Cytokines induced in vitro by Mycoplasma mycoides ssp. mycoides, large colony type

Septicemic disease in goats and sheep caused by the large colony type of Mycoplasma mycoides ssp. mycoides has clinical and pathological features in common with septic endotoxemia. We studied the ability of the mycoplasma to induce mediators of biological responses to endotoxin, such as TNF alpha, IL-1 alpha, IL-6 and nitric oxide. Heat-killed suspensions of mycoplasmas in a concentration of 25 micrograms protein ml-1 induced all mediators in macrophages from peritoneal cavity and bone marrow of both C3H/HeN (LPS responsive) and C3H/HeJ (LPS low-responsive) mice. Lipopolysaccharide (LPS) in a concentration of 100 ng ml-1 induced the mediators in C3H/HeN derived macrophages, only. Simultaneous stimulation of macrophages with interferon-gamma enhanced the secretion of nitric oxide (measured as nitrite) but not the cytokines. We conclude that heat-killed Mycoplasma mycoides ssp. mycoides, large colony type, has cytokine inductive activity similar to bacterial endotoxin, but with an induction mechanism different from LPS.

Characterization and purification of a mycoplasma membrane-derived macrophage-activating factor

A highly hydrophobic component derived from the membrane of Mycoplasma capricolum has been characterized, purified and assessed for its ability to activate macrophages to tumor cytotoxicity. Initially, crude membranes were evaluated for their solubility in a wide range of solvents. Despite differential solubility in the various solvents, the mycoplasma membranes retained their ability to potentiate macrophage tumor cytotoxicity. Mycoplasma membranes were further characterized by appraising their macrophage-activating ability subsequent to various chemical treatments: cleavage of ester and thioester bonds, oxidation of vicinal hydroxyl groups, and exposure to a broad range of pH. Only strong alkaline treatment (pH > 12) caused a reduction in mycoplasma membrane activity; all other chemical treatments were inconsequential. With potential therapeutic applications in mind, mycoplasma membranes were subjected to various physical treatments including heating, freezing/thawing, sonication, lyophilization and storage. The ability of the membranes to induce macrophage activation was stably maintained following all these treatments. Purification of membranes was initiated by a chloroform/methanol lipid extraction. Macrophage-activating ability was found predominantly in the interphase. Proteolytic cleavage with trypsin increased specific activity at least sixfold. Trypsinized fractions were solubilized in 2-chloroethanol and gel filtration was performed on a hydroxylated Sephadex LH-60 column. The active fraction from this column had a further tenfold increase in specific activity. Subsequent rounds of reverse-phase HPLC on this fraction yielded three to four peaks absorbing at 280 nm, of which only one had macrophage-activating ability.

Closed head injury triggers early production of TNF alpha and IL-6 by brain tissue

In a model of closed head injury (CHI) in the rat we have shown the activation of phospholipase A2 and the production of eicosanoids after injury: at 15 min, mainly 5-hydroxyeicosatetraenoic acid (5-HETE), and at 24 h, mainly prostaglandin E2. The present study was designed to test whether CHI can also trigger the production of cytokines in the brain. CHI was induced in ether-anesthesized rats by a weight-drop device falling over the exposed skull covering the left hemisphere, 1-2 mm lateral to the midline in the midcoronal plane. In the posttraumatic period (1-24 h), the rats were decapitated, cortical tissue from the injured zone of the contused and contralateral hemispheres was removed and sonicated, and cytokine activity was assessed. Whereas no tumor necrosis factor alpha (TNF alpha) activity was found in normal brain tissue, it was detectable in the contused hemisphere (approximately of 72 +/- 50 pg/mg protein) as early as 1 h post-CHI. TNF alpha levels increased at 2 h, peaked at 4 h, (approximately of 609 +/- 540 pg/mg protein), and declined thereafter. At parallel intervals, only low levels of TNF alpha were detected in the contralateral hemisphere. In normal brain, interleukin-6 (IL-6) was nondetectable. Following CHI, high levels of IL-6 were present, although their accumulation lagged behind that of TNF alpha by 2-4 h, peaking at 8 h (62 +/- 31 ng/mg protein). We suggest that the rapid production of TNF alpha and IL-6 following CHI is a local inflammatory response of brain tissue to primary insult.

Mycoplasma triggering of nitric oxide production by central nervous system glial cells and its inhibition by glucocorticoids

The same cytokines that have been implicated in the pathology of central nervous system (CNS) inflammatory diseases and demyelinating diseases are also associated with the induction of nitric oxide (NO) production by macrophages and other somatic cells. Recently we have showed that mycoplasma can trigger the production of tumor necrosis factor (TNF)alpha and eicosanoids in rat astrocytes. In the present study, the effect of mycoplasma on NO production in rat glial cells was assessed. The addition of 10 micrograms/ml of membranes derived from M. capricolum (sheep isolate), M. fermentans (human isolate), or lipopolysaccharide (LPS) led to a 15- to 20-fold increase in NO production. The glucocorticoids dexamethasone and corticosterone, but not progesterone, markedly inhibited NO production. The addition of glucocorticoid prior or conjointly with the activator prevented large amounts of NO from being formed. Even when glucocorticoids were added 5 or 24 h after activation, effective inhibition of NO production was obtained. Thus, it is likely that glucocorticoids exert some of their ameliorating effects in neurological diseases by reducing the production of NO, cytokines and prostaglandins in the CNS.

Trophoblasts protect the inner cell mass from macrophage destruction

Several mechanisms have been suggested to account for the survival of the semiallogeneic fetus in the maternal uterus. However, no data are available to explain how the blastocyst resists the high number of macrophages in the uterus at the time of implantation. The present study examines the in vitro development of murine 3.5-day-old syngeneic or semiallogeneic blastocysts in the presence of nonactivated or lipopolysaccharide (LPS)-activated macrophages. It was found that the in vitro development of blastocysts was undisturbed by the presence of nonactivated or LPS-activated macrophages. The outgrowing trophoblasts were not only nonadhesive to the macrophages but also repelled them actively, thus preventing them from reaching the inner cell mass (ICM). Removing the zona pellucida by use of pronase or killing the ICM by irradiation did not alter the repulsion of macrophages by the trophoblasts. On the other hand, removal of the trophectoderm by antibody and complement treatment rendered the macrophages adhesive and destructive to the ICM. Four of 15 ICM (27%) were destroyed by nonactivated macrophages, and all of the ICM (15/15) were destroyed by LPS-activated macrophages. It is noteworthy that the addition of colchicine, cytochalasin B, proteinase inhibitors, anti-transforming growth factor-beta (TGF beta) antibodies, and indomethacin had no effect on the repulsion of macrophages by the trophoblasts. Therefore, it seems that microtubular proteins, microfilaments, extracellular matrix-degrading enzymes, TGF beta, and prostaglandins are not involved in the repulsion process. These results indicate that trophoblasts protect the ICM from the destructive action of macrophages by a repulsion mechanism of an as yet unknown nature.

Stimulation of tumor necrosis factor-alpha production by mycoplasmas and inhibition by dexamethasone in cultured astrocytes

Elevated levels of tumor necrosis factor-alpha (TNF alpha) and other cytokines and eicosanoids in the central nervous system (CNS) have been noted in several human neurologic diseases, including multiple sclerosis and AIDS dementia. Recently it was shown that glial cells, especially astrocytes, are a major source of cytokines and eicosanoids. In the present study we have shown that astrocytes derived from fetal rat brain triggered by mycoplasmas produce TNF alpha and prostaglandin E2 (PGE2). Addition of mycoplasma (Mycoplasma capricolum isolated from sheep and M. fermentans KL-4 from human) at a concentration of 1-50 micrograms protein/ml (2 x 10(7)-10(9) colony forming units/ml), as well as lipopolysaccharide (5 micrograms/ml), led to a 200-500-fold increase in TNF alpha and a 2.5-4.5-fold increase in PGE2 production. Preincubation of the cells with the synthetic glucocorticoid, dexamethasone (2 x 10(-5)-2 x 10(-8) M), as well as with the natural hormone, corticosterone, markedly inhibited the secretion of both TNF alpha and PGE2. Thus, mycoplasmas can be added to the wide variety of agents that stimulate glial cells to produce cytokines and eicosanoids, and may contribute to various CNS pathological manifestations. In addition, the ability of glucocorticoids to inhibit particularly the stimulated productions of TNF alpha and PGE2 may explain at least in part the therapeutic benefit of these agents in CNS inflammation and demyelination.

Polymyxin B reduces total parenteral nutrition-associated hepatic steatosis by its antibacterial activity and by blocking deleterious effects of lipopolysaccharide

Overgrowth of Gram-negative bacteria as a result of total parenteral nutrition (TPN) and bowel rest could be responsible for the release of a variety of hepatotoxic substances such as endotoxin or tumor necrosis factor (TNF) and the ensuing TPN-associated liver function derangements. Polymyxin B is an effective antimicrobial agent as well as a blocking agent for endotoxin (lipopolysaccharide) activity and TNF production. In the present study we compared the oral and intravenous effects of polymyxin in rats receiving TPN in an attempt to define these two possible mechanisms of action of polymyxin on TPN-associated hepatic steatosis. Both oral, as well as intravenous polymyxin B, significantly reduced total hepatic fat and triglyceride accumulation in TPN rats, more so in the intravenous group exhibiting close to control levels. Both polymyxin-treated groups exhibited significantly lower Gram-negative bacterial counts in the cecum, with the oral group exhibiting a lower count than the IV group. The spontaneous production of TNF by peritoneal macrophages was markedly increased in rats receiving TPN and very close to being undetected in both groups receiving TPN and polymyxin. We believe polymyxin B protects the liver during TPN by both its antimicrobial effect which prevents overgrowth of gut Gram-negative bacteria and the subsequent translocation of endotoxin, and by its specific antilipopolysaccharide activity which, in the present study, completely abolished hepatic steatosis and TNF production during TPN.

Mycoplasma fermentans (incognitus strain) induces TNF alpha and IL-1 production by human monocytes and murine macrophages

We have demonstrated that Mycoplasma fermentans (incognitus strain), as well as M. fermentans KL4, PG 18 and IM 1 strains have the ability to activate human peripheral blood monocytes and murine macrophages of two inbred strains to secrete a high level of tumor necrosis factor alpha (TNF alpha) in a dose-dependent manner. Secretion of interleukin-1 (IL-1) was also stimulated following the incubation of human monocytes with the organism. We suggest that cytokine secretion following infection with M. fermentans (incognitus strain) that was detected in AIDS patients may contribute to the pathological manifestations, including cachexia, in this disease.

The unique killing of embryo-derived teratocarcinoma cells by nonactivated murine macrophages is not due to a lack of H-2 antigen expression

It is well documented that activated macrophages, but not nonactivated ones, kill tumor cells in vitro without damaging normal cells. We, however, have previously shown that embryo-derived teratocarcinoma cells (F9, P19, PCC4) are efficiently killed by nonactivated macrophages as well as by activated ones. Whereas other tumor cells are killed extracellularly by macrophages, we found that F9 teratocarcinoma cells are phagocytosed alive by macrophages and subsequently killed intracellularly by a process dependent on intact lysosomal function. Neither the H-2 antigens nor the mRNAs for the alpha-chain and beta 2-microglobulin are detectable in embryo-derived teratocarcinoma cells. An obvious explanation for this unique killing is that the nonactivated macrophages recognize and kill these cells due to their lack of class I MHC antigen expression, assuming that class I MHC gene products on the target cells switch off the cytolytic machinery of nonactivated macrophages. Our present findings demonstrate that there is no correlation between H-2 antigen expression on tumor cells and their susceptibility to killing by macrophages. Retinoic acid-differentiated F9 cells and P19 cells expressing H-2 antigen after exposure to MAF (IFN-gamma) were sensitive to the killing by nonactivated macrophages. Hybrids that arose from fusion of P19 teratocarcinoma cells with embryonal normal fibroblasts (C57BL/6), which displayed the morphology of embryonal carcinoma stem cells and expressed H-2 antigens, were also sensitive to the killing by nonactivated macrophages. On the other hand, the H-2-negative testicular 402AX teratocarcinoma cells and K1735P melanoma cells were both resistant to the killing by nonactivated macrophages. We concluded that the unique killing of embryo-derived teratocarcinoma cells by nonactivated murine macrophages is not related to a lack of H-2 antigen expression.

Radiation leukemia virus (RadLV)-induced leukemogenesis is associated with an increased number and activity of thymic macrophages

The radiation leukemia virus (RadLV) is a chronic leukemia retrovirus that induces thymic lymphomas in C57BL/6 mice after a latency of 3 to 6 months. During the pre-leukemic (PL) period, the number of thymic macrophages gradually increased up to 100 fold. Of the cells in a RadLV-induced lymphoma, 0.3% were large macrophages packed with infected lymphoma cells. These thymic lymphoma macrophages (TLM) also ingested RadLV-induced lymphoma cells in vitro. Cultured RadLV-induced lymphoma lines could activate and fix C3 fragments through the alternative complement pathway (ACP). C3-bound lymphoma cells elicited an oxidative burst (OB) response in TLM but not in bone-marrow macrophages (BMM). However, IL4 treatment of BMM rendered them capable of responding with an OB following triggering by C3-opsonized cells. Thymic macrophages (TM) responded moderately with OB to C3-opsonized cells and this response was elevated if the TMs were treated by rIL4. The OB reaction of the TLMs could be partially inhibited by anti-LFA-I or anti-MALA-2 antibodies, and was completely inhibited by anti-CR3 antibodies. These results suggest that IL4 can prime macrophages for triggering an OB reaction and that the interaction between C3-opsonized cells and IL4-primed macrophages is mediated primarily through CR3.

Thymocyte maturation following interaction with thymus-derived macrophages

Murine C57BL/6 thymocytes were cultivated together with syngeneic thymus-derived macrophages (TDM phi) for up to 96 hr to determine whether TDM phi participate in thymocyte maturation. The expression level of H-2b and Thy-1.2 antigens served as thymocyte differentiation surface markers as analyzed by flow cytometry. Indirect immunofluorescent staining profiles of the thymocytes demonstrate a dramatic increase in H-2b expression and a profound decrease in Thy-1.2 expression during cultivation with TDM phi. A similar phenomenon was observed when enriched populations of immature thymocytes were cocultivated with TDM phi. These changes were not observed when thymocytes were cultivated alone or with trypsin-treated TDM phi; neither were they observed when cortisone-resistant thymocytes manifesting mature characteristics were cultivated together with TDM phi. These findings suggest that interaction of thymocytes with TDM phi, involving binding and engulfment, results in the appearance of mature thymocyte subsets.

Partial biochemical characterization of spiroplasma membrane component inducing tumor necrosis factor alpha

We have recently found that membranes of Spiroplasma spp. strain MQ-1 (hereafter referred to as MQ-1) induce both tumor necrosis factor alpha (TNF alpha) secretion by bone marrow macrophages and blast transformation of lymphocytes via a mechanism different from that operated by bacterial lipopolysaccharide (LPS). This report presents evidence indicating that the MQ-1-derived membrane component(s) which activates bone marrow macrophages to secrete TNF alpha is, at least in part, protein. This conclusion is supported by our findings that TNF alpha secretion was reduced following exposure of MQ-1 membranes to elevated temperatures, extreme acidic pH treatment and incubation with protease K or pronase. Furthermore, following lipid extraction of MQ-1 membranes, most of both induction of TNF alpha secretion and blast transformation activities appeared in the 'protein' fraction. When membranes were chromatographed on a phenyl-Sepharose column, two major peaks were obtained, one containing most of the TNF alpha induction activity and the other the mitogenic activity. Neither peak coeluted with the peak of bulk membrane lipids. The possibility that the spiroplasma membrane component inducing TNF alpha secretion is acylated protein is discussed.

In vitro induction of tumor necrosis factor alpha, tumor cytolysis, and blast transformation by Spiroplasma membranes

Membranes of Spiroplasma sp. strain MQ-1 (hereafter referred to as MQ-1) were potent inducers of tumor necrosis factor alpha (TNF alpha) secretion and of blast transformation. Specific anti-recombinant murine TNF alpha antibodies markedly inhibited macrophage-mediated tumor cytolysis of A9 fibrosarcoma target cells following activation by MQ-1 membranes. Thus, TNF alpha plays a major role in mediation of tumor cytolysis induced by MQ-1 membranes, which is similar to its role in lipopolysaccharide (LPS)-induced tumor cytolysis. Two findings, however, suggested that the mechanism of macrophage activation by MQ-1 membranes differs from that by LPS: (a) macrophages, taken from C3H/HeJ mice showing a low responsiveness to LPS, were activated by MQ-1 membranes to enhanced TNF alpha secretion, resulting in a high-level tumor cytolysis compared with the negligible tumor cytolysis induced by LPS; and (b) MQ-1 membranes and LPS synergized to highly augment TNF alpha secretion by macrophages of C57BL/6 mice. MQ-1 membranes were capable of inducing blast transformation of murine lymphocytes as well. In addition, they activated human monocytes to secrete high levels of TNF alpha. Further studies need to be carried out using in vivo models to evaluate the therapeutic potential of MQ-1 membranes in the treatment of malignant diseases.

Characterization and localization of human placental mononuclear phagocytes by monoclonal antibodies and other cell markers

Human first-trimester placental macrophages were identified in frozen tissue sections and following cultivation as single-cell suspensions. In placental sections, the placental macrophages were shown to be the main cell type of the stromal zone. They clearly expressed macrophage-specific antigen, identified by EBM 11 monoclonal antibody; all were positive for nonspecific esterase (NSE), and 40-60% of them expressed the HLA-DR antigen. These cells were negative to anti-alpha-human chorionic gonadotropin (HCG) monoclonal antibody. After preparation of the placental single-cell suspension, the macrophages comprised about 15% of the total cell population. This fact was established by the following findings: 15% of the placental cell suspension expressed the antigen recognized by the OKM5 antibody, 16% expressed the EBM 11 antigen, 13% expressed the HLA-DR antigen, 5.5% were positive for NSE, and 24% phagocytosed Staphylococcus albus. Thirty percent of the cells were positive to anti-alpha-HCG antibody. After 48 h of cultivation in defined tissue culture conditions, the macrophages comprised more than 95% of the culture cell population. In the 48-h cultures, 98% of the cells expressed the macrophage antigens identified by OKM5 and EBM 11 monoclonal antibodies, and 75% expressed the HLA-DR antigen; 95% were positive for NSE, 88% of the cells had Fc receptors, and 95% of them were able to phagocytose S. albus. Only 0.1% of the culture cell population were positive to anti-alpha-HCG monoclonal antibody. These assays showed that the placental macrophages are the main cells in the placental stroma. In our tissue culture conditions, incubation time was shown to be the dominant factor in selecting macrophages over other cell types, most probably because the macrophages are the only placental cells that have the ability to adhere to plastic dishes. This observation might also explain the contradictory results in various studies over the last decade, regarding the incidence of the placental mononuclear macrophages.

Mycoplasma capricolum membranes induce tumor necrosis factor alpha by a mechanism different from that of lipopolysaccharide

Heat-inactivated (60 degrees C, 45 min) Mycoplasma capricolum strain JR cells activate murine macrophages to secrete high levels of tumor necrosis factor alpha (TNF alpha) and to lyse tumor target cells efficiently. Fractionation of the intact M. capricolum cells, obtained from cells harvested at the exponential phase of growth, shows that their capacity to induce TNF alpha secretion by macrophage resides exclusively in the membrane fraction. The macrophage-mediated cytolysis following activation by M. capricolum membranes was significantly inhibited by specific anti-recombinant murine TNF alpha antibodies. M. capricolum membranes are a potent inducer of TNF alpha as the commonly used bacterial lipopolysaccharide, indicated by their dose-response curve for macrophage activation. Our study further showed that M. capricolum membranes and lipopolysaccharide synergize to augment TNF alpha secretion by C57BL/6-derived macrophages markedly. Moreover, lipopolysaccharide-unresponsive C3H/HeJ-derived macrophages, were pronouncedly activated by M. capricolum membranes, which do not contain lipopolysaccharide. These findings suggest that the mechanism by which M. capricolum membranes activate macrophages differs from that of lipopolysaccharide. Results of preliminary experiments show that human monocytes as well secrete TNF alpha following activation by M. capricolum membranes. Thus, in contrast with the prohibitive toxicity of lipopolysaccharide to animals and humans, M. capricolum membranes, which contain no lipopolysaccharide and are nontoxic in nature, may be of therapeutic value in the treatment of cancer.

Engulfment and intracellular killing of F9 teratocarcinoma cells by non-activated murine macrophages

Activated macrophages kill several types of tumor cells in vitro, whereas non-activated macrophages lack this capacity. We, however, observed that non-activated macrophages efficiently kill F9 teratocarcinoma as well as other teratocarcinoma cell lines. Dexamethasone, a glucocorticoid known to prevent macrophage activation, did not perturb the killing of F9 teratocarcinoma cells. Neither tumor necrosis factor alpha, nor the reactive oxygen intermediates, i.e. hydrogen peroxide, superoxide anion, and hydroxyl radical, nor serine proteases participated in this killing, shown by employing various agents which interfere with their production, secretion, or function. Using acridine orange/ethidium bromide vitality staining, the F9 teratocarcinoma cells were shown to be phagocytized alive by macrophages and subsequently killed intracellularly. Intact lysosomal function is required for the killing of F9 cells, as the lysosomotropic drugs chloroquine and ammonium chloride markedly inhibited this killing without perturbing their engulfment. The signal transduction pathway induced in the macrophages upon interaction with F9 teratocarcinoma cells seems to differ from that induced by macrophage activation. Neither the protein kinase C inhibitors polymyxin B and H-7 [1-(5-isoquinolinylsulfonyl)-2-methyl piperazine] nor the protein kinase C activator phorbol 12-myristate-13-acetate affected the killing of F9 cells. However, chlorpromazine (a powerful inhibitor of calmodulin), dibutyryl cAMP (a cAMP analog), and prostaglandin E2 inhibited the macrophage-mediated killing of F9 cells. In vivo studies indicate that an increased number of macrophages at the F9 tumor inoculation site (the peritoneal cavity) as a result of elicitation by thioglycollate prevents F9 tumor development. Our findings indicate that non-activated macrophages kill teratocarcinoma cells using a mechanism which differs from that employed by activated macrophages in the killing of other tumor cells.

Production of a factor (CIF) from normal fibroblast cells inhibiting tumor necrosis factor/cachectin production

Murine embryonic fibroblast cells produce a factor designated cytotoxin-inhibiting factor (CIF) which inhibits tumor necrosis factor (TNF) and interleukin 1 production as well as tumoricidal activity by lipopolysaccharide-activated macrophages. This study determines the physiologic conditions of CIF production in serum-free medium. CIF production was largely dependent upon the presence of lipopolysaccharide. A quantitative correlation between fibroblast cell number, lipopolysaccharide concentration, and incubation time was established. Evidence is presented that CIF inhibited the production or release of TNF. CIF did not destroy TNF after production and release nor did it sequester secreted TNF. The supernatant fluids which inhibited TNF production did not suppress the capability of resting macrophages to phagocytize opsonized sheep erythrocytes, suggesting that only functions expressed in the activated state are inhibited.

Tumor necrosis factor as a mediator of Mycoplasma orale-induced tumor cell lysis by macrophages

We and other investigators have previously demonstrated that mycoplasmas induce macrophage-mediated lysis of tumor cells, but the mechanism responsible for this process had, thus far, not been clarified. We now report that addition of either viable or heat-killed Mycoplasma orale to murine macrophages induces a cytolytic activity which, due to its neutralization by a specific antiserum against murine cloned recombinant tumor necrosis factor (rTNF), was identified as TNF-mediated. Both thioglycollate-elicited peritoneal macrophages and the normal macrophages cloned from our JBM phi 1.1 bone-marrow-derived cell line effectively produced TNF at levels similar to, or higher than, those obtained in the presence of high concentrations of lipopolysaccharide (LPS). Four other mycoplasma species demonstrated a varied capacity to induce TNF production by macrophages. Elevated TNF levels were also observed during macrophage-mediated cytolysis of murine A9 fibrosarcoma cells in the presence of either M. orale or LPS. Addition of the specific antiserum against rTNF at a concentration which neutralized all TNF activity in the co-cultures partially inhibited concomitant A9 cell killing. We can, therefore, conclude that M. orale induces TNF production which is, at least partially, responsible for subsequent tumor cell killing.

Interaction between thymocytes and thymus-derived macrophages. II. Engulfment of thymocytes by macrophages

A high percentage (80-90%) of immature thymocytes were engulfed by syngeneic thymus-derived macrophages (TDM phi) following cocultivation for 3 days. Elimination occurred via internalization of thymocytes by the macrophages. We unequivocally demonstrated the presence of many live thymocytes inside the TDM phi by means of specific staining. Mature PNA- thymocytes were phagocytized to a lower degree than immature thymocytes, and T splenocytes were not eliminated at all. Bone marrow-derived macrophages internalized immature thymocytes to a degree similar to TDM phi. Since thymocyte survival was not at all affected by M phi culture supernatants alone, we conclude that cell to cell contact is necessary for thymocyte elimination. To identify the surface molecules which participate in internalization of thymocytes by the macrophages, both cell types were pretreated with a variety of agents. Treatment of thymocytes with tunicamycin (N-glycosylation inhibitor) and anti-Lyt-2 mAb decreased their elimination by M phi. Similarly, treatment of M phi with neuraminidase, trypsin, and anti-Ia mAb markedly suppressed their capacity to engulf thymocytes. On the other hand, thymocyte elimination was unaffected by (1) cell cultivation in syngeneic serum rather than heterologous serum; (2) use of allogeneic rather than syngeneic thymocytes; and (3) use of X-irradiated M phi and LPS-activated M phi rather than nontreated M phi.

Interaction between thymocytes and thymus-derived macrophages. I. Surface components participating in mutual recognition

Incubation of C57BL/6 thymus-derived macrophages (TDM phi) with syngeneic thymocytes resulted in binding of thymocytes to macrophages and rosette formation. Up to 60% of the TDM phi formed rosettes with thymocytes after 6 hr of interaction at 4 degrees C. Rosette formation of the immature PNA+ thymocyte fraction was up to fivefold higher than that of PNA- and cortisone-resistant thymocytes. Pretreatment of PNA- thymocytes with neuraminidase enhanced thymocyte binding to macrophages up to sevenfold, whereas a marked reduction of rosette formation was seen following (1) incubation of thymocytes with tunicamycin; (2) incubation of macrophages with 20 mM D-galactose, GLCNaC, or GalNaC; (3) treatment of macrophages or thymocytes with trypsin; (4) treatment of macrophages with anti-1-Ab mAb and its F(ab')2 fragment; (5) treatment of thymocytes with anti-Lyt-2.2 mAb; and (6) addition of EDTA and EGTA to the interacted two cell populations.

Modulation of lipoprotein lipase activity in mouse peritoneal macrophages by recombinant human tumor necrosis factor

Thioglycollate-elicited mouse peritoneal macrophages spontaneously secrete lipoprotein lipase during culture. Exposure of the cultures to 50 ng/ml of recombinant human tumor necrosis factor (rTNF) for 48 h resulted in a 69% reduction in lipoprotein lipase activity in the culture medium with a concomitant decrease in cellular enzyme activity. The decrease in enzyme activity was not the result of rTNF-dependent reduction in the total protein synthesis, since the presence of rTNF did not affect [3H]leucine incorporation into cellular proteins. The effect of rTNF on lipoprotein lipase was reversible; upon TNF withdrawal, enzyme activity returned to basal levels after 60 h. The reduction of lipoprotein lipase in rTNF-treated cultures could be completely prevented by preincubation with a specific antiserum against recombinant human TNF. The late onset of decrease of lipoprotein lipase (LPL) activity suggests that rTNF might induce a mediator, which in turn suppresses LPL production. While rTNF was very effective in reducing lipoprotein lipase activity in mouse peritoneal macrophages, it did not affect lipoprotein lipase activity when added to the murine J774 cell line and to CT2 macrophage-like cells, a variant of the J774 cell line.

Prostaglandin E2-mediated suppression of human maternal lymphocyte alloreactivity by first-trimester fetal macrophages

We examined the immunosuppressor role of the first-trimester human placental macrophages on maternal lymphocyte alloreactivity in vitro to determine whether these macrophages and their secreted mediator(s) participate in the local immunoregulation at the fetomaternal interface. These cells were tested for their effects on maternal lymphocyte proliferation when added alone (as stimulator cells) or as regulator cells in 1) a one-way mixed lymphocyte reaction for six days and 2) lymphocyte cultures in the presence of phytohemagglutinin for three days. We detected a dose-dependent inhibition of phytohemagglutinin-induced proliferation with 48-hour macrophage culture supernatants. Placental macrophages, when used alone as stimulator cells in the mixed lymphocyte reaction, were stimulatory only at a concentration of 1-2%; the stimulation was abolished at a concentration of 10%. When used as regulator cells, they showed a strong inhibition of mixed lymphocyte reaction and phytohemagglutinin-induced proliferation at a concentration of 10%; no effect was observed at a concentration of 1-2%. The major class of the molecules mediating the suppression was identified as prostaglandins, primarily prostaglandin E2 (PGE2), based on the findings that the presence of indomethacin (10(-5) M) or various dilutions of an anti-PGE2 antibody abrogated suppression substantially or completely. Prostaglandin E2 levels measured in the mixed leukocyte culture wells containing placental macrophages as regulator cells correlated positively with the macrophage dose and its suppressive effect.

Killing of virus-induced thymic lymphoma cells obtained from primary and early transfers by activated macrophages

Primary radiation leukemia virus-induced thymic lymphomas were susceptible to both stasis and lysis mediated by LPS-activated macrophages. During short in vitro cultivation the X-irradiation-induced lymphoma (PIR-2) cells were killed to the same extent (up to 99%) by LPS-activated macrophages as the RadLV lymphoma cells at effector:target ratios of 10:1 and 1:1. Since RadLV is not present in PIR-2 lymphoma cells, its presence is not required for susceptibility of the lymphoma cells to macrophage-mediated killing. Our data also disprove the assumption that artifacts of long-term cultivation in vitro are responsible for tumor susceptibility to cytostasis and cytolysis mediated by activated macrophages.