Wnt7b regulates placental development in mice.

Secreted Wnt proteins regulate many developmental processes in multicellular organisms. We have generated a targeted mutation in the mouse Wnt7b gene. Homozygous Wnt7b mutant mice die at midgestation stages as a result of placental abnormalities. Wnt7b expression in the chorion is required for fusion of the chorion and allantois during placental development. The alpha4 integrin protein, required for chorioallantoic fusion, is not expressed by cells in the mutant chorion. Wnt7b also is required for normal organization of cells in the chorionic plate. Thus, Wnt7b signaling is central to the early stages of placental development in mammals.

Vascular cell adhesion molecule-1 augments adenovirus-mediated gene transfer.

We have reported that adenovirus-mediated gene transfer is augmented in the endothelium of atherosclerotic blood vessels. We observed that vascular cell adhesion molecule-1 (VCAM-1) shares some homology with the coxsackievirus and adenovirus receptor. Because VCAM-1 is upregulated on atherosclerotic endothelial cells, we hypothesized that VCAM-1 may act as an auxiliary receptor to augment adenovirus-mediated gene transfer. To test this hypothesis, stable NIH 3T3 cell lines that constitutively express VCAM-1 on the cell surface were generated. Recombinant adenovirus 5 (Ad5), which contains the reporter ss-galactosidase gene, was used to compare Ad5 infection in VCAM-1(+) and parental NIH 3T3 cells. Total ss-galactosidase activity and the number of transgene-positive cells were 6- to 10-fold and 5-fold higher, respectively, in VCAM-1(+) than in VCAM-1(-) cells. Ad5 binding to VCAM-1(+) cells was increased by 3-fold over VCAM-1(-) cells. Soluble VCAM-1 protein, present during infection or viral binding, reduced ss-galactosidase activity in VCAM-1(+) cells in a dose-dependent manner. Taken together, we conclude that VCAM-1 can mediate adenovirus binding and infection. This may explain, in part, the previous finding that adenovirus-mediated gene transfer is augmented in atherosclerotic arteries.

Alpha 4 beta 1 integrin/VCAM-1 interaction activates alpha L beta 2 integrin-mediated adhesion to ICAM-1 in human T cells.

Modulation of integrin affinity and/or avidity provides a regulatory mechanism by which leukocyte adhesion to endothelium is strengthened or weakened at different stages of emigration. In this study, we demonstrate that binding of high-affinity alpha 4 beta 1 integrins to VCAM-1 strengthens alpha L beta 2 integrin-mediated adhesion. The strength of adhesion of Jurkat cells, a human leukemia T cell line, or MnCl2-treated peripheral blood T cells to immobilized chimeric human VCAM-1/Fc, ICAM-1/Fc, or both was quantified using parallel plate flow chamber leukocyte detachment assays in which shear stress was increased incrementally (0.5-30 dynes/cm2). The strength of adhesion to VCAM-1 plus ICAM-1, or to a 40-kDa fragment of fibronectin containing the CS-1 exon plus ICAM-1, was greater than the sum of adhesion to each molecule alone. Treatment of Jurkat or blood T cells with soluble cross-linked VCAM-1/Fc or HP2/1, a mAb to alpha 4, significantly increased adhesion to ICAM-1. These treatments induced clustering of alpha L beta 2 integrins, but not the high-affinity beta 2 integrin epitope recognized by mAb 24. Up-regulated adhesion to ICAM-1 was abolished by cytochalasin D, an inhibitor of cytoskeletal rearrangement. Taken together, our data suggest that the binding of VCAM-1 or fibronectin to alpha 4 beta 1 integrins initiates a signaling pathway that increases beta 2 integrin avidity but not affinity. A role for the cytoskeleton is implicated in this process.

Vascular cell adhesion molecule-1 expression by hematopoiesis-supporting stromal cells is not essential for lymphoid or myeloid differentiation in vivo or in vitro.

Binding of myeloid and lymphoid precursors to stromal cells in bone marrow has been suggested to be mediated through alpha 4 integrins (alpha 4) and vascular cell adhesion molecule-1 (VCAM-1) expressed on hematopoietic progenitors and stromal cells, respectively. It has not been shown, however, how essential the VCAM-1/ alpha 4 interaction is for hematopoiesis in vivo and whether or not other adhesion pathways can provide similar functional binding between stromal cells and hematopoietic progenitors. We addressed this issue by analyzing myeloid and lymphoid differentiation in vivo in mice with VCAM-1-null or -hypomorphic mutations and in vitro in long-term hematopoietic cultures with stromal cell clones from wild-type mice, which express or do not express VCAM-1. Mice bearing VCAM-1 mutations had no gross hematopoietic insufficiencies in the myeloid or lymphoid compartments and the distribution of myeloid progenitors between bone marrow and periphery was normal. In Dexter type long-term bone marrow cultures from mutant mice, the formation of supportive stromal cell layers and myeloid proliferation and differentiation were not affected by the absence of VCAM-1. Long-term maintenance and proliferation of clonable pre-B cells, cobblestone formation and differentiation to IgM-secreting, mature B cells was equally possible on VCAM-1+ and VCAM-1- stromal cell clones. We conclude from our data that VCAM-1 is not essential for the functional interaction between hematopoietic progenitors and stromal cells required for myeloid and B-lymphoid development in vivo or in vitro.

The omega-3 fatty acid docosahexaenoate reduces cytokine-induced expression of proatherogenic and proinflammatory proteins in human endothelial cells.

The mechanisms by which dietary fatty acids can modulate atherogenesis and inflammation are poorly understood. Induction in endothelial cells of adhesion molecules for circulating leukocytes and of inflammatory mediators by cytokines probably contributes to the early phases of atherogenesis and inflammation. We report here that incorporation into cellular lipids of docosahexaenoic acid (DHA), a specific fatty acid of the omega 3 family, decreases cytokine-induced expression of endothelial leukocyte adhesion molecules, secretion of inflammatory mediators, and leukocyte adhesion to cultured endothelial cells. DHA, but not eicosapentaenoic acid, decreased in a dose- and time-dependent fashion the expression of vascular cell adhesion molecule 1 (VCAM-1) induced by interleukin (IL)-1, tumor necrosis factor (TNF), IL-4, or bacterial lipopolysaccharide, with half-maximum inhibition at < 10 mumol/L. This reduction required prolonged (24- to 96-hour) exposure of endothelial cells to DHA and correlated with the degree of DHA incorporation into cellular lipids. DHA also limited cytokine-stimulated endothelial cell expression of E-selectin and intercellular adhesion molecule 1 and the secretion of IL-6 and IL-8 into the medium but not the surface expression of constitutive surface molecules. Cyclooxygenase inhibition did not block the effect of DHA on VCAM-1. In parallel with reduced surface VCAM-1 protein expression, DHA reduced VCAM-1 mRNA induction by IL-1 or TNF. DHA treatment also reduced the adhesion of human monocytes and of monocytic U937 cells to cytokine-stimulated endothelial cells. These properties of DHA may contribute to antiatherogenic and anti-inflammatory effects of omega 3 fatty acids.

Inducible expression of vascular cell adhesion molecule-1 by vascular smooth muscle cells in vitro and within rabbit atheroma.

Vascular cell adhesion molecule-1 (VCAM-1), a mononuclear leukocyte adhesion molecule, is expressed in cultured vascular endothelial cells activated by cytokines and is induced in rabbit aortic endothelium in vivo within 1 week after initiation of an atherogenic diet. We now demonstrate that vascular smooth muscle cells can also express VCAM-1 in rabbit atherosclerotic lesions in vivo and in response to cytokines in vitro. Immunohistochemical staining of aortas from rabbits fed a 0.3% cholesterol-containing diet revealed that a portion of smooth muscle cells within intimal foam cell-rich lesions expressed VCAM-1. The intimal VCAM-1-expressing cells localized predominantly in regions above the internal elastic lamina. These VCAM-1-positive cells had the typical spindle shape of smooth muscle cells but had reduced alpha-actin expression in comparison to normal medial smooth muscle cells, and did not bear markers for endothelium, macrophages, and T cells. In culture, rabbit aortic smooth muscle cells expressed VCAM-1 mRNA and protein in a time- and concentration-dependent fashion when exposed to interferon-gamma or Gram-negative bacterial lipopolysaccharide. Cultured human vascular smooth muscle cells also expressed VCAM-1 mRNA and protein in response to lipopolysaccharide, interferon-gamma, and interleukin-4. The monokines interleukin-1 alpha and tumor necrosis factor-alpha did not induce VCAM-1 expression in either rabbit or human vascular smooth muscle cells. Inducible VCAM-1 expression by vascular smooth muscle cells in vivo during hypercholesterolemia and in vitro in response to certain cytokines suggests a broader range of VCAM-1 functions in vascular biology than heretofore appreciated.

Glomerular epithelial cells produce endothelin-1.

Endothelin-1, a vasoactive peptide originally isolated from vascular endothelial cell culture supernatants, has constricting or mitogenic effects on smooth muscle and glomerular mesangial cells. Whether or not cultured rat glomerular epithelial cells synthesize endothelin-1 was assessed. Under basal culture conditions, the synthesis and release of endothelin-1 peptide by glomerular epithelial cells was time dependent, reaching 0.231 +/- 0.017 pg/1,000 cells at 24 h. For comparison, unstimulated bovine pulmonary artery endothelial cells and rat mesangial cells produced 0.982 +/- 0.237 and 0.004 +/- 0.002 pg of endothelin-1 peptide/1,000 cells per 24 h, respectively. In addition to endothelin-1 peptide, unstimulated glomerular epithelial cells expressed preproendothelin-1 mRNA. Transforming growth factor-beta, complement C5b-9, thrombin, and phorbol myristate acetate significantly enhanced endothelin-1 peptide synthesis in glomerular epithelial cells (45, 15, 55, and 25% above basal levels at 24 h, respectively), whereas epidermal growth factor had no effect. Thrombin and phorbol myristate acetate appeared to stimulate endothelin-1 peptide by activating protein kinase C, because the protein kinase inhibitor 1-(5-isoquinolinyl-sulfonyl)-3-methyl-piperazine abolished the thrombin- and phorbol myristate acetate-induced rise in endothelin-1 but had no effect on basal production. The stimulatory effect of thrombin was also markedly diminished in glomerular epithelial cells that had been depleted of protein kinase C by prolonged preincubation with a high dose of phorbol myristate acetate. Thus, glomerular epithelial cells may be an important source of endothelin-1, which might influence glomerular vasoconstriction or proliferation of target cells, particularly in the presence of proinflammatory molecules in the glomerulus.

Lysophosphatidylcholine, a component of atherogenic lipoproteins, induces mononuclear leukocyte adhesion molecules in cultured human and rabbit arterial endothelial cells.

Accumulation of monocyte-derived foam cells in focal areas of the arterial intima is one of the key events in early atherogenesis. We have examined the effect of lysophosphatidylcholine (lyso-PC; lysolecithin), a major phospholipid component of atherogenic lipoproteins, on the expression of adhesion molecules for monocytes, such as vascular cell adhesion molecule-1 (VCAM-1) and intercellular adhesion molecule-1 (ICAM-1), in cultured human and rabbit arterial endothelial cells. Cultured rabbit aortic endothelial cells treated with lyso-PC showed increased mRNA and cell surface expression of VCAM-1 and ICAM-1, which was associated with increased adhesion of monocytes and monocyte-like cells (THP-1, U937). In cultured human iliac artery endothelial cells, lyso-PC similarly induced both VCAM-1 and ICAM-1, whereas in umbilical vein endothelial cells only ICAM-1 was up-regulated. In all endothelial cells examined, the effect of lyso-PC on E-selectin (endothelial-leukocyte adhesion molecule-1) expression was negligible, thus differentiating this stimulus from other endothelial activators, such as interleukin 1, tumor necrosis factor, or lipopolysaccharide. We conclude that lyso-PC can selectively induce VCAM-1 and ICAM-1 in arterial endothelial cells and that this action, in addition to its monocyte chemoattractant activity, may play an important role in monocyte recruitment into atherosclerotic lesions.

Nucleotide sequence of rat vascular cell adhesion molecule-1 cDNA.

Vascular cell adhesion molecule 1 (VCAM-1) is an inducible transmembrane protein which is expressed by vascular endothelium following cytokine activation. VCAM-1 mediated the adhesion of certain blood leukocytes and tumor cells via the interaction with its counter-receptor, the integrin VLA4. When initially cloned from interleukin-1 (IL-1) stimulated human umbilical vein endothelial cells, VCAM-1 was reported to contain six immunoglobulin-like domains. However, subsequent cDNA clones and structural analysis of the human gene evealed an alternatively spliced seventh immunoglobulin domain. This seven domain form appears to be the predominant transcript in IL-1 activated endothelium. In this report, the cloning and nucleotide sequence of rat VCAM-1 is described.

Gene structure, chromosomal location, and basis for alternative mRNA splicing of the human VCAM1 gene.

Vascular cell adhesion molecule 1 (VCAM-1) is a cell surface glycoprotein adhesive for certain blood leukocytes and tumor cells, which is expressed by activated endothelium in a variety of pathologic conditions including atherosclerosis. Genomic clones encoding the VCAM1 gene were isolated and the organization of the gene was determined. The gene, which is present in a single copy in the human genome, contains 9 exons spanning approximately 25 kilobases of DNA. Exons 2-8 contain C2 or H-type immunoglobulin domains. At least two different VCAM-1 precursors can be generated from the human gene as a result of alternative mRNA splicing events, which include or exclude exon 5. A consensus TATAA element is located upstream of the transcriptional start site. The VCAM1 promoter contains consensus binding sites for NF-kappa B, the GATA family of transcription factors, as well as an AP1 site. The VCAM1 gene was assigned to the 1p31-32 region of chromosome 1 based on the analysis of human-mouse hybrid cell lines and in situ hybridization. Structural analysis of the human VCAM1 gene provides the basis for alternative mRNA splicing and an initial approach to elucidating the regulation of VCAM-1 expression.

Hypercholesterolemia enhances macrophage recruitment and dysfunction of regenerated endothelium after balloon injury of the rabbit iliac artery.

BACKGROUND: We studied the effects on and possible interaction of balloon denudation and hypercholesterolemia on large arteries in the rabbit with special regard to structure and vascular reactivity. METHODS AND RESULTS: New Zealand White rabbits fed a 1% cholesterol diet or a standard diet for 14 weeks underwent balloon denudation of the left iliac artery 4 weeks before death. Both the balloon-injured and the control iliac arteries were harvested for in vitro studies of vascular reactivity, for immunohistochemical staining with monoclonal antibodies directed at smooth muscle cells and macrophages, and for scanning electron microscopy. Balloon injury caused intimal smooth muscle proliferation with little macrophage infiltration and was followed by recovery of endothelium-dependent vasodilator function within 4 weeks. Hypercholesterolemia caused macrophage-rich lesions confined to the intima with moderate impairment of endothelial vasodilator function. Balloon injury in the setting of hypercholesterolemia caused intimal smooth muscle cell proliferation and intense macrophage infiltration throughout the arterial wall and severe impairment of endothelial vasodilator function. Scanning electron microscopy confirmed regrowth of the endothelium in all balloon-injured vessels. In the balloon-injured arteries of hypercholesterolemic animals, the regenerated endothelium exhibited areas of atypical morphology not seen after balloon injury or hypercholesterolemia alone. CONCLUSIONS: The present study shows that balloon injury, hypercholesterolemia, and their combination cause distinct lesions and functional disturbances. An arterial balloon injury in the setting of hypercholesterolemia produces a diffuse inflammatory response that is accompanied by a sustained impairment of endothelial function and a marked proliferative response.

Tumor cell surface alpha 4 beta 1 integrin mediates adhesion to vascular endothelium: demonstration of an interaction with the N-terminal domains of INCAM-110/VCAM-1.

Hematogenous metastasis involves adhesive interactions between blood-borne tumor cells and the vessel wall. By the use of in vitro assays, the adhesion of human melanoma, osteosarcoma, and kidney carcinoma (but not colon carcinoma) cell lines was shown to involve the cytokine-inducible endothelial cell surface protein inducible cell adhesion molecule 110 (INCAM-110) and the alpha 4 beta 1 integrin, molecules normally involved in endothelial-leukocyte interactions. Tumor adhesion to human endothelial cell monolayers was increased 1.9- to 8.2-fold by endothelial activation with the cytokine tumor necrosis factor (TNF) and inhibited by the anti-INCAM-110 monoclonal antibody (mAb) E1/6. Each of these tumor cells expressed members of the beta 1 integrin family of adhesion molecules, and antibodies to the alpha 4 and beta 1 integrin subunits inhibited tumor-endothelial adhesion (48-87% inhibition). A cDNA encompassing the three N-terminal Ig-like domains of vascular cell adhesion molecule 1 (VCAM-1) encoded a protein recognized by the anti-INCAM-110 mAb E1/6 and, when captured onto plastic, supported melanoma cell adhesion by an alpha 4 integrin-dependent mechanism. In contrast to mAb E1/6, a second anti-INCAM-110 mAb Hu8/4 neither inhibited adhesion to activated endothelium nor bound the first three Ig-like domains of INCAM-110/VCAM-1. These data indicate that the adherence of several human tumors to activated endothelium is mediated by an interaction of alpha 4 beta 1 integrin and the N-terminal Ig-like domains of endothelial INCAM-110/VCAM-1. Tumor acquisition of the alpha 4 integrin subunit and endothelial expression of INCAM-110 may affect the frequency and distribution of metastasis.

Alternative splicing of human VCAM-1 in activated vascular endothelium.

Vascular cell adhesion molecule 1 (VCAM-1)/inducible cell adhesion molecule 110 is a mononuclear leukocyte-selective adhesion molecule, expressed on vascular endothelium following activation by certain cytokines or endotoxin. This inducible transmembrane protein and member of the immunoglobulin gene superfamily was previously reported to contain six immunoglobulinlike domains. Using the polymerase chain reaction, a VCAM-1 cDNA was obtained from mRNA of interleukin-1 (IL-1)-treated cultured human umbilical vein endothelial cells (HUVEC). The cDNA clone contained an additional 276 base-pair (bp) domain, located between domains 3 and 4. This new domain is most homologous to the existing N-terminal domain (domain 1). The internal 276-bp region is encoded by a single exon of the human VCAM-1 gene, indicating that the two forms of mRNA arise by alternative splicing. Both forms of VCAM-1 mRNA were detected by polymerase chain reaction in IL-1-stimulated HUVEC, although the seven-domain form appeared predominant. On the surface of HUVEC only a 110-kd polypeptide, consistent with the seven-immunoglobulinlike domain form of VCAM-1, was detectable by immunoprecipitation. Alternative splicing of the VCAM-1 gene in cytokine-activated endothelium may generate functionally distinct cell-surface adhesion molecules.

Persistent dysfunction of regenerated endothelium after balloon angioplasty of rabbit iliac artery.

This study investigated the vasodilator function of endothelium that regenerated after balloon angioplasty and the relation of this function to the extent of vascular injury and to subsequent intimal proliferation. Balloon angioplasty was performed in the left iliac artery of 47 New Zealand White rabbits. Vascular responses were examined in vitro 2 and 4 weeks after a "severe" injury (3.0-mm balloon) or a "moderate" injury (2.5-mm balloon). Both degrees of balloon injury caused complete endothelial denudation. Endothelial regrowth 2 weeks after either injury was confirmed histologically. Although the regenerated cells had irregular sizes and polygonal shapes and lacked the typical alignment in the direction of blood flow, immunocytochemical staining for factor VIII-related antigen identified these cells as endothelium. To study the vasodilator function of regenerated endothelium, rings of balloon-injured and control (contralateral) iliac arteries were suspended in organ chambers for recording of isometric force. Endothelium-dependent relaxation of balloon-injured vessels to acetylcholine and to the calcium ionophore A23187 were reduced at 2 and at 4 weeks after severe injury. After moderate injury, endothelium-dependent relaxations to these agents were reduced at 2 weeks but had normalized by 4 weeks. Endothelium-independent relaxation to sodium nitroprusside, however, was preserved in all study groups. Morphometric analysis revealed an inverse correlation between the degree of intimal thickening and maximal relaxation to acetylcholine (r = 0.45, p less than 0.01). Thus, there is a persistent attenuation of receptor- and nonreceptor-mediated endothelium-dependent relaxations after arterial injury. The regenerated cells have an altered morphological appearance, but staining for factor VIII-related antigen confirms their endothelial origin. The degree and duration of endothelial dysfunction depends on the severity of the initial injury and is related to the extent of intimal thickness.

Protein synthesis dependent and independent mechanisms of neutrophil emigration. Different mechanisms of inflammation in rabbits induced by interleukin-1, tumor necrosis factor alpha or endotoxin versus leukocyte chemoattractants.

Inflammation constitutes the body's principal mode of defense against infection and other harmful agents. Neutrophil leukocytes are the primary effector cells in this process. The role of protein synthesis in neutrophil emigration into acute inflammatory lesions was examined. Local intradermal injections of actinomycin D, cycloheximide or puromycin could inhibit in a dose- and time-dependent manner neutrophil emigration induced by interleukin-1, tumor necrosis factor alpha or endotoxin, but not by the leukocyte chemoattractants C5a des arg (zymosan-activated plasma), n-formyl-methionyl-leucyl-phenylalanine or leukotriene B4. Maximal inhibition, measured at the time of peak emigration, was greater than 90%. The onset of neutrophil emigration induced by the cytokines or by endotoxin was delayed by 30 to 60 minutes in comparison to the leukocyte chemoattractants. These results demonstrate at least two mechanisms of neutrophil emigration: one with a slower onset and dependence on local RNA transcription and translation and the other rapid in onset and independent of protein synthesis.

Neutrophil leukocyte emigration induced by endotoxin. Mediator roles of interleukin 1 and tumor necrosis factor alpha 1.

The hypothesis that cytokines mediate neutrophil emigration induced by endotoxin (LPS) was studied by examining the potency, the kinetics of neutrophil emigration, and the tachyphylaxis of intradermal sites with IL-1, TNF-alpha and LPS. Human rIL-1 alpha and IL-1 beta, synthetic lipid A, and LPS were several orders of magnitude more potent than human rTNF. The kinetic profiles of neutrophil emigration induced by IL-1 alpha, TNF, and LPS were characterized by minimal emigration in the first 30 min, followed by rapid and transient emigration. After the injection of LPS, the onset and the time at which the rate of emigration was maximal consistently appeared 30 min later than IL-alpha or TNF, suggesting that neutrophil emigration in response to LPS was mediated by a locally generated cytokine. IL-1 and TNF were then examined as potential secondary mediators of LPS-induced emigration by comparing the patterns of tachyphylaxis between LPS and IL-1 alpha or TNF; i.e., the magnitude of neutrophil emigration into inflammatory sites was compared with sites injected 6 h previously (desensitizing injections) with a cytokine or with LPS. Tachyphylaxis was dose dependent with each and also between the IL-1 species; therefore, when tachyphylaxis between the cytokines and LPS was examined, relatively higher doses were selected for the desensitizing injections than for the test injections. With this approach, desensitizing injections of IL-1 alpha diminished the neutrophil accumulation after LPS, and LPS also desensitized sites to IL-1 alpha. However, tachyphylaxis was not observed between TNF and LPS, or between TNF and IL-1 alpha. These data suggest that IL-1, but not TNF, is a potential mediator of LPS-induced neutrophil emigration.

Acute inflammation and a Shwartzman-like reaction induced by interleukin-1 and tumor necrosis factor. Synergistic action of the cytokines in the induction of inflammation and microvascular injury.

A Shwartzman-like reaction was elicited in rabbits by preparing the skin with intradermal injections of recombinant human tumor necrosis factor alpha (TNF alpha) and recombinant human interleukin-1 (IL-1 alpha or beta). The animals were challenged intravenously with endotoxin or by intravascular activation of complement with immune complexes or zymosan 18 hours later and were sacrificed after another 2 hours. Animals challenged with saline did not develop Shwartzman-like reactions. The sites prepared with endotoxin or with either form of IL-1 plus TNF alpha developed visible hemorrhage, whereas sites injected with either IL-1 or TNF alpha alone did not. Hemorrhage and microthrombosis were quantitated with 59Fe-labeled erythrocytes and 111In-platelets for 2 hours after the intravenous challenge, and the findings confirmed the observations made on gross inspection. Dermal sites prepared with the cytokines and challenged intravenously with endotoxin, immune complexes, or zymosan exhibited some diffuse hemorrhage and an intense erythrocyte extravasation around distended vessels, along skin appendages, and the panniculus carnosus muscle. The lumens of many large and postcapillary venules contained aggregates of platelets and leukocytes. These changes were superimposed on those seen at prepared sites (leukocyte infiltration). By electron microscopy fibrin was demonstrable in association with the formed elements of the blood. Histologic examination of the 18-hour-old preparative lesions or 20-hour-old lesions of saline-"challenged" animals revealed accumulation of leukocytes in the dermis, predominantly neutrophils. This accumulation was sparse at sites treated with only IL-1 or TNF alpha, but very intense at sites treated with both IL-1 and TNF alpha or with endotoxin. These observations were confirmed quantitatively by measuring the accumulation of 51Cr-labeled neutrophils for 2 hours after injection. The potency of IL-1 alpha was comparable to that in our earlier report, and TNF alpha was about three log times less potent. Sites treated with both IL-1 alpha and TNF alpha resulted in 69% greater neutrophil emigration than the additive response elicited by each cytokine. The reported findings implicate a synergism between IL-1 and TNF alpha in the induction of both the inflammatory reaction (preceding the Shwartzman reaction) and the thrombohemorrhagic component of the Shwartzman reaction proper.

Acute inflammation in gram-negative infection: endotoxin, interleukin 1, tumor necrosis factor, and neutrophils.

Experimental bacterial infection of the dermis induced with gram-negative microorganisms is associated with an acute inflammatory reaction, which represents the principal local defense against spread of the infection. When the inflammatory reaction is quantitated with radiolabeled cells and proteins, the kinetics resemble acute inflammation induced with other agents, such as immune complexes or chemotaxins. There is an interrelationship between the components or events of the inflammatory reaction; inasmuch as vascular injury is neutrophil-dependent, neutrophils must migrate to the site where the bacteria multiply. In neutropenic animals there is no such emigration and bacterial multiplication is not inhibited. The microorganisms shed endotoxin, which in turn induces secretion of interleukin 1 (IL 1) and probably tumor necrosis factor. Endotoxin is the most potent agent (10(-15) mol vs. 10(-12) mol of C5ades Arg) capable of inducing a neutrophil influx. Desensitization or tachyphylaxis of the tissues (probably of postcapillary venular endothelium) to IL 1 seems to control cessation of the neutrophil influx (also in vitro evidence). Phagocytosis of the bacteria by neutrophils is associated with release of oxygen radicals and lysosomal proteases from the neutrophils. These are instrumental in eliciting microvascular injury, which is characterized by enhanced vasopermeability, hemorrhage, and thrombosis.