help button home button Am J Pathol JNCI
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Purchase Article
Right arrow View Shopping Cart
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 Sultana, S.
Right arrow Articles by Skalli, O.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Sultana, S.
Right arrow Articles by Skalli, O.
(American Journal of Pathology. 1998;153:1157-1168.)
© 1998 American Society for Investigative Pathology

Regular Articles

Effects of Growth Factors and Basement Membrane Proteins on the Phenotype of U-373 MG Glioblastoma Cells as Determined by the Expression of Intermediate Filament Proteins

Sadath Sultana, Rixin Zhou, Madhu S. Sadagopan and Omar Skalli

From the Department of Anatomy and Cell Biology, University of Illinois at Chicago, Chicago, Illinois


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
Various growth factors and basement membrane proteins have been implicated in the pathobiology of astrocytomas. The goal of this study was to determine the relative contribution of these two factors in modulating the phenotype of U-373 MG glioblastoma cells as determined by the expression of the intermediate filament proteins glial fibrillary acidic protein, vimentin, and nestin. For these determinations, cells plated in serum-free medium were treated either with growth factors binding to tyrosine kinase receptors including transforming growth factor-{alpha}, epidermal growth factor, platelet-derived growth factor-AA, basic fibroblast growth factor, and insulin-like growth factor-1 or with basement membrane proteins including collagen IV, laminin, and fibronectin. The changes in the expression levels of intermediate filament proteins in response to these treatments were analyzed by quantitation of immunoblots. The results demonstrate that collagen IV and growth factors binding to tyrosine kinase receptors decrease the glial fibrillary acidic protein content of U-373 MG cells. Growth factors binding to tyrosine kinase receptors also decrease the vimentin content of these cells but do not affect their nestin content. On the other hand, basement membrane proteins decrease the nestin content of U-373 MG cells but do not affect their vimentin content. The significance of these results with respect to the role played by different factors in modulating the phenotype of neoplastic astrocytes during tumor progression is discussed.


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
Growth factors as well as extracellular matrix proteins are instrumental in regulating the degree of differentiation of most cell types. It is interesting in this respect that the malignant transformation of astrocytes is accompanied by the secretion of various growth factors binding to tyrosine kinase receptors as well as by increased interactions between neoplastic astrocytes and basement membrane (BM) proteins. Little is known, however, about the respective contributions of these events in modulating the degree of differentiation of neoplastic astrocytes.

Growth factors that can be secreted by neoplastic astrocytes include transforming growth factor-{alpha} (TGF-{alpha}),1-3 platelet-derived growth factor (PDGF),3-7 basic fibroblast growth factor (bFGF),3,8-10 and insulin-like growth factor-1 (IGF-1).11-14 These growth factors bind to specific tyrosine kinase receptors which can also be expressed by neoplastic astrocytes, suggesting that autocrine and/or paracrine stimulation of astrocytes plays an important role in the pathobiology of astrocytic tumors. This notion is also supported by the fact that in neoplastic astrocytes some of these tyrosine kinase receptors may be overexpressed or constitutively activated as a consequence of various genetic alterations.4,5,15-17 In fact, one of the most frequent genetic alterations found in neoplastic astrocytes is the amplification of the epidermal growth factor (EGF) receptor.15-17 In addition to the activation of various tyrosine kinase receptors, the malignant transformation of astrocytes frequently entails increased interactions between neoplastic astrocytes and BM proteins such as collagen IV, laminin, and fibronectin. This is due to overexpression of integrin receptors by neoplastic astrocytes,18 as well as to blood vessel proliferation resulting in augmentation of the area of contact between neoplastic astrocytes and the BM surrounding these vessels.

In this study we have determined the relative contributions of growth factors and BM proteins in regulating the degree of differentiation of neoplastic astrocytes as assessed by the expression of the intermediate filament proteins glial fibrillary acidic protein (GFAP), vimentin and nestin. GFAP is usually considered to be one of the best markers of astrocytic differentiation as its expression is largely specific of differentiated astrocytes.19-23 In contrast, vimentin and nestin are present in the precursor cells of astrocytes but not in mature astrocytes.24-29 The utility of intermediate filament proteins in assessing the degree of astrocytic differentiation is further underscored by the fact that during neoplasia, as astrocytes progress toward a higher degree of malignancy, their intermediate filament protein composition becomes increasingly similar to that of undifferentiated astrocytes. Indeed, in astrocytic tumors the proportion of cells expressing GFAP and the GFAP content of these tissues decrease with increasing tumor grade.30-35 In addition, vimentin and nestin are reexpressed by neoplastic astrocytes.29,36-38 While vimentin is found in neoplastic astrocytes irrespective of their tumor grade,37,38 the proportion of neoplastic astrocytes expressing nestin augments with increasing malignancy.29,36

We used the glioblastoma cell line U-373 MG for our studies because we determined that it is the only one of four astrocytoma cell lines tested that co-expresses GFAP, vimentin, and nestin. Expression of these proteins was examined in cells plated in serum-free, chemically defined medium (CDM) to which TGF-{alpha}, EGF, PDGF-AA, bFGF, or IGF-1 were added. Alternatively, the cells were plated in CDM on Petri dishes coated with collagen IV, laminin, or fibronectin. The changes in the protein levels of GFAP, vimentin, and nestin in response to these different conditions were monitored by quantitation of immunoblots. The results of our study demonstrate that growth factors binding to tyrosine kinase receptors as well as collagen IV decrease the GFAP content of U-373 MG astrocytoma cells, resulting in a phenotype similar to that of neoplastic astrocytes in vivo. Because GFAP is a marker of astrocytic differentiation, this suggests that in astrocytic tumors growth factors binding to tyrosine kinase receptors cooperate with some BM proteins to decrease the degree of differentiation of neoplastic astrocytes. Our results with vimentin and nestin, however, suggest that in astrocytic tumors the cytoskeletal phenotype of neoplastic astrocytes is also modulated by agents other than the growth factors and BM proteins tested here.


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Growth Factors and Commercial Antibodies

Growth factors were purchased from Gibco Co. (Grand Island, NY) and included human recombinant TGF-{alpha}, murine natural EGF, human recombinant PDGF-AA, human recombinant bFGF, and human recombinant IGF-I.

GFAP or vimentin mouse monoclonal antibodies were purchased from Sigma (St. Louis, MO) and mouse monoclonal anti-nestin 40126 was obtained from Chemicon (Temecula, CA). Rabbit anti-bovine brain GFAP was purchased from Incstar (Stillwater, MN). The secondary antibodies were obtained from Kirkegaard & Perry (Gaithersburg, MD) and included affinity purified, human serum adsorbed, goat anti-mouse or anti-chicken or anti-rabbit IgGs conjugated to peroxidase for immunoblotting, or either fluorescein or rhodamine for immunofluorescence.

Cell Culture

U-373 MG, U-87 MG and U-118 MG human glioblastoma cells and Hs 683 human glioma cells were obtained from the American Type Culture Collection (ATCC, Manassas, VA). To maintain frozen stocks, each cell line was grown and trypsinized following the recommendation of the ATCC.

Cells from frozen stocks were thawed and plated at a density of 5 x 104 cells/cm2 in serum-free CDM consisting of minimal essential medium with Earle's salts supplemented with 406 mg/L L-alanyl-L-glutamine (GlutaMax, Gibco), 10 µg/ml insulin, 5.5 µg/ml transferrin, 6.7 ng/ml sodium selenite, 100 U/ml penicillin, and 100 µg/ml streptomycin. L-Alanyl-L-glutamine was added as a stable substitute for L-glutamine, and insulin, transferrin, and selenium were included for optimal viability of cells in CDM.39 The medium was changed the day after plating and appropriate growth factors were added at a final concentration of 10 ng/ml. At the same time an identical volume of vehicle was added to control cultures.

In experiments involving BM proteins, the cells were plated in CDM as described above but using Petri dishes (Becton Dickinson, Bedford, MA) coated with either human fibronectin (10 µg/cm2) purified from human plasma or coated with mouse collagen IV (15 µg/cm2) or mouse laminin (15 µg/cm2), both purified from Engelbreth-Holm-Swarm tumors.

Production and Affinity Purification of Nestin Antibodies

The immunogen used to raise nestin antibodies was a 14-amino-acid-long peptide which sequence is unique to nestin (as determined by a SwissProt database search), but common to rat and human nestin (for comparison of the two sequences see Ref. 40 ). A cysteine residue was added to the C terminus of that peptide to enable conjugation to carrier proteins. The peptide, QEFLQARTPTLASTC, was synthesized at the Biochemistry Core Facility of the University of Illinois at Chicago. The peptide was conjugated to keyhole limpet hemocyanin and to bovine serum albumin with m-maleimidobenzoyl-N-hydroxysuccinimide ester as previously described.41 Immunization of hens with the nestin peptide-keyhole limpet hemocyanin conjugate was performed following standard procedures.41

IgGs were purified from the egg yolk of immunized hens by precipitation with 14% (w/v) ammonium sulfate. Affinity purification of anti-nestin IgG was carried out on nestin peptide-bovine serum albumin conjugate covalently bound to aminophenylthioether cellulose (APT Transa-Bind paper, Schleicher and Schuell, Keene, NH) as previously described,42 except that antibodies binding to the peptide were eluted with 5 mol/L NaI. Eluted antibodies were immediately run through a 0.45-mµ pore filter (Millipore, Bedford, MA) and dialyzed against several volumes of phosphate-buffered saline (PBS) consisting of 6 mmol/L phosphate buffer, 170 mmol/L NaCl, 3 mmol/L KCl, pH 7.4.

Immunofluorescence

Cells plated on coverslips were washed briefly in PBS and fixed for 5 minutes in acetone at -20°C. All antibodies were diluted at 1:20 in PBS. Cells were incubated with primary and secondary antibodies for 30 minutes each. Each incubation was followed by three 5-minute washes in PBS. After the last wash, the cells were mounted in Fluoromount G (Southern Biotechnology Associates, Birmingham, AL).

Microscopic observations were carried out with an Axiophot microscope (Carl Zeiss, Oberkochen, Germany) equipped with epi-illumination and specific filters for rhodamine and fluorescein. Photographs were taken on Tmax 400 black and white films (Kodak, Rochester, NY).

Sample Preparation for Electrophoresis and Determination of Protein and DNA Concentration

Cells from one 75-cm2 Petri dish were washed twice with PBS and scraped off with a rubber policeman in 50 µl of sample buffer consisting of 3% sodium dodecyl sulfate (SDS), 8 mol/L urea, 60 mmol/L Tris/HCl (pH 6.8), 10% (v/v) glycerol, 5 mmol/L EDTA, 0.001% (w/v) bromophenol blue, 1 mmol/L phenylmethylsulfonylfluoride, 1 mmol/L N-p-tosyl-L-arginine metylester. The viscosity of the cell lysate was reduced by sonication. Boiling the sample was avoided because it would have led to protein carbamylation by the urea present in the sample buffer.43 ß-Mercaptoethanol was added to each sample to a final concentration of 3% (v/v) after completion of the protein and DNA measurements. This is because reducing agents interfere with the assays we used to measure DNA and proteins.44,45

Protein concentration measurements were done with the bicinchoninic acid (BCA) assay45 following the enhanced protocol described by the manufacturer (Pierce, Rockford, IL). Bovine serum albumin was used for the standard curve. Protein concentration values of samples containing cells plated on BM proteins were corrected to account for these extraneous proteins. To perform this correction, we prepared samples containing only BM proteins by scraping off BM protein-coated dishes with sample buffer. Sample buffer was then added to these samples to equalize their volume with that of samples containing cells plated on the same BM proteins. The protein concentration of samples containing only a given BM protein were subtracted from that of samples containing cells plated on that BM protein.

DNA concentration measurements were performed with the diphenylamine method44 using salmon sperm DNA for the standard curve. Briefly, 5 µl of sample was added to 400 µl of 10% perchloric acid and heated at 70°C for 20 minutes. To the mixture were then added 800 µl of 1.5% (w/v) diphenylamine in pure acetic acid and 40 µl of 0.2% (v/v) acetaldehyde in water. Samples were incubated overnight in the dark at 25°C and absorbance was measured at 600 nm. DNA concentration values were used to estimate the concentration of cells in each sample by using the formula cells/ml = (µg DNA/ml)/(3.4 x 10-6). This formula is based on the fact that the DNA content of a human diploid nucleus is about 3.4 pg.46

Gel Electrophoresis and Immunoblotting

Gel electrophoresis of proteins was performed on 7.5% SDS-polyacrylamide gels as previously described.47 Staining of electrophoresed proteins was performed with 0.05% (w/v) Coomassie Brilliant Blue in 50% methanol and 10% acetic acid. Destaining was done with 10% acetic acid.

For immunoblotting, the proteins separated by SDS-polyacrylamide gel electrophoresis were electrophoretically transferred onto nitrocellulose membranes48 at 60 V overnight at 4°C. Non-specific protein binding sites on the nitrocellulose blots were blocked with PBS containing 5% nonfat milk (PBS-milk). The primary antibodies were diluted in PBS-milk at the following dilutions: 1:5,000 for mouse monoclonal anti-GFAP and mouse monoclonal anti-vimentin and 20 µg/ml for affinity purified hen anti-nestin IgG. Incubation of the nitrocellulose blots with these antibodies was performed for 4 hours at room temperature. The blots were then washed with PBS plus 0.1% Nonidet P-40 and incubated for 3 hours at room temperature with the appropriate peroxidase-conjugated antibodies diluted at 1:1000 in PBS-milk. After washes in PBS plus 0.1% NP-40, the peroxidase activity of the nitrocellulose-bound secondary antibodies was detected with the chromogenic substrate 3,3',5,5'-tetramethlybenzine (Kirkegaard & Perry).

Quantitation of GFAP, Vimentin, and Nestin in Whole Cell Extracts of U-373 MG Cells

The GFAP, vimentin, and nestin protein content of U-373 MG cells treated with growth factors or plated on BM proteins (experimental samples) was compared to that of cells plated in CDM (control sample). For these comparisons, either equal amounts of protein or an equal number of cells (ie, equal amounts of DNA) were loaded for control and experimental samples on the same gel and transferred onto nitrocellulose paper, allowing for direct comparison between control and experimental samples. To quantify GFAP and vimentin, 3 µg of proteins or 3 x 104 cells (corresponding to 0.1 µg of DNA) were loaded on the gel for each sample. Ten times more material was loaded on gels used to quantify nestin, as this protein is less abundant than GFAP and vimentin in U-373 MG cells.

The density of the bands corresponding to GFAP, vimentin, or nestin on immunoblots was measured by scanning these immunoblots with a flat-bed, 2400-dpi resolution scanner (HP ScanJet 4c, Hewlett-Packard, Palo Alto, CA) and by analyzing the scanned images with the Sigma Gel software (Jandel Scientific, Chicago). Only density values falling within the linear portion of the density range of the scanner were considered. For each experiment, the percentage of change in GFAP, vimentin, or nestin protein content relative to control was calculated as follows: 100x(1-(density of the band in experimental sample/density of the band in control sample)) if the change was a decrease or: 100 x (1 + (density of the band in experimental sample/density of the band in control sample)) if the change was an increase. For each condition, the values of at least seven different experiments were used to calculate means ± SEM and to perform Student's t-test or one-way analysis of variance.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
Intermediate Filament Protein Composition of Various Human Astrocytoma Cells Lines and Specificity of Nestin Antibodies

To select for our studies an astrocytoma cell line expressing GFAP, vimentin, and nestin, we conducted a series of immunoblotting experiments to examine the presence of these proteins in U-373 MG, U-87 MG, U-118 MG and Hs 683 human astrocytoma cells plated in CDM. All four cell lines were found to express vimentin, which appeared on immunoblots as a 57-kd band reacting with vimentin antibodies (Figure 1) . In contrast, immunoblotting experiments showed that GFAP is expressed only in U-373 MG cells (Figure 1) .



View larger version (60K):
[in this window]
[in a new window]
  		Figure 1. a: Coomassie Blue-stained SDS-polyacrylamide gel loaded with whole cell extracts of U-373 MG (lane 1), U-87 MG (lane 2), U-118 MG (lane 3 ), and Hs 683 (lane 4) cells. Molecular mass are indicated in kilodaltons at the left of the gel. b-d: Blots of gels similar to a incubated with the following primary antibodies: anti-nestin (b), anti-GFAP (c), and anti-vimentin (d). Note that U373-MG is the only cell line to co-express GFAP and nestin.

 
The presence of nestin in U-373 MG, U-87 MG, U-118 MG and Hs 683 human astrocytoma cells was examined with an antibody raised against a 14 amino acid peptide derived from the C-terminal domain of nestin and conjugated to keyhole limpet hemocyanin (see Materials and Methods). This antibody was specific for nestin because on immunoblots of P6 rat cerebellum it labeled only a 240-kd protein comigrating with nestin as determined with the nestin monoclonal antibody 40126 (data not shown). In contrast to the nestin monoclonal antibody 401,29 our antibody also reacted with human nestin. This was demonstrated by the fact that on immunoblots of U-373 MG cells, which have been shown previously to express nestin,29 our antibody only labeled a 280-kd band (Figure 1) . The difference in electrophoretic mobility between human and rat nestin is in good agreement with that reported in other studies29 and is likely due to the extensive sequence differences that exist between rat and human nestin.40 Immunofluorescence experiments confirmed the presence of nestin in U-373 MG cells (Figure 2) . Double immunofluorescence with nestin and vimentin antibodies also showed that in U-373 MG cells the overall subcellular distribution of nestin is very similar to that of vimentin, consisting of a network of wavy cytoplasmic filaments (Figure 2) . However, nestin antibodies seem to yield more diffuse cytoplasmic fluorescence than vimentin antibodies. While nestin is expressed by U-373 MG cells, that protein could not be detected by either immunoblotting (Figure 1) or immunofluorescence in U-87 MG, U-118 MG, and Hs 683 astrocytoma cells.



View larger version (137K):
[in this window]
[in a new window]
  		Figure 2. Double immunofluorescent staining of U-373 MG glioblastoma cells with vimentin (a) and nestin (b) antibodies. Nestin antibodies stain a filamentous network similar to that stained by anti-vimentin. Scale bar = 10 µm.

 
Changes in the Protein Levels of GFAP, Vimentin, and Nestin in Response to TGF-{alpha}, EGF, PDGF-AA, bFGF, and IGF-1

U-373 MG cells were chosen for these experiments because they co-express GFAP, vimentin, and nestin (see above). The cells were plated at a low density (5 x 104 cells/cm2) to minimize the influence of cell-cell contacts on cell differentiation. In addition, the cells were plated in medium without serum, but supplemented with insulin, transferrin, and selenium. The addition of these three components to the medium was essential for good viability and attachment of U-373 MG cells in the absence of serum. Under these conditions, most U-373 MG cells attached to the Petri dish and extended several short processes, resulting in a spiny cell shape. After the addition of growth factors to the medium, the cells became more flattened and extended additional short processes.

We determined by two methods the changes in the protein levels of GFAP, vimentin, and nestin in growth factor-treated U-373 MG cells compared to controls. In the first method, these changes were determined by measuring GFAP, vimentin, and nestin as percentage of total proteins using blots obtained from gels loaded with equal amount of proteins for control and experimental samples. In the second method, the changes were determined by measuring the amount of GFAP, vimentin, and nestin per cell using blots obtained from gels loaded with equal number of cells for control and experimental samples. In all cases there was no statistically significant difference between the values obtained with these two methods (Figures 3–5) . This shows the reliability of our protein and DNA assays and also indicates that modifications in the protein levels of GFAP, vimentin, and nestin reflect changes specific to the turnover of each of these three proteins rather than changes due to increase or decrease in the total amount of protein per cell.



View larger version (44K):
[in this window]
[in a new window]
  		Figure 3. Effect of growth factors on GFAP protein levels in U-373 MG cells. a: Blot corresponding to a gel loaded with equal amounts of proteins; b: blot corresponding to a gel loaded with equal number of cells. In a and b, the lane labeled C shows the result for the control sample, and subsequent lanes show the results for cells treated with different growth factors as indicated above each lane. c: Histogram showing the changes in GFAP protein levels in growth factor treated cells relative to control cells. Values are calculated from seven different experiments and are expressed as means ± SEM. Black bars represent values determined by comparing blots obtained from gels loaded with equal amounts of proteins, and gray bars represent values determined by comparing blots obtained from gels loaded with equal number of cells. Changes that are statistically different from control are indicated by * (P <= 0.001) or # (P = 0.005).

 


View larger version (39K):
[in this window]
[in a new window]
  		Figure 4. Effect of growth factors on vimentin protein levels in U-373 MG cells. a: Blot corresponding to a gel loaded with equal amounts of proteins; b: blot corresponding to a gel loaded with equal number of cells. In a and b, the lane labeled C shows the result for the control sample, and subsequent lanes show the results for cells treated with different growth factors as indicated above each lane. c: Histogram showing the changes in vimentin protein levels in growth factor treated cells relative to control cells. Values are calculated from seven different experiments and are expressed as means ± SEM. Black bars represent values determined by comparing blots obtained from gels loaded with equal amounts of proteins, and gray bars represent values determined by comparing blots obtained from gels loaded with equal number of cells. Changes that are statistically different from control are indicated by * (P <= 0.001), # (P <= 0.005) or ## (P = 0.013).

 


View larger version (33K):
[in this window]
[in a new window]
  		Figure 5. Effect of growth factors on nestin protein levels in U-373 MG cells. a: Blot corresponding to a gel loaded with equal amounts of proteins; b: blot corresponding to a gel loaded with equal number of cells. In a and b, the lane labeled C shows the result for the control sample and subsequent lanes show the results for cells treated with different growth factors as indicated above each lane. c: Histogram showing the changes in nestin protein levels in growth factor treated cells relative to control cells. Values are calculated from seven different experiments and are expressed as means ± SEM. Black bars represent values determined by comparing blots obtained from gels loaded with equal amounts of proteins, and gray bars represent values determined by comparing blots obtained from gels loaded with equal number of cells. Changes that are statistically different from control are indicated by * (P = 0.022) or # (P = 0.015).

 
The most dramatic effect of growth factors was the induction of a large and statistically significant (P <= 0.001 to P = 0.005) decrease in the GFAP levels of U-373 MG cells compared to untreated cells (Figure 3) . This decrease was most pronounced with TGF-{alpha} and EGF (ligands of the EGF receptor), which reduced GFAP protein levels by about 50%. PDGF-AA, IGF-1, and bFGF induced a decrease in GFAP levels that was 25–35% of the control levels. The decrease induced by IGF-1 and bFGF was significantly different (P < 0.005) from that induced by TGF-{alpha} and EGF. The decrease in GFAP protein levels obtained after treatment of U-373 MG cells with PDGF-AA was intermediate between that obtained with TGF-{alpha} and EGF on the one hand and with IGF-1 and bFGF on the other (Figure 3) , but was not statistically different from the decrease induced by any of these growth factors. Immunofluorescence experiments showed that the decrease in GFAP levels following treatment with growth factors was not due to an increase in the number of GFAP negative cells (about 20% in control and growth factor treated cells), but rather to a diminution in the amount of GFAP per cell as shown by markedly reduced fluorescence intensity in growth factor treated cells (data not shown).

When compared to controls, vimentin protein levels in U-373 MG cells treated with TGF-{alpha}, EGF, PDGF-AA, bFGF, and IGF-1 decreased by 15–25% (Figure 4) . This decrease was statistically significant for all growth factors (P < 0.005, except for bFGF where P = 0.013) (Figure 4) . When compared to each other, there was no statistically significant difference between the effect of the different growth factors on the protein levels of vimentin.

Nestin protein levels were not significantly altered by EGF, bFGF, and IGF-1 (Figure 5) . TGF-{alpha} increased nestin protein levels in some experiments (Figure 5) , but not in others. Because of the large SEM resulting from this variability, the 17% average increase in nestin protein levels in TGF-{alpha}-treated U-373 MG cells was not found to be statistically significant. PDGF-AA led to a decrease of approximately 15% in the protein levels of nestin; this decrease was at the limit of statistical significance (0.013 <= P <= 0.022) (Figure 5) .

Changes in Protein Levels of GFAP, Vimentin, and Nestin in Response to Collagen IV, Laminin, and Fibronectin

To examine the role of BM proteins in regulating the intermediate filament composition of U-373 MG cells, these cells were plated at a low density (5 x 104 cells/cm2) in CDM on regular Petri dishes (control) or on Petri dishes coated with collagen IV, laminin, or fibronectin. Under all conditions the cells extended several processes. However, the processes extended by cells plated on BM proteins were much longer and thinner than those extended by control cells (data not shown).

The changes relative to control in the protein levels of GFAP, vimentin, and nestin in U-373 MG cells plated on BM proteins were determined as described above for our experiments with growth factors. There was no statistically significant difference between the values obtained by comparing equal amounts of proteins for control and experimental samples and those obtained by comparing equal numbers of cells for control and experimental samples (Figures 6–8) . This indicates that changes in the protein levels of GFAP, vimentin, and nestin of U-373 MG cells plated on different BM proteins reflect modifications in the turnover of these proteins. It also indicates the reliability of the correction made to take into account the presence of extraneous proteins in samples containing cells plated on BM proteins (see Materials and Methods).



View larger version (42K):
[in this window]
[in a new window]
  		Figure 6. Effect of BM proteins on GFAP protein levels in U-373 MG cells. a: Blot corresponding to a gel loaded with equal amounts of proteins; b: blot corresponding to a gel loaded with equal number of cells. In a and b, the lane labeled C shows the result for the control sample, and subsequent lanes show the results for cells plated on collagen IV (COL), laminin (LMN), or fibronectin (FN). c: Histogram showing the changes in GFAP protein levels in cells plated on BM proteins relative to control cells. Values are calculated from seven different experiments and are expressed as means ± SEM. Black bars represent values determined by comparing blots obtained from gels loaded with equal amounts of proteins, and gray bars represent values determined by comparing blots obtained from gels loaded with equal number of cells. Changes that are statistically different from control are indicated by * (P <= 0.001).

 


View larger version (30K):
[in this window]
[in a new window]
  		Figure 7. Effect of BM proteins on vimentin protein levels in U-373 MG cells. a: Blot corresponding to a gel loaded with equal amounts of proteins; b: blot corresponding to a gel loaded with equal number of cells. In a and b, the lane labeled C shows the result for the control sample, and subsequent lanes show the results for cells plated on collagen IV (COL), laminin (LMN), or fibronectin (FN). c: Histogram showing the changes in vimentin protein levels in cells plated on BM proteins relative to control cells. Values are calculated from seven different experiments and are expressed as means ± SEM. Black bars represent values determined by comparing blots obtained from gels loaded with equal amounts of proteins, and gray bars represent values determined by comparing blots obtained from gels loaded with equal number of cells. Changes are not statistically different from control.

 


View larger version (46K):
[in this window]
[in a new window]
  		Figure 8. Effect of BM proteins on nestin protein levels in U-373 MG cells. a: Blot corresponding to a gel loaded with equal amounts of proteins, b: blot corresponding to a gel loaded with equal number of cells. In a and b, the lane labeled C shows the result for the control sample, and subsequent lanes show the results for cells plated on collagen IV (COL), laminin (LMN), or fibronectin (FN). c: Histogram showing the changes in nestin protein levels in cells plated on BM proteins relative to control cells. Values are calculated from seven different experiments and are expressed as means ± SEM. Black bars represent values determined by comparing blots obtained from gels loaded with equal amounts of proteins, and gray bars represent values determined by comparing blots obtained from gels loaded with equal number of cells. Changes that are statistically different from control are indicated by * (P <= 0.001) or # (0.001<P <= 0.005).

 
Our results show that BM proteins either decreased or did not alter the levels of intermediate filament proteins in U-373 MG cells. In the case of GFAP, different BM proteins affected differently the levels of that protein. Thus, collagen IV decreased GFAP levels by about 30% compared to control, a statistically significant decrease (P <= 0.001). The plating of U-373 MG cells on laminin and fibronectin, on the other hand, did not correlate with a statistically significant change in the protein levels of GFAP (Figure 6) . Similarly, the protein levels of vimentin were not altered by plating U-373 MG cells on BM proteins (Figure 7) . BM proteins, however, strongly modulated nestin levels in U-373 MG cells. When compared to control, these levels were decreased by 45–55% by collagen IV and laminin and by about 25% by fibronectin (Figure 8) . These changes were statistically significant, with P <= 0.001 for cells plated on collagen IV or laminin and P = 0.005 for cells plated on fibronectin (Figure 8) . In addition, the decrease in nestin protein levels induced by plating U-373 MG cell on fibronectin was statistically different (P < 0.05) from that induced by plating the cells on collagen IV or laminin.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
Two prominent events associated with the malignant transformation of astrocytes are the activation of tyrosine kinase receptors by growth factors and the increased interactions between neoplastic astrocytes and BM proteins at the level of proliferating capillaries. The goal of our study was to determine the contribution of these two events in determining the cytoskeletal phenotype of neoplastic astrocytes. Immunohistochemical studies of astrocytic tumors have shown that this phenotype is characterized by the presence of vimentin in cells of all tumor grades,37,38 with decreasing levels of GFAP30-35 and increasing levels of nestin29,36 as the tumor grade increases. Our results demonstrate that in vitro under serum free conditions, collagen IV and growth factors binding to tyrosine kinase receptors decrease the GFAP content of U-373 MG glioblastoma cells. This suggests that these agents may be responsible for the progressive loss of GFAP that occurs in astrocytic tumors. We also report here that growth factors binding to tyrosine kinase receptors decrease the vimentin content of U-373 MG cells, and that BM proteins including collagen IV, laminin, and fibronectin decrease the nestin content of these cells. This suggests that in astrocytic tumors factors other than those tested here are responsible for inducing and/or increasing vimentin and nestin expression.

Studies of the genetic changes giving rise to astrocytic tumors have led to the recognition that glioblastoma multiforme can be subdivided into at least two types. Type I is characterized by p53 mutations and type II is characterized by the amplification of the EGF receptor gene.17,49 U-373 MG cells appear to be similar to type I cells since they exhibit mutations in the p53 gene50-52 but do not overexpress the EGF receptor.53,54 Hence, the cytoskeletal changes taking place in U-373 MG cells in response to growth factors and BM proteins may be typical of type I glioblastoma cells but not necessarily of glioblastoma cells with other genetic alterations. However, based on our finding that a large decrease in GFAP content occurs in U-373 MG cells treated with EGF or TGF-{alpha}, it is reasonable to propose that these growth factors may produce an even more dramatic decrease in the GFAP content of neoplastic astrocytes overexpressing the EGF receptor. Because bFGF55 and EGF56 decrease GFAP expression in primary cultures of rat astrocytes, it is also possible that decreased GFAP expression is a general response of astrocytes, whether normal or neoplastic, to growth factors binding to tyrosine kinase receptors.

Because the growth factors used in our study bind to tyrosine kinase receptors, our results show that, in U-373 MG cells, a common effect of the activation of these receptors is a decrease in GFAP levels. The decrease in GFAP was greater in cells treated with TGF-{alpha}, EGF, or PDGF than in cells treated with bFGF or IGF-1. Since each of these growth factors was added to the culture medium to the same final concentration, this raises the possibility that U-373 MG cells possess more receptors for TGF-{alpha}, EGF, and PDGF than for bFGF and IGF-1. U-373 MG cells have been found to overexpress the {alpha}-chain of the PDGF receptor57 and to contain an average number of EGF receptors (for which both EGF and TGF-{alpha} are ligands).53,54 However, we do not know the number of bFGF or IGF-1 binding sites present at the surface of U-373 MG cells. It is therefore difficult to determine at the present time whether quantitative differences seen in the effect of TGF-{alpha}, EGF, PDGF, bFGF, and IGF-1 on GFAP protein levels are due to differences in the number and/or affinity of the receptors specific for each of these growth factors or to subtle variations in the signaling pathway downstream of these receptors.

Low levels of GFAP are considered to reflect a low degree of astrocytic differentiation.22 The results of our experiments therefore indicate that stimulation of U-373 MG glioblastoma cells by various growth factors binding to tyrosine kinase receptors leads to a less differentiated phenotype. This notion is also supported by several in vitro studies that have shown that various growth factors enhance the proliferation and invasive properties of neoplastic astrocytes.15 Interestingly, neoplastic astrocytes in tumors can secrete and possess receptors for TGF-{alpha},1,3 PDGF,3-7 bFGF,3,8-10 and IGF-1,11,12,14 and the levels of expression of these growth factors and their receptors increase with increasing tumor grade.3 Although no direct comparison has been made between levels of expression of GFAP and those of growth factors in astrocytic tumors, it is interesting to notice that in these tumors GFAP protein levels decrease with increasing degree of astrocytic maligancy.30-35 Taken together with our in vitro results, these in vivo findings suggest that in astrocytic tumors the degree of differentiation of neoplastic astrocytes may decrease following autocrine or paracrine stimulation by growth factors such as TGF-{alpha}, PDGF, bFGF, and IGF-1. Additionally, loss of astrocytic differentiation could arise as a consequence of genetic alterations that frequently affect the EGF receptor in high-grade astrocytomas and may result in the constitutive activation of this receptor in some cases.15-17 That such mutations could decrease the degree of astrocytic differentiation is suggested by our finding that stimulation of the EGF receptor occurring through EGF or TGF-{alpha} elicits a sharp decrease in the GFAP protein levels of U-373 MG cells.

We report that under serum-free conditions the plating of U-373 MG cells on collagen IV correlates with a statistically significant decrease in their GFAP content. Laminin and fibronectin elicited a modest decrease in the GFAP content of these cells, but this decrease was not statistically significant. Previous studies have demonstrated that the motility as well as the adhesive and invasive properties of U-373 MG cells are enhanced more effectively by collagen than by fibronectin or laminin.18,58 The preferential response of U-373 MG to collagen may be due to the particular profile of integrin receptors expressed by these cells.58 Indeed, in vitro the motility and/or invasive behavior of astrocytoma cells in response to various BM proteins has been shown to depend on the integrin receptors expressed by these cells.59-62 In the brain, the integrin profile of astrocytes may also influence their response to BM proteins. BM proteins occur only exceptionally around normal and neoplastic astrocytes and therefore, in both normal brain and astrocytic tumors, the interactions between astrocytes and BM proteins are restricted to the BM of blood vessels.63-65 However, during neoplasia the integrin profile of astrocytes is modified, and it has been suggested that these modifications could alter the behavior of these cells in response to BM proteins.18 Changes in the expression of integrins by astrocytes during neoplasia include the overexpression of {alpha}3 and ß1 integrins and the de novo expression of {alpha}6, {alpha}v, and ß3 integrins.18 Interestingly, this integrin composition is very similar to that of U-373 MG cells.18 Collagen IV in vitro diminishes the GFAP content of U-373 MG cells, which suggests that in astrocytic tumors the interaction of neoplastic cells with collagen IV at the level of the BM of blood vessels may lead to a decrease in GFAP protein levels.

Our results suggest that in the tumor environment, growth factors binding to tyrosine kinase receptors and at least some BM proteins may cooperate to reduce the level of GFAP expression by neoplastic astrocytes. Other factors, such as cytokines, could also contribute to this decline. TNF, for instance, is present in some astrocytic tumors66,67 and has been shown to decrease GFAP levels in primary cultures of astrocytes.68 One should be aware, however, that in the complex tumor environment the effect of various factors on the phenotype of neoplastic astrocytes may be different from that seen in vitro. For instance, although collagen IV decreases GFAP levels in U-373 MG glioblastoma cells in vitro (this study), in astrocytic tumors GFAP immuno-peroxidase staining is increased in several neoplastic astrocytes contacting small blood vessels or invading leptomeningeal connective tissue.69 It is therefore possible that BM proteins other than those tested here increase the GFAP content of neoplastic astrocytes in vivo. Alternatively, changes in GFAP immuno-peroxidase staining may not always reflect actual changes in GFAP protein content as measured biochemically.70-72

The response of neoplastic astrocytes to different agents may also be affected by serum proteins present in the tumor environment in necrotic or hemorrhagic areas. This is suggested by the finding that in vitro, EGF appears to enhance GFAP expression in neonatal or in malignant astrocytes grown in the presence of serum73,74 but decreases GFAP levels in the absence of serum.56 Similarly, the presence of serum in the culture medium diminishes the mitogenic response of neoplastic astrocytes to EGF and PDGF.75 It is also intriguing that collagen IV increases GFAP levels in U-373 MG cells grown in the presence of serum76 but decreases these levels under serum-free conditions (this study). These discrepancies may also be due to the degree of cell confluence at the time of sampling because in our study the cells were still sparse at the time of sampling, whereas in an earlier study76 they were confluent. At any rate, the use of serum-free culture conditions under well defined cell density conditions is important to distinguish the contribution of individual factors to the phenotype of neoplastic astrocytes.

The results of various transfection studies suggest that alterations in GFAP expression levels, such as those seen in astrocytic tumors or in U-373 MG cells in response to growth factors and collagen IV, could have important consequences for the malignant behavior of neoplastic astrocytes. For instance, the transfection of an antisense GFAP cDNA in astrocytoma cells abolished GFAP expression and resulted in increased proliferation and invasive properties of the transfected cells.77 The transfected cells also exhibited reduced adhesion to the substratum77 and decreased ability to extend cytoplasmic processes when grown in the presence of neurons.78 Conversely, when astrocytoma cells lacking GFAP were induced to synthesize that protein by transfection of a sense GFAP cDNA, the transfected cells gained the ability to extend cytoplasmic processes and displayed a reduced rate of proliferation as well as a decreased ability to form colonies in soft agar.79

The importance of GFAP in maintaining some of the differentiated features of astrocytes is also supported by the fact that this protein appears to be the prime target of growth factor action compared to the other intermediate filament proteins expressed by U-373 MG cells. Thus, the decrease in the vimentin content of U-373 MG cells in response to growth factors acting on tyrosine kinase receptors was smaller than that of GFAP. The nestin content of U-373 MG cells was not altered by EGF, bFGF, and IGF-1 and was only marginally affected by TGF-{alpha} and PDGF-AA. Northern blotting experiments will help to determine whether or not TGF-{alpha} and PDGF-AA may effect nestin expression because changes in mRNA levels often precede and are more pronounced than changes in protein levels. BM proteins had no effect on vimentin levels in U-373 MG cells but decreased nestin protein levels more prominently than they reduced GFAP levels. This demonstrates that within the same cell type different intermediate filament proteins are differently regulated, perhaps to fulfill separate functional requirements. The impact of changes in vimentin or nestin expression on the malignant behavior of astrocytoma cells is not known. It is, however, interesting to note that in breast carcinoma cells, the induction of vimentin synthesis by transfection of a vimentin cDNA enhances the motility and invasiveness of transfected cells.80

During development there is an inverse correlation between GFAP expression on one hand and vimentin and nestin expression on the other hand.20,23-25,27-29,81 In this context it is intriguing that in U-373 MG cells the decrease in GFAP levels in response to growth factors and collagen IV is paralleled by either a decrease or a lack of change in the protein levels of nestin and vimentin. In other situations, however, there is no inverse correlation between GFAP and vimentin/nestin expression. For instance, the few astrocytoma cell lines found to date to express nestin also contain GFAP29 (this study). GFAP remains also co-expressed with vimentin in subpopulations of astrocytes in the adult brain.28,82,83 Perhaps the most striking example of the fact that GFAP expression levels are not always inversely correlated with vimentin and nestin expression levels is provided by reactive astrocytes, which contain high amounts of GFAP84 together with nestin85,86 and vimentin.84 This suggests that the expression of vimentin and nestin may not always correlate with an undifferentiated astrocytic phenotype but could also depend on conditions resulting in cellular stress, such as brain damage. It also remains to be seen whether some or all of the cells expressing nestin in astrocytic tumors could be reactive astrocytes. In any case, our results suggest that factors other than those tested here up-regulate vimentin and nestin in astrocytic tumors. Further studies of the expression of these two proteins in neoplastic astrocytes may help in identifying such factors. Vimentin expression, for instance, could be controlled by factors involved in the early stages of the malignant transformation of astrocytes, as suggested by the finding that vimentin is expressed in astrocytic tumors of all grades.37,38

In addition to furthering our understanding of the mechanisms responsible for tumor progression, identifying the factors involved in the establishment of a malignant phenotype is important to the development of rational therapeutic strategies. In this respect, our study suggests that inhibition of the signaling pathway lying downstream of tyrosine kinase receptors may impact the phenotype of neoplastic astrocytes more than inhibition of a particular tyrosine kinase receptor. This is because even if the neoplastic cells in astrocytic tumors can be stimulated by growth factors binding to different tyrosine kinase receptors,3 all of these growth factors seem to induce similar phenotypic changes in neoplastic astrocytes, as suggested by our results. However, in astrocytic tumors inhibition of tyrosine kinase receptors signaling pathways may be necessary but not sufficient to bring about a more differentiated astrocytic phenotype, as suggested by our finding that at least some BM proteins affect the astrocytic phenotype in a manner similar to that of growth factor binding to tyrosine kinase receptors.


   Acknowledgements
 
We thank Mrs. Virginia Kriho for excellent technical assistance during the early phase of this study.


   Footnotes
 
Address reprint requests to Dr. Omar Skalli, Department of Anatomy and Cell Biology, University of Illinois at Chicago, 808 S. Wood Street, M/C 512, Chicago, IL 60612. E-mail: oskalli{at}uic.edu

Supported by National Institutes of Health Grant NS-35317.

Accepted for publication June 25, 1998.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Maruno M, Kovach JS, Kelly PJ, Yanagihara T: Transforming growth factor-{alpha}, epidermal growth factor receptor, and proliferating potential of benign and malignant gliomas. J Neurosurg 1991, 75:97-102[Medline]
  2. Samuels V, Barrett JM, Bockman S, Pantazis CG, Allen MB: Immunocytochemical study of transforming growth factor expression in benign and malignant gliomas. Am J Pathol 1989, 134:895-902
  3. Van der Valk P, Lindeman J, Kamphorst W: Growth factor profiles of human gliomas. Do non-tumour cells contribute to tumour growth in glioma? Ann Oncol 1997, 8:1023-1029[Abstract/Free Full Text]
  4. Fleming TP, Saxena A, Clark WC, Robertson JT, Oldfield EH, Aaronson SA, Ali IU: Amplification and/or overexpression of platelet derived growth factor receptors and epidermal growth factor receptors in human glial tumors. Cancer Res 1992, 52:4550-4553[Abstract/Free Full Text]
  5. Hermanson M, Funa K, Hartman M, Claesson Welsh L, Heldin CH, Westermark B, Nister M: Platelet-derived growth factor and its receptors in human glioma tissue: expression of messenger RNA and protein suggests the presence of autocrine and paracrine loops. Cancer Res 1992, 52:3213-3219[Abstract/Free Full Text]
  6. Maxwell M, Naber SP, Wolfe HJ, Galanopoulos T, Hedley-Whyte ET, Black PM, Antoniades HN: Coexpression of platelet-derived growth factor (PDGF) and PDGF receptor gene by primary human astrocytomas may contribute to their development and maintenance. J Clin Invest 1990, 86:131-140
  7. Nister M, Westermark B: Clonal variation in the production of a PDGF-like protein and expression of corresponding receptors in human gliomas. Cancer Res 1986, 46:332-340[Abstract/Free Full Text]
  8. Morrison RS, Yamaguchi F, Saya H, Bruner JM, Yahanda AM, Donehower LA, Berger M: Basic fibroblast growth factor and fibroblast growth factor receptor I are implicated in the growth of human astrocytomas. J Neuro-Oncol 1994, 18:207-216[Medline]
  9. Paulus W, Grothe C, Sensebrenner M, Janet T, Baur I, Graf M, Roggendorf W: Localization of basic fibroblast growth factor, a mitogen and angiogenic factor in human brain tumors. Acta Neuropathol 1990, 79:418-423[Medline]
  10. Takahashi JA, Mroi H, Fukumoto M, Igarashi K, Jaye M, Oda Y, Kikuchi H, Hatanaka M: Gene expression of fibroblast growth factor in human gliomas and meningiomas: demonstration of cellular source of basic fibroblast growth factor mRNA and peptide in tumor tissues. Proc Natl Acad Sci USA 1990, 87:5710-5714[Free Full Text]
  11. Glick RP, Unterman TG, Van der Woude M, Zollner Blaydes L: Insulin and insulin-like growth factors in the central nervous system tumors. PartV: Production of insulin-like growth factors I and II in vitro. J Neurosurg 1992, 77:445-450[Medline]
  12. Merrill MJ, Edwards N: Insulin-like growth factor-I receptors in human glial tumors. J Clin Endocr Metab 1990, 71:199-209[Abstract]
  13. Sandberg NA, Stahlbom PA, Reinecke M, Collins VP, von Holst H, Sara V: Characterization of insulin-like growth factor 1 in human primary brain tumors. Cancer Res 1993, 53:2475-2478[Abstract/Free Full Text]
  14. Sara VR, Prissel P, Sjogren B, Persson L, Boethius J, Enberg G: Enhancement of insulin-like growth factor and receptors in glioblastoma. Cancer Lett 1986, 32:229-234[Medline]
  15. Finn PA, Bjerkvig R, Pilkington GJ: The role of growth factors in the malignant and invasive progression of intrinsic brain tumors. Anticancer Res 1997, 17:4163-4172[Medline]
  16. Funari FB, Huang HJS, Cavenee WK: Molecular biology of malignant degeneration of astrocytoma. Pediatric Neurosurg 1996, 24:41-49
  17. von Deimling A, Louis DN, Wiestler OD: Molecular pathways in the formation of gliomas. Glia 1995, 15:328-338[Medline]
  18. Paulus W, Baur I, Schuppan D, Roggendorf W: Characterization of integrin receptors in normal and neoplastic human brain. Am J Pathol 1993, 143:154-163[Abstract]
  19. Ho C-L, Liem RKH: Intermediate filaments in the central nervous system: implication in tumors. Canc Metast Rev 1996, 15:483-497
  20. Landry CF, Ivy GO, Brown IR: Developmental expression of glial fibrillary acidic protein mRNA in the rat brain analyzed by in situ hybridization. J Neurosci Res 1990, 25:194-203[Medline]
  21. Levitt P, Cooper ML, Rakic P: Early divergence and changing proportions of neuronal and glial precursor cells in the primate cerebral ventricular zone. Dev Biol 1983, 96:472-484[Medline]
  22. Rutka JT, Murakami M, Dirks PB, Hubbard SL, Becker LE, Fukuyama K, Jung S, Tsugu A, Matsuzawa K: Role of glial filaments in cells and tumors of glial origin: a review. J Neurosurg 1997, 87:420-430[Medline]
  23. Voigt T: Development of glial cells in the cerebral wall of ferrets: direct tracing of their transformation from radial glia into astrocytes. J Comp Neurol 1989, 289:74-88[Medline]
  24. Dahl D, Rueger DC, Bignami A, Weber K, Osborn M: Vimentin, the 57,000 dalton protein of fibroblast filaments, is the major cytoskeletal component in immature glia. Eur J Cell Biol 1981, 24:191-196[Medline]
  25. Dahlstrand J, Lardelli M, Lendahl U: Nestin mRNA expression correlates with the central nervous system progenitor cell state in many, but not all, regions of developing central nervous system. Dev Br Res 1995, 84:109-129
  26. Lendahl U, Zimmerman LB, McKay RDG: CNS stem cells express a new class of intermediate filament protein. Cell 1990, 60:585-595[Medline]
  27. Pixley SK, De Vellis J: Transition between radial glia and mature astrocytes studied with a monoclonal antibody to vimentin. Dev Br Res 1984, 15:201-209
  28. Schnitzer J, Franke WW, Schachner M: Immunocytochemical demonstration of vimentin in astrocytes and ependymal cells of developing and adult mouse nervous system. J Cell Biol 1981, 90:435-447[Abstract/Free Full Text]
  29. Toyama T, Lee VMY, Rorke LB, Marvin M, McKay RDG, Trojanowski JQ: Nestin expression in embryonic human neuroepithelium and in human neuroepithelial tumor cells. Lab Invest 1992, 66:303-313[Medline]
  30. Duffy PE, Huang YY, Rapport MM, Graf L: Glial fibrillary acidic protein in giant cell tumors of brain and other gliomas. Acta Neuropathol (Berl.) 1980, 52:51-57[Medline]
  31. Jacque CM, Vinner C, Kujas M, Raoul M, Racadot J, Baumann NA: Determination of glial fibrillary acidic protein (GFAP) in human brain tumors. J Neurol Sci 1978, 35:147-155[Medline]
  32. Jacque CM, Kujas M, Poreau A, Raoul M, Collier P, Racadot J, Baumann N: GFA and S100 protein levels as index for malignancy in human gliomas and neurinomas. J Natl Canc Inst 1979, 62:479-483
  33. Tascos NA, Parr J, Gonatas NK: Immunohistochemical study of the glial fibrillary acidic protein in human neoplasms of the central nervous system. Hum Pathol 1982, 13:454-458[Medline]
  34. Van der Meulen JDM, Houthoff HJ, Ebels EJ: Glial fibrillary acidic protein in human gliomas. Neuropathol Appl Neurobiol 1978, 4:177-190[Medline]
  35. Velasco ME, Dahl D, Roessmann U, Gambetti P: Immunohistochemical localization of glial filbrillary acidic protein in human glial neoplasms. Cancer 1980, 45:484-494[Medline]
  36. Dahlstrand J, Collins PV, Lendhal U: Expression of the class VI intermediate filament nestin in human central nervous tumors. Cancer Res 1992, 52:5334-5341[Abstract/Free Full Text]
  37. Herpers MJHM, Ramaekers FCS, Alderweireldt J, Moesker O, Slooff J: Co-expression of glial fibrillary acidic protein and vimentin-type intermediate filaments in human astrocytomas. Acta Neuropathol (Berl.) 1986, 70:333-339[Medline]
  38. Yung WK, Luna M, Borit A: Vimentin and glial fibrillary acidic protein in human brain tumors. J Neuro-Oncol 1985, 3:35-38[Medline]
  39. Bottenstein J, Hayashi I, Hutchings S, Masui H, Mather J, McClure DB, Ohasa S, Rizzino A, Sato G, Serrero G, Wolfe R, Wu R: The growth of cells in serum-free hormone-supplemented media. Methods Enzymol 1979, 58:94-109[Medline]
  40. Dahlstrand J, Zimmerman LB, McKay RDG, Lendahl U: Characterization of human nestin gene reveals a close evolutionary relationship to neurofilaments. J Cell Sci 1992, 103:589-597[Abstract]
  41. Harlow E, Lane D: Antibodies: A Laboratory Manual. 1988:pp 82-83 Cold Spring Harbor Laboratory Press Cold Spring Harbor, NY
  42. Olmsted JB: Affinity purification of antibodies from diazotized paper blots of heterogenous protein samples. J Biol Chem 1981, 256:11955-11957[Abstract/Free Full Text]
  43. Stark GR, Stein WH, Moore S: Reactions of the cyanate present in aqueous urea with amino acids and proteins. J Biol Chem 1960, 235:3177-3181[Free Full Text]
  44. Burton K: Determination of DNA concentration with diphenylamine. Methods Enzymol 1968, 12(part B):163-166
  45. Smith PK, Krohn RI, Hermanson GT, Mallia AK, Gartner FH, Provenzano MD, Fujimoto EK, Goeke NM, Olson BJ, Klenk DC: Measurement of proteins using bicinchoninic acid. Anal Biochem 1985, 150:76-85[Medline]
  46. Lewin B: Gene Expression, vol. 2, 2nd ed. New York, Wiley, 1980, p 958
  47. Laemmli UK: Cleavage of structural proteins during the assembly of the head of the bacteriophage T4. Nature 1974, 227:680-685
  48. Towbin M, Staehelin Y, Gordon J: Electrophoretic transfer of proteins from acrylamide gels to nitrocellulose sheets: procedure and some applications. Proc Natl Acad Sci USA 1985, 76:4350-4357
  49. von Deimling A, von Ammon K, Schoenfeld D, Wiestler OD, Seizinger BR, Louis DN: Subsets of glioblastoma multiforms defined by molecular genetic analysis. Brain Pathol 1993, 3:19-26[Medline]
  50. Godbout R, Miyakoshi J, Dobler KD, Andison R, Matsuo K, Allalunis TM, Takabe H, and Day R III: Lack of expression of tumor suppressor genes in human malignant glioma cell lines. Oncogene 7:1879–1884
  51. Van Meir EG, Kikuchi T, Tada M, Li H, Diserens AC, Wojcik BE, Su Huang HJ, Friedman T, de Tribolet N, Cavenee WK: Analysis of the p53 gene and its expression in human glioblastoma cells. Cancer Res. 54:649–652
  52. Ullrich SJ, Mercer WE, Appella E: Human wild-type p53 adopts a unique conformational and phosphorylation state in vivo during growth arrest of glioblastoma cells. Oncogene 1992, 7:1635-1643[Medline]
  53. Bigner SH, Humphrey PA, Wong AJ, Vogelstein B, Mark J, Friedman J, Bigner D: Characterization of the epidermal growth factor receptor in human glioma cell lines and xenografts. Cancer Res 1990, 50:8017-8022[Abstract/Free Full Text]
  54. Sang UH, Espiritu OD, Kelley PY, Klauber MR, Hatton JD: The role of the epidermal growth factor receptor in human gliomas: I. the control of cell growth. J Neurosurg 1995, 82:841-846[Medline]
  55. Reilly JF, Maher PA, Kumari VG: Regulation of astrocyte GFAP expression by TGF-ß1 and FGF-2. Glia 1998, 22:202-210[Medline]
  56. Avola R, Condorelli DF, Surrentino S, Turpeenoja L, Costa A, Giuffrida Stella AM: Effect of epidermal growth factor and insulin on DNA, RNA and cytoskeletal protein labeling in primary rat astroglial cell cultures. J. Neurosci Res 1988, 19:230–238
  57. Nister M, Claesson-Welsh L, Eriksson A, Heldin CH, Westermark B: Differential expression of platelet-derived growth factor receptors in human malignant glioma cell lines. J Biol Chem 1991, 266:16755-16763[Abstract/Free Full Text]
  58. Ohnishi T, Arita N, Hiraga S, Taki T, Izumoto S, Fukushima Y, Hayakawa T: Fibronectin-mediated cell migration promotes glioma cell invasion through chemokinetic activity. Clin Exp Metastasis 1997, 15:538-546[Medline]
  59. Berens ME, Rief MD, Loo MA, Giese A: The role of the extracellular matrix in human astrocytoma migration and proliferation studied in a microliter scale assay. Clin Exp Metastasis 1994, 12:405-415[Medline]
  60. Giese A, Rief MD, Loo MA, Berens ME: Determinants of human astrocytoma migration. Cancer Res 1994, 54:3897-3904[Abstract/Free Full Text]
  61. Haugland HK, Tysnes BB, Tysnes OB: Adhesion and migration of human glioma cells are differently dependent on extracellular matrix molecules. Anticancer Res 1997, 17:1035-1043[Medline]
  62. Paulus W, Tonn JC: Basement membrane invasion of glioma cells mediated by integrin receptors. J Neurosurg 1994, 80:515-519[Medline]
  63. Bellon G, Caulet T, Cam Y, Pluot M, Poulin G: Immunohistochemical localization of macromolecules of the basement membrane and extracellular matrix of human gliomas and meningiomas. Acta Pathol (Berl.) 1985, 66:245-252
  64. Jones TR, Ruoslahti E, Schold SC, Bigner D: Fibronectin and glial fibrillary acidic protein expression in normal human brain and anaplastic human gliomas. Cancer Res 1982, 42:168-177[Abstract/Free Full Text]
  65. Paulus W, Roggendorf W, Schuppan D: Immunohistochemical investigation of collagen subtypes in human glioblastomas. Virchows Arch A Pathol Anat 1988, 413:325-332
  66. Bethea JR, Chung IY, Sparacio GY, Benveniste EN: Interleukin-1 induction of tumor necrosis-{alpha} gene expression in human astroglial cells. J Neuroimmunol 1992, 36:179-191[Medline]
  67. Schneider J, Hofman FM, Apuzzo ML, Hinton DR: Cytokines and immunoregulatory molecules in malignant glial neoplasms. J Neurosurg 1992, 77:265-173[Medline]
  68. Selmaj K, Shafit-Zagardo B, Aquino DA, Farooq M, Raine CS, Norton WT, Brosnan CF: Tumor necrosis factor-induced proliferation of astrocytes from mature brain is associated with down-regulation of glial fibrillary acidic protein mRNA. J Neurochem 1991, 57:823-830[Medline]
  69. Herpers MJHM, Budka H, McCormick D: Production of glial fibrillary acidic protein (GFAP) by neoplastic cells: adaptation to the microenvironment. Acta Neuropathol (Berl.) 1984, 64:33-338
  70. Aquino DA, Chiu FC, Brosnan CF, Norton WT: Glial fibrillary acidic protein increases in the spinal cord of Lewis rats with acute experimental autoimmune encephalitis. J Neurochem 1988, 51:1085-1096[Medline]
  71. Eng LF, Ghirnikar RS: GFAP and astrogliosis. Brain Pathol 1995, 4:229-237
  72. Norton WT, Aquino DA, Hozumi I, Chiu F-C, Brosnan CF: Quantitative aspects of gliosis: a review. Neurochem Res 1992, 17:877-885[Medline]
  73. Avola R, Magri G, Ingrao L, Insirello L, Carpano P, Nicoletti G, Condorelli DF, Ragusa N, Giuffrida Stella AM: Effect of EGF on DNA labeling in rat cerebellar immature astrocytes maintained under different culture conditions: presence or absence of polylysine, serum or both. Ann NY Acad Sci 1991, 633:540-542[Medline]
  74. Sang U H, Espiritu OD, Kelley PY, Klauber MR, Hatton JD: The role of the epidermal growth factor receptor in human gliomas: II. The control of glial process extension and the expression of glial fibrillary acidic protein. J Neurosurg 1995, 82:847-857[Medline]
  75. Pollack I, Randall MS, Kristofik MP, Kelly RH, Selker RG, Vertosick FT: Response of malignant glioma cell lines to epidermal growth factor and platelet-derived growth factor in a serum-free medium. J Neurosurg 1990, 73:106-112[Medline]
  76. Rutka JT, Giblin JR, Apodaca G, DeArmond SJ, Stern R, Rosenblum ML: Inhibition of growth and induction of differentiation in a malignant glioma cell line by normal leptomeningeal extracellular matrix proteins. Cancer Res 1987, 3515–3522
  77. Toda M, Miura M, Asou H, Toya S, Uyemura K: Cell growth suppression of astrocytoma C6 cells by glial fibrillary acidic protein cDNA transfection. J Neurochem 1994, 1978, 63:1975
  78. Weinstein DE, Shelanski ML, Liem RKH: Suppression by antisense mRNA demonstrates a requirement for the glial fibrillary acidic protein in the formation of stable astrocytic processes in response to neurons. J Cell Biol 1991, 112:1205-1213[Abstract/Free Full Text]
  79. Rutka JT, Smith SL: Transfection of human astrocytoma cells with glial fibrillary acidic protein complementary cDNA: analysis of expression, proliferation and tumorigenicity. Cancer Res 1993, 53:3624-3631[Abstract/Free Full Text]
  80. Hendrix MJC, Seftor EA, Seftor REB, Trevor KT: Experimental co-expression of vimentin and keratin intermediate filaments in human breast cancer cells results in phenotypic interconversion and increased invasive behavior. Am J Pathol 1997, 150:483-495[Abstract]
  81. Yachnis AT, Rorke L, Lee VMY, Trojanowski JQ: Expression of neuronal and glial polypeptides during histogenesis of the human cerebellar cortex including observations on the dentate nucleus. J Comp Neurol 1993, 334:356-369[Medline]
  82. Calvo JL, Carbonell AL, Boya J: Coexpression of vimentin and glial fibrillary acid protein in astrocytes of the adult rat optic nerve. Brain Res 1990, 532:355-357[Medline]
  83. Shaw G, Osborn M, Weber K: An immunofluorescence microscopical study of the neurofilament triplets proteins, vimentin and glial fibrillary acidic protein within the adult rat brain. Eur J Cell Biol 1981, 26:68-82[Medline]
  84. Bignami A, Dahl D: Gliosis. Neuroglia: Edited by H Kettenmann and BR Ransom. 1995, :pp 843-858 Oxford University Press, Oxford
  85. Clarke SR, Shetty AK, Bradley JL, Turner DA: Reactive astrocytes express the embryonic intermediate neurofilament nestin. Neuroreport 1994, 5:1885-1888[Medline]
  86. Frisen J, Johansson CB, Torok C, Risling M, Lendahl U: Rapid, widespread, and long-lasting induction of nestin contributes to the generation of glial scar tissue after CNS injury. J Cell Biol 1995, 131:453-464[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 J. Histochem. Cytochem.Home page
E. Bertelli, M. Regoli, L. Fonzi, R. Occhini, S. Mannucci, L. Ermini, and P. Toti
Nestin Expression in Adult and Developing Human Kidney
J. Histochem. Cytochem., April 1, 2007; 55(4): 411 - 421.
[Abstract] [Full Text] [PDF]

Home page
 J. Cell Sci.Home page
R. Jing, U. Wilhelmsson, W. Goodwill, L. Li, Y. Pan, M. Pekny, and O. Skalli
Synemin is expressed in reactive astrocytes in neurotrauma and interacts differentially with vimentin and GFAP intermediate filament networks
J. Cell Sci., April 1, 2007; 120(7): 1267 - 1277.
[Abstract] [Full Text] [PDF]