Rodney J. Dilley

Distinct Patterns of Transforming Growth Factor-β Isoform and Receptor Expression in Human Atherosclerotic Lesions Colocalization Implicates TGF-β in Fibrofatty Lesion Development

BACKGROUNDnSome animal studies suggest that transforming growth factor-beta (TGF-beta) protects vessels from atherosclerosis by preventing intima formation, but others indicate a role in vessel proteoglycan accumulation and lipoprotein retention. To distinguish between these possibilities in humans, immunohistochemical studies were performed examining the coexpression of TGF-beta isoforms and the TGF-beta receptors ALK-5 and TbetaR-II in aorta during the various stages of atherosclerotic lesion development.nnnMETHODS AND RESULTSnThe spatial relationships between TGF-beta1, TGF-beta3, ALK-5, and TbetaR-II expression were compared in aortic segments from 21 subjects. Nonatherosclerotic intima contained predominantly TGF-beta1, low concentrations of TbetaR-II, and barely detectable amounts of ALK-5. In contrast, fatty streaks/fibrofatty lesions contained high concentrations of both TGF-beta isoforms. Smooth muscle cells (SMCs), macrophages, and foam cells of macrophage and SMC origin contributed to these high levels. These lesions also contained high, colocalized concentrations of ALK-5 and TbetaR-II. Despite fibrous plaques containing TGF-beta1, its receptors were at detection limits. We found no evidence for truncated TbetaR-II expression in either normal intima or the various atherosclerotic lesions.nnnCONCLUSIONSnTGF-beta appears to be most active in lipid-rich aortic intimal lesions. The findings support the hypothesis that TGF-beta contributes primarily to the pathogenesis of lipid-rich atherosclerotic lesions by stimulating the production of lipoprotein-trapping proteoglycans, inhibiting smooth muscle proliferation, and activating proteolytic mechanisms in macrophages.

Inhibition of Protein Tyrosine Kinases Attenuates Increases in Expression of Transforming Growth Factor-β Isoforms and Their Receptors Following Arterial Injury

Transforming growth factor-beta 1 (TGF-beta 1) has been implicated in neointima formation in mechanically injured vessels and in restenosis after angioplasty. To further understand the significance of TGF-beta s in neointima formation, we examined the temporal expression of three TGF-beta isoforms (-beta 1, -beta 2, and -beta 3), their receptors (ALK-2, ALK-5, and T beta RII), and two putative TGF-beta responses (elevations in alpha v and beta 3 integrin mRNAs) in balloon catheter-injured rat carotid arteries and their dependency on tyrosine kinase activity. Using a standardized reverse transcriptase-polymerase chain reaction assay optimized to estimate mRNA levels, we observed distinct patterns of mRNA regulation for TGF-beta 1, -beta 2, and -beta 3 during the 48 hours immediately after injury, which were localized to the vessels media. TGF-beta 1 mRNA increased 10-fold during this time while TGF-beta 3 mRNA also increased almost 2-fold. There were also increases in mRNAs encoding the TGF-beta type I receptors ALK-5 and ALK-2, as well as the type II receptor (T beta RII). Eight hours after the injury, mRNA levels for ALK-2 and ALK-5 were on average 2-fold higher; mRNA encoding the type II receptor increased approximately 3-fold by 24 hours. There were also associated increases in TGF-beta 1, TGF-beta 3, ALK-5, and T beta RII immunoreactive peptide levels. Peak increases in mRNAs for integrins alpha v and beta 3 averaged approximately 2-fold and 2.5-fold, respectively. Perivascular administration of the tyrosine kinase inhibitor genistein at the time of vessel injury markedly (> 85%) inhibited elevations in mRNAs encoding TGF-beta 1, TGF-beta 3, T beta RII, and the two integrins alpha v and beta 3, while application of its inactive chemically similar homologue daidzein did not prevent the injury-induced elevations in mRNA levels. Since the increases in integrins alpha v and beta 3 mRNA could be theoretically attributed to TGF-beta actions despite being dependent on tyrosine kinase activity, we examined whether the observed elevations in integrins alpha v and beta 3 were due to TGF-beta 1 secretion, using cultured rat carotid artery smooth muscle cells. TGF-beta 1 neutralizing antibodies specifically inhibited elevations in integrins alpha v and beta 3 mRNAs due to platelet-derived growth factor-BB and fibroblast growth factor-2. We conclude that multiple components of the TGF-beta system in vessels are activated following injury and influence expression of integrin receptors important for smooth muscle cell migration. Activation of the TGF-beta system appears to be highly dependent on tyrosine kinases.

Low Blood Flow After Angioplasty Augments Mechanisms of Restenosis Inward Vessel Remodeling, Cell Migration, and Activity of Genes Regulating Migration

Abstract —The predominant cause of restenosis after angioplasty is now thought to be inward remodeling, but the mechanisms responsible are unknown. Remodeling in normal vessels is regulated by the endothelium in response to altered shear stress. Although the endothelium is often damaged by angioplasty, restenosis rates after angioplasty have been correlated with impaired coronary flow. Thus, we examined how increases or decreases in blood flow through balloon catheter–injured rat carotid arteries affect vessel morphometry (4, 10, and 28 days), cell migration (4 days), and levels of promigratory mRNAs (2 and 10 days). After 28 days, the luminal area in vessels with low blood flow was significantly less than in those with normal and high blood flow (0.17±0.01 [low] versus 0.24±0.06 [normal] versus 0.30±0.02 [high] mm2, P <0.01), predominantly because of accentuated inward remodeling (or reduced area within the external elastic lamina; 0.42±0.02 [low] versus 0.54±0.07 [normal] versus 0.53±0.04 [high] mm2, P <0.05). Low flow also enhanced smooth muscle cell migration 4 days after injury by 90% above normal and high flows (P <0.01). Two days after injury, low flow significantly increased levels of mRNAs encoding promigratory peptides (integrin &agr;v&bgr;3, transforming growth factor-&bgr;1, CD44v6, MDC9, urokinase plasminogen activator receptor, and &bgr;-inducible gene h3); these changes persisted 10 days after injury and were localized to the neointima. Low blood flow may promote restenosis after angioplasty because of its adverse effect on vessel remodeling, and it is associated with the augmented expression of multiple genes central to cell migration and restenosis.

Angiotensin-Converting Enzyme Inhibition Abolishes Medial Smooth Muscle PDGF-AB Biosynthesis and Attenuates Cell Proliferation in Injured Carotid Arteries Relationships to Neointima Formation

BACKGROUNDnACE inhibitors can attenuate the development of intimal fibrocellular lesions after balloon catheter vessel injury, but the mechanisms responsible are unknown.nnnMETHODS AND RESULTSnTo evaluate how basic fibroblast growth factor (FGF-2) and the platelet-derived growth factor (PDGF) isoforms are affected by ACE inhibition in injured rat carotid arteries in relation to smooth muscle cell (SMC) proliferation, we examined the effects of oral perindopril on FGF-2 and PDGF isoform levels in carotid arteries 2 days after balloon catheter injury. [3H]Thymidine incorporation into medial and intimal SMCs was also assessed. Uninjured vessels contained two forms of FGF-2, with molecular weights of 18 and 22 kD, and PDGF-AA. Two days after injury, FGF-2 and PDGF-AA levels were markedly reduced, but high levels of PDGF-AB became apparent when the SMCs were proliferating. Perindopril completely abolished the biosynthesis of PDGF-AB but had little effect on residual FGF-2. This was accompanied by a 25% reduction in medial SMC proliferation. Neointimal cell proliferation 10 days after injury was unaffected by perindopril, although neointima size was reduced by 30%. Commencing perindopril treatment 4 days after the injury confirmed that early events associated with effects on medial SMCs were the major contributors to the attenuated neointimal lesions.nnnCONCLUSIONSnThe ability of ACE inhibitors such as perindopril to attenuate neointima formation and growth in balloon catheter-injured rat carotid arteries is dependent on early events in the media, the inhibition of SMC PDGF-AB biosynthesis and attenuation of proliferation. Neointima formation in similarly injured vessels containing SMCs that are either unresponsive to PDGF-AB or exhibit an ACE-independent profile of growth factor biosynthesis responses may account for the ineffectiveness of ACE inhibition in some species.

Mechanical strain stimulates a mitogenic response in coronary vascular smooth muscle cells via release of basic fibroblast growth factor

Mechanical strain has been shown to induce mitogenesis in a rat neonatal vascular smooth muscle (VSM) cell line in a response mediated predominantly by transcription, expression, and release of platelet-derived growth factor (PDGF). We examined the effect of cyclic mechanical strain and growth factor production on mitogenic response in ovine coronary artery smooth muscle cells. Vascular smooth muscle cells were cultured from explants of left anterior descending (LAD) coronary arteries from young sheep. Cells for experiments were grown on wells with silicone-elastomer bottoms, and subjected to strain (60 cycles/min) using a vacuum actuated strain device. Tritiated thymidine incorporation was used as a measure of DNA synthesis. Cell membrane damage was assessed with differentially permeable nuclear staining dyes. We observed an increase in tritiated thymidine incorporation in response to strain with a temporal response identical to that observed in response to exogenous growth factors (PDGF-BB and basic fibroblast growth factor [bFGF]). Supernatant medium obtained from stretched cells induced a twofold increase in DNA synthesis in unstretched cells. The mitogenic response was abolished by monoclonal antibodies to bFGF, but not by antibodies to PDGF-AB. Studies of fluorescent dye exclusion indicated the stretching protocol caused no cell membrane damage. Thus, mechanical strain is an important stimulus for growth factor release in coronary VSM cells. The mitogenic response is mediated by release of bFGF.

Fenofibrate modifies human vascular smooth muscle proteoglycans and reduces lipoprotein binding

Aims/hypothesisVascular disease in type 2 diabetes is associated with an up-regulation of atherogenic growth factors, which stimulate matrix synthesis including proteoglycans. We have examined the direct actions of fenofibrate on human vascular smooth muscle cells (VSMCs) and have specifically investigated proteoglycan synthesis and binding to LDL.MethodsProteoglycans synthesised by human VSMCs treated with fenofibrate (30xa0µmol/l) were assessed for binding to human LDL using a gel mobility shift assay, metabolically labelled with [35S]-sulphate and quantitated by cetylpyridinium chloride. They were then assessed for electrophoretic mobility by SDS-PAGE, for size by gel filtration, for sulphation pattern by fluorophore-assisted carbohydrate electrophoresis, and for glycosaminoglycan (GAG) composition by enzyme digestion.ResultsProteoglycans synthesised in the presence of fenofibrate showed an increase in the half-maximum saturation concentration of LDL from 36.8±12.4xa0µg/ml to 77.7±17xa0µg/ml under basal conditions, from 24.9±4.6xa0µg/ml to 39.1±6.1xa0µg/ml in the presence of TGF-β1, and from 9.5±4.4xa0µg/ml to 31.1±3.4xa0µg/ml in the presence of platelet-derived growth factor/insulin. Fenofibrate treatment in the presence of TGF-β1 inhibited the incorporation of [35S]-sulphate into secreted and cell-associated proteoglycans synthesised by human VSMCs by 59.2% (p<0.01) and 39.8% (p<0.01) respectively. The changes in sulphate incorporation following treatment with fenofibrate were associated with a concentration-related increase in the electrophoretic mobility due to a reduction in GAG length. There was no change in the sulphation pattern; however, there was an alteration in the disaccharide composition of the GAGs.Conclusions/interpretationFenofibrate modifies the structure of vascular proteoglycans by reducing the length of the GAG chains and GAG composition, resulting in reduced binding to human LDL, a mechanism which may lead to a reduction of atherosclerosis and cardiovascular disease in people with diabetes treated with fenofibrate.

Vascular Hypertrophy in Renal Hypertensive Spontaneously Hypertensive Rats

Vascular smooth muscle cells isolated from spontaneously hypertensive rats (SHR) replicate faster in vitro than do cells from Wistar-Kyoto (WKY) rats, suggesting that the vascular hypertrophy seen early in the life of SHR might be at least partially caused by abnormal cellular growth properties in vivo. To test whether specific growth stimuli produce more extensive hypertrophy in SHR than WKY rats, we compared their cardiovascular growth responses to two-kidney, one clip renal hypertension. Six-week-old animals were subjected to either renal artery clipping or sham operation. Four weeks after renal artery clipping, there was a proportionately smaller rise in systolic blood pressure in SHR than WKY rats (21% and 44%, respectively); however, the overall level of systolic blood pressure achieved in the two rat strains differed by less than 10 mm Hg (4%). Limitations in the blood pressure responses of SHR to renal artery clipping were not due to inadequate development of left ventricular hypertrophy, as this was greater in SHR than WKY rats; however, aortic hypertrophy was similar in both strains. Aortic DNA content changes in SHR were consistent with a significant hyperplasia of medial smooth muscle cells, whereas in WKY rats, there was cellular hypertrophy. Small and medium-sized arteries of the mesenteric vasculature were also hypertrophied in SHR, and the medial cross-sectional area increased by 63% and 114%, respectively, compared with increases of only 15% and 23% in WKY rats. Strain differences between the sham-operated rat groups were small.(ABSTRACT TRUNCATED AT 250 WORDS)

High glucose potentiates mitogenic responses of cultured ovine coronary smooth muscle cells to platelet derived growth factor and transforming growth factor-β1

Macrovascular complications in diabetes are associated with exaggerated growth responses of vascular smooth muscle cells. We studied the effect of high glucose media on the growth responses of vascular smooth muscle cells from the left anterior descending (LAD) coronary artery of young sheep. Experiments were conducted in DMEM containing 5.5 or 25 mmol/l glucose and mitogenic responses assessed by 3H-thymidine incorporation. In the absence of growth factors there was a slight and variable response to high glucose but the maximum response to platelet derived growth factor-bb (PDGF-bb) (100 ng/ml) was increased more than 2-fold. Transforming growth factor-beta1 (1 ng/ml) caused a 100% increase of the PDGF-bb response in both normal and high glucose media. The acute stimulatory effect of high glucose was not affected by pre-incubation of the cells for 24 h in the high glucose medium. The mitogenic response occurring in the presence of PDGF-bb and high glucose was totally inhibited by the tyrosine kinase inhibitors (imatinib and genistein) and could not be mimicked by increasing diacylglycerol in low glucose media with the diacylglycerol kinase inhibitor, R59949. In conclusion, high glucose, per se, only very weakly stimulates smooth muscle cell growth but it interacts positively to potentiate the responses to the vascular derived growth factors PDGF and TGF-beta1. The effect of high glucose is transduced via receptor tyrosine kinases and may not involve diacylglycerol that is subject to diacylglycerol kinase catabolism. The data provide explanations for the accelerated vascular smooth muscle cell proliferation in diabetes.

Left ventricular remodelling impacts on coronary flow reserve in hypertensive patients: is there a vascular mechanism?

Left ventricular hypertrophy is an important independent risk factor for cardiovascular morbidity and mortality [1–3], and is associated with arterial hypertension [3] in both treated and untreated patients [4]. The intrinsic factors that increase the risk (i.e. the pathophysiologic mechanisms by which hypertrophic remodelling of the myocardium contributes to cardiovascular related death) remain unknown [5–6]. One factor is likely to be the increased prevalence of ischaemic damage and resultant fibrosis in the myocardium, leading to dysrhythmia, diastolic dysfunction and heart failure [5–7]. Increased myocardial mass leads to increased oxygen consumption while at the same time reducing coronary blood flow reserve (the ratio of coronary flow at maximal dilatation to flow at rest) to impair delivery of blood to the affected region, resulting in ischaemia [8]. Coronary flow reserve may be impaired very early in hypertension, before hypertrophy is apparent, and thus may be causative for subsequent ischaemia and fibrosis [9]. Coronary endothelial dysfunction is common in hypertensive patients, resulting in an ineffective vasodilator function [10]. However, reduced coronary flow reserve is largely the result of changes in minimal coronary resistance which are identified in a pharmacologically dilated vasculature, and therefore independent of vascular tone [7–11]. Thus, structural changes in the coronary vasculature are most likely to be the major contributors to impaired coronary flow reserve and these may be qualitatively similar to the well described effects of hypertension on the peripheral circulation.

729-2 Macrophage-Monocyte Invasion is Associated with Greater Cardiac Hypertrophy in Hypertensive Rats

Cardiac hypertrophy is an adaptive response to increased wall stress in hypertension. Although the mechanisms underlying this adaptive response are not clearly understood, a genetic predisposition appears to playa role. We sought to examine the hypothesis that an inflammatory response occurs in the heart early in the course of development of hypertension. We measured blood pressure, heart weight and cardiac macrophage-monocyte invasion in 6 week old SHR and WKY rats with renal hypertension (1 Kidney-1 Clip, 1K1C) or sham-operated controls (SHAM). Macrocyte-monophage invasion was measured as the extent of immunohistochemical staining with ED1, a cytoplasmic antigen in this cell type. Measurements were made at 7 and 21 days post procedure. Renal hypertensive rats, both SHR and WKY, had higher blood pressures at both time points than SHR sham, which in turn had higher blood pressures that WKY sham rats. Heart weight/body weight ratios were highest in SHR-1 K1 C rats, followed by SHR SHAM and WKY 1K1C, and least in WKY sham rats. ED1 staining was also highest in SHR 1K-lC rats, followed by SHR sham rats, and was considerably lower in WKY rats, 1K-1C or sham. Data atthe 21 day time point are shown below: BP (mmHg) Heart wt/body wt (mg/g) ED1 (counts/HPF) SHR 1K1C 205 4.3 20.0 SHR SHAM 170 3.4 15.0 WKY 1K1C 203 3.7 12.5 WKY SHAM 143 2.8 12.5 Thus, significant monocyte-macrophage invasion is a hitherto unrecognized feature of cardiac hypertrophy in SHR rats. We speculate that growth factors released by such cells may contribute to hypertensive cardiac hypertrophy.