Purification and sequence analysis of two rat tissue inhibitors of metalloproteinases.

Two protein inhibitors of metalloproteinases (TIMP) were isolated from medium conditioned by the clonal rat osteosarcoma line UMR 106-01. Initial purification of both a 30-kDa inhibitor and a 20-kDa inhibitor was accomplished using heparin-Sepharose chromatography with dextran sulfate elution followed by DEAE-Sepharose and CM-Sepharose chromatography. Purification of the 20-kDa inhibitor to homogeneity was completed with reverse-phase high-performance liquid chromatography. The 20-kDa inhibitor was identified as rat TIMP-2. The 30-kDa inhibitor, although not purified to homogeneity, was identified as rat TIMP-1. Amino terminal amino acid sequence analysis of the 30-kDa inhibitor demonstrated 86% identity to human TIMP-1 for the first 22 amino acids while the sequence of the 20-kDa inhibitor was identical to that of human TIMP-2 for the first 22 residues. Treatment with peptide:N-glycosidase F indicated that the 30-kDa rat inhibitor is glycosylated while the 20-kDa inhibitor is apparently unglycosylated. Inhibition of both rat and human interstitial collagenase by rat TIMP-2 was stoichiometric, with a 1:1 molar ratio required for complete inhibition. Exposure of UMR 106-01 cells to 10(-7) M parathyroid hormone resulted in approximately a 40% increase in total inhibitor production over basal levels.     

E-cadherins identified in osteoblastic cells: Effects of parathyroid hormone and extracellular calcium on localization

The presence and regulation of cadherin localization in osteoblastic cells were examined. Monoclonal antibody (ECCD-1) that interferes with E-cadherin function prevented cell adhesion in UMR 106-H5 rat osteosarcoma cells and non-tumorigenic mouse calvarial MC3T3-E1 cells, whereas CCL39 fibroblast adhesion was not affected. Immunofluorescent antibodies (ECCD-2 and polyclonal L-CAM P1) revealed cadherins are localized along the osteoblastic cell-cell boundaries. Exposure of UMR 106-H5 cells to bovine parathyroid hormone (1–84) (PTH; 10 ng/ml × 1 hr) or low calcium medium (1.0 – 0.025 Mn) produced cellular retraction accompanied by intense immunofluorescence for cadherins throughout cells with a corresponding loss of punctate localization at remaining cell-cell adhesion points. Western immunoblot analysis indicated 108 kd and 115 kd cadherins are present, with a smaller 29.5 kd band that became predominantly associated with the cytosolic fraction of cells treated with parathyroid hormone or lowered calcium. The results demonstrate E-like cadherins are present in osteoblastic cells and implicate a regulatory role for parathyroid hormone and calcium in cadherin function and localization.     

Regulation of connective tissue collagenase production: stimulators from adult and fetal epidermal cells

We have examined the ability of primary adult rabbit skin cells to regulate collagenase production in vitro. Dermal cells constitutively produce collagenase in culture, and enzyme production by these cells can be influenced by epithelial cells. Co-culture with skin epidermal cells resulted in more enzyme production by dermal cells, whereas co- culture with corneal epithelial cells yielded less enzyme activity. Connective tissue cells from a different source, cornea, also produced collagenase when co-cultured with skin epidermal cells, although the stromal cells alone made no enzyme. The drug cytochalasin B had very little influence on collagenase production by dermal cells, either alone or in co-culture with epidermal cells, but did significantly potentiate enzyme production by corneal stromal cells responding to epidermal effector molecules. Epidermal-cell-conditioned medium from both fetal and adult rabbit skin was a potent source of stimulators (apparent mol wt 20,500 and 55,000) of connective-tissue-cell collagenase production. Stimulator production by epidermal cultures was cell density dependent. Optimal production of stimulators occurred in adult cultures containing 10(6) epidermal cells/ml of medium, and in fetal cultures containing 10(5) cells/ml. Inhibitors of connective tissue cell enzyme production were not detected in conditioned medium from either adult or fetal epidermal cells.     

Serotonin: an inducer of collagenase in myometrial smooth muscle cells.

Rat myometrial smooth muscle cells in culture actively produce collagenase in medium containing fetal bovine serum, but not in medium containing newborn bovine serum or containing fetal serum adsorbed with dextran-coated charcoal. A dialyzable molecule has been isolated from fetal bovine serum, which restores the ability of the smooth muscle cells to produce collagenase. The molecule has been purified and identified as serotonin (5-hydroxytryptamine). Cells cultured in medium depleted of serotonin for 3 days fail to produce collagenase, as assessed both enzymatically and immunologically. Addition of serotonin promptly restores the ability of the cells to produce the enzyme. The EC50 for serotonin is approximately 2 microM; maximum stimulation of collagenase production is observed at 5 microM. The response is specific for serotonin: a wide variety of compounds tested, either related to serotonin or of potential reproductive significance, were without effect in the induction of collagenase production by the cells. No changes in DNA content, general protein synthesis, or cellular collagen production were observed as a consequence of serotonin depletion or restoration, suggesting a selective effect of the compound on collagenase production. The effect of serotonin was also selective to myometrial smooth muscle cells; collagenase-producing fibroblasts from skin and cervix displayed no serotonin requirement for enzyme production. Studies using specific agonists or antagonists for a variety of serotonin receptor subtypes suggest that the 5-HT-2 receptor mediates the serotonin induction of collagenase in these cells. Preliminary evidence indicates that cultured human myometrial smooth muscle cells are also dependent upon serotonin for collagenase production. The evidence in this study suggests the possibility that serotonin serves as a signal to begin the massive collagen degradation that occurs in the postpartum uterus.     

Interactive regulation of signalling pathways in bone cells: possible modulation of PGE2-stimulated adenylyl cyclase activity by protein kinase C

We have reported previously that tumour-promoting phorbol esters modulate both basal and vasoactive intestinal polypeptide (VIP)-stimulated adenylyl cyclase activity in GH3 (an established pituitary cell line). Here, we probe the receptor and cell specificity of this response. Experiments were performed in the presence of isobutylmethylxanthine. Unlike the response in GH3 cells, the tumour-promoting phorbol ester (tetradecanoyl phorbol acetate (TPA] did not affect either basal adenylyl cyclase activity nor VIP-stimulated activity in the rat osteosarcoma subclones UMR 106-01 and UMR 106-06. In addition, the cyclase responses to parathyroid hormone (PTH), and, in the case of UMR 106-06, to calcitonin were unaffected by tumour-promoting phorbol ester. However, prostaglandin E2-stimulated cyclase activity in both of these subclones was attenuated in a dose-dependent manner.     

Modulation of L-type Ca2+ channels in UMR 106 cells by parathyroid hormone-related protein.

The effects of rat parathyroid hormone-related protein (rPTHrP) and bovine and rat parathyroid hormone (bPTH and rPTH) on L-type Ca2+ channels in UMR 106 cells were investigated using the patch clamp technique. rPTHrP increased the whole cell L-type Ca2+ channel currents and the increase was concentration dependent. rPTHrP, at a concentration of 62.5 nM, increased the L-type Ca2+ channel current by 122+/-25%. bPTH was less potent. A concentration of 7.5 microM bPTH increased the current by 99+/-24%. Results obtained with rPTH were similar to those obtained using bPTH. Single channel measurements, using the cell-attached version of the patch clamp technique, showed an increase in both the number of channel openings and the mean open time when the cells were exposed to rPTHrP. This suggested that rPTHrP affected the gating of L-type Ca2+ channels in UMR 106 cells. This study demonstrates that the actions of bPTH and rPTHrP in UMR cells are mediated in part by extracellular Ca2+ entry. PTHrP, a paracrine agent important in development, is more potent in regulating Ca2+ entry than PTH.     

Regulation of the collagenase-3 receptor and its role in intracellular ligand processing in rat osteoblastic cells

We have previously described a specific, saturable receptor for rat collagenase-3 in the rat osteosarcoma cell line, UMR 106-01. Binding of rat collagenase-3 to this receptor is coupled to the internalization and eventual degradation of the enzyme and correlates with observed extracellular levels of the enzyme. In this study we have shown that decreased binding, internalization, and degradation of 125I-rat collagenase-3 were observed in cells after 24 h of parathyroid hormone treatment; these activities returned to control values after 48 h and were increased substantially (twice control levels) after 96 h of treatment with the hormone. Subcellular fractionation studies to identify the route of uptake and degradation of collagenase-3 localized intracellular accumulation of 125I-rat collagenase-3 initially in Golgi-associated lysosomes and later in secondary lysosomes. Maximal lysosomal accumulation of the radiolabel and stimulation of general lysosomal activity occurred after 72 h of parathyroid hormone treatment. Preventing fusion of endosomes with lysosomes (by temperature shift, colchicine, or monensin) resulted in no internalized 125I-collagenase-3 in either lysosomal fraction. Treatment of UMR cells with the above agents or ammonium chloride decreased excretion of 125I-labeled degradation products of collagenase-3. These experiments demonstrated that degradation of collagenase-3 required receptor-mediated endocytosis and sequential processing by endosomes and lysosomes. Thus, parathyroid hormone regulates the expression and synthesis of collagenase-3 as well as the abundance and functioning of the collagenase-3 receptor and the intracellular degradation of its ligand. The coordinate changes in the secretion of collagenase-3 and expression of the receptor determine the net abundance of the enzyme in the extracellular space.     

Degrading activity for human parathyroid hormone [PTH-(1-84)] in rat osteoblast-like osteosarcoma cell line UMR106

The degrading activity for human parathyroid hormone [hPTH-(1-84)] was studied in a rat osteoblast-like osteosarcoma cell line UMR106. At 37 C,UMR106 cells degraded hPTH-(1-84) into fragments in a time-dependent manner, which was shown by a radioimmunoassay with the use of antibody recognizing the C-terminal and middle regions of PTH molecule, whereas the degradation was completely suppressed at 4 C and failed to occur in the absence of the cells. The Lineweaver-Burk plot of this degrading activity at 37 C showed a fairly good linearity and gave a Km value of 5.1 X 10(-7) M. Reverse-phase high-performance liquid chromatography (HPLC) analysis of immunoreactive PTH fragments in the medium disclosed two peaks aside from intact PTH, indicating a limited PTH-hydrolyzing activity of UMR106 cells cleaving the molecule between at least two separate positions. This study suggests the possible involvement of osteoblasts on the metabolism of intact PTH.     

Plasminogen activator regulation in osteoblasts: parathyroid hormone inhibition of type-1 plasminogen activator inhibitor and its mRNA.

In order to determine the mechanism by which parathyroid hormone (PTH) stimulates plasminogen activator (PA) activity in rat osteoblasts, we investigated the effect of human PTH(1-34) [hPTH(1-34)] on the synthesis of mRNAs for tissue-type PA (tPA), urokinase-type PA (uPA), and PA inhibitor-1 (PAI-1), and on release of PA activity and PAI-1 protein in both normal rat calvarial osteoblasts and UMR 106-01 osteogenic sarcoma cells. hPTH(1-34) (0.25-25 nM) decreased PAI-1 mRNA and protein, and increased PA activity in both cell types in a dose-dependent manner with ED50 of about 1 nM for both responses. Forskolin and isobutylmethylxanthine also stimulated PA activity and decreased PAI-1 protein and mRNA in both cell types. hPTH(1-34) did not show any consistent effect on tPA and uPA mRNA in calvarial osteoblasts, but a modest (two-fold) increase of both mRNAs was observed in UMR 106-01 cells treated with 25 nM hPTH(1-34). However, when protein synthesis was inhibited with 100 microM cycloheximide, the increase of tPA and uPA mRNA by hPTH(1-34) was enhanced in UMR 106-01 cells and became evident in calvarial osteoblasts. Fibrin autography also revealed that hPTH(1-34) increases tPA and uPA activity, especially after cycloheximide treatment in UMR 106-01 cells. These results strongly suggest that PTH increases PA activity predominantly by decreasing PAI-1 protein production through a cyclic adenosine monophosphate (cAMP)-dependent mechanism in rat osteoblasts. The reduction of PAI-1 protein by PTH results in enhanced action of both tPA and uPA, and would contribute to the specific roles of these PAs in bone.     

Atrial natriuretic factor receptors and stimulation of cyclic GMP formation in normal and malignant osteoblasts

Synthetic rat atrial natriuretic factor (Ile-ANF-26) stimulated cyclic GMP formation by up to several hundred-fold in osteoblast-rich cultures from newborn rat calvaria and in clonal osteogenic sarcoma cells (UMR 106-01) which are phenotypically osteoblast. ANF had no effect on the cyclic AMP response to parathyroid hormone in the same cells. Specific, high-affinity binding sites for ANF were identified in both cell types, with K_d and receptor numbers in normal osteoblasts of 1.2 ± 0.1 × 10⁻¹⁰ M and 42 ± 4 × 10³ per cell, and in UMR 106-01 cells of 1.4 ± 0.1 × 10⁻¹⁰M and 22 ± 4 × 10³per cell.     

Relaxin, oxytocin, and prostaglandin effects on progesterone secretion from bovine luteal cells during different stages of gestation

To determine the effects of relaxin, oxytocin, and prostaglandin F2 alpha on progesterone secretion, bovine luteal cells from different stages of gestation were dispersed in Medium 199 with 200 units/ml penicillin, 1.0% kanamycin, 0.5% bovine serum albumin, and 400 units/ml collagenase. Cells (10(5) were cultured in 400 microliters of Dulbecco's modified Eagle's medium and Ham's F-12 medium containing fetal bovine serum and antibiotics, in Falcon multiwell plates, in a humidified environment of 95% O2 and 5% CO2 at 37 degrees C. Cells were cultured for 24 hr without treatment and thereafter with medium-hormone replacement every 24 hr. Progesterone was quantified from unextracted media by radioimmunoassay. Basal progesterone secretion after 24 hr was 1.81 +/- 0.14, 1.76 +/- 0.17, 0.54 +/- 0.49, and 0.57 +/- 0.21 pg/ml per viable luteal cell from 145-, 165-, 185-, and 240-day-old corpora lutea, respectively. Basal progesterone secretion increased (P less than 0.05) with time in culture. Relaxin induced a dose-dependent (greater than 100 ng/ml) increase in progesterone release, compared with the controls. Oxytocin and prostaglandin F2 alpha induced greater release (P less than 0.05) of progesterone than relaxin at all stages of gestation, but progesterone release was dependent on the stage of gestation and the duration in culture. Luteinizing hormone (100 ng/ml) stimulated whereas 17 beta-estradiol (50 ng/ml) inhibited progesterone secretion by luteal cells at all stages of gestation examined. Relaxin obliterated the prostaglandin- and oxytocin-induced progesterone secretion by bovine luteal cells from 145 to 214 days of gestation. Thus, relaxin, cloprostenol, and oxytocin regulate progesterone production by cultured bovine luteal cells, but hormone secretion was dependent on the stage of gestation.     

Identification of immunoreactive sites in bovine parathyroid cells to antibodies raised against the NH2-terminal sequence of parathyroid hormone.

The NH2-terminal sequence of bovine parathyroid hormone (1-84) was localized with different immunocytochemical methods on the light and electron microscopic level in bovine parathyroid glands and in isolated bovine parathyroid parenchymal cells. The peroxidase labeled staphylococcal protein A and the peroxidase anti-peroxidase method were found to be advantageous for light and electron microscopic localization, respectively. Reaction product was found light microscopically in the cytoplasma of the parenchymal cells and electron microscopically largely over the secretion granules of the parenchymal cells. The immunoreactive sites were subsequently identified to represent only intact parathyroid hormone (1-84) by gel electrophoresis derived enzyme linked immunosorbent assay.     

The biological properties of bovine parathyroid hormone (1–41), a fragment generated from the native hormone by human leukocytes

Native bovine parathyroid hormone (bPTH) was found to be readily cleaved with human leukocyte elastase to yield the fragments bPTH(1–41) and bPTH(42–84). These were then isolated by reverse-phase HPLC and characterised by gas-phase sequencing and amino acid analysis. The biological activities of these fragments were assessed in an adenylate cyclase bioassay using the rat osteosarcoma cell line UMR106. bPTH(1–41) was found to have approximately twice the molar potency of the native hormone from which it was derived, bPTH(42–84) had no biological activity and did not modulate the adenylate cyclase response to these cells to the native hormone. The possible physiological significance of these observations is discussed.     

Osteoblast collagenase: collagenase synthesis by clonally derived mouse osteogenic (MC3T3-E1) cells

Latent collagenase was isolated by heparin-Sepharose affinity chromatography from the culture medium of clonally derived mouse osteogenic (MC3T3-E1) cells. Collagenase synthesis by MC3T3-E1 cells was significantly stimulated by the addition of parathyroid hormone to the serum-containing culture medium. The cellular origin of the isolated collagenase was confirmed by demonstrating the characteristic 3/4 and 1/4 fragments of collagen alpha-chain, as well as inhibition of the enzyme by anti-mouse bone collagenase antibody.     

1,25-Dihydroxyvitamin D3 stimulates the synthesis of matrix gamma-carboxyglutamic acid protein by osteosarcoma cells. Mutually exclusive expression of vitamin K-dependent bone proteins by clonal osteoblastic cell lines

Several clonal rat osteosarcoma cell lines were tested for the ability to express and secrete matrix Gla protein (MGP), a small vitamin K-dependent protein found in bone and cartilage. Two independently derived cell lines, UMR 106-01 and ROS 25/1, expressed MGP mRNA and secreted MGP antigen identical in size with that found in bone. No MGP message could be detected in ROS 17/2 and 2/3 cells, cell lines previously shown to synthesize the other known vitamin K-dependent bone protein, bone Gla protein (BGP), and no BGP mRNA could be detected in the cell lines which synthesize MGP. Since UMR 106-01 and ROS 17/2 are presently the best characterized clonal osteoblastic cell lines, the discovery of the mutually exclusive expression of MGP and BGP by these cell lines indicates that osteosarcoma cells can be fixed in different phenotypic states and that MGP and BGP should be useful markers for the analysis of phenotypic expression in bone. Treatment of UMR 106-01 cells with 1,25-dihydroxyvitamin D3 (1,25(OH)2D3) dramatically increased MGP mRNA within 4 h and, by 24 h, increased MGP secretion 15-fold. This is only the second example of a bone matrix protein whose synthesis is dramatically increased by vitamin D, the first being the 6-fold stimulation of BGP synthesis by 1,25(OH)2D3 in ROS 17/2 cells. The discovery that MGP and BGP are similarily regulated by 1,25(OH)2D3 was unexpected since the two proteins differ markedly in structure, physical properties, and tissue distribution. Since the synthesis of MGP is rapidly and dramatically increased by 1,25(OH)2D3, it is probable that MGP plays a role in the normal bone response to the hormone. MGP may also be the vitamin K-dependent protein whose abnormal synthesis in the Warfarin-treated animal modifies the bone response to 1,25(OH)2D3.     

Effect of cortisol or adrenocorticotropic hormone on luteinizing hormone secretion by pig pituitary cells in vitro

The effect of cortisol or adrenocorticotropic hormone (ACTH) on basal and gonadotropin-releasing hormone (GnRH)-induced secretion of luteinizing hormone (LH) was studied in vitro using dispersed pig pituitary cells. Pig pituitary cells were dispersed with collagenase and DNAase and then grown in McCoy's 5a medium containing 10% dextran charcoal-pretreated horse serum and 2.5% fetal calf serum for 3 days. Cells were preincubated with cortisol or ACTH before GnRH was added. When pituitary cells were incubated with 400 micrograms cortisol/ml medium for 6 h or longer, increase basal secretion of LH was observed. However, GnRH-induced LH release was reduced by cortisol. The degree of this reduction was dependent on cortisol, and a concentration of cortisol higher than 100 micrograms/ml was needed. Cortisol also inhibited the 17 beta-estradiol-induced increase in GnRH response. ACTH-(1-24), ACTH-(1-39), or porcine ACTH had no influence on GnRH-induced LH secretion. Our results show that cortisol can act directly on pig pituitary to inhibit both normal and estradiol-sensitized LH responsiveness to GnRH.     

Regulation of the phosphate (Pi) concentration in UMR 106 osteoblast-like cells: Effect of Pi,Na+ and K+

Osteoblast-like cells possess Na-dependent transporters which accumulate orthophosphate (Pi) from the extracellular medium. This may be important in bone formation. Here we describe parallel measurements of Pi uptake and cellular [Pi] in such cells from the rat (UMR 106–01 and UMR 106–06) and human (OB), and in non-osteoblastic human fibroblasts (Detroit 532 (DET)). In UMR 106–01, cellular [Pi] was weakly dependent on extracellular [Pi] and higher than expected from passive transport alone. [³²Pi]-uptake was inhibited by Na deprivation, but paradoxically increased on K deprivation. With Na, 87 per cent of cellular ³²P was found in organic phosphorus pools after only 5 min. Na deprivation also decreased cellular [Pi], in both UMR 106–01 and DET, but the decrease was smaller than that in [³²Pi]-uptake. Ouabain decreased [³²Pi]-uptake and cellular [Pi] in DET, but not in UMR 106–01. Regulation of cellular [Pi] is therefore at least partly dependent on Na/Pi co-transport, but this does not seem to be an exclusive property of osteoblasts.