Intraperitoneal host cellular responses and in vivo killing of Bacteroides fragilis in a bacterial containment chamber.

A bacterial containment chamber was used to evaluate the peritoneal cellular response to Bacteroides fragilis during intraperitoneal challenge. This containment system was also used to determine the fate of bacteria within the peritoneal cavities of animals immunized, either actively or through adoptive transfer of cells or cell lysates, with the capsular polysaccharide of B. fragilis. This system demonstrated that the dominant cell types in the peritoneal cavities within 48 h of implantation of the containment chambers containing B. fragilis were neutrophils and macrophages. However, the early cellular response in immunized animals included an increase in the lymphocyte population within 4 h of challenge which was not detected in naive animals. In immunized animals, a later dramatic increase in the lymphocyte population at approximately 4 to 6 days following implantation of the containment chambers occurred. This increase in the lymphocyte population in immunized animals coincided with a decline in the viable bacterial counts within the chambers from 10(8) to 10(9) CFU/ml to less than 10(2) CFU/ml. A similar decline was not seen in naive animals challenged in the same manner. Killing of B. fragilis within containment chambers occurred when spleen cells, T cells, or lysates of T cells from actively immunized animals were passively transferred to naive recipient animals. It was shown that the factor responsible for bacterial killing was not antibody mediated, since bacteria contained within dialysis sacs with an exclusion of 50 kilodaltons were still killed in this model. Moreover, removal of T cells from adoptively transferred cell populations before transfer abrogated the decline in viable bacterial populations. The postulated mechanisms by which this bacterial killing occurred are discussed.

Ontogenic development of autoantibody repertoires in spleen and peritoneal cavity of normal mice: examples of T cell-dependent and -independent reactivities.

The ontogenic development of B cell clonal precursors (BCP) reactive to bromelain-treated, syngeneic erythrocytes (BrMRC) and to single-stranded DNA has been studied by limiting dilution of both spleen and peritoneal cells. It was found that the frequency of anti-BrMRC BCP in the spleen is very low up to 4 weeks of age and slowly increases thereafter, to reach adult levels by 6-10 weeks. In the peritoneal cavity, no such BCP can be found before 2 weeks, but they occur at a very high frequency already by 3 weeks of age. Injection of adult, normal syngeneic T cells at birth has no apparent effect on the representation of anti-BrMRC BCP in the peritoneal cavity, but brings these to adult levels or even higher in the spleen already at 3 weeks of age. Accordingly, adult athymic (nude) mice contain normal frequencies of BrMRC-specific BCP in the peritoneal cavity but are devoid of such clones in the spleen. In contrast, the frequency of anti-DNA BCP is very high throughout postnatal development in both spleen and peritoneal cavity, of normal and athymic mice, in both resting and naturally activated splenic B cell compartments, and it is independent of T cell transfers into nude animals. These results indicate the role of T cells in the establishment of some clonal specificities in the adult, splenic autoreactive B cell repertoire.

Cloricromene improves survival rate and peritoneal macrophage function in splanchnic artery occlusion shock in rats.

Splanchnic artery occlusion (SAO) shock, induced by a transient occlusion of splanchnic arteries for 45 min, was performed in male rats, treated with vehicle or cloricromene, a coumarin derivative, 15 min before surgery. Survival rate, plasma levels of myocardial depressant factor (MDF), macrophage phagocytosis and killing of Candida albicans, and thromboxane B2 (TxB2) synthesis by peritoneal macrophages were evaluated. Of the SAO-shocked animals, 10% survived for 6 hr after the release of the occlusion of the splanchnic arteries, whereas none of the sham-shocked rats died. Peritoneal macrophages of shocked animals exhibited decreased phagocytosis (24.7 +/- 2.7%) and killing (8.0 +/- 2.1%) and increased TxB2 levels (3.23 +/- 0.27 ng/ml) with respect to those collected from sham-shocked animals (phagocytosis 48.8 +/- 3.0%; killing 16.5 +/- 2.4%; TxB2 0.30 +/- 0.18 ng/ml). MDF was also increased (114.3 +/- 21.5 U/ml) compared with sham-shocked animals (31.5 +/- 3.7 U/ml). Cloricromene, given intravenously (i.v.) at doses of 1, 2, and 4 mg/kg, significantly increased survival rate and lowered MDF in shocked rats. Lower doses (0.25 and 0.5 mg/kg/i.v.) were without effect. Doses that were able to reduce mortality partially reverted shock-induced macrophage impairment of phagocytosis, killing of C. albicans, and TxB2 synthesis. In addition, cloricromene (5, 10, and 25 microM) added in vitro to peritoneal macrophages, collected from shocked rats, significantly enhanced their phagocytic activity depressed by shock.

Elevated levels of GM-CSF and IL-1 in the serum, peritoneal and pleural cavities of GM-CSF transgenic mice.

Levels of granulocyte-macrophage colony-stimulating factor (GM-CSF) in the peritoneal and pleural cavity fluid of two lines of GM-CSF transgenic mice were abnormally high (up to 120,000 U/ml), often exceeded the elevated serum GM-CSF levels in these mice and correlated with the increased number of macrophages present. In the peritoneal fluid, the only significant elevations of IL-1 levels were seen in moribund male-line transgenic mice. In contrast, IL-1 levels in pleural cavity fluid of male-line transgenic mice were clearly higher than those in littermate control mice or female-line transgenic mice. In male-line transgenic mice, IL-1 levels in both peritoneal and pleural cavities correlated with local macrophage numbers. Endotoxin was detectable in the peritoneal cavity fluid from some mice of all types but did not correlate with elevated IL-1 levels. No correlation was observed between levels of GM-CSF or IL-1 in the peritoneal cavity and the development of fibrotic nodules in the peritoneal cavity or gut congestion, two lesions common in male-line GM-CSF transgenic mice. The data suggest that the elevated levels of IL-1 in GM-CSF transgenic mice may be the consequence of stimulation by GM-CSF of IL-1 production by the elevated numbers of macrophages in these mice.

Implication of activation of intraperitoneal macrophages with bio-degradable microspheres.

To estimate the possibility that biological degradable starch microspheres (DMS) activate abdominal or intraperitoneal macrophages (IMP), two sizes of DMS (Spherex, Pharmacia, Sweden) were injected into the peritoneum of the ICR mice of 4 to 8 weeks of age. Three days after the injection, peritoneal fluid was collected and incubated for one hour at 37 degrees C under 5% CO2. The cells which adhered to the petri dish were IMP, to which DMS was added for 18 hrs. The cultured IMP were observed by scanning electron microscope (SEM) and the ratio of the active type to the total number of IMP was counted as an index of the effect of DMS to IMP. The activation effect of DMS on the incubated IMP was significant in the group which was cultured with 2 microns DMS after the 45 microns DMS injection. That indicated the possible DMS function as a potential IMP activating factor (MAF).

Spontaneous cytotoxicity of macrophages against pancreatic islet cells.

Activated peritoneal macrophages were found to lyse syngeneic [3H]leucine-labeled pancreatic islet cells or rat insulinoma cells after 15 h of coculture at 37 degrees C. Lysis was verified by electron microscopic analysis. Islet cell lysis was dependent on the T:E ratio and was comparable with P815 and L929 tumor cells used as targets. The cytotoxic activity was localized in the adherent fraction of Corynebacterium parvum activated peritoneal cells and was destroyed by incubation of cells with macrophage-toxic silica particles. Syngeneic thyrocytes and hepatocytes were found to be resistant to the cytolytic action of activated macrophages. It has been shown previously that macrophages contribute to pancreatic islet inflammation. The present in vitro analysis demonstrates that macrophages can function as effector cells in islet destruction.

Enhancement of immunoglobulin G responses in mice against hepatitis B virus surface antigen, influenza virus hemagglutinin vaccine, and tetanus toxoid by 6-O-acylated muramyl dipeptides.

The adjuvant activity of chemically synthesized 6-O-acylated muramyl dipeptides (MDP) was tested in aqueous form. The activity was assessed by determining immunoglobulin G (IgG) titers in sera of mice immunized with hepatitis B virus surface antigen, influenza virus hemagglutinin (HA) vaccine, or tetanus toxoid with an enzyme-linked immunosorbent assay. Administration of 6-O-acyl-MDP analogs with antigens induced marked enhancement of primary and secondary IgG antibody responses and maintained high antibody levels for at least 7 weeks. Among the analogs tested, an MDP methyl ester carrying a 6-O-3-hexadecanoyl-oxytetradecanoyl group (compound 309) exhibited the most intensive adjuvant activity. Its activity was stronger than that of 6-O-2-tetradecylhexadecanoyl (B3O)-MDP used as a positive control. However, accumulation of peritoneal cells and activation of peritoneal macrophages by compound 309 was weaker than that by 6-O-B30-MDP, suggesting that 309 as an immunoadjuvant is more suitable for vaccination in terms of its stronger enhancement of antibody formation and lower induction of inflammatory response than 6-O-B30-MDP.

Neutrophil-macrophage cooperation in the host defence against mycobacterial infections.

CD-1 mice inoculated intraperitoneally with Mycobacterium avium, M. bovis, M. microti or M. kansasii showed a persistent peritoneal granulocytosis (above 10(6) cells, i.e. more than 15% of total cells) throughout the 3 month period of infection studied. By contrast, in mice inoculated with the non-pathogenic M. aurum or with heat-killed M. avium the number of granulocytes decreased progressively after the first 15 days. No mycobacteria were found in granulocytes except in the first 2 days of infection. The mycobacteria-induced chronic granulocytosis was accompanied by phagocytosis of granulocytes by macrophages. Throughout the 3 months of infection, macrophages were found to contain intracellular lactoferrin. Macrophages with lactoferrin were also found in subcutaneous infection caused by M. marinum and in systemic infection caused by M. avium or M. kansasii. The in vitro activity of mouse peritoneal macrophages against M. avium and M. microti was increased after ingestion of granulocyte material by macrophages. These results lead us to propose that granulocytes participate in the host response to mycobacterial infections, not as phagocytes but rather through an indirect mechanism, as a source for the macrophages of molecules involved in antimicrobial mechanisms (e.g., lactoferrin and myeloperoxidase) lacking in the mature macrophage.

Immune regulation of the L5178Y murine tumor-dormant state: induction of cell- and soluble factor-mediated inhibition of tumor cell growth by tumor necrosis factor and gamma-interferon.

Small concentrations of recombinant murine tumor necrosis factor (rMuTNF) synergized with recombinant murine gamma-interferon (rMuIFN-gamma) to inhibit tumor cell growth in peritoneal cell (PC) cultures from tumor-dormant mice. A soluble inhibitor of tumor cell growth was produced in the rMuTNF-treated PC cultures, but there was no synergistic enhancement of production of this inhibitor by the addition of rMuIFN-gamma. Treatment of both plastic-adherent PC and nonadherent PC cultures which contained L5178Y cells with rMuTNF + rMuIFN-gamma resulted in inhibition of tumor cell growth. However, in the absence of L5178Y cells, rMuTNF + rMuIFN-gamma induced antitumor cytotoxic activity in the plastic-adherent but not in the plastic nonadherent PC. These results indicate that the antitumor activity of rMuTNF + rMuIFN-gamma in PC from tumor-dormant mice involves both cell- and soluble factor-mediated mechanisms.

CA 125 in peritoneal washings and fluid: correlation with plasma CA 125 and peritoneal cytology.

Sixty-four women with ovarian cancer and 46 controls with benign gynecologic conditions underwent cytologic and CA 125 evaluations of peritoneal fluid or peritoneal washings at laparotomy. These parameters were correlated with preoperative disease status, intraoperative findings, and preoperative plasma CA 125 levels to determine their value in assessing occult disease. No false-positive cytology reports were observed. Cytology specimens were positive in 20 of 23 patients (87%) with clinical evidence of disease who had peritoneal fluid present, but in only 18 of 29 (62%) of similar patients with no peritoneal fluid present (P less than .05). CA 125 values were elevated in 16 of 23 (69.5%) and 15 of 29 (52%) of these samples, respectively (P greater than .05). Levels of CA 125 in peritoneal fluid washings correlated poorly with the presence of obvious intraperitoneal cancer and had questionable reliability when used to evaluate patients clinically free of disease. Positive peritoneal cytology reflected the presence of ovarian cancer, but its absence did not mean that an objective response to chemotherapy treatment had occurred. Disease status correlated best with physical examination and circulating levels of CA 125.

In vivo interaction of bilirubin with the cells of the immune system in mice: an ultrastructural electronmicroscopic study.

Morphological evidence revealed the presence of round-shaped particles after application of bilirubin. The possibility of polymerization of bilirubin-albumin complexes is discussed.

Expression of VH11-gene family in hybridoma collections from peritoneum and spleen: differential correlation with BrMRBC reactivity.

Hybridoma collections from spleen or peritoneal cells of newborn or adult individuals were screened by RNA hybridization for expression of the VH11-gene family using a V-region probe VCP12, which encodes anti-BrMRBC antibodies. No VH11 expression was observed in hybridomas derived from newborn spleen cells in either BALB/c, NZB or (CBA/N x BALB/c) F1 mice (0/93). Adult NZB and BALB/c spleen cell collections contained only one hybridoma expressing VH11 (1/242). Interestingly, however, the VH11-positive hybridoma showed no anti-BrMRBC reactivity, while one anti-BrMRBC clone in the same collection expressed a Q52 VH gene. In contrast, hybridomas derived from peritoneal cells showed an absolute correlation between expression of VH11 genes and anti-BrMRBC reactivity (15/32). The high expression in the peritoneal cavity of such cells is likely the result of local positive selection.

Regulation of insulin and interleukin-1 release after Propionibacterium acnes-induced macrophage activation in mice.

The administration of a potent activator of macrophages (M phi), Propionibacterium acnes, in nondiabetic mice was associated with the release of significant amounts of interleukin-1 (IL-1) in the peritoneal cavity and plasma within 4 hours after treatment. Shortly before IL-1 peaks were observed, the levels of pancreatic insulin, [3H]leucine-proinsulin, and insulin/total protein ratio were elevated, and followed by a transient but marked hyperinsulinemia at 4 hours after treatment. A single dose of recombinant murine IL-1 in mice was also associated with a 2- to 9-fold increase in the levels of insulin in the pancreas and plasma at 4 hours after treatment. During the period of observation after the administration of P. acnes, plasma glucose levels in treated mice were significantly less than in parallel controls. Mild hypoglycemia was observed at 7 to 10 days posttreatment. Although circulating IL-1-like activity could not be detected in plasma 1 to 10 days after P. acnes treatment, this activity was measured in activated peritoneal and liver M phi. IL-1-like activity (specific activity: 276 units/mg protein) was detected in plasma, after it was chromatographed on a Sephadex G-150 column to remove proteins with higher molecular weight. Peritoneal and liver M phi from P. acnes mice were also able to elaborate significant amounts of IL-1-like activity in their supernatants with or without Escherichia coli lipopolysaccharide. At the same time, total protein synthesis and insulin content in the pancreas in P. acnes mice were significantly lower than the parallel control (p less than 0.01). These results suggest that P. acnes-induced M phi activation in mice was associated with the modulation of insulin release and glucose homeostasis which may be attributed to the accumulation and release of IL-1 by activated M phi.

Effects of rat and human intestinal lamina propria cells on viability and muscle establishment of Trichinella spiralis newborn larvae.

Although eosinophils and other inflammatory cells from the circulation and peritoneal cavity can damage Trichinella spiralis newborn larvae (NBL) in vitro, the cytotoxic potential of cells from the intestinal lamina propria, a site that may be the first line of defense against NBL migration, is unknown. Accordingly, we examined the interaction between NBL and isolated intestinal lamina propria cells (ILPC), including an enriched eosinophil population, from rats and humans. Rat ILPC killed NBL in vitro only after a prolonged incubation of 6 days. However they strongly adhered to NBL after only 4 hr incubation and prevented muscle establishment of NBL injected intravenously. Human ILPC showed similar adherence as rat ILPC but no killing was seen at the incubation time tested (36 hr).

Inflammatory and immunological responses to murine cytomegalovirus in resistant CBA mice.

The resistance of CBA mice to MCMV is associated with the resistance of H-2k cells to infection in vitro, high NK and virus-specific DTH responses, and minimal accumulation of cytostatic peritoneal macrophages. This study investigates the functional capacity of lymphoid cells from infected CBA mice, using this strain as a model for successful control of CMV. Splenic viral replication was high 1-3 days p.i. and cell numbers were depressed, but T and B cells frequencies were maintained. The remaining spleen cells were hyporesponsive in culture and accessory cell function was marginally deficient. By 7 days p.i., virus titres declined, responsiveness increased and the residual defect was associated with cytostatic macrophages. The lymph nodes did not atrophy, exhibited low levels of viral replication and proliferative capacity was retained. The beige mutation did not affect the local response to intraperitoneal infection, but spleen cell numbers and responsiveness declined progressively. The results suggest the spleen may contribute to CMV disease by the replication of virus in susceptible cells until the NK response reduces virus titres and hence limits acute virus-induced immunosuppression. Macrophage-mediated suppression persisted in the spleen during recovery so clonal expansion of protective virus-primed T cells may occur predominantly in the lymph nodes.

Induction of nonspecific killer cells by delayed-type hypersensitivity against soluble protein antigens in murine peritoneal cavities.

We induced nonspecific killer cells in the local site of delayed-type hypersensitivity against keyhole limpet hemocyanin or ovalbumin. Delayed-type hypersensitivity was induced in the peritoneal cavities of mice, and peritoneal exudate cells (PEC) were collected. These PEC were found to have killer activity toward SP2 and YAC-1 cells (target cells susceptible to natural killer cells) by 4-h 51Cr-release assays. The induction of killer activity in PEC was observed in parallel with the eliciting of delayed-type hypersensitivity in the peritoneal cavity, in which the killer activity was maximum 24-48 h after the antigen challenge, but was not induced in nu/nu mice and was induced in an antigen-specific way. These killer cells did not adhere to nylon wool and had Thy1 and asialo-GM1 antigens on their surfaces. Their precursor cells were also asialo-GM1-positive. These findings indicate that the killer cells probably belong to the NK cell lineage. Results of tumor challenge experiments showed that these killer cells had an antitumor effect in vivo as well as in vitro.

A major peritoneal reservoir of precursors for intestinal IgA plasma cells.

Studies presented examine the origin of IgA plasma cells in B lineage chimeric mice constructed by reconstituting lethally irradiated mice with a mixture of syngeneic bone marrow cells and peritoneal cells from Ig heavy chain allotype congenic donors. In these mice, essentially all B cells in spleen and Peyer's patches are derived from the bone marrow donor; however Ly-1 B lineage cells which have been mainly detected in the peritoneum are derived from the peritoneal cell donor. Surprisingly, roughly half of the IgA plasma cells in the lamina propria of the gut are also derived from the peritoneal cell donor, suggesting an important role for peritoneally-derived B cells in the mucosal immune response.

Eosinophilia in mouse regulated by multiple factors following repeated stimulations with ascaris eggs into the peritoneal cavity.

Repeated injections of viable eggs of Ascaris suum into the peritoneal cavity of BALB/c mice induced marked peritoneal eosinophilia (9.3 X 10(6) cells/mouse, 52.8%) without infection. Extracted antigen (Asc) and eggs that had been killed also were able to induce marked eosinophilia; however, their levels were significantly lower (p less than 0.001). This experimental system was able to demonstrate how various multiple factors cumulatively affected the degree of peritoneal eosinophilia. These factors included: mechanical stimulation, caused by the repeated peritoneal lavages (4.4 X 10(5) cells/mouse), against a background level of peritoneal eosinophils of 4.5 X 10(4) cells/mouse; T-cell-independent stimulation, observed when the viable eggs were injected repeatedly into athymic nude mice (nu/nu; 1.6 X 10(6) cells/mouse), while on the other hand, neither eggs that had been killed nor Asc displayed any T-cell-independent augmentation of eosinophilia; and, T-cell-dependent augmentation of peritoneal eosinophilia, observed in heterozygous athymic mice (nu/+), in thymocyte-reconstituted nu/nu, and in normal BALB/c mice. Egg culture supernatant could not act on bone marrow cells to support eosinophil colony formation. In conclusion, only viable eggs were considered prime inducers, consisting of T-independent and T-dependent parts exerting cumulative effects.