Heparin facilitates the extraction of tissue fibronectin.

Extraction of fibronectin from two human tissues, lung parenchyma and placental villi, was facilitated by the incorporation of heparin into extraction media. The effect of heparin was additive to the effect of urea which is known to extract fibronectin. These experiments provide further evidence that fibronectin and glycosaminoglycans are associated in connective tissues and the use of heparin forms the basis for a simple method for extraction and quantitation of tissue fibronectin.

Cold-insoluble globulin (fibronectin) in connective tissues of adult human lung and in trophoblast basement membrane.

Cold-insoluble globulin (CIG), which is immunochemically indistinguishable from the fibroblast surface protein known as large external transformation-sensitive glycoprotein and fibronectin, was detected immunologically in connective tissue fractions from adult human lung. The fractions tested were (a) intact parenchyma, (b) acidic structural glycoproteins (ASG) extracted from lung parenchyma with 0.3 M acetic acid, and (c) isolated alveolar basement membrane (ABM). For comparison with ABM, preparations of human glomerular basement membrane and human trophoblast basement membrane (TBM) were tested. CIG was not detected in glomerular basement membrane but was present in large amounts in TBM. The CIG antigen could be solubilized from the parenchyma and from ABM by collagenase digestion which indicates that CIG occurs in lung connective tissue in association with collagen. Fibrinogen antigenic determinants were present in the ASG fraction, but the question of whether CIG and fibrin(ogen) are associated in lung connective tissue requires further study. When CIG was quantified by electroimmunoassay, intact lung parenchyma contained approximately equal to 0.4% CIG, ASG contained 3-4.5% CIG, ABM contained 0.1-0.9% CIG and TBM contained 1.5%-7.2% Cg. the evidence suggests that CIG is a chemical constituent of lung connective tissue matrix where it may influence the function of alveoli.

Presence of fibronectin in basement membranes and acidic structural glycoproteins from human placenta and lung.

Cold-insoluble globulin was detected in both trophoblast and alveolar basement membrane preparations, whereas it was not detected in GBM preparations. Acidic structural glycoproteins from both placental villi and lung parenchyma, which were extracted with 0.3 M acetic acid and recovered by adjusting the pH to 4.7, also contained CIg. Fractions of TBM, solubilized by either dilute alkali (0.01 N NaOH), by reduction and alkylation of the disulfide bonds, or by 0.3 M acetic acid extraction, all contained the antigen and possessed properties similar to those of ASG. The ASG fractions also reacted with antifibrinogen, but proof that the two types of determinants occur on a given molecular species is lacking at present. Purified collagenase solubilized CIg from ABM, from lung parenchyma, and from the stroma of placental villi, and this finding is strong evidence for an association of CIg with collagen in these connective tissues.

Hyaluronan affects extravascular water in lungs of unanesthetized rabbits.

We have determined whether changes in lung hyaluronan content affect extravascular water in lungs of unanesthetized rabbits. Three groups of experiments were performed. In group 1 (n = 12), no infusions were given; in group 2, nine pairs of rabbits received either intravenous hyaluronidase (750 U.kg-1.min-1) or an equivalent volume of saline; in group 3, nine pairs of rabbits received either hyaluronidase or saline, followed by intravenous saline infusion amounting to 24% of body weight. At the end of each experiment, one lung was analyzed for extravascular lung water by the wet-dry method. Except for group 3, in all animals the other lung was analyzed for hyaluronan content by a method that involved hydrolyzing lung hyaluronan with fungal hyaluronidase to release reducing N-acetyl glucosamine groups, which were quantified. In group 1, lung hyaluronan, which varied from 50 to 159 micrograms/g dry wt (mean 106 +/- 35 micrograms/g dry wt), significantly correlated with variation in extravascular lung water (mean 4.2 +/- 0.3 g/g dry wt). In group 2 rabbits given hyaluronidase, lung hyaluronan was 40% lower and extravascular lung water was 14.6% lower than in paired controls (P less than 0.01). In group 3, volume expansion did not affect lung water, except after hyaluronidase when lung water was 47% higher than paired controls. We conclude that in the lung the content of hyaluronan is one of the determinants of extravascular water content.

The role of hyaluronan in the pulmonary alveolus.

The duplex nature of the lining of the pulmonary alveolus has long been appreciated. It appears that surfactant is present at the interface with air where it prevents the collapse of the alveolus by lowering surface tension and that the surfactant rests on an aqueous subphase. This subphase has enough structure to form a smooth, continuous surface over the projections of the epithelial cells and because of its hydrophilic nature it attracts the polar heads of surfactant phospholipids. The chemical composition of the subphase has not been addressed. Type II cells in the wall of the alveolus are specialized to produce surfactant and they also secrete hyaluronan (hyaluronic acid) into the subphase. In solution, molecules of hyaluronan appear to be flexible coils which self-aggregate. The resulting solutions are quite viscous and exhibit non-Newtonian behavior. Hyaluronan binds to cell surface receptors and to proteins in the extracellular matrix. The networks formed with self-aggregated hyaluronan with or without proteins create gels whose properties depend largely upon the molecular weight of the hyaluronan and its concentration. Hyaluronan is also known to interact with phospholipids and has hydrophobic regions which could bind to the hydrophobic surfactant proteins B and C. The working hypothesis presented herein states that hyaluronan interacts with itself and with proteins in the subphase to form a hydrophilic gel. At the epithelial cell layer the components are concentrated due to tethered HA molecules and the gel smooths over cell projections. At the air interface the components are so dilute that a layer which is essentially water is present. The surfactant phospholipids spread on the water. Direct interactions of HA and surfactant phospholipids may also occur and contribute to the stability of the surfactant layer.

Quantification of tissue fibronectin from terminal villi of placenta.

Tissue fibronectin (TFn) was solubilized from the terminal villi of perfused human placentas by sequential chemical extractions and plasmin digestion. Alternatively, plasmin digestion of intact tissue solubilized all the TFn, which amounted to 1.8-2.9% of the dry weight of the villi. Concomitantly, 69% of the tissue was solubilized. The non-equilibrium competitive e.l.i.s.a. (enzyme-linked immunoabsorbent assay), in which the TFn was immunologically identical with plasma fibronectin (PFn), was used for the quantification of TFn. This study demonstrates that the bulk of TFn can be obtained in a form that can be quantified by e.l.i.s.a. and that TFn is immunologically identical with PFn. Thus the fibronectin molecule is not significantly altered as it is incorporated into the connective-tissue matrix and could exchange with PFn.

Evidence that fibronectin-immunoglobulin complexes occur normally in plasma.

This study was done to ascertain whether complexes of plasma fibronectin (PFn) and immunoglobulins (Igs) are normally present in plasma. PFn preparations from plasmas of six random human donors were analyzed for IgG, IgM, and IgA by competitive ELISA procedures. To look for direct evidence of circulating complexes, two of the plasma samples were chromatographed on Sephacryl S-300. PFn prepared from a pool of three plasmas by cryoprecipitation was studied by crossed immunoelectrophoresis to detect PFn-Ig complexes. The Ig content of PFn ranged from 0.78% to 9.0% IgG; 0.25% to 4.9% IgM, and 0.15% to 0.71% IgA. The relative amounts were IgG > IgM > IgA in every sample, but the total Ig content of PFn varied with the plasma donor. The Sephacryl S-300 chromatogram of the plasma from which Ig-rich PFn was made showed distortion of the PFn peak by Ig peaks. PFn prepared by cryoprecipitation contained PFn-immunoglobulin complexes. Plasma from adult beagle dogs was used to prepare canine PFn. IgM was detected in canine PFn by SDS-Page and by double diffusion experiments against antiserum to canine IgM. By crossed immunoelectrophoresis, a PFn-IgM complex was detected in canine plasma. The cryoprecipitate recovered from a Sephacryl S-300 chromatogram of canine plasma contained equivalent amounts of IgM and PFn. These data suggest that complexes of Igs with PFn are normally present in plasma.

Human basement membrane antigens from lung, placenta and kidney.

Three human basement membranes, glomerular basement membrane (GBM) from renal cortex, alveolar basement membrane (ABM) from lung parenchyma and trophoblast basement membrane (TBM) from the terminal villi of placenta have been isolated by sieving and sonication techniques. Canine GBM and ABM were also prepared. There were marked differences among the membranes from human tissues. Compared to GBM, TBM had very little collagen but contained high concentrations of charged amino acids. ABM was intermediate in composition between GBM and TBM and contained desmosine and isodesmosine indicative of the presence of elastin. Canine ABM (c-ABM) did not contain desmosine or isodesmosine. In the canine system an antigen was detected in ABM which was not present in GBM. The membrane preparations were analyzed for fibronectin content using a specific antiserum to fibronectin. This glycoprotein could not be detected in GBM whereas it was present in ABM in amounts up to 0.8% and in TBM in amounts as high as 7.2%. All the membranes induced the formation of precipitating antibodies in rabbits. Soluble material obtained from the membranes by alkali extraction, reduction of disulfide bonds, enzymatic digestion with elastase, plasmin or collagenase provided immunologically reactive fragments. These soluble fragments gave reactions of identity among the three basement membranes in immunodiffusion reactions in gels with antisera raised to all three BMs. The finding that plasmin digests basement membranes suggests that it may play a role in connective tissue remodeling. The fact that elastase degrades basement membranes provides an endogenous system for injury which may be triggered by infections.

Lung hyaluronan as assayed with a biotinylated hyaluronan-binding protein.

This study was designed to define how hyaluronan (HA) is bound in lung tissue. Aliquots of lyophilized hamster lungs were extracted with 0.5 M NaCl (associative conditions) or 4 M guanidine . HCL (Gu . HCl) (dissociative conditions) or with water. Aliquots were also digested with Pronase in phosphate-buffered saline (PBS) or in dilute Tris buffer. The nanogram amounts of solubilized HA were quantified by an inhibition assay based on the specificity of binding of HA to biotinylated HA-binding protein (B-HABP) rather than radioactive HA-binding protein. Lung HA was readily soluble. More than 80% of it was solubilized by one extraction with either 0.5 M NaCl or 4 M guanidine . HCl. Almost half of it was solubilized by two brief (15-min) water washes. After three extractions under associative conditions only 5% of the total HA remained insoluble and could exist in structural proteoglycan aggregates. However, HA is present in lung in more than one situation, as was discerned in Pronase digestion experiments. Digestion of lung tissue with Pronase solubilized total lung HA. In PBS all the HA was detected, but in dilute Tris buffer 52% of the HA solubilized was not available for combination with the B-HABP and was presumed to be bound to another lung component. Overall, the data suggest that lung HA is free to engage in water transport and to provide a protective coating for elastin and collagen fibers.

Early changes in lung tissue hyaluronan (hyaluronic acid) and hyaluronidase in bleomycin-induced alveolitis in hamsters.

Intratracheal instillation of bleomycin in hamsters initiates a series of events that mimic human interstitial pulmonary fibrosis. Because glycosaminoglycans and particularly hyaluronan (hyaluronic acid, HA), may play an important role in the extracellular matrix response to early injury and subsequent fibrosis, this study was undertaken to define the early time course of changes in HA and hyaluronidase. Hamsters were given either 1 unit bleomycin sulfate in 0.2 ml saline or 0.2 ml saline (control), and randomly selected animals from both groups were killed at Days 3, 5, 6, 7, 9, and 17. Glycosaminoglycan fractions prepared from lung tissue of individual animals were analyzed for HA. The maximal HA content was reached 6 days after instillation of bleomycin and was 14.6-fold the normal value. The weight of injured lungs was 2.3-fold the control value. Thus, the increase in HA content was 30-fold. By Day 7 the HA content had dropped sharply. It then declined gradually to approximately double control values at Day 17. The specific activity of lysosomal hyaluronidase was the same in bleomycin-treated lungs and control lungs. Total units of the enzyme were increased in injured lungs, even at the time of maximal HA content, indicating active turnover of HA. The maximal HA content occurs prior to the rise in collagen and elastin biosynthesis. This observation in addition to the magnitude of the increase and its abrupt decline suggest that HA may be an important initiating factor for pathologic changes in lung extracellular matrix components.

The fibronectin content of canine lungs is increased in bleomycin-induced fibrosis.

Fibronectin (Fn), a high molecular weight glycoprotein, was found to constitute 0.43% of the normal adult beagle dog lung. The tissue Fn (TFn) was solubilized by sequential chemical extractions and quantified by ELISA (enzyme-linked immunoabsorbent assay). Subsequent plasmin digestion did not appear to solubilize significantly more Fn. Since 70% of the lung tissue was solubilized by the extractions and plasmin digestions, the TFn quantified represented the bulk of lung Fn. The TFn was identical to plasma fibronectin in the ELISA and one can infer that the Fn molecule is not significantly altered as it is incorporated into the lung connective tissue matrix. Lungs from beagles in which fibrosis had been induced with bleomycin contained 0.99% Fn, more than a twofold increase over normal. In the ELISA TFn from fibrotic lungs gave an inhibition curve of the same shape as did TFn from normal lungs. Thus, Fn from fibrotic lungs is not different qualitatively from Fn from normal lungs in any way detectable with this antiserum. The TFn content of plasmin digests of intact lung was less than that of extracts, which was the converse of results obtained on placenta (B. A. Bray (1985) Biochem. J. 226, 811-815). This difference between lung TFn and placental TFn may be due to differences in degree of glycosylation, which determines susceptibility to proteases.

Human cervical and hamster oral epithelial neoplasia--colposcopic similarities.

Epithelial dysplasias and squamous carcinomas were experimentally produced in the cheek pouches of Syrian hamsters. These lesions were observed and photographed in the living animals with the aid of a stereoscopic microscope at magnifications of 20 and 40 times. Very apparent similarities were noted in the vascular patterns and epithelial configurations between the hamster oral dysplasias and carcinomas and the corresponding lesions seen in published cases of cervical intraepithlial neoplasia and invasive squamous carcinoma. It is speculated that these similarities reflect common factors in the interaction between neoplastic squamous epithelium and its blood supply. These findings are important in understanding the biology of cervical neoplasia and may be useful in the diagnosis and management of human oral neoplasia.

Association of fibrinogen and microfibrils with trophoblast basement membrane.

Two non-collagenous placental antigens have been detected both in membranes isolated from terminal villi by sieving and sonication and in acetic acid extracts of the villi. One of the antigens is apparently fibrinogen based on immunologic and chemical data. The second, detectable with anti-membrane antiserum after absorption with fibrinogen remained with the fibrinogen-like antigen during isoelectric precipitation at pH 4.7, electrophoresis at pH 8.6 and chromatography on DEAE cellulose. Fractions of villi in which the two antigens occur comprise more than a third of the weight of the villi. The fibrinogen-related antigen was concentrated in placental basement membranes compared to kidney and lung basement membranes. A third antigen common to the membranes, to glomerular basement membrane and to alveolar basement membrane was also detected.

Synthesis of crosslinked elastin by a mesothelial cell culture.

Synthesis of crosslinked elastin was measured in cultures of pleural mesothelial cells. Results indicate that mesothelial cells are a rich source of crosslinked elastin and may therefore be useful for in vitro studies of this connective tissue component.

Ligation of endothelial alpha v beta 3 integrin increases capillary hydraulic conductivity of rat lung.

Complement-mediated pulmonary edema results from increases in lung capillary hydraulic conductivity (Lp), possibly by receptor-mediated mechanisms. We considered the Lp effects of vitronectin and the vitronectin-containing complement complex SC5b-9, which ligate the integrin alpha v beta 3. Vitronectin, SC5b-9, and SC5b-9-enriched zymosan-activated serum all rapidly increased Lp, as determined by the split-drop technique in single lung capillaries of rat lung. The Lp increases were inhibited by a monospecific (LM609) and a polyclonal (R838) antibody against the alpha v beta 3 integrin but not by an irrelevant monoclonal antibody isotype matched with LM609, by a monoclonal antibody against the alpha v beta 5 integrin, or by preimmune rabbit serum. Vitronectin monomers failed to increase Lp. The tyrosine kinase blockers genistein and methyl 2,5-dihydroxycinnamate caused significant concentration-dependent inhibitions of Lp increases due to vitronectin and zymosan-activated serum. By contrast, the protein kinase C blocker calphostin C had no major effect. We conclude that (1) multivalent ligation of the luminally located alpha v beta 3 integrin of lung capillary endothelium increases transcapillary liquid flux, and (2) the dominant signal transduction pathway for this effect occurs through tyrosine kinase activation.