucosa.

During peritonitis, bacterial adherence is the initial step in a series of events that include mucosal infection, bacterial translocation, organ dysfunction, and death. Adherent Escherichia coli levels increase in response to stress. This study was designed to assess the adherence of E. coli to the cecal mucosa after cecal ligation and puncture (CLP) of increasing severity and to determine whether a relationship exists between adherence of bacteria and mortality. Sham surgery, sterile peritonitis (thioglycollate administration), lethal CLP (18-gauge double-puncture), and nonlethal CLP (23-gauge single-puncture) were performed on Swiss Webster mice and compared with normal mice or before CLP (time 0). Specimens of bowel tissue were harvested, and serial log dilutions of homogenized specimens or bowel contents were plated and cultured on media selective for determination of individual bacterial species. Low levels of E. coli and Proteus mirabilis adhered to the mucosa of unmanipulated controls; however, adherence of both species increased significantly by 18 hours after both lethal and nonlethal CLP. After 18 hours, adherent E. coli levels increased by greater than 5 x 10(6)-fold compared to unmanipulated controls, whereas P. mirabilis levels decreased. After nonlethal CLP, adherent P. mirabilis increased 3 x 10(6)-fold compared to unmanipulated animals, whereas E. coli levels did not increase after 24 hours. Sterile peritonitis had little effect on bacterial adherence. Higher levels of adherent E. coli in the cecum correlate with the increased mortality observed after lethal CLP. Higher levels of adherent P. mirabilis appear to prevent the overgrowth of adherent E. coli following nonlethal CLP. Our data indicate that E. coli plays a key role in mortality from polymicrobial peritonitis and that Proteus may be antagonistic to E. coli in murine peritonitis.

A novel model of pneumonia from intraperitoneal injection of bacteria.

BACKGROUND: Pneumonia remains a major clinical problem in the surgical patient. Experimental modeling by intratracheal injection of bacteria is not consistently reproducible. In an attempt to produce peritonitis by Klebsiella, we found evidence of pneumonia on autopsy and further developed this approach as a new experimental model. METHODS: Male Swiss Webster mice were given intraperitoneal (IP) injections of Klebsiella pneumoniae serotype 2 in different doses and this was compared with similar doses given intravenously (IV). A dose dependent survival curve was generated. Subsequently, 10(3) colony forming units (CFU) of bacteria were used in further experiments. Blood, peritoneal fluid and lung tissue were collected at time points up to 72 hours after injection and were cultured for levels of bacteria. Lung weights and myeloperoxidase levels were also measured. RESULTS: Intraperitoneal administration of Klebsiella was uniformly lethal with as few as 10(2) bacteria. Lung weight increased after IP Klebsiella, and all animals became bacteremic within 24 hours correlating with high bacterial levels in the lung. Conversely, most animals (72%) survived IV injection of bacteria, and were able to clear bacteria from the blood and lung. CONCLUSIONS: We found that this model produced no clinically apparent peritonitis after 48 hours, but uniformly resulted in histopathologic changes of pneumonia by 24 hours. Survival time was related to initial dose of Klebsiella and there was a linear correlation between bacterial levels in the blood and lung. This model is reproducible, simple to perform, and the severity is easy to manipulate.

Natural killer cell activation primes macrophages to clear bacterial infection.

Natural killer (NK) cells are major cytokine producers during bacterial sepsis, but their precise role is undefined. This study investigates the effect of NK cell depletion with and without prior activation on macrophage function and bacterial clearance during cecal ligation and puncture. Two different NK cell-depleting antibodies were used: anti-asialo-GM1 (GM1), a nonactivating antibody, and anti-NK1.1 (NK1.1), an NK cell-activating antibody. C57BL/6 mice were NK depleted with either GM1 or NK1.1 by intraperitoneal injection 7 and 3 days before experimentation. Control animals received isotype immunoglobulin G. Depletion was confirmed by flow cytometry. Bacterial levels in peritoneal washout, blood, and liver were determined 4 hours after cecal ligation and puncture. Macrophage activation was measured by phagocytosis ability and by production of nitric oxide and interleukin-6. Depletion with GM1 resulted in significantly higher bacterial levels at 4 hours, whereas depletion with NK1.1 had the opposite effect of significantly decreasing bacterial levels. Macrophage phagocytosis ability was significantly increased in mice depleted with NK1.1 compared with those mice depleted with GM1. We conclude that activation of NK cells improves bacterial clearance by priming macrophages to help clear a subsequent bacterial challenge. Macrophages are less able to clear bacteria when NK cells are depleted without activation. NK cells are therefore important in bacterial clearance through interactions with macrophages.

Natural killer cells participate in bacterial clearance during septic peritonitis through interactions with macrophages.

Natural killer (NK) cells have a well-established role in host defense against viral infections and malignancies. However, their function in bacterial infection and sepsis is poorly defined. We hypothesized that NK cells, as a major producer of interferon-gamma during sepsis, would be important in host defense against bacterial infections. Cecal ligation and puncture (CLP) was performed on Swiss Webster mice depleted of NK cells by pretreatment with anti-asialo GM1 and control mice given immunoglobulin G (IgG) antibody. NK cell-depleted mice had significantly higher anaerobic bacterial counts in the liver and peritoneal lavage fluid, as well as higher aerobic counts in the liver and blood 4 h after CLP. Macrophage phagocytosis, nitric oxide production, and interleukin (IL)-6 levels at 4 h were also decreased in mice depleted of NK cells compared with controls. Greater neutrophil influx into the peritoneum, indicated by higher myeloperoxidase levels, was also seen in NK cell-depleted mice. At 8 and 18 h after CLP, bacterial counts were similar between groups, and overall survival rates were not significantly different. Peritoneal IL-12 levels significantly increased by 18 h in normal mice, but not in NK cell-depleted animals. Our data suggest that NK cells participate in the early local and systemic eradication of bacteria and regulation of IL-12 during polymicrobial sepsis. These effects are likely due to their interactions with macrophages.

Vaccinia virus complement control protein increases early bacterial clearance during experimental peritonitis.

BACKGROUND: Complement is one of the first immunological pathways activated in peritonitis. It functions to initiate and augment the innate immune response. Complement activation has also been shown to contribute to multiple organ failure after sepsis. Vaccinia virus complement control protein (VCP) is an immunomodulatory protein encoded by vaccinia virus and binds complement components C3b and C4b of the complement cascade to inhibit both the classical and alternative pathways of complement activation. This study investigates the effect of complement inhibition by recombinant (r) VCP on bacterial clearance after cecal ligation and puncture (CLP). METHODS: Swiss Webster mice were intravenously given either 20 mg/kg rVCP in 0.2 mL of normal saline, or 0.2 mL of normal saline alone, at the time of CLP. After 4 and 18 h, samples of peritoneal washout, blood, liver, and lung were collected for bacteriology, myeloperoxidase (MPO) assay for neutrophil accumulation, differential cell counts, and interleukin (IL)12 ELISA. Statistical analysis was by Mann-Whitney U test for bacteriology, and analysis of variance (ANOVA) for MPO and IL-12 concentrations. RESULTS: Aerobic and anaerobic bacterial levels were significantly lower at 4 h after treatment with rVCP (p < 0.05) in peritoneal lavage, blood, and liver compared with controls. There were no differences in bacterial levels at 18 h. There were no differences in myeloperoxidase concentrations or in the differential cell counts between the groups at either 4 or 18 h after CLP. IL-12 concentrations in serum or peritoneal washout were also not different. CONCLUSIONS: rVCP enhances early bacterial clearance in mice after CLP, although not through neutrophil recruitment, as MPO concentrations and cell counts were not different. rVCP may, however, increase neutrophil function potentially by prevention of accumulation of complement factors that inhibit leukocytes. Further studies will be needed to elucidate this pathway.

Genetic background determines susceptibility during murine septic peritonitis.

BACKGROUND: The tolerance of mouse strains to cecal ligation and puncture (CLP), a clinically relevant model of sepsis, can vary greatly. We compared the immune response and bacterial eradication during CLP in two mouse strains with different susceptibilities to the lethal effects in an effort to understand alterations in tolerance. MATERIALS AND METHODS: CLP of increasing severity was performed on Swiss Webster mice. Interleukin (IL)-12 levels, bacterial counts, and myeloperoxidase were determined. We then compared the same parameters in Swiss Webster and in BALB/c mice and determined survival for both mouse strains after CLP. RESULTS: Bacterial counts locally and systemically as well as serum IL-12 correlated with the severity of CLP in Swiss Webster mice. Lung myeloperoxidase increased with increasing severity CLP, while peritoneal myeloperoxidase decreased. Following CLP, one-half of the Swiss Webster mice survived versus none of the BALB/c mice. Despite worsened survival, BALB/c mice had lower bacterial counts and similar IL-12 levels compared to Swiss Webster mice. Myeloperoxidase and IL-6 levels were similar between experimental groups. CONCLUSIONS: Swiss Webster and BALB/c mice have significantly different susceptibilities to the lethal effects of CLP, and this difference may be related to IL-12 responsiveness.