Fibroblast growth factor-2 (FGF-2) has been implicated in vascular smooth muscle cell (SMC) migration, a key process in vascular disease. We demonstrate here that FGF-2 promotes SMC motility by altering beta1 integrin-mediated interactions with the extracellular matrix (ECM). FGF-2 significantly increased surface expression of alpha2beta1, alpha3beta1, and alpha5beta1 integrins on human SMCs, as assessed by flow cytometry. The greatest increase was for the collagen-binding alpha2beta1 integrin. Despite this, FGF-2 did not increase SMC adhesion to type I collagen but instead promoted SMC elongation and SMC motility. The latter was evaluated by using a microchemotaxis chamber and by digital time-lapse video microscopy. Although FGF-2 was not chemotactic for human SMCs, cells preincubated with FGF-2 displayed a 3.1-fold increase in migration to the undersurface of porous type I collagen-coated membranes and a 2.1-fold increase in migration speed on collagen. Furthermore, chemotaxis to platelet-derived growth factor-BB on collagen was significantly greater in SMCs exposed to FGF-2. FGF-2-induced elongation and migration on collagen were inhibited by a blocking anti-alpha2beta1 antibody; however, SMC adhesion to collagen was unaffected. SMC migration on fibronectin was also enhanced by FGF-2, although less prominently: migration through porous membranes increased 1.8-fold, and migration speed increased 1.3-fold. Also, FGF-2 completely disassembled the smooth muscle alpha-actin-containing stress fiber network contemporaneously with the change in integrin expression and cell shape. We conclude that (1) exogenous FGF-2 promotes SMC migration and potentiates chemotaxis to PDGF-BB; (2) the promigratory effect of FGF-2 is especially prominent on type I collagen and is mediated by upregulation of alpha2beta1 integrin; and (3) FGF-2 disassembles actin stress fibers, which may promote differential utilization of alpha2beta1 integrin for motility but not adhesion. This dynamic SMC-ECM interplay may be an important mechanism by which FGF-2 facilitates SMC motility in vivo.

Genetic and biochemical studies have provided convincing evidence that the 5 noncoding region (5 NCR) of hepatitis C virus (HCV) is highly conserved among viral isolates worldwide and that translation of HCV is directed by an internal ribosome entry site (IRES) located within the 5 NCR. We have investigated inhibition of HCV gene expression using antisense oligonucleotides complementary to the 5 NCR, translation initiation codon, and core protein coding sequences. Oligonucleotides were evaluated for activity after treatment of a human hepatocyte cell line expressing the HCV 5 NCR, core protein coding sequences, and the majority of the envelope gene (E1). More than 50 oligonucleotides were evaluated for inhibition of HCV RNA and protein expression. Two oligonucleotides, ISIS 6095, targeted to a stem-loop structure within the 5 NCR known to be important for IRES function, and ISIS 6547, targeted to sequences spanning the AUG used for initiation of HCV polyprotein translation, were found to be the most effective at inhibiting HCV gene expression. ISIS 6095 and 6547 caused concentration-dependent reductions in HCV RNA and protein levels, with 50% inhibitory concentrations of 0.1 to 0.2 M. Reduction of RNA levels, and subsequently protein levels, by these phosphorothioate oligonucleotides was consistent with RNase H cleavage of RNA at the site of oligonucleotide hybridization. Chemically modified HCV antisense phosphodiester oligonucleotides were designed and evaluated for inhibition of core protein expression to identify oligonucleotides and HCV target sequences that do not require RNase H activity to inhibit expression. A uniformly modified 2-methoxyethoxy phosphodiester antisense oligonucleotide complementary to the initiator AUG reduced HCV core protein levels as effectively as phosphorothioate oligonucleotide ISIS 6095 but without reducing HCV RNA levels. Results of our studies show that HCV gene expression is reduced by antisense oligonucleotides and demonstrate that it is feasible to design antisense oligonucleotide inhibitors of translation that do not require RNase H activation. The data demonstrate that chemically modified antisense oligonucleotides can be used as tools to identify important regulatory sequences and/or structures important for efficient translation of HCV.

No abstract

Abstract-Angiotensin II is an established regulator of vascular tone and smooth muscle cell (SMC) growth. However, there are little data about its effect on collagen synthesis by SMCs and none regarding the mechanism of such an effect. We studied the effect of angiotensin II on collagen production by human arterial SMCs, using uptake of [ 3 H]proline into collagenase-digestible proteins, and by ribonuclease protection assay for mRNA encoding the pro␣1 chain of type I collagen, the major collagen in arteries. This revealed a dose-dependent increase in relative collagen synthesis rate and a dose-dependent increase in pro␣1(I) collagen mRNA abundance, with the half-maximal effect at 1.7 nmol/L. Angiotensin II-stimulated collagen expression was associated with a 6-fold increase in transforming growth factor-␤ (TGF-␤) production and was inhibited by a neutralizing antibody to TGF-␤. Both collagen production and TGF-␤ release were inhibited by the AT 1 -specific antagonist, losartan, but not by the AT 2 receptor antagonist, PD123319. To determined if tyrosine phosphorylation was functionally linked to collagen synthesis, we studied the effect of 2 mechanistically distinct inhibitors of tyrosine kinase, genistein, and tyrphostin A25. These inhibitors abrogated angiotensin II-mediated procollagen mRNA expression and angiotensin II-mediated TGF-␤ production, whereas the inactive homolog tyrphostin A1 had no effect. We conclude that angiotensin II stimulates collagen production in human arterial SMCs via the AT 1 receptor and an autocrine loop of TGF-␤, induction of which requires tyrosine phosphorylation.