Spontaneous and agonist-induced 86Rb efflux from rabbit aortic smooth muscle cells in culture: a comparison with fresh tissue

The role of heparin in angiogenesis

A series of experiments is described in which: (i) mast cells were found to accumulate at a tumour site before the ingrowth of new capillaries; (ii) heparin released by mast cells increased the migration of capillary endothelial cells in vitro; and (iii) heparin enhanced tumour angiogenesis in vivo. These experiments led to the discovery that protamine and platelet factor 4 are angiogenesis inhibitors. This finding suggests a central role for heparin or related glycosaminoglycans in the growth regulation of capillary blood vessels.

Inhibition of neovascularization by a cartilage factor

Neovascularization of developing, repairing or neoplastic tissues is regulated, at least partially, by a family of proteins of low molecular mass (1000 less than mol mass less than 50 000 Da) which can be extracted from avascular tissues, such as hyaline cartilage, aorta or bladder epithelium, by mild salt solutions. These extractable proteins, functionally defined as anti-invasion factor (AIF), act as local regulators for some of the major mechanistic pathways by which endothelial cells are thought to invade tissues during neovascularization, mainly by matrix-degrading enzymes and by increased rates of migration and proliferation. AIF contains a spectrum of proteinase (collagenase) inhibitory activities, as well as an endothelial cell growth inhibitor. The endothelial cell growth inhibitor is directed against actively dividing endothelial cells in culture but has no effect on endothelial cell monolayers or any other cell lines tested. In tumours, the AIF-derived endothelial cell growth inhibitor may limit tumour growth to less than 2 mm in diameter by inhibiting tumour neovascularization.

Contraction of vascular smooth muscle in cell culture

The use of cultured vascular smooth muscle cells for the study of events related to excitation and contraction of smooth muscle has been limited by the inability to reliably induce contractile responses after subculturing of the cells. This limitation has been overcome by the cell culture preparation described herein. We demonstrate that appropriate responses to both smooth muscle agonists and vasodilators were preserved in cells that were serially subcultured. Fetal bovine pulmonary artery and aortic cell cultures were established following enzymatic dispersion of the medial portion of freshly harvested vessels. At various times after isolation, cells were transferred to microscope coverslips coated with a polymerized silicone preparation (polydimethyl siloxane). Tension forces generated by the cells were manifested as wrinkles and distortions of this flexible growth surface. Visual evidence of cell contraction in the form of increased wrinkling was documented for cells exposed to angiotensin II, carbachol, and KCl. Decreases in cell tension occurred following treatment with isoproterenol, and those relaxing effects were overcome by subsequent treatment with the agonist carbachol. The contractile responses did not diminish with prolonged maintenance in culture or repeated subculturing. Phosphorylation of the light chains on the contractile protein myosin was also measured as a biochemical index of agonist-induced contraction. Cells depolarized with KCl or exposed to carbachol showed increased myosin phosphorylation when analyzed by 2-dimensional gel electrophoresis. The responses remained intact through 7 passages and 9 weeks in culture. These results show that cultured vascular smooth muscle cells do not necessarily undergo a phenotypic modulation with loss of contractility under prolonged maintenance in culture.

Effect of acute serum depletion on Na+-K+ homeostasis in cultured human skin fibroblasts

In order to elucidate changes in cell transport behavior of cultured human skin fibroblasts in response to acute serum depletion, we performed uptake and washout of 22Na+ and 86Rb+ as well as measurements of the intracellular Na+ and K+ levels in the presence and absence of ouabain. Pronounced and lasting increase in cellular Na+ and decrease in K+ were observed after removal of fetal bovine serum (FBS) from the medium. The sum of the Na+ and K+ contents (nEq/10(5) cells) was lower in FBS-free medium (mean +/- SD; 17.3 +/- 2.2) than in FBS-containing medium (26.2 +/- 3.8; P less than .02). Simultaneously, a decrease in cellular water volume was detected in the FBS-free medium. The cation uptake and washout data suggest that FBS removal primarily renders the cells more permeable to Na+ and K+ with a secondary stimulation of the ouabain-sensitive Na+ extrusion mechanism. FBS at a concentration of 0.2% prevented approximately 50% of the maximal increase in the 86Rb+ washout rate constant associated with FBS depletion. Ouabain (2 microM) produced an increase in the 86Rb+ washout rate constant. This effect was substantially larger in cells subjected to medium without FBS (from 0.0303 to 0.2500 min-1) than in fibroblasts incubated in medium with FBS (from 0.0107 to 0.0487 min-1). The cellular K+ content was drastically reduced by ouabain to a level not different in medium with or without FBS (33.9 +/- 4.5 to 1.75 +/- 0.38 and 16.7 +/- 1.4 to 1.4 +/- 0.13 nEq/10(5) cells, respectively). The 22Na+ washout data exhibited a three-exponential pattern. Analytical solutions of the washout data by means of two models (serial and parallel) with three compartments showed that FBS depletion resulted in increase of the size of all three compartments. It is concluded that in cultured human skin fibroblasts, FBS is essential to the maintenance of a normal Na+ and K+ homeostasis. The removal of FBS results in dramatic permutation of this homeostasis that develops within minutes and lasts for hours.

Sodium 22+ washout from cultured rat cells

The washout of Na+ isotopes from tissues and cells is quite complex and not well defined. To further gain insight into this process, we have studied 22Na+ washout from cultured Wistar rat skin fibroblasts and vascular smooth muscle cells (VSMCs). In these preparations, 22Na+ washout is described by a general three-exponential function. The exponential factor of the fastest component (k1) and the initial exchange rate constant (kie) of cultured fibroblasts decrease in magnitude in response to incubation in K+-deficient medium or in the presence of ouabain and increase in magnitude when the cells are incubated in a Ca++-deficient medium. As the magnitude of the kie declines (in the presence of ouabain) to the level of the exponential factor of the middle component (k2), 22Na+ washout is adequately described by a two-exponential function. When the kie is further diminished (in the presence of both ouabain and phloretin) to the range of the exponential factor of the slowest component (k3), the washout of 22Na+ is apparently monoexponential. Calculations of the cellular Na+ concentrations, based on the 22Na+ activity in the cells at the initiation of the washout experiments, and the medium specific activity agree with atomic absorption spectrometry measurements of the cellular concentration of this ion. Thus, all three components of 22Na+ washout from cultured rat cells are of cellular origin. Using the exponential parameters, compartmental analyses of two models (in parallel and in series) with three cellular Na+ pools were performed. The results indicate that, independent of the model chosen, the relative size of the largest Na+ pool is 92-93% in fibroblasts and approximately 96% in VSMCs. This pool is most likely to represent the cytosol.

Expression of smooth muscle-specific alpha-isoactin in cultured vascular smooth muscle cells: relationship between growth and cytodifferentiation

The relationship between growth and cytodifferentiation was studied in cultured rat aortic smooth muscle cells (SMCs) using expression of the smooth muscle (SM)-specific isoactins (Vanderkerckhove, J., and K. Weber, 1979, Differentiation, 14:123-133) as a marker for differentiation in these cells. Isoactin expression was evaluated by: (a) measurements of fractional isoactin content and synthesis ([35S]methionine incorporation) by densitometric evaluation of two-dimensional isoelectric focusing sodium dodecyl sulfate gels, and (b) immunocytological examination using SM-specific isoactin antibodies. Results showed the following: (a) Loss of alpha-SM isoactin was not a prerequisite for initiation of cellular proliferation in primary cultures of rat aortic SMCs. (b) alpha-SM isoactin synthesis and content were low in subconfluent log phase growth cells but increased nearly threefold in density-arrested postconfluent cells. Conversely, beta-nonmuscle actin synthesis and content were higher in rapidly dividing subconfluent cultures than in quiescent postconfluent cultures. These changes were observed in primary and subpassaged cultures. (c) alpha-SM actin synthesis was increased by growth arrest of sparse cultures in serum-free medium (SFM; Libby, P., and K. V. O'Brien, 1983, J. Cell. Physiol., 115:217-223) but reached levels equivalent to density-arrested cells only after extended periods in SFM (i.e., greater than 5 d). (d) SFM did not further augment alpha-SM actin synthesis in postconfluent SMC cultures. (e) Serum stimulation of cells that had been growth-arrested in SFM resulted in a dramatic decrease in alpha-SM actin synthesis that preceded the onset of cellular proliferation. These findings demonstrate that cultured vascular SMCs undergo differential expression of isoactins in relation to their growth state and indicate that growth arrest promotes cytodifferentiation in these cells.

Investigation of the preferred Mg(II)-adenine-nucleotide complex at the active site of ectonucleotidases in intact vascular cells using phosphorothioate analogues of ADP and ATP

In the presence of Mg2+ the ecto-(nucleoside diphosphatase) on intact vascular endothelial or smooth muscle cells in culture selectively catabolizes the PS diastereoisomer of adenosine 5'-[alpha-thio]diphosphate, (PS)-ADP [alpha S], and the ecto-(nucleoside triphosphatase) selectively catabolizes the PS isomer of adenosine 5'-[beta-thio]triphosphate, (PR)-ATP[beta S], but exhibits no selectivity towards ATP[alpha S] isomers. In the presence of Cd2+ selectivity to ADP[alpha S] and to ATP[beta S] isomers is reversed; in the presence of Co2+, selectivity is lost. We conclude that each enzyme preferentially recognises the lambda (screw-sense) bidentate Mg(II)-nucleotide complex at its active site.

Characterization of ectonucleotidases on vascular smooth-muscle cells

We compared the properties of the ectonucleotidases (nucleoside triphosphatase, EC 3.6.1.15; nucleoside diphosphatase, EC 3.6.1.6; 5'-nucleotidase, EC 3.1.3.5) in intact pig aortic smooth-muscle cells in culture with the properties that we previously investigated for ectonucleotidases of aortic endothelial cells [Cusack, Pearson & Gordon (1983) Biochem. J. 214, 975-981]. In experiments with nucleotide phosphorothioate diastereoisomers, stereoselective catabolism of adenosine 5'-[beta-thio]triphosphate, but not of adenosine 5'-[alpha-thio]triphosphate, by the triphosphatase and stereoselective catabolism of adenosine 5'-[alpha-thio]diphosphate by the diphosphatase were found, as occurs in endothelial cells. In contrast with endothelial ecto-5'-nucleotidase, the smooth-muscle-cell enzyme catabolized adenosine 5'-monophosphorothioate (AMPS) to adenosine: the affinity of the enzyme for AMPS was greater than for AMP, and Vmax for AMPS was about one-sixth that for AMP. In both cell types AMPS was an apparently competitive inhibitor of AMP catabolism by 5'-nucleotidase. The relative rates of catabolism of nucleotide enantiomers in which the natural D-ribofuranosyl moiety is replaced by an L-ribofuranosyl moiety were similar to those in endothelial cells. No ectopyrophosphatase activity was detected in smooth-muscle cells, in contrast with endothelial cells, where modest activity is present.

Cultured rat vascular smooth muscle cells: extracellular calcium and Na+-K+ regulation

This study explores the relationship between extracellular calcium (Cao) and Na+-K+ regulation as it particularly pertains to the activity of the Na+ pump in cultured vascular smooth muscle cells (VSMCs) originating from Sprague-Dawley rats. As compared with cells incubated in media containing 0.5, 2.0, or 4.0 mM calcium, when the Na pump is active, VSMCs incubated in a Ca-deficient medium show a marked increase in intracellular sodium and no significant change in intracellular potassium. Associated with the rise in intracellular sodium there is an augmented activity of the Na pump. When the Na pump is inhibited, VSMCs incubated in either high-Ca medium (Cao = 4.0 mM) or Ca-deficient medium manifest a greater decline in intracellular potassium than cells incubated in media containing 0.5 or 2.0 mM calcium. Furthermore, when the Na pump is inhibited, VSMCs incubated in a Ca-deficient medium exhibit higher intracellular sodium levels in comparison with their counterparts incubated in media containing calcium. Flux experiments indicate that the aforementioned changes reflect increased membrane permeabilities to Na+ and K+. It is concluded that by regulating the permeability of the VSMC membrane, Cao plays an important role in the intracellular Na+-K+ homeostasis and that its effect on the Na pump is mediated via perturbations in the intracellular Na+ and K+ concentrations.

Effects of alterations in cell phenotype and hypokalemia on sodium-potassium pump activity in rabbit vascular smooth muscle

We investigated in cell culture, how alterations in phenotype accompanying proliferation of rabbit aortic smooth muscle and chronic hypokalemia could affect the Na,K pump. Total rubidium-86 uptake as well as ouabain and frusemide-sensitive uptake into cells was measured in physiological salts solution (PSS), PSS containing 5% foetal calf serum and PSS containing foetal calf serum plus 15 microM monensin. In physiological salts solution 90% of the rubidium-86 uptake into contractile or synthetic state cells was frusemide-sensitive and less than 8% ouabain-sensitive. Total and frusemide-sensitive rubidium-86 uptakes, measured in PSS or PSS containing foetal calf serum were similar in cells cultured and maintained in the contractile phenotype, cells in the synthetic phenotype and those which had recently reverted from the synthetic to contractile phenotype. When cells were sodium loaded in the presence of monensin and foetal calf serum, ouabain-sensitive rubidium-86 uptake was 50% higher in cells which were maintained in culture in the contractile phenotype. Frusemide-sensitive rubidium-86 uptake was similar in each cell phenotype. To examine how cell culture in hypokalemic media would affect the Na,K pump, we determined ouabain-sensitive rubidium-86 uptake in the presence of monensin plus foetal calf serum in cells incubated for 24 hours in low and normal potassium containing culture media. Ouabain-sensitive uptake was 20% higher in cells cultured in a 0.76 mM potassium medium than in those cultured in 5.4 mM potassium medium. Frusemide-sensitive rubidium-86 uptake was unaffected. These results demonstrate that 'maximal' Na,K pump activity in sodium-loaded cells is reduced when cells change from the contractile to synthetic phenotype. This reduction appears only very slowly reversible when cells revert from the synthetic to contractile phenotype. Prolonged hypokalemia increases 'maximal' activity of the Na,K pump.

Age-dependent decrease of the passive Rb+ and K+ permeability of the nerve cell membranes in rat brain cortex as revealed by in vivo measurement of the Rb+ discrimination ratio

Young, adult and old male CFY rats (2, 12 and 24 mth of age, respectively) were treated with a daily dose of 30 mg RbCl/100 g body weight, in form of aqueous solution injected intraperitoneally for 14 days. A considerable part of the intracellular K+-content of the body was replaced by Rb+ during this treatment. After cessation of the RbCl injections, a relative steady state came into being in each age group, called Rb+-release period. During this period Rb+ and K+ contents of the blood serum and the cisternal CSF were measured by atomic absorption spectrophotometry, and of the intracellular space of brain cortical cells by energy-dispersive X-ray microanalysis. Ultrastructural features of the brain cortex were also checked by transmission electron microscopy. For X-ray microanalysis, the L-line of Rb at 1.694 keV energy was used at 10 kV accelerating voltage in a scanning electron microscope equipped with an EDAX System F. Rb+ and K+ concentrations were obtained for the cellular dry mass and converted into wet concentrations on the basis of intracellular water contents known from former experiments. Rb+-replacement of K+ did not cause any ultrastructural alteration in the brain cortex. However, the Rb+ accumulation displayed a very significant age-dependent increase: at the beginning of release, adult and old rats had 32.6 and 44.7 mM Rb+ in their intracellular water as against the 8.6 mM found in the young group, and similar proportional difference persisted during 20 days of the release. Rb+ discrimination ratios (DR) calculated either for the blood or the CSF displayed very considerable age-dependent increase: the values of the adult and old groups were 191 and 242% of the young one, indicating that the passive Rb+ (and K+) permeability of the nerve cell membrane decreases throughout the life span of rats. These results give further support to the membrane hypothesis of aging.

Loss of receptor-mediated 86Rb efflux from pig aortic endothelial cells in culture

The responsiveness of freshly-isolated and subcultured pig aortic endothelial cells to adenosine triphosphate (ATP), bradykinin and ionophore A23187 was compared by monitoring agonist-induced 86Rb efflux. ATP, bradykinin and ionophore A23187 stimulated 86Rb efflux from freshly-isolated cells. ATP and bradykinin, which act via specific receptors, were less effective at inducing 86Rb efflux from subcultured cells but ionophore A23187 was as effective on subcultured as on freshly-isolated cells. These results suggest that pig aortic endothelial cells, when subcultured, lose receptors for ATP and bradykinin and/or develop an abnormality in the coupling-mechanism between receptor-occupation and calcium-mobilization.

Stimulation of endothelial prostacyclin production plays no role in endothelium-dependent relaxation of the pig aorta

Stimulation of prostacyclin production by pig aortic endothelial cells adhering to microcarrier beads superfused in columns, or 3H release from cells prelabelled with [3H]-arachidonate, was studied in response to a range of agents that induce endothelium-dependent vascular relaxation. Bradykinin, adenosine triphosphate (ATP) and ionophore A23187 each stimulated release of prostacyclin from unlabelled cells and of 3H from prelabelled cells but acetylcholine did not. Bradykinin induced a parallel, dose-dependent increase in 3H release and 86Rb efflux, measured simultaneously from columns of aortic endothelial cells preloaded with 86Rb and [3H]-arachidonate. The rank-order of effectiveness at inducing both 3H and 86Rb release, measured simultaneously from columns of aortic endothelial cells prelabelled with 86Rb and [3H]-arachidonate and challenged with maximal doses of each agonist, was: A23187 greater than bradykinin greater than ATP. The similarity between agonist-induced 3H release (from cells prelabelled with [3H]-arachidonate) and 86Rb efflux indicates that a common mechanism may be responsible, and the effectiveness of ionophore A23187 suggests that a rise in the intracellular level of calcium may be involved. The lack of effect of acetylcholine on release of prostacyclin from unlabelled cells or of 3H from cells prelabelled with [3H]-arachidonate provides further evidence that acetylcholine acts on endothelial cells by a mechanism that does not involve calcium mobilisation. Although bradykinin, ATP and ionophore A23187 each induced release of prostacyclin from aortic endothelial cells, prostacyclin did not relax the pig aorta. Furthermore, endothelium-dependent relaxation was unaffected by pretreating aortic strips with aspirin. It therefore appears that neither prostacyclin nor any other cyclo-oxygenase product mediates endothelium-dependent relaxation of the pig aorta.