Proteomic Investigation of Glioblastoma Cell Lines Treated with Wild-Type p53 and Cytotoxic Chemotherapy Demonstrates an Association between Galectin-1 and p53 Expression Maja Puchades,† Carol L. Nilsson,*,‡ Mark R. Emmett,‡ Kenneth D. Aldape,§ Yongjie Ji,§ Frederick F. Lang,§ Ta-Jen Liu,§ and Charles A. Conrad§ Institute of Neuroscience and Physiology, Sahlgrenska Academy, Go¨teborg University, SU/Mo¨lndal, SE-43180 Mo¨lndal, Sweden, National High Magnetic Field Laboratory, Florida State University, 1800 E. Paul Dirac Dr., Tallahassee, Florida, and M.D. Anderson Cancer Center, 1515 Holcombe Blvd., Houston, Texas 77030 Received June 21, 2006
Abstract: Global protein analysis of treated and untreated glioblastoma cell lines was performed. Proteomic analysis revealed the identity of proteins that were significantly modulated by the treatment with wild-type TP53 and the cytotoxic chemotherapy SN38. In particular, galectin-1 was found to be negatively regulated by transfection with TP53 and further down-regulated by SN38. Expression level changes were confirmed by Western blot. Subsequent analysis of several high-grade glioma cell lines demonstrated very high levels of galectin-1, regardless if the cell lines contained mutant or wild-type TP53. High expression of galectin-1 in a human orthotopic murine tumor model was also detected by immunohistochemistry and revealed a consistent pattern of preferential expression in peripheral or leading tumor edges. Further examination of galectin-1 expression through microarray analysis in tumor materials from patients confirmed galectin-1 as a valuable biomarker and possible therapeutic target. These results demonstrate the utility of using proteomic approaches to interrogate and identify potential useful targets for cancer therapy by evaluating specific tumor responses, either positive or negative, to various therapies. Keywords: proteomics ‚ glioblastoma ‚ galectin-1 ‚ Western blot ‚ microarrays
Introduction Gliomas are the most common type of primary brain tumors. More than 12 000 new cases are diagnosed in the United States each year (Central Brain Tumor Registry in the United States, 2002). Of these, approximately 60-70% represents glioblastoma multiforme (GBM), which are essentially universally fatal tumors. There are many factors that complicate effective treatment for these tumors, such as their resistance to undergo * To whom correspondence should be addressed. Tel.: (850) 644-9861; fax: (850) 644-1366; e-mail:
[email protected]. † Go ¨ teborg University. ‡ Florida State University. § M.D. Anderson Cancer Center. 10.1021/pr060302l CCC: $37.00
2007 American Chemical Society
apoptosis as well as their invasive nature. In fact, a correlation seems to exist between cells that are actively invasive and mobile and their inability to trigger apoptosis after being exposed to either ionizing radiation or chemotherapy.1 It is also observed that blocking the invasive phenotype of malignant glioma cells in vitro increases the cells’ sensitivity to radiation and chemotherapy agents and, subsequently, to the triggering of apoptosis. Furthermore, it is well-established that transfecting glioma cell lines with wild-type tumor protein p53 (TP53 or p53) will trigger brisk apoptosis if a cell line harbors mutant p53, whereas the same transfection to cell lines which harbor wild-type p53 (wt p53) will contain a reduction or elimination of invasion and mobility.2,3 Ad-p53 refers to an adenoviral vector construct, which carries the gene for TP53 tumor suppressor protein; this gene is contained within the E1 region of the virus, rendering the virus incapable of self-replication. Regulation of the Fas/CD95 pathway is suggested to be partly responsible for Ad-p53 induction of apoptosis in glioma cells depending upon their p53 status.4 Lang et al.5 combined adenovirus transfection of wt p53 treatment of apoptosis resistant cells with chemotherapeutic DNA-damaging agents and discovered that p53 status and cytotoxic agents are important. Their work demonstrated that transfection of wild-type p53 into glioma cells which harbor wild-type p53, followed by treatment with the cytotoxic agent Irinotecan (CPT-11) or its metabolite SN-38 (7-ethyl-10hydroxycamptothecin), results in modest apoptosis and G2 arrest, whereas the reverse order produces almost complete G2 arrest followed by almost complete apoptosis of over 90% of cells. However, the exact mechanism of this activation is still unknown. To study differently expressed proteins in glioma cell lines with mutant p53 (cell lines which maintain the apoptotic phenotype) as compared with wild-type p53 (cells that have an apoptosis-resistant phenotype), we chose to use the common glioma cell line U87 MG (which carries the wild-type p53 genotype). By studying the wild type or apoptosis-resistant phenotype cell line through a proteomic approach, we hope to gain insights as to the potential mechanisms which render these cells resistant to apoptosis and to identify differentially expressed proteins. Through this approach, we hope to identify proteins that may be therapeutically useful targets, which could potentially trigger apoptosis within these particular genotype tumor cell lines. Journal of Proteome Research 2007, 6, 869-875
869
Published on Web 12/20/2006
technical notes
Proteomics of Human Glioblastoma
Table 1. Proteins Differentially Expressed in U87 MG Cells Treated with wt-p53 Prior to SN-38 Compared to Cells Treated with an Empty Vector and Then with SN-38 spot
acc nb. Swiss prot
protein identity
theor. Mw(Da)
theor. pI
no. pep.
seq. cov.%
2009 5303 5506 6302
P09382 P31944 P51570 P11021
galectin-1 caspase 14 galactokinase 1 GRP-78
14 706 27 947 42 702 72 333
5.34 5.44 6.04 5.07
7 4 9 12
70 11 21 20
Proteomic methods have been used as screening tools to discover new biomarkers in breast, lung, or ovary cancer,6,7 and there is an ongoing effort from several research laboratories to build cancer-specific proteomic databases; however, few reports show applications in the field of brain tumors.8 Recently, some proteomic studies on glioblastoma have been published, mainly focused on the discrimination of different subtypes of glioblastoma tumors.9-12 Understanding the molecular biology of gliomas is thus a clear and urgent necessity; the determination of the effectors of apoptotic resistance mechanisms in malignant gliomas could lead to future targeted therapies.
Experimental Section Cell Culture. The glioma cell line U87 MG (WT p53, ATCC #HTB-14), used in all cellular experiments, was grown in the presence of DMEM-F12 media supplemented with 10% FBS (Cell Gro, Mediatech, Herndon, VA) in a humidified CO2 incubator at 5% CO2. Cell cultures were grown in 150-mm dishes to 90% confluency. Additionally, to probe the expression of galectin-1 in other glioma cell lines (Figure 3, Supporting Information), several cell lines were grown in the same manner, namely, U343 (WT p53), LNB19 (mutant p53), LNB229 (mutant p53), U251 E4 (which has mutant p53 and full-length EGFR stably integrated), U87-viii (WT p53, gift from Dr. Bigner’s laboratory from Duke’s department of neuropathology), U251 HF (mutant p53), and D54 (WT p53, the latter two gifts from the laboratory of Peter Steck). Treatment of Cell Lines. Cell cultures U87 MG were treated with adenoviruses (therapeutic Ad-p53 or Dl-312 control adenovirus vector) or cytotoxic chemotherapy (SN-38) either alone, in combination, or in different sequences. Cell cultures were treated for 24 h with SN-38 at final concentration of 0.1 µM (stock solution of 10 mM). Cell cultures were also treated with either controlled virus Dl-312 at 1:100 MOI (multiplicity of infection) from a stock virus titered at 2.8 × 1011 pfu/mL (plaque-forming units/mL) or test virus which contained wildtype p53 gene inserted within the E1 region of the adenovirus vector (Ad-p53). This was similarly used at 1:100 MOI obtained from a stock of preparation of 2.2 × 1010 pfu/mL. Cell cultures that were treated with a combination of drug and virus included a total incubation time of 48 h allowing 24 h for each agent. Cells were washed three times with room-temperature phosphate-buffered saline (PBS) between treatments. Prior to viral infection, the cells were placed in serum-free media for 1 h to ensure adequate absorption of virus to the cells. Sample Preparation for 2D-PAGE. U87 MG glioblastoma cells (106 cells) were harvested before confluence, were washed four times with phosphate-buffered saline (PBS, 58 mM Na2HPO4 × 2 H2O, 17 mM Na2PO4 × H2O, 68 mM NaCl, pH 7.4), and were pelleted by centrifugation. The protein pellets were dissolved in 100 µL lysis buffer (9 M urea, 4% CHAPS (3-((3cholamidopropyl) dimethylammonio)-1-propanesulfonate), 1% dithiothreitol (DTT), and protease inhibitor (complete antiprotease solution, Roche Diagnostics, Mannheim, Germany) 870
Journal of Proteome Research • Vol. 6, No. 2, 2007
mass spec.
MALDI MS/MS MS/MS MS/MS
mascot score
levels
88 136 352 543
V v V V
and were homogenized with a plastic pestle. The samples were left 1 h at room temperature and were centrifugated at 2000g for 10 min to remove cell debris. The supernatant was transferred to new tubes. Protein concentration was measured on each sample with a protein kit (RD-DC, Bio-Rad, Hercules, CA). Two-Dimensional Gel Electrophoresis. All gels were run in duplicates. Two hundred fifty micrograms of protein was mixed with a rehydration buffer that contained 9 M urea, 35 mM tris, 42 mM DTT, 2% CHAPS, 0.66% sodium dodecyl sulfate (SDS), 2% IPG buffer, and bromophenol blue. The first dimension was performed with IPG strips, (pH 5-8, 17 cm [Bio-Rad]) with a Protean IEF Cell (Bio-Rad). The focusing was complete at 80 000 Vh, at 20 °C. After equilibration, the second dimension separation was performed on large 8-16% tris polyacrylamide gels (Bio-Rad) at 200 V for 5.5 h. Gel Staining and Imaging. The gels were stained with a fluorescent protein stain (SYPRO Ruby, Bio-Rad) according to the supplier’s protocol. Image acquisition and analysis were performed in a Fluor-S MultiImager (Bio-Rad). The optical density of spots was proportional to protein concentration. The protein spots were detected, quantified, and matched with the PD-Quest 2D-gel analysis software, v.7.1.1 (Bio-Rad). The protein spots from all gels were matched and their spot volumes were determined. The integrated optical densities of all spots within a gel that were matched to the reference standard image spots were summed, and the summed values were compared as a basis of normalization. Only statistically significant results (Student’s t-test < 0.02) were considered. The two sets of samples were compared by use of a Boolean tool. Only proteins detected in both sets of samples are presented in Table 1. Preparation of Samples for Mass Spectrometry. Digestion of proteins in-gel with trypsin has been previously described in detail.13 Briefly, dried gel pieces were rehydrated with porcine trypsin (Promega Corporation, Madison, WI), and the peptides were extracted with formic acid (FA) and acetonitrile (ACN). The protein digest was vacuum centrifuged and dissolved in 10 µL 0.1% FA (v/v). The samples were applied to the MALDI probe (AnchorChip, Bruker Daltonics, Bremen, Germany) as previously described.14 Mass Spectrometry and Database Searches. Matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry (MS) analysis was performed in a Reflex II MALDI-TOF mass spectrometer (Bruker-Franzen Analytik GmbH, Bremen, Germany). The spectra were acquired in reflectron mode at an accelerating voltage of 20 kV. Mass spectra were initially calibrated by external calibration using a mixture of known peptides (bradykinin fragment 1-7, [M + H]+ ) 747.40; angiotensin II, [M + H]+ ) 1046.54; somatostatin-14, [M + H]+ ) 1637.72; ACTH 18-39 (human), [M + H]+ ) 2465.20; and somatostatin-28, [M + H]+ ) 3147.47). A second calibration was provided by internal calibration with two autodigestion products of porcine trypsin (AA 100-107, [M + H]+ ) 842.5100 and AA 50-69, [M + H]+ ) 2211.1046). Monoisotopic m/z
technical notes values were submitted to the database search tool MASCOT (http://www.matrixscience.com) to search the NCBI nonredundant protein database. Carbamidomethylation of cysteine residues and oxidation of methionine residues were considered. A mass deviation of 100 ppm and one missed cleavage by trypsin were tolerated, Homo sapiens was specified and a significance threshold of p < 0.05 was chosen. Samples for which unambiguous protein identities could not be determined were analyzed in a hybrid linear ion trap FTICR MS (LTQ-FT, Thermo Electron, Bremen, Germany), equipped with a 7 T magnet. Nano-LC of the tryptic peptides was performed in a 20 cm × 50 µm i.d. fused silica column packed in-house with ReproSil-Pur C18-AQ porous (120 Å) C18bonded particles (Dr. Maisch GmbH, Ammerbuch, Germany). Sample injections (2 µL) were made with an HTC-PAL autosampler (CTC Analytics AG, Zwingen, Switzerland) connected to an Agilent 1100 binary pump (Agilent Technologies, Palo Alto, CA). The peptides were trapped on a precolumn (4.5 cm × 100 or 130 µm i.d.) packed with 3 µm C18-bonded particles (Hydrosphere,120 Å, YMC Co. Ltd., Kyoto, Japan) in a valveswitching configuration. The voltage applied was +1.4 kV and the eluent was electrosprayed from the emitter tip. The gradient was 0-50% ACN, starting with 0.2% FA in water (100 nL/min) for 40 min. The linear ion trap was operated in data-dependent mode, to automatically switch between MS and MS/MS acquisition. Survey MS spectra (from m/z 400 to 1600) were acquired in the FT-ICR, and the four most abundant doubly or triply protonated ions in each FT-scan were selected for MS/ MS in the linear ion trap followed by detection in the ion trap. Peptide mass values were matched to protein sequences by Mascot (Matrix Science, London). Carbamidomethylation of cysteine residues and oxidation of methionine residues were considered and a precursor mass deviation of 5 ppm was tolerated. One-Dimensional Western Blot. Cell cultures were harvested and lysed in SDS sample buffer. The protein levels were determined by using Bio-Rad protein assay reagent (Catalog #500-00016). Samples (10 µg) were loaded onto a 12% SDS gel (10 × 10 cm) and were run at 120 V for 90 min. The gel was electroblotted on a nitrocellulose membrane (Hybond ECL, Amersham Biosciences). After the gel was transferred to the membrane, it was blocked with 10% nonfat milk Tris-buffered saline (TBS, pH 7.6). The membrane was incubated overnight at 4 °C with a primary antibody for galectin-1 (H45 Rabbit polyclonal, Santa Cruz Biotechnology Inc., Santa Cruz, CA) at a dilution of 1:2000. The membrane was then incubated at ambient temperature for 1 h with HRP-conjugated secondary antibody (goat antirabbit polyclonal + IgG-HRP, Santa Cruz Biotechnology, Inc.) at a dilution of 1:5000. Proteins were detected by incubating the membrane with Western lighting chemiluscence for 1 h at ambient temperature. The films were exposed for 1-10 seconds (Hyblot CL, Denville Scientific, Inc.). The specificity of the galectin-1 antibody was verified by peptide mapping of protein bands (data not shown). Two-Dimensional Gel Western Blot. U87 MG cells were grown to 90% confluency and were either harvested alone or infected with Ad-p53 1:100 MOI for 24 h. The cells were harvested using a cell lysis buffer containing 9 M urea, 4% CHAPS, 3.5 mM Tris-HCl at pH 7.4, and 66 mM dithiothreitol. This cell lysate was then purified using a cleanup kit (Bio-Rad) according to the manufacturer’s instructions. Samples containing 150 µg of protein were analyzed. The sample was loaded on IPG strips (7 cm, pH 5-8, Bio-Rad). The protein sample
Puchades et al.
was rehydrated in sample buffer (Ready Prep 2-D Starter Kit, Bio-Rad). Isoelectric focusing was performed in a Protean IEF cell unit (Bio-Rad) according to the manufacturer’s recommendation. The second dimension was performed in a 4-12% Bis-Tris gel (NUPAGE, Invitrogen Life Technologies, Carlsbad, CA) in MOPS [3-(N-morpholino) propanesulfonic acid, 4.19 grams/L] and 1 mM ethylene diamine tetraacetic acid (EDTA). The separation proceeded at 120 V for 80 min. The gel was electroblotted in an identical fashion to the one-dimensional Western blot previously described, with the exception that the membrane used for the two-dimensional electroblotting was from Amersham. Treatment with siRNA. Cell cultures of U87 MG were grown in six-well dishes and were planted initially with 1 × 106 cells. The cells were then transfected with galectin-1 siRNA (Santa Cruz Biotchnology, Santa Cruz, CA). A total of 5 µL of siRNA per well was used together with 5 µL of Lipofectamne 2000 per well (Invitrogen Life Technologies, Carlsbad, CA) and was incubated at 37 °C for 24 h. The cells were then washed with room-temperature PBS and were trypsinized with CMF buffer (Hank’s cell dissociation buffer, Gibco BRL, Invitrogen). Microarray Analysis of Patient Samples for Galectin-1. All of the tumor samples were obtained through the University of California San Francisco Medical School, Department of Pathology in accordance with university policy. Patient consent to participation was obtained prior to surgery. All patient identification information remained confidential and no patient identification or demographic data was used. Specimens which had adequate amount of tumor which was not contaminated with large amounts of normal brain were included. Care was made to not isolate material that was near regions of necrosis. Preparation of RNA, sample processing, and initial analysis of this data set have been previously described.15 The microarray chip was U95Av2 (Affymetrix, Santa Clara, CA). The data were mined to assess galectin-1 expression compared to patient outcome. Cell Viability Assays. U87 MG cell lines were grown to 95% monolayer confluence. The cells were trypsinized and harvested with 0.25% trypsin/EDTA, were plated in six-well tissue culture plates, and were allowed to adhere overnight at 37 °C in 5% CO2 humidified incubators. Cells were incubated at 37 °C for 3 days and cell viability by cellular respiration was determined using an MTT assay [3-(4, 5-methylthiazole-2-yl)-5-(3-carboxymethoxy-phenyl)-2-(4-sulfonyl)-2H-tetrazolium)] (Promega, Madison, WI) according to the manufacturer’s protocol. The cell survival fraction was measured at each drug concentration as the ratio of absorbance at 490 nm. This calculation was normalized for background absorbance of the culture medium alone. Animal Studies. Cell implantation and adenoviral treatment were performed as described previously.16 Briefly, implants were placed in four each 6-8-week-old female NuNu mice (Harlan, Indianapolis, IN) by using screw-guide hardware with coordinates of 1 mm anterior and 1.2 mm lateral to the bregma. The mice were allowed to heal for 7 days, after which U87 MG glioma cells (5 × 105 cells in 10 µL of PBS) were injected at a depth of 4 mm in the region of the putamen. After 14 days, or if the mice displayed signs of neurological dysfunction, the animals were sacrificed by CO2 inhalation, and the brain with tumor was harvested for histopathologic examination and immunohistochemical staining. To induce growth of grafts subcutaneously, animals were injected in the flank with U87 MG glioma cells and the tumors were allowed to reach Journal of Proteome Research • Vol. 6, No. 2, 2007 871
technical notes
Proteomics of Human Glioblastoma
Figure 1. Photograph of a typical 2D gel, pH 5-8, of U87 MG cells. Two treatments were compared to detect changes due to wt-p53 transfection prior to SN-38 in those cells. Panel A is showing up-regulated (squares) and down-regulated proteins (circles) when cells are treated with wt-p53 prior to SN-38 compared to cells treated with an empty vector and then SN-38. Panel B represents a zoomed window of galectin-1 shown on gels treated with an empty vector and then SN-38.
approximately 300-400 mm3 before sacrifice and tumor harvest. All procedures were performed by trained veterinary medicine staff and were approved by the Institutional Animal Care and Use Committee. Animals were housed and maintained in laminar flow cabinets under specific pathogen-free conditions in facilities approved by the American Association for Accreditation of Laboratory Animal Care (AAALAC certification #000183, most recently reviewed March, 1998) and in accordance with current regulations and standards of the U. S. Department of Agriculture, U. S. Department of Health and Human Services, and National Institutes of Health. The laboratories at The University of Texas M.D. Anderson Cancer Center, where these animal studies were conducted, fully endorse the concepts within the “Guide for the Care and Use of Laboratory Animals” as promulgated by the National Institutes of Health. The institution has an approved animal assurance (#A-3343-01) on file with the OPRR. Immunohistochemical Analysis of Xenograft Tumor Sections. Animal brains were fixed in formalin and were embedded in paraffin, and sections were prepared after initial baking at 60 °C for 30 min. The sections were blocked with 0.3% H2O2 and 100% methanol for 30 min and were rinsed in 10 mM PBS 872
Journal of Proteome Research • Vol. 6, No. 2, 2007
with 0.2% Triton X-100. The sections were treated for 20 min with Triton-X 100 and PBS (1:50 ratio) and then were incubated with anti-galectin-1, diluted 1:2000 (Santa Cruz Biotechnology). Sections were incubated with secondary antibodies at a 1:50 dilution at ambient temperature for 1 h. Staining was performed with 3,3′-amino diamino-benzidine tablets (Sigma). The sections were counterstained with 0.01% methanol green. Statistical Analysis. Statistical analysis was performed by Prism’s Graph Pad software. Comparative data were calculated as means ( standard deviations (SD). Statistical analysis was performed using Student’s t-test (two-tailed), and statistical significance was defined as P < 0.05. Wilcoxon Summed Rank tests were used to analyze patient samples as related to relative galectin-1 expression from the Affymetrix U95Av2 data sets.
Results and Discussion By use of proteomic analysis, we identified proteins that were involved in a phenotypic change in high-grade glioma cell line U87 MG under the influence of transfection with wild-type p53 and additional treatment of cytotoxic chemotherapy with SN38 (Figure 1, Table 1). Of all the proteins identified, one protein,
technical notes
Puchades et al.
Figure 2. Immunohistochemistry staining of both xenograft tumors implanted intercranially (panel A) and subcutaneously (panel B). The panels demonstrate staining of galectin-1 primarily in the cytoplasmic regions but also to a lesser degree in the nuclei. In both cases, the staining appears to be more prominent in the leading edge or peripheral aspect of the expanding tumor mass. The images are shown at both 10 and 20× magnification. Counterstaining with hematoxylin and eosin is provided (V ) ventricle, In ) invasive leading edge, T ) tumor, SC ) subcutaneous tissue, and NL ) normal brain).
galectin-1, was found to be dramatically modulated when the cells were treated with wt p53 and SN-38. Spot 2009 (Figure 1) corresponded to galectin-1. In particular, spot 2009 (Figure 1) increased when cells were treated with SN-38 alone. The combined treatment of wt-p53 transfection prior to SN-38 resulted in a notable down-regulation of galectin-1. Galectin-1 is a member of a lectin family defined by two properties: a carbohydrate-recognizing domain and a β-galactoside affinity.17 Galectin-1 is an ubiquitous protein with many receptors and is involved in biological functions such as cell adhesion, cell proliferation, tumor metastasis, apoptosis, and immunoregulatory effects.18,19 In cancer, galectin-1 expression has been detected in several types of tumors, such as endometrial, prostate, head, and neck squamous cell carcinomas.20 In brain tumors, galectin-1 was shown to be expressed
in all human glioma types.21,22 Expression of galectin-1 was shown to be associated with malignancy and poor prognosis.23 Two other proteins were found to be modulated by wt-p53 transfection prior to SN-38 treatment, galactokinase 1 and GRP78. Galactokinase is a major enzyme for galactose metabolism and catalyzes the phosphorylation of D-galactose. Galactokinase has been linked to ovarian cancer24 but not to glioblastoma. GRP-78 was found as a fragment of approximately 29 kDa as revealed by MS/MS analysis. GRP-78 up-regulation has been implicated in hepatocellular carcinoma and lung cancer.25,26 Caspase 14 is a developmentally regulated protease27 that has been shown to be altered in epithelial malignancies. To confirm that p53 was able to modulate the expression of galectin-1, both one- and two-dimensional Western blot analyses were performed (Figures 1 and 2, Supporting Information). Journal of Proteome Research • Vol. 6, No. 2, 2007 873
technical notes
Proteomics of Human Glioblastoma
Figure 3. In vitro staining of U87 MG cells after treatment with ad-p53. Control panels A and B demonstrate monolayers of cells stained without anti-Galectin-1 antibody (Panel A) and with anti-Galectin-1 antibody (Panel B). (Note: the prominent staining of Galectin-1 within a cytoplasm of Panel B). Panel C represents monolayer U87 MG cells treated with wild-type p53 (with a multiplicity of infection at 100× using Ad-p53 for 24 h). As can be seen, a reduction of the overall staining for Galectin-1 is evident after these cells were subsequently stained with anti-Galectin-1 antibodies.
Galectin-1 expression was modulated by p53 alone but not to as high a degree as with the combined treatment. Treatment of the U87 cell line with p53 demonstrated a reduction of the expression of galectin-1, and the addition of both p53 and SN38 appears to decrease the amount of galectin-1 even more dramatically (Figure 1, Supporting Information). It is also evident that SN-38 by itself is not able to reduce the level of galectin-1. In addition, the two-dimensional Western blot demonstrates almost full abrogation of the enlarged spot corresponding to the appropriate region for galectin-1, following treatment with wild-type p53 transfection (Figure 2, Supporting Information). Because of the dramatic modulation of galectin-1 by p53 and cytotoxic treatment with SN-38, we investigated the presence of galectin-1 in other high-grade glioma cell lines, regardless of their p53 status (mutant or wild-type). All of the high-grade glioma cell lines that we investigated displayed high levels of galectin-1 (Figure 3, Supporting Information), which indicates that the protein is up-regulated during the course of glioma development. We investigated whether tumors growing intracranially (Figure 2A) or subcutaneously (Figure 2B) in xenograft mouse models would also demonstrate the expression of galectin-1. Both subcutaneous and intracranical tumors do indeed express high amounts of galectin-1, and the expression of the protein appears to be more prominent at the leading edge of the expanding tumor. In particular, the leading and invasive edges of subcutaneous tumors appear to express high levels of galectin-1. Intracellular galectin-1 staining of cells exposed to 100 MOIs of Ad-p53 demonstrated a reduction of cytoplasmic expression of galectin-1 when treated with wild-type p53 for 24 h (Figure 3A-C). Also noted within these cell cultures was a decrease in the overall mobility of these cells (data not shown). The viability of cell cultures was evaluated after reducing the level of galectin-1 by the administration of siRNA to galectin-1 transcripts. The data (Figure 4) demonstrated a clear increase in sensitivity to the cyototoxic chemotherapy SN38 after siRNA treatment. It is currently unclear how the reduction of galectin-1 can lead to increased sensitivity to topoisomerase inhibitors like SN-38. It is also interesting that simultaneous treatment of siRNA and recombinant galectin-1 protein partially abrogates the increase in sensitivity. 874
Journal of Proteome Research • Vol. 6, No. 2, 2007
Figure 4. Demonstrates the reduction in cell viability to knockdown of galectin-1 by siRNA to galectin-1 by MTT assay. The reduction of cell viability to progressive doses of SN-38 (0-50 nM) is shown with an additive effect with the reduction of galectin-1 and cytotoxic treatment.
Finally, the discovery of galectin-1 up-regulation was investigated regarding its possible clinical relevance (Figure 4, Supporting Information). Microarray analysis of a panel of samples from patients with high-grade gliomas demonstrated that patients who had poor survival (defined as surviving 2 years, showed a positive correlation between increased galectin-1 expression and poor patient survival. The data is in agreement with the decreased cell viability data demonstrated in Figure 4. Taken together, the data may indicate that galectin-1 could be used as a therapeutic target to increase cell sensitivity to cytotoxic agents.
Conclusion We investigated the use of proteomic analysis to identify key proteins or protein expression signatures in the high-grade glioma cell line U87 MG, which harbors wild-type p53. These cell lines undergo significant apoptosis when treated with exogenous wild-type p53 and cytotoxic chemotherapy.5 Our investigation identified several proteins, which were either upor down-regulated. Of particular interest to us was the large
technical notes change in the expression level of galectin-1, which led to further analysis of this particular protein within glioma cells lines, in particular, the representative glioma cell line, U87 MG. Subsequent analysis confirmed that the expression of wild-type p53 significantly reduces the expression level of galectin-1. This result was confirmed by Western Blot analysis with the addition of SN-38, either before wild-type p53 treatment or with SN-38 treatment after wild-type p53 treatment. The change in galectin-1 expression pattern reflected the number of cells undergoing apoptosis and implicated this protein as one of the mediators of apoptosis in this particular cell line. In addition, we were able to demonstrate that several highgrade cell lines routinely used in laboratories expressed a very high amount of galectin-1. We also demonstrated that galectin-1 appears to be expressed when the cells lines are grown in a xenograft model; specifically, the protein appears to be involved in the leading edge or periphery of the expanding tumor regardless of location. We also have shown in in vitro studies that the down-regulation of galectin-1 results in greater sensitivity to cytotoxic chemotherapies, which could have direct implications to clinical therapies that target galectin-1. Finally, the expression of galectin-1 in a set of clinically relevant samples was evaluated through microarray analysis. This analysis revealed a statistically significant relationship between high expression of galectin-1 and poor survivorship in patients with high-grade glioblastoma compared to patients who survived longer (defined as > 2 years) and who had lower overall expression levels of galectin-1 mRNA. Taken together, these results suggest that galectin-1 is a relevant target to downregulate in a clinical pharmacological setting to improve overall survival for patients suffering high-grade gliomas. Abbreviations: AA, amino acid; Acc nb, accession number; ACN, acetonitrile; Adp53, adenoviral construct that carries the gene TP53; CPT-11, Irinotecan, a cytotoxic agent; 2D, twodimensional; D1-312, an “empty” or control adenovirus vector; FBS, fetal bovine serum; FT-ICR, Fourier transform ion cyclotron resonance; GBM, glioblastoma multiforme; GRP-78, glucoseregulated protein 78; MALDI-TOF, matrix-assisted laser desorption/ionization time-of-flight; MOI, multiplicity of infection; MS, mass spectrometry; p53, tumor suppressor protein; no., number; PBS, phosphate-buffered saline; pep, peptides; SDS, sodium dodecyl sulfate; seq, sequence; SN-38, metabolite of Irinotecan, 7-ethyl-10-hydroxycamptothecin; spec., spectrometry; theor, theoretical; TP53, gene for tumor suppressor protein.
Acknowledgment. The authors thank Carina Sihlbom at Go¨teborg University for providing FT-ICR analysis. Financial support from the ICR facility at the National High Magnetic Field Laboratory (NSF DMR-00-84173), the Swedish Medical Research Council, Lundgrens Vetenskapsfond, Adlerbertska research foundation, Assar Gabrielssons foundation Go¨teborg, Sweden, the Swedish Society of Medicine, and the Swedish Foundation for International Cooperation in Research and Higher Education (STINT) is gratefully acknowledged. The 7T FT-ICR MS at Go¨teborg University was purchased through a grant from Knut and Alice Wallenberg Foundation (Carol L. Nilsson). Also, The Sorensen Family Fund from the Celebrate Life Charitable Organization is gratefully acknowledged for their generous support to the Brain Tumor Center at MDACC. Supporting Information Available: 1D Western blot analyses of galectin-1 expression in U87 MG cells under different treatments and in nine high-grade glioma cell lines,
Puchades et al.
2D Western blot analysis of U87 MG cells after treatment with ad-p53, and figure showing the possible clinical relevance of galectin-1 upregulation. This material is available free of charge via the Internet at http://pubs.acs.org.
References (1) Mariani, L.; Beaudry, C.; McDonough, W. S.; Hoelzinger, D. B.; Demuth, T.; Ross, K. R.; Coons, S. W.; Watts, G.; Trent, J. M.; Wei, J. S.; Giese, A.; Berens, M. E. J. Neurooncol. 2001, 53, 161-176. (2) Gomez-Manzano, C.; Fueyo, J.; Kyritsis, A. P.; Steck, P. A.; Roth, J. A.; McDonnell, T. J.; Steck, K. D.; Levin, V. A.; Yung, W. K. Cancer Res. 1996, 56, 694-699. (3) Nielsen, L. L.; Maneval, D. C. Cancer Gene Ther. 1998, 5, 52-63. (4) Cerrato, J. A.; Yung, W. K.; Liu, T. J. Neurooncology 2001, 3, 113122. (5) Lang, F. F.; Shono, T.; Gilbert, M. R. Neurooncology 2002, 4, 323324. (6) Alaiya, A. A.; Roblick, U. J.; Franzen, B.; Bruch, H. P.; Auer, G. J. Chromatogr., B 2003, 787, 207-222. (7) Celis, J. E.; Gromov, P.; Gromova, I.; Moreira, J. M.; Cabezon, T.; Ambartsumanian, N.; Grigorian, M.; Lukanidin, E.; Thor Straten, P.; Guldberg, P.; Bartkova, J.; Bartek, J.; Lukas, J.; Lukas, C.; Lykkesfeldt, A.; Jaattela, M.; Roepstorff, P.; Bolund, L.; Orntoft, T.; Brunner, N.; Overgaard, J.; Sandelin, K.; Blichert-Toft, M.; Mouridsen, H.; Rank, F. E. Mol. Cell Proteomics 2003, 2, 369377. (8) Zheng, P. P.; Kros, J. M.; Sillevis-Smitt, P. A. E.; Luider, T. M. Front Biosci. 2003, 8, d451-463. (9) Furuta, M.; Weil, R. J.; Vortmeyer, A. O.; Huang, S.; Lei, J.; Huang, T. N.; Lee, Y. S.; Bhowmick, D. A.; Lubensky, I. A.; Oldfield, E. H.; Zhuang, Z. Oncogene 2004, 23, 6806-6814. (10) Odreman, F.; Vindigni, M.; Gonzales, M. L.; Niccolini, B.; Candiano, G.; Zanotti, B.; Skrap, M.; Pizzolitto, S.; Stanta, G.; Vindigni, A. J. Proteome Res. 2005, 4, 698-708. (11) Vogel, T. W.; Zhuang, Z.; Li, J.; Okamoto, H.; Furuta, M.; Lee, Y. S.; Zeng, W.; Oldfield, E. H.; Vortmeyer, A. O.; Wei, J. S. Clin. Cancer Res. 2005, 11, 3624-3632. (12) Zhang, R.; Tremblay, T. L.; McDermid, A.; Thibault, P.; Stanimirovic, D. Glia 2003, 42, 194-208. (13) Davidsson, P.; Westman-Brinkmalm, A.; Nilsson, C. L.; Lindbjer, M.; Paulson, L.; Andreasen, N.; Sjogren, M.; Blennow, K. Neuroreport 2002, 13, 611-5. (14) Gobom, J.; Schuerenberg, M.; Mueller, M.; Theiss, D.; Lehrach, H.; Nordhoff, E. Anal. Chem. 2001, 73, 434-438. (15) Nigro, J.; Misra, A.; Zhang, L.; Smirnov, I.; Colman, H.; Griffin, C.; Ozburn, N.; Chen, M.; Pan, E.; Koul, D.; Yung, W. K.; Feuerstein, B. G.; Aldape, K. D. Cancer Res. 2005, 65, 1678-1686. (16) Lal, S.; Lacroix, M.; Tofilon, P.; Fuller, G. N.; Sawaya, R.; Lang, F. F. J. Neurosurg. 2000, 92, 326-333. (17) Barondes, S. H.; Castronovo, V.; Cooper, D. N.; Cummings, R. D.; Drickamer, K.; Feizi, T.; Gitt, M. A.; Hirabayashi, J.; Hughes, C.; Kasai, K. Cell 1994, 76, 597-598. (18) Hernandez, J. D.; Baum, L. G. Glycobiology 2002, 12, 127R-136R. (19) Elola, M. T.; Chiesa, M. E.; Alberti, A. F.; Mordoh, J.; Fink, N. E. J. Biomed. Sci. 2005, 12, 13-29. (20) Danguy, A.; Camby, I.; Kiss, R. Biochim. Biophys. Acta 2002, 1572, 285-293. (21) Camby, I.; Belot, N.; Lefranc, F.; Sadeghi, N.; de Launoit, Y.; Kaltner, H.; Musette, S.; Darro, F.; Danguy, A.; Salmon, I.; Gabius, H. J.; Kiss, R. J. Neuropathol. Exp. Neurol. 2002, 61, 585-596. (22) Rorive, S.; Belot, N.; Decaestecker, C.; Lefranc, F.; Gordower, L.; Micik, S.; Maurage, C. A.; Kaltner, H.; Ruchoux, M. M.; Danguy, A.; Gabius, H. J.; Salmon, I.; Kiss, R.; Camby, I. Glia 2001, 33, 241-255. (23) Yamaoka, K.; Mishima, K.; Nagashima, Y.; Asai, A.; Sanai, Y.; Kirino, T. J. Neurosci. Res. 2000, 59, 722-730. (24) Cramer, D. W.; Greenberg, E. R.; Titus-Ernstoff, L.; Liberman, R. F.; Welch, W. R.; Li, E.; Ng, W. G. Cancer Epidemiol. Biomarkers Prev. 2000, 9, 95-101. (25) Uramoto, H.; Sugio, K.; Oyama, T.; Nakata, S.; Ono, K.; Yoshimastu, T.; Morita, M.; Yasumoto, K. Lung Cancer 2005, 49, 5562. (26) Luk, J. M.; Lam, C. T.; Siu, A. F.; Lam, B. Y.; Ng, I. O.; Hu, M. Y.; Che, C. M.; Fan, S. T. Proteomics 2006, 6, 1049-1057. (27) Krajewska, M.; Kim, H.; Shin, E.; Kennedy, S.; Duffy, M. J.; Wong, Y. F.; Marr, D.; Mikolajczyk, J.; Shabaik, A.; Meinhold-Heerlein, I.; Huang, X.; Banares, S.; Hedayat, H.; Reed, J. C.; Krajewski, S. Clin. Cancer Res. 2005, 11, 5462-5471.
PR060302L Journal of Proteome Research • Vol. 6, No. 2, 2007 875
