Comparative Proteomic Profiles of Meningioma Subtypes

Research Article

Hiroaki Okamoto,1,3 Jie Li,1 Alexander O. Vortmeyer,1 Howard Jaffe,2 Youn-Soo Lee,4 Sven Gla¨sker,1 Tae-Sung Sohn,1 Weifen Zeng,5 Barbara Ikejiri,1 Martin A. Proescholdt,6 Christina Mayer,6 Robert J. Weil,1,5 Edward H. Oldfield,1 and Zhengping Zhuang1

1Surgical Neurology Branch and 2Protein/Peptide Sequencing Facility, National Institute of Neurological Disorders and Stroke, NIH, Bethesda, Maryland; 3Department of Neurosurgery, Faculty of Medicine, Saga University, Saga, Japan; 4Department of Clinical Pathology, College of Medicine, The Catholic University of Korea, Seoul, Korea; 5Brain Tumor Institute, Cleveland Clinic Foundation, Cleveland, Ohio; and 6Department of Neurosurgery, Regensburg University Medical Center, Regensburg, Germany

Abstract WHO grade 1 tumors (4.7-7.2% of meningiomas; refs. 1–3). Meningiomas are classified into three groups (benign, However, sampling issues and imperfect histopathologic features atypical, and anaplastic) based on morphologic character- of these tumors can make exact grading difficult. Some markers, istics. Atypical meningiomas, which are WHO grade 2 tumors, such as vimentin, are frequently present in meningiomas (1, 4) but and anaplastic meningiomas, which are WHO grade 3tumors, do not correlate specifically with grade. exhibit an increased risk of recurrence and premature death The origins of atypical and anaplastic meningiomas are still compared with benign WHO grade 1 tumors. Although unclear. It is well known that some high-grade tumors arise atypical and anaplastic meningiomas account for <10% of from low-grade tumors. For instance, Vogelstein et al. showed all of meningiomas, it can be difficult to distinguish them accumulating molecular alterations that paralleled clinical pro- from benign meningiomas by morphologic criteria alone. We gression in colorectal tumors (5). It has also been reported that glioblastomas consist of two different types of tumors with used selective tissue microdissection to examine 24 human de novo meningiomas and did two-dimensional gel electrophoresis to different pathogenesis ( and secondary tumors) in which determine protein expression patterns. Proteins expressed low-grade gliomas develop through a multistep process (6, 7). differentially by meningiomas of each WHO grade were Generally, meningiomas are also thought to have the potential to identified and sequenced. Proteomic analysis revealed protein progress from low grade to high grade, as colorectal carcinomas expression patterns unique to WHO grade 1, 2, and 3 or secondary glioblastomas (1, 8, 9). Weber et al. documented a meningiomas and identified 24 proteins that distinguish each genetic model of progression in meningiomas by accumulated subtype. Fifteen proteins showed significant changes in alterations (8). However, it is still unclear whether high-grade meningiomas progress from benign meningiomas, or arise expression level between benign and atypical meningiomas, de novo whereas nine distinguished atypical from anaplastic meningi- primarily as more aggressive tumors (9). Thus, specific omas. Differential protein expression was confirmed by markers that would reliably coincide with meningiomas of a Western blotting and immunohistochemistry. We established particular grade and clinical behavior, and with pathogenetic differential proteomic profiles that characterize and distin- mechanisms, would be useful. Proteomics is a broad protein profile screening approach that guish meningiomas of increasing grades. The proteins and proteomic profiles enhance understanding of the pathogene- permits a direct analysis of proteins differentially or uniquely sis of meningiomas and have implications for diagnosis, expressed by a cell or a tissue type. Proteomic research has been developed recently in the post-genome era. Using a combination of prognosis, and treatment. (Cancer Res 2006; 66(20): 10199-204) two-dimensional gel electrophoresis, mass spectrometry, and bioinformatics, one can now readily scan a particular cell type Introduction and establish protein expression patterns. Recent studies have Meningiomas are common central nervous system (CNS) tumors shown that specific proteomic patterns and proteins can distin- that arise from the coverings of the brain and the spinal cord, with guish subtypes or grades of human brain tumors (10, 11). We set out an incidence of between 13% and 26% of all primary intracranial to use selective tissue microdissection to obtain a pure population tumors (1). Most meningiomas (>90%) are slowly growing benign of meningioma tumor cells to determine whether we could identify tumors and correspond histologically to grade 1 according to the proteins and protein patterns that distinguish WHO grade 1, 2, WHO classification of tumors of the CNS (1). Anaplastic and 3 meningiomas. meningiomas are WHO grade 3 neoplasms and are associated with a high risk for recurrence and a less favorable prognosis; Materials and Methods fortunately, they are rare (1-2.8% of all meningiomas; refs. 1–3). An intermediate group, the atypical meningiomas (WHO grade 2), also Patients and clinical data. Twenty-four meningiomas were collected at exhibit an increased risk of recurrence and death compared with surgery, immediately snap-frozen in liquid nitrogen, and stored at À80jC until analyzed. Tumors were classified according to the 2000 WHO Histologic Classification (1): 10 benign meningiomas (grade 1), 9 atypical meningiomas (grade 2), and 5 anaplastic meningiomas (grade 3). Tissues Note: H. Okamoto and J. Li contributed equally to this study. and clinical information were obtained as part of an Institutional Review Requests for reprints: Zhengping Zhuang, Surgical Neurology Branch, National Board–approved study at the Cleveland Clinic Foundation, Cleveland, OH Institute of Neurological Disorders and Stroke, NIH, Building 10, Room 5D 37, 10 and the University of Regensburg, Regensburg, Germany. Additional Center Drive, Bethesda, MD 20892-1414. Phone: 301-534-8445; Fax: 301-402-0380; independent 30 paraffin slides were obtained from the Cleveland Clinic E-mail: [email protected]. n I2006 American Association for Cancer Research. Foundation for immunohistochemical studies, including benign ( = 10), doi:10.1158/0008-5472.CAN-06-0955 atypical (n = 10), and anaplastic (n = 10) tumors. www.aacrjournals.org 10199 Cancer Res 2006; 66: (20). October 15, 2006

Tissue and sample preparation. Selective tissue microdissection was Results done, as previously described, to avoid normal brain or areas of necrosis, hemorrhage, and inflammation (10). Microdissected tissues were dissolved Two-dimensional gel electrophoresis. Two-dimensional gel in extraction buffer II containing 8 mol/L urea, 4% (w/v) Bio-Lyte 4/7, and electrophoresis of the 10 benign, nine atypical, and five anaplastic 2 mmol/L tributyl phosphine (Bio-Rad, Hercules, CA); vigorously vortexed; meningiomas was done over a pH range of 4 to 7 (Fig. 1). We and centrifuged in a microcentrifuge. The supernatant was combined with compared proteomic patterns between the benign and atypical a rehydration buffer containing rehydration buffer [8 mol/L urea, 2% meningiomas and between the atypical and anaplastic meningi- CHAPS, 50 mmol/L DTT, and 0.2% (w/v) Bio-Lyte 4/7 ampholytes; Bio- omas. Consistently expressed protein patterns and individual Rad], IPG buffer (Amersham Biosciences, Piscataway, NJ), and bromophe- proteins on two-dimensional gels within each group were selected nol blue and subsequently rehydrated overnight with Immobiline Drystrips with statistically significant values by t-test (P < 0.05) using two- (pH 4/7, 11 cm; Amersham Biosciences) on a Reswelling Tray (Amersham dimensional gel analysis software (Proteomweaver, Definiens, Biosciences). Munich). Twenty-four proteins that were differentially expressed Two-dimensional gel electrophoresis. Isometric focusing for the first dimensional electrophoresis was done with a Multiphore II Electrophoresis in all members of each tumor grade were isolated and identified System (Amersham Biosciences). The strips were subjected to high voltages by LC-MS/MS. at 300 to 3,500 V. Immobilized pH gradient (IPG) strips were equilibrated Comparison between benign and atypical meningiomas. with equilibration buffer I containing 6 mol/L urea, 2% SDS, 375 mmol/L First, we compared proteomic patterns and identified proteins Tris-HCL (pH 8.8), 20% glycerol, and 2% (w/v) DTT and buffer II containing that differentiated the 10 benign and 9 atypical meningiomas 6 mol/L urea, 2% SDS, 375 mmol/L Tris-HCL (pH 8.8), 20% glycerol, and (Table 1A and B). Fifteen distinguishing proteins that were highly 2.5% (w/v) iodoacetamide (Bio-Rad). Precast gels (ExcelGel SDS gradient gel; expressed in either the benign or atypical group were identified. Â Â pH 4-7 12-14%, 245 180 0.5 mm; Amersham Biosciences) were used for Six of the 15 distinguishing proteins were differentially overex- the second dimension of protein separation by a Multiphor II Flated System pressed in benign meningiomas (Table 1A). Meanwhile, 9 of the 15 (Amersham Biosciences) under a constant voltage of 700 V. A silver staining identified proteins were differentially up-regulated in atypical menin- kit (Amersham Biosciences) was used to detect protein spots according to the manufacturer’s instructions. All samples were run in duplicate. giomas (Table 1B). These proteins reliably and reproducibly dis- Image analysis and in-gel digestion. The intensities of protein spots on tinguished benign from atypical meningiomas. two-dimensional gels were analyzed with Proteomweaver according to the Comparison between atypical and anaplastic meningiomas. manufacturer’s protocol (Definiens, Munich, Germany). Statistically signi- Next, we isolated and identified nine proteins that differentiated ficant (P < 0.05) protein spots of interest were excised from the gel and the nine atypical and five anaplastic meningiomas (Table 2A and B). subjected to in-gel digestion with trypsin. Mass spectrometry. Peptides from in-gel digest were analyzed by liquid chromatography tandem mass spectrometry (LC-MS/MS) on a ProteomeX LC/MS system (ThermoElectron, San Jose, CA) operated in the high- throughput mode. Reversed-phase high-performance liquid chromatography was carried out using a BioBasic-18 column (0.18Â150 mm; Thermo- Electron) eluted at 1 to 2 AL/min with a gradient of 2% to 50% B over 30 minutes. Mobile phase A was H2O (0.1% FA), and mobile phase B was CH3CN (0.1% FA). Column effluent was analyzed on the LCQ Deca XP Plus (ThermoElectron) operating in the ‘‘Top Five’’ mode. Protein identification. Uninterpreted MS/MS spectra were searched against a human database using the BioWorks and SEQUEST programs (ThermoElectron). Protein identification was accepted when the MS/MS spectra of at least two peptides from the same protein exhibited at a minimum the default Xcorr versus charge values set by the program (for Z = 1,1.50; for Z = 2, 2.00; for Z = 3, 2.50). Western blot analysis. Meningioma samples were homogenized in T-PER Tissue Protein Extraction Reagent (Pierce, Rockford, IL) with protease inhibitor cocktail (Roche Molecular Biochemicals, Indianapolis, IN) and centrifuged; 40 Ag of each lysate were loaded to 4%-20% SDS- polyacrylamide gels (Invitrogen, Carlsbad, CA). Proteins were electropho- retically transferred to nitrocellulose membranes (Invitrogen), blocked with StartingBlock Blocking Buffer (Pierce), washed, and incubated with polyclonal antibodies against galectin-1 (1:500; R&D Systems, Minneapolis, MN), a-enolase, and RP/EB family, member 1 (EB1; 1:200; Santa Cruz Biotechnology, Inc., Santa Cruz, CA), and monoclonal antibodies against calponin, smooth muscle actin (1:200; DakoCytomation, Glostrup, Den- mark), tropomyosin, and h-actin (1:500; Sigma, Saint Louis, MO). These membranes were exposed to the secondary anti-goat antibody for galectin- 1, and a-enolase, anti-rabbit antibody for EB1, and anti-mouse antibody for calponin, smooth muscle actin, tropomyosin, and h-actin (1:20,000; Santa Cruz Biotechnology). Signals were detected by enhanced chemilumines- Figure 1. Two-dimensional gel images with selected distinguishing proteins in cence Substrate (Pierce). meningiomas. Two-dimensional and cropped gel images with identified proteins Immunohistochemical analysis. Calponin-1 was detected immunohis- in benign, atypical, and anaplastic meningiomas. A, representative two- tochemically on paraffin-embedded tumor sections using antigen retrieval. dimensional gels for benign (a and b), atypical (c and d), and anaplastic meningiomas (e and f). B, distinguishing proteins on the cropped gel images. The streptavidin-biotin complex method was applied. Primary antibody a, placental ribonuclease inhibitor; b, galectin-1; c, tropomyosin 1 a chain; incubation was done with monoclonal anti-calponin antibody (1:50; d, a-enolase. I, benign meningiomas; II, atypical meningiomas; III, anaplastic DakoCytomation). meningiomas.

Cancer Res 2006; 66: (20). October 15, 2006 10200 www.aacrjournals.org

Table 1.

Name Swiss- Location kDa Function P Intensity in two-dimensional gels Prot ID Benign Atypical Anaplastic

A. Up-regulated proteins in benign meningiomas compared with atypical meningiomas 1 Calponin-1 P51911 19p13.2 33.2 Actin-binding protein 0.046 0.252 0.18 0.19 2 Activator of 90-kDa heat shock O95433 14q24.3 38.2 Cochaperone 0.049 0.201 0.135 0.1 protein ATPase homologue 1 3 Serine-threonine kinase Q9Y3F4 12p12.3 38.4 Interactin with 0.033 0.314 0.184 0.277 receptor-associated protein receptor of TGF-h 4 Placental ribonuclease inhibitor P13489 11p15.5 49.8 RNase inhibiton 0.017 0.678 0.35 0.569 5 Vitamin D-binding protein P02774 4q13.3 53 Precursor of macrophage 0.022 0.735 0.374 0.628 activating factor 6 Apolipoprotein A-I P02647 11q23.3 30.8 Cholesterol transport 0.042 3.951 1.87 1.937

B. Up-regulated proteins in atypical meningiomas compared with benign meningiomas 1 Peroxiredoxin 6 P30041 1q25.1 29.4 Antioxidant 0.047 0.172 0.431 0.216 2 Galectin-1 P09382 22q13.1 14.6 h-galactoside-binding 0.04 0.111 0.275 0.312 protein 3 Retinoic acid-binding P29373 1q23.1 15.6 Regulation of retinoic 0.028 0.413 0.995 0.464 protein II, cellular acid signaling 4 Peroxiredoxin 2 P32119 19p13.13 21.9 Antioxidant 0.01 0.146 0.339 0.161 5 Voltage-dependent anion-selective P45880 10q22.2 38.1 Anti-apoptosis 0.02 0.171 0.391 0.257 channel protein 2 6 Actin, aortic smooth muscle P62736 10q23.31 42 Cytoskelton 0.017 0.125 0.285 0.13 7 Annexin A1 P04083 9q21.13 38.6 Calcium-binding 0.001 0.166 0.344 0.46 protein 8 Proteasome activator Q9UL46 14q11.2 27.2 Proteolysis 0.01 0.355 0.597 0.355 complex subunit 2 9 ATP synthase D chain, O75947 17q25.1 18.4 ATP synthase 0.049 0.705 1.045 0.504 mitochondrial

Six of the nine identified proteins were significantly up-regulated in Western blot and immunohistochemical analysis for verifi- atypical meningiomas compared with anaplastic meningiomas, cation of identified proteins. To validate the two-dimensional gel whereas three proteins were differentially overexpressed in electrophoresis results, we did Western blot analysis and immu- anaplastic meningiomas (Table 2A and B). nohistochemistry (Figs. 2 and 3). Subject to the availability of

Table 2.

Name Swiss- Location kDa Function P Intensity in two-dimensional gels Prot ID Benign Atypical Anaplastic

A. Up-regulated proteins in atypical meningiomas compared with anaplastic meningiomas 1 Microtubule-associated protein, Q15691 20q11.21 29.9 Stabilizing microtubules. 0.012 0.265 0.297 0.184 RP/EB family, member 1 Binding to APC 2 Enoyl-CoA hydratase, mitochondrial P30084 10q26.3 31.4 Fatty acid metabolism 0.039 0.583 0.635 0.378 3 Proteasome subunit a type 1 P25786 11p15.2 29.6 Proteolysis 0.028 0.218 0.243 0.132 4 Acid ceramidase Q13510 8p22 44.7 Catalysis of lysosomal 0.024 0.163 0.283 0.149 hydrolysis of ceramide 5 Selenium-binding protein 1 Q13228 10q23.31 52.3 May participate in 0.021 1.078 1.313 0.411 protein transport in Golgi 6 Tropomyosin 1 a chain P09493 15q22.2 32.7 Binding to actin filaments 0.027 0.332 0.264 0.082

B. Up-regulated proteins in anaplastic meningiomas compared with atypical meningiomas 1 a-Enolase P06733 1p36.23 47 Binding to the c-myc 0.002 0.498 0.61 1.993 promoter 2 a-1-antitrypsin P01009 14q32.13 46.7 Inhibition of serine 0.044 1.841 1.067 2.349 proteases 3 Fibrinogen gamma chain P02679 4q31.3 51.5 Platelet aggregation 0.039 0.247 0.371 0.666

www.aacrjournals.org 10201 Cancer Res 2006; 66: (20). October 15, 2006

suppression activity. Vitamin D–binding protein (DBP), significantly up-regulated in WHO grade 1 tumors, is a multifunctional protein that binds and transports vitamin D, acts as an actin scavenger, and is a precursor of macrophage-activating factor (MAF), which is a tumoricidal agent in various types of tumors (14–17). DBP possesses antiangiogenic activity, and cancer cells can inactivate DBP function by deglycosylation (14–17). Photo- dynamic therapy with DBP-derived MAF has been shown to have synergistic antitumor effect in squamous cell carcinoma tumor models (18). DBP is inactivated in several other cancers as well (14–17). Another protein that was identified, calponin-1, is an actin-associated protein originally isolated from bovine aorta and chicken gizzard (19, 20). Horiuchi et al. showed that calponin may function as a tumor suppressor gene in human uterine leiomyosarcomas (19). Thus, loss of calponin-1 may potentiate Figure 2. Verification of expression of six identified proteins by Western blotting. or accelerate progression from a low-grade meningioma into a Lanes 1 to 4, benign meningiomas; lanes 5 to 7, atypical meningiomas; lanes 8 and 9, anaplastic meningiomas. As a loading control, h-actin expression is higher-grade tumor. included. Several of the nine proteins that were significantly overexpressed in WHO grade 2 meningiomas have potential cancer-related functions. Galectin-1 mediates adhesions between cells and with the extracellular matrix and has been shown to play a role in antibodies, we did Western blot analysis on selected proteins that tumorigenesis and to enhance metastatic potential (21). It was also were differentially expressed in the comparisons between benign reported to play a role in synaptogenesis, axon guidance, and cell and atypical meningiomas, with antibodies against galectin-1, process fasciculation in the brain (22, 23). Expression of galectin-1 calponin-1, and smooth muscle actin as well as to several proteins enhances the activity of cell migration in human gliomas (23, 24). that were significantly expressed in the comparison between Gliomas with overexpression of galectin-1 behave in a more atypical and anaplastic meningiomas with antibodies against malignant fashion. Antisense against galectin-1 inhibits growth of enolase, EB1, and tropomyosin (Fig. 2). Western blot for h-actin 9L glioma cells (25). Paz et al. found that interaction of galectin-1 was used as an internal control. Western blotting consistently and Ras, which is associated with cell proliferation in meningio- confirmed the differential protein expression determined by two- mas, was necessary for membrane fixation of Ras and subsequent dimensional gel electrophoresis. Finally, we did immunohisto- cell transformation (26). This suggests that galectin-1 could be a chemical staining for the expression of calponin on independent paraffin sections of 30 meningiomas, which further validated the differential proteomic findings in this diverse series of meningiomas (Fig. 3).

Discussion Most meningiomas are benign, WHO grade 1, and tend to follow indolent clinical course. However, atypical and anaplastic meningiomas, WHO grade 2 and 3 tumors, respectively, exhibit more aggressive behavior and are associated with poorer outcome. Atypical meningiomas are characterized histologically by several of the following features: increased cellularity, small cells with nuclear pleomorphism, increased mitotic activity (MIB-1 staining of nearly 5% compared with <0.1% for grade 1 tumors), an uninterrupted pattern of sheet-like growth, and occasional foci of necrosis (1, 12). Anaplastic, WHO grade 3 meningiomas exhibit morphologically more aggressive histologic features with higher mitotic activity (mean MIB-1 index is f15%; ref. 1) than atypical meningiomas. The median survival of patients with this subgroup is <2 years due to aggressive, invasive, and recurrent tumors (1, 13). Although there has been keen interest in identifying adjuvant markers that will be of use in meningiomas grading, none have been identified that have consistent use. Therefore, we applied proteomic tools to identify proteins that distinguish meningiomas of varying grades. We identified 15 proteins that are differentially expressed between benign and atypical meningiomas (Table 1A and B). Six Figure 3. Immunohistochemical detection of calponin. Representative examples of immunohistochemistry with anti-calponin antibody for benign (A), of these proteins were up-regulated in benign meningiomas, some atypical (B), and anaplastic meningioma (C) confirm abundant expression of the of which have previously been shown to have potential tumor protein in the benign but not in the other groups. Original magnification, Â200.

Cancer Res 2006; 66: (20). October 15, 2006 10202 www.aacrjournals.org

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 2006 American Association for Cancer Research. Proteomic Profiling of Meningiomas potential therapeutic target in these meningiomas, either indirectly, ization analyses have identified gene expression patterns associ- through interactions with Ras (e.g., using farnesyltransferase ated with different grades in 30 cases of meningiomas (42). inhibitors) or directly. Wrobel et al. found that expression of proteins within the insulin- We also identified two antioxidant proteins, peroxiredoxin 2 like growth factor and WNT signaling pathways were increased in and 6, in grade 2 tumors. Resistance to oxidation is a feature of atypical and anaplastic meningiomas and were associated with many tumors, which results in improved tolerance for hypoxia allelic losses on chromosome 10 or 14 (42). Simon et al. reported and resistance to chemotherapy. Antisense of peroxiredoxin 2 that deletion of 1p32-pter, 10q24-qter, and 14q24-q32 are more enhances cisplatin-induced or radiation-induced cell death in common in atypical and anaplastic meningiomas than WHO cancer cells (27, 28). Finally, two more identified proteins (voltage- grade 1 tumors (43). Mihaila et al. showed that losses on dependent anion-selective channel protein 2 and annexin A1) chromosome 10p15.1 and 10q26.3 were associated with higher- also may play roles in the genesis or progression to atypical grade meningiomas, and LOH at 10q26.1 were directly correlated meningiomas (29, 30). with shorter survival (44). Some of the proteins that we identified When comparing the proteomic profiles of atypical and are located on these chromosomes, such as activator of 90-kDa anaplastic meningiomas, we identified nine proteins that were heat shock protein ATPase homologue 1 and enoyl-CoA hydratase expressed differentially in one grade of tumor or the other. Six (Tables 1 and 2). These findings confirm and extend previous were up-regulated in atypical meningiomas compared with genetic and expression profiling studies. anaplastic meningiomas (Table 2A). Selenium-binding protein 1, Expressions of some identified proteins are consistently up- which was significantly up-regulated in WHO grade 2 meningi- regulated/down-regulated parallel to the progression of the grade omas, is one of the three mammalian selenium-containing in meningiomas, such as tropomyosin 1 a chain and a-enolase, proteins (31, 32). Supplementation of selenium has been pro- but most of the proteins are not. Several possibilities may explain posed to reduce total cancer incidence and cancer mortality with this. The development of most tumors is thought to be initiated prostate, lung, and colorectal cancers (33). Reduced expression of by the clonal expansion of a single cell that possesses alterations selenium-binding protein 1 was observed in poorly differentiated that convey growth advantage. With tumors that possess more lung adenocarcinomas and is associated with poor prognosis than one grade of malignancy (compared with normal histology (32). Another protein, EB1, a microtubule-associated protein, of the cell or organ of origin), it is unclear if in every case tumors interacts with the tumor suppressor gene adenomatous polyposis of a high grade arose from a lower-grade tumor or if they began coli (APC) to contribute to microtubule stability (34). Loss of as a higher-grade tumor. Glioblastomas, for example, arise in EB1’s interaction with APC could result in genetic instability and either fashion (10). The situation with meningiomas is unclear. trigger tumorigenesis (34). Tropomyosin-1 is an actin-binding Although for most grade 1 tumors, alterations in NF2 seem protein that is associated with stability of actin filaments and cell generally to play a necessary, although not sufficient, role in motility (35). Expression of tropomyosin-1 is decreased in some tumorigenesis. The fact that higher-grade meningiomas may types of tumors such as breast cancers compared with normal resemble sarcomas suggests that this is one possibility. Al-Mefty tissues (36). Hughes et al. reported an inverse relationship et al. documented that with benign meningiomas, which recurred between expression of tropomyosin and grades in human brain as high-grade tumors, the original tumor already possessed tumors (mainly astrocytic tumors), and that tropomyosin could cytogenetic alterations, suggesting the potential to progress (9). be a potential marker for distinguishing the grade of human Rempel et al. identified a secreted, extracellular matrix-associated brain tumors (37). In contrast, a-enolase, a potent glycolytic protein (SPARC) as a possible marker of invasive behavior. enzyme, was one of the three proteins that were up-regulated in Although SPARC expression was detected in all grade z2 anaplastic meningiomas. Expression of a-enolase has a reverse meningiomas, it was found only in grade 1 tumors that developed correlation with tumor differentiation in human hepatocellular an invasive phenotype (45). This suggests that proteins or protein carcinomas and is also associated with tumor size and patterns, such as some of those that we have identified, may serve invasiveness (38). In getting to meningiomas, a-enolase may to identify lower-grade tumors that are likely to be invasive or serve as a similar marker of differentiation status in WHO grade behave more aggressively. 3 tumors. In summary, we used selective tissue microdissection to obtain a Previous investigators have identified a series of genetic pure population of meningioma cells and submitted it to two- alterations that characterize nearly all meningiomas. Loss of dimensional gel electrophoresis and definitive LC-MS/MS protein heterozygosity (LOH) on chromosome 22, which contains the sequencing to identify proteomic profiles and proteins that neurofibromatosis type 2 (NF2) tumor suppressor gene, is the distinguish WHO grade 1, 2, and 3 meningiomas. A relatively small most common genetic abnormality in sporadic meningiomas number of proteins were sufficient to differentiate meningiomas of (1, 8, 39). Mutation, deletion, and hypermethylation of NF 2 leads each grade. This method is simple, accurate, and reliable and may to depression, or complete loss, of expression of the protein prove useful to gain a broader understanding of the genesis and merlin (40). In atypical meningiomas, additional genetic alter- progression of meningiomas. ations include losses on chromosomes 1p, 6q, 10, 14q, and 18q and gains on 1q, 9q, 12q, 15q, 17q, and 20q (1, 8). Anaplastic meningiomas show supplementary changes, which include deletions along 9p and amplification of 17q (1, 8). Muller et al. Acknowledgments showed that monosomy 1p and loss of expression of tissue Received 3/14/2006; revised 4/14/2006; accepted 7/13/2006. nonspecific alkaline phosphatase (ALPL), located on 1p36, were Grant support: Intramural Research Program of the NIH, National Institute of strongly correlated with WHO grade (41), which suggests that Neurological Disorders and Stroke. ALPL The costs of publication of this article were defrayed in part by the payment of page may be a tumor suppressor gene in meningiomas. charges. This article must therefore be hereby marked advertisement in accordance Furthermore, cDNA microarray and comparative genomic hybrid- with 18 U.S.C. Section 1734 solely to indicate this fact. www.aacrjournals.org 10203 Cancer Res 2006; 66: (20). October 15, 2006

References maf) inhibits angiogenesis and tumor growth in mice. 33. Duffield-Lillico AJ, Reid ME, Turnbull BW, et al. Neoplasia 2003;5:32–40. Baseline characteristics and the effect of selenium 1. Louis DN, Scheithauer BW, Budka H, von Deimling A, 17. Gomme PT, Bertolini J. Therapeutic potential of supplementation on cancer incidence in a randomized Kepes JJ. Meningiomas. In: Kleihues P, Cavenee WK, vitamin D-binding protein. Trends Biotechnol 2004;22: clinical trial: a summary report of the Nutritional editors. Tumours of the nervous system. Lyon: IAPC 340–5. Prevention of Cancer Trial. Cancer Epidemiol Bio- Press; 2000. p. 176–84. 18. Korbelik M, Naraparaju VR, Yamamoto N. Macro- markers Prev 2002;11:630–9. 2. Mahmood A, Caccamo DV, Tomecek FJ, Malik GM. phage-directed immunotherapy as adjuvant to photo- 34. Morrison EE, Wardleworth BN, Askham JM, Atypical and malignant meningiomas: a clinicopatho- dynamic therapy of cancer. Br J Cancer 1997;75:202–7. Markham AF, Meredith DM. EB1, a protein which logical review. Neurosurgery 1993;33:955–63. 19. Horiuchi A, Nikaido T, Taniguchi S, Fujii S. Possible interacts with the APC tumour suppressor, is associated 3. Modha A, Gutin PH. Diagnosis and treatment of role of calponin h1 as a tumor suppressor in human with the microtubule cytoskeleton throughout the cell atypical and anaplastic meningiomas: a review. Neuro- uterine leiomyosarcoma. J Natl Cancer Inst 1999;91: cycle. Oncogene 1998;17:3471–7. surgery 2005;57:538–50; discussion -50. 790–6. 35. Varga AE, Stourman NV, Zheng Q, et al. Silencing of 4. Ng HK, Wong AT. Expression of epithelial and 20. Takahashi K, Yamamura H. Studies and perspectives the Tropomyosin-1 gene by DNA methylation alters extracellular matrix protein markers in meningiomas. of calponin in smooth muscle regulation and cancer tumor suppressor function of TGF-h. Oncogene 2005;24: Histopathology 1993;22:113–25. gene therapy. Adv Biophys 2003;37:91–111. 5043–52. 5. Vogelstein B, Fearon ER, Hamilton SR, et al. Genetic 21. Perillo NL, Marcus ME, Baum LG. Galectins: versatile 36. Raval GN, Bharadwaj S, Levine EA, et al. Loss of alterations during colorectal-tumor development. modulators of cell adhesion, cell proliferation, and cell expression of tropomyosin-1, a novel class II tumor N Engl J Med 1988;319:525–32. death. J Mol Med 1998;76:402–12. suppressor that induces anoikis, in primary breast 6. Kleihues P, Ohgaki H. Primary and secondary glio- 22. Zanetta JP. Structure and functions of lectins in the tumors. Oncogene 2003;22:6194–203. blastomas: from concept to clinical diagnosis. Neuro- central and peripheral nervous system. Acta Anat 37. Hughes JA, Cooke-Yarborough CM, Chadwick NC, oncol 1999;1:44–51. (Basel) 1998;161:180–95. et al. High-molecular-weight tropomyosins localize to 7. Kleihues P, Burger PC, Collins VP, E.W. N, Ohgaki H, 23. Camby I, Belot N, Lefranc F, et al. Galectin-1 the contractile rings of dividing CNS cells but are absent Cavenee WK. Glioblastoma. In: Kleihues P, Cavenee WK, modulates human glioblastoma cell migration into the from malignant pediatric and adult CNS tumors. Glia editors. Tumours of the nervous system. Lyon: IAPC brain through modifications to the actin cytoskeleton 2003;42:25–35. Press; 2000. p. 29–39. and levels of expression of small GTPases. J Neuropathol 38. Takashima M, Kuramitsu Y, Yokoyama Y, et al. 8. Weber RG, Bostrom J, Wolter M, et al. Analysis of Exp Neurol 2002;61:585–96. Overexpression of alpha enolase in hepatitis C virus- genomic alterations in benign, atypical, and anaplastic 24. Rorive S, Belot N, Decaestecker C, et al. Galectin-1 is related hepatocellular carcinoma: association with meningiomas: toward a genetic model of meningioma highly expressed in human gliomas with relevance for tumor progression as determined by proteomic analysis. progression. Proc Natl Acad Sci U S A 1997;94:14719–24. modulation of invasion of tumor astrocytes into the Proteomics 2005;5:1686–92. 9. Al-Mefty O, Kadri PA, Pravdenkova S, Sawyer JR, brain parenchyma. Glia 2001;33:241–55. 39. Ruttledge MH, Sarrazin J, Rangaratnam S, et al. Stangeby C, Husain M. Malignant progression in 25. Yamaoka K, Mishima K, Nagashima Y, Asai A, Sanai Y, Evidence for the complete inactivation of the NF2 gene meningioma: documentation of a series and analysis Kirino T. Expression of galectin-1 mRNA correlates with in the majority of sporadic meningiomas. Nat Genet of cytogenetic findings. J Neurosurg 2004;101:210–8. the malignant potential of human gliomas and expres- 1994;6:180–4. 10. Furuta M, Weil RJ, Vortmeyer AO, et al. Protein sion of antisense galectin-1 inhibits the growth of 9 40. Lomas J, Bello MJ, Arjona D, et al. Genetic and patterns and proteins that identify subtypes of glioblas- glioma cells. J Neurosci Res 2000;59:722–30. epigenetic alteration of the NF2 gene in sporadic toma multiforme. Oncogene 2004;23:6806–14. 26. Paz A, Haklai R, Elad-Sfadia G, Ballan E, Kloog Y. meningiomas. Genes Chromosomes Cancer 2005;42: 11. Iwadate Y, Sakaida T, Hiwasa T, et al. Molecular Galectin-1 binds oncogenic H-Ras to mediate Ras 314–9. classification and survival prediction in human membrane anchorage and cell transformation. Onco- 41. Muller P, Henn W, Niedermayer I, et al. Deletion of gliomas based on proteome analysis. Cancer Res 2004; gene 2001;20:7486–93. chromosome 1p and loss of expression of alkaline 64:2496–501. 27. Park SH, Chung YM, Lee YS, et al. Antisense of phosphatase indicate progression of meningiomas. Clin 12. Perry A, Gutmann DH, Reifenberger G. Molecular human peroxiredoxin II enhances radiation-induced cell Cancer Res 1999;5:3569–77. pathogenesis of meningiomas. J Neurooncol 2004;70: death. Clin Cancer Res 2000;6:4915–20. 42. Wrobel G, Roerig P, Kokocinski F, et al. Microarray- 183–202. 28. Yo YD, Chung YM, Park JK, Ahn CM, Kim SK, Kim HJ. based gene expression profiling of benign, atypical and 13. Perry A, Scheithauer BW, Stafford SL, Lohse CM, Synergistic effect of peroxiredoxin II antisense on anaplastic meningiomas identifies novel genes associ- Wollan PC. ‘‘Malignancy’’ in meningiomas: a clinico- cisplatin-induced cell death. Exp Mol Med 2002;34:273–7. ated with meningioma progression. Int J Cancer 2005; pathologic study of 116 patients, with grading implica- 29. Cheng EH, Sheiko TV, Fisher JK, Craigen WJ, 114:249–56. tions. Cancer 1999;85:2046–56. Korsmeyer SJ. VDAC2 inhibits BAK activation and 43. Simon M, von Deimling A, Larson JJ, et al. Allelic 14. Koga Y, Naraparaju VR, Yamamoto N. Antitumor mitochondrial apoptosis. Science 2003;301:513–7. losses on chromosomes 14, 10, and 1 in atypical and effect of vitamin D-binding protein-derived macrophage 30. Oh P, Li Y, Yu J, et al. Subtractive proteomic mapping malignant meningiomas: a genetic model of meningio- activating factor on Ehrlich ascites tumor-bearing mice. of the endothelial surface in lung and solid tumours for ma progression. Cancer Res 1995;55:4696–701. Proc Soc Exp Biol Med 1999;220:20–6. tissue-specific therapy. Nature 2004;429:629–35. 44. Mihaila D, Jankowski M, Gutierrez JA, et al. 15. Kanda S, Mochizuki Y, Miyata Y, Kanetake H, 31. Behne D, Kyriakopoulos A. Mammalian selenium- Meningiomas: loss of heterozygosity on chromosome Yamamoto N. Effects of vitamin D(3)-binding protein- containing proteins. Annu Rev Nutr 2001;21:453–73. 10 and marker-specific correlations with grade, recur- derived macrophage activating factor (GcMAF) on 32. Chen G, Wang H, Miller CT, et al. Reduced selenium- rence, and survival. Clin Cancer Res 2003;9:4443–51. angiogenesis. J Natl Cancer Inst 2002;94:1311–9. binding protein 1 expression is associated with poor 45. Rempel SA, Ge S, Gutierrez JA. SPARC: a potential 16. Kisker O, Onizuka S, Becker CM, et al. Vitamin D outcome in lung adenocarcinomas. J Pathol 2004;202: diagnostic marker of invasive meningiomas. Clin Cancer binding protein-macrophage activating factor (DBP- 321–9. Res 1999;5:237–41.

Cancer Res 2006; 66: (20). October 15, 2006 10204 www.aacrjournals.org

Hiroaki Okamoto, Jie Li, Alexander O. Vortmeyer, et al.

Cancer Res 2006;66:10199-10204.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/66/20/10199

Cited articles This article cites 43 articles, 9 of which you can access for free at: http://cancerres.aacrjournals.org/content/66/20/10199.full#ref-list-1

Citing articles This article has been cited by 3 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/66/20/10199.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/66/20/10199. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.