Stimulation of plasminogen activator expression and induction of DNA synthesis by microtubule-disruptive drugs

Linking cell structure to gene regulation: signaling events and expression controls on the model genes PAI-1 and CTGF

The microtubule and microfilament cytoskeletal systems as well as cell-to-cell contacts and cell-matrix interactions are critical regulators of cell structure and function. Alterations in cell shape profoundly influence signaling events and gene expression programs that impact a spectrum of biological responses including cell growth, migration and apoptosis. These same pathways also contribute to the progression of several important pathologic conditions (e.g., arteriosclerosis, vascular fibrosis, and endothelial dysfunction). Indeed, hemodynamic forces in the vascular compartment are established modifiers of endothelial and smooth muscle cell cytoarchitecture and orchestrate complex genetic and biological responses in concert with contributions from the extracellular matrix (ECM), growth factors (e.g., EGF, and TGF-beta) and cell adhesion receptors (e.g., integrins, and cadherins). The profibrotic matricellular proteins plasminogen activator inhibitor-1 (PAI-1) and connective tissue growth factor (CTGF) are prominent members of a subset of genes the expression of which is highly responsive to cell shape-altering stimuli (i.e., disruption of the actin-based and microtubule networks, shear strain and cyclic stretch). Since both PAI-1 and CTGF are major mediators of cardiovascular fibrotic disease, understanding cell structure-linked signaling cascades provides potential avenues for focused therapy. It is increasingly evident that growth factor receptors (EGFR) are activated by changes in cytoarchitecture and that the "repressive state" of certain signaling proteins (e.g., SMAD, and Rho-GEFs) is maintained by sequestration on cell structural networks. Functional repression can be relieved by cytoskeletal perturbations (e.g., in response to treatment with network-specific drugs) resulting in activation of signaling cascades (e.g., Rho, and MAPK) with associated changes in gene reprogramming. Recent studies document a complex network of both similar and unique signaling control elements leading to the induction of PAI-1 and CTGF in response to modifications in cell shape. The purpose of this review is to highlight our current understanding of "cell deformation"-responsive signaling cascades focusing on the potential value of targeting such pathways, and their model response genes (e.g., PAI-1, and CTGF), as a therapeutic option for the treatment of fibrotic diseases.

Perturbation of the actin cytoskeleton induces PAI-1 gene expression in cultured epithelial cells independent of substrate anchorage

Perturbation of cellular architecture with agents that alter cytoskeletal organization provides a means to assess the relationship between cell shape and gene expression. Induced transcription of the plasminogen activator inhibitor type-1 (PAI-1) gene in serum-free cultures of normal rat kidney (NRK-52E) cells following disruption of actin microfilament structures with cytochalasin D (CD) provides a simple model to probe mechanisms underlying shape-related expression control. Transition from the typical flat epithelial cell shape to an "arborized" phenotype was a concomitant of the PAI-1 inductive response. Stimulated expression occurred rapidly (i.e., within 2 h of CD addition), involved increases in both PAI-1 mRNA abundance and de novo protein synthesis, and was dependent upon the concentration of CD used. A series of culture conditions were designed (e.g., use of bacteriological surfaces, poly-HEMA coated surfaces, maintenance in suspension on agarose) to discriminate cell shape from adhesive influences on CD-stimulated PAI-1 expression. Cytoskeletal disruption, and not simply changes in cell shape, was a critical aspect of CD-mediated PAI-1 expression in NRK cells cultured under serum-free conditions; induced expression was independent of substrate anchorage. Low concentrations of CD (1-2 microM) failed to cause cell arborization or increase either relative PAI-1 mRNA/protein abundance levels suggesting, however, that cell rounding may be a necessary but not sufficient aspect in CD-mediated PAI-1 induction. Transfection of PAI-1 promoter-CAT reporter constructs into NRK cells followed by stimulation with CD or serum additionally indicated that CD-induced PAI-1 expression did not utilize the same functional complement of serum-responsive promoter sequences, thus, further defining differences in the growth factor- and cytoskeletal-mediated pathways of PAI-1 gene regulation.

Cell shape-dependent pathway of plasminogen activator inhibitor type-1 gene expression requires cytoskeletal reorganization

Synthesis of plasminogen activator inhibitor type-1 (PAI-1), a major physiological modulator of plasmin generation, is regulated by growth factors and changes in cell shape. To evaluate the specific relationship between PAI-1 gene expression and cytoarchitecture, serum-free cultures of quiescent rat kidney (NRK) cells were exposed to cytochalasin D (CD) at concentrations that disrupt microfilament structure. Treatment with 1-10 microM CD resulted in an increased 1) incidence of rounded cells, 2) relative PAI-1 mRNA content, and 3) fraction of PAI-1 protein-expressing cells. Abrupt increases in each response were evident at a final concentration of 5 microM CD. Maximal levels of induced PAI-1 transcripts (18-fold that of control) occurred 4 hours post-CD addition and declined thereafter but remained elevated (by at least tenfold) for 24 hours. Assessment of the metabolic requirements for CD-induced PAI-1 expression by using the protein synthesis inhibitors puromycin and cycloheximide indicated that PAI-1 transcripts were regulated in a complex manner in response to CD. The predominant mode of induction reflected secondary (protein synthesis-dependent) metabolic processes, although a minor, albeit significant, primary (protein synthesis-independent) pathway was also evident. PAI-1 mRNA levels in NRK cells maintained in serum- and CD-free agarose suspension culture were low or undetectable. Relative abundance of PAI-1 transcripts in suspended cells cultured in the presence of CD, however, closely approximated that of plastic-adherent, CD-treated cells (13-fold over control). NRK cells in suspension culture with or without CD were morphologically identical, remaining spherical and unattached. It appears, therefore, that cell rounding alone is not a sufficient stimulus to induce PAI-1 expression in quiescent NRK cells and that perturbation of the actin skeleton as a consequence of CD treatment is a critical event in the inductive response. A protein tyrosine kinase is likely involved in the CD-mediated signal-transduction cascade, since induced PAI-1 expression can be down-regulated by genistein and herbimycin A but not by calphostin C or tyrphostin B46.

Host cell and malarial targets for docetaxel (Taxotere) during the erythrocytic development of Plasmodium falciparum

The microtubular stabilizing agent docetaxel (Taxotere) is known to inhibit the intraerythrocytic development of Plasmodium falciparum. To investigate the mechanism(s) of inhibition, we analyzed the structural organization of the mitotic spindle by immunofluorescence and electron microscopy. When 30 microM docetaxel was applied for five hours on ring forms, alterations in the mitotic spindles leading to abnormal nuclear divisions were observed. At the trophozoite- and schizont-stage, docetaxel pulses prevent mitosis by stabilizing microtubular structures associated with the mitotic apparatus, giving abnormal spindles. However, this inhibition did not interfere with parasite DNA synthesis indicating the absence of a checkpoint that couples exit from mitosis with proper spindle assembly as observed in higher eukaryotic cells. In parallel, intraerythrocytic concentration of docetaxel was measured in parasitized erythrocytes, after incubation of cells with 3H-docetaxel for five hours. It was found to be 14-fold increased at the ring-stage of infected erythrocytes compared to normal ones, 170-fold increased at the trophozoite-stage and 1,500-fold increased at the schizont-stage. Our data show that, even though the overall intracellular concentration of docetaxel is low in docetaxel-pulsed rings, the agent might be sufficient to disturb the spindle organization. However, the existence of targets for docetaxel other than mitotic spindle microtubules, i.e. erythrocyte membrane components could interfere with mitotic spindle formation.

The degradation of collagen in pig synovium in vitro and the effect of colchicine

Colchicine induced a rapid destruction of the collagenous matrix of pig synovial explants in culture in the presence of serum. The most efficacious doses were 0.01-0.1 micrograms/ml (2.5 x 10(-8) M - 2.5 x 10(-7) M). The histological progression of the tissue breakdown induced by colchicine was very similar, although faster, to that described for other agents (Fell et al., 1986), with cells having basophilic nuclei accumulating in areas of fibril degradation. The loss of collagen correlated with an increase in collagenase production and at the peak of resorption (6 to 8 days) active collagenase was present in the culture media. Immunocytochemical methods demonstrated active collagenase bound to collagen fibrils after only 4 days in culture, before significant collagen degradation could be observed histologically. Collagen breakdown was completely inhibited by cortisol, and partially inhibited by indomethacin: only the inhibition by indomethacin could be reversed by exogenous prostaglandin E2. Vinblastine at a higher dose was as effective as colchicine but the lumicolchicines, which do not disrupt microtubules, were ineffective. Although the precise mechanism of action of colchicine is unknown, this culture system provides a useful in vitro model for increasing our understanding of the cellular mechanisms of tissue breakdown and for elucidating the roles of active collagenase and related metalloproteinases.

Effect of colchicine on collagen synthesis by liver fibroblasts in murine schistosomiasis

Colchicine, an antimicrotubular agent, was shown to block the transcellular movement of certain structural macromolecules such as collagen. In the present study, the effect of colchicine on collagen synthesis and secretion by monolayer cultures of fibroblasts from livers of mice infected with Schistosoma mansoni was investigated. The effect of colchicine on proliferation of these fibroblasts was studied as well. Collagen and non-collagen protein synthesis was measured by incubating cultures with [14C]proline and measuring the incorporation of radioactivity into these protein fractions in both culture media and cell layers. Proliferation was measured by [3H]thymidine uptake. The isolated fibroblasts actively formed collagen and secreted most of it into the culture medium; 10-20% of the collagenase-sensitive radioactive protein remained in the cell layer. The addition of colchicine to culture medium led to selective inhibition of collagen formation with negligible effects on non-collagen protein synthesis. Fibroblast proliferation was also reduced by colchicine treatment. Both inhibition of collagen synthesis and inhibition of fibroblast proliferation were dose-dependent. Comparison of medium and cell layer collagen radioactivity confirmed inhibition of synthesis rather than only inhibition of secretion. These data suggest that colchicine has a specific effect on synthesis of collagen and proliferative activity by fibroblasts from S. mansoni-infected liver and may, therefore, be useful in modulating schistosomal hepatic fibrosis.

Imbalance of total cellular nucleotide pools and mechanism of the colchicine-induced cell activation

Treatment with colchicine or vinblastine, both inhibitors of microtubule assembly, renders quiescent 3T3 cells in an "activated state" as evidenced by induction of DNA synthesis and other criteria. Microtubule disassembly caused by colchicine or vinblastine brings about a dramatic expansion of total cellular UTP pools with a concomitant diminution in total cellular ATP pools, thus resulting in a marked imbalance in total cellular nucleotide pools. Colchicine and vinblastine also stimulate total cellular RNA synthesis without enhancing uridine phosphorylation, suggesting that these drugs affect the G1 phase of the cell cycle at a point beyond the enhancement of uridine phosphorylation that usually accompanies mitogenic stimulation of quiescent mammalian cells. The markedly expanded cellular UTP pools appear to be necessary for initiation of the colchicine-stimulated DNA synthesis because decreasing cellular UTP pools by addition of D-glucosamine results in a selective inhibition of DNA synthesis in the colchicine-stimulated, but not control, cells. Furthermore, D-glucosamine exerts its inhibitory effect only when it is present in the cultures within the first 14 hr after colchicine treatment. When added at 21 hr, D-glucosamine still decreases cellular UTP pools, but it is no longer inhibitory for DNA synthesis, which commences 14-16 hr after colchicine stimulation. Taxol, an antitumor drug, prevents microtubule disassembly and also blocks such events as expansion of total cellular UTP pools and stimulation of RNA and DNA synthesis, indicating that microtubule depolymerization acts as a primary event initiating the process of cell activation induced by colchicine.

Colchicine-induced modulation of collagenase in human skin fibroblast cultures. I. Stimulation of enzyme synthesis in normal cells

Microtubule-active agents affect the secretion of a variety of proteins, including collagenase. To gain insight into the mechanisms involved in this process, we examined the effects of colchicine on the synthesis, secretion, and activity of human skin collagenase. When added to monolayer cultures of human skin fibroblasts, 10(-6) M colchicine produced a mean 3-fold increase in trypsin-activatable collagenase in the culture medium. Stimulation was not observed with lumicolchicine. The enhanced accumulation of collagenase was dose-dependent with 10(-9), 10(-8), 10(-7), and 10(-6) M colchicine giving collagenase activities/mg protein that were 100 +/- 6%, 165 +/- 20%, 186 +/- 34%, and 297 +/- 62% of control, respectively. Although the effect on collagenase was seen under conditions independent of cellular growth (i.e., in serum-free medium), maximum stimulation occurred in subconfluent cultures. The colchicine-induced increase in activity was paralleled by an increase in immunoreactive enzyme protein, suggesting stimulation of enzyme synthesis. The catalytic efficiency of the enzyme (activity per unit immunoreactive protein) was unchanged, however, indicating that a structurally normal enzyme was being synthesized. To examine the process in more detail, the biosynthesis of 3H-labeled collagenase was quantitated in these cultures by specific immunoprecipitation. Although 10(-6) M colchicine produced no increase in total protein synthesis, an increased rate of collagenase synthesis was seen after only 1.5 hr. These data suggest that colchicine has a specific effect on the synthesis of collagenase and may be a useful probe for studying its regulation.