The effect of fibroblast growth factor 15 deficiency on the development of high fat diet induced non-alcoholic steatohepatitis

Non-alcoholic steatohepatitis (NASH) is a form of non-alcoholic fatty liver disease (NAFLD) characterized by steatosis, inflammation, and fibrosis often associated with metabolic syndrome. Fibroblast growth factor 15 (FGF15), an endocrine factor mainly produced in the distal part of small intestine, has emerged to be a critical factor in regulating bile acid homeostasis, energy metabolism, and liver regeneration. We hypothesized that FGF15 alters the development of each of the listed features of NASH. To test this hypothesis, four-week old male Fgf15^-/- and their corresponding wild-type (WT) mice were fed either a high fat diet (HFD) or a control chow diet for six months. The results confirmed that HFD feeding for six months in WT mice recapitulated human NASH phenotype, including macrovesicular steatosis, inflammation, and fibrosis. Whereas FGF15 deficiency had no effect on the severity of liver steatosis or inflammation, it was associated with decreased liver fibrosis. Furthermore, FGF15 deficiency resulted in abnormal bile acid homeostasis, increased insulin resistance, increased HFD-induced serum triglycerides, decreased inductions of hepatic cholesterol content by HFD, and altered gene expression of lipid metabolic enzymes. These data suggest that FGF15 improves lipid homeostasis and reduces bile acid synthesis, but promotes fibrosis during the development of NASH.

Toxicodynamics of rigid polystyrene microparticles on pulmonary gas exchange in mice: implications for microemboli-based drug delivery systems

The toxicodynamic relationship between the number and size of pulmonary microemboli resulting from uniformly sized, rigid polystyrene microparticles (MPs) administered intravenously and their potential effects on pulmonary gas exchange were investigated. CD-1 male mice (6-8 weeks) were intravenously administered 10, 25 and 45 μm diameter MPs. Oxygen hemoglobin saturation in the blood (SpO(2)) was measured non-invasively using a pulse oximeter while varying inhaled oxygen concentration (F(I)O(2)). The resulting data were fit to a physiologically based non-linear mathematical model that estimates 2 parameters: ventilation-perfusion ratio (V(A)/Q) and shunt (percentage of deoxygenated blood returning to systemic circulation). The number of MPs administered prior to a statistically significant reduction in normalized V(A)/Q was dependent on particle size. MP doses that resulted in a significant reduction in normalized V(A)/Q one day post-treatment were 4000, 40,000 and 550,000 MPs/g for 45, 25 and 10 μm MPs, respectively. The model estimated V(A)/Q and shunt returned to baseline levels 7 days post-treatment. Measuring SpO(2) alone was not sufficient to observe changes in gas exchange; however, when combined with model-derived V(A)/Q and shunt early reversible toxicity from pulmonary microemboli was detected suggesting that the model and physical measurements are both required for assessing toxicity. Moreover, it appears that the MP load required to alter gas exchange in a mouse prior to lethality is significantly higher than the anticipated required MP dose for effective drug delivery. Overall, the current results indicate that the microemboli-based approach for targeted pulmonary drug delivery is potentially safe and should be further explored.

Nitric oxide and peroxynitrite in ozone-induced lung injury

One of the hallmarks of the inflammatory response associated with tissue injury is the accumulation of macrophages at sites of damage. These cell types release proinflammatory cytokines and cytotoxic mediators to destroy invading pathogens and initiate wound repair. However, when produced in excessive amounts, these macrophage-derived mediators may actually contribute to tissue injury. This process involves both direct damage to target tissues and amplification of the inflammatory response. One group of macrophage-derived mediators of particular interest are reactive nitrogen intermediates including nitric oxide and peroxynitrite which have been implicated in tissue injury induced by a variety oftoxicants. Our laboratory has been investigating the role of reactive nitrogen intermediates in lung injury induced by oxidants such as ozone. Inhalation of ozone causes epithelial cell damage and Type II cell hyperplasia. This is associated with an accumulation of activated macrophages in the lower lungs which we have demonstrated contribute to toxicity. To analyze the role of macrophage-derived reactive nitrogen intermediates in ozone toxicity, we used transgenic mice lacking the gene for inducible nitric oxide synthase (NOSII). Treatment of wild type control animals with ozone (0.8 ppm) for 3 hr resulted in an increase in bronchoalveolar lavage (BAL) fluid protein reaching a maximum 24-48 hr after exposure. This was correlated with increased expression of NOSII protein and mRNA by alveolar macrophages and increased production of nitric oxide as well as peroxynitrite. Ozone inhalation also resulted in the appearance of nitrotyrosine residues in the lungs, an in vivo marker of peroxynitrite-induced damage. In contrast, in NOSII knockout mice, BAL protein was not increased demonstrating that these mice were protected from ozone-induced epithelial injury. Moreover, alveolar macrophages from the transgenic mice did not produce nitric oxide or peroxynitrite even after ozone inhalation. There was also no evidence for the formation of nitrotyrosine in lung tissue. These data indicate that ozone-induced lung injury is mediated by reactive nitrogen intermediates.

Tissue injury following inhalation of fine particulate matter and hydrogen peroxide is associated with altered production of inflammatory mediators and antioxidants by alveolar macrophages

Hydrogen peroxide (H(2)O(2)) is present in the atmosphere at concentrations known to induce cell and tissue damage. However, inhaled H(2)O(2) vapor should not reach the lower lung due to its high water solubility. It has been suggested that hygroscopic components of particulate matter (PM) may transport H(2)O(2) into the lower lung and induce tissue injury and this was investigated. Ammonium sulfate [(NH(4))(2)SO(4)] was selected as a model for fine atmospheric PM. Treatment of female Sprague-Dawley rats with (NH(4))(2)SO(4) (429 or 215 microg/m(3); 0.3-0.4 microm mass median diameter) or H(2)O(2) (10, 20, or 100 ppb) alone or in combination for 2 h had no major effect on bronchoalveolar lavage fluid cell number or viability or on protein content or lactate dehydrogenase levels, either immediately or 24 h after exposure, relative to air-exposed rats. However, electron microscopy revealed increased numbers of neutrophils in pulmonary capillaries adhered to the vascular endothelium in rats treated with the combination of (NH(4))(2)SO(4) + H(2)O(2). Exposure of rats to (NH(4))(2)SO(4) + H(2)O(2) also resulted in tumor necrosis factor-alpha (TNF-alpha) production by alveolar macrophages. This was observed immediately and 24 h after exposure. Immediately after inhalation of (NH(4))(2)SO(4) + H(2)O(2), a transient increase in production of superoxide anion by alveolar macrophages was observed. In contrast, nitric oxide production by cells from rats exposed to (NH(4))(2)SO(4) + H(2)O(2) or H(2)O(2) alone was decreased, and this persisted for 24 h. Decreases in nitric oxide may be due to superoxide anion-driven formation of peroxynitrite. In this regard, nitrotyrosine, an in vivo marker of peroxynitrite, was detected in lung tissue after exposure of rats to (NH(4))(2)SO(4) + H(2)O(2) or H(2)O(2). We also found that expression of the antioxidant enzyme heme oxygenase-1 by stimulated alveolar macrophages was increased following exposure of rats to (NH(4))(2)SO(4) + H(2)O(2). Taken together, these studies demonstrate that the biological effects of inhaled fine PM are augmented by H(2)O(2). Moreover, tissue injury induced by fine PM may be related to altered production of cytotoxic mediators by alveolar macrophages.

Mechanisms underlying reduced responsiveness of neonatal neutrophils to distinct chemoattractants

Potential mechanisms underlying impaired chemotactic responsiveness of neonatal neutrophils were investigated. Two distinct chemoattractants were compared: bacterially derived N-formyl-methionyl-leucyl-phenylalanine (fMLP) and a unique chemotactic monoclonal antibody, designated DL1.2, which binds to a neutrophil antigen with an apparent molecular mass of 120 kDa. Chemotaxis of neutrophils toward fMLP, as well as DL1.2, was reduced in neonates when compared with adult cells. This did not appear to be a result of decreased fMLP receptor or DL1.2 antigen expression by neonatal neutrophils. fMLP, but not DL1.2, induced a rapid increase in intracellular calcium in adult and neonatal cells, which reached a maximum within 30 s. The calcium response of cells from neonates to fMLP was reduced when compared with adult cells, and an unresponsive subpopulation of neonatal neutrophils was identified. NF-kappaB nuclear binding activity induced by fMLP and DL1.2, as well as expression of the p65 NF-kappaB subunit and IkappaB-alpha, was also significantly reduced in neonatal cells, when compared with adult cells. In contrast, although fMLP, but not DL1.2, activated p42/44 and p38 mitogen-activated protein (MAP) kinases in neutrophils, no differences were observed between adults and neonates. Chemotaxis of adult and neonatal neutrophils toward fMLP and DL1.2 was also blocked to a similar extent by inhibitors of phosphatidylinositol 3-kinase, as well as an inhibitor of NF-kappaB. These findings indicate that reduced chemotactic responsiveness in neonatal neutrophils is a result of, at least in part, aberrations in chemoattractant-induced signaling. However, the biochemical pathways mediating this defect appear to be related to the specific chemoattractant.

Functional heterogeneity in liver and lung macrophages

Although initially considered merely "scavenger cells" that participate in immunologic responses only after B and T lymphocytes have performed their biological tasks, more recent evidence suggests that macrophages play a key role in host defense as well as in the maintenance of normal tissue structure and function. For macrophages to perform their biological functions, they must be activated. This involves up-regulation of an array of signaling pathways resulting in altered gene expression and increased biochemical and functional activity. Macrophages have been identified in almost all tissues of the body. However, the basal activity of these cells, as well as their ability to respond to inflammatory mediators, varies considerably with their location. In addition, even within a particular tissue, there is evidence of macrophage heterogeneity. The largest populations of macrophages in the body are located in the liver and lung. Because of the unique attributes of these tissues, hepatic and pulmonary macrophages play essential roles not only in nonspecific host defense but also in the homeostatic responses of these tissues. In this review, the functional and biochemical activities of macrophages localized in the liver and lungs are compared. Evidence suggests that these represent distinct cell populations with unique functions and responsiveness to inflammatory agents.

Minimal amidine structure for inhibition of nitric oxide biosynthesis 2 2Abbreviations: NOS1, neuronal nitric oxide synthase; NOS2, inducible nitric oxide synthase; l-NMMA, NG-monomethyl-l-arginine; and NIO, l-N5-(iminoethyl)ornithine

Pharmacological modulation of nitric oxide synthase activity has been achieved using structural analogs of arginine. In the present studies, we demonstrated that the minimal amidine structure required for enzymatic inhibition is formamidine. We found that the production of nitric oxide by primary cultures of rat hepatocytes and several mouse and human cell lines, including RAW 264.7 macrophages, PAM 212 keratinocytes, G8 myoblasts, S180 sarcoma, CX-1 human colon cells, and GH3 rat pituitary cells, was inhibited in a concentration- and time-dependent manner by formamidine. Formamidine was 2- to 6-fold more effective in inhibiting nitric oxide production in cells expressing inducible nitric oxide synthase (NOS2) than in a cell line expressing calcium-dependent neuronal nitric oxide synthase (NOS1). Whereas formamidine had no effect on gamma-interferon-induced expression of nitric oxide synthase protein, its enzymatic activity was blocked. Kinetic analysis revealed that formamidine acts as a simple competitive inhibitor with respect to arginine (K(i) formamidine approximately 800 microM). Using a polarographic microsensor to measure real-time flux of nitric oxide release from RAW 264.7 macrophages, formamidine was found to require 30-90 min to inhibit enzyme activity, suggesting that cellular uptake of the drug may limit its biological activity. Our data indicate that formamidine is an effective inhibitor of nitric oxide production. Furthermore, its low toxicity may make it useful as a potential therapeutic agent in diseases associated with the increased production of nitric oxide.

Distinct roles of tumor necrosis factor-alpha and nitric oxide in acute liver injury induced by carbon tetrachloride in mice

Macrophages are known to release a number of different inflammatory mediators with cytotoxic potential. In the present studies we analyzed the role of two macrophage-derived mediators, tumor necrosis factor-alpha (TNF-alpha) and nitric oxide, in liver injury induced by carbon tetrachloride (CCl4). Treatment of mice with CCl4 resulted in a dose- and time-dependent induction of centrilobular hepatic necrosis. This was observed within 12 h with 0.3 ml/kg CCl4 and was correlated with increases in serum transaminase levels. CCl4 administration also caused increases in hepatic TNF-alpha mRNA expression and serum TNF-alpha levels, as well as inducible nitric oxide synthase (NOS II) protein expression in the liver. To study the role of TNF-alpha and nitric oxide in hepatotoxicity, we used knockout mice lacking the gene for the 55-kDa TNF-alpha receptor (TNFR1/p55), the TNF-alpha cytokine, or NOS II. We found that CCl4 was significantly less effective in inducing hepatotoxicity in mice lacking TNFR1/p55 or the TNF-alpha cytokine. In contrast, CCl4-induced liver injury was increased in knockout mice lacking the gene for NOS II. This was associated with an increase in hepatic TNF-alpha mRNA expression and serum TNF-alpha levels. These data suggest that the hepatoprotective effects of nitric oxide in this model may be due in part to inhibition of TNF-alpha.

Prooxidant and antioxidant functions of nitric oxide in liver toxicity

In response to tissue damage and inflammation induced by a variety of xenobiotics including acetaminophen, carbon tetrachloride, ethanol, galactosamine, and endotoxin, as well as disease states such as viral hepatitis, and postischemic and regenerative injury, the liver produces large quantities of nitric oxide. Indeed, nearly all cell types in the liver including hepatocytes, Kupffer cells, stellate cells, and endothelial cells have the capacity to generate nitric oxide. Thus, these cells, as well as infiltrating leukocytes, may indirectly augment tissue injury. In many models of liver damage, nitric oxide and its oxidation products such as peroxynitrite contribute to the injury process by directly damaging the tissue or by initiating additional immunologic reactions that result in damage. In some models, nitric oxide donors or peroxynitrite can mimic the cytotoxic actions of liver toxins. Moreover, agents that prevent the generation of nitric oxide or antioxidants that bind reactive nitrogen intermediates, or knockout mice with reduced capacity to produce nitric oxide, are protected from xenobiotic-induced tissue injury. In contrast, there have been reports that blocking nitric oxide production enhances xenobiotic-induced tissue injury. This has led to the concept that nitric oxide either inactivates proteins critical for xenobiotic-induced tissue injury or acts as an antioxidant, reducing cellular levels of cytotoxic reactive oxygen intermediates. Whether or not nitric oxide or secondary oxidants generated from nitric oxide act as mediators of tissue injury or protect against toxicity is likely to depend on the precise targets of these reactive nitrogen intermediates, as well as levels of superoxide anion present and the extent to which tissue injury is mediated by reactive oxygen intermediates. In addition, as toxicity is a complex process involving a variety of cell types and many soluble mediators, the contribution of each of these factors must be taken into account when considering the role of nitric oxide as a determinant of tissue injury.

Role of macrophages and inflammatory mediators in chemically induced toxicity

Macrophages are critical cellular effectors of nonspecific host defense. They are also potent secretory cells releasing an array of mediators including proinflammatory and cytotoxic cytokines and growth factors, bioactive lipids, hydrolytic enzymes and reactive oxygen and nitrogen intermediates, each of which has been implicated in tissue injury. The research in our laboratories has focused on analyzing the role of macrophages in chemically induced injury in the lung and the liver. In both these tissues, a localized accumulation of macrophages is observed following toxicant exposure. This is directly correlated with the generation of cytotoxic inflammatory mediators at these sites. Moreover, when macrophage functioning is blocked, pulmonary and hepatic injury induced by toxicants such as ozone or acetaminophen is prevented. These findings provide direct support for our hypothesis that macrophages contribute to tissue injury. Approaches using pharmacologic inhibitors and transgenic animals are currently being used to evaluate the specific macrophage-derived products involved in the pathogenic process. Our results suggest that the extent to which a particular mediator contributes to injury depends on the nature of the toxicant, the target tissue, and quantities of the mediator produced.

The toxicology of inhaled nitric oxide

Inhaled nitric oxide is a targeted pulmonary vasodilator that improves clinical outcomes for newborn patients with persistent pulmonary hypertension of the newborn, and may be effective in treating some premature patients with acute respiratory distress syndrome or lung disease of prematurity. Nitric oxide is now recognized as playing an important role in the regulation of diverse physiological processes. However, the pharmacological properties of inhaled nitric oxide are not easy to separate from its toxicological effects. For example, the intended effect of inhaled nitric oxide, vasodilation in the lung, is mediated, in part, by increased cellular cyclic GMP (cGMP). However, increased cGMP can also interfere with normal cellular proliferation. Nitric oxide has also been shown to cause DNA strand breaks and/or base alterations that are potentially mutagenic. Inhaled nitric oxide can rapidly react with oxygen in the lung to form nitrogen dioxide, which is a potent pulmonary irritant. Nitric oxide also reacts with superoxide anion to form peroxynitrite, a cytotoxic oxidant that can interfere with surfactant functioning. The overall effect of inhaled nitric oxide in potentiating or attenuating inflammation and oxidative damage in diseased lung is dependent on the dose administered. Furthermore, despite rapid inactivation by circulating hemoglobin, inhaled nitric oxide exerts effects outside the lung, including blocking platelet aggregation, causing methemoglobinemia, and possibly inducing extrapulmonary vasodilation. The toxicology of inhaled nitric oxide is not completely understood and must be considered in the design of protocols for its safe and effective clinical use.

Pharmacologic therapy of persistent pulmonary hypertension of the newborn

Persistent pulmonary hypertension of the newborn (PPHN) is a potentially life-threatening condition characterized by a failure of pulmonary vascular resistance to decrease adequately during the transition to extrauterine life. Inhaled nitric oxide, a vasodilator that acts selectively on the pulmonary circulation, has revolutionized the treatment of this condition. However, inhaled nitric oxide has not proven effective in all patients, particularly those with congenital diaphragmatic hernias or meconium aspiration syndrome. Furthermore, large clinical trials of inhaled nitric oxide have failed to demonstrate significant differences in mortality between nitric oxide-treated and control infants with PPHN. Other therapeutic approaches to PPHN have been limited by a relative lack of specificity for the pulmonary circulation, and have received much less attention. Pharmacologic approaches, including pulmonary surfactants, prostacyclin, endothelin antagonists, Ca(2+)-channel blockers, magnesium sulfate, and tolazoline, have exhibited varying degrees of efficacy in lowering pulmonary vascular pressures in humans and/or animals. A number of these agents are also effective when used in combination. For example, phosphodiesterase inhibitors have been reported to act synergistically with inhaled nitric oxide. Surfactants also appear to be useful in PPHN, particularly in patients with congenital diaphragmatic hernia, when used in combination with other therapies. Surfactant lavage and other novel therapies may also be effective in combination therapy of meconium aspiration syndrome. Further studies should be directed at defining the optimal therapies in specific clinical settings. Validation of multiple therapeutic modalities for PPHN, including inhaled nitric oxide, will allow for rational, combined vasodilator strategies that are specific for the underlying pathophysiology in each patient.

Functional heterogeneity of rat hepatic and alveolar macrophages: effects of chronic ethanol administration

Chronic ethanol consumption is associated with increased incidence of hepatic and pulmonary infections. To determine if this is correlated with altered macrophage activity, we analyzed the functional properties of cells isolated sequentially from the liver and lung of rats fed a liquid diet containing ethanol (35% of calories) or malto-dextrin control for 9-12 weeks. Hepatic and alveolar macrophages from control animals were found to exhibit distinct morphologic and functional properties. Thus, hepatic macrophages were highly vacuolated and appeared larger and more irregular in shape than alveolar macrophages. These cells also displayed greater phagocytic activity and random migration. In contrast, lung macrophages produced more superoxide anion and nitric oxide, and exhibited enhanced chemotactic activity toward the complement fragment C5a. Whereas administration of ethanol to rats for 9-12 weeks resulted in decreased chemotaxis and superoxide anion production by alveolar macrophages, cell adhesion molecule expression was reduced in hepatic macrophages. Nitric oxide production and inducible nitric oxide synthase protein expression were decreased in both macrophage populations. These effects were not observed after 3-6 weeks of ethanol administration to rats. Our results suggest that changes in macrophage functioning may play a role in decreased host defense following chronic ethanol exposure.

Characterization of the inflammatory response to biomaterials using a rodent air pouch model

Using a rodent air pouch, the inflammatory responses to biomaterials with distinct physical properties and chemical compositions were compared. The polymers examined were expanded poly(tetrafluoroethylene) (ePTFE), silicone, low-density polyethylene (LDPE), poly(L-lactic acid) (PLLA), poly(desaminotyrosyl-tyrosine ethyl carbonate) [poly(DTE carbonate)], and poly(desaminotyrosyl-tyrosine benzyl carbonate) [poly(DTBzl carbonate)]. We found that implantation of disks (4.5-4.8 mm) of these materials into rodent air pouches for 2 days had no effect on the number or type of cells recovered relative to sham controls. With each of the materials, macrophages were the predominant cell type identified (60-75%), followed by granulocytes (20-25%) and lymphocytes (10%). Implantation of poly(DTE carbonate), ePTFE, LDPE, or poly(DTBzl carbonate) into the pouches for 2 days caused an increase in release of superoxide anion by the pouch cells. Cells from pouches containing poly(DTE carbonate) also released more hydrogen peroxide and were more phagocytic. In contrast, PLLA and silicone had no effect on the functional activity of cells recovered from the pouches. Prolonging the implantation time of poly(DTE carbonate) or PLLA to 7 days did not alter the number or type of cells isolated from the pouches. However, cells from pouches containing poly(DTE carbonate) for 7 days continued to produce increased quantities of superoxide anion relative to sham control pouch cells. These results suggest that the air pouch model is a highly sensitive method and therefore useful for evaluating the functional responses of inflammatory cells to biomaterials.

An exposure system to study the effects of water-soluble gases on PM-induced toxicity

An aerosol generation and exposure system to evaluate the role of water-soluble gases in particulate matter (PM)-induced injury was designed, built, and validated by generating test atmospheres to study the role of hydrogen peroxide in PM-induced toxicity. In this system, particle number concentration, size distribution, hydrogen peroxide concentration, and water concentration can all be varied. An ammonium sulfate aerosol with mass median diameter 0.46 +/- 0.01 microm was used as a model atmospheric aerosol because ammonium sulfate is a major component of the fine aerosol, and the water uptake of ammonium sulfate aerosol is well characterized. The following four test atmospheres were generated: (1) ammonium sulfate aerosol, (2) an aerosol containing hydrogen peroxide and ammonium sulfate, (3) vapor-phase hydrogen peroxide, and (4) particle-free air. All test atmospheres were maintained at a relative humidity of 85%. Particle size distribution, number concentration, total hydrogen peroxide concentration, temperature, and relative humidity were measured continuously in the exposure chamber. The gas-particle partitioning of hydrogen peroxide was calculated using total hydrogen peroxide concentration, the Henry's law constant for hydrogen peroxide in water, and aerosol water content. We found that the aerosol generation system produced stable concentrations throughout the 2-hour exposures.

Nitric oxide in the lung: therapeutic and cellular mechanisms of action

Nitric oxide is produced by many cell types in the lung and plays an important physiologic role in the regulation of pulmonary vasomotor tone by several known mechanisms. Nitric oxide stimulates soluble guanylyl cyclase, resulting in increased levels of cyclic GMP in lung smooth muscle cells. The gating of K+ and Ca2+ channels by cyclic GMP binding is thought to play a role in nitric oxide-mediated vasodilation. Nitric oxide may also regulate pulmonary vasodilation by direct activation of K+ channels or by modulating the expression and activity of angiotensin II receptors. Administration of nitric oxide by inhalation has been shown to acutely improve hypoxemia associated with pulmonary hypertension in humans and animals. This is presumably due to its ability to induce pulmonary vasodilation. Inhaled nitric oxide improves oxygenation and reduces the need for extracorporeal membrane oxygenation in term and near-term infants with persistent pulmonary hypertension. However, long-term benefits to these infants have been difficult to demonstrate. In other pathologic conditions, such as prematurity and acute respiratory distress syndrome, short-term benefits have not been shown conclusively to outweigh potential toxicities. For example, high-dose inhaled nitric oxide decreases surfactant function in the lung. Inhaled nitric oxide also acts as a pulmonary irritant, causing priming of lung macrophages and oxidative damage to lung epithelial cells. Conversely, protective effects of nitric oxide have been described in a number of pathological states, including hyperoxic and ischemia/reperfusion injury. Nitric oxide has also been reported to protect against oxidative damage induced by other reactive intermediates, including superoxide anion and hydroxyl radical. The dose and timing of nitric oxide administration needs to be ascertained in clinical trials before recommendations can be made regarding its optimal use in patients.

Inhibition of macrophages with gadolinium chloride alters intercellular adhesion molecule-1 expression in the liver during acute endotoxemia in rats

Cell adhesion molecules are important for localized accumulation of phagocytes at sites of tissue damage. In the present studies, we analyzed the effects of blocking hepatic macrophages on expression of beta2 integrins and intercellular adhesion molecule-1 (ICAM-1) adhesion molecules on liver cells during acute endotoxemia. Flow cytometric analysis revealed distinct subpopulations of macrophages from control animals that varied on the basis of their size and density. In contrast, hepatocytes and endothelial cells were relatively homogeneous. Treatment of rats with endotoxin (5 mg/kg, intravenously) resulted in a time-dependent increase in the percentage of small, dense macrophages and a progressive loss of larger, less-dense cells. In contrast, no major effects were observed on the physical properties of hepatocytes or endothelial cells. ICAM-1 was found to be constitutively expressed on endothelial cells and hepatocytes, as well as on macrophages. Induction of acute endotoxemia resulted in a time-dependent increase in ICAM-1 expression on hepatocytes, which was observed within 3 hours and reached a maximum after 24 hours. An increase in ICAM-1 expression was also observed on endothelial cells and on macrophages at 3 hours, followed by a decrease at 24 to 48 hours. Macrophages and endothelial cells also constitutively expressed beta2 integrins. Induction of acute endotoxemia had no effect on beta2 integrin expression by these cells. Pretreatment of rats with gadolinium chloride (GdCl3), a macrophage inhibitor known to block endotoxin-induced liver injury, abrogated the effects of endotoxin on ICAM-1 expression by hepatocytes and macrophages. In contrast, ICAM-1 expression on endothelial cells increased. Interestingly, treatment of rats with GdCl3 alone resulted in a marked increase in expression of ICAM-1 on endothelial cells and hepatocytes, and of beta2 integrins on macrophages and endothelial cells. Taken together, these data suggest that ICAM-1 is involved in mediating macrophage adherence and accumulation in the liver during endotoxemia. Furthermore, macrophages appear to regulate expression of this cell adhesion molecule on parenchymal cells.

Role of inflammatory cytokines and nitric oxide in hepatic and pulmonary toxicity

Exposure of humans and experimental animals to inhaled irritants such as ozone, induces an acute inflammatory response and lung injury. We hypothesize that macrophage-derived inflammatory cytokines and cytotoxic mediators contribute to the pathogenic process. Treatment of rats with ozone (2 ppm, 3 h) results in damage to the alveolar epithelium and increased protein in lung lavage fluid. This is associated with an increase in the number of macrophages in the lung. We found that these cells are activated to release the proinflammatory cytokine tumor necrosis factor-alpha (TNF alpha) which has been implicated in tissue injury. Following ozone inhalation, alveolar macrophages also produce increased amounts of the cytotoxic mediator, nitric oxide. This response is time-dependent and correlated with expression of inducible nitric oxide synthase (NOS II) protein and mRNA. Inhibition of macrophages with gadolinium chloride abrogates ozone-induced inflammation, mediator production and tissue injury. These data demonstrate, that macrophages and mediators they release contribute to irritant-induced lung injury. Ozone inhalation also caused alterations in the liver, including increased nitric oxide production and protein synthesis suggesting that ozone induces an acute phase response. We speculate that this is mediated by cytokines such as TNF alpha produced by alveolar macrophages. In this regard we noted increased expression of TNF alpha in both lung and liver tissue. Thus cytokines produced locally by macrophages following toxicant exposure may exert pathophysiologic effects outside the target organ.

Regulation of alveolar macrophage and type II cell DNA synthesis: effects of ozone inhalation

A characteristic reaction of the lung to inhaled ozone is an increase in the number of type II epithelial cells and alveolar macrophages (AMs). In the present study, we analyzed mechanisms regulating this response. Acute exposure of rats to ozone (2 parts/million, 3 h) induced expression of proliferating cell nuclear antigen, a marker of cellular proliferation, in both type II cells and AMs. This was maximum 48 h after ozone inhalation. Type II cells and AMs isolated from treated rats at this time also incorporated significantly more [3H]thymidine ([3H]TdR) than cells from control animals. When type II cells and AMs were cocultured, a synergistic increase in [3H]TdR uptake was observed. This appeared to be due to increased DNA synthesis by both cell types. Thus [3H]TdR incorporation by type II cells and AMs cocultured with mitomycin C-treated AMs and type II cells, respectively, was elevated compared with cells cultured alone. Type II cells and AMs plated onto tissue culture inserts, as well as culture supernatants from these cells, were found to stimulate DNA synthesis in AMs and type II cells, respectively. In addition, crude membrane preparations from these cells exhibited growth-promoting activity. Thus the mitogenic effects of both cell types appeared to be mediated by soluble factors and membrane-associated molecules. Ozone inhalation resulted in an increase in the mitogenic activity of AMs treated with mitomycin C and plated on tissue culture inserts toward type II cells and of type II cell culture supernatants toward AMs. These data suggest that type II cell and AM proliferation contributes to the regulation of the number of cells in the lung under normal homeostatic conditions and after ozone-induced injury. Moreover, type II cells and AMs produce paracrine mediators that contribute to cellular proliferative responses.

Increased nitric oxide synthase in the lung after ozone inhalation is associated with activation of NF-kappa B

Acute inhalation of ozone is associated with a inflammatory response characterized by the accumulation of macrophages at sites of tissue injury. These cells, along with resident alveolar epithelial cells, become activated and release cytotoxic and proinflammatory mediators, such as nitric oxide (.NO), that we speculate contribute to toxicity. In these studies we analyzed mechanisms regulating increased .NO synthase activity in lung macrophages and type II cells after ozone inhalation. Brief exposure of rats to ozone (2 ppm for 3 hr) resulted in an increase in .NO production by alveolar macrophages as well as type II cells in response to the inflammatory mediators lipopolysaccharide and interferon gamma. These effects were apparently due to increased expression of inducible .NO synthase (iNOS) protein and mRNA, which were evident in vitro and in situ in histologic sections. .NO production and iNOS protein expression by both macrophages and epithelial cells were blocked by pyrrolidine dithiocarbamate (PDTC), an agent that inhibits activity of nuclear transcription factor kappa B (NF-kappa B). Cells from ozone-treated animals were less sensitive to the effects of PDTC than cells from control animals. Using electrophoretic mobility shift assays, we measured NF-kappa B binding activity in nuclear extracts of cells from control and ozone-exposed animals. Treatment of rats with ozone resulted in a time-dependent increase in NF-kappa B binding activity in both cell types, reaching a maximum in cells isolated 12 to 24 hr after ozone inhalation. Taken together, these data suggest that changes in the activity of NF-kappa B signaling are important in the response of lung macrophages and type II epithelial cells to cytokines after ozone inhalation.

Inhaled nitric oxide primes lung macrophages to produce reactive oxygen and nitrogen intermediates

Inhaled nitric oxide is a selective pulmonary vasodilator used for the treatment of pulmonary hypertension. The potential adverse effects of inhaled nitric oxide are unknown and represent the focus of the present studies. Whereas inhalation of nitric oxide (10 to 100 ppm, 5 h) by Balb/c mice had no effect on the number or type of cells recovered from the lung, a dose-related increase in bronchoalveolar lavage protein was observed, suggesting that nitric oxide induces alveolar epithelial injury. To determine if this was associated with altered alveolar macrophage activity, we quantified production of reactive oxygen and nitrogen intermediates by these cells. Interferon-gamma, alone or in combination with lipopolysaccharide (LPS), induced expression of inducible nitric oxide synthase (iNOS) protein and nitric oxide production by alveolar macrophages. Cells from mice exposed to 20 to 100 ppm nitric oxide produced significantly more nitric oxide and expressed greater quantities of iNOS than cells from control animals. Superoxide anion production and peroxynitrite generation by alveolar macrophages were also increased after exposure of mice to nitric oxide. This was correlated with increased antinitrotyrosine antibody binding to macrophages in histologic sections. Taken together, these data demonstrate that inhaled nitric oxide primes lung macrophages to release reactive oxygen and nitrogen intermediates. Increased production of these mediators by macrophages following inhalation of nitric oxide may contribute to tissue injury.

Distinct inflammatory responses of adherent vascular lung neutrophils to pulmonary irritants

The nature and the extent of the damage that occurs in the lung following exposure to pulmonary irritants vary with the pathogenic agent. In the present studies we determined if this was due to unique functional responses of adherent vascular neutrophils to different irritants. Because of their location within the lung, these cells may be more relevant than circulating neutrophils to the pathophysiology of irritant-induced lung injury. For our studies we used two model irritants, ozone and endotoxin, which cause distinct pathologic effects in the lung. Treatment of rats with ozone resulted in a transient increase (2-fold) in the number of adherent vascular neutrophils in the lung which was maximum 2 hr after exposure and returned to control levels by 12 hr. In contrast, following endotoxin administration, 10-fold greater numbers of adherent neutrophils were recovered from the lung. Moreover, cell number remained elevated 3-fold for up to 48 hr. Unstimulated neutrophils isolated 2-12 hr after endotoxin treatment of rats produced 3 times more superoxide anion than cells from ozone-treated rats. Cells isolated 12-48 hr after endotoxin administration were also sensitized to produce more nitric oxide than cells from ozone-treated rats and to express inducible nitric oxide synthase protein. These data demonstrate that endotoxin and ozone induce distinct patterns of accumulation and functional changes in adherent vascular neutrophils in the lung which may contribute to different pathological processes observed following exposure to these pulmonary irritants.

Role of nitric oxide in acetaminophen-induced hepatotoxicity in the rat

Acetaminophen is a mild analgesic and antipyretic agent known to cause centrilobular hepatic necrosis at toxic doses. Although this may be due to a direct interaction of reactive acetaminophen metabolites with hepatocyte proteins, recent studies have suggested that cytotoxic mediators produced by parenchymal and nonparenchymal cells also contribute to the pathophysiological process. Nitric oxide is a highly reactive oxidant produced in the liver in response to inflammatory mediators. In the present studies we evaluated the role of nitric oxide in the pathophysiology of acetaminophen-induced liver injury. Treatment of male Long Evans Hooded rats with acetaminophen (1 g/kg) resulted in damage to centrilobular regions of the liver and increases in serum transaminase levels, which were evident within 6 hours of treatment of the animals and reached a maximum at 24 hours. This was correlated with expression of inducible nitric oxide synthase (iNOS) protein in these regions. Hepatocytes isolated from both control and acetaminophen-treated rats were found to readily synthesize nitric oxide in response to inflammatory stimuli. Cells isolated from acetaminophen-treated rats produced more nitric oxide than cells from control animals. Production of nitric oxide by cells from both control and acetaminophen-treated rats was blocked by aminoguanidine, a relatively specific inhibitor of iNOS. Arginine uptake and metabolism studies revealed that the inhibitory effects of aminoguanidine were due predominantly to inhibition of iNOS enzyme activity. Pretreatment of rats with aminoguanidine was found to prevent acetaminophen-induced hepatic necrosis and increases in serum transaminase levels. This was associated with reduced nitric oxide production by hepatocytes. Inhibition of toxicity was not due to alterations in acetaminophen metabolism since aminoguanidine had no effect on hepatocyte cytochrome P4502E1 protein expression or N-acetyl-p-benzoquinone-imine formation. Taken together, these data demonstrate that nitric oxide is an important mediator of acetaminophen-induced hepatotoxicity.

Macrophage and interleukin-1 induced nitric oxide production and cytostasis in hamster tumor cells varying in malignant potential

Nitric oxide has been shown to contribute to cytotoxicity in mouse and rat tumor cells. In these studies we examined the role of nitric oxide in cytostasis in hamster tumor cells varying in their malignant potential. Spontaneously transformed hamster embryonic fibroblasts (STHE cells) with low metastatic activity produced significantly greater amounts of nitric oxide in response to interleukin-1 (IL-1) or lipopolysaccharide (LPS)-activated hamster alveolar macrophages (HAM) than did tumor cell lines with high experimental metastatic activity (HET-SR, HET-SR1, STHE-83/20 cells). HET-SR cells, which exhibit low spontaneous metastastic activity, also produced relatively high levels of nitric oxide in response to IL-1, whereas the response of the spontaneously metastatic lines, HET-SR1 and STHE-83/20 cells, was low. IL-1 and HAM also induced cytostasis in nitric oxide-producing STHE and HET-SR cells. However, the nitric oxide synthase inhibitor, N(G)-monomethyl-L-arginine (L-NMMA), had no effect on this activity. These findings, together with the observation that anti-tumor necrosis factor alpha antibody prevented HAM-mediated cytostasis in all of the tumor cell lines demonstrate that nitric oxide is not involved in hamster macrophage-induced tumor cell growth suppression. In contrast to HAM, rat alveolar macrophages, which produced nitric oxide in response to LPS, exerted similar levels of cytostasis toward all of the hamster tumor cell variants, an action that was blocked by L-NMMA in HET-SR, HET-SR1, and STHE-83/20 cells. Thus production of nitric oxide by hamster tumor cells is inversely correlated with their malignant potential. However, nitric oxide does not appear to be involved in IL-1- or HAM-mediated cytostasis toward hamster tumor cells.

Cytostasis is required for IL-1 induced nitric oxide production in transformed hamster fibroblasts

Interleukin-1 (IL-1) is known to inhibit proliferation in some tumor cells. This proinflammatory cytokine also induces nitric oxide production in a variety of cell types. In the present studies we determined if nitric oxide is involved in IL-1 induced growth inhibition in spontaneously transformed hamster embryonic fibroblasts (STHE cells). Both IL-1 alpha and IL-1 beta were found to stimulate nitric oxide production and to reduce 3H-thymidine (TdR) incorporation in high density cultures of STHE cells. However, maximal cytostasis was observed at least 24 h before significant amounts of nitric oxide accumulated in the cultures. In addition, doses of IL-1 which were too low to stimulate nitric oxide synthesis were effective in inducing cytostasis. Furthermore, in low density cultures of STHE cells, IL-1 inhibited DNA synthesis without inducing nitric oxide production. The nitric oxide synthase inhibitor NG-monomethyl-1-arginine (L-NMMA) had no effect on proliferation of cells plated at low density. In contrast, L-NMMA treatment resulted in a 40-60% reduction in IL-1 induced cytostasis in high density cultures. Neutralizing antibodies to IL-1 were found to completely block IL-1 induced cytostasis and nitric oxide production in cells plated at both densities. Although anti-IL-1 alpha and anti-IL-1 beta antibodies were highly specific and did not cross react, anti-tumor necrosis factor-alpha (TNF-alpha) antibody was able to partially suppress activation of STHE cells by both IL-1 alpha and IL-1 beta. These data suggest a potential involvement of endogenous TNF-alpha in IL-1 induced cytostasis and nitric oxide production. Exponentially growing STHE cells produced six-times less nitric oxide than non-proliferating cells. A ten-fold excess of 1-arginine was found to stimulate nitric oxide synthesis, an action that was independent of the rate of cellular proliferation. Taken together these data suggest that nitric oxide is not a major mediator of IL-1 induced cytostasis in STHE cells. Moreover, cytostasis appears to be required for nitric oxide synthesis in these cells.

Role of nitric oxide in hematosuppression and benzene-induced toxicity

It is becoming increasingly apparent that nitric oxide plays a multifunctional role in regulating inflammatory processes in the body. Although nitric oxide and its oxidation products are cytotoxic toward certain pathogens, they can also cause tissue injury and suppress proliferation. Cytokines and growth factors released at sites of inflammation or injury stimulate both immune and nonimmume cells to produce nitric oxide. Nowhere in the body is this more detrimental than in the bone marrow, for the continuous production of hematopoietic precursors is essential for normal blood cell maturation. Our laboratories have discovered that, in response to inflammatory mediators, bone marrow cells readily produce nitric oxide. Nitric oxide production is enhanced by hematopoietic growth factors including interleukin-3, macrophage colony stimulating factor, and granulocyte-macrophage colony-stimulating factor. When bone marrow cells produce nitric oxide, hematopoiesis is impaired, an effect that is potentiated by colony-stimulating factors. Treatment of mice with benzene, which suppresses bone marrow cell development, was found to markedly enhance the ability of bone marrow cells to produce nitric oxide in response to inflammatory mediators alone and in combination with hematopoietic growth factors. Taken together, these data suggest that nitric oxide may be an important mediator of benzene-induced bone marrow suppression.

Mechanisms regulating macrophage-induced nitric oxide production by spontaneously transformed hamster fibroblasts

Nitric oxide has been implicated as an important effector molecule involved in tumor cell growth and cytotoxicity. In these studies we examined mechanisms regulating nitric oxide production by hamster tumor cells. Cocultures of hamster alveolar macrophages (HAM) and spontaneously transformed hamster embryonic fibroblasts (STHE cells) produced significant quantities of nitric oxide in response to lipopolysaccharide (LPS). Culture supernatants from HAM treated with LPS also stimulated nitric oxide production by STHE cells, whereas tumor cell culture supernatants had no effect on HAM. These data, together with the findings that paraformaldehyde treatment of STHE cells, but not macrophages, completely abrogated nitric oxide production in the cocultures demonstrate that the tumor cells were the source of this mediator. In contrast to STHE cells, STHE-83/20 cells, a highly malignant variant, did not produce nitric oxide in response to HAM or HAM culture supernatants even in the presence of LPS. Both anti-tumor necrosis factor-alpha (TNF-alpha) and anti-interleukin-1alpha (IL-1alpha) antibodies inhibited HAM-induced nitric oxide production by STHE cells. However, the kinetics of their effects were different. Moreover, although the nitric oxide stimulating activity in HAM culture supernatants was abrogated by anti-TNF-alpha antibody, it was only minimally reduced by anti-IL-1alpha antibody. These data demonstrate that TNF-alpha and IL-1alpha play distinct roles in induction of nitric oxide synthesis in STHE cells. HAM were also found to suppress proliferation of STHE cells, an effect that was inhibited by anti-TNF-alpha antibody, but not NG-monomethyl-L-arginine, which blocks nitric oxide synthase. Abrogation of macrophage-induced cytostasis in STHE cells by anti-TNF-alpha antibody was associated with decreased nitric oxide production. Thus TNF-alpha released by macrophages may indirectly activate STHE cells for nitric oxide synthesis by suppressing tumor cell proliferation.

Osteopontin inhibits nitric oxide production and cytotoxicity by activated RAW264.7 macrophages

Osteopontin (OPN), a secreted acidic phosphoglycoprotein found in many tissues and body fluids, is produced in increased amounts in response to certain infections and after malignant transformation. In this study we examined the action of OPN on macrophage cytotoxicity and nitric oxide (NO) synthesis. A human OPN cDNA was cloned into the bacteriophage T7-based vector, pET8C, and the encoded protein purified from an induced culture of Escherichia coli carrying the plasmid. Recombinant OPN inhibited NO production by macrophage-like RAW264.7 cells stimulated with lipopolysaccharide plus interferon-gamma. OPN also inhibited the cytolytic activity of the activated macrophages toward NO-sensitive P815 mastocytoma cells, an action that was blocked by the NO synthase inhibitor, NG-monomethyl-L-arginine. Inhibition of NO production correlated with an OPN-dependent decrease in the abundance of inducible NO synthase mRNA. The shape of the dose-response curve, with a maximal effect over a narrow range of OPN concentrations, suggested a complex interaction of OPN with cell surface receptors. Our data support the hypothesis that tumor-cell-derived OPN functions to protect the tumor cells from macrophage-mediated destruction.

Macrophages, Inflammatory Mediators, and Lung Injury

Macrophages and the various inflammatory mediators they release have been implicated in lung injury induced by a number of different pulmonary toxicants. Exposure of humans or experimental animals to toxic doses of xenobiotics such as ozone, bleomycin, or mineral dusts results in an accumulation of macrophages in the lung. These cells are activated to release increased amounts of proinflammatory and cytotoxic mediators such as hydrogen peroxide, nitric oxide, peroxynitrite, bioactive lipids, interleukin-1, and tumor necrosis factor-alpha. Each of these mediators has the capacity to induce tissue injury directly and/or augment the inflammatory response. When animals are treated with agents that block macrophage functioning and/or mediator release, pulmonary injury induced by agents such as ozone or endotoxin is abrogated. Conversely, treatment of animals with macrophage activators enhances toxicant-induced lung damage. These data provide direct support for a role of macrophages and inflammatory mediators in pulmonary toxicity.

Inhibition of ozone-induced nitric oxide synthase expression in the lung by endotoxin

Inhalation of the pulmonary irritant ozone is associated with an accumulation of macrophages in the lung. These cells, along with type II epithelial cells, are activated to release increased quantities of hydrogen peroxide and nitric oxide, two reactive mediators that have been implicated in tissue injury. In the present studies we determined whether pretreatment of rats with bacterially derived endotoxin, which modulates oxidant levels in tissues, could abrogate the effects of ozone on lung injury and nitric oxide production. Acute exposure of rats to ozone (2 parts per million, 3 h) resulted in nitric oxide production in the lung as measured by electron paramagnetic resonance spin trapping. This was correlated with expression of inducible nitric oxide synthase (iNOS) mRNA in the lung as determined by in situ hybridization. Particularly high levels of iNOS were evident in alveolar macrophages and type II cells. Alveolar macrophages isolated from ozone-treated rats also expressed increased iNOS mRNA and protein as measured by Northern and Western blotting, respectively, and produced more nitric oxide compared with cells from air-exposed animals. Treatment of rats with endotoxin (5 mg/kg, intravenously), 30 min prior to ozone, was found to abrogate ozone-induced increases in iNOS mRNA and protein expression, as well as nitric oxide production by alveolar macrophages. This was associated with a reduction in ozone-induced tissue injury as determined by levels of lung lavage fluid protein. Ozone inhalation also resulted in a reduction in intracellular glutathione in alveolar macrophages, an effect that was blocked by endotoxin administration. Taken together, these data provide evidence that the protective effects of endotoxin against ozone-induced injury are mediated, at least in part, by alterations in levels of lung oxidants and antioxidants.

Stimulation of nitric oxide production in rat lung lavage cells by anti-Mac-1beta antibody: effects of ozone inhalation

Acute inhalation of the pulmonary irritant ozone is associated with an inflammatory response characterized by increased numbers of macrophages in the lung that release elevated quantities of nitric oxide. The accumulation of phagocytes in the lung is dependent on expression of leukocyte adhesion molecules including Mac-1. In the present studies, we determined whether activation of the Mac-1 receptor is involved in regulating nitric oxide production by lung phagocytes, and whether this response is modified following acute ozone inhalation. Cells were isolated from the lung by bronchoalveolar lavage 48 h after exposure of female Sprague-Dawley rats to air or ozone (2 parts per million, for 3 h). Anti-Mac-1beta antibody, but not anti-Mac-1alpha antibody, stimulated nitric oxide production by cells from both air- and ozone-exposed animals. Cells from ozone-exposed rats produced more nitric oxide and expressed greater quantities of inducible nitric oxide synthase mRNA than did cells from air-exposed animals. Production of nitric oxide in response to anti-Mac-1beta was also found to be augmented by cross-linking of the Mac-1beta receptor. Pretreatment of lavage cells with granulocyte/macrophage colony-stimulating factor (GM-CSF), which activates phagocytes, enhanced the expression of Mac-1beta and increased anti-Mac-1beta-induced nitric oxide production by the cells. Lavage cells from ozone-exposed animals were more responsive to GM-CSF than were cells from control animals. Taken together, these data suggest that the Mac-1beta adhesion molecule may contribute to phagocyte activation and mediator release during ozone-induced inflammatory reactions in the lung.

Sinusoidal lining cells and hepatotoxicity

Treatment of experimental animals with toxic doses of acetaminophen, carbon tetrachloride, phenobarbital, galactosamine, or endotoxin results in an accumulation of macrophages in the liver. These mononuclear phagocytes, as well as hepatic endothelial cells and stellate cells, are activated to release increased amounts of proinflammatory and cytotoxic mediators including hydrogen peroxide, superoxide anion, nitric oxide, bioactive lipids, interleukin 1, platelet activating factor, and tumor necrosis factor alpha. Each of these mediators has the capacity to induce tissue injury directly and/or augment the inflammatory response. When animals are treated with agents that block macrophage functioning and/or mediator release, xenobiotic-induced hepatotoxicity is reduced. In contrast, treatment of animals with macrophage activators augments toxicant-induced liver damage. These data provide direct support for a role of macrophages and inflammatory mediators in hepatotoxicity.

Taurine protects rat bronchioles from acute ozone-induced lung inflammation and hyperplasia

Ozone is a potent respiratory irritant known to induce lung injury in humans and experimental animals. The present studies determined if ozone-induced lung inflammation was modified by pretreatment of animals with taurine, a detoxifying antioxidant. Rats were pretreated for 10 days with 5% taurine in their drinking water (controls received water only) prior to exposure to 2 ppm ozone for 3 h. At 2, 6, 12, 24, 48, and 72 h after ozone exposure, rats were anesthetized and the lungs were perfusion-fixed for histopathologic evaluation. An additional group of rats was used to examine bronchoalveolar lavage cell counts and hydroxyproline levels. A count of bronchoalveolar lavage cells 48 h after ozone exposure showed significantly fewer total inflammatory cells and fewer polymorphonuclear leukocytes accompanied by a reduction in hydroxyproline in the lavage fluid of ozone-exposed rats pretreated with taurine compared to rats that did not receive taurine. Light microscopy revealed an inflammatory cell infiltrate in the lungs of rats exposed to ozone. This was followed by focal hyperplasia in the terminal and respiratory bronchioles. Rats pretreated with taurine and then exposed to ozone showed none of these alterations. In addition, although there was a significant reduction in cell proliferation as measured by DNA precursor incorporation in the lungs of rats pretreated with taurine prior to ozone exposure compared to unsupplemented rats, the distribution of labeled cells was the same in taurine supplemented and unsupplemented groups. Also, significantly higher levels of taurine were found in the plasma, whole blood, and lavage fluid of rats pretreated with dietary taurine compared to rats that received water only. The results suggest that supplemental taurine protects rat lung epithelium from acute ozone-induced lung inflammation and hyperplasia.

Unique patterns of regulation of nitric oxide production in fibroblasts

The pathways regulating rat and mouse embryonic and lung fibroblast nitric oxide production were analyzed in an attempt to evaluate the potential role of these cells in nonspecific host defense and inflammation. Interleukin-1 beta (IL-1 beta) was found to be the strongest single activator in all types of fibroblasts examined. In addition, lipopolysaccharide (LPS) was synergistic with IL-1 beta or tumor necrosis factor-alpha (TNF-alpha) in induction of nitric oxide synthesis. These patterns of responsiveness are not observed in macrophages and may be significant in initiation of early host defense processes, before specific interferon-gamma (IFN-gamma)-mediated immune responses have become operative. Rat and mouse fibroblasts were also found to produce nitric oxide when primed with IFN-gamma and simultaneously treated with IL-1, TNF-alpha, or LPS. The doses of IFN-gamma effective in priming fibroblasts for nitric oxide production were as low as 1-10 U/ml. Furthermore, effective triggering doses of LPS, TNF-alpha, and IL-1 were 10 ng/ml, 100 U/ml, and 0.2 ng/ml, respectively. These results demonstrate that fibroblasts are activated more readily to produce nitric oxide than interstitial macrophages and may be the major source of this mediator in tissues. Immunohistochemical studies demonstrated that fibroblasts are heterogeneous with respect to inducible nitric oxide synthase expression with the majority of cells not involved in the response. Fibroblasts were also found to be distinct from macrophages in their sensitivity to the suppressive effects of transforming growth factor-beta, which in fibroblasts inhibited both IFN-gamma plus LPS- and IFN-gamma plus TNF-alpha-induced nitric oxide production. At the stage of growth crisis, a dramatic increase in nitric oxide production was observed in rat fibroblasts in response to IFN-gamma or TNF-alpha that may be directly correlated with cellular senescence. Taken together, our data suggest that mouse and rat fibroblasts are potential effectors in both IFN-gamma-dependent and -independent nitric oxide-mediated processes and that the patterns regulating nitric oxide metabolism in these cells are distinct from those of macrophages.

Inhibition of macrophages with gadolinium chloride abrogates ozone-induced pulmonary injury and inflammatory mediator production

Acute inhalation of toxic doses of ozone (O3) induces macrophage accumulation in the lung and the release of cytotoxic and proinflammatory mediators. To evaluate the role of macrophages and their mediators in the pathophysiologic response of the lung to O3, we examined the effects of the macrophage inhibitor, gadolinium chloride (GdCl3), on O3-induced inflammation, mediator production, and lavage fluid protein levels. Rats were pretreated with GdCl3 (7 mg/kg, intravenously) or control 24 h prior to exposure to air or O3 (2 parts per million, 3 h). Animals were killed 48 h after exposure. GdCl3 pretreatment of rats was found to abrogate O3-induced increases in the number of cells, as well as the amount of protein recovered in bronchoalveolar lavage fluid. Following GdCl3 pretreatment of rats, lung lavage cells consisting of > 90% macrophages were found to produce significantly less nitric oxide and express less inducible nitric oxide synthase (iNOS) when compared to cells from rats exposed to O3. O3-induced alterations in superoxide anion production by alveolar macrophages, both in vitro and in situ, were also attenuated by GdCl3 pretreatment of rats. In addition, increases in tumor necrosis factor alpha (TNF-alpha) and fibronectin in lung tissue induced by O3 were reduced. Taken together, these data provide support for the hypothesis that macrophages contribute to the pathogenesis of O3-induced lung injury.

Hepatic nitric oxide production following acute endotoxemia in rats is mediated by increased inducible nitric oxide synthase gene expression

In the present studies, we analyzed the effects of acute endotoxemia on hepatocyte nitric oxide production and functional activity. Treatment of rats with 5 mg/kg of lipopolysaccharide (LPS), which induces acute endotoxemia, caused an increase in nitric oxide production in the liver, as measured by electron paramagnetic spin trapping, which was evident within 6 hours. This was associated with expression of inducible nitric oxide synthase (iNOS) messenger (m) RNA in hepatocytes and in sinusoidal cells throughout the liver lobule. Acute endotoxemia also caused alterations in hepatic structure, including hypertrophy, vacuolization, and chromosomal emargination, however these changes were not apparent for 24 to 48 hours. Hepatocytes isolated from endotoxemic rats released increased amounts of nitric oxide, measured by nitrite production, in response to interferon gamma (gamma-IFN) alone or in combination with LPS, tumor necrosis factor alpha, macrophage-colony stimulating factor, granulocyte/macrophage-colony stimulating factor, or hepatocyte growth factor. These results show that hepatocytes are sensitized by acute endotoxemia to respond to inflammatory mediators and growth factors. Increased nitrite production by hepatocytes was due to increased expression of iNOS mRNA and protein and was correlated with the time following induction of acute endotoxemia. Thus, cells isolated 48 hours after induction of acute endotoxemia released significantly more nitrite than cells recovered after 6 hours, a response that was not due to alterations in hepatocyte viability. Hepatocytes isolated from endotoxemic rats also exhibited a marked increase in proliferative capacity when compared with cells from control rats. Nitric oxide production by hepatocytes in vitro was associated with inhibition of cell growth and protein synthesis, which was reversed by the nitric oxide synthase inhibitor, NG-monomethyl-l-arginine (L-NMMA). Agarose gel electrophoresis showed extensive cytoplasmic DNA fragmentation in hepatocytes treated with LPS and gamma-IFN, a characteristic of apoptosis, which was also reversed by L-NMMA. These results, together with our findings that treatment of rats with an inhibitor of nitric oxide synthase partially reversed the structural alterations in the liver associated with acute endotoxemia suggest that nitric oxide may contribute to the pathophysiologic response to this bacterially derived toxin.

Modulation of macrophage functioning abrogates the acute hepatotoxicity of acetaminophen

Acetaminophen is a mild analgesic and antipyretic agent that is safe and effective when taken in therapeutic doses. Ingestion of overdoses, however, may lead to acute liver failure accompanied by centrilobular degeneration and necrosis. Although the toxicity of acetaminophen is generally thought to be caused by direct interaction of its reactive metabolites with cellular macromolecules, recent studies have suggested that nonparenchymal cells also may contribute to tissue injury indirectly through the release of cytotoxic mediators. We analyzed the potential role of hepatic macrophages in acetaminophen hepatotoxicity by examining the effects of modulating the activity of these cells on tissue injury. Treatment of male Long Evans Hooded rats with acetaminophen (800 mg/kg) was found to induce extensive centrilobular hepatic necrosis. Pretreatment of the rats with either dextran sulfate or gadolinium chloride, two inhibitors of hepatic macrophage functioning, completely blocked hepatic necrosis, as well as increases in serum transaminase levels induced by acetaminophen. Interestingly, treatment of rats with the macrophage activator, lipopolysaccharide (LPS), also reduced tissue injury induced by acetaminophen. To exclude the possibility that the effects of gadolinium chloride, dextran sulfate, or LPS were due to alterations in acetaminophen metabolism, we analyzed the effects of these agents on various pharmacokinetic properties of this analgesic. Dextran sulfate and gadolinium chloride had no effect on the half-life of a low dose of acetaminophen (20 mg/kg), or on the activity of any of its individual pathways of metabolism, including the formation of acetaminophen-mercapturic acid.(ABSTRACT TRUNCATED AT 250 WORDS)

Flow cytometry reveals subpopulations of murine epidermal cells that are refractory to induction of cytochrome P-4501A1 by beta-naphthoflavone

Topical application of beta-naphthoflavone to CD-1 mice induced an 87-fold increase in epidermal 7-ethoxyresorufin O-dealkylation activity per cell and a many-fold increase in epidermal cytochrome P-4501A1 (CYP1A1) concentration. Flow cytometric analysis of individual epidermal cells from acetone-treated and beta-naphthoflavone-treated mice using a monoclonal antibody for CYP1A1 indicated that 50% to 60% of the isolated epidermal cells were refractory to beta-naphthoflavone induction of CYP1A1. Examination of the differences between responsive and nonresponsive epidermal cells from beta-naphthoflavone-treated mice revealed that 70% of the low CYP1A1-containing cells (noninduced) separated by flow cytometry were basal cells and only 12% were suprabasal differentiated cells. In contrast, about 50% of the high CYP1A1-containing induced cells separated by flow cytometry from the epidermis of mice treated with beta-naphthoflavone were suprabasal cells and 35% were basal cells. These results indicate that topical application of beta-naphthoflavone increased the level of CYP1A1 in about 80% of the separated suprabasal cells and in about 35% of the separated basal cells.

Macrophages and inflammatory mediators in tissue injury

Tissue injury induced by a diverse group of xenobiotics appears to involve both direct and indirect damage to target cells. Thus, while chemicals may act directly on target cells resulting in toxicity, they may also act indirectly by recruiting and activating resident and inflammatory tissue macrophages. Macrophages are potent secretory cells that release an array of mediators, including proinflammatory and cytotoxic cytokines and growth factors, bioactive lipids, hydrolytic enzymes, reactive oxygen intermediates, and nitric oxide--each of which has been implicated in the pathogenesis of tissue injury. The potential role of macrophages and their mediators in tissue injury has been extensively investigated in the lung and the liver. In both of these tissues, xenobiotics induce localized macrophage accumulation and mediator release. Furthermore, when macrophage functioning is blocked, pulmonary and hepatic injury-induced agents such as ozone, bleomycin, acetaminophen, carbon tetrachloride, and galactosamine are reduced. These data provide direct support for the hypothesis that macrophages and the mediators they release contribute to xenobiotic-induced tissue injury.

Modulation of macrophage functioning abrogates the acute hepatotoxicity of acetaminophen

Acetaminophen is a mild analgesic and antipyretic agent that is safe and effective when taken in therapeutic doses. Ingestion of overdoses, however, may lead to acute liver failure accompanied by centrilobular degeneration and necrosis. Although the toxicity of acetaminophen is generally thought to be caused by direct interaction of its reactive metabolites with cellular macromolecules, recent studies have suggested that nonparenchymal cells also may contribute to tissue injury indirectly through the release of cytotoxic mediators. We analyzed the potential role of hepatic macrophages in acetaminophen hepatotoxicity by examining the effects of modulating the activity of these cells on tissue injury. Treatment of male Long Evans Hooded rats with acetaminophen (800 mg/kg) was found to induce extensive centrilobular hepatic necrosis. Pretreatment of the rats with either dextran sulfate or gadolinium chloride, two inhibitors of hepatic macrophage functioning, completely blocked hepatic necrosis, as well as increases in serum transaminase levels induced by acetaminophen. Interestingly, treatment of rats with the macrophage activator, lipopolysaccharide (LPS), also reduced tissue injury induced by acetaminophen. To exclude the possibility that the effects of gadolinium chloride, dextran sulfate, or LPS were due to alterations in acetaminophen metabolism, we analyzed the effects of these agents on various pharmacokinetic properties of this analgesic. Dextran sulfate and gadolinium chloride had no effect on the half-life of a low dose of acetaminophen (20 mg/kg), or on the activity of any of its individual pathways of metabolism, including the formation of acetaminophen-mercapturic acid.(ABSTRACT TRUNCATED AT 250 WORDS)

Distinct actions of benzene and its metabolites on nitric oxide production by bone marrow leukocytes

Benzene is a widely used industrial solvent known to cause bone marrow depression. This is associated with increased production of reactive oxygen metabolites and nitric oxide by bone marrow phagocytes, which have been implicated in hematotoxicity. Benzene metabolism to phenolic intermediates appears to be an important factor in bone marrow toxicity. In the present studies, we compared the effects of benzene and several of its metabolites on nitric oxide production by murine bone marrow leukocytes. Bone marrow cells readily produced nitric oxide in response to the inflammatory mediators lipopolysaccharide (LPS) and interferon-gamma (IFN-gamma). Treatment of mice with benzene (800 mg/kg), or its metabolites hydroquinone (100 mg/kg), 1,2,4-benzenetriol (25 mg/kg), or p-benzoquinone (2 mg/kg), at doses that impair hematopoiesis, sensitized bone marrow leukocytes to produce increased amounts of nitric oxide in response to LPS and IFN-gamma. Granulocyte-macrophage colony-stimulating factor (GM-CSF) and macrophage colony-stimulating factor (M-CSF) augmented bone marrow leukocyte production of nitric oxide induced by inflammatory mediators. Benzene, as well as its metabolites, markedly increased the sensitivity of the cells to both GM-CSF and M-CSF. Cells from hydroquinone- or 1,2,4-benzenetriol-treated mice were significantly more responsive to the inflammatory cytokines and growth factors than cells isolated from benzene- or p-benzoquinone-treated mice, suggesting that the phenolic metabolites of benzene are important biological reactive intermediates. Because nitric oxide suppresses cell growth and can be metabolized to mutagens and carcinogens, the ability of benzene and its metabolites to modulates its production in the bone marrow may be important in their mechanism of action.

Distinct biochemical responses of hepatic macrophages and endothelial cells to platelet-activating factor during endotoxemia

Acute endotoxemia is associated with activation of hepatic macrophages and endothelial cells. These cells release a variety of inflammatory mediators that have been implicated in tissue injury. In the present studies, we analyzed the biochemical responses of these cells to platelet-activating factor (PAF), a lipid autacoid released during hepatic inflammatory responses. To induce acute endotoxemia, rats were injected intravenously with lipopolysaccharide (LPS). Using the calcium sensitive fluorescent indicator dye Indo-1, we found that PAF induced a rapid and transient increase in intracellular calcium in both hepatic macrophages and endothelial cells. Induction of acute endotoxemia resulted in an increase in the amount of calcium mobilized by both cell types. Although endothelial cells from control rats were less responsive to PAF than macrophages, these cells were more sensitive to in vivo endotoxin. PAF was also found to cause a rapid decrease in intracellular pH in hepatic macrophages that was quantified by fluorescence image analysis using the pH sensitive dye SNAFL-calcein. This decrease occurred more rapidly in macrophages from endotoxemic rats. In cells from both control and endotoxemic rats, the effects of PAF on intracellular pH were inhibited by the specific PAF antagonist triazolam. In contrast to hepatic macrophages, PAF had no effect on intracellular pH in endothelial cells from either control or endotoxemic rats. Ligand binding studies demonstrated that both hepatic macrophages and endothelial cells possess high affinity binding sites for PAF. Macrophages expressed 6- to 7-fold more binding sites/cell than endothelial cells and exhibited a higher Kd. Whereas treatment of rats with LPS had no effect on the Kd for PAF binding to macrophages or on the number of binding sites, a significant increase in both of these receptor characteristics was observed in endothelial cells. Taken together, the present data suggest that the biochemical responses of endothelial cells and macrophages to PAF are distinct. Furthermore, cellular activation induced by PAF in endothelial cells appears to be independent of changes in intracellular pH.

Lymphocyte-mediated nitric oxide production by rat endothelial cells

Lymphocyte migration from the blood to sites of tissue injury is mediated, in part, through the interaction of these cells with endothelial cells lining the vessel walls. The ability of endothelial cells to produce nitric oxide may be important in this process. We found that the addition of the nonspecific lymphocyte activators lipopolysaccharide (LPS) or concanavalin A (Con A) to rat hepatic endothelial cell cultures from control or endotoxemic rats markedly enhanced the ability of these cells to produce nitric oxide. In contrast, wheat germ agglutinin (WGA) and phytohemagglutinin (PHA) had no effect on nitric oxide release. Coculture of endothelial cells with lymphocyte-rich preparations of rat thymocytes or splenocytes stimulated endothelial cell nitric oxide production. This response was enhanced by LPS or Con A and to a lesser extent by WGA or PHA. In contrast to endothelial cells, thymocytes and splenocytes did not produce nitric oxide either in the presence or absence of lymphocyte mitogens. Increased production of nitric oxide by endothelial cells in response to lymphocytes and lymphocyte mitogens was due, at least in part, to increased expression of protein for an inducible form of nitric oxide synthase, as measured by Western blotting. Stimulation of endothelial cell nitric oxide production by thymocytes and splenocytes was inhibitable by the specific nitric oxide synthase inhibitor NG-monomethyl-L-arginine and dependent on cell-cell contact. Thus, nitric oxide production by endothelial cells was reduced when the lymphocytes were physically separated from the endothelial cells using cell culture inserts. We hypothesize that nitric oxide released by endothelial cells increases vascular permeability, thereby allowing the extravasation of lymphocytes into the surrounding tissue, a process that may be important in inflammation, tissue injury, and/or wound healing.

Induction of hepatic Ito cell nitric oxide production after acute endotoxemia

Nitric oxide is a highly reactive mediator released in the liver by hepatocytes, Kupffer cells and endothelial cells during endotoxin-induced inflammation. In this study we determined whether Ito cells also produce nitric oxide after exposure to endotoxin. For induction of endotoxemia, rats were injected intravenously with Escherichia coli lipopolysaccharide (2.5 mg/kg). Ito cells were isolated from the animals 48 hr later by means of in situ perfusion of the liver with protease and collagenase followed by purification on an arabinogalactan gradient. Ito cells from untreated and endotoxemic rats were found to produce low levels of nitric oxide in response to interferon-gamma. In both cell types, this response depended on L-arginine and was blocked by NG-monomethyl-L-arginine, a specific nitric oxide synthase inhibitor. Cells from rats treated with endotoxin produced significantly more nitric oxide than did cells from untreated animals; this was due, at least in part, to increased expression of protein for an inducible form of nitric oxide synthase. These cells also responded to stimulation with lipopolysaccharide in vitro, as well as the combination of interferon-gamma and lipopolysaccharide, which was synergistic in stimulating nitric oxide production. Tumor necrosis factor-alpha and macrophage colony-stimulating factor were also found to stimulate nitric oxide production by Ito cells from endotoxemic rats. In addition, in these cells, tumor necrosis factor-alpha synergized with interferon-gamma in inducing nitric oxide production. The combination of interferon-gamma and lipopolysaccharide was also found to inhibit Ito cell DNA synthesis, as measured on the basis of [3H]-thymidine uptake.(ABSTRACT TRUNCATED AT 250 WORDS)

Production of nitric oxide and peroxynitrite in the lung during acute endotoxemia

Nitric oxide is a short-lived cytotoxic mediator that has been implicated in the pathogenesis of endotoxin-induced tissue injury and septic shock. In the present studies we determined whether this mediator is produced in the lung during acute endotoxemia. We found that intravenous injection of rats with bacterially derived lipopolysaccharide (LPS), a condition that induces acute endotoxemia, caused a time-dependent increase in inducible nitric oxide synthase (iNOS) mRNA expression in the lung, which reached a maximum after 24 h. This was correlated with nitric oxide production in the lung as measured by electron paramagnetic spin trapping, which was detectable within 6 h. Alveolar macrophages (AMs) and interstitial macrophages (IMs) isolated from rats 6-12 h after induction of acute endotoxemia were also found to exhibit increased nitric oxide production in response to in vitro stimulation with interferon-gamma (IFN-gamma) and LPS measured by nitrite accumulation in the culture medium. The effects of acute endotoxemia on nitric oxide production by these cells were, however, transient and returned to control levels by 24 h in AMs and 36 h in IMs. Interestingly, although nitrite accumulation in the culture medium of IMs isolated 48 h after induction of acute endotoxemia and stimulated with low concentrations of IFN-gamma and LPS was reduced, when compared with cells from control animals, these cells, as well as AMs, continued to express high levels of iNOS protein and mRNA. This was correlated with increased peroxynitrite production by the cells. Peroxynitrite has been shown to act as a nitrating agent and can generate nitrotyrosine residues in proteins. Using a specific antibody and immunohistochemistry, we found evidence of nitrotyrosine residues in sections of lungs 48 h after treatment of rats with endotoxin. These data suggest that nitric oxide produced by IMs and AMs can react with superoxide anion to form peroxynitrite. Taken together, the present studies demonstrate that AMs and IMs are activated following acute endotoxemia to produce reactive nitrogen intermediates and that both cell types contribute to inflammatory responses in the lung.

Pulmonary and hepatic effects of inhaled ozone in rats

Nitric oxide is a highly reactive molecule that has been implicated in host defense and in tissue injury. In the present studies we analyzed the effects of brief exposure of rats to inhaled ozone on production of this mediator by lung macrophages and type II epithelial cells. We found that ozone exposure (1-2 ppm, 3 hr) induced a marked increase in spontaneous nitric oxide production by alveolar (AM) and interstitial macrophages, as well as type II cells. These effects were apparently due to increased expression of inducible nitric oxide synthase protein and mRNA, which was evident in vitro in isolated cells and in situ in histologic sections. Macrophages and epithelial cells from ozone-treated rats were also sensitized to produce increased amounts of nitric oxide in response to inflammatory cytokines such as interferon-gamma, a response that was also mediated by inducible nitric oxide synthase. Unexpectedly, we also discovered that brief inhalation of ozone caused dramatic effects on the liver, including increased production of nitric oxide by hepatocytes and enhanced protein synthesis. These data suggest that this inhaled irritant induces an acute phase response. Additional studies indicate that AM from ozone-treated rats produced significantly more tumor necrosis factor-alpha and interleukin-1 than did cells from control animals. Elevated levels of tumor necrosis factor-alpha were also noted immunohistochemically in both lung and liver tissue. These results indicate that the extrapulmonary effects of ozone may be mediated by inflammatory cytokines released by activated lung macrophages.(ABSTRACT TRUNCATED AT 250 WORDS)

Distinct patterns of nitric oxide production in hepatic macrophages and endothelial cells following acute exposure of rats to endotoxin

Hepatic macrophages and endothelial cells play an important role in the clearance of endotoxin from the portal circulation. These cells are activated by endotoxin to release reactive mediators including superoxide anion, hydrogen peroxide, and nitric oxide, which have been implicated in hepatic inflammation and tissue injury. In the present studies we analyzed mechanisms regulating the production of nitric oxide by hepatic macrophages and endothelial cells following in vivo exposure to endotoxin. Rats were injected intravenously with Escherichia coli lipopolysaccharide (LPS, 5 mg/kg). Cells were isolated from the animals 48 h later by in situ perfusion of the liver with collagenase and pronase followed by differential centrifugation and centrifugal elutriation. We found that macrophages and endothelial cells from both untreated and endotoxin-treated rats readily synthesized nitric oxide following in vitro stimulation with interferon-gamma (IFN-gamma) and LPS alone and in combination. This response was dependent on l-arginine and was blocked by two nitric oxide synthase inhibitors, NG-monomethyl-l-arginine and l-canavanine. Macrophages produced more nitric oxide in response to LPS or LPS plus IFN-gamma than endothelial cells. In addition, nitric oxide production by both cell types in response to LPS plus IFN-gamma was increased after treatment of rats with endotoxin. Macrophages appeared to be more sensitive than endothelial cells to the in vivo effects of this inflammatory stimulus. Northern and Western blot analysis demonstrated that nitric oxide production by macrophages and endothelial cells in response to LPS plus IFN-gamma was due to increased expression of an inducible form of nitric oxide synthase (iNOS) mRNA and protein. Using fluorescence image analysis, iNOS protein was found to be localized in the cytoplasm of the cells. Treatment of rats with endotoxin was associated with increased expression of iNOS protein in the macrophages. The phorbol ester 12-O-tetradecanoyl-phorbol-13-acetate (TPA) also stimulated nitric oxide production by macrophages and endothelial cells from endotoxin-treated rats, although not as effectively as LPS and IFN-gamma. Macrophages were more responsive than endothelial cells to TPA. Furthermore, depletion of the cells of glutathione using buthionine sulfoximine had no major effect on nitric oxide production by macrophages but resulted in small but significant inhibition in endothelial cells. This suggests that this sulfhydryl-containing tripeptide does not regulate intracellular levels of reactive nitrogen intermediates in activated macrophages.(ABSTRACT TRUNCATED AT 400 WORDS)