Role of fibroblast growth factor receptors (FGFR) and FGFR like-1 (FGFRL1) in mesenchymal stromal cell differentiation to osteoblasts and adipocytes.

Fibroblast growth factors (FGF) and their receptors (FGFRs) regulate many developmental processes including differentiation of mesenchymal stromal cells (MSC). We developed two MSC lines capable of differentiating to osteoblasts and adipocytes and studied the role of FGFRs in this process. We identified FGFR2 and fibroblast growth factor receptor like-1 (FGFRL1) as possible actors in MSC differentiation with gene microarray and qRT-PCR. FGFR2 and FGFRL1 mRNA expression strongly increased during MSC differentiation to osteoblasts. FGF2 treatment, resulting in downregulation of FGFR2, or silencing FGFR2 expression with siRNAs inhibited osteoblast differentiation. During adipocyte differentiation expression of FGFR1 and FGFRL1 increased and was down-regulated by FGF2. FGFR1 knockdown inhibited adipocyte differentiation. Silencing FGFR2 and FGFR1 in MSCs was associated with decreased FGFRL1 expression in osteoblasts and adipocytes, respectively. Our results suggest that FGFR1 and FGFR2 regulate FGFRL1 expression. FGFRL1 may mediate or modulate FGFR regulation of MSC differentiation together with FGFR2 in osteoblastic and FGFR1 in adipocytic lineage.

Selective augmentation of histone H1 phosphorylation sites by interaction of poly(ADP-ribose) polymerase and cdc2-kinase: comparison with protein kinase C.

The molecular interactions between PARP I, cdc2-kinase, PKC and histone H1 were determined with the aid of the common phosphate acceptor function of histone H1 to both kinases. PKC phosphorylates both histone H1 and PARP I and PARP I augments the acceptor function of histone H1. When both acceptors (PARP I and histone H1) are present an apparent distributive phosphorylation of both acceptors takes place. In contrast, cdc2-kinase only phosphorylates histone H1, and the activation of this reaction by PARP I does not involve PARP I-cdc2-kinase binding only PARP I-histone H1 association. Since the phosphorylation of histone H1 by PKC is a model reaction with no apparent physiologic consequences, the PARP I activated phosphorylation of histone H1 by cdc2-kinase, by contrast, reflects a physiologically meaningful regulation of the linker histone by a cyclin dependent kinase (cdc2-kinase). The increased phosphorylation of histone H1 by cdc2-kinase following PARP I-histone H1 binding results in the appearance of new phosphorylated histone H1 polypeptides as measured by proteolytic digestion and re-electrophoresis of cdc2-kinase phosphorylated polypeptides, indicating a probable conformational change in histone H1, following PARP I binding. The cell biologic significance of this reaction in PARP I ligand-induced enzyme induction is briefly analysed.

Molecular interactions between poly(ADP-ribose) polymerase (PARP I) and topoisomerase I (Topo I): identification of topology of binding.

The molecular interactions of poly(ADP-ribose) polymerase I (PARP I) and topoisomerase I (Topo I) have been determined by the analysis of physical binding of the two proteins and some of their polypeptide components and by the effect of PARP I on the enzymatic catalysis of Topo I. Direct association of Topo I and PARP I as well as the binding of two Topo I polypeptides to PARP I are demonstrated. The effect of PARP I on the 'global' Topo I reaction (scission and religation), and the activation of Topo I by the 36 kDa polypeptide of PARP I and catalytic modifications by poly(ADP-ribosyl)ation are also shown. The covalent binding of Topo I to circular DNA is activated by PARP I similar to the degree of activation of the 'global' Topo I reaction, whereas the religation of DNA is unaffected by PARP I. The geometry of PARP I-Topo I interaction compared to automodified PARP I was reconstructed from direct binding assays between glutathione S-transferase fusion polypeptides of Topo I and PARP I demonstrating highly selective binding, which was correlated with amino acid sequences and with the 'C clamp' model derived from X-ray crystallography.

Activation of topoisomerase I by poly [ADP-ribose] polymerase.

Poly(ADP-ribose) polymerase (PARP I) and Topoisomerase I (Topo I) were reisolated from calf thymus to eliminate cross contamination as tested by immunotransblots. The specific activity of Topo I was greatly increased by added PARP I, following saturation kinetics. Recombinant PARP I and isolated PARP I at final purity were indistinguishable in terms of their activation of Topo I. There was a coincidence of experimentally obtained binding constants and computer generated values based on the kinetic model, indicating that the association of PARP I and Topo I is rate limiting in the catalytic activation of Topo I by PARP I. Polypeptide domains of PARP I that are required for protein-protein binding and protein-DNA binding also activate Topo I. Fluorescence resonance energy transfer between fluorophor-labeled PARP I and Topo I was demonstrated. The binding of Topo I to circular SV40 DNA, assayed either by the formation of a) the sum of non-covalently and covalently attached Topo I to DNA or b) by the covalently bound transient intermediate in the presence of camptothecin, was augmented when PARP I protein was bound to SV40 DNA. These binding experiments provide a molecular basis for the kinetic activation of Topo I by PARP I inasmuch as the increased superhelicity of SV40 DNA induced by PARP I may facilitate the formation of a more Topo I-DNA complex that increases the rate of the DNA breakage-reunion cycle of Topo I catalysis.

Unusual potentiation by vinca alkaloids of the cytostatic and cytocidal action of methyl-3,5-diiodo-4-(4'-methoxyphenoxy) benzoate (DIME) and its nonhydrolyzable ethanone analog (DIPE) on MDA-MB-231 human mammary cancer cells.

Drug interaction between DIME or DIPE ¿1-[3, 5-diiodo-4-(4'-methoxyphenoxy)-phenyl]-ethanone¿ with vincristine and vinblastine on the growth rate of MDA-MB-231 human mammary cancer cells was determined by the median effect kinetic method. Mutually exclusive cellular binding sites were identified kinetically and isobologram analyses showed potentiation. The combind effect of 0.75 MICROM DIME and 2 nM vincristine demonstrated a nearly type of mutual activation. It was shown that the nonhydrolyzable DIME derivative DIPE is equivalent to DIME, but because of its biological stability is a preferred drug candidate. Vinblastine-DIME cooperative action is similar to that of vincristine-DIME (or DIPE). Activation of caspase 3 by both DIME and vincristine is greatly potentiated when both drugs are added simultaneously in a given proportion. We propose that following a primary binding of DIME and vinca alkaloids to microtubules, an as yet unrecognized mutual activation of caspase 3 apoptotic path is initiated, explaining DNA fragmentation and cell death. A subpopulation of cancer cells, capable of slow growth at 1.5 microM DIME was identified. This cell type was also killed by the DIME-vincristine drug combination.

Isolation and identification of a proteinase from calf thymus that cleaves poly(ADP-ribose) polymerase and histone H1.

A proteinase was isolated from calf thymus that degraded pADPRT, histone H1 and alpha-casein in a Ca(2+)-dependent manner. In a five-step procedure, a homogenous proteinase was obtained with a subunit structure of 80 and 30 kDa. The amino-acid homology of an internal sequence as well as kinetic and inhibitor assays identified the proteinase as calpain I. It is suggested that even though the general substrate alpha-casein is widely used for the assaying of calpains, more appropriately physiological cellular components (pADPRT and histone H1) specify the thymus proteinase.

Potential chemotherapeutic activity of 4-iodo-3-nitrobenzamide. Metabolic reduction to the 3-nitroso derivative and induction of cell death in tumor cells in culture.

A C-nitroso prodrug, 4-iodo-3-nitrobenzamide, was synthesized, and its action on a variety of tumor cells of human and animal origin tested. This prodrug was reduced transiently by tumor cells to 4-iodo-3-nitrosobenzamide at a very low rate, which was, however, sufficient to kill tumor cells. The final reduction product was 4-iodo-3-aminobenzamide, and no intermediates accumulated. No toxicity could be observed in hamsters even at 200 mg/kg, given i.p. daily for 7 days. The chemical reactivity of both 4-iodo-3-nitrosobenzamide and its noniodinated homolog with reduced ascorbate yielded the hydroxylamines. With glutathione, 4-iodo-3-aminobenzamide was formed, suggesting glutathione sulfinic acid formation. Confirming earlier studies, 4-iodo-3-nitrosobenzamide inactivated poly(ADP-ribose) polymerase by zinc ejection from the first zinc finger of this nuclear protein. The iodinated nitroso compound was more effective than its iodine-free analog. Selective tumoricidal action appeared to correlate with the reduction of the nitro group to nitroso in tumor cells, and with the previously described subsequent induction of tumor apoptosis by the C-nitroso intermediate. These processes were accelerated by buthionine sulfoximine, which diminishes cellular GSH.

Identification of domains of poly(ADP-ribose) polymerase for protein binding and self-association.

Cellular proteins extracted from normal and cancer cells bind polymerizing ADP-ribose transferase (pADPRT) on nitrocellulose membrane transblots. Histones at 1 mg/ml concentration completely prevent the binding of pADPRT to cellular proteins, indicating that the binding of histones to pADPRT sites competitively blocks the association of pADPRT to proteins other than histones. The direct binding of pADPRT to histones is shown by cross-linking with glutaraldehyde. The COOH-terminal basic histone H1 tail binds to the basic polypeptide domain of pADPRT. The basic domain present in the NH2-terminal part of core histones is the probable common structural feature of all core histones that accounts for their binding to pADPRT. Two polypeptide domains of pADPRT were identified, by way of CNBr fragments, to bind histones. These two domains are located within the 64-kDa fragment of pADPRT and are contiguous with the polypeptide domains that were shown to participate in self-association of pADPRT, ending at the 606th amino acid residue. The polypeptide domains of pADPRT which participate in DNA binding are thus shown to associate also with other proteins. Intact pADPRT binds to both the zinc-free or zinc-reconstituted basic polypeptide fragments of pADPRT. Histones activate auto-poly(ADP)-ribosylation of pADPRT by increasing the number of short oligomers on pADPRT. This reaction is also dependent in a biphasic manner on the concentration of pADPRT. Histones in solution are only marginally poly(ADP)-ribosylated but are good polymer acceptors when incorporated into artificial nucleosome structures.

Induction of endonuclease-mediated apoptosis in tumor cells by C-nitroso-substituted ligands of poly(ADP-ribose) polymerase.

6-Nitroso-1,2-benzopyrone and 3-nitrosobenzamide, two C-nitroso compounds that inactivate the eukaryotic nuclear protein poly(ADP-ribose) polymerase [NAD+:poly(adenosine diphosphate D-ribose) ADP-D-ribosyltransferase, ADPRT, EC 2.4.2.30] at one zinc-finger site, completely suppressed the proliferation of leukemic and other malignant human cells and subsequently produced cell death. Tumoricidal concentrations of the drugs were relatively harmless to normal bone marrow progenitor cells and to superoxide formation by neutrophil granulocytes. The cellular mechanism elicited by the C-nitroso compounds consists of apoptosis due to DNA degradation by the nuclear calcium/magnesium-dependent endonuclease. This endonuclease is maintained in a latent form by poly(ADP-ribosyl)ation, but inactivation of ADPRT by C-nitroso drugs derepresses the DNA-degrading activity. ADPRT is thus identified as a critical regulatory enzyme component of a DNA-binding multiprotein system that plays a central function in defining DNA structures in the intact cell.

Inhibition of HIV-1 IIIb replication in AA-2 and MT-2 cells in culture by two ligands of poly (ADP-ribose) polymerase: 6-amino-1,2-benzopyrone and 5-iodo-6-amino-1,2-benzopyrone.

The effects of two adenosine diphosphoribose transferase (ADPRT) enzyme inhibitory ligands, 6-amino-1,2-benzopyrone and its 5-iodo-derivative, were determined in AA-2 and MT-2 cell cultures on the replication of HIV-1 IIIb, assayed by an immunochemical test for the HIV protein p24, and syncytium formation, characteristic of HIV-infected cells. Intracellular concentrations of both drugs were sufficient to inhibit poly(ADP-ribose) polymerase activity within the intact cell. Both drugs inhibited HIV replication parallel to their inhibitory potency on ADPRT, but distinct differences were ascertained between the two cell lines. In AA-2 cells both p24 and syncytium formation were depressed simultaneously, whereas in MT-2 cells only syncytium formation was inhibited by the drugs, and the p24 production, which remained unchanged during viral growth, was unaffected. Both drugs only moderately depressed the growth rate of the AA-2 and MT-2 cells and there was no detectable cellular toxicity. Results suggest the feasibility of the development of a new line of ADPRT ligand anti-HIV drugs that fundamentally differ in their mode of action from currently used chemotherapeutics.

Inhibitory binding of adenosine diphosphoribosyl transferase to the DNA primer site of reverse transcriptase templates.

Purified adenosine diphosphoribose transferase protein binds to RNA-DNA hybrid templates of reverse transcriptase at the DNA primer site and inhibits RT activity of HIV and MMu RTs. This action is prevented by auto-poly-ADP-ribosylation of the transferase but is reinduced by inhibitory ligands of the enzyme.

Destabilization of Zn2+ coordination in ADP-ribose transferase (polymerizing) by 6-nitroso-1,2-benzopyrone coincidental with inactivation of the polymerase but not the DNA binding function.

6-Nitroso-1,2-benzopyrone, an oxidation product of 6-amino-1,2-benzopyrone, binds to the DNA-recognizing domain of the ADP-ribose transferase protein and preferentially destabilizes Zn2+ from one of the two zinc finger polypeptide complexes present in the intact enzyme, as determined by the loss of 50% of 65Zn2+ from the 65Zn(2+)-isolated protein molecule, coincidental with the loss of 99% of enzymatic activity. The 50% zinc-deficient enzyme still binds to a DNA template, consisting of a 17-mer DNA primer annealed to M13 positive strand, resulting in the blocking of DNA synthesis by the Klenow fragment of Pol I. Auto-poly-ADP-ribosylated ADP-ribose transferase, which is the probable physiological state of this protein in intact cells, does not bind to primer-template DNA and does not block DNA synthesis by the Klenow fragment. On the basis of this in vitro model it is proposed that molecules which inhibit or inactivate ADP-ribose transferase in intact cells can induce significant alteration in DNA structure and replication.

Evidence for the participation of histidine residues located in the 56 kDa C-terminal polypeptide domain of ADP-ribosyl transferase in its catalytic activity.

Purified ADPRT protein was inactivated by the histidine specific reagent diethylpyrocarbonate, binding to two histidine residues, or by a relatively histidine selective photoinactivation method. Inactivation with up to 1.3 mM diethylpyrocarbonate was reversible by hydroxylamine. Enzymatic inactivation coincided with the loss of binding capacity of the enzyme protein to benzamide affinity matrix but not to DNA cellulose. Labelled diethylpyrocarbonate was identified exclusively in the 56 kDa carboxyl-terminal polypeptide where 2 out of 13 histidine residues were modified by this reagent. It is proposed that histidine residues in the 56 kDa polypeptide may participate as initiator sites for polyADP-ribosylation.

Macromolecular association of ADP-ribosyltransferase and its correlation with enzymic activity.

The macromolecular self-association of ADP-ribosyltransferase protein in solution was studied by several experimental techniques: quantitative gel filtration, electrophoretic analyses in non-denaturing gels, and cross-linking the enzyme protein with glutaraldehyde, dimethyl pimelimidate, dimethyl suberimidate, dimethyl 3,3'-dithiobisproprionimidate and tetranitromethane. The self-association of the polypeptide components obtained by plasmin digestion was also determined by using the above cross-linking agents. Monomers and cross-linked dimers of the enzyme protein, possessing enzymic activity, were separated in non-denaturing gels by electrophoresis. The basic polypeptide fragments, exhibiting molecular masses of 29 kDa and 36 kDa, self-associated, whereas the polypeptides with molecular masses of 56 kDa and 42 kDa associated only to a negligible extent, indicating that the peptide regions that also bind DNA and histones are probable sites of self-association in the intact enzyme molecule. Macromolecular association of the enzyme was indicated by a protein-concentration-dependent red-shift in protein fluorescence. The specific enzymic activity of the isolated ADP-ribosyltransferase depended on the concentration of the enzyme protein, and at 2.00 microM concentration the enzyme was self-inhibitory. Dilution of the enzyme protein to 30-40 nM resulted in a large increase in its specific activity. Further dilution to 1-3 nM coincided with a marked decrease of specific activity. Direct enzymic assays of electrophoretically separated monomers and cross-linked dimers demonstrated that the dimer appears to be the active molecular species that catalyses poly(ADP-ribose) synthesis. The NAD+ glycohydrolase activity of the enzyme was also dependent on protein concentration and was highest at 1-3 nM enzyme concentration, when polymerase activity was minimal, indicating that the monomeric enzyme behaved as a glycohydrolase, whereas poly(ADP-ribosyl)ation of enzyme molecules was maximal when the enzyme tends to be self-associated to the dimeric form.

Binding of adenosine diphosphoribosyltransferase to the termini and internal regions of linear DNAs.

Binding mechanisms of ADPR-transferase to restricted double-stranded DNA fragments of SV40 and pBR322 DNA were determined by nuclease protection techniques. Top and bottom strands of double-stranded DNA were identified by specific labeling with 32P. Protection against specific exonucleases identified binding of ADPR-transferase to DNA termini, whereas binding to internal regions of linear DNAs was probed by protection against endonucleases. ADPR-transferase protein protected against exonucleolytic attack from lambda exo and exoIII in all DNA fragments tested, demonstrating that the enzyme protein binds indiscriminately to all DNA termini. Extending earlier results [Sastry, S.S., & Kun, E. (1988) J. Biol. Chem. 263, 1505-1512], the modifying effect of the binding of ADPR-transferase to DNA induced changes in DNA conformation, as evident from altered pause sites that appeared following digestion of DNA fragments by lambda exonuclease in the presence of ADPR-transferase. In contrast to the nonselective binding of ADPR-transferase to DNA termini, ADPR-transferase conferred protection endonuclease attack (DNase I and micrococcal nuclease) only to the 209-bp EcoRI-PstI SV40 DNA fragment. These results indicate that binding of ADPR-transferase to relatively rare internal regions of restricted DNA fragments exhibits some degree of specificity. Specificity of binding appears to be related to the coincidental relative A+T-rich regions in DNA, and to DNA bending, both identified in the 209-bp SV40 DNA fragment. Synthetic polydeoxyribonucleotides containing dA-dT bind ADPR-transferase stronger than polydeoxyribonucleotides containing dG-dC. It was deduced from endonuclease protection patterns that binding of the enzyme protein leaves no defined footprints on the 209-bp SV40 DNA fragment, but there is significant modification of DNA structure following binding of the enzyme protein. Methylation protection assays and the prevention of the binding of ADPR-transferase to T4 DNA by its glucosylation indicate that the enzyme binds in the major groove of DNA. The 36-kDa A peptide fragment of ADPR-transferase [Buki, K. G., & Kun, E. (1988) Biochemistry 27, 5990-5995] exhibits the same protection against endonucleolytic enzymes as the intact ADPR-transferase molecule.

Polypeptide domains of ADP-ribosyltransferase obtained by digestion with plasmin.

Proteolysis by plasmin inactivates bovine ADP-ribosyltransferase; therefore, enzymatic activity depends exclusively on the intact enzyme molecule. The transferase was hydrolyzed by plasmin to four major polypeptides, which were characterized by affinity chromatography and N-terminal sequencing. Based on the cDNA sequence for human ADP-ribosyltransferase enzyme [Uchida, K., Morita, T., Sato, T., Ogura, T., Yamashita, R., Noguchi, S., Suzuki, H., Nyunoya, H., Miwa, M., & Sugimura, T. (1987) Biochem. Biophys. Res. Commun. 148, 617-622], a polypeptide map of the bovine enzyme was constructed by superposing the experimentally determined N-terminal sequences of the isolated polypeptides on the human sequence deduced from its cDNA. Two polypeptides, the N-terminal peptide (Mr 29,000) and the polypeptide adjacent to it (Mr 36,000), exhibited binding affinities toward DNA, whereas the C-terminal peptide (Mr 56,000), which accounts for the rest of the transferase protein, bound to the benzamide-Sepharose affinity matrix, indicating that it contains the NAD+-binding site. The fourth polypeptide (Mr 42,000) represents the C-terminal end of the larger C-terminal fragment (Mr 56,000) and was formed by a single enzymatic cut by plasmin of the polypeptide of Mr 56,000. The polypeptide of Mr 42,000 still retained the NAD+-binding site. The plasmin-catalyzed cleavage of the polypeptide of Mr 56,000-42,000 was greatly accelerated by the specific ligand NAD+. Out of a total of 96 amino acid residues sequenced here, there were only 6 conservative replacements between human and bovine ADP-ribosyltransferase.

Isolation of adenosine diphosphoribosyltransferase by precipitation with reactive red 120 combined with affinity chromatography.

The DNA-associating enzyme, adenosine diphosphoribosyltransferase, has been isolated from calf thymus by selective precipitation with a solution of dihydroxy Reactive Red 120, followed by extraction of the enzyme from the precipitate with 2 M KCl and an on-line train of three successive column chromatographic steps, including a final 3-aminobenzamide-Sepharose 4B affinity chromatography. The method yields 8-9 mg of more than 95% homogeneous enzyme protein per kilogram starting material and requires about 3 working days. This dye precipitation method is distinct from affinity precipitation, since it involves the binding of the dye to both nonspecific sites and the substrate and DNA sites of the transferase as indicated by enzyme inhibition by dihydroxy Reactive Red 120 at both enzyme sites.

Interspecific interaction of aortic endothelial and smooth muscle cells.

The proliferation stimulating or inhibitory effect of aortic endothelial cells on aortic smooth muscle cells and vice versa was studied in cell culture by using a newly developed co-cultivation technique. The effect was registered as the incorporation of labelled thymidine detected by liquid scintillation counting. The experiments were performed in intra and interspecific combinations, using bovine, pig or rat aortic cell cultures. The endothelial cells stimulated smooth muscle cell proliferation in a species specific way. In interspecies combinations either inhibition or no effect was observed. The smooth muscle cells usually had an inhibitory effect on the proliferation of endothelial cells except for the pig smooth muscle cells which had a stimulatory effect in both inter and intraspecies combinations. Co-cultivation of smooth muscle or endothelial cells with cells of the same type resulted mostly in inhibition or no effect in both intra and interspecies combinations. This phenomenon was particularly conspicuous with endothelial cells. The different stimulatory and inhibitory effects produced by these cells may play an important role in the regulation of neovascularization and in reendothelization of denuded intimal areas.

Metabolism of non-enzymic glycosylated low density lipoprotein by mini pig aortic endothelial cells.

[U-14C]-glucose was incorporated non-enzymatically into the protein moiety of human low density lipoprotein. The incorporation was time and glucose concentration dependent. Investigated mini pig aortic endothelial cells showed that the glycosylated low density lipoprotein was bound, internalized and degraded significantly less than that of the native one measured with [125-I]-low density lipoprotein.

A low-Mr chemoattractant for vascular endothelial cells.

The formation of new blood vessels occurs by sprouting from previously existing microvasculature. The process involved directed migration of the vascular endothelial cells towards chemical signals released from the target tissue. We have used the Boyden chemotaxis chamber method to identify chemotactic signals for fetal bovine vascular endothelial cells. Human placenta organ cultures produce a high-Mr chemoattractant for the endothelial cells from which a low-Mr factor can be liberated with trichloroacetic acid treatment and ethanol extraction. This activity was isolated from extracts of human placenta using Sephadex LH-20, Amberlite XAD-2, and silica gel thin-layer chromatography. The Mr of the factor is less than 400, it is lipophilic and resistant to proteolytic enzymes. The factor induces chemotactic migration of both aortic endothelial cells and capillary endothelial cells from the retina, but has no effect on fibroblasts or leukocytes suggesting a specific function of the compound for the vascular endothelial cells.