| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Biochemistry |
Division of Neurosurgery [G. Y. G.] and Departments of Neurobiology [L. S., T. M.], Neuropathology [C. L. G.], and Neurology [S. S. R.], University of Alabama at Birmingham, Birmingham, Alabama 35294
 |
ABSTRACT |
|---|
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Tissue invasion by malignant cells is a multistep process (2) . Central to this is the ability of transformed cells to crawl through the extracellular matrix, a process that requires the coordination between a force generating intracellular "motor" and an attachment mechanism to the extracellular matrix. Previous studies have focused on a number of extracellular factors that contribute to or modulate invasive behavior. These include growth factors, extracellular matrix proteins, integrins, and matrix metalloproteinases (3, 4, 5, 6, 7, 8, 9) . By contrast, relatively few studies have focused on the intracellular machinery that powers invasive behavior. This machinery is contained within the cytoskeleton, and includes microtubules, actin-containing microfilaments, and an array of mechanochemical enzymes collectively referred to as "molecular motors" (10) . The best-studied example of the latter is myosin II, the cytoplasmic analogue of smooth and skeletal muscle myosin (11) . Both pharmacological and molecular biological methods have demonstrated the importance of myosin II in powering a variety of motile behaviors, including growth cone motility, fibroblast locomotion, and astrocyte process outgrowth and contractile responses (12, 13, 14, 15, 16, 17, 18, 19) . Thus, myosin II represents a logical target for the development of anti-invasive strategies in glioma therapy.
The activity of myosin II is controlled by phosphorylation of a serine residue on its regulatory light chain by the enzyme MLCK3 (11 , 20) . Phosphorylation of the RLC by MLCK activates myosin II. This action is opposed by dephosphorylation of the RLC, which is mediated by a specific myosin light chain phosphatase. Thus, the degree of activity of myosin II in a cell is controlled by the relative balance of activities of MLCK and myosin light chain phosphatase.
Studying the role of myosin II in tissue culture has been aided by the availability of several drugs, all derivatives of fungal metabolites, that are highly specific competitive inhibitors of ATP binding to the active site of MLCK. We have used these MLCK inhibitors to examine the contribution of myosin II to the motility and invasive behavior of malignant gliomas. We have found that these drugs are potent and specific inhibitors of motility, as measured by in vitro assays, and exert their effects at nontoxic doses. These results highlight the importance of myosin II in glioma invasion and suggest that it may serve as an important target in the future development of anti-invasion strategies for the treatment of malignant glial neoplasms.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Human Glioma Cell Line.
U251MG glioma cells were obtained as a gift from Dr. Darell D. Bigner (Duke University, Durham, NC) and have been described previously (21)
. DMEM mixed 50:50 with Ham’s Nutrient Mixture F-12 (DMEM/F12; Mediatech, Inc.) supplemented to 2.6 mM L-glutamine and 7% FBS was used for culture of glioma cells at 37°C and 7.5% CO2. FBS (Atlanta Biologicals, Atlanta, GA) was heat inactivated (56°C, 45 min). Culture media and sera were determined to have <100 pg/ml of endotoxin by the Limulus amoebocyte lysate assay (Associates of Cape Cod, Inc., Woods Hole, MA). 3[H] Thymidine was purchased from Amersham Corp. (Arlington Heights, IL). This cell line has been passed >600 times and was maintained in 75- or 150-cm2 plastic tissue culture flasks in complete culture. Antibiotics were not used routinely, and these glioma cells were negative for Mycoplasma by a two-stage nested primer PCR (American Type Culture Collection, Rockville, MD).
Haptotaxis Assay.
Two-well Boyden-type chambers with 8-µm pore polycarbonate filters were coated with human vitronectin on the lower surface, as described (22, 23, 24)
. Recently subcultured, subconfluent U251MG cells were harvested with 1.5 mM EDTA in PBS, counted, and suspended in DMEM + 0.1% BSA + 1.5 mM L-glutamine (migration buffer) at 400,000 cells/ml. Aliquots of 100 µl were plated in the upper chamber and incubated at 37°C in 5% CO2 for 4 or 12 h. Migrating cells were fixed and stained with crystal violet as described (22, 23, 24)
and quantified by counting 10 random 1 mm2 fields in a 1-cm grid on an inverted microscope. The effects of MLCK inhibitors were examined in this assay by adding equal concentrations of drug to the upper and lower chambers at the time of plating. Each experiment was accompanied by a control consisting of a polycarbonate filter coated on the lower surface with ovalbumin, and this consistently demonstrated no migration.
Scrape Motility Assay.
U251 MG cells were grown to confluency on 12-mm glass coverslips. The cell monolayer was mechanically scarred with a sterile pipette tip, and coverslips were placed on the stage of a LU-CB-1 tissue chamber equipped with a NP-2 incubator (Nikon, Japan), which maintained the temperature at 37°C and at an atmosphere of 95% O2 and 5% CO2. Cells were visualized by a Nikon Diaphot inverted microscope with phase contrast optics and a x20 objective. Images were captured on a time-lapse VHS video recorder/player and digitized off-line using an Axon Imaging Workbench 2 (Version 2.10; Axon Instruments) software. ML7 (10 µM), KT5926 (2 µM), and DMSO (0.1%; used as a vehicle for KT5926) were added directly to the growth media at the time of scraping. Afterward, media were not changed throughout the recording time.
Measurement of RLC Phosphorylation.
Measurement of the effects of ML7 on RLC phosphorylation was performed on U251 MG cell lysates by using urea gel electrophoresis as described (16)
. Briefly, cells were collected in a 10% trichloroacetic acid solution supplemented with 10 mM DTT. Pellets were rinsed with ether and suspended in sample buffer containing 8.0 M urea, 9.0 mM DTT, 20 mM glycine, and 55 mM Tris base. Samples were run on a 1.5% Bis, 30% acrylamide, 40% glycerol gel, followed by electrotransfer to a nitrocellulose membrane. The membrane was incubated overnight with a rabbit polyclonal anti-RLC antibody (kindly provided by Dr. James T. Stull, University of Texas Southwestern Medical Center Dallas, TX). RLC isoforms were then visualized by using an alkaline phosphatase secondary antibody (Vector, Burlingame, CA).
Alamar Blue Viability Assay.
Glioma cells were plated at 1 x 104 cells/well in 96-well plates in 0.2 ml of complete culture medium and incubated (37°C, 7.5% CO2) for up to 72 h. Twelve h after addition of MLCK inhibitors or DMSO vehicle, AlamarBlue (40 µl; Alamar Biosciences, Sacramento, CA) was added to each well, the plates incubated for 120–180 min, and cell viability based on metabolic conversion of the deep blue dye to a pink color was assessed spectrophotometrically in situ. Absorbances of supernatants were measured at a wavelength of 562 nm, and background absorbance at 590 nm was subtracted using a semiautomatic plate reader (model EL310; Bio-Tek Instruments, Inc., Winooski, VT). Multiple experiments have confirmed that AlamarBlue does not interfere with detection of [3H]thymidine uptake (25
, 26)
. All experiments were repeated a minimum of three times.
[3H]Thymidine Incorporation
Glioma cell proliferation was measured by [3H]thymidine incorporation. Cells were plated at 1 x 104 cells/well in 96-well plates in 90 µl of complete culture medium and incubated (37°C, 7.5% CO2) for up to 72 h. Cells were incubated with [3H] thymidine (41 Ci/mmol; 0.5 µCi/well) for an additional 15 h. Cells were detached by 30 min incubation at 37°C in cell release solution (0.1% trypsin, 2.15 mM EDTA, and 0.4 M sucrose, pH 7.5, 40 µl/well). Cells were then harvested with a semiautomatic cell harvester (Skatron, Inc., Sterling, VA) onto scintillant-impregnated Ready Filter (Beckman, Fuller, CA) and lysed with 3% acetic acid. Filter discs were transferred to scintillation vials, and radioactivity was measured in a liquid scintillation spectrometer (LKB Rackbeta 1211, Helsinki, Finland). Experiments were performed with at least four replicates, and results were expressed as mean cpm. SE was usually <10%.
Cell Adhesion Assay.
Glioma cells were passaged two days before the experiment at 5 x 106 cells in a T75 cm2 flask. On the day of the experiment, cells were detached with 0.53 mM EDTA and 0.01% KCl in PBS, washed twice in serum-free medium, and adjusted to 2 x 105 viable cells/ml in serum-free medium. Dynatech Immunolon II 96-well flat-bottomed EIA plates were coated overnight at 4°C with 10 µg/ml vitronectin or 0.2% BSA in PBS, rinsed three times in PBS, and blocked with 0.2%BSA (37°C, 2 h). Cells (2 x 104) were added in a volume of 100 µl to vitronectin-coated and control wells and allowed to bind for 1 h (37°C, 5% CO2). Triplicate samples were run for each concentration of drug. Unbound cells were removed by rinsing each well three times with 200 µl of warmed DMEM/F12 + 0.1% BSA, and the wells were refilled with 100 µl of DMEM/F12 + 0.1% BSA. AlamarBlue dye (25 µl) was added to each well, and the plate was incubated at 37°C, 5% CO2 for 4 h. The capacity of the adherent cells to convert the dark blue soluble formazan dye to a reddish pink soluble product was used to indicate the relative number of cells in each well. Detailed experiments using a graded number of U251MG cells (1, 2, 4, 8, 10, 12, 15, or 20 x 103) plated in medium with 0.1% BSA or 7% FBS demonstrated a linear relation between cells plated and A562–590 nm (a measure of cells bound) and no difference in the numbers of cells binding in the two media. As might be expected, washing cells plated in 0.1% BSA removed an average of 30 ± 8% more cells than washing cells plated in 7% FBS. However, the relation between cells plated and cells bound remained linear (with different slopes) for both groups. Values obtained with inhibitors were expressed as a percentage of cells binding in DMEM/F12 + 0.1% BSA only.
Fluorescence Microscopy.
For examining the effects of MLCK inhibitors on actin distribution, U251 MG cells cultured on glass coverslips were treated with KT5926 or ML7 for 1 h and were then fixed with 4.0% (w/v) PBS-buffered paraformaldehyde for 5 min and permeabilized with 0.15% Triton X-100 in PBS for 5 min. Permeabilized cells were then incubated with a 1:500 dilution of rhodamine phalloidin (Molecular Probes) in blocking buffer (2% BSA in PBS) for 2 h at room temperature in the dark. Cells were then washed three times for 10 min each with PBS and visualized with a Leica fluorescence microscope. Images were captured and digitized with ImaCapt software (Imagraph Corporation, Chelmsford, MA) and with a DEI-750 CCD camera.
For examining the effects of MLCK inhibitors on focal adhesion and integrin distribution, U251MG cells were allowed to attach to coverslips coated previously with vitronectin (10 µg/ml) in serum-free DMEM overnight. Prior to immunofluorescent analysis, ML7 (10 µM), KT5926 (2 µM), or the vehicle control (DMSO) were preincubated with the cells for 1 h at 37°C and 5% CO2. Subsequently, the cells were fixed with 3% buffered paraformaldehyde, permeabilized with 0.15% Triton X-100 in PBS (60 s), reacted with mAb anti-
v (LM142, 1:500 of ascites) or with mAb anti-vinculin (1:500 of ascites) for 30 min, washed, reacted with FITC-conjugated goat anti-mouse IgG, and coverslipped with Fluoromount. Fluorescence was visualized with a Leitz DMRB microscope equipped for fluorescence.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
demonstrates the average number of cells migrating per mm2 after 4 h of incubation in chambers in the absence of drug and in the presence of KT5926, KT5823, and echistatin. As can be seen from the figure, KT5926 reduces Transwell migration in this assay by nearly 95%, whereas KT5823 has no appreciable effect. By comparison, echistatin, which blocks integrin-mediated attachment to vitronectin (27)
, reduced migration by 65%.
|
for KT5926 and Fig. 2B
for ML7. Data presented in these figures is for 12-h incubations. Shorter incubation periods (4 h) gave qualitatively similar results but with larger SD. Data were fit to a Michaelis-Menten dependence of cell count on drug concentration using a curve-fitting procedure (DeltaGraph Pro 3). Both KT5926 and ML7 act as competitive inhibitors of ATP binding to the active site of MLCK (28
, 29)
. Assuming a 1:1 correspondence between MLCK inhibition and inhibition of in vitro haptotaxis, the relationship between drug concentration and inhibition of haptotaxis is as follows:
 |
|
, as evidenced by the large apparent ratio of phosphorylated to unphosphorylated RLC. The limited resolution of the urea gel system prevents a precise measure of the number of phosphorylated species present. However, at least three and possibly four bands of phosphorylated RLC are evident, which correspond to mono- and diphosphorylation of the two isoforms (arrows). By contrast, treatment for 3 h with ML7 at doses that abolish haptotaxis and motility leads to a marked reduction in RLC phosphorylation. This is demonstrated by the near disappearance of the more mobile, phosphorylated species.
|
Cell movement was monitored with a scrape motility assay. As demonstrated in Fig. 4A
, U251MG cells can be seen to move toward and fill in an 
150-µm-wide cell free zone over a 24-h incubation period. When this assay is performed in the presence of 2 µ M KT5926 (Fig. 4B)
or 10 µM ML7 (Fig. 4C)
, this process is completely blocked.
|
shows, both KT5926 and ML7 partially inhibit attachment over the same range of concentrations that block in vitro haptotaxis. By contrast, KT5823 has little effect in this assay. Similar results have been reported previously in lung carcinoma cells attaching to fibronectin (29)
.
|
Effects of MLCK Inhibitors on Integrin, Focal Adhesion, and Microfilament Distribution.
To determine whether inhibition of MLCK affected cytoskeletal organization in cells adherent to vitronectin, we performed immunofluorescence microscopy of U251 MG cells stained with anti-vinculin (a marker of focal adhesions) or with an antibody to the v subunit of the vß3 integrin. As Figs. 6
and 7
demonstrate, untreated U251MG cells demonstrate focal adhesion localization of both proteins, which are located both centrally and peripherally, consistent with the appearance of mature focal adhesions (A, arrows). Addition of either ML7 (panel B) or KT5926 (panel C) produces a redistribution of focal adhesions toward the cell cortex (arrows).
|
|
shows, untreated U251 cells demonstrate both prominent stress fibers as well as F-actin in a submembranous layer and in membrane ruffles, a pattern that has been described in cultured astrocytes (37)
. Treatment of cells with KT5926 (Fig. 8B)
or ML7 (Fig. 8C)
for 3 h leads to a marked reduction in the number of stress fibers, in formation of punctate, centrally located actin staining, and with relative preservation of F-actin in membrane ruffles and along the plasma membrane. Pretreatment of cells with either drug for 24 h, followed by replacement with drug-free medium for an additional 24 h, was performed to determine reversibility of this effect. Cells were then fixed and stained with rhodamine phalloidin, and this revealed a normal complement of stress fibers (data not shown).
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The processes of intracellular force generation require an intact cytoskeleton, and cytoskeletal components that are essential for this process include actin and myosin II. Molecular genetic, immunological, and pharmacological interventions have been used to demonstrate that myosin II plays a central role in the formation of a cleavage furrow in cytokinesis, in receptor internalization, in secretion, in wound closure, in generation of contractile forces, and in directed cell movement (12, 13, 14, 15, 16, 17, 18, 19 , 38 , 39) . More relevant studies of the importance of myosin II to the central nervous system have been published recently (16 , 40 , 41) and have demonstrated the importance of myosin II in growth cone motility and in astrocyte contractility. In this study, we have extended these observations to a tumor derived from astrocytes, glioblastoma multiforme.
Myosin II, like its smooth muscle counterpart, is regulated by calmodulin-dependent phosphorylation of the regulatory light chain (11
, 20)
. The importance of myosin II regulation in cell function is underscored by the multiple signal transduction pathways that control RLC phosphorylation. These include mitogen-activated protein kinase, which regulates MLCK activity in a phosphorylation-dependent manner, as well as the G protein Rho, which can induce RLC phosphorylation through its effects on Rho kinase and myosin light chain phosphatase (42, 43, 44)
. Phosphorylation allows myosin II to interact with actin in a force-productive manner (20)
, and dephosphorylation of the regulatory light chain by a myosin light chain phosphatase inactivates this system. The balance of activities of MLCK and myosin light chain phosphatase determine how motile a cell is. Thus, drugs that inhibit MLCK, such as KT5926 and ML7, would be expected to lead to an eventual dephosphorylation of the myosin regulatory light chains and inactivation of the contractile apparatus. In this study, we have used KT5926 and ML7 to determine whether inhibition of myosin function indeed blocks motile-dependent behavior in U251MG, a human glioblastoma cell line. As demonstrated in Figs. 1
and 2
, both drugs are potent inhibitors of glioma motility and invasiveness, as measured in vitro. Other protein kinases, including calmodulin-kinase II and protein kinase C, can be inhibited by these drugs. However, the Ki for protein kinase C for KT5926 is 723 nM, which is >40-fold higher than that for MLCK. Likewise, the Ki of ML7 for protein kinase C is 42 µ M, which is 140-fold higher than that for MLCK. Furthermore, although calmodulin-kinase II is effectively inhibited by KT5926, it is essentially unaffected by ML7. Thus, the dose-response relationship seen with these agents together can only be explained by their specific inhibition of MLCK. This is also supported by urea gel electrophoresis of U251MG lysates (Fig. 3)
, which provides direct evidence that ML7 at doses that abolish haptotaxis and migration also leads to a significant reduction in RLC phosphorylation. In addition to phosphorylation at serine 19 and threonine 18 by MLCK, the RLC can be phosphorylated by protein kinase C at serine 1, and this phosphorylation reaction inhibits actomyosin contractility (45)
. Thus, the small amount of phosphorylation still present at high doses of ML7 (Fig. 3)
may reflect the effects of residual protein kinase C activity.
Although the scrape motility assay is sensitive to the motile behavior of the glioma cells, the in vitro haptotaxis assay used in this study depends not only on cell motility but also on the ability of glioma cells to detect a gradient of vitronectin, crawl toward it, and attach to a vitronectin-coated membrane. This in turn may require gliomas to bind vitronectin through their membrane-bound integrins, internalize these complexes, and recycle the vitronectin receptors. Localization of membrane-bound receptors and their internalization are myosin II-dependent processes (18) , and thus, myosin II may also be responsible for the ability of U251MG cells to sample their environment and migrate toward a chemical gradient. This may explain why KT5926 and ML7 also inhibited attachment of U251MG cells to a vitronectin-coated surface and redistributed focal adhesions to the cell cortex. The latter distribution is seen during early cytoskeletal organization, when focal adhesions are immature.
Immunofluorescence microscopy has demonstrated that stress fibers terminate at focal adhesions, and both direct and indirect interactions exist between specific components of focal adhesions and F actin (46
, 47)
. Stress fibers contain both actin and myosin II (46
, 47) , and if myosin II activity were necessary for stress fiber integrity, then its inhibition with KT5926 or ML7 might be expected to disrupt not only focal adhesions but also stress fibers as well. As shown in Fig. 8
, this is in fact the case, and these results are consistent with other studies that have demonstrated the necessity of myosin II for normal stress fiber localization (29)
.
The results of this study thus demonstrate that myosin II has several important functions in supporting the process of invasion by glioma cells. Myosin II drives cell motility, but in addition, its activity may be necessary for sampling the environment for chemoattractant cues and attachment to the extracellular matrix. However, myosin II is not a homogeneous population of molecular motors. There are two major isoforms of myosin II, referred to as IIA and IIB (41) , and they differ in their intracellular localization and tissue distribution (46) . Both isoforms are expressed in human glioblastoma surgical specimens at levels that are severalfold higher than in normal brain.4 Because the activity of both myosin II isoforms is controlled by regulatory light chain phosphorylation, it is not possible to deduce which isoform is responsible for glioma invasiveness on the basis of pharmacological studies with MLCK inhibitors. Such investigations will likely require more specific methods of suppressing myosin II function, such as utilization of isoform-specific antisense oligonucleotides (48) . Nevertheless, the results of this study make it clear that myosin II plays a central role in glioma invasiveness and could therefore serve as a target for the development of invasion-inhibiting therapies.
|
FOOTNOTES |
|---|
2 To whom requests for reprints should be addressed, at Department of Neurology, University of Alabama at Birmingham, UAB Station, Birmingham, AL 35294. Fax: (205) 975-7546; E-mail: s_rosenfeld{at}email.neuro.uab.edu
3 The abbreviations used are: MLCK, myosin light chain kinase; RLC, regulatory light chain; FBS, fetal bovine serum.
4 S. S. Rosenfeld and G. Y. Gillespie, manuscript in preparation.
Received 11/23/98. Accepted 3/ 4/99.
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
. Neurosurgery, 36: 1165-1171, 1995.[Medline]
vß3 integrin: adhesion mechanism for transformed glial cells. J. Clin. Invest., 88: 1924-1932, 1991.
5ß1 is required for internalization of vitronectin by integrin vß3. J. Biol. Chem., 272: 2736-2743, 1997.This article has been cited by other articles:
|
|
 |
|
  S. R. Wylie and P. D. Chantler Myosin IIC: A Third Molecular Motor Driving Neuronal Dynamics Mol. Biol. Cell, September 1, 2008; 19(9): 3956 - 3968. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Beadle, M. C. Assanah, P. Monzo, R. Vallee, S. S. Rosenfeld, and P. Canoll The Role of Myosin II in Glioma Invasion of the Brain Mol. Biol. Cell, August 1, 2008; 19(8): 3357 - 3368. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Bazzaro, A. Santillan, Z. Lin, T. Tang, M. K. Lee, R. E. Bristow, I.-M. Shih, and R. B.S. Roden Myosin II Co-Chaperone General Cell UNC-45 Overexpression Is Associated with Ovarian Cancer, Rapid Proliferation, and Motility Am. J. Pathol., November 1, 2007; 171(5): 1640 - 1649. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. E. Gonzalez, E. A. Peterson, L. M. Privette, J. L. Loffreda-Wren, L. M. Kalikin, and E. M. Petty High SEPT9_v1 Expression in Human Breast Cancer Cells Is Associated with Oncogenic Phenotypes Cancer Res., September 15, 2007; 67(18): 8554 - 8564. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. S. Rosenfeld, J. Xing, L.-Q. Chen, and H. L. Sweeney Myosin IIB Is Unconventionally Conventional J. Biol. Chem., July 18, 2003; 278(30): 27449 - 27455. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Watson, A. Perry, V. Budhjara, C. Hicks, W. D. Shannon, and K. M. Rich Gene Expression Profiling with Oligonucleotide Microarrays Distinguishes World Health Organization Grade of Oligodendrogliomas Cancer Res., March 1, 2001; 61(5): 1825 - 1829. [Abstract] [Full Text]
|
|
|
|
 |
|
 R. Eddy, L. Pierini, F Matsumura, and F. Maxfield Ca2+-dependent myosin II activation is required for uropod retraction during neutrophil migration J. Cell Sci., January 4, 2000; 113(7): 1287 - 1298. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Cancer Research | Clinical Cancer Research |
| Cancer Epidemiology Biomarkers & Prevention | Molecular Cancer Therapeutics |
| Molecular Cancer Research | Cancer Prevention Research |
| Cancer Prevention Journals Portal | Cancer Reviews Online |
| Annual Meeting Education Book | Meeting Abstracts Online |
) and ML7 (
) inhibit attachment by 65–75% over the same range of concentrations that inhibit U251MG in vitro haptotaxis (Fig. 2)
), an inhibitor of cyclic GMP-dependent protein kinase, has minimal effect. Bars, SD.
