Biomedical subjects

H Wang

Publications and source records attributed to H Wang.

At least 1,549 records · Page 86Linked to original sources

PubMed1997

Determination of Association Constants for Cyclodextrin-Surfactant Inclusion Complexes: A Numerical Method Based on Surface Tension Measurements

Inclusion complexes of beta-cyclodextrin (beta-CD) with sodium octyl sulfonate (C8As), sodium dodecyl sulfonate (C12As), and sodium hexadecyl sulfonate (C16As) in aqueous solutions are studied by surface tension measurement at the air/water interface at different temperatures. At fixed concentrations of the surfactants, the surface tension increases with an increase in beta-CD concentration to a maximum value, at which it holds. The surface tension curves of the surfactants in the presence of beta-CD are higher than those in the absence of beta-CD. The curves rise higher with the increase in beta-CD concentration for each surfactant. The apparent critical micelle concentrations (CMC*) of the surfactants vary linearly with beta-CD concentration. The CMC* and surface tension values (including those after the CMC*) for the same system decrease with increase in temperature. A numerical method based on surface tension measurements is developed to determine the association constants for 1:1 inclusion complexes. This method is very reliable and easy to perform. The results demonstrate that the longer the hydrophobic tail of the surfactant, the greater the association constant with beta-CD, and that for the same surfactant the association constant is higher at lower temperatures.

Journal Article↗

PubMed1997

High affinity rheumatoid factor transgenic B cells are eliminated in normal mice.

Although systemic autoimmune diseases can be accompanied by multiple autoantibodies, certain specificities are dominant. Presumably, these specificities and their cognate Ags have properties that make them particularly amenable to autoimmune induction. Rheumatoid factors (RFs) are a dominant class of autoantibodies in rheumatoid arthritis and certain other autoimmune syndromes. To study the regulation of RFs in normal and autoimmune animals, we previously created a RF Ig transgenic model based on an RF isolated from an autoimmune MRL/lpr mouse. Using this model, called AM14, we were surprised to find that normal mice do not regulate disease-related RF B cells. This raised the question of whether RFs in general are not susceptible to tolerance induction, perhaps due to the unique properties of serum IgG and its FcRs. Alternatively, RFs can be tolerized, and the disease-related RFs are below the affinity threshold for such tolerance. To distinguish these possibilities, we generated a second RF transgenic model with the same specificity but much higher affinity than AM14. We found that, in contrast to AM14, high affinity RF B cells are subject to central tolerance, showing that there is not an absolute defect in RF B cell tolerance, but, rather, that RF B cell tolerance is affinity dependent even in normal animals. This is also the first model in which a disease-related specificity has been shown clearly to delete in a system in which Ag-positive and negative mice can be produced and compared.

Animals↗

PubMed1997

Phosphorylation and regulation of the Na+/H+ exchanger through mitogen-activated protein kinase.

We examined mitogen-activated protein kinase-mediated phosphorylation and activation of the Na+/H+ exchanger isoform type 1. A rabbit skeletal muscle extract was fractionated by FPLC chromatography. Four main fractions had the ability to phosphorylate the carboxyl-terminal region of NHE1. Western blot analysis and immunoprecipitation showed that three of these were associated with MAP kinase-dependent phosphorylation. Phosphorylation studies using purified MAP kinase showed that the region involved was the carboxyl-terminal 178 amino acids of the protein and that the stoichiometry was 1 phosphate/mol of protein. In-gel kinase assays showed that cytosolic extracts from smooth muscle cells also phosphorylate the carboxyl-terminal of NHE1 and that the MAP kinase-dependent phosphorylation could be activated by PDGF and AngII. Mutant cell lines with an inducible dominant negative MAP kinase showed decreased serum activation of Na+/H+ exchange but normal hypertonic activation of the protein. The results show that MAP kinase is intimately involved in regulation of the Na+/H+ exchanger, possibly through phosphorylation of one amino acid of the carboxyl-terminal cytosolic domain.

Animals↗

PubMed1997

BRCA1 proteins are transported to the nucleus in the absence of serum and splice variants BRCA1a, BRCA1b are tyrosine phosphoproteins that associate with E2F, cyclins and cyclin dependent kinases.

BRCA1, a familial breast and ovarian cancer susceptibility gene encodes nuclear phosphoproteins that function as tumor suppressors in human breast cancer cells. Previously, we have shown that overexpression of a BRCA1 splice variant BRCA1a accelerates apoptosis in human breast cancer cells. In an attempt to determine whether the subcellular localization of BRCA1 is cell cycle regulated, we have studied the subcellular distribution of BRCA1 in asynchronous and growth arrested normal, breast and ovarian cancer cells using different BRCA1 antibodies by immunofluorescence and immunohistochemical staining. Upon serum starvation of NIH3T3, some breast and ovarian cancer cells, most of the BRCA1 protein redistributed to the nucleus revealing a new type of regulation that may modulate the activity of BRCA1 gene. We have also characterized two new variant BRCA1 proteins (BRCA1a/p110 and BRCA1b/ p100) which are phosphoproteins containing phosphotyrosine. Immunofluorescence and Western blotting analysis indicate cytoplasmic and nuclear localization of BRCA1a and BRCA1b proteins. To elucidate the biological function of BRCA1, we created a bacterial fusion protein of glutathione-transferase (GST) and BRCA1 zinc finger domain and detected two cellular proteins with molecular weights of approximately 32 and 65 kD, one of which contains phosphotyrosine designated p32 and p65 BRCA1 interacting proteins (BIP) that specifically interact with BRCA1. Western blot analysis of BIP with cyclins/CDKs and E2F antisera indicated association with cdc2, cdk2, cdk4, cyclin B, cyclin D, cyclin A and E2F-4 but not with cdk3, cdk5, cdk6, E2F-1, E2F-2, E2F-3, E2F-5 and cyclin E. Furthermore, we have also demonstrated a direct interaction of in vitro translated BRCA1a and BRCA1b proteins with recombinant cyclin A, cyclin B1, cyclin D1, cdc2, cdk2 and E2F fusion proteins in vitro. Taken together these results seem to suggest that BRCA1 could be an important negative regulator of cell cycle that functions through interaction with E2F transcriptional factors and phosphorylation by cyclins/cdk complexes with the zinc ring finger functioning as a major protein-protein interaction domain. If the interactions we observe in vitro is also seen in vivo then it may be possible that lack or impaired binding of the disrupted BRCA1 proteins to E2F, cyclins/CDKs in patients with mutations in the zinc finger domain could deprive the cell of an important mechanism for braking cell proliferation leading to the development of breast and ovarian cancers.

3T3 Cells↗

PubMed1997

Determination of rate and equilibrium constants for the reactions between electron transfer mediators and proteins by linear sweep voltammetry.

Redox proteins undergo measurable charge transfer at electrodes only under special circumstances, while they readily take part in electron transfer reactions with mediators in solution. Advantage was taken of the latter fact to develop a new method to study the kinetics and equilibria of protein-mediator electron transfer reactions. It was shown that rate and equilibrium constants for the electron exchange between electron transfer mediator and the protein can be obtained from the analysis of the perturbation of the linear sweep voltammetry (LSV) response of the mediator due to the presence of the protein. The experiments were carried out under conditions where the protein does not interact with the electrode. Theoretical data obtained by digital simulation are presented to show the conditions under which rate and equilibrium constants are accessible by the LSV technique. The electron transfer reactions between ferri- and ferrocytochrome c and N,N,N',N'-tetramethylphenylenediamine and the corresponding radical cation in phosphate-buffered saline (0.04 M phosphate, pH 7.4, 0.1 M NaCl) buffer were selected to demonstrate the technique. These studies resulted in an equilibrium constant equal to 1.0 and forward and reverse rate constants equal to 1.6 x 10(4) M-1 s-1. The data available from this method include forward and reverse rate constants for electron transfer and the formal potential for the protein redox couple.

Animals↗

PubMed1997

Serine-578 is a major phosphorylation locus in human plasma plasminogen.

It has been reported that human plasminogen (HPg) exists in plasma in a phosphorylated form. We now document that both major glycoforms of plasma HPg contain a phosphoserine residue in their latent protease chains, as revealed by quantitative protein phosphate determinations and 31P-NMR analysis. The sequence location of the phosphoserine residue was established by time-of-flight matrix-assisted laser desorption ionization with delayed extraction mass spectrometric analysis of peptides resulting from complete tryptic and cyanogen bromide digests of the latent protease chain of HPg. Confirmation of the presence of organic phosphate in the identified peptide was obtained by determination of the resulting mass shift after treatment of the peptide with alkaline phosphatase. The data show that Ser578 is a major phosphorylation site in HPg.

Amino Acid Sequence↗

PubMed1997

The association of cardiovascular disease risk factors with abdominal obesity in Canada. Canadian Heart Health Surveys Research Group.

OBJECTIVE: To assess the degree of association of abdominal obesity with blood pressure and plasma lipid levels and to determine which anthropometric measures of obesity are most closely associated with these cardiovascular risk factors. DESIGN: Population-based, cross-sectional surveys. SETTING: Five Canadian provinces (Alberta, Manitoba, Ontario, Quebec and Saskatchewan) between 1989 and 1992. PARTICIPANTS: A probability sample of 16,007 men and women aged 18 to 74 was selected using health insurance registration files in each province and invited to participate. A complete set of measurements was available for 8974 (56%) adults. OUTCOME MEASURES: Initially, simple correlation analyses by age and sex were performed between the anthropometric variables-body mass index, waist circumference (WC), hip circumference (HC), ratio of waist to hip circumference (WHR)- and cardiovascular disease risk variables-systolic blood pressure (SBP), diastolic blood pressure (DBP), levels of total cholesterol (TC), low density lipoprotein (LDL) cholesterol, high density lipoprotein (HDL) cholesterol and triglycerides (TRIG) and the TC/HDL ratio. Canonical correlation analyses were performed to determine the multivariate associations between the anthropometric and risk variables. RESULTS: The simple correlations between anthropometric variables and cardiovascular disease risk variables were highest for SBP; moderate for DBP, HDL, TRIG and TC/HDL; and lowest for LDL and TC. Of the anthropometric variables, WC demonstrated the greatest correlations with the risk variables. The first canonical correlations were significant (p < 0.0001) in men (0.58) and women (0.61) of all ages. Of the anthropometric variables, WC consistently demonstrated the highest loading values in the first canonical variable in men (0.56) and women (0.59). Of the risk variables in both sexes, the loadings of TRIG were generally the largest, those of HDL, SBP, DBP intermediate and those of LDL the smallest. In men, the strength of these associations generally decreased with age, whereas in women they peaked in the 35-54 year age group. CONCLUSION: Considerable association was seen between measures of abdominal obesity and blood pressure and plasma lipid levels. WC is the measure of abdominal obesity most highly correlated with these cardiovascular disease risk factors.

Abdomen↗

PubMed1997

Structure of ribgrass mosaic virus at 2.9 A resolution: evolution and taxonomy of tobamoviruses.

Ribgrass mosaic virus (RMV) is a member of the tobamovirus group of plant viruses. The structure has been determined at 2.9 A resolution by fiber diffraction methods, and refined by molecular dynamics methods to an R-factor of 0.095. The carboxyl-carboxylate interactions that drive disassembly in tobamoviruses are present in RMV, but are very different from those in other tobamoviruses. RMV has some of the structural features of a subgroup I tobamovirus, a smaller number from subgroup II, and a number that appear to be unique to the RMV cluster of viruses. The structural studies confirm the evolutionary and taxonomic separation of the RMV cluster from both subgroup I and subgroup II tobamoviruses.

Amino Acid Sequence↗

PubMed1997

Up-regulation of nuclear genes in response to inhibition of mitochondrial DNA expression in chicken cells.

Vertebrate cells depleted of (rho0) mitochondrial DNA (mtDNA) exhibited phenotypic traits that differed from the parental (rho+) cells. To isolate genes whose expression is associated with mtDNA depletion, we constructed cDNA libraries from mRNAs isolated from chicken rho+ cells transformed by the MC29 (v-myc-containing) retrovirus and from rho0 cells developed by long-term exposure of the rho+ cells to ethidium bromide (EtdBr). Through subtractive hybridization procedures, three genes, elongation factor 1 alpha (EF- 1 alpha), beta-actin and v-myc were identified and found to be up-regulated in rho0 cells. In addition, Northern analysis demonstrated that the mRNA content for GAPDH was also elevated in rho0 cells. Run-on transcription assays and mRNA stability studies in the presence of actinomycin D indicated that elevated expression of these four genes depends, at least in part, upon increased rate of transcription. Other regulatory mechanisms contribute to the elevated expression of the transcripts in rho0 cells, as suggested by cycloheximide enhancement of the accumulation of the mRNAs for EF-1 alpha and beta-actin in rho0 cells, but not in parental rho+ cells. Moreover, inhibition of mtDNA replication and transcription by EtdBr and inhibition of translation on mitoribosomes by chloramphenicol also increased the expression of the four genes in parental rho+ cells, thus mimicking the situation in rho0 cells. These data suggest that information encoded within mtDNA participates in the regulation of nuclear genes in chicken cells.

Actins↗

PubMed1997

Paroxysmal discharges in the EL mouse, a genetic model of epilepsy.

The EL/Suz (EL) mouse is a strain that is highly susceptible to convulsive seizures after repeated sensory stimulation. Its control strain, DDY/Jc1 (DDY), is less susceptible under similar conditions. The seizure prone phenotype is the result of differences at several genetic loci. In vivo electrical recordings from the seizure prone EL mouse brain have shown that the appearance of abnormal discharges in the hippocampus are critical to the onset of generalized seizures, indicating that the hippocampus plays an important role in EL mouse seizure activity. In the present study, electrophysiological differences between EL and DDY mice (9-15 weeks of age) were examined by comparing field potentials recorded from the dentate granule cell layer of hippocampal brain slices from mice that had not been stimulated to induce seizures. In control physiological solution, no significant differences were observed in characteristics of perforant path evoked field potentials or in paired pulse depression of evoked field potentials using 20 to 300 ms interstimulus intervals. After 60 min of disinhibition following bicuculline (10 microM) exposure, however, prolonged large amplitude potentials, paroxysmal discharges, were evoked by perforant path stimulation in the dentate gyrus of EL mice but were absent in the DDY strain. Paroxysmal discharges were curtailed by APV and were similar to responses recorded from the dentate gyrus in hippocampal brain slices from temporal lobe epileptic patients. The field response to hilar stimulation was identical in both strains and was composed of a single population spike before and after bicuculline exposure. Mossy fiber terminals were not present in the molecular layer of either strain. We propose that the mechanisms leading to a greater likelihood of paroxysmal discharge generation in EL mouse may be important in the development and/or generation of epileptic seizures in this mouse strain and may be a significant phenotypic difference between the EL mouse and its parent strain.

Animals↗

PubMed1997

Glycosaminoglycans can influence fibroblast growth factor-2 mitogenicity without significant growth factor binding.

Fibroblast growth factors are important heparin binding, mitogenic proteins. The binding site in heparin and heparan sulfate for fibroblast growth factor-2 (basic fibroblast growth factor) has been described as rich in glucosamine-2-sulfate 1-->4 linked to iduronic acid-2-sulfate. The glucosamine residue in the heparin binding site is also 6-sulfated. A new glycosaminoglycan, acharan sulfate, has been chemically modified to prepare a polysaccharide, N-sulfoacharan sulfate, consisting of glucosamine-2-sulfate 1-->4 linked to iduronic acid-2-sulfate. Acharan sulfate binds very weakly to fibroblast growth factor-2 while N-sulfoacharan sulfate binds with nearly the same affinity as heparin. Mitogenicity studies were performed using heparan sulfate-free cells stably transfected with fibroblast growth factor receptor-1. Acharan sulfate inhibits heparin's enhancement of fibroblast growth factor-2 mitogenic activity, without affecting cell viability, while N-sulfoacharan sulfate shows heparin-like activity but at a greatly reduced level. These results suggest additional mechanisms not requiring high affinity glycosaminoglycan binding to fibroblast growth factor-2 may be important in its mitogenic activity.

Animals↗

PubMed1997

Immobilization of Polymer-Stabilized Noble Metal Colloids and Their Catalytic Properties for Hydrogenation of Olefins

Polymer-stabilized noble metal colloids were efficiently immobilized on silica by the addition of organic acids under mild conditions. The function of organic acids in the immobilization was studied by infrared spectroscopy. Transmission electron micrographs indicate that the immobilized colloids have a controlled particle size and size distribution. They serve as catalysts in the hydrogenation of cyclohexene and cyclopentadiene, the results of which show that this new type of immobilized colloid has high selectivity and good stability.

Journal Article↗

PubMed1997

Cyclin D1 overexpression promotes cardiomyocyte DNA synthesis and multinucleation in transgenic mice.

D-type cyclin/cyclin-dependent kinase (CDK) complexes regulate transit through the restriction point of the cell cycle, and thus are required for the initiation of DNA synthesis. Transgenic mice which overexpress cyclin D1 in the heart were produced to determine if D-type cyclin deregulation would alter myocardial development. Cyclin D1 overexpression resulted in a concomitant increase in CDK4 levels in the adult myocardium, as well as modest increases in proliferating cell nuclear antigen and CDK2 levels. Flow cytometric and morphologic analyses of dispersed cell preparations indicated that the adult transgenic cardiomyocytes had abnormal patterns of multinucleation. Histochemical analyses confirmed a marked increase in number of cardiomyocyte nuclei in sections prepared from the transgenic mice as compared with those from control animals. Tritiated thymidine incorporation analyses revealed sustained cardiomyocyte DNA synthesis in adult transgenic hearts.

Animals↗

PubMed1997

Elevated urokinase-type plasminogen activator receptor expression in a colon cancer cell line is due to a constitutively activated extracellular signal-regulated kinase-1-dependent signaling cascade.

The urokinase-type plasminogen activator receptor (u-PAR) facilitates extracellular matrix degradation in part by accelerating plasmin formation at the cell surface. We previously reported that u-PAR expression is elevated in colon cancer cell lines characterized by their in vitro invasive capacity. Since, u-PAR expression is increased by a variety of growth factors, which signal through the extracellular signal-regulated kinases 1 and 2 (ERK1/ERK2), we determined if these mitogen-activated protein kinases (MAPKs) regulate u-PAR expression in two cultured colon cancer cell lines. An in-gel kinase assay showed that ERK1 activity was considerably higher in RKO cells, which display > or = 10(5) receptors/cell, than the GEO cells which have approximately 10(4) urokinase receptors per cell. The expression of either an ERK-inactivating phosphatase (CL100), or a kinase-defective ERK1, decreased the activity of a u-PAR promoter-driven CAT reporter in RKO cells. Immune complex kinase assays indicated that the constitutive ERK1 activity in RKO cells was largely a result of an activated MEK1. Further, treatment of RKO cells with a specific inhibitor (PD 098059) of MEK1 activation, which diminished ERK1 activity, reduced the amount of urokinase specifically bound to the cell surface and this was associated with reduced laminin degradation. The expression of a dominant negative c-Raf-1 also reduced u-PAR promoter activity suggesting that MEK1 activation involved an activator at, or upstream, of this serine-threonine kinase. Transfection of the u-PAR-deficient GEO cells with a constitutively activated MEK1 expression construct up-regulated u-PAR promoter activity. Similarly treatment of GEO cells with a phosphatase inhibitor (sodium vanadate) caused a dose-dependent increase in ERK1 activity which paralleled increased cell surface binding of urokinase. Taken together, these data suggest that elevated u-PAR expression, in at least a sub-population of colon cancer, is partly a consequence of a constitutively activated ERK-1-dependent signaling cascade.

Calcium-Calmodulin-Dependent Protein Kinases↗

PubMed1997

Raf-1 kinase and exoenzyme S interact with 14-3-3zeta through a common site involving lysine 49.

14-3-3 proteins are a family of conserved dimeric molecules that bind to a range of cellular proteins involved in signal transduction and oncogenesis. Our solution of the crystal structure of 14-3-3zeta revealed a conserved amphipathic groove that may allow the association of 14-3-3 with diverse ligands (Liu, D., Bienkowska, J., Petosa, C., Collier, R. J., Fu, H., and Liddington, R. (1995) Nature 376, 191-194). Here, the contributions of three positively charged residues (Lys-49, Arg-56, and Arg-60) that lie in this Raf-binding groove were investigated. Two of the charge-reversal mutations greatly (K49E) or partially (R56E) decreased the interaction of 14-3-3zeta with Raf-1 kinase, whereas R60E showed only subtle effects on the binding. Interestingly, these mutations exhibited similar effects on the functional interaction of 14-3-3zeta with another target protein, exoenzyme S (ExoS), an ADP-ribosyltransferase from Pseudomonas aeruginosa. The EC50 values of 14-3-3zeta required for ExoS activation increased by approximately 110-, 5-, and 2-fold for the K49E, R56E, and R60E mutants, respectively. The drastic reduction of 14-3-3zeta/ligand affinity by the K49E mutation is due to a local electrostatic effect, rather than the result of a gross structural alteration, as evidenced by partial proteolysis and circular dichroism analysis. This work identifies the first point mutation (K49E) that dramatically disrupts 14-3-3zeta/ligand interactions. The parallel effects of this single point mutation on both Raf-1 binding and ExoS activation strongly suggest that diverse associated proteins share a common structural binding determinant on 14-3-3zeta.

14-3-3 Proteins↗

PubMed1997

Spermine inhibits proinflammatory cytokine synthesis in human mononuclear cells: a counterregulatory mechanism that restrains the immune response.

The local production of proinflammatory cytokines mediates the host response to inflammation, infection, and injury, whereas an overexpression of these mediators can injure or kill the host. Recently, we identified a class of multivalent guanylhydrazone compounds that are effective inhibitors of proinflammatory cytokine synthesis in monocytes/macrophages. The structure of one such cationic molecule suggested a molecular mimicry with spermine, a ubiquitous endogenous biogenic amine that increases significantly at sites of inflammation and infection. Here, we addressed the hypothesis that spermine might counterregulate the innate immune response by downregulating the synthesis of potentially injurious cytokines. When spermine was added to cultures of human peripheral blood mononuclear cells stimulated with lipopolysaccharide (LPS), it effectively inhibited the synthesis of the proinflammatory cytokines tumor necrosis factor (TNF), interleukin-1 (IL-1), IL-6, MIP-1alpha, and MIP-1beta. The inhibition of cytokine synthesis was specific and reversible, with significant inhibition of TNF synthesis occurring even when spermine was added after LPS. The mechanism of spermine-mediated cytokine suppression was posttranscriptional and independent of polyamine oxidase activity. Local administration of spermine in vivo protected mice against the development of acute footpad inflammation induced by carrageenan. These results identify a distinct molecular counterregulatory role for spermine in downregulating the monocyte proinflammatory cytokine response.

Animals↗

PubMed1997

Different developmental arrest points in RAG-2 -/- and SCID thymocytes on two genetic backgrounds: developmental choices and cell death mechanisms before TCR gene rearrangement.

To analyze the early development of T cell precursors in the absence of TCR gene rearrangement, recombinase-activating gene-deficient (RAG-2 -/-) thymocytes were compared with thymocytes from SCID mice on the C.B-17 (BALB) and B6 genetic backgrounds. RAG-2 -/- thymocytes accumulate as quiescent cells with a heat-stable Ag (HSA)-positive CD25+ CD44- c-kit(low) phenotype, resembling normal cells just before selection for functional TCR beta-chain expression. CD44 and c-kit progressively down-regulate in the HSA+ subset, providing a background-independent and TCR-independent developmental clock. On this basis, compared with RAG-2 -/- thymocytes, SCID thymocytes 1) arrest at more heterogeneous, and generally earlier, stages; 2) accumulate to lower overall cell numbers; and 3) maintain higher populations of cycling and activated G1 cells, showing both increased responsiveness and increased cell death. B6-SCID thymocytes appear to die particularly early. Low levels of Fas were observed on "advanced" HSA+ SCID thymocytes but not on any RAG-2 -/- thymocytes, suggesting a potential difference in activation state or mechanism of death. In both RAG-2 -/- and SCID thymocytes, there are also two discrete subsets of HSA(low) CD25- CD44+ c-kit+ cells: a Sca-1+ CD44++ CD122- NK1.1- putative progenitor subset and an NK-like Sca-1- CD44+(+) CD122+ NK1.1+ subset. The absolute cell numbers in these HSA(low) subsets and the extent of NK cell differentiation, measured by perforin expression, are nearly constant in all the mutant strains analyzed, in contrast to the HSA+ CD25+ population, which was expanded in the RAG-2 -/-. Thus, the SCID thymocytes appear to undergo a normal generation but a premature death as compared with the RAG-2 -/- thymocytes.

Animals↗

PubMed1997

Basic fibroblast growth factor causes growth arrest in MCF-7 human breast cancer cells while inducing both mitogenic and inhibitory G1 events.

Basic fibroblast growth factor (bFGF), a classical mitogen in fibroblasts and endothelial cells, inhibits the proliferation of MCF-7 and other human breast cancer cell lines. To explain this paradoxic effect, we investigated the effects of bFGF on cyclins and protein members of cyclin complexes that exert positive and negative control on the progression of cells through the G1 phase of the cell cycle. bFGF induced an increase in cyclin D1, cyclin E, and cyclin-dependent kinase 4 (cdk4) protein levels in a bFGF dose-dependent manner. However, bFGF also induced a heat-stable, transferable cytoplasmic factor in MCF-7 cells that inhibited the histone H1 kinase activity of reconstituted cyclin E-cdk2 and cyclin A-cdk2 complexes from Mv1Lu mink lung epithelial cells. The appearance of this inhibitor correlated with a bFGF dose- and time-dependent increase in the levels of cdk inhibitor p21(WAF1/CIP1) mRNA and protein. The increase in the level of p21(WAF1/CIP1) was associated with the disappearance of the rapidly migrating, activated form of cdk2 from cell lysates, dephosphorylation of the retinoblastoma protein (Rb), and a decrease in cyclin A levels. These changes were represented in the cyclin D1 and E complexes by an increased association with p21(WAF1/CIP1), proliferating cell nuclear antigen (PCNA), and the inactive form of cdk2, without an absolute change in cellular PCNA levels and by a switch in the association of cyclin D1 complexes with the hyperphosphorylated form to the dephosphorylated form of Rb. These experiments demonstrate that stimulation of MCF-7 cells with bFGF, although resulting in up-regulation of G1 proteins responsible for mitogenic events, also induces a concomitant decrease in cyclin A levels and an increase in p21(WAF1/CIP1) mRNA and protein and results in inactivation of cdk2, dephosphorylation of Rb, and a segregation of PCNA to the G1 cyclin complexes. The dual, conflicting signaling by bFGF results in a net inhibitory phenotype in these cells. These experiments suggest a pleiotropic role for bFGF in breast cancer.

Blotting, Western↗