Cancer Research Targets Advances in Breast Cancer Research
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
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Werbowetski-Ogilvie, T. E.
Right arrow Articles by Del Maestro, R. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Werbowetski-Ogilvie, T. E.
Right arrow Articles by Del Maestro, R. F.
[Cancer Research 66, 1464-1472, February 1, 2006]
© 2006 American Association for Cancer Research

Cell, Tumor, and Stem Cell Biology

Isolation of a Natural Inhibitor of Human Malignant Glial Cell Invasion: Inter {alpha}-Trypsin Inhibitor Heavy Chain 2

Tamra E. Werbowetski-Ogilvie1, Nathalie Y.R. Agar1, Roberta M. Waldkircher de Oliveira1, Damien Faury3, Jack P. Antel2, Nada Jabado3 and Rolando F. Del Maestro1

1 Brain Tumour Research Centre and 2 Neuroimmunology Unit, Montreal Neurological Institute, McGill University and 3 Department of Pediatrics, Montreal Children's Hospital Research Institute, Montreal, Quebec, Canada

Requests for reprints: Rolando Del Maestro, Brain Tumour Research Centre, Montreal Neurological Institute and Hospital, 3801 Rue University, Montreal, Quebec, Canada H3A 2B4. Phone: 514-398-5791; Fax: 514-398-2811; E-mail: rolando.delmaestro{at}mcgill.ca.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Malignant central nervous system (CNS) tumors, such as glioblastoma multiforme, invade the brain and disrupt normal tissue architecture, making complete surgical removal virtually impossible. Here, we have developed and optimized a purification strategy to isolate and identify natural inhibitors of glioma cell invasion in a three-dimensional collagen type I matrix. Inter {alpha}-trypsin inhibitor heavy chain 2 (ITI H2) was identified from the most inhibitory fractions and its presence was confirmed both as a single protein and in a bikunin-bound form. Stable overexpression in U251 glioma cells validated ITI H2's strong inhibition of human glioma cell invasion together with significant inhibition of cell proliferation and promotion of cell-cell adhesion. Analysis of primary human brain tumors showed significantly higher levels of ITI H2 in normal brain and low-grade tumors compared with high-grade gliomas, indicating an inverse correlation with malignancy. The phosphatidylinositol 3-kinase/Akt signaling cascade seemed to be one of the pathways involved in the effect of ITI H2 on U251 cells. These findings suggest that reduction of ITI H2 expression correlates with brain tumor progression and that targeting factors responsible for its loss or restoring the ITI supply exogenously may serve as potential therapeutic strategies for a variety of CNS tumors. (Cancer Res 2006; 66(3): 1464-72)


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Glioblastoma multiforme is the most common malignant central nervous system (CNS) tumor in adults (1). It is characterized by extensive necrosis, a high mitotic index, endothelial cell proliferation, and nuclear pleomorphism. Invading glioblastoma cells rapidly infiltrate and disrupt normal tissue architecture, making complete surgical removal virtually impossible (2). This diffuse infiltrative growth leads to early recurrence and poor patient prognosis. Consequently, the invasive cascade is an important therapeutic target for glioblastoma multiforme.

There are many known positive regulators of malignant brain tumor cell proliferation and invasion; however, little is known about negative modulators of malignant glial cell invasion (1, 3–5). Our laboratory has provided evidence for a secreted chemorepellent that directs glioma cell invasion (5). Conditioned medium from both U251 and C6 spheroids significantly inhibited invasion of both spheroid types when implanted into three-dimensional collagen type I gels, providing evidence for a potential inhibitory or kinetic effect as well (5). Because this model can be used as a functional assay to identify endogenous and/or serum-derived inhibitors/repellents of glioma invasion, a purification strategy was designed and optimized to isolate these inhibitors/repellents from the conditioned medium of glioma spheroids.

This study presents the isolation and identification of inter {alpha}-trypsin inhibitor heavy chain 2 (ITI H2) as a strong natural inhibitor of brain tumor invasion. Overexpression of ITI H2 in the U251 glioma cell line confirmed its inhibitory role in malignant glioma invasion and revealed an inhibitory effect on proliferation with a concomitant increase in cell adhesion. Moreover, expression levels of ITI H2 in primary human tumors and cultured human glioma cell lines were inversely correlated with malignancy, indicating a possible role for the ITI family members in CNS tumor progression. Our results present ITI H2 both as a potential indicator of tumor malignancy and as a novel target for therapeutic intervention that merits further functional characterization.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Cell culture and activity assay. C6 (murine glioblastoma) cells were cultured in DMEM supplemented with 10% fetal bovine serum (FBS), 125 units/mL penicillin G, 125 µg/mL streptomycin sulfate, and 2.2 µg/mL amphotericin B (Fungizone) and seeded into spinner culture flasks as described (5, 6). Tumor spheroids were implanted into 48-well culture dishes containing 500-µL aliquots of a collagen type I solution (Vitrogen 100; COHESION, Palo Alto, CA) and cell invasion was assessed daily as previously described (5, 6). All culture reagents were obtained from Life Technologies Invitrogen (Burlington, Ontario, Canada) unless otherwise stated. Cell lines were purchased from the American Type Culture Collection (ATCC; Rockville, MD).

Protein purification. Conditioned medium was collected from C6 spheroids in spinner culture between 2 and 5 weeks, and cell debris was removed by ultracentrifugation for 60 minutes at 45,000 rpm. Conditioned medium was then concentrated 30x by ultrafiltration using a stirred cell equipped with a YM-10 membrane (Millipore, Etobicoke, Ontario, Canada). A wash buffer containing 20 mmol/L MOPS (pH 7.2) and an elution buffer containing 20 mmol/L MOPS (pH 7.2) and 1.5 mol/L NaCl were prepared. The concentrated C6 conditioned medium was reconstituted in 100 mL of 2 mmol/L MOPS wash buffer, passed through a 0.2-µm filter, and loaded onto a Resource Q anion exchange column (Amersham Biosciences, Baie d'Urfe, Quebec, Canada) equilibrated with 20 mmol/L MOPS wash buffer. Chromatography was done on an AKTA fast protein liquid chromatography (Amersham Biosciences). The column was operated at a flow rate of 2.5 mL/min. Bound proteins were eluted with a linear gradient of 0 to 1.5 mol/L NaCl over 30 column volumes, re-equilibrated in DMEM without serum, and applied over C6 spheroids in triplicate. Fractions having an inhibitory effect on invasion were pooled, collected, and reconstituted in 10 mmol/L sodium phosphate buffer (pH 7) and loaded onto a 5-mL HiTrap Heparin HP column (Amersham Biosciences) equilibrated with the same buffer. Fractions were eluted with a linear gradient of 10 mmol/L sodium phosphate + 2 mol/L NaCl (pH 7) over 45 column volumes, and then re-equilibrated in DMEM without serum. Fractions were then applied over C6 spheroids in triplicate and those that had an inhibitory effect on invasion were selected for further analysis. Inhibitory fractions from both columns were resolved using SDS-PAGE on a 10% gel. Gels were silver stained according to established procedures (Amersham Biosciences). Protein concentrations were determined using the Bio-Rad Protein assay dye reagent concentrate (Bio-Rad, Mississauga, Ontario, Canada).

Mass spectrometry analysis of purified fractions. Tryptic digest and tandem mass spectrometry (MS) analysis were conducted on pooled fractions with the most inhibition activity from the HiTrap Heparin HP column (McGill University and Genome Quebec Innovation Centre, Montreal, Quebec, Canada). Mascot scores were obtained and significance was determined as the minimum threshold required to be considered a nonrandom assignment (7).

Cloning and stable cell line production. pCMV Sport6 vectors containing the ITI H2 and bikunin full-length cDNAs were obtained from ATCC. The BglII and EcoRI restriction sites were used to subclone the ITI H2 and bikunin full-length constructs into the pLPCX retroviral expression vector (Clontech, Palo Alto, CA). The new constructs were introduced into the U251 human glioblastoma cell line following the guidelines of the manufacturer (Retroviral Gene Transfer and Expression User Manual, BD Biosciences, Palo Alto, CA). Cells were cultured in selection medium containing 1 µg/mL puromycin (Sigma-Aldrich Co. Canada Ltd., Oakville, Ontario, Canada) for 3 weeks, and colonies were then pooled to generate polyclonal cell lines. Two distinct polyclonal cell lines were produced from independent rounds of retroviral infections for each DNA construct with separate sets of cells. Confirmation of stable ITI H2 and bikunin overexpression was assessed using reverse transcription-PCR (RT-PCR; see below). See Supplementary Table S2 (1–3) for primer details.

Hanging-drop aggregates and time-lapse videomicroscopy. Confluent cultures of wild-type U251 cells, U251 pLPCX cells, and U251 cells stably overexpressing ITI H2 and bikunin were trypsinized and hanging-drop aggregates were prepared (5, 8–10). Aggregates were implanted into collagen type I gels and invasion was assessed (5, 6). Six hours after implantation, aggregates were imaged for the following 20 hours using time-lapse equipment and Northern Eclipse 6.0 software as previously described (11, 12).

Proliferation assay. To evaluate the effect of ITI H2 and bikunin overexpression on U251 glioma cell proliferation, monolayers of U251 wild-type, U251 pLPCX, U251 ITI H2, and U251 bikunin cells were plated on six-well culture dishes (40,000 per well and three replicates for each time point). Cells were counted on days 2 and 4 using a Coulter Z Series counter (Beckman-Coulter, Inc., Miami, FL) and doubling time for each cell line was calculated.

Cell attachment assay. U251 pLPCX and ITI H2–overexpressing cells were seeded at 40,000 per well in six-well dishes in triplicate. After 1, 2, 3, and 4 hours at 37°C, the wells were imaged and the number of attached cells was calculated for both empty vector and ITI H2 cells (5, 11, 12).

Antibody production and Western blot analysis. An anti-ITI H2 polyclonal antibody was generated against the COOH-terminal region of the protein by injecting rabbits with the synthetic peptide PGKDPEKPEASMEVK coupled to KLH (Sheldon Biotechnology Centre, McGill University, Montreal, Canada). SDS-PAGE gel wells were loaded each with 1 µg protein from each purified fraction. The ITI H2–specific polyclonal antibody was used at a concentration of 1:500. The antirabbit bikunin antibody (1:2,000) has been generously donated by Dr. Bo Akerstrom (Lund University, Lund, Sweden) and a commercial antibody against all ITI heavy and the light chain (1:2,000) was also used to validate ITI H2 enrichment (Dako Cytomation, Inc., Mississauga, Ontario, Canada). The horseradish peroxidase–conjugated goat anti-rabbit immunoglobulin G secondary antibody (1:5,000) was used and visualized by enhanced chemiluminescence (Perkin-Elmer Life Sciences, Inc., Markham, Ontario, Canada).

For analysis of Akt and p44/42 mitogen-activated protein kinase (MAPK) expression, U251 pLPCX and ITI H2 cells were lysed in radioimmunoprecipitation assay buffer containing 1 Complete Mini protease inhibitor tablet (Roche Applied Science, Laval, Quebec, Canada), 20 µg/mL phenylmethylsulfonyl fluoride, and 0.4 µmol/L sodium orthovanadate phosphastase inhibitor. Twenty micrograms of protein for each sample were subjected to SDS-PAGE. The phospho-AKT (Ser473; 1:1,000), AKT (1:1,000), p44/42 MAPK (Thr202/Tyr204; 1:2,000), and MAPK (1:2,000) antibodies were obtained from Cell Signaling Technology (Beverly, MA). Donkey anti-rabbit and anti-mouse secondary antibodies (1:10,000) were used.

Immunocytochemistry. To examine intracellular localization of ITI H2 in the U251 ITI H2 stable cell line, immunocytochemistry was done. Briefly, U251 ITI H2 cells were plated on collagen-coated coverslips for 48 hours (BioCoat, BD Biosciences, Discovery Labware, Mississauga Ontario, Canada). The ITI H2–specific polyclonal antibody was used at a concentration of 1:50 in 0.5% bovine serum albumin 0.02% Triton X-100/PBS. Preimmune serum and staining without primary antibody were used as negative controls. Donkey anti-rabbit Alexa 555 secondary antibody was used. Cells were colabeled for actin by adding Phalloidin Alexa Fluor 488 (1:400; Molecular Probes, Eugene, OR) to the secondary antibody mixture. Coverslips were mounted and imaged using a LSM 510 confocal scanning laser microscope at x63 magnification using the helium-neon (543 nm) and argon (488 nm) lasers (Carl Zeiss, Toronto, Ontario, Canada).

Human brain tumor samples. Twenty-seven human brain tumor samples were obtained from the Department of Molecular Pathology and Neurology, University of Lodz tissue bank (Poland), Simmelweiss University tissue bank (Hungary), Centre de Neuro-pathologie, Clermont-Ferrand tissue bank (France), the Brain Tumor Tissue Bank London Health Sciences Centre (London, Ontario, Canada), and the Brain Tumour Research Centre Tissue Bank at the Montreal Neurological Institute (Montreal, Quebec, Canada). Samples of human fetal astrocytes obtained at 12 to 16 weeks of gestation following Canadian Institutes of Health Research guidelines were acquired from the Albert Einstein College of Medicine Human Fetal Tissue Repository (Bronx, NY; ref. 13). A sample of normal brain was also analyzed and all relevant clinical information for the tumors was available.

RT-PCR of malignant glioma cell lines and primary human brain tumors. The endogenous expression of ITI H2 and ß-actin mRNA in U251, U87, U343, and U373 glioma cells lines and primary human brain tumors was examined by RT-PCR. Total RNA was extracted from malignant glioma cell lines using RNeasy kit (Qiagen, Mississauga, Ontario, Canada). For primary human brain tumors, total RNA was extracted from frozen samples using the TRIzol Reagent (Life Technologies). First-strand cDNA was synthesized using the First-Strand cDNA Synthesis Kit (Amersham Biosciences). See Supplementary Table S2 (1, 4) for primer details. ß-Actin was used as an internal control. The following PCR conditions were used: 95°C for 2 minutes, 80°C for 2 to 4 minutes (during which Taq is added), and 30 cycles of 95°C for 60 seconds, 55°C (60°C for ß-actin) for 30 seconds, and 72°C for 60 seconds followed by 10 minutes at 72°C.

Immunohistochemistry. Normal brain and brain tumor samples embedded in paraffin were sectioned in 7-µm intervals and samples were subjected to antigen retrieval and placed in a Dako Autostainer (Dako Diagnostics Canada, Inc., Mississauga, Ontario, Canada). The ITI H2 antibody and preimmune serum (1:100 dilution) were used for immunohistochemical staining. Samples were counterstained in hematoxylin for 1 minute, mounted using a water-soluble mountant, and imaged using a standard light microscope at x40 magnification.

Statistical analysis. All tests were done using SPSS Graduate Pack 9.0 statistical software (SPSS, Inc., Chicago, IL). Descriptive statistics including mean and SE along with one-way ANOVAs, independent sample two-tailed t tests, and Tukey's test for multiple comparisons were used to determine significant differences. P < 0.05 was considered significant.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Inhibitory fractions contain enriched protein bands involved in invasion. To isolate either secreted or serum-derived inhibitors, we concentrated C6 conditioned medium and applied it to a Resource Q anion exchange column. Fractions were collected and applied to C6 astrocytoma spheroids implanted in collagen gels to assay for inhibition of cell invasion. Under control conditions, the average invasive rate was 10.2 µm/h (Fig. 1A). Fractions 16 to 21 consistently had an inhibitory effect on invasion (Fig. 1A). Fractions 18 to 19 yielded the highest inhibitory effect on cell invasion and induced a significant decrease of the average invasive rate for C6 spheroids from 10.2 to 6.2 µm/h (Tukey's test, P < 0.001; Fig. 1A). Fractions 16 to 21 were pooled and applied to a HiTrap Heparin HP column. Fractions were then collected and applied to C6 spheroids implanted in collagen gels. Fractions 1 to 8 (comprising the proteins that did not bind to the column) and 17 to 22 displayed the highest inhibitory effect on invasion with an average invasive rate of 6.85 µm/h (Fig. 1B). Silver staining of inhibitory fractions revealed the presence of two enriched bands at ~90 and 140 kDa (Fig. 1C). MS analysis identified ITI H2 as the most abundant protein from a pooled sample of inhibitory fractions 1 to 8 and 17 to 22 from the HiTrap Heparin column (see Supplementary Table S1). Mascot scores >38 were considered significant at P < 0.05.


Figure 1
View larger version (33K):
[in this window]
[in a new window]
  		Figure 1. Purified fractions from Resource Q and HiTrap Heparin HP columns contain enriched proteins involved in invasion. A, quantification of inhibitory effect of fractions obtained from a Resource Q anion exchange column. Note the maximum inhibition of C6 spheroid invasion rate with fractions 16 to 17, 18 to 19, and 20 to 21. Bars, SE. **, P < 0.01; ***, P < 0.001. Representative of three independent experiments. B, quantification of inhibitory effect of fractions obtained from a HiTrap Heparin HP column. Note the maximum inhibition of C6 spheroid invasion rate with fractions 1 to 8, 17 to 19, and 20 to 22. Bars, SE. *, P < 0.05; **, P < 0.01; ***, P < 0.001. Representative of three independent experiments. C, silver-stained PAGE gel of C6 conditioned medium, DMEM + 10% FBS, and inhibitory fractions from the Resource Q anionic exchange and HiTrap Heparin HP column. Note enrichment of two bands running at ~140 and 90 kDa.

 
Western blot analysis confirms the presence of ITI H2 and ITI H2-bikunin complexes in purified fractions. To confirm the presence of ITI H2 in the purified fractions, we designed an antibody against ITI H2 and used Western blot analysis to compare bands obtained with those using a commercial antibody against all ITI chains and a bikunin-specific antibody. Western blot analysis confirmed the presence of an ITI H2 band at ~90 kDa and an ITI H2-bikunin band at ~140 kDa (Fig. 2A). Another band running at ~260 kDa suggested a combination of multiple heavy chains and bikunin (Fig. 2A). This is consistent with previous reports showing that the ITI heavy chain family members are characterized by a unique covalent bond between the heavy chains and the chondroitin sulfate chain of bikunin (14). Therefore, ITI heavy chain-bikunin complexes will be present on reducing PAGE gels. Taken together, these results confirm that the enriched proteins in our purified inhibitory fractions were ITI H2 and combinations of ITI H2 and/or other heavy chains and bikunin. They do not indicate whether ITI H2 alone is inhibiting C6 cell invasion or whether the serine protease inhibitory activity of bikunin is having a concomitant effect on the cells.


Figure 2
View larger version (37K):
[in this window]
[in a new window]
  		Figure 2. Western blot analysis confirms presence of ITI H2 in inhibitory fractions and ITI H2 localizes to perinuclear regions, membrane ruffles, and lamellipodia in U251 cells overexpressing ITI H2. A, Western blot analysis of purified inhibitory fractions from Resource Q and HiTrap Heparin columns. For the commercial ITI antibody, enriched bands at 260, 140, and 90 kDa were detected. The bikunin antibody detected a 260-kDa and a 140-kDa band, and the ITI H2 anti-serum detected a 140-kDa and a 90-kDa band. Note that all bands correspond to the same molecular weights using all three antibodies and represent various combinations of ITI heavy chains and bikunin. B, RT-PCR analysis of ITI H2, bikunin, and ß-actin expression in U251 stable cell lines. C, immunocytochemistry of U251 ITI H2 cells labeled with ITI H2 antibody (left), phalloidin Alexa 488 for actin (middle), and image overlay (right). Note the localization of ITI H2 in perinuclear regions, membrane ruffles, and thin, extensive lamellipodia. Actin colocalizes with ITI H2 in membrane ruffles and lamellipodia. Bar, 10 µm.

 
Stable overexpression of ITI H2 in U251 cells is localized to perinuclear regions and lamellipodia. We next wanted to validate the role of ITI H2 as an inhibitor of glioma cell motility by stably overexpressing both ITI H2 and the light chain bikunin in U251 cells. Infection of U251 cells with individual ITI H2 and bikunin chains should allow for the distinction between the effect of ITI H2 alone and/or the serine protease inhibitory activity of bikunin. To ensure that the effect of ITI H2 expression in U251 cells is not due to secondary effects arising from selection in culture, we generated two independent polyclonal cell lines for each DNA construct. For each round of infection and selection, the U251 ITI H2 cell line showed phenotypic changes in proliferation and attachment, which were not seen in the control U251 pLPCX or bikunin cell lines. RT-PCR analysis of stable cell lines showed that U251 ITI H2 and bikunin cell lines expressed only the respective ITI H2 and bikunin mRNAs whereas U251 cells infected with the control empty vector pLPCX did not have any ITI mRNA expression (Fig. 2B). Further characterization of U251 ITI H2 was done using immunocytochemistry to examine intracellular localization (Fig. 2C). U251 cells expressing ITI H2 showed increased ITI H2 localization in the perinuclear regions as well as in thin, extensive lamellipodia and membrane ruffles (Fig. 2C, left). Membrane ruffle and lamellipodia ITI H2 expression seemed to colocalize with actin (Fig. 2C, right).

ITI H2 inhibits U251 invasion in three-dimensional collagen gels. Because ITI H2 was originally identified from our functional screen for inhibitors of glioma cell invasion in vitro, it is important to confirm its role as an invasion inhibitor in our stable U251 ITI H2 cell line. We therefore implanted spheroids from U251 cells infected with the empty vector pLPCX, ITI H2, and bikunin in collagen type I gels, and monitored invasion for 3 days. Within the first 24 hours, cell invasion of U251 ITI H2 spheroids was inhibited by ~55% compared with pLPCX control spheroids and bikunin spheroids (Fig. 3A and C). The inhibitory effect was maintained after 3 days in culture (Fig. 3B and D). These results suggest that ITI H2 inhibits cell invasion in three dimensions. They do not differentiate between the effects of ITI H2 on individual cell invasive rate, adhesion, or proliferation.


Figure 3
View larger version (81K):
[in this window]
[in a new window]
  		Figure 3. Overexpression of ITI H2 inhibits U251 cell invasion in type I collagen gels. A, still photographs of U251 pLPCX, ITI H2, and bikunin spheroids 24 hours after implantation in collagen type I gels. Bar, 250 µm. B, still photographs of U251 pLPCX, ITI H2, and bikunin spheroids 72 hours after implantation. Bar, 250 µm. C to D, quantification of invasion distances for pLPCX, ITI H2, and bikunin spheroids after 24 hours (C) and 72 hours (D) in collagen type I gels. Note that invasion distances for ITI H2 spheroids are significantly lower than both pLPCX and bikunin spheroids. N = 21, N = 20, and N = 19 for U251 pLPCX, ITI H2, and bikunin spheroids, respectively, at both 24 and 72 hours. Bars, SE. ***, P < 0.001.

 
ITI H2 inhibits U251 cell proliferation without causing cell death. To examine the effect of ITI H2 and bikunin overexpression on cell proliferation in our U251 glioma model, we cultured U251 wild-type and U251 cells infected with control pLPCX, ITI H2, and bikunin in six-well dishes for 4 days. The total number of cells on day 2 and 4 was calculated for each cell line and results were converted to a doubling time. The doubling time of U251 ITI H2 cells was significantly higher than those of U251 wild-type, pLPCX, and bikunin cells, indicating a lower proliferation rate. The calculated in vitro doubling time for ITI H2 cells was 32.9 hours, and this was significantly different from the 25.6-, 26.8-, and 26-hour doubling times for U251 wild-type, pLPCX, and bikunin, respectively (Tukey's test, P < 0.01; Fig. 4A). There was no significant difference between U251 wild-type, pLPCX, and bikunin cell lines (Tukey's test; Fig. 4A).


Figure 4
View larger version (27K):
[in this window]
[in a new window]
  		Figure 4. Overexpression of ITI H2 in U251 cells inhibits proliferation without causing cell death. A, quantification of doubling time for U251 wild-type, pLPCX, ITI H2, and bikunin stable cell lines after 4 days in monolayer culture. Data represent three pooled independent trials. Bars, SE. **, P < 0.01. B, live-dead cell death/cytotoxicity assay for U251 pLPCX and ITI H2 cells in monolayer culture for 4 days. Top two rows, pLPCX and ITI H2 cells under normal culture conditions. Bottom two rows, pLPCX and ITI H2 cells pretreated with 0.1% saponin as a control for cell death/cytotoxicity. Bar, 200 µm.

 
To validate the role of ITI H2 as an inhibitor of glioma cell proliferation and not a promoter of cell cytotoxicity or death, a live-dead assay was done on control U251 cells infected with pLPCX and ITI H2 cells in monolayer culture over 4 days. As a control for possible induction of cell death, both infected cell lines were pretreated with 0.1% saponin for 10 minutes. U251 pLPCX and U251 ITI H2 cells that received no detergent treatment showed no evidence of cell death or cytotoxicity after 4 days in monolayer culture (Fig. 4B, top). All cells pretreated with 0.1% saponin died (Fig. 4B, bottom). Taken together, these results suggest that, in addition to its role in glioma cell invasion, ITI H2 significantly inhibits glioma cell proliferation without increasing cell death or cytotoxicity.

ITI H2 increases cell attachment in monolayers. To test the hypothesis that the inhibitory effect on cell proliferation may be accompanied by an increase in cell attachment, we cultured U251 pLPCX and ITI H2 cells on monolayer culture and counted the number of attached cells as a percentage of the total number of cells at 1, 2, 4, and 6 hours. After 4 to 6 hours, ITI H2 cells maintained a slightly more "spread-out" appearance on the plate and exhibited a small increase in the number of fanlike lamellipodia (see Supplementary Fig. S1A, insets). One and two hours after plating, there was a small increase in the number of attached cells for ITI H2 cultures; however, the effect was only significant at 4 hours (68% attached for U251 pLPCX and 80% attached for U251 ITI H2) and 6 hours [83% for U251 pLPCX and 88% attached for U251 ITI H2; P < 0.01 at 4 hours and P < 0.05 at 6 hours (t test); see Supplementary Fig. S1B]. These results suggest that the inhibitory effect on cell proliferation may be accompanied by an increase in cell attachment.

ITI H2 inhibits individual glioma cell velocity while increasing cell-cell adhesion. Previous results using U251 ITI H2 spheroids only examined average invasion of the entire cell population. To directly characterize the effect of ITI H2 on individual cell velocity and spheroid detachment, U251 pLPCX and U251 ITI H2 spheroids were implanted into collagen gels, allowed to invade for 6 hours, and then imaged using time-lapse videomicroscopy for the following 20 hours. This time period was chosen based on previous results obtained for doubling time. Because the doubling time for the pLPCX and ITI H2 cells is ~26.8 and 32.9 hours, respectively, significant differences in cell proliferation are not expected during the 20-hour video. The average invasion of U251 ITI H2 spheroids compared with U251 pLPCX spheroids over 20 hours is shown in Fig. 5A (see also Supplementary Videos 1 and 2). Quantification of individual cell invasive rates showed an average cell velocity of 15.5 and 8.8 µm/h for pLPCX and ITI H2 cells, respectively (N = 30 cells for each cell line tested; Fig. 5B). The inhibitory effect on cell velocity was also seen for detached U251 ITI H2 cells imaged 24 to 44 hours postimplantation (data not shown). In addition to the effect on individual cell velocity, 60% of U251 pLPCX cells had detached from the spheroids within 8 hours; however, only 17% of ITI H2 cells had detached, suggesting a role for ITI H2 in cell-cell adhesion (N = 30 cells for each cell line; Fig. 5C). There was a small inhibitory effect on ITI H2 cell proliferation but the effect was not statistically significant (Fig. 5C). These results further validate ITI H2 as an inhibitor of glioma cell invasion and show an additional potential role for ITI H2 in cell-cell adhesion.


Figure 5
View larger version (57K):
[in this window]
[in a new window]
  		Figure 5. Time-lapse videomicroscopy confirms ITI H2 as an inhibitor of individual cell velocity and promoter of cell-cell adhesion. A, still photographs depicting cell invasion for U251 pLPCX (top) and U251 ITI H2 (bottom) over 20 hours. Note that the representative U251 pLPCX spheroid (top) invades approximately twice the distance of the representative U251 ITI H2 spheroid (bottom). Bar, 250 µm. B, quantification of individual cell invasive rates for U251 pLPCX and U251 ITI H2 after 20 hours. N = 30 cells for each cell line. Bars, SE. ***, P < 0.001. C, quantification of individual cell proliferation and spheroid detachment for pLPCX and ITI H2 cells. N = 30 cells for each cell line. ***, P < 0.001.

 
ITI H2 expression is down-regulated in high-grade brain tumors. To explore the significance of our findings to primary brain tumors, we assessed the expression of ITI H2 in pilocytic astrocytoma, glioblastoma multiforme, low- and high-grade oligodendroglioma, meningioma, normal brain, and human fetal astrocyte cultures using RT-PCR. Representative samples are shown in Fig. 6. ITI H2 mRNA was detected for both cultures of primary human fetal astrocytes and normal adult brain (Fig. 6A). ITI H2 was present in 4 of 5 (80%) pilocytic astrocytoma but in only 2 of 11 (22%) glioblastoma multiforme samples (Fig. 6A). A similar trend was observed for oligodendroglioma tumors with 3 of 4 (75%) low-grade oligodendroglioma expressing ITI H2 mRNA and only 1 of 2 anaplastic oligodendroglioma exhibiting faint bands (Fig. 6A). Similarly, 4 of 5 (80%) nonmetastatic, low-grade meningioma expressed ITI H2. RT-PCR analysis was also conducted for glioma cell lines and faint levels of ITI H2 were detected for both U251 and U87 (Fig. 6B). No expression was detected in U373 (Fig. 6B). ITI H2 mRNA levels were higher in U343 (Fig. 6B). Interestingly, invasion rates of U343 spheroids in collagen type I gels are lower than those of U87, U251, and U373 spheroids.4


Figure 6
View larger version (58K):
[in this window]
[in a new window]
  		Figure 6. Expression of ITI H2 is down-regulated in high-grade brain tumors. A, RT-PCR analysis of ITI H2 and ß-actin mRNA levels in human fetal astrocyte cultures (HFA), normal brain (NB), pilocytic astrocytoma (PA), glioblastoma multiforme (GBM), meningioma (MG), low-grade oligodendroglioma (LO), and anaplastic oligodendroglioma (AO). B, RT-PCR analysis of ITI H2 and ß-actin mRNA levels in U251, U87, U343, and U373 glioma cell lines. Note the higher ITI H2 mRNA levels in low-grade tumors and normal brain samples. C, immunohistochemical analysis of ITI H2 protein expression in normal brain (top) and glioblastoma multiforme (bottom) representative paraffin sections. Samples were stained with anti-ITI H2 antibody and preimmune serum as a negative control. D, Western blot analysis of pAKT and p44/42 MAPK expression in U251 control pLPCX and ITI H2 cell lines. Note the down-regulation of pAKT in U251 ITI H2.

 
To examine ITI H2 protein expression patterns in primary human brain tumors and normal brain, we did immunohistochemical analysis. Normal brain samples stained with the ITI H2 antibody showed a strong perinuclear/cytoplasmic signal and a homogeneous staining of the extracellular matrix (Fig. 6C, top). In contrast, highly cellular glioblastoma multiforme samples showed practically no cellular localization and the observed ITI H2 signal was typically a weak and sporadic staining of the extracellular matrix (Fig. 6C, bottom). In two glioblastoma multiforme samples, we observed isolated clonal cell populations expressing ITI H2, but this pattern was very distinct from the homogeneous staining seen in normal brain and other lower-grade tumor samples (data not shown).

Taken together, these results suggest that ITI H2 is expressed in normal brain and low-grade CNS tumors including pilocytic astrocytoma and meningioma. This expression seems to be lost in higher-grade tumors including glioblastoma multiforme, suggesting that its loss is associated with an increased malignant potential.

ITI H2 down-regulates Akt phosphorylation. To explore possible signaling cascades associated with ITI H2–mediated inhibition of glioblastoma invasion and proliferation, we did Western blot analysis using phosphospecific antibodies for Akt and p44/42 MAPK. Results are shown in Fig. 6D. Compared with empty vector controls, overexpression of ITI H2 resulted in a down-regulation of pAKT but had no effect on p44/42 MAPK (Fig. 6D). These results suggest that the phosphatidylinositol 3-kinase/Akt pathway plays a role in the effects of ITI H2 on glioblastoma cell invasion and proliferation.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Over the last 20 years, there have been very few new developments in therapeutic strategies targeting malignant brain tumors. In this era of time-efficient production of cumulative lists of protein and gene players associated with a disease or cellular process of interest, one is still confronted with having to rationally choose a key player for further investigation. To avoid this dilemma, we opted to design and optimize a purification strategy to isolate natural inhibitors of glioma cell invasion from conditioned medium of C6 spheroids. In the present study, we have purified, identified, and validated ITI H2 as a natural inhibitor of glioma cell velocity and proliferation with a concomitant positive effect on cell-cell adhesion to produce an overall strong inhibition of human glioma cell invasion in vitro. The quantity and purity of our sample enabled us to identify ITI H2 by MS analysis of tryptic digest products as the most abundant protein in our purified fractions from conditioned medium.

The ITI proteoglycan family is an example of a protein-glycosaminoglycan-protein complex derived from alternative combinations of multiple heavy chains (ITI H1-H4) and one common light chain, bikunin, encoded by four genes on three different chromosomes (14, 15). Interestingly, the ITI H2 gene resides on the human chromosome 10p15 and deletions of portions of chromosome 10 are typically associated with malignant glioma progression (4, 15). Recently, a fifth heavy chain, ITI H5, has been identified and structural analysis has revealed early divergence from a common ITI ancestor (16).

The ITI heavy chain family members are characterized by a unique covalent bond between the heavy chains and the chondroitin sulfate chain of bikunin (14). In our study, silver staining of purified fractions revealed the presence of two enriched protein bands at 140 and 90 kDa. These bands correspond with those obtained from Western blot analysis using our antibody raised against the heavy chain 2, a bikunin antibody, and a commercial antibody that recognizes the various ITI chains, and confirm the presence of individual ITI H2 and a bikunin-bound form in the conditioned medium. Although ITI is principally produced by the liver and secreted into serum, individual heavy chain and bikunin expression have been detected in other tissues and are produced as components of ITI-related proteins in these tissues (17, 18). In particular, high levels of individual ITI H2 mRNA have also been detected in brain, adrenal gland, kidney, and lung (17, 18).

ITI H2 was identified in our functional selection for inhibitors of glioma cell invasion in type I collagen gels; however, this does not rule out the possibility that other heavy chain family members (H1, H3, H4, and H5) may also play a role in glioma cell motility and proliferation. For example, overexpression of ITI H1 and H3 induced a significant decrease in lung tumor metastasis number whereas ITI H2 had no effect (19). Recently, a cleavage fragment of ITI H4 has been identified as a marker for early-stage ovarian cancer and plasma bikunin levels are indicative of a favorable prognosis for ovarian cancer (20, 21). In addition, ITI H5 is down-regulated in breast cancer, thus further implicating the ITI heavy chains as tumor suppressor genes (16).

Using immunocytochemistry analysis of stable U251 ITI H2 cells, we showed that ITI H2 localizes to perinuclear regions as well as to thin extensive lamellipodia and membrane ruffles. This intracellular expression does not rule out the possibility that ITI H2 is secreted or is secreted in a modified form. We have not detected ITI H2 in the serum-free medium from our U251 stable cell line.5 It has been shown that proteolytic processing of ITI heavy chains during biosynthesis varies greatly with cell type, suggesting that cleavage is not autocatalytic and further reiterating the structural complexity of the ITI family (22). Colocalization of ITI H2 with actin in U251 membrane ruffles and lamellipodia suggests a relationship with cytoskeletal dynamics and a role in the control of invasion.

Our results show that overexpression of the individual ITI H2 chain decreases invasion of U251 glioma cells in collagen type I matrices whereas stable bikunin overexpression does not affect U251 cell proliferation or invasion. This indicates an exclusive role for ITI H2 in the control of cell invasion that is independent from bikunin. In addition to the invasion assays used here, bikunin has also been previously examined in glioma cell migration models where transient overexpression of placental bikunin alone did not significantly affect migration despite an inverse correlation of primary tumor mRNA levels with malignancy (23). Therefore, it is reasonable to assume that the ITI H2 isolated and identified in our purified fractions is individually responsible for the inhibitory effects on glioma cell invasion rather than the covalently bound bikunin.

An interesting finding in the presented study was the concomitant effect of ITI H2 on glioma cell proliferation and cell adhesion both in three-dimensional collagen gels and in monolayer cultures. Previous studies assessing the role of other molecules in glioma proliferation and motility have also shown an increase in cell attachment on a variety of brain extracellular matrices with a delay in tumor cell growth corresponding to the log phase of the growth curve (24). Overexpression of molecules that promote cell-cell adhesion (i.e., neural cell adhesion molecule) decreases glioma cell motility in vitro (25). Cadherins, as another example, can be classified as tumor suppressors and are associated with differentiation and contact inhibition of growth and motility (26). In our study, glioma cells overexpressing ITI H2 were less able to detach from the spheroid, and consequently, invasion was inhibited within the first 24 hours postimplantation. However, even individual cells that had already detached from the spheroid exhibited slower invasive rates. The additional inhibitory effect of ITI H2 on U251 cell doubling time suggests that ITI H2 targets multiple biological processes, making it an interesting therapeutic target.

An important finding in our study was the higher ITI mRNA and protein levels in lower-grade tumors and normal brain. Studies have also implicated other ITI chains, such as bikunin and an ITI H4 cleavage fragment, as biomarkers for early-stage ovarian cancer and are thus positive prognostic factors for this disease (20, 21). Our results show higher ITI H2 levels not only in normal brain and low-grade gliomas, such as pilocytic astrocytoma and oligodendroglioma, but also in other CNS tumors such as meningioma. Meningiomas are slow-growing, typically benign tumors that usually present themselves as WHO grade 1 (27). Interestingly, ITI H2 mRNA levels also seemed higher in cultures of human fetal astrocytes compared with samples of adult normal brain, suggesting a potential role for ITI H2 in neural development. The relationship of ITI H2 levels with tumor progression combined with the effects on proliferation, invasion, and adhesion in vitro suggest that ITI H2 may potentially serve as a biomarker for a variety of low-grade CNS tumors.

The signaling cascades associated with ITI heavy chain (H1-H5) overexpression are completely unknown. To investigate the possible intracellular pathways that could be involved with ITI H2 activity in our model, we considered two distinct pathways already known to participate in glioblastoma progression: the phosphatidylinositol 3-kinase/Akt and Ras pathways (28). The phosphatidylinositol 3-kinase pathway has been implicated in glioblastoma invasion, proliferation, and apoptosis (29–31) and down-regulation of pAKT by antisense urokinase-type plasminogen activator has also been shown to inhibit glioblastoma migration and growth without inducing apoptosis (29). Similarly, our ITI H2 cell lines showed inhibition of invasion and proliferation without a concomitant survival effect, together with down-regulation of pAKT, thus suggesting a role for the phosphatidylinositol 3-kinase/Akt pathway in our model. Further investigations will be necessary to unravel the other signaling pathways that are also involved in ITI H2 effects on glioblastoma cells.

The results presented here are the first to show an inhibitory role for ITI H2 in glioma cell invasion and proliferation while increasing cell attachment and adhesion. The positive correlation of invasion with proliferation indicates that, in our model, the two events are not mutually exclusive and that ITI H2 can be used to clinically target both processes. The higher ITI H2 levels in normal brain, fetal astrocytes, low-grade gliomas, and meningiomas suggest that ITI H2 may function as a tumor suppressor and potential biomarker for low-grade tumors. In our opinion, the classic biochemical approach presented here to identify a key player in the development of CNS tumor malignancy is quintessential to the significance of the outcome of the study. Targeting factors responsible for ITI H2 loss or restoring the supply exogenously may serve as potential therapeutic strategies for a variety of high-grade invasive CNS tumors.


   Acknowledgments
 
Grant support: Goals for Lily and the Alex Pavanel Family, the Franco Di Giovanni, the Barbara Jacobson and Carol Cuthbertson, and the Raymonde and Tony Boeckh Brain Tumour Research Funds; Canadian Institutes of Health Research Doctoral Research Award (T.E. Werbowetski-Ogilvie), Postdoctoral Fellowship from Fonds de la Recherche en Sante du Quebec (N.Y.R. Agar), and Chercheur Boursier Salary award from Fonds de la Recherche en Sante du Quebec (N. Jabado).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Carmen Sabau, Dr. Barry Bedell, Lily Li, and Dr. Marie-Christine Guiot for technical assistance (Montreal Neurological Institute, Montreal, Quebec); Dr. Hugh Bennett for advice and access to the AKTA fast protein liquid chromatography system (Sheldon Biotechnology Centre, Montreal, Quebec, Canada); and Dr. Marcos Di Falco and Susan James for MS analysis and technical assistance (McGill University and Genome Quebec Innovation Centre, Montreal, Quebec).


   Footnotes
 
Note: T.E. Werbowetski-Ogilvie and N.Y.R. Agar contributed equally to this work. R.F. Del Maestro holds the William Feindel Chair of Neurooncology and is a Killam Scholar at the Montreal Neurological Institute.

Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

4 Unpublished results. Back

5 Unpublished data. Back

Received 6/ 2/05. Revised 10/ 3/05. Accepted 12/ 2/05.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Chicoine M, Silbergeld DL. Mitogens as motogens. J Neurooncol 1997;35:249–57.[CrossRef][Medline]
  2. Kleihues P, Burger PC, Collins VP, Newcomb EW, Ohgaki H, Cavenee WK. Glioblastoma. In: Kleihues PC, Cavenee WK, editors. WHO classification of tumors: pathology and genetics of tumors of the nervous system. Lyon (France): Int. Agency for Res. On Cancer; 2000. p. 29–39.
  3. Mueller MM, Werbowetski T, Del Maestro RF. Soluble factors in glioma invasion. Acta Neurochir (Wien) 2003;145:999–1008.[Medline]
  4. Wechsler-Reya R, Scott MP. The developmental biology of brain tumors. Annu Rev Neurosci 2001;24:385–428.[CrossRef][Medline]
  5. Werbowetski T, Bjerkvig R, Del Maestro RF. Evidence for a secreted chemorepellent that directs glioma cell invasion. J Neurobiol 2004;60:71–88.[CrossRef][Medline]
  6. Tamaki M, McDonald W, Amberger VR, Moore E, Del Maestro RF. Implantation of C6 astrocytoma spheroid into collagen type I gels: invasive, proliferative, and enzymatic characterizations. J Neurosurg 1997;87:602–9.[Medline]
  7. Perkins DN, Pappin DJ, Creasy DM, Cottrell JS. Probability-based protein identification by searching sequence databases using mass spectrometry data. Electrophoresis 1999;20:3551–67.[CrossRef][Medline]
  8. Corcoran A, Del Maestro RF. Testing the "go or grow" hypothesis in human medulloblastoma cell lines in two and three dimensions. Neurosurgery 2003;53:174–84.[Medline]
  9. Corcoran A, De Ridder LI, Del Duca D, et al. Evolution of the brain tumor spheroid model: transcending current model limitations. Acta Neurochir (Wien) 2003;145:819–24.[CrossRef][Medline]
  10. Del Duca D, Werbowetski T, Del Maestro RF. Spheroid preparation from hanging drops: characterization of a model of brain tumor invasion. J Neurooncol 2004;67:295–303.[CrossRef][Medline]
  11. Angers-Loustau A, Hering R, Werbowetski TE, Kaplan DR, Del Maestro RF. SRC regulates actin dynamics and invasion of malignant glial cells in three dimensions. Mol Cancer Res 2004;2:595–605.[Abstract/Free Full Text]
  12. Del Maestro R, Shivers R, McDonald W, Del Maestro A. Dynamics of C6 astrocytoma invasion into three-dimensional collagen gels. J Neurooncol 2001;53:87–98.[CrossRef][Medline]
  13. Ruffini F, Arbour N, Blain M, Olivier A, Antel JP. Distinctive properties of human adult brain-derived myelin progenitor cells. Am J Pathol 2004;165:2167–75.[Abstract/Free Full Text]
  14. Zhuo L, Hascall VC, Kimata K. Inter-{alpha}-trypsin inhibitor, a covalent protein-glycosaminoglycan-protein complex. J Biol Chem 2004;279:38079–82.[Free Full Text]
  15. Diarra-Mehrpour M, Bourguignon J, Sesboue R, et al. Human plasma inter-{alpha}-trypsin inhibitor is encoded by four genes on three chromosomes. Eur J Biochem 1989;179:147–54.[Medline]
  16. Himmelfarb M, Klopocki E, Grube S, et al. ITI H5, a novel member of the inter-{alpha}-trypsin inhibitor heavy chain family is down-regulated in breast cancer. Cancer Lett 2004;204:69–77.[Medline]
  17. Mizushima S, Nii A, Katao K, Uemura A. Gene expression of the two heavy chains and one light chain forming the inter-{alpha}-trypsin-inhibitor in human tissues. Biol Pharm Bull 1998;21:167–9.[Medline]
  18. Chan P, Risler JL, Raguenez G, Salier JP. The three heavy-chain precursors for the inter-{alpha}-inhibitor family in mouse: new members of the multicopper oxidase protein group with differential transcription in liver and brain. Biochem J 1995;306:505–12.
  19. Paris S, Sesboue R, Delpech B, et al. Inhibition of tumor growth and metastatic spreading by overexpression of inter-{alpha}-trypsin inhibitor family chains. Int J Cancer 2002;97:615–20.[CrossRef][Medline]
  20. Zhang Z, Bast RC, Jr., Yu Y, et al. Three biomarkers identified from serum proteomic analysis for the detection of early stage ovarian cancer. Cancer Res 2004;64:5882–90.[Abstract/Free Full Text]
  21. Matsuzaki H, Kobayashi H, Yagyu T, et al. Plasma bikunin as a favorable prognostic factor in ovarian cancer. J Clin Oncol 2005;23:1463–72.[Abstract/Free Full Text]
  22. Thogersen IB, Enghild JJ. Biosynthesis of bikunin proteins in the human carcinoma cell line HepG2 and in primary human hepatocytes. J Biol Chem 1995;270:18700–9.[Abstract/Free Full Text]
  23. Hamasuna R, Kataoka H, Meng JY, et al. Reduced expression of hepatocyte growth factor activator inhibitor type-2/placental bikunin (HAI-2/PB) in human glioblastomas: implication for anti-invasive role of HAI-2/PB in glioblastoma cells. Int J Cancer 2001;93:339–45.[CrossRef][Medline]
  24. Rempel SA, Golembieski WA, Fisher JL, Maile M, Nakeff A. SPARC modulates cell growth, attachment and migration of glioma cells on brain extracellular matrix proteins. J Neurooncol 2001;53:149–60.[CrossRef][Medline]
  25. Prag S, Lepekhine EA, Kolkova K, et al. NCAM regulates cell motility. J Cell Sci 2002;115:283–92.[Abstract/Free Full Text]
  26. Demuth T, Berens ME. Molecular mechanisms of glioma cell migration and invasion. J Neurooncol 2004;70:217–28.[CrossRef][Medline]
  27. Perry A, Gutmann DH, Reifenberger G. Molecular pathogenesis of meningiomas. J Neurooncol 2004;70:183–202.[CrossRef][Medline]
  28. Guha A, Mukherjee J. Advances in the biology of astrocytomas. Curr Opin Neurol 2004;17:655–62.[CrossRef][Medline]
  29. Chandrasekar N, Mohanam S, Gujrati M, Olivero WC, Dinh DH, Rao JS. Down-regulation of uPA inhibits migration and PI3K/Akt signaling in glioblastoma cells. Oncogene 2003;22:392–400.[CrossRef][Medline]
  30. Kubiatowski T, Jang T, Lachyankar MB, et al. Association of increased phosphatidylinositol 3-kinase signaling with increased invasiveness and gelatinase activity in malignant gliomas. J Neurosurg 2001;95:480–8.[Medline]
  31. Shi Q, Bao S, Maxwell JA, et al. Secreted protein acidic, rich in cysteine (SPARC), mediates cellular survival of gliomas through AKT activation. J Biol Chem 2004;279:52200–9.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Molecular Cancer TherapeuticsHome page
T. Demuth, L. B. Reavie, J. L. Rennert, M. Nakada, S. Nakada, D. B. Hoelzinger, C. E. Beaudry, A. N. Henrichs, E. M. Anderson, and M. E. Berens
MAP-ing glioma invasion: Mitogen-activated protein kinase kinase 3 and p38 drive glioma invasion and progression and predict patient survival
Mol. Cancer Ther., April 1, 2007; 6(4): 1212 - 1222.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Werbowetski-Ogilvie, T. E.
Right arrow Articles by Del Maestro, R. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Werbowetski-Ogilvie, T. E.
Right arrow Articles by Del Maestro, R. F.

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