Andrew C. Newby

Serum-induced proliferation of rabbit aortic smooth muscle cells from the contractile state is inhibited by 8-Br-CAMP but not 8-Br-cGMP

A method is described for the quantification of vascular smooth muscle cell growth from individual explants of contractile rabbit aortic tunica media. The precision of the method probably depends on regular explant geometry (1-mm squares) and pooling sufficient explants. Serum-induced growth was quantified by measurements of ATP concentration, incorporation of [3H]thymidine and DNA concentration. The possible effects of endogenous vasodilator agents on growth were investigated by using lipid soluble analogues of their second messengers, namely 8-Br-cAMP and 8-Br-cGMP, which are known to relax rabbit aortic strips. Cell growth was inhibited concentration-dependently by 8-Br-cAMP but not 8-Br-cGMP (0.01-1 mM). The effect of 8-Br-cAMP was reversible, and also occurred when addition was delayed until after growth had commenced. The results imply that endogenous vasodilators such as prostacyclin, adenosine and adrenaline, which increase cAMP concentration, may normally suppress smooth muscle cell growth, whereas nitric oxide and atriopeptins, which increase cGMP concentration, may not.

Extracellular matrix degrading metalloproteinases in the pathogenesis of arteriosclerosis

We review the importance of extracellular matrix remodelling to the processes of vascular smooth muscle cell migration and proliferation that contribute to morphogenesis of the atherosclerotic plaque. In particular, the role of the matrix degrading metalloproteinase (MMP) family is discussed. This family of neutral, ZN(2+)-requiring enzymes are capable, in principle, of degrading all matrix proteins. Their activity is tightly controlled, however, at the level of synthesis of the inactive zymogens, activation by limited proteolysis and binding to endogenous inhibitor proteins (TIMPs). Direct evidence is presented for the involvement of MMPs in proliferation and outgrowth of vascular smooth muscle cells from explants of rabbit aorta in vitro. This was obtained using two structurally-unrelated inhibitors of MMPs, Ro 31-4724 and Ro 31-7467, both of which inhibited proliferation of cells in a concentration-dependent manner, Ro 31-4724 also inhibited outgrowth. Rabbit aortic smooth muscle cells were further shown to release MMPs, namely a 95 and a 72 kDa gelatinases that were inhibited by Ro 31-4724 and Ro 31-7467. The evidence suggests that degradation of basement membrane by gelatinase is required for proliferation and outgrowth of these cells. The implications of these findings for the pathogenesis and treatment of atherosclerosis are also discussed.

Endothelium-derived relaxing factor.

Abstract This article reviews what is known of endothelium-derived relaxing factor and its possible physiologic and pathophysiologic roles. This relaxing factor is now thought to be nitric oxide or a ready source of it. It acts as an endogenous nitrovasodilator, stimulating soluble guanvlate cyclase to increase cyclic guanosine monophosphate (GMP) levels in vascular smooth muscle and platelets, with consequent relaxant and anti-aggregatory effects (predominantly when stimulated through receptor-operated channels). Its actions are thus synergistic with those of cyclic adenosine monophosphate (AMP)-mediated stimulation (for example, adenosine, prostacyelin). Endothelium-derived relaxing factor is unstable and is thought lo act only very locally in vivo. Its release is continuous in the basal state and is stimulated by a number of neuropeptides and by agents released during platelet activation and thrombosis—with large differences in activity among different vessels. Endothelium-derived relaxing factor activity is also flow related, thereby coordinating vasomotor behavior in an intact vascular tree in response to changes in flow. Endothelium-derived relaxing factor activity is reduced in several pathologic states, including atherosclerosis.

An essential role for platelet-derived growth factor in neointima formation in human saphenous vein in vitro

The role of platelet-derived growth factor (PDGF), a potent vascular smooth muscle cells (SMC) mitogen and chemoattractant, was investigated during neointima formation in human saphenous vein organ culture. PDGFA and B messenger ribonucleic acid (mRNA) expression was detected by RNase protection assay and in situ hybridisation and PDGF protein by immunocytochemistry. The expression of PDGFA and B mRNA was low in veins before culture while PDGF protein was detected in all cell types. A neointima consisting of densely packed SMC developed after 14 days of culture. The dense packing and high expression of PDGFA and B mRNA in neointimal SMC led to higher PDGF protein concentrations in the neointima, the role of which was examined by culturing with neutralising anti-(human PDGF) antibodies. The anti-PDGF antibodies significantly reduced neointimal thickness by approximately 66% and the number of neointimal cells by approximately 50%, without affecting neointimal or medial proliferation indices or cell viability. These results suggest that PDGF played an essential role in SMC migration into the neointima in human saphenous vein.

Development of recombinant adenoviruses that drive high level expression of the human metalloproteinase-9 and tissue inhibitor of metalloproteinase-1 and -2 genes: Characterization of their infection into rabbit smooth muscle cells and human MCF-7 adenocarcinoma cells

Remodelling of the extracellular matrix resulting from increased secretion of metalloproteinase enzymes (MMPs) is implicated in many pathological conditions, including rheumatoid arthritis, restenosis following balloon angioplasty, atherosclerosis and cancer cell invasion and metastasis. Clear definition of the normal and pathological function of individual MMPs will benefit from approaches that use gene transfer to produce increases in MMP levels that mimic those observed in pathological conditions. Similarly, gene transfer methods leading to controlled increases in levels of the tissue inhibitor of metalloproteinases (TIMPs) will help to define the function of MMPs both in vitro and in vivo. Gene transfer of TIMPs may also have therapeutic potential in pathological conditions where inhibition of MMP activity may be beneficial. We have used the adenovirus serotype 5 vector system to generate replication-deficient recombinant adenoviruses capable of expressing the MMP-9, TIMP-1 or -2 genes. High level expression is driven by the cytomegalovirus major immediate early promoter (CMV IEP). Efficient and selective over-production of each recombinant protein was shown by immunofluorescence in either rabbit smooth muscle cells (SMC) or human MCF-7 adenocarcinoma cells. High level secretion directly dependent on the multiplicity of infection (MOI) was observed for each functional transgene by gelatin zymography. Using a quantitative ELISA assay, levels of recombinant TIMP-1 were detected when SMC were infected with as low as three plaque forming units (pfu) of virus per cell in vitro. A linear increase in TIMP-1 secretion was observed up to 1000 pfu/cell of virus (0.75 ng/10(4) cells/24 h at 3 pfu/cell to 1243 ng/10(4) cells/24h at 1000 pfu/cell). Similar levels of secretion of MMP-9 and TIMP-2 were observed by Western blot analysis using the same MOI of adenovirus. Thus, recombinant adenoviruses are an efficient and flexible system for high level expression of MMPs and TIMPs and will be useful tools in the study of matrix remodelling in vivo and in vitro.

Methylene blue but not changes in cyclic GMP inhibits resting and bradykinin-stimulated production of prostacyclin by pig aortic endothelial cells.

1 Primary cultures of pig aortic endothelial cells produced 6‐keto‐prostaglandin F1α (6‐keto PGF1α), the stable breakdown product of prostacyclin, both in the resting state and in response to bradykinin. The rise in 6‐keto‐PGF1α production induced by bradykinin (1–100 nm) was concentration‐dependent. 2 Treating endothelial cells with the inhibitor of soluble guanylate cyclase, methylene blue (0.1–20 μm) produced an irreversible reduction in resting and bradykinin (0.1 μm)‐stimulated production of 6‐keto‐PGF1α with an IC50 of 0.5 ± 0.1 μm. Treating endothelial cells with haemoglobin (10 μm) had no effect on resting or bradykinin (0.1 μm)‐stimulated production of 6‐keto‐PGF1α. 3 Two stimuli that elevate the level of guanosine 3′:5′‐cyclic monophosphate (cyclic GMP) in endothelial cells, 8‐bromo cyclic GMP (30 μm) and atriopeptin II (0.1 μm), each had no effect on resting or bradykinin (0.1 μm)‐stimulated production of 6‐keto‐PGF1α. Furthermore, treating endothelial cells with either 8‐bromo cyclic GMP (30 μm) or atriopeptin II (0.1 μm) had no effect on the ability of methylene blue (20 μm) to inhibit resting or bradykinin (0.1 μm)‐stimulated production of 6‐keto‐PGF1α. 4 Adding arachidonic acid (1 μm) to endothelial cells led to a marked stimulation of 6‐keto‐PGF1α production. Treating cells with either methylene blue (20 μm) or the cyclo‐oxygenase inhibitor, flurbiprofen (10 μm), inhibited both resting and arachidonic acid (1 μm)‐induced production of 6‐keto‐PGF1α. 5 In pig aortic endothelial cells methylene blue appears to block prostacyclin production by a mechanism independent of inhibition of soluble guanylate cyclase. Care should be exercised when using methylene blue as a selective inhibitor of endothelium‐derived relaxing factor due to its additional ability to block production of the other endothelium‐derived vasodilator, prostacyclin.

Increased secretion of gelatinases A and B from the aortas of cholesterol fed rabbits: relationship to lesion severity

Basement membrane degrading metalloproteinases (gelatinases) have been implicated in the regulation of vascular smooth muscle cell migration and proliferation in culture and during neointima formation in vivo. We compared the expression and activation of gelatinases A and B in explants derived from the arch, mid and distal portions of thoracic aortas of normal rabbits and those given a 1% cholesterol-containing diet for 8 weeks. Neointimal/medial ratio was less than 0.01 in normal rabbits but was significantly increased by cholesterol feeding in the arch (1.08 +/- 0.26), mid (0.75 +/- 0.28) and distal (0.32 +/- 0.12) portions of the aorta (mean +/- S.E.M., n = 6), and to a significantly (P < 0.05) greater extent in the arch and mid than distal portions. Secretion of gelatinase B measured by densitometric scanning of zymograms was undetectable from normal aortas, but was significantly increased by cholesterol feeding in the arch (0.16 +/- 0.06), mid (0.26 +/- 0.08) and distal (0.11 +/- 0.05) portions (optical density units, n = 6, each P < 0.05 versus normal diet). The increase in gelatinase B expression was localised by in situ hybridisation to neointimal vascular smooth muscle cells, macrophages and endothelial cells. Secretion of pro-gelatinase A was detected from normal aortas; it was increased by cholesterol feeding from the arch (4.0 versus 2.8, P < 0.05) and mid (3.6 versus 2.8, P < 0.05) but not distal portions of the aorta (1.8 versus 1.2, n.s.). Similar results were obtained for active gelatinase A secretion from the arch (0.50 versus 0.28, P < 0.05) and mid (0.47 versus 0.23, P < 0.05) but not distal portions (0.19 versus 0.20, n.s.). Increases in pro- and active gelatinase A secretion therefore paralleled the severity of atheroma formation. The results imply that increased basement membrane turnover mediated by gelatinases occurs during cholesterol induced atherosclerosis formation.

Sites of adenosine formation, action and inactivation in the brain.

Two extreme and non-exclusive hypotheses can be put forward to account for the largely inhibitory actions of adenosine on neural activity. Adenosine formation may be linked to the balance between formation and degradation of cytoplasmic ATP and hence only be indirectly related to neurotransmission. Adenosine formed by this route must then diffuse out of the cell to reach extracellular receptor sites. It may then act on the adenosine forming cells, neighbouring neurones or other cells including the cerebral vasculature to restore energy balance. Alternatively, adenosine formation may be directly linked to neurotransmission, probably by secretion of adenine nucleotides followed by ecto-nucleotidase activity. This adenosine then acts directly at extracellular receptors to modulate the activity of the neurones from which it arises. Localization of the sites of adenosine formation, action and inactivation in the brain could help, in principle, to distinguish these two hypotheses, or to define the neurones to which each hypothesis applies. Nagy and co-workers (Neurochem. Int.16, 211-221, 1990) review what has been learnt of the location in the brain of adenosine receptors, the nucleoside transporter and adenosine deaminase. In this critique, we attempt to place these important elements of the pathway of adenosine formation, action and inactivation in a wider perspective.

Inhibition of rabbit aortic smooth muscle cell proliferation by selective inhibitors of protein kinase C

1 We studied the effect of two structurally‐related, selective inhibitors of protein kinase C, Ro 31–8220 and Ro 31–7549, on the reinitiation of proliferation in quiescent first passage rabbit aortic smooth muscle cells in response to (a) the direct activator of protein kinase C, phorbol dibutyrate (PDBu), (b) platelet‐derived growth factor (PDGF), (c) a combination of PDGF and 5‐hydroxytryptamine (5‐HT) or (d) serum. 2 Ro 31–8220 and Ro 31–7549 concentration‐dependently inhibited prliferation in response to each mitogen. The inhibitory potency (IC50) of Ro 31–8220 and Ro 31–7549, respectively, was similar against proliferation induced by PDBu (0.55 and 1.1 μm), PDGF (0.6 and 0.9 μm), PDGF and 5‐HT (0.68 and 1.1 μm), although slightly less against serum (1.7 and 5 μm). The effects of the protein kinase C inhibitors on proliferation could not be ascribed to cytotoxicity. Neither Ro 31–8220 nor Ro 31–7549 (0.3‐3 μm) inhibited PDGF receptor tyrosine phosphorylation. 3 The results show that Ro 31–8220 and Ro 31–7549 are potent inhibitors of smooth muscle cell proliferation in response to a direct activator of protein kinase C, the defined growth factors, PDGF and 5‐HT, and the complex mixture of mitogens in serum. Protein kinase C activation thus appears to be an important growth transducing mechanism for each of these agents.

Evidence for extracellular deamination of adenosine in the rat heart

1. In rat heart perfused with adenosine (10(-6) M), dilazep (10(-4) M) inhibited incorporation of adenosine into nucleotides (an index of nucleoside transport and phosphorylation) to a greater extent (70%) than metabolism to inosine and uric acid (40%) and actually increased the recovery of inosine to 30% of the adenosine infused. 2. Extrapolating for complete inhibition of transport suggested that 60% of adenosine metabolism was intracellular and 40% extracellular. 3. Static incubations of atria also gave an estimate for extracellular metabolism of 40%. 4. Adenosine deaminase was localised by immunocytochemistry to the extracellular surface of endothelial cells of small coronary arteries. 5. Extracellular deamination may explain the lack of effect of nucleoside transport inhibitors on responses to adenosine in rat heart.