QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) All Versions of this Article: mol.105.019075v1 69/4/1207    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lu, H. Articles by Schwartz, E. L. Search for Related Content PubMed PubMed Citation Articles by Lu, H. Articles by Schwartz, E. L.

Original Article

The Microtubule Binding Drug Laulimalide Inhibits Vascular Endothelial Growth Factor-Induced Human Endothelial Cell Migration and Is Synergistic when Combined with Docetaxel (Taxotere)

Department of Oncology, Albert Einstein College of Medicine, Bronx, New York

Received September 21, 2005; accepted January 13, 2006

Abstract

Laulimalide, a natural product from marine sponges, is a microtubule-stabilizing agent that binds to tubulin at a site distinct from that of the taxoids. In the present study, we found that laulimalide inhibited human umbilical vein endothelial cell (HUVEC) tubule formation and vascular endothelial growth factor (VEGF)-induced HUVEC migration, key components of the angiogenic process. These occurred at concentrations substantially lower than that which inhibited HUVEC proliferation. When combined, laulimalide and docetaxel (Taxotere) synergistically inhibited migration and tubule formation, but their combined effect on proliferation was antagonistic. Possible mechanism(s) by which laulimalide inhibited VEGF-induced HUVEC migration were explored. Similar to docetaxel, laulimalide had no effect on the VEGF-induced tyrosine phosphorylation of the VEGF receptor Flk-1/KDR (VEGFR-2). Low concentrations of laulimalide substantially blocked subsequent VEGFR-2 downstream events, as did docetaxel, including the phosphorylation of the Tyr³⁹⁷ and Tyr⁴⁰⁷ residues of focal adhesion kinase (FAK), the association of VEGFR-2 with FAK and Hsp90, and the Tyr³¹ phosphorylation of paxillin. Laulimalide inhibited integrin activation; however, compared with docetaxel, it had a weaker inhibitory effect on the VEGF-induced association of VEGFR-2 with the ₅₁ integrin. Compared with docetaxel, laulimalide more potently caused a reduction in the constitutive levels (i.e., in the absence of VEGF) of phosphorylated paxillin and more potently inhibited the association of RhoA with the ₅₁ integrin. In conclusion, although both docetaxel and laulimalide inhibited integrin-associated signaling pathways that mediated VEGF-induced cell migration, their actions on the signaling cascade seemed not to be identical. These complementary actions could account for their synergistic effects on HUVEC.

Endothelial cell migration plays an essential role in angiogenesis, and it is mediated by focal adhesions, structures that connect the extracellular matrix (ECM) with the plasma membrane and the underlying actin cytoskeletal network (Stokes and Lauffenburger, 1991). ECM proteins serve as ligands for cell surface integrins, and the attachment of cells to the ECM results in the clustering of integrin receptors and initiates the recruitment of additional cytoplasmic proteins to the focal adhesion complex, including structural and catalytically active signaling proteins. Focal adhesions are dynamic structures, and their formation and breakdown are regulated by many different extracellular stimuli, including VEGF (Abedi and Zachary, 1997; Rousseau et al., 1997). Formation of these complexes has been shown to require the activity of the small Rho GTP-binding proteins (Soga et al., 2001; Zeng et al., 2002). VEGF is a multifunctional cytokine that stimulates endothelial cell proliferation and migration, increases microvascular permeability, and is required for tumorigenesis as well as angiogenesis (Millauer et al., 1993; Ferrara, 2002). The biological effects of VEGF on endothelial cells are mediated through the activation of its receptor tyrosine kinases, including Flt-1 (VEGFR-1) and Flk-1/KDR (VEGFR-2), with the latter responsible for VEGF-induced endothelial cell migration (Millauer et al., 1993; Waltenberger et al., 1994). Signaling events that follow VEGFR-2 phosphorylation include the activation of integrins, the phosphorylation of proteins involved in cell migration signaling including focal adhesion kinase (FAK), and the recruitment of growth factor receptors, integrins, and downstream effectors into newly formed focal adhesions (Avraham et al., 2003; Masson-Gadais et al., 2003; Le Boeuf et al., 2004).

Microtubule-binding agents such as colchicine, nocodazole, vinblastine, paclitaxel (Taxol), and docetaxel (Taxotere) have been shown to inhibit the migration of a range of cell types, including fibroblasts, monocytes, endothelial cells, and cell lines derived from lymphomas, melanomas, and prostate carcinomas (Goldman, 1971; Zakhireh and Malech, 1980; Stearns and Wang, 1992; Liao et al., 1995; Hotchkiss et al., 2002). Laulimalide, a natural product that was first isolated from Pacific marine sponges, is a microtubule-stabilizing agent that binds to the tubulin polymer at a site that is distinct from that of the taxoids (Mooberry et al., 1999; Pryor et al., 2002). Laulimalide has been shown to be active against tumor cells that were resistant to the taxanes, whether resistance was due to overexpression of P-glycoprotein or to mutations in the -tubulin gene (Mooberry et al., 1999). In the present study, we found that laulimalide, similar to docetaxel, inhibited VEGF-induced endothelial cell migration at concentrations lower than those required to inhibit endothelial cell proliferation or stabilize microtubules. The purpose of the present study was to identify potential mechanism(s) by which laulimalide inhibited endothelial cell migration, to compare these actions with those of docetaxel, and to evaluate the nature of the interaction of laulimalide and docetaxel on endothelial cells.

Materials and Methods

Reagents. Laulimalide was kindly provided by Scott Nelson and Billy Day, University of Pittsburgh (Pittsburgh, PA). Docetaxel was from Aventis Pharmaceuticals (Parsippany, NJ).

Cell Culture. Human umbilical vein endothelial cells (HUVEC) were purchased from GlycoTech, Inc. (Rockville, MD) by arrangement of the Angiogenesis Resources Branch of the Developmental Therapeutics Program (National Cancer Institute, Bethesda, MD). Culture medium consisted of MCDB131 with 2% FBS (Invitrogen, Carlsbad, CA), 10 ng/ml epidermal growth factor, 12 µg/ml endothelial cell growth supplement, 1 µg/ml hydrocortisone, 10 units/ml heparin, 2 mM L-glutamine, 5 units/ml penicillin G, and 5 µg/ml streptomycin sulfate (all from Sigma, St. Louis, MO).

Endothelial Cell Proliferation Assay. HUVEC (10⁴) were placed in each well of a 96-well plate. After allowing for attachment overnight, the laulimalide was added, and the number of cells after 6 days was determined by staining with sulforhodamine B (Sigma).

Assay of in Vitro Tubule Formation. The spontaneous formation of capillary-like structures by HUVEC on a basement membrane matrix preparation, Matrigel (BD Biosciences, San Jose, CA), was used to assess angiogenic potential. Twelve-well plates (Costar, Cambridge, MA) were coated with Matrigel (10 mg/ml) according to the manufacturer's instructions, and HUVEC (2 x 10⁵ cells/well) were seeded and incubated at 37°C for 60 min. Laulimalide was added, and in vitro tubule formation was photographed after the cultures were incubated at 37°C for 24 h. Quantitation was done by counting the number of tubules having branch points at both ends in three low-power (10x) fields, a sensitive descriptor that has been shown to measure endothelial cell reorganization into a capillary-like network (Guidolin et al., 2004).

Endothelial Cell Migration Assay. The migration assay used a modified Boyden chamber. Confluent HUVEC monolayers were cultured with nongrowth factor-containing media for 48 h before harvesting with Cell Dissociation solution (Sigma). Harvested cells were suspended at 10⁶/ml in M199 with 1% serum, and 10⁵ cells seeded into transwell inserts (8-µm pore; Costar) precoated with 10 µg/ml fibronectin. Inserts containing HUVEC were placed into a 24-well plate containing 700 µl of M199 with 1% serum and incubated for 1 h at 37°C. HUVEC migration was stimulated by addition of the VEGF (10 ng/ml) to the lower well of the Boyden chamber. The effects of laulimalide, docetaxel, or the combination of the two compounds on endothelial migration was observed by their addition to the lower chamber. After 5 h, HUVEC were stained with 10 µM Cell Tracker Green (Molecular Probes, Eugene, OR) for 30 min at 37°C, and the upper membrane of the insert was swabbed to remove nonmigrated cells. Inserts were washed with PBS, fixed in formaldehyde, and mounted on microscope slides. HUVEC migration was quantitated by counting the number of cells in five random fields (x100) per insert.

Some experiments used the CytoSelect 24-well cell migration assay kit (Cell Biolabs, San Diego, CA) to quantitate migration on the entire filter rather than by counting individual cells. In this assay, after swabbing to remove nonmigrated cells, the inserts were placed in cell staining solution for 10 min and then shaken with 200 µl of extraction solution. HUVEC migration was quantitated by measuring the optical density at 560 nm. Results from the two assays were comparable, and data presented are a combination of both.

Immunofluorescent Staining. HUVEC were seeded on fibronectin (10 µg/ml)-covered chamber slides. After 4 h of incubation in serum-free M199 medium, the cells were treated with serial-diluted laulimalide for 1 h and then stimulated with VEGF. The cells were fixed in 4% paraformaldehyde for 15 min, permeabilized with 0.5% Triton X-100 in PBS for 5 min at room temperature, and treated with blocking buffer (1% bovine serum albumin in PBS) for 1 h at room temperature. FAK (red) was visualized with a rabbit anti-FAK antibody (1:200; Santa Cruz Biotechnology, Santa Cruz, CA) and a Cy-3 conjugated anti-rabbit antibody. Tubulin (green) was visualized with an anti--tubulin monoclonal antibody (Sigma) and a fluorescein isothiocyanate-conjugated anti-mouse antibody. An engineered monovalent antibody that recognizes integrin _V3 in its high-affinity state (WOW-1 Fab) (Pampori et al., 1999; kindly provided by Dr. S. Shattil, Department of Medicine, University of California, San Diego) or a mouse anti-integrin _V3 antibody (Santa Cruz Biotechnology) was used to assess the effects of VEGF and laulimalide on integrin activation.

Immunoprecipitation and Western Blots. Confluent HUVEC, seeded on fibronectin (10 µg/ml)-covered dishes and starved overnight in M199 with 1% FBS, were treated with serial dilutions of laulimalide or docetaxel for 1 h. Cells were stimulated with VEGF (10 ng/ml) for the indicated times. Monolayers were washed twice in PBS and treated with immunoprecipitation lysis buffer [50 mM Tris, pH 7.4, 150 mM NaCl, 1% (v/v) Nonidet P-40, 1 mM phenylmethylsulfonyl fluoride, 1 mg/ml aprotinin, and 1 mg/ml leupeptin]. Cell suspensions were incubated on ice for 30 min and clarified by centrifugation. For immunoprecipitation, protein content was determined by the bicinchoninic acid method (Pierce), and 300 µg of total protein was incubated overnight at 4°C with protein G-agarose beads coated with saturating amounts of antibodies to integrin _V3 (Santa Cruz Biotechnology), integrin ₅₁ (Chemicon International), or VEGFR-2 (Santa Cruz). The resulting immune complexes were recovered after centrifugation by boiling for 3 min in SDS-PAGE loading buffer. For direct immunoblotting of whole cell lysates, cells were lysed in a modified radioimmunoprecipitation assay lysis buffer (Biosource International, Camarillo, CA).

For immunoblotting, aliquots of whole-cell lysates (30 µg) or isolated immunocomplexes were separated by SDS-PAGE under reducing conditions using 10% polyacrylamide gels. Proteins were transferred onto polyvinylidene difluoride membrane and analyzed by immunoblotting using antibodies against VEGFR-2, FAK, paxillin, integrin ₁ (all from Santa Cruz), RhoA, VEGFR-2 phosphorylated on Tyr⁹⁵¹, FAK phosphorylated on various tyrosine residues (Tyr³⁹⁷, Tyr⁴⁰⁷, Tyr⁵⁷⁶, Tyr⁵⁷⁷, and Tyr⁸⁶¹), and paxillin phosphorylated on Tyr³¹ (all from Biosource International). Antibodies against total FAK, paxillin, and -actin were used as control for loading when using whole-cell lysates, whereas antibodies against integrins ₁ and VEGFR-2 were used to quantitate protein loading for the immunoprecipitation reactions.

Drug Combination Effect Analyses. A computerized version of the combined effects method was used to analyze the nature of the interaction of laulimalide and docetaxel on HUVEC migration, tubule formation, and proliferation. Details of the method and its statistical basis have been described previously (Chou and Talalay, 1984; Bible and Kaufmann, 1997). For these experiments, the two agents were used either alone or together at several concentrations, in all cases at fixed molar ratios based on their IC₅₀ values. The ratio used in the migration and tubule assays was 1:1 (docetaxel/laulimalide), whereas that in the proliferation assay was 1:6.25. Calcusyn software (Biosoft, Cambridge, UK) was then used to calculate the combination index (CI) for each concentration tested, whereby CI values less than 1 indicated synergy, equal to 1 indicated additivity, and those greater than 1 indicating antagonism in the interaction of the drugs.

Results

Laulimalide and Docetaxel Synergistically Inhibited HUVEC Migration and the Formation of Capillary-like Structures at Low Concentrations that Did Not Inhibit HUVEC Proliferation. Sprouting angiogenesis encompasses successive phases of microvessel formation, neovessel growth, and neovessel stabilization (Vailhé et al., 2001). Steps in the neovessel growth process include the migration of endothelial cells from the parent vessel toward an angiogenic factor, proliferation of endothelial cells behind the front of migration, and the organization of the endothelial cells into capillary-like structures. The creation of capillary-like structures in vivo and in vitro involves the remodeling and pruning (via apoptosis) of the endothelial cells, the formation of lumens, and the formation of loops by anastomoses (Vailhé et al., 2001). These multistep processes can be recapitulated with in vitro assays, which were used in this study to evaluate the effects of laulimalide and docetaxel. HUVEC migration in a modified Boyden chamber is a chemotactic model of migration representative of tumor-induced endothelial cell migration. Stimulation of HUVEC along a directional gradient of VEGF resulted in migration to the underside of the membrane, and this was inhibited in a concentration-dependent manner by laulimalide with an IC₅₀ of approximately 0.01 nM, comparable with that of docetaxel (Fig. 1A).

View larger version (29K): [in this window] [in a new window]   Fig. 1. Inhibition of endothelial cell migration, capillary-like tubule formation, and proliferation by laulimalide and docetaxel. A, for cell migration, HUVEC were starved in M199 media lacking growth supplements for 18 h and then were harvested and seeded into transwell inserts (8-µm pore) precoated with 10 µg/ml fibronectin. Inserts containing HUVEC were placed in a 24-well plate containing M199 with 1% serum and incubated for 1 h at 37°C. Docetaxel () or laulimalide () was added to the upper chamber for 1 h, and HUVEC migration was then initiated by the addition of VEGF (10 ng/ml) to the lower chamber. After an additional 5 h, migration of HUVEC to the underside of the transwell insert was quantitated as described under Materials and Methods. Data are from three experiments and are expressed relative to the VEGF-alone control. B, for tubule formation in vitro, HUVEC (2 x 105/well) were seeded onto Matrigel (10 mg/ml)-coated 12-well plates and incubated at 37°C for 60 min. The indicated concentrations of docetaxel () or laulimalide () were added, and cells were incubated for an additional 24 h. Endothelial tubules were photographed and quantitated; only intact tubules with branch points at both ends were counted as positive. Data are from three experiments, and are expressed relative to the vehicle (DMSO)-treated control. C, for cell proliferation, HUVEC cells were placed into a 96-well plate, and after allowing for cell attachment overnight, the indicated concentrations of docetaxel () or laulimalide () were added. The number of cells was determined after 3 days by staining with sulforhodamine B. Data are means of four experiments and are expressed as a percentage of vehicle-control. D, E, and F, combination effects of laulimalide and docetaxel were determined for cell migration (D), tubule formation (E), and proliferation (F). Each assay was done as described in A–C and data for the two drugs, used individually and in combination, obtained. Drugs were combined at equipotent concentrations, which for migration and tubule formation was a molar ratio of 1:1 and for proliferation was a molar ratio of 1:6.25 (docetaxel/laulimalide). Drug effects are calculated as the combination index and are expressed as a function of the fraction affected in the combination treatment. CI values less than 1 indicate that the drug combination had a synergistic interaction, values at or near 1 indicate the combination was additive, and values greater than 1 indicate an antagonistic interaction. Migration data are means of eight experiments, tubule formation are the means of two experiments, and proliferation data are the means of four experiments.

View larger version (156K): [in this window] [in a new window]   Fig. 2. Inhibition of endothelial cell capillary-like tubule formation by laulimalide and docetaxel. Tubule formation on Matrigel was done as described in the legend to Fig. 1B. Representative low-power magnification pictures are shown for HUVEC treated with vehicle control (A), laulimalide at concentrations of 0.01 nM (B), 0.1 nM (C), and 1 nM (D), docetaxel at 0.1 nM (E), or a combination of laulimalide and docetaxel, each at 0.1 nM (F).

Laulimalide Inhibited VEGF-Induced Cell Polarization, Focal Adhesion Formation, Radial Microtubule Growth, and Integrin _V3 Activation. Cell migration is a coordinated process consisting of adhesion at the leading edge and detachment at the rear of the cell. Adhesion to ECM involves structures heterogeneous with respect to size, composition, and orientation to actin filaments, the largest and tightest structures of which are focal adhesions. Focal adhesions link the actin cytoskeleton to the ECM by integrin receptor complexes. These processes can be readily visualized in endothelial cells as VEGF and other chemotactic factors initiate a series of morphologic changes. HUVEC were plated on fibronectin-coated coverslips, stimulated with VEGF in the absence or presence of laulimalide for 1 h, and then fixed and labeled for both tubulin (green) and FAK (red) (Fig. 3). Control endothelial cells predominantly had a typical polygonal shape with a relatively uniform distribution of FAK within the cytoplasm and on the cell surface and a poorly visible microtubule skeleton. As can be seen in Fig. 3, VEGF induced cell spreading, flattening, and polarization and the formation of membrane protrusions, precursors to nascent lamellipodia. The size and number of FAK-containing focal adhesions increased in VEGF-treated cells, and FAK and the focal contacts appeared to redistribute to the protruding edges of the cell membranes. The microtubules became more visible and grew radially toward the expanding regions of the cell periphery. All of the VEGF-induced changes were inhibited in a concentration-dependent manner by laulimalide, with substantial inhibition seen at 0.1 nM and complete inhibition at 1 nM. It was only at the highest concentration tested, 10 nM, at which both HUVEC migration and proliferation were inhibited, that the cells acquired the rounded morphology and condensed microtubules typically seen at cytotoxic concentrations.

View larger version (39K): [in this window] [in a new window]   Fig. 3. Laulimalide inhibited VEGF-induced endothelial cell polarization, assembly of focal contacts into focal adhesions, and activation of integrin V3. For the staining of focal adhesions, HUVEC were seeded on fibronectin, starved in M199 for 4 h, treated with laulimalide for 1 h at the indicated concentrations, and stimulated with VEGF (10 ng/ml) for 4 h. FAK (red) was visualized with a rabbit anti-FAK antibody (1:200; Santa Cruz) and a Cy-3 conjugated anti-rabbit antibody. Tubulin (green) was visualized with an anti--tubulin monoclonal antibody (Sigma) and a fluorescein isothiocyanate-conjugated anti-mouse antibody. For the staining of integrins, HUVEC were seeded on fibronectin and starved in Dulbecco's modified Eagle's medium for 6 h, treated with laulimalide for 1 h, and stimulated with VEGF (50 ng/ml) for 10 min. "Activated" integrin V3 was determined by staining with an engineered antibody (WOW-1 Fab) that recognizes integrin V3 in its high-affinity state; total integrin was stained using a monoclonal V3 antibody. The results are representative of two independent experiments.

Integrins mediate cell adhesion and signaling and therefore play a critical role in cell migration. Their primary function is ligand binding, and this is regulated in part by conformational changes in the heterodimers that increase integrin affinity for their ECM ligands and thereby cause integrin activation. Previous work showed that VEGF-stimulated HUVEC migration is mediated through two distinct integrins, _V3 and ₅₁, and VEGF stimulates the recruitment of activated, high-affinity integrins to the leading edge of migrating endothelial cells (Byzova et al., 2000; Hotchkiss et al., 2003). To investigate whether laulimalide was capable of suppressing integrin activation, we used WOW-1, an engineered, monovalent, ligand-mimetic Fab fragment that reacts selectively with _V3 when it is activated and in a high affinity state (Pampori et al., 1999). As illustrated in Fig. 3, WOW-1 staining in control cells exhibited a nuclear distribution, and this was most likely to be due to nonspecific interactions. VEGF treatment resulted in _V3 activation, as judged by the more intense WOW-1 staining including around the cell periphery. Incubation of the cells with laulimalide blocked the VEGF-induced integrin _V3 activation, and as before, partial inhibition was observed at 0.1 nM, and complete inhibition at 1 nM laulimalide. In contrast, neither VEGF nor laulimalide had any effect on the levels or distribution of total _V3 integrins (Fig. 3).

Laulimalide Had No Effect on VEGF-Induced VEGFR-2 Phosphorylation but Did Inhibit FAK and Paxillin Phosphorylation. The cellular actions of VEGF are triggered by its binding to its cognate receptors, with VEGFR-2 playing the predominant role in endothelial cell migration. Thus, tyrosine phosphorylation of VEGFR-2 represents one of the first events in the signaling cascade whereby VEGF induces cell migration. To determine whether VEGFR-2 activation was affected by laulimalide, HUVEC were seeded on fibronectin, starved in serum-free medium overnight, treated with laulimalide for 1 h, and stimulated with VEGF (10 ng/ml) for 5 min. Using an antibody that was specific for Tyr⁹⁵¹-phosphorylated VEGFR-2, we found that laulimalide had no effect on VEGF-induced receptor phosphorylation compared with cells stimulated with VEGF alone (Fig. 4A). Identical results were obtained when total VEGFR-2 was immunoprecipitated and then probed with an antiphosphotyrosine antibody (not shown).

View larger version (46K): [in this window] [in a new window]   Fig. 4. Laulimalide did not affect VEGFR-2 tyrosine phosphorylation but did inhibit the association of VEGFR-2 with Hsp90 and FAK and the tyrosine phosphorylation of FAK and paxillin. Confluent HUVEC seeded on fibronectin (10 µg/ml) were starved overnight before being treated with docetaxel or laulimalide for 1 h at the indicated concentrations. Cells were then stimulated with VEGF (10 ng/ml) for 5 min (A), 10 min (B), or 30 min (C and D), the times having been determined in pilot studies to produce maximal phosphorylation of each respective protein; control cells (-) were not treated with VEGF or drug. Changes in the levels of phosphorylated Tyr951 (pY951) of VEGFR-2 (A), Tyr397, Tyr407, Tyr576, Tyr577, and Tyr861 of FAK (B), and Tyr31 of paxillin (D) were visualized by immunoblots using phosphorylation-specific antibodies. Total FAK and paxillin levels were also determined. For immunoprecipitations (C), cell lysates from control, VEGF, and laulimalide+VEGF-treated cells were immunoprecipitated (IP) with an antibody to VEGFR-2, and then analyzed by SDS-PAGE. Changes in the levels of Hsp90 and FAK, associated with VEGFR-2, were visualized by probing with the indicated antibodies. Total VEGFR-2 levels were also determined.

Integrins have short cytoplasmic tails that lack catalytic activity, and thus their ability to transmit outside-in cell signaling is mediated by cytoplasmic proteins that localize to sites of clustered and activated integrins. One such protein is FAK, a 125-kDa cytoplasmic tyrosine kinase that is localized to focal adhesions where it acts to integrate growth factor and integrin signals (Hanks et al., 1992). FAK becomes phosphorylated at six different tyrosine residues after the engagement of integrins with ECM proteins, and these phosphorylations are important for cell migration (Sieg et al., 2000). Because laulimalide inhibited VEGF-induced HUVEC migration and focal adhesion formations, we next examined its effect on FAK phosphorylation using phosphorylation site-specific antibodies. Phosphorylation of FAK on Tyr³⁹⁷ (the autophosphorylation site) occurred rapidly upon VEGF stimulation, reaching a maximum at 5 min and declining to basal level by 30 min (not shown). There were smaller increases in VEGF-induced FAK phosphorylation at tyrosine residues 407, 576, 577, and 861 (Fig. 4B). Laulimalide most prominently inhibited phosphorylation of FAK on Tyr³⁹⁷, with a substantial decrease observed at 0.01 nM and a reduction to near baseline levels at 0.1 nM laulimalide (Fig. 4B). It also blocked VEGF-induced phosphorylation at Tyr⁴⁰⁷, although this site seemed to be somewhat less sensitive to inhibition by laulimalide. Laulimalide had no effect on the more modest increases in FAK phosphorylation at tyrosines 576 and 861.

FAK is required for the downstream signaling from VEGFR-2 in endothelial cells, and its actions are mediated by the heat shock protein Hsp90, which acts as a bridging protein to stabilize the physical association of FAK and VEGFR-2 (Le Boeuf et al., 2004). The interaction of Hsp90 and FAK with VEGFR-2 was lost in cells treated with laulimalide (Fig. 4C), and this action of the drug probably prevents signal transduction from the VEGF receptor and could account for its inhibitory effects. FAK also plays a key role in the dynamic reorganization of the cytoskeletal network that precedes cell migration. Its phosphorylation provides sites for the interaction with a number of focal adhesion-associated proteins, and we next examined one of these, paxillin. Paxillin is phosphorylated in VEGF-treated cells on Tyr³¹, and this occurred maximally at 30 min after VEGF addition, indicating, as expected, that it occurred downstream and probably as a consequence of FAK phosphorylation (not shown). The phosphorylation of paxillin is dependent upon FAK phosphorylation, and because the latter was inhibited by laulimalide, we expected that paxillin phosphorylation would likewise be inhibited. This can be seen in Fig. 4D, where the increase in paxillin phosphorylation was blocked by low concentrations of laulimalide.

Laulimalide Suppressed the Basal Tyrosine Phosphorylation Level of Paxillin. There was a low level of tyrosine phosphorylation of both FAK and paxillin observed under control conditions (in the absence of VEGF), and our investigations of the effect of laulimalide suggested that the drug could be also reducing this basal level, in addition to blocking the VEGF-induced effect. Evaluation of this possible drug action could be done using a longer exposure of the immunoblots. As shown in Fig. 5, laulimalide prominently inhibited the basal phosphorylation of paxillin on Tyr³¹, with a substantial decrease observed at 0.01 nM. Docetaxel showed a substantially weaker effect, with only a modest decrease observed at 1 nM (Fig. 5). Neither laulimalide nor docetaxel affected the basal tyrosine phosphorylation level of the autophosphorylation site of FAK (Tyr³⁹⁷) (not shown).

View larger version (36K): [in this window] [in a new window]   Fig. 5. Laulimalide inhibited basal levels (in the absence of VEGF) of tyrosine phosphorylation of paxillin. HUVEC were seeded on fibronectin (10 µg/ml) and starved overnight in M199 (0.1% FBS) before treatment with docetaxel or laulimalide for 1 h at the indicated concentrations. Changes in paxillin phosphorylation were visualized by immunoblotting with an antibody to phosphotyrosine residue 31. Protein loading was evaluated by reprobing the membranes for total paxillin and actin.

Association of Integrin ₅₁ with Signaling Pathways: Docetaxel but Not Laulimalide Blocked Association with VEGFR-2, Whereas Laulimalide but Not Docetaxel Blocked Association with RhoA.

View larger version (63K): [in this window] [in a new window]   Fig. 6. Differential effects of laulimalide and docetaxel: the association of VEGFR-2 with integrin 51 was blocked by docetaxel but not laulimalide, whereas the association of RhoA with integrin 51 was blocked by laulimalide but not by docetaxel. Confluent HUVEC were seeded on fibronectin (10 µg/ml), starved overnight, and treated with the indicated concentrations of docetaxel (Taxotere) or laulimalide for 1 h. Cells were then stimulated with VEGF (10 ng/ml) for 30 min. Cell lysates from control cells and cells treated with VEGF, docetaxel+VEGF, and laulimalide+VEGF were immunoprecipitated (IP) with antibodies to integrin 51 or a control IgG as indicated and then analyzed by SDS-PAGE. Changes in the levels of VEGFR-2 (A) and RhoA (B) associated with integrin 51 were visualized by probing with the indicated antibodies. Total integrin 1 levels and RhoA and actin levels (C) were also determined.

The Rho family of small GTPases is involved in the regulation of several components of cell migration, including the development of cell polarization, the assembly of focal adhesions, the formation of directional cell protrusions, and the rapid reorganization of actin filaments. They also regulate the local stabilization of microtubules at the leading edge of migrating cells in a process that requires integrin-mediated activation of FAK (Palazzo et al., 2004). VEGF induced the activation of RhoA and its recruitment to the cell membrane of endothelial cells, and RhoA has been shown to be required for VEGFR-2-mediated endothelial cell migration (Zeng et al., 2002; van Nieuw Amerongen et al., 2003). We examined the effect of laulimalide and docetaxel on the association of RhoA with integrins, and as shown in Fig. 6B, found that laulimalide strongly blocked the association of RhoA with integrin ₅₁, an effect that was apparent at 0.01 nM laulimalide. Total RhoA levels in the cell lysates were not affected (Fig. 6C). As in its effects on paxillin, the actions of laulimalide on RhoA could be distinguished from those of docetaxel, which only weakly reduced the RhoA-integrin association.

Discussion

Angiogenesis, the process of sprouting of capillaries from preexisting blood vessels, plays an important role in the process of tumor growth and metastasis. Endothelial cell activation, migration, proliferation, and differentiation are major cellular events in this process. Microtubule-interfering agents were among the first of the cytotoxic chemotherapeutic drugs to be reported to have an antiangiogenic effect, and most members of this class of agents have been shown to have this activity (Goldman, 1971; Zakhireh and Malech, 1980; Liao et al., 1995; Belotti et al., 1996; Hotchkiss et al., 2002). Given the multiple roles that microtubules play in cell migration, there are potentially several sites and mechanisms by which microtubule-disrupting compounds could be acting to block cell motility. Some of the earlier studies, in which the high concentrations of these agents used caused near complete microtubule breakdown, are probably not relevant to the effects of microtubule-stabilizing drugs observed in this report or to the concentrations of these drugs which occur clinically. Rather, we have found inhibition of endothelial cell migration in vitro at low concentrations that did not affect gross microtubule structure. These observations suggested that the drugs are either having more subtle effects on microtubule plasticity and dynamics, or that they were acting at sites distinct from, or in addition to, their effects on microtubules.

A mechanism-based screening of marine sponges led to the identification of a macrocyclic lactone, laulimalide, with microtubule stabilizing properties remarkably similar to those of the taxanes (Mooberry et al., 1999; Gapud et al., 2004). Consistent with this observation, in this report we found that laulimalide inhibited human endothelial cell migration and tubule formation with a potency equal to that of docetaxel. Despite these similarities, previous studies found that laulimalide does not bind to the taxoid site on -tubulin, and when used at suboptimal concentrations, acted synergistically with paclitaxel in inducing tubulin assembly in cell-free systems (Pryor et al., 2002; Gapud et al., 2004). Computational analyses suggested that although laulimalide could bind to the taxoid site, its preferred binding site on the -tubulin dimer was over a mobile region of -tubulin, the B9-B10 loop extension, a site distinct from that which binds the taxanes and colchicine (Pineda et al., 2004). This explains why laulimalide was active against tumor cells made resistant to the taxanes by mutations in the -tubulin gene, because the amino acids involved in the mutations were not predicted to interact with laulimalide according to this model (Pineda et al., 2004). Laulimalide was also active in cells that were resistant to the taxanes because of overexpression of P-glycoprotein, suggesting that there were significant differences in the nature of the interaction of laulimalide with P-glycoprotein as well, compared with the taxanes. Our studies of the molecular actions of laulimalide suggested that although it might share, with the taxanes, similar target protein(s) for the inhibition of cell migration, it also interacts with these proteins in a manner that differs from that of the taxanes. It is presumed that it is these differences that contribute to the synergistic inhibition of migration and tubule formation by the combination of laulimalide and docetaxel found in this study.

Inhibition of HUVEC proliferation by laulimalide occurred with an IC₅₀ of approximately 4 nM, which was similar to its IC₅₀ for the inhibition of proliferation of several tumor cell lines (Mooberry et al., 1999). More interesting was the observation that tubule formation and VEGF-induced migration of HUVEC was shown to be reduced by laulimalide at low subnanomolar concentrations, suggesting that laulimalide and related compounds could have therapeutically relevant antiangiogenic actions. In addition to being a potent inhibitor of VEGF's effects, these actions will probably apply to effects on endothelial cells that are mediated by other angiogenic factors, as laulimalide inhibited paxillin phosphorylation and tubule formation in the absence of VEGF. The data obtained demonstrated that laulimalide inhibited signal transduction pathways involved in the initiation of endothelial cell migration. Inhibition probably occurred at a point subsequent to the binding of VEGF and the activation of the VEGF receptor, as laulimalide did not affect VERGFR-2 tyrosine phosphorylation. Given the complexity and interconnected nature of these downstream signaling events, it is not possible to precisely pinpoint the site of action of laulimalide, although the effect of laulimalide to block the association of VEGFR-2 with Hsp90 and FAK is an early event and thus is a likely site of action (Rousseau et al., 2000; Le Boeuf et al., 2004).

Our data demonstrated that laulimalide, similar to docetaxel, suppressed the activation of integrins upon treatment with VEGF. Integrins, which are specific protein receptors for ECM, are activated as a result of an increase in their binding affinity and by their clustering at the cell surface (avidity), and the inhibition of their activation by laulimalide has consequences for two related pathways (Parsons et al., 2000). Data suggest that the VEGF receptor forms complexes with the integrins in a manner that promotes reciprocal cross-talk and cooperative actions of the VEGF- and ECM-receptors to enhance their respective cellular responses (Soldi et al., 1999; Byzova et al., 2000; Wijelath et al., 2002). In contrast to the effect of docetaxel, however, laulimalide did not block the association of VEGFR-2 with the integrins, suggesting that this cross-talk was not a likely site of laulimalide action. Activation of integrins also leads to the subsequent activation, via protein phosphorylation, of FAK and other focal adhesion-associated proteins, including paxillin, c-src, and p130^CAS (Parsons et al., 2000). FAK does not have intrinsic enzyme activity but rather serves a scaffold protein for the assembly of other regulatory molecules. Thus, it was anticipated that the inhibition of VEGF-induced FAK activation by laulimalide would also lead to the inhibition of downstream FAK-dependent proteins, and this was found to be the case for the FAK-binding protein paxillin. That laulimalide also potently inhibited the basal level of paxillin activation was unexpected, however, and suggested that laulimalide could have an additional and independent action on paxillin or on a paxillin-related pathway. This action also provides an indication as to how laulimalide and related compounds might inhibit migration that is not initiated by VEGF or by the activation of other receptor tyrosine kinases.

Studies have implicated Rho family members in the cross-talk between microtubules and actin that is required for the regulation of cell motility (Waterman-Storer et al., 1999). During migration, microtubules emanate out from the center of the cell toward the leading edge, where their plus ends exhibit dynamic instability (random changes between periods of growth and shortening). The dynamic instability of the microtubules can activate signal-transduction cascades at the leading edge of the cell, with depolymerization causing an increase in the level of GTP-bound RhoA, whereas the polymerization resulting in activation of the related GTPase Rac1 (Waterman-Storer et al., 1999). Ultimately, there is a localized stabilization of the microtubules at the leading edge of cells undergoing migration, a process that has been shown to be FAK- and Rho-dependent (Palazzo et al., 2004). Thus, it is possible that by stabilizing microtubules in general or at specific sites within the cell, laulimalide and related agents impair the development of polarization in the cell, a process that is required for their directional motility. The ability of laulimalide to prevent integrin and FAK activation also would be anticipated to affect the Rho-mediated signal pathways, as Rho signaling is facilitated by FAK (Palazzo et al., 2004). On the other hand, if one action of laulimalide is to block the activation of members of the Rho family via an effect on the microtubules, then this could explain the effect of laulimalide on FAK and the integrins. RhoA kinase activation has been reported to be involved in the activation of a tyrosine kinase that is responsible for phosphorylating Tyr⁴⁰⁷ of FAK (Hanks and Polte, 1997), whereas Rac has been shown to recruit high affinity integrins to the lamellipodia in endothelial cells (Kiosses et al., 2001).

In summary, laulimalide and docetaxel had several effects on cell signaling that might contribute to their inhibitory effects on endothelial cell migration and tubule formation. These molecular effects probably represent novel sites of action of microtubule-binding drugs, distinct from their well documented effects on mitosis, apoptosis, and cell proliferation. The synergistic inhibitory effects of laulimalide and docetaxel on endothelial cell migration may be due to their differing and/or complementary actions on signaling pathways participating in endothelial cell migration.

Footnotes

This work was supported by National Cancer Institute grant CA98456.

Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org.

doi:10.1124/mol.105.019075.

ABBREVIATIONS: ECM, extracellular matrix; VEGF, vascular endothelial growth factor; VEGFR, vascular endothelial growth factor receptor; FAK, focal adhesion kinase; HUVEC, human umbilical vein endothelial cell(s); M199, medium 199; FBS, fetal bovine serum; PBS, phosphate-buffered saline; PAGE, polyacrylamide gel electrophoresis; CI, combination index.

Address correspondence to: Edward L. Schwartz, Department of Oncology, Albert Einstein College of Medicine, Montefiore Medical Center, 111 East 210th Street, Bronx, NY 10467. E-mail: eschwart{at}aecom.yu.edu

References

Abedi H and Zachary I (1997) Vascular endothelial growth factor stimulates tyrosine phosphorylation and recruitment to new focal adhesions of focal adhesion kinase and paxillin in endothelial cells. J Biol Chem 272: 15442-15451.[Abstract/Free Full Text]

Avraham HK, Lee TH, Koh Y, Kim TA, Jiang S, Sussman M, Samarel AM, and Avraham S (2003) Vascular endothelial growth factor regulates focal adhesion assembly in human brain microvascular endothelial cells through activation of the focal adhesion kinase and related adhesion focal tyrosine kinase. J Biol Chem 278: 36661-36668.[Abstract/Free Full Text]

Belotti D, Vergani V, Drudis T, Borsotti P, Pitelli MR, Viale G, Giavazzi R, and Taraboletti G (1996) The microtubule-affecting drug paclitaxel has antiangiogenic activity. Clin Cancer Res 2: 1843-1849.[Abstract]

Bible KC and Kaufmann SH (1997) Cytotoxic synergy between flavopiridol (NSC 64980, L86–8275) and various antineoplastic agents: the importance of sequence of administration. Cancer Res 57: 3375-3380.[Abstract/Free Full Text]

Byzova TV, Goldman CK, Pampori N, Thomas KA, Bett A, Shattil SJ, and Plow EF (2000) A mechanism for modulation of cellular responses to VEGF: activation of the integrins. Mol Cell 6: 851-860.[Medline]

Chou TC and Talalay P (1984) Quantitative analysis of dose-effect relationships: the combined effects of multiple drugs or enzyme inhibitors. Adv Enzyme Regul 22: 27-55.[CrossRef][Medline]

Ferrara N (2002) Role of vascular endothelial growth factor in physiologic and pathologic angiogenesis: therapeutic implications. Semin Oncol 29: 10-14.[Medline]

Gapud EJ, Bai R, Ghosh AK, and Hamel E (2004) Laulimalide and paclitaxel: a comparison of their effects on tubulin assembly and their synergistic action when present simultaneously. Mol Pharmacol 66: 113-121.[Abstract/Free Full Text]

Goldman RD (1971) The role of three cytoplasmic fibers in BHK-21 cell motility. I. Microtubules and the effects of colchicine. J Cell Biol 51: 752-762.[Abstract/Free Full Text]

Guidolin D, Vacca A, Nussdorfer GG, and Ribatti D (2004) A new image analysis method based on topological and fractal parameters to evaluate the angiostatic activity of docetaxel by using the Matrigel assay in vitro. Microvasc Res 67: 117-124.[CrossRef][Medline]

Hamm-Alvarez SF, Alayof BE, Himmel HM, Kim PY, Crews AL, Strauss HC, and Sheetz MP (1994) Coordinate depression of bradykinin receptor recycling and microtubule-dependent transport by taxol. Proc Natl Acad Sci USA 91: 7812-7816.[Abstract/Free Full Text]

Hanks SK, Calalb MB, Harper MC, and Patel SK (1992) Focal adhesion protein-tyrosine kinase phosphorylated in response to cell attachment to fibronectin. Proc Natl Acad Sci USA 89: 8487-8491.[Abstract/Free Full Text]

Hanks SK and Polte TR (1997) Signaling through focal adhesion kinase. Bioessays 19: 137-145.[CrossRef][Medline]

Hotchkiss KA, Ashton AW, Mahmood R, Russell RG, Sparano JA, and Schwartz EL (2002) Inhibition of endothelial cell function in vitro and angiogenesis in vivo by docetaxel (Taxotere): association with impaired repositioning of the microtubule organizing center. Mol Cancer Ther 1: 1191-1200.[Abstract/Free Full Text]

Hotchkiss KA, Ashton AW, and Schwartz EL (2003) Thymidine phosphorylase and 2-deoxyribose stimulate human endothelial cell migration by specific activation of the integrins ₅₁ and _V3. J Biol Chem 278: 19272-19279.[Abstract/Free Full Text]

Kaverina I, Rottner K, and Small JV (1998) Targeting, capture and stabilization of microtubules at early focal adhesions. J Cell Biol 142: 181-190.[Abstract/Free Full Text]

Kiosses WB, Shattil SJ, Pampori N, and Schwartz MA (2001) Rac recruits high-affinity integrin alphavbeta3 to lamellipodia in endothelial cell migration. Nat Cell Biol 3: 316-320.[CrossRef][Medline]

Le Boeuf F, Houle F, and Huot J (2004) Regulation of vascular endothelial growth factor receptor 2-mediated phosphorylation of focal adhesion kinase by heat shock protein 90 and Src kinase activities. J Biol Chem 279: 39175-39185.[Abstract/Free Full Text]

Liao G, Nagasaki T, and Gundersen GG (1995) Low concentrations of nocodazole interfere with fibroblast locomotion without significantly affecting microtubule level: implications for the role of dynamic microtubules in cell locomotion. J Cell Sci 108: 3473-3483.[Abstract]

Masson-Gadais B, Houle F, Laferriere J, and Huot J (2003) Integrin alphavbeta3, requirement for VEGFR2-mediated activation of SAPK2/p38 and for Hsp90-dependent phosphorylation of focal adhesion kinase in endothelial cells activated by VEGF. Cell Stress Chaperones 8: 37-52.[CrossRef][Medline]

Millauer B, Wizigmann-Voos S, Schnurch H, Martinez R, Moller NP, Risau W, and Ullrich A (1993) High affinity VEGF binding and developmental expression suggest Flk-1 as a major regulator of vasculogenesis and angiogenesis. Cell 72: 835-846.[CrossRef][Medline]

Mooberry SL, Tien G, Hernandez AH, Plubrukarn A, and Davidson BS (1999) Laulimalide and isolaulimalide, new paclitaxel-like microtubule-stabilizing agents. Cancer Res 59: 653-660.[Abstract/Free Full Text]

Palazzo AF, Eng CH, Schlaepfer DD, Marcantonio EE, and Gundersen GG (2004) Localized stabilization of microtubules by integrin- and FAK-facilitated Rho signaling. Science (Wash DC) 203: 836-839.

Pampori N, Hato T, Stupack DG, Aidoudi S, Cheresh DA, Nemerow GR, and Shattil SJ (1999) Mechanisms and consequences of affinity modulation of integrin _V3 detected with a novel patch-engineered monovalent ligand. J Biol Chem 274: 21609-21616.[Abstract/Free Full Text]

Parsons JT, Martin KH, Slack JK, Taylor JM, and Weed SA (2000) Focal adhesion kinase: a regulator of focal adhesion dynamics and cell movement. Oncogene 19: 5606-5613.[CrossRef][Medline]

Pineda O, Farras J, Maccari L, Manetti F, Botta M, and Vilarrasa J (2004) Computational comparison of microtubule-stabilising agents laulimalide and peloruside with taxol and colchicine. Bioorg Med Chem Lett 14: 4825-4829.[CrossRef][Medline]

Pryor DE, O'Brate A, Bilcer G, Diaz JF, Wang Y, Kabaki M, Jung MK, Andreu JM, Ghosh AK, Giannakakou P, et al. (2002) The microtubule stabilizing agent laulimalide does not bind in the taxoid site, kills cells resistant to paclitaxel and epothilones and may not require its epoxide moiety for activity. Biochemistry 41: 9109-9115.[CrossRef][Medline]

Rousseau S, Houle F, Kotanides H, Witte L, Waltenberger J, Landry J, and Huot J (2000) Vascular endothelial growth factor (VEGF)-driven actin-based motility is mediated by VEGFR2 and requires concerted activation of stress-activated protein kinase 2 (SAPK2/p38) and geldanamycin-sensitive phosphorylation of focal adhesion kinase. J Biol Chem 275: 10661-10672.[Abstract/Free Full Text]

Rousseau S, Houle F, Landry J, and Huot J (1997) p38 MAP kinase activation by vascular endothelial growth factor mediates actin reorganization and cell migration in human endothelial cells. Oncogene 15: 2169-2177.[CrossRef][Medline]

Sieg DJ, Hauck CR, Ilic D, Klingbeil CK, Schaefer E, Damsky CH, and Schlaepfer DD (2000) FAK integrates growth-factor and integrin signals to promote cell migration. Nat Cell Biol 2: 249-256.[CrossRef][Medline]

Soga N, Namba N, McAllister S, Cornelius L, Teitelbaum SL, Dowdy SF, Kawamura J, and Hruska KA (2001) Rho family GTPases regulate VEGF-stimulated endothelial cell motility. Exp Cell Res 269: 73-87.[CrossRef][Medline]

Soldi R, Mitola S, Strasly M, Defilippi P, Tarone G, and Bussolino F (1999) Role of alphavbeta3 integrin in the activation of vascular endothelial growth factor receptor-2. EMBO (Eur Mol Biol Organ) J 18: 882-892.[CrossRef][Medline]

Stearns ME and Wang M (1992) Taxol blocks processes essential for prostate tumor cell (PC-3 ML) invasion and metastases. Cancer Res 52: 3776-3781.[Abstract/Free Full Text]

Stokes CL and Lauffenburger DA (1991) Analysis of the roles of microvessel endothelial cell random motility and chemotaxis in angiogenesis. J Theor Biol 152: 377-403.[CrossRef][Medline]

Vailhé B, Vittet D and Feige JJ (2001) In vitro models of vasculogenesis and angiogenesis. Lab Investig 81: 439-452.[Medline]

van Nieuw Amerongen GP, Koolwijk P, Versteilen A, and van Hinsbergh VWM (2003) Involvement of RhoA/Rho kinase signaling in VEGF-induced endothelial cell migration and angiogenesis in vitro. Arterioscler Thromb Vasc Biol 23: 211-217.[Abstract/Free Full Text]

Waltenberger J, Claesson-Welsh L, Siegbahn A, Shibuya M, and Heldin CH (1994) Different signal transduction properties of KDR and Flt1, two receptors for vascular endothelial growth factor. J Biol Chem 269: 26988-26995.[Abstract/Free Full Text]

Waterman-Storer CM, Worthylake RA, Liu BP, Burridge K, and Salmon ED (1999) Microtubule growth activates Rac1 to promote lamellipodial protrusion in fibroblasts. Nat Cell Biol 1: 45-50.[CrossRef][Medline]

Wijelath ES, Murray J, Rahman S, Patel Y, Ishida A, Strand K, Aziz S, Cardona C, Hammond WP, Savidge GF, et al. (2002) Novel vascular endothelial growth factor binding domains of fibronectin enhance vascular endothelial growth factor biological activity. Circ Res 91: 25-31.[Abstract/Free Full Text]

Wittmann T and Waterman-Storer CM (2001) Cell motility: can Rho GTPases and microtubules point the way? J Cell Sci 114: 3795-3803.[Abstract/Free Full Text]

Zakhireh B and Malech HL (1980) The effect of colchicine and vinblastine on the chemotactic response of human monocytes. J Immunol 125: 2143-2153.[Medline]

Zeng H, Zhao D, and Mukhopadhyay D (2002) KDR stimulates endothelial cell migration through heterotrimeric G protein Gq/11-mediated activation of a small GTPase RhoA. J Biol Chem 277: 46791-46798.[Abstract/Free Full Text]

Zhou X, Li J, and Kucik DF (2001) The microtubule cytoskeleton participates in control of 2 integrin avidity. J Biol Chem 276: 44762-44769.[Abstract/Free Full Text]

This article has been cited by other articles:

				J. Murtagh, H. Lu, and E. L. Schwartz Taxotere-Induced Inhibition of Human Endothelial Cell Migration Is a Result of Heat Shock Protein 90 Degradation Cancer Res., August 15, 2006; 66(16): 8192 - 8199. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) All Versions of this Article: mol.105.019075v1 69/4/1207    most recent Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Lu, H. Articles by Schwartz, E. L. Search for Related Content PubMed PubMed Citation Articles by Lu, H. Articles by Schwartz, E. L.