Demonstration of structural differences between the two subunits of human-plasma fibronectin in the carboxy-terminal heparin-binding domain

Structural differences between the two subunits of human plasma fibronectin were studied by analyzing the carboxy-terminal heparin-binding domain (Hep-2). Two fragments (29 kDa and 38 kDa) derived from the Hep-2 domain were purified from thermolysin-digested human plasma fibronectin. Identical NH2-terminal sequences were obtained for both fragments through 16 Edman cycles. Neither domain contained the 90-amino-acid extra domain which is predicted by cDNA analysis of the cellular form of fibronectin. We have examined the primary structures of the 29-kDa and 38-kDa Hep-2 domains produced from the two chains of plasma fibronectin by analyzing the tryptic peptides by fast atom bombardment/mass spectrometry and comparison with the predicted fragments deduced from the corresponding cDNA-derived peptide sequences. Peptides that were unique to each domain were further characterized by microsequence analysis. The two domains showed identical amino acid sequences through 274 residues, followed by a region of variability. The 29-kDa domain contains 279 amino acids with an estimated relative molecular mass (Mr) of 30,460. This domain is located in the heavy chain of plasma fibronectin and contains three repeats of type III sequences plus a portion of the connecting segment (IIICS) region. The 38-kDa domain contains 359 amino acids and one O-linked glycosyl unit for an estimated Mr of 39,263. This domain is from the light chain of plasma fibronectin and contains four repeats of type III sequences with the deletion of the entire 120-amino-acid IIICS area. Secondary structure analysis by Chou/Fasman and circular dichroism reveals extensive beta-sheet structure for these domains. Key sulfhydryl and glycosylation sites are located near the mRNA splice junctions for the two chains. It is postulated that the splice junctions are adjacent to a flexible domain joining two regions of extensive beta-sheet structure.     

Monoclonal antibodies in the analysis of fibronectin isoforms generated by alternative splicing of mRNA precursors in normal and transformed human cells

Recent results showing that a single fibronectin gene can give rise to several different mRNAs by alternative splicing have offered an explanation for fibronectin polymorphism. Here we report on monoclonal antibodies that show specificity for a fibronectin segment (ED) that can be included or omitted from the molecule depending on the pattern of splicing of the mRNA precursors. Using these monoclonals, we have quantitatively analyzed the expression of the ED sequence in human fibronectin from different sources. The results demonstrated that, at the protein level, the ED segment is not expressed in plasma fibronectin and that, in fibronectin from the tissue culture medium of tumor-derived or simian virus-40-transformed human cells, the percentage of fibronectin molecules containing the ED segment is about 10 times higher than in fibronectin from normal human fibroblasts. These results suggest that in malignant cells the mechanisms that regulate the splicing of mRNA precursors are altered.     

Characterization of DNA sequences that mediate nuclear protein binding to the regulatory region of the Pisum sativum (pea) chlorophyl a/b binding protein gene AB80: identification of a repeated heptamer motif

Two protein factors binding to the regulatory region of the pea chlorophyl a/b binding protein gene AB80 have been identified. One of these factors is found only in green tissue but not in etiolated or root tissue. The second factor (denominated ABF-2) binds to a DNA sequence element that contains a direct heptamer repeat TCTCAAA. It was found that presence of both of the repeats is essential for binding. ABF-2 is present in both green and etiolated tissue and in roots and factors analogous to ABF-2 are present in several plant species. Computer analysis showed that the TCTCAAA motif is present in the regulatory region of several plant genes.     

Adhesion of human dermal reticular fibroblasts on complementary fragments of fibronectin: aging in vivo or in vitro

Attachment, spreading, and microfilament reorganization have been evaluated in human dermal reticular fibroblasts isolated from the inner, upper aspect of the arm of a newborn male (RET5 cells) and a 78-year-old male (RET8 cells). Substrata were tested using a set of complementary fragments from individual polypeptide chains of human plasma fibronectin (pFN) or cellular FNs (cFN). With both cell classes, fragments containing the C-terminal heparin-binding (HepII) domain only elicited linear bundles of microfilaments in spreading cells but no stress fibers; fragments containing the RGDS-dependent cell-binding (CellI) domain elicited only partial spreading with condensations of F-actin at ruffling membranes and at other regions along the plasma membrane. The minimum sequence required to obtain responses identical to those on intact pFN (broad spreading with extensive stress fiber formation) was found in fragment 155 (F155) from the beta chain of pFN; F155 contains both HepII and CellI domains. In contrast, the analogous fragment from the alpha chain of pFN (F145) was notably less effective for generating stress fibers. This evidence along with the better attachment, spreading, and microfilament bundle formation on the HepII fragment from the beta chain than the analogous fragment from the alpha chain indicates that the extra type III homology unit permits more effective interaction of beta chain fragments with cell-surface heparan sulfate proteoglycan and possibly integrin (binding efficiency to the substratum was similar for fragments from both chains). Therefore, alternatively spliced sequences that neighbor binding domains can play significant roles in the interaction of the domain with cell-surface receptors of dermal fibroblasts. Comparison of RET5 responses with those of RET8 cells has identified changes in adhesive mechanisms as cells undergo "aging" processes. Attachment and microfilament bundle formation were far more effective for RET5 cells than for RET8 cells on any of the HepII fragments. Conversely, RET8 cells were far more sensitive to an RGDS-containing peptide in their medium on CellI fragments than RET5 cells. These results together indicate that in vivo aging leads to greater dependence upon cell-surface integrin binding and less dependence upon heparan sulfate proteoglycan binding for responses on FN matrices. When RET5 cells entered senescence (in vitro aging), they also became much more sensitive to peptide A. On several fragments and on intact pFN, RET8 cells generated very thick stress fibers that were observed only on one fragment with RET5 cells.(ABSTRACT TRUNCATED AT 400 WORDS)     

Differences in domain structure between human fibronectins isolated from plasma and from culture supernatants of normal and transformed fibroblasts. Studies with domain-specific antibodies

The domain structure of human fibronectins isolated from plasma and from the conditioned medium of normal and transformed fibroblasts was analyzed by limited proteolysis and S-cyanylation followed by immunostaining of released fragments with five kinds of antibodies, each specific for one functional domain. The results indicate that all three human fibronectins are composed of the same set of functional domains aligned in the same topological order. However, the following clear differences were found in specific fragments released from plasma fibronectin (pFN) and those released from fibronectin of normal (N-cFN) and transformed fibroblasts (T-cFN). Two fragments (Mr = 70,000 and 60,000) were released from the COOH-terminal region of pFN by cathepsin D. These fragments represent the COOH-terminal heparin-binding (Hep-2) and fibrin-binding (Fib-2) domains. The corresponding fragments released from both N-cFN and T-cFN by cathepsin D had much larger molecular weights (Mr = 100,000 and 83,000-74,000) than those from pFN. The fragments from the Fib-2 domain alone, however, did not show any difference among all three FNs. The internal region, from the gelatin-binding (Gel) domain through the Hep-2 domain, of N-cFN and T-cFN was released as a Mr = 210,000 fragment upon mild trypsin digestion. The corresponding fragment from pFN was released as a Mr = 185,000 fragment. The COOH-terminal half, including the Hep-2 domain, of both N-cFN and T-cFN was released by S-cyanylation as Mr = 160,000-145,000 fragments, which are 25,000-20,000 larger than the corresponding fragments of pFN. These results clearly indicate that the Hep-2 domain of N-cFN and T-cFN is 30,000-20,000 daltons larger than the same domain of pFN. Although various fragments released from N-cFN and T-cFN showed a similar pattern, there were minor differences. Thermolysin fragments derived from the Hep-2 domain of N-cFN were clearly distinguishable from those from T-cFN. Three groups of fragments with Mr = 40,000, 35,000-32,000, and 30,000 were released from N-cFN, while only the 35,000-32,000 fragment was released from T-cFN. The Mr = 44,000/60,000 thermolysin fragments representing the Gel domain and the Mr = 210,000/165,000 tryptic fragments representing the internal domains of T-cFN were slightly, but consistently, larger than those of N-cFN.(ABSTRACT TRUNCATED AT 400 WORDS)     

Structural differences in the cell binding region of human fibronectin molecules isolated from cultured normal and tumor-derived human cells

Fibronectins isolated from human plasma (pFN) and from the conditioned media of normal (N-cFN) and tumor (T-cFN) human cells were compared by cathepsin D digestion followed by immunostaining of released fragments with the monoclonal antibody 3E3, specific for the cell binding site. Two different staining patterns were obtained, one specific for pFN and N-cFN, the second common to fibronectins from the 3 different kinds of tumors studied. This indicates structural differences between N-cFN and T-cFN in the cell binding region of the fibronectin molecule.     

Spontaneous binding ofYersinia enterocolitica to murine leukocytes

Twelve strains ofYersinia enterocolitica were examined for their ability to bind spontaneously to murine leukocytes. Each of eight HeLa cell invasive strains exhibited nonselective binding to peritoneal leukocytes, lymph node leukocytes, and thymocytes, whereas four noninvasive strains lacked binding properties. Like the HeLa cell invasion, the binding ofY. enterocolitica to leukocytes was much less efficient for bacteria grown at 37°C than for bacteria grown at 22°C. The binding properties were not influenced by the virulence plasmid that codes for Vwa⁺ phenotype. This leukocyte binding test is proposed as a simple assay for invasive properties ofY. enterocolitica.     

The 21-aminosteroid U-74389F increases the number of glial fibrillary acidic protein-expressing astrocytes in the spinal cord of control and wobbler mice

Summary 1. Wobbler mice suffer an autosomal recessive mutation producing severe motoneuron degeneration and dense astrogliosis, with increased levels of glial fibrillary acidic protein (GFAP) in the spinal cord and brain stem. They have been considered animal models of amyotrophic lateral sclerosis and infantile spinal muscular atrophy. 2. Using Wobbler mice and normal littermates, we investigated the effects of the membrane-active steroid Lazaroid U-74389F on the number of GFAP-expressing astrocytes and glucocorticoid receptors (GR). Lazaroids are inhibitors of oxygen radical-induced lipid peroxidation, and proved beneficial in cases of CNS injury and ischemia. 3. Four days after pellet implantation of U-74389F into Wobbler mice, hyperplasia and hypertophy of GFAP-expressing astrocytes were apparent in the spinal cord ventral and dorsal horn, areas showing already intense astrogliosis in untreated Wobbler mice. In control mice, U-74389F also produced astrocyte hyperplasia and hypertophy in the dorsal horn and hyperplasia in the ventral-lateral funiculi of the cord. 4. Givenin vivo U-74389F did not change GR in spinal cord of Wobbler or control mice, in line with the concept that it is active in membranes but does not bind to GR. Besides, U-74390F did not compete for [³H]dexamethasone binding when addedin vitro. 5. The results suggest that stimulation of proliferation and size of GFAP-expressing astrocytes by U-74389F may be a novel mechanism of action of this compound. The Wobbler mouse may be a valuable animal model for further pharmacological testing of glucocorticoid and nonglucocorticoid steroids in neurodegenerative diseases.