Expression and characterization of human tissue inhibitor of metalloproteinases-1 in a baculovirus-insect cell system.

The baculovirus expression system was used to overexpress human tissue inhibitor of metalloproteinases-1 (TIMP-1). Approximately 5 mg of recombinant TIMP-1 was produced per 10(9) Sf9 insect cells infected with the recombinant baculovirus. The optimum time point for the production of biologically active rTIMP-1 was 20 hours postinfection. TIMP-1 activity was demonstrated by a soluble collagenase inhibition assay and reverse zymography. The baculovirus system has the advantage over previously described methods for generating human rTIMP-1 in that it generates large amounts of a fully glycosylated and active protein.

Studies on the effects of laminin, E-8 fragment of laminin and synthetic laminin peptides PA22-2 and YIGSR on matrix metalloproteinases and tissue inhibitor of metalloproteinase expression.

BACKGROUND: Based on a previous observation of laminin-mediated increase in type IV collagenolytic activity of human melanoma (A-2058) and fibrosarcoma (HT-1080) cell lines (Turpeenniemi-Hujanen T, et al. Laminin increases the release of type IV collagenase from malignant cells. J Biol Chem 1986;261:1883-1889), the goal of this study was to identify the proteinases involved. EXPERIMENTAL DESIGN: A soluble type IV collagenase assay and substrate gel electrophoresis were used to assess the effect of laminin and laminin peptides on type IV collagenolytic activity. RESULTS: The laminin-mediated increase in type IV collagenolytic activity was not due to augmented expression or induction of three known type IV collagenolytic matrix metalloproteinases (MMP), i.e., 72-kilodalton gelatinase/type IV collagenase (MMP-2), stromelysin (MMP-3), and 92 kilodalton gelatinase/type IV collagenase (MMP-9). Furthermore, laminin did not modulate the expression of the tissue inhibitors of metalloproteinases (TIMP), TIMP-1 and TIMP-2. The E-8 fragment of laminin and the YIGSR laminin peptide had no effect on type IV collagenolytic or MMP/TIMP activities. However, the IKVAV containing PA22-2 laminin peptide selectively stimulated type IV collagenolytic activity of the A-2058 melanoma cell line, although it did not modulate MMP/TIMP activity. Laminin from three different sources of the Engelbreth-Holm-Swarm tumor was found to contain type IV collagenolytic activity. When laminin was added to harvested culture supernatants of the A-2058 and HT-1080 cell lines, the increase in type IV collagenolytic activity was comparable with that observed in supernatants from cells incubated with laminin for 48 hours. Analysis of the laminin preparations revealed five MMP forms ranging in molecular weight from approximately 58 to 105 kilodalton, which may represent latent and active forms of MMP-2 and MMP-9. CONCLUSIONS: These findings suggest that proteinases present in the Engelbreth-Holm-Swarm laminin may account for most, if not all, of the observed laminin-mediated increase in type IV collagenolytic activity. However, the PA22-2-mediated increase in type IV collagenolytic activity of the A-2058 melanoma line remains to be elucidated.

Identification of the 72-kDa (MMP-2) and 92-kDa (MMP-9) gelatinase/type IV collagenase in preparations of laminin and Matrigel.

EDTA inhibitable type IV collagenolytic activity copurified with laminin preparations from the Engelbreth-Holm-Swarm (EHS) tumor. Several gelatinolytic and type IV collagenolytic matrix metalloproteinase (MMP) species were visualized in EHS laminin from three different sources by gelatin and type IV collagen substrate gel electrophoresis. Incubation with 4-aminophenylmercuric acetate and trypsin suggested that laminin contained both active and latent MMPs. EHS-derived reconstituted basement membrane, Matrigel, was found to possess an MMP profile identical to that of laminin. The presence of 72-kDa (MMP-2) and 92-kDa (MMP-9) gelatinases/type IV collagenases was demonstrated in laminin and Matrigel preparations by Western blot analysis. A rough quantitation of MMP-2 and MMP-9 in 30 micrograms of laminin and 100 micrograms of Matrigel was between 0.3 and 0.6 ng. The presence of these contaminants must be considered in experiments addressing the effects of EHS laminin or Matrigel on cell behavior and, in particular, stimulation of cellular proteolytic activity.

Immunomagnetic separation as a final purification step of liver endothelial cells.

We describe a fast and reproducible method that can be used as a final step in obtaining pure populations of liver endothelial cells. This method employs endothelial cell specific lectin covalently bound to magnetic polystyrene beads (Dynabeads). Evonymus europaeus agglutinin (EEA)-coated Dynabeads were used to purify monkey liver endothelium from Percoll gradient separated nonparenchymal cells. EEA-coated beads were also successfully used to purify monkey aortic endothelial cells. The endothelial cells grew to confluence as a cobblestonelike monolayer, expressed Factor VIII related antigen, and incorporated acetylated-low density lipoprotein. The magnetic beads seemed not to modify the normal properties of the isolated endothelium, thus facilitating their use in experimental studies. This immunomagnetic separation technique may be applicable for purification of endothelial cells from a wide variety of tissue sources.

Immunomagnetic separation as a final purification step of liver endothelial cells.

We describe a fast and reproducible method that can be used as a final step in obtaining pure populations of liver endothelial cells. This method employs endothelial cell specific lectin covalently bound to magnetic polystyrene beads (Dynabeads). Evonymus europaeus agglutinin (EEA)-coated Dynabeads were used to purify monkey liver endothelium from Percoll gradient separated nonparenchymal cells. EEA-coated beads were also successfully used to purify monkey aortic endothelial cells. The endothelial cells grew to confluence as a cobblestonelike monolayer, expressed Factor VIII related antigen, and incorporated acetylated-low density lipoprotein. The magnetic beads seemed not to modify the normal properties of the isolated endothelium, thus facilitating their use in experimental studies. This immunomagnetic separation technique may be applicable for purification of endothelial cells from a wide variety of tissue sources.

Effect of phorbol ester and cytokines on matrix metalloproteinase and tissue inhibitor of metalloproteinase expression in tumor and normal cell lines.

The effect of the tumor promoter 12-O-tetradecanoylphorbol-13-acetate (TPA) and the cytokines interleukin-1 (IL-1) and tumor necrosis factor-alpha (TNF-alpha) on matrix metalloproteinases (MMP) and metalloproteinase inhibitors was studied in a variety of human cell lines. Expression of the mammalian collagenase (MMP-1), 72-kD gelatinase/type IV collagenase (MMP-2), stromelysin (MMP-3), 92-kD gelatinase/type IV collagenase (MMP-9), and tissue inhibitors of metalloproteinases (TIMP-1 and TIMP-2) was assessed by zymography and Northern blot analysis. MMP-2 and TIMP-2 activities were refractory to TPA, IL-1 and TNF-alpha treatment in most of the cell lines. In contrast, MMP-3, MMP-9 and TIMP-1 activities were markedly stimulated by TPA in most of the tumor cell lines and human umbilical vein endothelial cells (HUVEC), whereas the fibroblast lines were minimally stimulated or unresponsive to TPA. The MMP-3, MMP-9 and TIMP-1 stimulation in response to IL-1 and TNF-alpha treatment was detected in some of the tumor cell lines and HUVEC. The increase in activity was less marked than in TPA. A breast carcinoma cell line, MDA-MB-231, which did not express MMP-2, had high expression of MMP-3 and MMP-9 which were unaffected by TPA and cytokine treatment. Northern blot analysis of MMP and TIMP mRNA expression reflected the zymogram findings for most of the cell lines. TPA-mediated stimulation of MMP-1 was similar to that of MMP-3 and MMP-9. Exceptions were the fibroblast cell lines which showed either a much more marked mRNA response of MMP-9 to TPA than observed at protein level, or a high constitutive MMP-9 mRNA when MMP-9 activity was not detectable by zymography. TPA-mediated stimulation of MMP-9 and TIMP-1 activity was blocked by staurosporine, an inhibitor of protein kinase C (PKC). A non-PKC-activating phorbol ester, 4 alpha-phorbol-12,13-didecanoate, did not stimulate MMP-9 and TIMP-1 activity. TPA treatment caused the increased expression of c-fos containing AP-1-specific binding activity in selected tumor cell lines. This activity was maximal at 6 h. An association was observed between AP-1 binding activity and increased expression of MMP-1, MMP-3 and MMP-9, which possess TPA-responsive elements (TRE). TPA-sensitive MMPs and TIMP-1 were variably stimulated by biologically relevant cytokines, such as IL-1 and TNF-alpha.(ABSTRACT TRUNCATED AT 400 WORDS)

Ras levels and metalloproteinase activity in normal versus neoplastic rat mammary tissues.

We have previously reported that activated ras oncogenes can simultaneously switch on the metastatic phenotype and increased capability to degrade type IV collagen. Here the relationship between c-H-ras, metalloproteinase expression and metastatic behavior was studied in N-nitrosomethylurea (NMU)-induced rat mammary carcinomas, which are known to possess activated c-H-ras. When comparing normal rat breast tissue to mammary carcinomas there was no direct relationship between ras DNA levels and neoplastic changes. Furthermore, there were no consistent differences between metastatic and non-metastatic carcinomas, or between primary tumors and metastases. The NMU-induced rat mammary carcinomas expressed two major gelatinolytic metalloproteinases (gelatinases) of 65 and 92 kD, but only the 65 kD gelatinase was detected in normal breast tissue and a rat fibroma. Type IV collagenolytic activity per 5 micrograms of protein was two to three times higher in the mammary carcinomas than in the normal breasts, whereas the primary tumors did not differ from the corresponding metastases. This study shows that ras amplification is not necessary for development of the malignant or metastatic phenotype in the NMU-induced rat mammary carcinoma model. We have also found that induction of p21 ras protein synthesis in a v-H-ras transfected NIH/3T3 (433) cell line, containing a glucocorticoid promoter, does not lead to an increase in metastatic capacity.

Regulation of glutamate dehydrogenases in nit-2 and am mutants of Neurospora crassa.

The regulation of the glutamate dehydrogenases was investigated in wild-type Neurospora crassa and two classes of mutants altered in the assimilation of inorganic nitrogen, as either nitrate or ammonium. In the wild-type strain, a high nutrient carbon concentration increased the activity of reduced nicotinamide adenine dinucleotide phosphate (NADPH)-glutamate dehydrogenase and decreased the activity of reduced nicotinamide adenine dinucleotide (NADH)-glutamate dehydrogenase. A high nutrient nitrogen concentration had the opposite effect, increasing NADH-glutamate dehydrogenase and decreasing NADPH-glutamate dehydrogenase. The nit-2 mutants, defective in many nitrogen-utilizing enzymes and transport systems, exhibited low enzyme activities after growth on a high sucrose concentration: NADPH-glutamate dehydrogenase activity was reduced 4-fold on NH(4)Cl medium, and NADH-glutamate dehydrogenase, 20-fold on urea medium. Unlike the other affected enzymes of nit-2, which are present only in basal levels, the NADH-glutamate dehydrogenase activity was found to be moderately enhanced when cells were grown on a low carbon concentration. This finding suggests that the control of this enzyme in nit-2 is hypersensitive to catabolite repression. The am mutants, which lack NADPH-glutamate dehydrogenase activity, possessed basal levels of NADH-glutamate dehydrogenase activity after growth on urea or l-aspartic acid media, like the wild-type strain, and possessed moderate levels (although three- to fourfold lower than the wild-type strain) on l-asparagine medium or l-aspartic acid medium containing NH(4)Cl. These regulatory patterns are identical to those of the nit-2 mutants. Thus, the two classes of mutants exhibit a common defect in NADH-glutamate dehydrogenase regulation. Double mutants of nit-2 and am had lower NADH-glutamate dehydrogenase activities than either parent. A carbon metabolite is proposed to be the repressor of NADH-glutamate dehydrogenase in N. crassa.

Purification and characterization of homogeneous assimilatory reduced nicotinamide adenine dinucleotide phosphate-nitrate reductase from Neurospora crassa.

Neurospora crassa wild type STA4 NADPH-nitrate reductase (NADPH : nitrate oxidoreductase, EC 1.6.6.3) has been purified 5000-fold with an overall yield of 25--50%. The final purified enzyme contained 4 associated enzymatic activities: NADPH-nitrate reductase, FADH2-nitrate reductase, reduced methyl viologen-nitrate reductase and NADPH-cytochrome c reductase. Polyacrylamide gel electrophoresis yielded 1 major and 1 minor protein band and both bands exhibited NADPH-nitrate and reduced methyl viologen-nitrate reductase activities. SDS gel electrophoresis yielded 2 protein bands corresponding to molecular weights of 115 000 and 130 000. A single N-terminal amino acid (glutamic acid) was found and proteolytic mapping for the two separated subunits appeared similar. Purified NADPH-nitrate reductase contained 1 mol of molybdenum and 2 mol of cytochrome b557 per mol protein. Non-heme iron, zinc and copper were not detectable. It is proposed that the Neurospora assimilatory NADPH-nitrate reductase consists of 2 similar cytochrome b557-containing 4.5-S subunits linked together by one molybdenum cofactor. A revised electron flow scheme is presented. p-Hydroxymercuribenzoate inhibition was reversed by sulfhydryl reagents. Inhibitory pattern of p-hydroxymercuribenzoate and phenylglyoxal revealed accessible sulfhydryl and arginyl residue(s) as functional group(s) in the earlier part of electron transport chain as possibly the binding site of NADPH or FAD.

Induction and repression of nitrate reductase in Neurospora crassa.

Synthesis of wild-type Neurospora crassa assimilatory nitrate reductase is induced in the presence of nitrate ions and repressed in the presence of ammonium ions. Effects of several Neurospora mutations on the regulation of this enzyme are shown: (i) the mutants, nit-1 and nit-3, involving separate lesions, lack reduced nicotinamide adenine dinucleotide (NADPH)-nitrate reductase activity and at least one of three other activities associated with the wild-type enzyme. The two mutants do not require the presence of nitrate for induction of their aberrant nitrate reductases and are constitutive for their component nitrate reductase activities in the absence of ammonium ions. (ii) An analog of the wild-type enzyme (similar to the nit-1 enzyme) is formed when wild type is grown in a medium in which molybdenum has been replaced by vanadium or tungsten; the resulting enzyme lacks NADPH-nitrate reductase activity. Unlike nit-1, wild type produced this analog only in the presence of nitrate. Contaminating nitrate does not appear to be responsible for the observed mutants' activities. Nitrate reductase is proposed to be autoregulated. (iii) Mutants (am) lacking NADPH-dependent glutamate dehydrogenase activity partially escape ammonium repression of nitrate reductase. The presence of nitrate is required for the enzyme's induction. (iv) A double mutant, nit-1 am-2, proved to be an ideal test system to study the repressive effects of nitrogen-containing metabolites on the induction of nitrate reductase activity. The double mutant does not require nitrate for induction of nitrate reductase, and synthesis of the enzyme is not repressed by the presence of high concentrations of ammonium ions. It is, however, repressed by the presence of any one of six amino acids. Nitrogen metabolites (other than ammonium) appear to be responsible for the mediation of "ammonium repression."

Interactions of trace metals in mouse and rat tissues; zinc, chromium, copper, and manganese with 13 other elements.

Tissues of rats and mice fed a nonessential metal in drinking water for life were analyzed for the essential metals chromium, copper, manganese and zinc. The study involved 505 rats and 843 mice. Livers, lungs, hearts, kidneys and spleens were pooled in groups according to age at death, averaging 5 for rats and 8 for mice, in order to provide adequate sample weight. Copper was significantly higher in livers of rats fed tin, germanium, niobium and zirconium than in controls. Similarly, niobium was associated with deposition of manganese in heart and zinc deposition in liver. Chromium levels were depressed in heart, kidney and spleen by germanium. In mice the greatest effects occurred when indium and rhodium were fed, all four essential trace metals exhibiting raised levels principally in kidney but also in heart and spleen. Chromium levels were raised in all organs but heart when hexavalent chromium was fed. From these data it is apparent that the ingestion of a nonessential metal can enhance the retention of an essential trace metal, perhaps thus avoiding toxicity from the nonessential one.

Invitro formation of assimilatory reduced nicotinamide adenine dinucleotide phosphate: nitrate reductase from a Neurospora mutant and a component of molybdenum-enzymes.

An active Neurospora-like assimilatory NADPH-nitrate reductase (EC 1.6.6.2), which can be formed in vitro by incubation of extracts of nitrate-induced Neurospora crassa mutant nit-1 with extracts of (a) certain other nonallelic nitrate reductase mutants, (b) uninduced wild type, or (c) xanthine oxidizing and liver aldehyde-oxidase systems was also formed by combination of the nit-1 extract with other acid-treated enzymes known to contain molybdenum. These molybdenum enzymes included (a) nitrogenase, or its molybdenum-iron protein, from Clostridium, Azotobacter, and soybeannodule bacteroids, (b) bovine liver sulfite oxidase, (c) respiratory formate-nitrate reductase from Escherichia coli, (d) NADH-nitrate reductase from foxtail grass (Setaria faberii), and (e) FADH(2)- and reduced methyl viologennitrate reductase preparations from certain Neurospora mutants. Several molybdenum-amino-acid complexes, as possible catalytic models of nitrogenase, were inactive (as were some previously tested 20 nonmolybdenum enzymes) in place of the acid-treated molybdenum-containing enzymes. The results imply the existence of a molybdenum-containing component shared by the known molybdenum-enzymes.