Angiopeptin (BIM23014C) inhibits vascular smooth muscle cell migration in vitro through a G-protein-mediated pathway and is associated with inhibition of adenylyl cyclase and cyclic AMP accumulation.

Angiopeptin (AP: BIM23014C), a cyclic analogue of the peptide hormone somatostatin, inhibits intimal hyperplasia after balloon angioplasty. This inhibition has been attributed to a direct inhibitory effect on smooth muscle cell (SMC) proliferation. However, the SMC that proliferate in the intima and contribute to intimal hyperplasia arrive there by migrating from the injured media, suggesting that SMC migration may also play an important role in this process. Indeed, in the experiments we describe, AP inhibited the migration of rat aortic SMC cells (RA-SMC) in response to type I collagen, the predominant form of collagen in the vessel media, and did so dose dependently. RA-SMC migration was inhibited 70% in the presence of AP 100 nM. RA-SMC adhesion to type I collagen in these conditions was not inhibited, suggesting that AP does not interfere with RA-SMC recognition of type I collagen; instead, it blocks subsequent signaling events that are necessary for RA-SMC migration in response to type I collagen. AP inhibited the forskolin-stimulated accumulation of cyclic AMP by RA-SMC (35% at 30 nM). In addition, pertussis toxin (PT), which blocks Gi-mediated inhibition of adenylyl cyclase, blocked the inhibitory effect of AP on cyclic AMP (cAMP) accumulation and also blocked the inhibitory effect of AP on RA-SMC migration. These findings suggest that the inhibitory effect of AP on intimal hyperplasia is due at least in part to its effects on SMC migration and that these effects are mediated by a Gi-dependent pathway and may involve inhibition of adenylyl cyclase and cAMP accumulation.

Tumor cell adhesion to the extracellular matrix and signal transduction mechanisms implicated in tumor cell motility, invasion and metastasis.

The relationship between tumor cell adhesion to the extracellular matrix (ECM) and invasion and metastasis formation is one of the most intensively studied topics in cancer biology within the last 10-15 years. The aberrant molecular relationships between malignant tumor cells and their surrounding ECM have been implicated at virtually every stage of the metastatic process; ranging from steps that involve the local invasion of tumor cells away from the primary tumor to those that are involved in mediating extravasation through microvessel-associated basement membranes at the site(s) of metastasis formation. The complexity of tumor metastasis has required that a reductionist approach be taken in order to identify and relate specific molecular mechanisms involved in tumor cell adhesion to various aspects of tumor metastasis. The intensive research efforts into cell adhesion and tumor cell biology have generated many significant new concepts towards our understanding of the molecular aspects of tumor cell adhesion and metastasis. Our purpose in this article is to briefly summarize the relationship of ECM-stimulated tumor cell adhesion to the processes of tumor cell motility and invasion. This is followed by a discussion of certain aspects of signal transduction pathways that may impact on cell motility, with an emphasis on the relationship between phosphatidylinositol hydrolysis and actin polymerization, as well as certain GTP-binding protein-(G-protein) mediated events that could influence cytoskeletal organization and cell motility. Our emphasis is based on increasing evidence that implicates members of the signal transduction G-proteins in the motility and invasion of many normal and transformed cells.

Inositol and inositol 1,4,5-trisphosphate content of Down syndrome fibroblasts exhibiting enhanced inositol uptake.

Fibroblasts from individuals with Down syndrome (DS; trisomy 21) exhibit increased inositol uptake. Here we examine the relationship between this increase in uptake and mass levels of free inositol and inositol 1,4,5-trisphosphate (IP3) in DS fibroblasts. We report that human fibroblasts contain high levels of free inositol which are not significantly affected by the increase in inositol uptake associated with DS. In addition, increased uptake is accompanied by increased efflux of radiolabelled inositol from DS cells. Neither basal nor bradykinin-stimulated IP3 levels in DS cells differ significantly from normal values. This work highlights the usefulness of the DS cells in uncovering the role of transport across the plasma membrane in cellular inositol homeostasis.

High-affinity [3H]inositol uptake by dissociated brain cells and cultured fibroblasts from fetal mice.

The accumulation of [3H]inositol by mechanically dissociated brain cells and cultured skin fibroblasts from fetal mice was examined. Uptake by both tissues was strongly dependent on temperature and the presence of sodium ions. Brain and fibroblast uptake also responded similarly to inhibition by inositol isomers and phloridzin. At lower concentrations of inositol, both tissues exhibited high-affinity uptake kinetics with apparent Km values near 30 microM, similar to values observed previously in human fibroblasts and other cultured cells. The activity of brain high-affinity uptake was nearly an order of magnitude lower than that of fibroblasts, however, and was in part confounded by the presence of a low-affinity or simple diffusion system operating at inositol concentrations above 100 microM. Brain preparation from adult mice also showed evidence of high-affinity, Na+ dependent uptake, but its activity was significantly diminished relative to that of fetal brain preparations. Our results demonstrate that a high-affinity inositol transport system closely resembling that found in cultured cells is expressed in the developing mouse brain.

The role of G-protein in matrix-mediated motility of highly and poorly invasive melanoma cells.

Membranes from 2 K1735 murine melanoma clones of high invasive capacity show increased amounts of pertussis toxin (PT) substrate when compared to a weakly invasive cellular counterpart. Using a panel of specific G-protein antibodies, we identified Gi alpha 2 as the PT-sensitive G-protein uniquely abundant in highly invasive cells. In addition, RNA hybridization results confirm the immunoblot observations that Gi alpha 2 is present at higher levels in strongly invasive cells. This result suggests that the elevated expression of Gi alpha 2 in highly invasive cells is not entirely due to differences in either translational efficiency or protein degradation but is related to altered RNA transcriptional initiation, processing and/or degradation. ADP-ribosylation of Gi alpha-subunits by PT inhibited the fibronectin, laminin and collagen type-IV-stimulated motility of the 2 highly invasive clones, while PT treatment of cells from a poorly invasive clone resulted in little or no reduction of the fibronectin, laminin or collagen type-IV-stimulated lower motility. Furthermore, PT treatment of highly or poorly invasive K1735 clones does not result in any alteration in cellular cAMP accumulation, suggesting that the PT substrate is not linked with the adenylyl cyclase enzyme complex. The data suggest that a PT-sensitive G-protein, probably Gi alpha 2 regulates second messenger pathways that contribute to elevated motility in highly invasive K1735 cells.

Down's syndrome fibroblasts exhibit enhanced inositol uptake.

The inositol metabolism of Down's syndrome (DS, trisomy 21) skin fibroblasts was examined. We report that DS cells accumulated [3H]inositol 2-3-fold faster than did other aneuploid or diploid controls. In contrast, trisomy 21 did not affect the uptake of choline, serine or glucose. Kinetic analysis demonstrated an increased maximal velocity of high-affinity, Na(+)-dependent, inositol transport, consistent with the expression of higher numbers of transporters by DS cells. Enhanced uptake was accompanied by a proportional increase in the incorporation of radiolabelled inositol into phospholipid. We suggest that an imbalance of inositol metabolism may contribute to plasma membrane abnormalities characteristic of DS cells.

Melanotropin receptors of murine melanoma characterized in cultured cells and demonstrated in experimental tumors in situ.

Cultured melanoma cells are known targets for the pigment-inducing actions of melanotropins such as alpha-melanocyte-stimulating hormone (alpha-MSH). The objectives of the present studies were to determine the binding properties and functional relevance of MSH binding sites in a mouse melanoma cell line and to determine whether MSH receptors are expressed in situ. The binding properties of MSH receptors in intact cells of a highly metastatic, highly MSH-responsive mouse melanoma cell subline (B16-F10C23) were determined using a radiolabeled, biologically active preparation of the superpotent alpha-MSH analogue, [Nle4,D-Phe7]-alpha-MSH (125I-NDP-MSH). A single high-affinity class of binding site was detected (Kd for NDP-MSH, 5.6 x 10(-11) M; Kd for alpha-MSH, 2.6 x 10(-9) M as determined by Scatchard analysis and heterologous inhibition assays, respectively). alpha-MSH showed nearly identical concentration-response relationships in the radioreceptor assay (inhibition of 125I-NDP-MSH binding) and a bioassay (stimulation of intracellular cyclic AMP accumulation). Furthermore, the respective potencies of three melanotropins, NDP-MSH, alpha-MSH, and adrenocorticotropic hormone, in binding and biological assays were highly correlated. These results indicate that the 125I-NDP-MSH binding site represents the functional MSH receptor. Tumors were induced by inoculation of C57BL/6 mice with B16-F10C23 cells, and the presence of 125I-NDP-MSH binding sites was determined by in situ radiolabeling of frozen tissue sections followed by autoradiography. Specific MSH binding sites were distributed throughout the tumor tissue, but not in associated fibrovascular elements or in neighboring nonmelanoma tissues. As in cultured B16-F10C23 cells, melanotropins inhibited 125I-NDP-MSH binding to tissue sections in a concentration-dependent manner. These results support the hypothesis that functional MSH receptors are expressed in melanoma in situ, suggesting that the activities of melanoma cells in vivo may be subject to modulation by endogenous melanotropins. The methods described will be applicable for studies of the expression and regulation of MSH receptors in human melanoma and other target tissues.

G-protein involvement in matrix-mediated motility and invasion of high and low experimental metastatic B16 melanoma clones.

Membranes from a B16 murine melanoma clone of high experimental metastatic capacity show increased amounts of pertussis toxin (PT) substrate when compared to a low metastatic counterpart. Using specific antibodies, we identified Gi2 as the PT-sensitive G-protein uniquely abundant in highly metastatic cells. ADP ribosylation of a G-protein alpha subunit by PT decreased both the migration of tumor cells through Matrigel (Collaborative Research, Bedford, MA) and the fibronectin-, laminin-, and collagen type IV-mediated motility of a high metastatic clone. Treatment of cells from a low metastatic clone with PT did not alter either the relatively low invasive capacity or lower motility of these cells. While cholera toxin treatment of cells resulted in decreased invasion and motility of both high and low metastatic clones, there were significant qualitative and quantitative differences, when compared to the PT effects, which indicated that the two toxins were acting on different second messenger systems. PT treatment of B16 clones of high or low experimental metastatic capacity does not result in any alteration in cellular cyclic AMP accumulation suggesting that the PT substrate is not linked with the adenylyl cyclase enzyme complex. The data suggest that a PT-sensitive G-protein, possibly Gi2, regulates second messenger pathways that contribute to the metastatic capacity of B16 melanoma cells.

Growth rate dependence of differential incorporation of a guanosine triphosphate photoaffinity probe into the alpha subunit of a guanine nucleotide binding protein, Gs, from metastatic variants of B16 melanoma cells.

The signal transducing regulatory protein (Gs alpha) was examined in B16 melanoma clones of low (F1C29) and high (F10C23) experimental metastatic potential. Incorporation of the photoaffinity analogue, [8-azido-gamma-32P]GTP, into Gs alpha was decreased in F10C23 extracts when compared to F1C29. This difference disappeared when the photolabeling reaction was carried out at an elevated temperature which enhanced the rate of GTP exchange, suggesting functional differences in the ability of Gs alpha to bind or release GTP rather than dissimilar intracellular Gs alpha concentrations. Differential Gs alpha photolabeling occurred only during the period of rapid growth when F10C23 cells proliferated faster than F1C29 cells. During the recovery phase of growth immediately following plating and at confluence, periods in which F1C29 and F10C23 growth rates are similar, Gs alpha photolabeling between the two clones was equal. CMT lung carcinoma clones of differential metastatic potential grew at a uniform rate at all stages of growth and also exhibited equal Gs alpha photolabeling. F10C23 cells were more responsive to alpha-melanocyte-stimulating hormone stimulation of adenylate cyclase activity than F1C29 cells at all growth stages. These results confirm previously observed functional differences in Gs alpha between B16 metastatic variants and show that photolabeling differences in Gs alpha are related to growth rate.

Reconstitution of the Gs protein from B16 melanoma clones of high and low experimental metastatic potential into S49 cyc-membranes.

The ability of a series of B16 melanoma clones to form experimental lung metastases in syngeneic mice has been shown to correlate positively with adenylate cyclase activity. (Sheppard et al, Int. J. Cancer 37 (1986) 713-722). To begin to identify the components of the adenylate cyclase complex that account for enhanced enzyme activity in highly metastatic tumor populations, cholate extracts containing the GTP-binding protein GS from B16 melanoma clones of different metastatic capacities were reconstituted with membranes prepared from S49 cyc-, a variant lymphoma cell line that lacks GS function. The results revealed that extracts from a highly metastatic B16 clone (F10-C23) reconstituted significantly greater adenylate cyclase activities in S49 cyc- membranes than parallel preparations from a B16 clone (F1-C29) of low metastatic capacity. The data suggest that aberrations in GS function may contribute to the heightened responsiveness of adenylate cyclase observed in B16 melanoma clones of increased metastatic potential.

cAMP metabolism in B16 melanoma clones during the formation of experimental and spontaneous metastases.

The ability of B16 melanoma clones to form tumor colonies in the lung after i.v. injection (experimental metastases) correlates positively with their capacity to respond to activators of cyclic adenosine 3-, 5-monophosphate (cAMP) metabolism (such as melanocyte-stimulating hormone and forskolin). To investigate whether this relationship is causal, the cAMP responses of 4 B16 melanoma clones of differing colonizing potential have been examined in freshly established stock cell cultures, in pulmonary colonies in vivo and in cultures established from these lesions. In all cases, the cAMP responsiveness of the excised colonies (and cultures derived from them) mimicked the responsiveness of the cultures from which they arose. B16 clones exhibiting low cAMP responsiveness gave rise to few experimental metastases all of which were poorly responsive to activators of cAMP metabolism. Similarly, clones with high cAMP responsiveness formed multiple lung colonies which displayed a marked sensitivity to agents that stimulated cAMP production. Parallel experiments on the spontaneous metastatic behavior of the same clones revealed that the cAMP responsiveness of cells in the primary (intrafootpad) tumor and spontaneous metastatic lesions in the lung faithfully reflected the response profile of the original tumor-cell inoculum but no correlation was found between cAMP responsiveness and the capacity to form spontaneous metastases. These data suggest that cAMP-dependent events may influence the survival, arrest and organ colony formation by cells injected directly into the circulation but appear to be of little or no importance in determining the early event(s) involved in the evolution of spontaneous metastases prior to the entry of cells into the circulation.

Desensitization of LLC-PK1 cells by vasopressin results in receptor down-regulation.

The molecular mechanism of desensitization of antidiuretic hormone receptors is not well understood. Preincubation of LLC-PK1 cells with lysine vasopressin (LVP) (10(-6) M, 5 h) decreased subsequent LVP-stimulated cAMP accumulation in cells by 83% and reduced the Vmax of LVP-stimulated adenylate cyclase by 81%. Such preincubation also reduced by 90% the binding of [3H]LVP to both intact cells and isolated plasma membranes, suggesting a loss of vasopressin receptors. Both the reduction in cAMP response and the apparent loss of receptors showed similar dose and time dependence. Monensin (33 microM) did not alter [3H]LVP binding or stimulation of cAMP by LVP, nor did it prevent desensitization. However, membranes prepared from cells preincubated with LVP in the presence of monensin did not show a decrease in [3H]LVP binding. Forskolin preincubation, at 0.1, 1, 10 and 100 microM, did not alter [3H]LVP binding or accumulation of cellular cAMP by LVP, nor did it induce desensitization to LVP. Cells desensitized with varying LVP concentrations in the presence of 10 microM forskolin displayed the same loss of [3H]LVP binding and LVP responsiveness as observed in the absence of forskolin. LVP-desensitized cells, upon removal from LVP-containing medium, recovered cAMP responsiveness to LVP and specific binding of [3H]LVP at the same rate, achieving control levels after 50 h. Recovery was prevented by cycloheximide (25 micrograms/ml). These findings are consistent with a desensitization process involving LVP-mediated receptor internalization, and a recovery process requiring protein synthesis.

Differential solubilization of gamma-aminobutyric acid receptive sites from membranes of mammalian brain.

Sodium-dependent (+Na) and sodium-independent (-Na) receptive sites for gamma-aminobutyric acid (GABA) residing in or on frozen synaptic plasma membranes (SPM) of bovine cerebral cortex were characterized as to binding constants, pharmacologic specificities, and sodium dependence. The SPM fraction was then treated with various concentrations of Triton X-100 resulting in the loss of pharmacologic specificity, binding characteristics, and sodium dependence associated with +Na GABA receptive sites in SPM. The resulting junctional complex preparation (JC), i.e., a fraction enriched in junctional complexes, possessed only the pharmacologic specificity and binding constants associated with -Na receptive sites whether assayed in the presence or absence of 100 mM-NaCl. This is probably due to the detergent dispersal or solubilization of the +Na GABA receptive site. The binding constants, KD and Bmax, for -Na GABA binding in SPM were 170 nM and 4.4 pmol/mg protein, while in JC they were 186 nM and 3.7 pmol/mg protein. Under repeated washing the KD was reduced to 60 +/- 6.9 nM and the Bmax was reduced to 2.5 +/- 0.5 pmol/mg protein in JC, probably owing to the removal of endogenous ligand or inhibitor, and not to inhibition by residual Triton X-100. Multiple extraction with 0.1% or 0.5% Triton X-100 did not alter the KD or Bmax values for the binding of [3H]GABA to JC. Sodium-independent GABA binding was lost from JC membranes with the use of sodium deoxycholate, probably through solubilization.

Kinetic and pharmacologic characterization of gamma-aminobutyric acid receptive sites from mammalian brain.

Sodium-dependent (+Na) and sodium-independent (-Na) receptive sites for gamma-aminobutyric acid (GABA) have been characterized using synaptic plasma membranes from bovine and rat brain. Synaptic plasma membranes were prepared from either rat cerebellar cortex or calf cerebral cortex by discontinuous sucrose gradient flotation centrifugation of crude mitochondrial pellets, and assayed using equilibrium ligand binding assays to obtain the maximum binding capacity (Bmax) and the thermodynamic constant (KD). Values for KD from equilibrium studies were subsequently confirmed by kinetic analyses of association and dissociation reactions. The KD for +Na GABA binding (5.0 +/- 0.2 micron) corresponds to the apparent Michaelis constant for neuronal GABA transport (3.8 +/- 0.1 micron)22, while the KD for -Na binding (0.17 +/- 0.04 micron) agrees with that determined by Enna and Snyder for the putative postsynaptic receptor. Maximal binding activities of about 5 and 55 pmole/mg protein were obtained for -Na and +Na binding respectively. The pharmacologic specificities of the two sites were determined using competition binding studies. Nipecotic acid and diaminobutyric acid inhibit both synaptosomal GABA uptake (Ki approximately 25 micron and 120 micron respectively) and +Na binding of GABA to synaptic plasma membrane (IC50 approximately 40 micron and 350 micron respectively) but do not inhibit -Na binding. Bicuculline inhibits -Na [3H]GABA binding at low concentrations (IC50 approximately 15 micron), while affecting the uptake and +Na binding of [3H]GABA only at high concentrations (IC50 approximately 520 micron and 300 micron respectively). beta-Alanine inhibits the -Na binding site (IC50 approximately 100 micron), but is ineffective at the +Na binding site and does not interfere with synaptosomal uptake of GABA. Finally, chlorpromazine and N-ethylmaleimide inhibit the +Na binding, albeit at high concentrations (IC50 approximately 600 micron and 5 mM respectively) but are ineffective at the -Na binding site. From these results the -Na binding site is tentatively identified as a postsynaptic receptor and the +Na binding site is identified as the neuronal uptake receptive site.

Purification of Leucine tRNA Isoaccepting Species from Soybean Cotyledons: I. Benzoylated Diethylamino Cellulose Fractionation, N-Hydroxysuccinimide Modification, and Characterization of Product.

Transfer RNA from soybean (Glycine max) cotyledons was purified to homogeneity followed by the purification of the family of leucine tRNA via benzoylated diethylaminoethyl cellulose (BDC) chromatography. Nonacylated total purified tRNA was salicylhydroxamate (SHAM) modified by the phenoxyacetyl method and fractionated into three peaks on a BDC column. The first peak containing bulk tRNA with no hydrophobic character amounted to 78% of the added tRNA. The second peak containing 19% of the added tRNA and represents the tRNA with intrinsic hydrophobic properties. The third peak containing 3% of the tRNA represents the SHAM modified tRNA and nonspecifically modified tRNA. Transfer RNA peaks I and II were pooled and subsequently stoichiometrically acylated in two batches, one containing [(14)C]leucine while the other contained unlabeled leucine. The acylated tRNA was loaded on and step-eluted from a BDC column. The purified acylated-tRNA was phenoxyacetyl modified and following ethanol precipitation was fractionated on a BDC column. A double peak eluted from the column in the ethanol gradient contained 5.3% of the starting optical density and 85.3% of the starting counts per minute. Characterization of this leucine tRNA showed typical ultraviolet spectra properties and appeared to be homogeneous on a G-100 Sephadex column. The minimum purity of the tRNA was 32 to 35%. Finally, the acylated tRNA was chromatographed on an RPC-2 column giving six leucine isoaccepting tRNAs. The data indicate that leucine tRNA was highly purified without losing the integrity of the family of isoacceptors.

Purification of Leucine tRNA Isoaccepting Species from Soybean Cotyledons: II. RPC-2 Purification, Ribosome Binding, and Cytokinin Content.

Two of the six leucine isoaccepting tRNA species from soybean (Glycine max) cotyledons recognize U-beginning codons, and contain cytokinin moieties in their structure. These same two isoaccepting species have been shown to undergo quantitative changes in their relative amounts upon treatment with N(6)-benzyladenine in vivo. In addition a procedure has been developed for purification of the isoaccepting species of leucine tRNA from soybean cotyledons resulting in isoacceptors of minimum purity, calculated by amino acid acceptance capacity, of from 46 to 78% leucine tRNA.

Pentobarbital and synaptic high-affinity receptive sites for gamma-aminobutyric acid.

Electrophysiological investigations of others show that pentobarbital enhances the inhibitory inflences of gamma-aminobutyric acid (GABA). Specifically, receptor activation is amplified and prolonged, suggesting the presence of an increased number of GABA molecules in the synaptic cleft. Either inactivation of high-affinity GABA transport or alteration of post-synaptic GABA receptors might account for these influences of pentobarbital. In this sudy the effect of pentobarbital on high-affinity uptake and binding of GABA to synaptic receptive sites has been examined. Using synaptosomes and subsynaptosomal fractions of cerebral cortex and hippocampus, it si shown that concentrations of pentobarbital, exceeding 1 mM have no appreciable effect on GABA uptake or binding. Thus the synaptic influence of pentobarbital, evident at 0.1 mM in electrophysiologic experiments, must originate from mechanisms other than the high-affinity uptake or binding of GABA. Possible sites of action include the presynaptic release of GABA and the ionophores coupled with postsynapitc sites.