Growth factors, growth factor response elements, and the cardiac phenotype.

Fibroblast growth factors (FGF) and type beta-1 transforming growth factor (TGF beta 1) are pleiotropic regulatory peptides which are expressed in myocardium in a precise developmental and spatial program and are up-regulated, in the adult heart, by ischemia or a hemodynamic burden. The accumulation of trophic factors after aortic banding supports the hypothesis that autocrine or paracrine pathways might function to mediate, in part, the consequences of mechanical load. Our laboratory has demonstrated that cardiac muscle cells are targets for the action of peptide growth factors and, more specifically, that modulation of the cardiac phenotype by basic FGF (bFGF) and TGF beta 1 strongly resembles the induction of fetal cardiac genes--including skeletal alpha-actin (SkA), beta-myosin heavy chain, and atrial natriuretic factor--which are characteristic of pressure-overload hypertrophy. Unexpectedly, and despite effects like those of bFGF on five other cardiac genes, acidic FGF (aFGF) was found to repress, rather than stimulate, SkA transcription in neonatal cardiac muscle cells. The proximal 200 nucleotides of a heterologous SkA promoter were sufficient for basal tissue-specific transcription, for induction by bFGF, and for inhibition by aFGF. Thus, both positive and negative regulation by peptide growth factors can be localized to the proximal SkA promoter. Full promoter activity required each of three CC[A/T]6GG motifs similar to the serum response element (SRE) for activation of the c-fos proto-oncogene, as previously shown for SkA transcription in a skeletal muscle background. The most proximal SRE, SRE1, was sufficient in the absence of other SkA promoter sequences for efficient tissue-specific expression in cardiac myocytes (versus cardiac fibroblasts), and was stimulated by bFGF to the same extent as the full-length promoter and endogenous gene. Despite its ability to repress the SkA promoter, aFGF had no significant effect on SRE1. Both FGFs up-regulated the canonical fos SRE, to a comparable degree. Thus, SRE1 can discriminate between signals generated in cardiac myocytes by bFGF and aFGF. In cardiac myocyte extracts, two predominant proteins contact SRE1: serum response factor (SRF) and a second protein, F-ACT-1. Thus, serum response factor and F-ACT-1 are candidate trans-acting factors for basal transcription of the SkA gene in cardiac muscle cells and for induction of SkA by bFGF and, potentially, other trophic signals.

The vascular smooth muscle alpha-actin gene is reactivated during cardiac hypertrophy provoked by load.

Cardiac hypertrophy triggered by mechanical load possesses features in common with growth factor signal transduction. A hemodynamic load provokes rapid expression of the growth factor-inducible nuclear oncogene, c-fos, and certain peptide growth factors specifically stimulate the "fetal" cardiac genes associated with hypertrophy, even in the absence of load. These include the gene encoding vascular smooth muscle alpha-actin, the earliest alpha-actin expressed during cardiac myogenesis; however, it is not known whether reactivation of the smooth muscle alpha-actin gene occurs in ventricular hypertrophy. We therefore investigated myocardial expression of the smooth muscle alpha-actin gene after hemodynamic overload. Smooth muscle alpha-actin mRNA was discernible 24 h after coarctation and was persistently expressed for up to 30 d. In hypertrophied hearts, the prevalence of smooth muscle alpha-actin gene induction was 0.909, versus 0.545 for skeletal muscle alpha-actin (P less than 0.05). Ventricular mass after 2 d or more of aortic constriction was more highly correlated with smooth muscle alpha-actin gene activation (r = 0.852; P = 0.0001) than with skeletal muscle alpha-actin (r = 0.532; P = 0.009); P less than 0.0005 for the difference in the correlation coefficients. Thus, smooth muscle alpha-actin is a molecular marker of the presence and extent of pressure-overload hypertrophy, whose correlation with cardiac growth at least equals that of skeletal alpha-actin. Induction of smooth muscle alpha-actin was delayed and sustained after aortic constriction, whereas the nuclear oncogenes c-jun and junB were expressed rapidly and transiently, providing potential dimerization partners for transcriptional control by c-fos.

Desensitization of prostaglandin F2 alpha-stimulated inositol phosphate generation in NIH-3T3 fibroblasts transformed by overexpression of normal c-Ha-ras-1, c-Ki-ras-2 and c-N-ras genes.

The stimulation of inositol phosphate generation in control and ras-gene-transformed NIH-3T3 cells by prostaglandin F2 alpha (PGF2 alpha) was investigated. Compared with the control cells, a desensitization of the response was observed in cells transformed by the overexpression of N-, Ha-, or Ki-ras genes. This desensitization was without effect upon the concentration causing half-maximal effect (EC50), dissociation constant (Kd) or number of PGF2 alpha receptors. Inhibition of PG synthesis was without effect upon desensitization, demonstrating that the effect was not agonist-induced. Desensitization could be induced in NIH-3T3 cells by culturing under conditions where the cells were all in the exponential growth phase, or by a 12 h exposure to a C-kinase-activating phorbol ester. These results suggest that desensitization of certain agonist-induced inositol phospholipid responses in ras-transformed cells is a consequence of increased cell proliferation and associated amplification in C-kinase activity and is an indirect consequence of transformation by ras.

Activation of inositol phospholipid breakdown by prostaglandin F2 alpha without any stimulation of proliferation in quiescent NIH-3T3 fibroblasts.

Stimulation of NIH-3T3 cells with prostaglandin F2 alpha (PGF2 alpha) caused a dose- and time-dependent generation of inositol phosphates. The first detectable changes were in the levels of Ins(1,4,5)P3 and Ins(1,3,4,5)P4. Increases in Ins(1,3,4)P3, InsP2 and InsP were detected later, and only minor changes were observed in putative InsP5 or InsP6. The accumulation of inositol phosphates was synergistically increased by the addition of calf serum, whereas PGF2 alpha had no effects on cell proliferation in either the presence or the absence of calf serum. Stimulation of a different clone of NIH-3T3 cells (AmNIH-3T3) or Swiss 3T3 cells with PGF2 alpha resulted in both inositol phospholipid breakdown and cell proliferation. No differences were found in the characteristics of PGF2 alpha-stimulated inositol phosphate generation between the two clones of NIH-3T3 cells, nor was there any difference in receptor number of Kd. These results question the role of inositol phospholipid breakdown in mitogenesis and demonstrate significant differences in the biochemical properties of apparently the 'same' cells.

A controlled release papaverine tablet (Papacontin): a study in normal volunteers.

1. A controlled release tablet containing 300 mg of papaverine hydrochloride was compared with a capsule containing the same dose. 2. The in vitro release pattern showed a steady dissolution of the tablet, with release of the papaverine over 10 h, whereas the capsule release time was 40 min. 3. Plasma levels in volunteers who were given the tablet showed a rapid rise to therapeutic levels (150-300 ng/ml) within 1 h, and maintenance of these levels for 10-12 h, with no accumulation on repeated 12 hourly dosage. Administration of the capsule produced early peaking and a subsequent rapid fall in plasma levels.