Tumor necrosis factor alpha and interferon gamma modulate the expression of the vitronectin receptor (integrin beta 3) in human endothelial cells.

In this paper we show that tumor necrosis factor alpha (TNF alpha) and interferon gamma (IFN gamma) alter the expression of extracellular matrix receptors (integrins) in cultured human endothelial cells. Endothelial cells express at their surface integrins of the beta 1 and beta 3 groups that include receptors for fibronectin, collagen, laminin, and vitronectin. After treatment for 72 h with a combination of TNF alpha and IFN gamma, the level of the vitronectin receptor (alpha v beta 3) at the cell surface decreases by 70%, whereas the amounts of the beta 1 integrins remain unchanged. The decreased expression of the alpha v beta 3 complex at the cell surface is due to a selective effect of TNF alpha and IFN gamma on the regulation of the beta 3 subunit synthesis at the translational level. In fact, although the steady state levels of the mRNA for the beta 3 subunit are comparable in control and treated cells, the overall synthesis of the beta 3 subunit is decreased by a factor of 70%. No significant alteration of the synthesis of the companion alpha v subunit is detectable in cytokine-treated cells. As a consequence of the decreased expression of the receptor, cytokine-treated cells show decreased ability to adhere to vitronectin but adhere normally to fibronectin. These data show that two important inflammatory mediators, TNF alpha and IFN gamma, can modify the interaction of endothelial cells with the extracellular matrix by selectively altering the expression of specific cell surface integrin complexes.     

PAR-2 activation and LPS synergistically enhance inflammatory signaling in airway epithelial cells by raising PAR expression level and interleukin-8 release

Protease-activated receptors (PARs) are involved in the contribution of airway epithelial cells to the development of inflammation by release of pro- and anti-inflammatory mediators. Here, we evaluated in epithelial cells the influence of LPS and continuous PAR activation on PAR expression level and the release of the proinflammatory chemokine IL-8. We studied primary human small airway epithelial cells and two airway epithelial cell lines, A549 and HBE cells. LPS specifically upregulated expression of PAR-2 but not of PAR-1. Exposure of epithelial cells to PAR-1 or PAR-2 agonists increased the PAR-1 expression level. The PAR-2 agonist exhibited higher potency than PAR-1 activators. However, the combined exposure of epithelial cells to LPS and PAR agonists abrogated the PAR-1 upregulation. The PAR-2 expression level was also upregulated after exposure to PAR-1 or PAR-2 agonists. This elevation was higher than the effect of PAR agonists on the PAR-1 level. In contrast to the PAR-1 level, the PAR-2 level remained elevated under concomitant stimulation with LPS and PAR-2 agonist. Furthermore, activation of PAR-2, but not of PAR-1, caused production of IL-8 from the epithelial cells. Interestingly, both in the epithelial cell line and in primary epithelial cells, there was a potentiation of the stimulation of the IL-8 synthesis and release by PAR-2 agonist together with LPS. In summary, these results underline the important role of PAR-2 in human lung epithelial cells. Moreover, our study shows an intricate interplay between LPS and PAR agonists in affecting PAR regulation and IL-8 production.     

Thrombin induces proteinase-activated receptor-1 gene expression in endothelial cells via activation of Gi-linked Ras/mitogen-activated protein kinase pathway

We addressed the mechanisms of restoration of cell surface proteinase-activated receptor-1 (PAR-1) by investigating thrombin-activated signaling pathways involved in PAR-1 re-expression in endothelial cells. Exposure of endothelial cells transfected with PAR-1 promoter-luciferase reporter construct to either thrombin or PAR-1 activating peptide increased the steady-state PAR-1 mRNA and reporter activity, respectively. Pretreatment of reporter-transfected endothelial cells with pertussis toxin or co-expression of a minigene encoding 11-amino acid sequence of COOH-terminal Galphai prevented the thrombin-induced increase in reporter activity. Pertussis toxin treatment also prevented thrombin-induced MAPK phosphorylation, indicating a role of Galphai in activating the downstream MAPK pathway. Expression of constitutively active Galphai2 mutant or Gbeta1gamma2 subunits increased reporter activity 3-4-fold in the absence of thrombin stimulation. Co-expression of dominant negative mutants of either Ras or MEK1 with the reporter construct inhibited the thrombin-induced PAR-1 expression, whereas constitutively active forms of either Ras or MEK1 activated PAR-1 expression in the absence of thrombin stimulation. Expression of dominant negative Src kinase or inhibitors of phosphoinositide 3-kinase also prevented the MAPK activation and PAR-1 expression. We conclude that thrombin-induced activation of PAR-1 mediates PAR-1 expression by signaling through Gi1/2 coupled to Src and phosphoinositide 3-kinase, and thereby activating the downstream Ras/MAPK cascade.     

TGF beta 1 and TNF alpha potentiate nitric oxide production in astrocyte cultures by recruiting distinct subpopulations of cells to express NOS-2

Nitric oxide (NO) synthase-2 (NOS-2), a key source of NO at sites of neuroinflammation, is induced in astrocyte cultures treated with lipopolysaccharide (LPS) plus interferon-gamma (IFN gamma). A recent study examining the regulation of astrocytic NOS-2 expression demonstrated that transforming growth factor-beta1 (TGF beta 1) potentiated LPS plus IFN gamma-induced NOS-2 expression via expansion of the pool of astrocytes that express NOS-2. Results in the current report indicate that this population-based mechanism of increasing NOS-2 expression is not restricted to TGF beta 1, since it also accounts for the potentiation of NO production in astrocyte cultures by tumor necrosis factor-alpha (TNFalpha). In contrast to TGF beta 1, which required 24h preincubation for optimal potentiation of NO production, TNF alpha was maximally effective when added concurrently with LPS plus IFN gamma. Nevertheless, under conditions that optimally potentiated NO production, both cytokines recruited similar numbers of astrocytes to express NOS-2 (% NOS-2-positive cells after LPS plus IFN gamma alone or with TNFalpha or TGF beta 1 was 9.5+/-1.2, 25.3+/-2.9, and 32.4+/-3.0, respectively). Interestingly, stimulation of astrocytes in the presence of both TGF beta 1 and TNFalpha additively increased the number of astrocytes that expressed NOS-2 protein (% NOS-2-positive cells was 61.0+/-4.2) relative to each cytokine alone. Potentiation of NO production by either TNF alpha or TGF beta 1 was not ablated by neutralizing antibodies to TGF beta 1 or TNFalpha, respectively. Thus, the two cytokines act independently to recruit separate pools of astrocytes to express NOS-2. These results are consistent with the notion that astrocytes possess an innate heterogeneity with respect to responsiveness to these cytokines.     

Adult human vascular endothelial cells express the IL6 gene differentially in response to LPS or IL1

We investigated the regulation of IL6 biological activity, de novo synthesis, and mRNA levels in adult vascular endothelial cells (EC) by bacterial endotoxin or inflammatory cytokines. Cells incubated without stimulus released scant IL6 activity. IFN gamma, IL2, or PDGF did not augment IL6 release from EC. LPS, lipid A, and TNF increased IL6 release modestly (5 to 20-fold), while recombinant IL1s (rIL1s) stimulated this process 100 to 400-fold. Differential release of IL6 from EC treated with LPS or rIL1 continued for at least 144 hr. Exposure to LPS or rIL1 caused EC to synthesize IL6 de novo. EC secreted the newly synthesized IL6 into the supernatant, rather than retaining it within or bound to cells. EC accumulated IL6 mRNA after 3 hr of exposure to rIL1. However, we could only detect IL6 message in cells incubated with LPS under "superinduction" conditions with cycloheximide, consistent with lower levels of IL6 biological activity in response to LPS compared to IL1 stimulation. We propose that local production of IL6 by vascular EC, which comprise the barrier between tissues and the blood, may influence regional immune and inflammatory responses.     

Differential regulation of cytokine and chemokine production in lipopolysaccharide-induced tolerance and priming

LPS pretreatment of human pro-monocytic THP-1 cells induces tolerance to secondary LPS stimulation with reduced TNFalpha production. However, secondary stimulation with heat-killed Staphylococcus aureus (HKSa) induces priming as evidenced by augmented TNFalpha production. The pro-inflammatory cytokine, IFNgamma, also abolishes suppression of TNFalpha in LPS tolerance. The effect of LPS tolerance on HKSa and IFNgamma-induced inflammatory mediator production is not well defined. We hypothesized that LPS, HKSa and IFNgamma differentially regulate pro-inflammatory mediators and chemokine production in LPS-induced tolerance. THP-1 cells were pretreated for 24 h with LPS (100 ng/ml) or LPS (100 ng/ml) + IFNgamma (1 microg/ml). Cells were subsequently stimulated with LPS or HKSa (10 microg/ml) for 24 h. The production of the cytokines TNFalpha, IL-6, IL-1beta, and GMCSF and the chemokine IL-8 were measured in supernatants. LPS and HKSa stimulated TNFalpha (3070 +/- 711 pg/ml and 217 +/- 9 pg/ml, respectively) and IL-6 (237 +/- 8.9 pg/ml and 56.2 +/- 2.9 pg/ml, p < 0.05, n = 3, respectively) in control cells compared to basal levels (< 25 pg/ml). LPS induced tolerance to secondary LPS stimulation as evidenced by a 90% (p < 0.05, n = 3) reduction in TNFalpha. However, LPS pretreatment induced priming to HKSa as demonstrated by increased TNFalpha (2.7 fold, from 217 to 580 pg/ml, p < 0.05, n = 3 ). In contrast to suppressed TNFalpha, IL-6 production was augmented to secondary LPS stimulation (9 fold, from 237 to 2076 pg/ml, p < 0.01, n = 3) and also primed to HKSa stimulation (62 fold, from 56 to 3470 pg/ml, p < 0.01, n = 3). LPS induced IL-8 production and to a lesser extent IL-1beta and GMCSF. LPS pretreatment did not affect secondary LPS stimulated IL-8 or IL-1beta, although HKSa stimulation augmented both mediators. In addition, IFNgamma pretreatment reversed LPS tolerance as evidenced by increased TNFalpha levels while IL-6, IL-1beta, and GMCSF levels were further augmented. However, IL-8 production was not affected by IFNgamma. These data support our hypothesis of differential regulation of cytokines and chemokines in gram-negative- and gram-positive-induced inflammatory events. Such changes may have implications in the pathogenesis of polymicrobial sepsis.     

Cytokine and chemokine expression associated with steatohepatitis and hepatocyte proliferation in rats fed ethanol via total enteral nutrition

To determine the temporal relationship between alcohol-induced changes in cytokines and chemokines, development of liver pathology and stimulation of hepatocyte proliferation, male Sprague-Dawley rats were intragastrically fed low carbohydrate-containing ethanol (EtOH) diets via total enteral nutrition (TEN) for up to 49 d. Induction of EtOH metabolism and appearance of steatosis preceded development of oxidative stress, inflammation, and cell death. A transitory peak of tumor necrosis factor (TNFalpha) and interferon gamma (IFN gamma) was observed at 14 d followed by reduced expression of TNFalpha, IFN gamma and another Th1 cytokine IL-12 accompanied by reduced expression of the Th1 regulators T-bet and STAT4. After 35-49 d of EtOH, at a time when hepatocyte proliferation was stimulated, IL-12 returned to control values and a second peak of TNFalpha occurred. The Th2 cytokine IL-4 remained suppressed throughout the study and was accompanied by reductions in the Th2 regulator GATA3. There was no temporal effect of EtOH on expression of IL-6 or TGFbeta. IL-5 and IL-13 mRNA were undetectable. Chemokine CXCL-2 expression increased progressively up to 35 d and preceded the appearance of inflammatory infiltrates. These data suggest that steatosis, increased ethanol metabolism, a transient induction of the innate immune response and suppression of Th2 responses were acute consequences of ethanol treatment and were followed by suppression of Th1 responses. However, the majority of necrosis, apoptosis and a late peak of TNFalpha only occurred after 6-7 weeks of ethanol, coincided with the appearance of inflammatory infiltrates and were associated with stimulation of hepatocyte proliferation.     

Macrophage migration inhibitory factor is induced by thrombin and factor Xa in endothelial cells

Macrophage migration inhibitory factor (MIF), a proinflammatory cytokine, has been shown to play a role in wound-healing processes. In this study, we investigated whether protease-activated receptor (PAR)-1 and PAR-2 mediated MIF expression in human endothelial cells. Thrombin, factor Xa (FXa), and trypsin induced MIF expression in human dermal microvascular endothelial cells and human umbilical vein endothelial cells, but other proteases, including kallikrein and urokinase, failed to do so. Thrombin-induced MIF mRNA expression was significantly reduced by the thrombin-specific inhibitor hirudin. Thrombin receptor activation peptide-6, a synthetic PAR-1 peptide, induced MIF mRNA expression, suggesting that PAR-1 mediates MIF expression in response to thrombin. The effects of FXa were blocked by antithrombin III, but not by hirudin, indicating that FXa might enhance MIF production directly rather than via thrombin stimulation. The synthetic PAR-2 peptide SLIGRL-NH(2) induced MIF mRNA expression, showing that PAR-2 mediated MIF expression in response to FXa. Concerning the signal transduction, a mitogen-activated protein kinase kinase inhibitor (PD98089) and a nuclear factor (NF)-kappaB inhibitor (SN50) suppressed the up-regulation of MIF mRNA in response to thrombin, FXa, and PAR-2 agonist stimulation, whereas a p38 inhibitor (SB203580) had little effect. These facts indicate that up-regulation of MIF by thrombin or FXa is regulated by p44/p42 mitogen-activated protein kinase-dependent pathways and NF-kappaB-dependent pathways. Moreover, we found that PAR-1 and PAR-2 mRNA expression in endothelial cells was enhanced by MIF. Furthermore, we examined the inflammatory response induced by PAR-1 and PAR-2 agonists injected into the mouse footpad. As shown by footpad thickness, an indicator of inflammation, MIF-deficient mice (C57BL/6) were much less sensitive to either PAR-1 or PAR-2 agonists than wild-type mice. Taken together, these results suggest that MIF contributes to the inflammatory phase of the wound healing process in concert with thrombin and FXa via PAR-1 and PAR-2.     

Lipopolysaccharide pretreatment increases protease-activated receptor-2 expression and monocyte chemoattractant protein-1 secretion in vascular endothelial cells

Background

This study investigated whether lipopolysaccharide (LPS) increase protease-activated receptor-2 (PAR-2) expression and enhance the association between PAR-2 expression and chemokine production in human vascular endothelial cells (ECs).

Methods

The morphology of ECs was observed through microphotography in cultured human umbilical vein ECs (EA. hy926 cells) treated with various LPS concentrations (0, 0.25, 0.5, 1, and 2 μg/mL) for 24 h, and cell viability was assessed using the MTT assay. Intracellular calcium imaging was performed to assess agonist (trypsin)-induced PAR-2 activity. Western blotting was used to explore the LPS-mediated signal transduction pathway and the expression of PAR-2 and adhesion molecule monocyte chemoattractant protein-1 (MCP-1) in ECs.

Results

Trypsin stimulation increased intracellular calcium release in ECs. The calcium influx was augmented in cells pretreated with a high LPS concentration (1 μg/mL). After 24 h treatment of LPS, no changes in ECs viability or morphology were observed. Western blotting revealed that LPS increased PAR-2 expression and enhanced trypsin-induced extracellular signal-regulated kinase (ERK)/p38 phosphorylation and MCP-1 secretion. However, pretreatment with selective ERK (PD98059), p38 mitogen-activated protein kinase (MAPK) (SB203580) inhibitors, and the selective PAR-2 antagonist (FSLLRY-NH2) blocked the effects of LPS-activated PAR-2 on MCP-1 secretion.

Conclusions

Our findings provide the first evidence that the bacterial endotoxin LPS potentiates calcium mobilization and ERK/p38 MAPK pathway activation and leads to the secretion of the pro-inflammatory chemokine MCP-1 by inducing PAR-2 expression and its associated activity in vascular ECs. Therefore, PAR-2 exerts vascular inflammatory effects and plays an important role in bacterial infection-induced pathological responses.

     

Interferon gamma encapsulated into liposomes enhances the activity of monocytes and natural killer cells and has antiproliferative effects on tumor cells in vitro

The effects of human interferon gamma (IFN gamma) encapsulated into liposomes were investigated in vitro. Monocytes were induced to release a cytotoxic factor with either IFN gamma encapsulated into liposomes, free IFN gamma or lipopolysaccharide (LPS). If IFN gamma was applied in the liposomal form, less IFN activity was required to stimulate monocytes. Most of the cytotoxic factor was secreted during the first 4 h of stimulation. The cytotoxic factor in supernatants from PMNLs was completely neutralized by a monospecific polyclonal antiserum to tumor necrosis factor (TNF). Combining subthreshold doses of IFN gamma liposomes or IFN gamma with lipopolysaccharide synergistically enhanced the release of TNF. In fluorescence analysis, altered expression of the class II HLA-DR antigen on LeuM3 positive monocytes was induced with IFN gamma liposomes as well as with IFN gamma. Not only monocytes but also natural killer (NK) cells were stimulated to higher cytotoxicity by IFN gamma liposomes in a dose-dependent manner. In comparison with IFN gamma, the same amount of activity was necessary for adequate stimulation of NK-cells against the K562 target cells. Furthermore, the antiproliferative effects of IFN gamma liposomes and free IFN gamma on several human tumor cell lines was compared. Among several cell lines tested, U937 and A549 turned out to be sensitive to IFN gamma, and both cell lines reacted with 50% growth inhibition at a lower amount of gamma presented by liposomes than in the free form. These data show production of IFN gamma liposomes which possess immunomodulatory and antiproliferative activity in vitro. In several of the test systems studied, liposome-encapsulated IFN gamma was more effective than free IFN gamma.     

In situ localization of TNFalpha/beta,TACE and TNF receptors TNF-R1 and TNF-R2 in control and LPS-treated lung tissue

Tumor necrosis factor (TNF) has been implicated in several infectious and inflammatory lung diseases. Two closely related variants, TNFalpha and TNFbeta, elicit various cellular responses via two distinct TNF receptors, the 55-kDa TNF-R1 and the 75-kDa TNF-R2. Recently, a TNFalpha-converting enzyme (TACE) was described, which cleaves and releases the membrane-bound TNFalpha. In the present study in normal rat and human lung tissue, the constitutive expression of TNFalpha/beta, TACE and TNF-R1/R2 was investigated by immunohistochemical techniques. In addition, TNFalpha and TNFbeta mRNA were localized by in situ hybridization. Both TNFalpha and TNFbeta were detected in various lung cell types. Expression of TNFalpha was particularly prominent in bronchial epithelial cells and vascular smooth muscle cells, next to alveolar macrophages. Both in situ hybridization for TNFalpha message and TACE immunostaining matched this expression profile. TNFbeta-so far only known to be produced by lymphocytes-was demonstrated in alveolar macrophages, bronchial epithelial cells, vascular smooth muscle cells and endothelial cells at the protein and the message level. Both TNF receptors were detected, with TNF-R1 being prominent on bronchial epithelial cells and endothelial cells, and TNF-R2 being expressed by nearly all cell types. Following LPS stimulation in isolated rat lungs TNFalpha/beta signal intensity was largely reduced due to liberation of stored TNFalpha/beta, while TACE immunoreactivity remained unchanged or was enhanced, demonstrating increased TNF generation.We conclude that both TNFalpha and TNFbeta are constitutively expressed by several non-leukocytic cell types in the human and rat lung. In concert with the expression of TACE and the TNF receptors R1 and R2, this finding suggests in addition to the known role of the TNF system in inflammation physiological functions of the TNF system in different compartments of the adult lung, with the vasculature and the bronchial tissue being of particular interest in addition to the leukocyte/macrophage populations.     

Role of T cell-replacing factor (TRF) in the murine B cell differentiation: induction of increased levels of expression of secreted type IgM mRNA

T cell-replacing factor (TRF) is known to play a critical role in the regulation of B cell growth and differentiation. In this study, the role of TRF in the expression of mRNA for both IgM and IgG1 class was investigated. The TRF was purified from cellfree supernatants from a T cell hybridoma, B151K12. RNA was isolated from chronic B cell leukemia (BCL1) cells, DNP-KLH-primed B cells, or normal B cells cultured with or without LPS, and LPS plus TRF or LPS plus BSF-1. The steady state level of isotype-specific mRNA was assessed by Northern blot analysis with a mu-specific or a gamma 1-specific probe. It was demonstrated that BCL1 and purified B cells cocultured with TRF expresses increased levels (twofold and fourfold, respectively) of secreted forms of mu mRNA. Purified B cells from DNP-KLH-primed mice also expressed increased levels (twofold to fourfold) of mu as well as gamma 1 mRNA for secreted form by stimulation with TRF. Total expression of mu mRNA, however, was approximately threefold higher than that of gamma 1 mRNA. The stimulation of normal B cells with LPS plus TRF induced an increase in the levels of mu mRNA and gamma 1 mRNA expression, fourfold and threefold, respectively. However, the levels of gamma 1 mRNA expression was one-third of that induced in B cells stimulated with LPS plus BSF-1. These results indicate that TRF preferentially induces increased levels of secreted type of mu mRNA and induces less gamma 1 mRNA than BSF-1. The differential role of TRF from BSF-1 in the expression of Ig mRNA will be discussed.     

Interferon-gamma inhibits steroidogenesis and accumulation of mRNA of the steroidogenic enzymes P450scc and P450c17 in cultured porcine Leydig cells

The inhibitory effects of recombinant porcine interferon-gamma (IFN gamma) on human CG (hCG)-stimulated testosterone production, and on mRNA concentrations of cholesterol side-chain cleavage (P450scc) and 17 alpha-hydroxylase/C17-20lyase (P450c 17) were investigated using porcine primary Leydig cell culture as a model. After preincubation of Leydig cells for 24 h with 1000 pM IFN gamma, hCG-stimulated (10 ng/ml, 2 h) testosterone production was inhibited by 50%, whereas no significant changes were seen in hCG-stimulated cAMP production. Incubation with 10 microM 5-cholestene-3 beta,22(R)-diol or 10 microM 5-cholestene-3 beta,20 alpha-diol together with hCG (10 ng/ml, 2 h) reversed most of the inhibitory effect of IFN gamma, suggesting that IFN gamma inhibits P450scc activity, possibly by inhibiting the substrate (cholesterol) availability for P450scc. Incubation with IFN gamma also decreased basal concentrations of P450scc (45%) and P450c 17 (35%) mRNA, although these changes probably did not contribute to the decreased testosterone production. Long-term treatment with hCG (100 ng/ml, 24 h) increased P450scc mRNA (3- to 4-fold) and P450c 17 mRNA (4- to 5-fold) concentrations. Simultaneous treatment with IFN gamma attenuated these hCG-induced increases in P450scc mRNA (50%) and P450c 17 mRNA (40-100%) concentrations, as well as in testosterone production (77%). This inhibition of testosterone production could only be partly reversed by the hydroxylated cholesterol derivatives. This suggests that in addition to possible suppression of cholesterol availability, decreased P450scc and/or P450c 17 activities (through decreased mRNA concentrations) were also involved in the IFN gamma suppressed steroidogenic capacity of porcine Leydig cells during long-term hCG stimulation.     

Regulation of luteal prostaglandin F(2 alpha) production and its relevance to cell death: an in vitro study using rat dispersed luteal cells

We investigated the mechanism by which rat luteal cells produce prostaglandin F(2 alpha) (PGF(2 alpha)) and its relevance to cell death in vitro. Treatment with progesterone (P4) of dispersed luteal cells prepared from rats on day 9 of pseudopregnancy caused dose-dependent inhibition of PGF(2 alpha) secretion. Cytokines, tumor necrosis factor alpha (TNFalpha) or interferon gamma (IFN gamma) alone had no or modest regulatory effects. Arachidonyl trifluoromethyl ketone (AACOCF(3)), a specific group IVA phospholipase A(2) inhibitor, depressed both basal and cytokine-regulated PGF(2 alpha) production. A combination of TNFalpha and IFN gamma stimulated PGF(2 alpha) synthesis and cytotoxicity (both, P<0.05). Agonistic anti-Fas antibody challenge caused a significant cytotoxic effect but without affecting PGF(2 alpha) production. The present data suggest that P4 inhibits and TNFalpha and IFN gamma cooperatively stimulate PGF(2 alpha) release by rat luteal cells. They also suggest that luteal cell death induced by TNFalpha/IFN gamma and Fas stimulation seems to occur via distinct signaling pathways involving PGF(2 alpha) production.