Comparative Gene Expression Profile Analysis of Neurofibromatosis 1-Associated and Sporadic Pilocytic Astrocytomas1

Home , Fibrillary astrocytoma

[CANCER RESEARCH 62, 2085–2091, April 1, 2002] Comparative Gene Expression Profile Analysis of Neurofibromatosis 1-associated and Sporadic Pilocytic Astrocytomas1

David H. Gutmann,2 Nicole´M. Hedrick, Jun Li, Rakesh Nagarajan, Arie Perry, and Mark A. Watson Departments of Neurology [D. H. G., N. M. H., J. L.] and Pathology and Immunology [R. N., A. P., M. A. W.], Washington University School of Medicine, St. Louis, Missouri 63110

ABSTRACT a clinically benign fashion; however, a small but significant fraction of these tumors will continue to grow and cause loss of vision or Pilocytic astrocytomas (PAs) are WHO grade I brain tumors that do dysfunction of the hypothalamus. not typically progress to more malignant grades of astrocytoma. Whereas It has been reported that PAs express molecular markers shared there have been significant advances in the molecular genetics of high- grade astrocytomas, relatively little is known about the genetic changes with oligodendrocytes or O2A precursors. O2A progenitor cells rep- associated with PA formation. In an effort to better characterize these resent multipotent precursor cells that can mature into either type 2 low-grade neoplasms, we compared the gene expression profiles of six astrocytes or oligodendrocytes (5–7). Low-grade astrocytomas have sporadic and two neurofibromatosis 1-associated PAs with other tissues been shown to express myelin PLP and MBP (8, 9), as well as the and cell lines of both astrocytic and oligodendroglial origin. Hierarchical PEN-5 epitope (10), which is not detected in high-grade astrocytomas. cluster analysis of gene expression data clearly delineated PAs from These proteins are expressed in oligodendrocytes and O2A cells, as low-grade oligodendrogliomas and normal white matter. The two NF1- well as a large proportion of oligodendrogliomas, suggesting that PAs associated tumors and one of the sporadic PAs displayed expression may arise from a different cell of origin (e.g., O2A progenitor) than profiles that were more closely related to those of cultured normal human the higher-grade fibrillary astrocytoma. fetal astrocytes. However, PAs also expressed individual genes typically The increased incidence of PAs in NF1 patients suggests that the associated with oligodendroglial lineage (e.g., proteolipid protein and PMP-22). The expression patterns of specific genes (e.g., ApoD) were NF1 tumor suppressor gene is a critical growth regulator for astro- unique to PA tumors, whereas genetic changes characteristic of high- cytes. Previous work from our laboratory and others has demonstrated grade astrocytomas were not encountered. Differential expression of two loss of NF1 gene expression in NF1-associated PAs but not in spo- transcripts, neural cellular adhesion molecule and connexin-43, was con- radic tumors (11, 12). These results argue that NF1 loss is associated firmed at the protein level, suggesting that these cell adhesion molecules with the development of NF1-associated PAs and that other genes might be particularly important in the molecular pathogenesis of these must be important for the pathogenesis of sporadic PAs. Compared tumors. We conclude that PAs are genetically unique gliomas with gene with the more malignant fibrillary astrocytoma, relatively little is expression profiles that resemble those of fetal astrocytes and, to a lesser known about the molecular pathogenesis of PAs. Cytogenetic studies extent, oligodendroglial precursors. have demonstrated a variety of chromosomal aberrations, including gains of chromosomes 7, 8, 9, and 11 (13–15). Whereas high-grade INTRODUCTION astrocytomas demonstrate overexpression and/or amplification of the cyclin-dependent kinase 4 and the epidermal growth factor receptor PAs3 are the most common type of childhood central nervous genes, as well as losses of the p16, p53, and PTEN/MMAC1 tumor system astrocytoma. These tumors are classified by the WHO as grade suppressor genes, these genetic changes have not been identified in I neoplasms, and they typically behave in a benign fashion (1). PAs PAs (16, 17). appear as relatively discrete lesions on gross neuropathological ex- To better understand the molecular events associated with PA amination when compared with their more diffusely infiltrative fibril- pathogenesis, we used differential expression profiling of eight PAs, lary counterparts (WHO grades II-IV). Other important features in- including two from individuals affected with NF1. The objective of clude the presence of Rosenthal fibers, a frequent cystic component, this study was to define expression profiles characteristic of PAs and and a lower incidence of malignant progression. These tumors are to identify potential genes that might be important for PA pathogen- frequently encountered in the context of NF1, a common inherited esis. In addition, we sought to determine whether PAs had gene tumor predisposition syndrome (2). PAs are observed in 15–20% of expression profiles more similar to those of low-grade oligodendro- children with NF1 and most frequently involve the optic nerve, optic gliomas and normal white matter as compared with those of commit- chiasm, optic pathway, and hypothalamus. These tumors are also ted astrocytic precursors. Such a finding would provide support for the found in the brainstem, cerebral cortex, and cerebellum, the latter hypothesis of an O2A progenitor cell of origin for PAs. representing a site of reportedly increased aggressiveness when associated with NF1 (3). In the setting of NF1, PAs develop almost exclusively in children with a mean age at diagnosis of 4.5 years (4). MATERIALS AND METHODS As is true for the sporadic PAs, most NF1-associated PAs behave in Tissue Procurement and Sample Preparation. All tissue samples were collected by the Siteman Cancer Center Tissue Procurement Facility under an Received 9/5/01; accepted 1/30/02. approved protocol from the institution’s Human Studies Committee. Resected The costs of publication of this article were defrayed in part by the payment of page tumor tissue was immediately snap frozen in liquid nitrogen. Frozen tumor charges. This article must therefore be hereby marked advertisement in accordance with ␮ 18 U.S.C. Section 1734 solely to indicate this fact. specimens were embedded in freezing medium, sectioned at 5 m, and stained 1 Supported by NIH Grant NS36996 (to D. H. G.). with H&E. The histopathology of each collected specimen was reviewed to 2 To whom requests for reprints should be addressed, at Department of Neurology, confirm the adequacy of the sample (i.e., minimal contamination with non- Washington University School of Medicine, 660 South Euclid Avenue, Box 8111, St. neoplastic elements) and assess the extent of tumor necrosis and cellularity. Louis, MO 63110. Phone: (314) 362-7379; Fax: (314) 362-2388; E-mail: gutmannd@ ␮ neuro.wustl.edu. Subsequent 50- m serial sections from each banked frozen specimen were 3 The abbreviations used are: PA, pilocytic astrocytoma; NF1, neurofibromatosis 1; then cut and placed immediately into TRIzol reagent (Life Technologies, Inc.), O2A, oligodendrocyte-type 2 astrocyte; PLP, proteolipid protein; MBP, myelin basic and homogenized. NHA cells were obtained from Clonetics (Walkersville, protein; NHA, normal human fetal astrocyte; SADV, scaled average difference value; CV, MD), maintained in the recommended media, and passaged according to the coefficient of variation; SAGE, serial analysis of gene expression; PMP-22, peripheral myelin protein 22; EF1-␣2, elongation factor 1 ␣ 2; gp36, glycoprotein 36; ApoD, manufacturer’s specifications. NHA cells from passages five through eight Apolipoprotein D; N-CAM, cellular adhesion molecule identified in neurons. were used for RNA extraction using the TRIzol reagent. Total RNA was 2085

Fig. 1. Hierarchical cluster analysis of patient tumor specimens. Gene expression data were scaled, normalized, and subjected to hierarchical cluster analysis as described in the text. The dendrogram at the top displays the relationship of the samples based on their global pattern of gene expression. The samples could be divided into two primary clusters (A), two secondary clusters (B), and two tertiary clusters (C) that were maintained regardless of the clustering algorithm used. Below the dendrogram is a “heat map” representing gene expression with each of the 16 samples represented in a column, each of the 9628 genes represented as a horizontal line, and the relative expression of any one gene in any one sample in continuous linear grayscale from high (black) to low (white) expression. The table (bottom) provides basic diagnostic and demographic information from each of the tumor samples, as well as the number of genes scored as detected (% Genes ϭ P) by the GeneChip software algorithm and the average hybridization signal intensity of detected genes (Avg Diff P).

isolated using the manufacturer’s protocol. Extracted RNA was then further clustering, SADVs were normalized to a mean of 0 and an SD of 1. The purified by spin chromatography (RNeasy kit; Qiagen) following the manu- visualization in Fig. 1 was created using the average linkage of Euclidean facturer’s protocol. Purified RNA was quantitated by UV absorbance at 260 distance measures method using the programs CLUSTER and TREEVIEW5 and 280 nm and assessed qualitatively using an RNA LabChip and Bioanalyzer (19). For other comparative analyses of the filtered gene sets, flat text file data 2100 (Agilent). of Absolute Calls and SADVs were imported into DecisionSite 6.0 and Array Oligonucleotide Array Analysis. Analysis was performed by the Siteman Explorer Software (Spotfire). To distinguish technical variability from biolog- Cancer Center GeneChip Facility. Purified total RNA (10 ␮g) was spiked with ical differences in gene expression, a CV was calculated using SADVs for the a set of four synthetic, polyadenylated, and bacterial transcripts (Lys, Phe, Thr, 3Ј end probe sets directed to the internal standard bacterial transcripts (Lys, and Trp) diluted to defined copy numbers. Oligonucleotide probes for these Phe, Thr, and Trp). This calculation included values from 11 of the 16 samples, transcripts are present on all Affymetrix GeneChips such that monitoring the as 5 of the samples were not spiked with the transcript controls. The maximum expression level of these internal standards provides an indication of the total CV of the control transcripts (0.69) was then used as a threshold value for the technical variability associated with the experiment. Spiked RNA was con- genes listed in Fig. 2. To define genes with significant differences between verted to cDNA, purified, and then used as a template for in vitro transcription astrocytic and oligodendrocytic specimens and between normal human astro- of biotin-labeled antisense RNA. All protocols were performed as recom- cytes and PAs, SADVs were used in a standard t test, correcting Ps for multiple mended by the manufacturer (Affymetrix) and have been described elsewhere tests using permutation analysis. For SAGE analysis, the indicated tissue (18). Each biotinylated antisense RNA preparation (20 ␮g) was fragmented, SAGE tag libraries were examined using the Xprofiler tool,6 selecting genes assessed by gel electrophoresis, and placed in hybridization cocktail containing with Ͼ90% probability of being differentially represented in the two libraries four biotinylated hybridization controls (BioB, BioC, BioD, and Cre) as at Ն3-fold. HTML output from the analysis was imported into Microsoft recommended by the manufacturer. Samples were hybridized to Affymetrix Access. Unigene IDs corresponding to Affymetrix probe pair sets7 were linked Hu95A GeneChip arrays for 16 h. GeneChips were washed and stained using with Unigene IDs from Xprofiler output to create a concordant gene list. the instrument’s standard Eukaryotic GE Wash 2 protocol, using antibody- Western Blot. Frozen tumor specimens were homogenized in standard mediated signal amplification. The images from the scanned chips were protein lysis (radioimmunoprecipitation assay) buffer containing protease in- processed using Affymetrix Microarray Analysis Suite 4.0. The image from hibitors and the protein concentration determined by the Bradford method each GeneChip was scaled such that the average intensity value for all arrays (Bio-Rad Laboratories, Hercules, CA). Total protein (100 ␮g) was separated was adjusted to a target intensity of 1500. SADV and Absolute Call Data from by SDS PAGE as described previously (17). Proteins were transferred onto each GeneChip were exported as flat text files and used for further analysis. Immobilon membranes for Western blotting. The N-CAM (clone B11; mouse 4 The complete set of raw data will be made available at the Web site. monoclonal antibody; Sigma Chemical Co., St. Louis, MO), WA30 (merlin Data Analysis. Of the total of 12,651 gene sequences represented on the polyclonal antibody), and connexin-43 (rabbit polyclonal; Sigma Chemical array, hybridization control sequences and sequences scored as “A” (not Co.) antibodies were used according to the manufacturer’s recommendations detected) in all 16 samples were excluded from initial analysis. For the remaining 9,628 genes, all SADVs of Ͻ1 were set to 1. For hierarchical 5 Internet address: http://rana.lbl.gov/EisenSoftware.htm. 6 Internet address: http://www.ncbi.nlm.nih.gov/SAGE/. 4 Internet address: http://pathbox.wustl.edu/ϳmgacore. 7 Internet address: http://www.netaffx.com/index2.jsp. 2086

Fig. 2. Absolute gene expression levels in each of the 16 tumor samples of genes associated previously with astrocytomas or the astroglial phenotype. The numerical values in the table represent the GeneChip SADV metric. To the right of the table, values are represented pictorially in linear grayscale, with high expression levels in black and low expression levels in white. Genes are arranged by functional category and accompanied by corresponding GenBank accession numbers. Samples are arranged based on hierarchical cluster analysis differential expression (CV of gene expression across all 16 samples) exceeds CV of internal standards; A, statistically ,ء :described in Fig. 1. Notations after each gene name indicate significant difference (adjusted P Ͻ 0.05, t test) between oligodendrocytic (n ϭ 6) and astrocytic (n ϭ 10) specimens; P, statistically significant difference (adjusted P Ͻ 0.05, t test) between PAs (n ϭ 8) and normal human astrocytes (n ϭ 2). See text for details concerning the statistical methods. Expression values for the internal controls (Lys, Phe, Thr, and Trp) are included for reference at the bottom of the table. or our methods published previously (20). Tubulin (Sigma Chemical Co.; in each sample ranged from 45 to 56%, suggesting that the overall clone DM1A) was used as an internal control for protein quality (data not quality of each of the labeled targets and the resulting fluorescent shown). Development was accomplished using appropriate horseradish perox- signal obtained from each of the 16 GeneChips was comparable. The idase-conjugated secondary antibodies and enhanced chemiluminescence expression level of four different internal control transcripts (added to (Amersham, Arlington Heights, IL). 11 of the 16 RNA samples at four different, defined copy numbers) is shown in Fig. 2. The maximum CV of these values (0.69 for the low RESULTS copy number Trp transcript) was used as an estimate of the expected Sixteen low-grade and nonneoplastic astrocytic and oligodendro- technical variability associated with the study. glial samples were chosen for analysis (Fig. 1). This included 8 cases The hierarchical clustering relationship among all 16 specimens is of grade I PA, 2 cases of which were from patients with NF1. We displayed as a dendrogram in Fig. 1. Aligned with this dendrogram is chose normal human white matter as a source of brain tissue enriched information regarding pathological diagnosis, patient age at resection, in mature oligodendrocyte RNA. Cultured NHAs were selected as a gender, tumor location, and NF1 status. As might be expected, the model of committed astrocytic precursor cells. Three WHO grade II three oligodendrogliomas, three normal white matter specimens, and oligodendrogliomas were chosen to represent a comparable low-grade two cultures of normal human astrocytes each respectively clustered malignancy of presumed oligodendroglial precursor origin. All PA as highly related groups. The clustering relationship among the 16 tumors were carefully reviewed by an experienced neuropathologist samples was maintained regardless of the clustering method or metric (A. P.), and no obvious differences in tumor cellularity, tumor loca- used (data not shown). Furthermore, all eight PAs were closely related tion, patient age, or clinical behavior were observed. In addition, to each other and exhibited profiles of greater similarity to those of the patient 613 has no evidence of NF1 based on the NIH diagnostic normal human astrocyte cultures as compared with those of either criteria at the age of 13 years. normal white matter or low-grade oligodendrogliomas (Fig. 1A). As shown in Fig. 1, the number of genes scored as detected (“P”) Within the cluster of eight PAs, both tumors derived from NF1 2087

expression of connexin-43 was further verified at the protein level in selected tumors for which sufficient material was available for West- ern blot analysis (Fig. 3). The translation EF1-␣2 was overexpressed in the sporadic PA tumors and not the two NF1-associated astrocytomas. Other transcripts were only differentially expressed in the PA tumors relative to normal human astrocytes. These included N-CAM, L1-CAM, syndecan, ApoD, and SPARC. The overexpression of N- CAM was further confirmed at the protein level by Western blot analysis in selected tumors from which high quality protein lysates could be obtained (Fig. 3). To assist in determining whether these and other genes were of potential biological significance, we compared the expression profiles of our samples using oligonucleotide array technology with those identified by computational analysis of SAGE tags.6 Genes on the Affymetrix GeneChip were linked to Unigene ID numbers resulting Fig. 3. Western blot of N-CAM, merlin, and connexin-43 expression in PAs. Total protein (100 ␮g) from each sample was separated by SDS-PAGE and analyzed using from Xprofiler analysis of SAGE tags from a PA tumor (34-year-old specific antibodies as described in “Materials and Methods.” Tubulin was used as a male, Duke #H-1043) versus a culture of normal human astrocytes control for protein quality and loading (data not shown). The individual gene GeneChip (passage 5 NHA). As shown in Fig. 4, many additional genes differ- expression values for each tumor (from Fig. 2) are shown along the bottom of each gel. entially expressed by GeneChip were similarly represented in the proportion of SAGE tags identified in the Xprofiler output. The patients, and a third sporadic tumor (juvenile pilocytic astrocytoma expression of ApoD, ApoE, Sparc-like protein, IGFBP-2, and several 613) clustered closer to the normal astrocyte cultures than did the five other transcripts in PA tumors seen in this study was confirmed by other sporadic tumors (Fig. 1B). Further stratification revealed that SAGE analysis. Although the biological significance of these findings NF1-associated tumor #201 was more similar to normal human as- is still uncertain, the concordance of differential gene expression in trocyte cultures than any of the other primary tumors (Fig. 1C). these cell types by two different analytical methods suggest that these Histological review of this NF1-associated tumor did not reveal any differences are genuine. differences in cellular composition compared with the other PA tumors or the presence of contaminating nonneoplastic parenchyma to DISCUSSION account for its unique clustering profile. From the complete set of 9628 genes, we next examined a subset of PAs are low-grade astrocytic tumors that lack many of the genetic genes identified previously as being either overexpressed or lost in changes associated with their high-grade counterparts (16, 17). As higher grade astrocytomas or those associated with the astroglial these tumors rarely progress to WHO grade II astrocytomas, it has phenotype. Fig. 2 illustrates the expression patterns for these genes been suggested that they represent a distinct subtype of astrocytoma, (arranged by functional category) in each of the 16 samples. First, we perhaps arising from a different precursor cell than the fibrillary observed that many fibrillary astrocytoma-associated genes were nei- astrocytoma. In support of this notion, PAs express the PEN5 protein ther up- nor down-regulated in the PAs. This included the tumor (10) not expressed in type 1 astrocytes or high-grade astrocytomas but suppressor genes p53, retinoblastoma, PTEN/MMAC1, p19, p16, found in oligodendrocyte precursor cells and oligodendrogliomas. To mdm2, and cyclin-dependent kinase 4 expression. One exception was better characterize the specific genetic changes associated with PA merlin (the product of the NF2 gene), which was down-regulated in tumor formation, we initiated a gene expression profiling study on the PAs relative to normal human astrocyte cultures. This overall eight PAs. We proposed to determine whether the overall gene ex- pattern was confirmed on the protein level using merlin-specific pression profile (“molecular fingerprint”) and specific gene expres- antibodies on tumors from which high-quality protein lysates could be sion patterns of PAs most closely resembled astrocytic or oligoden- obtained (Fig. 3). droglial precursors. In addition, we wanted to identify specific Second, we evaluated the expression profiles for a number of gene transcripts associated with PAs. Lastly, we planned to determine transcripts expressed specifically in subsets of glial cells, such as whether specific gene expression profiles would distinguish NF1- oligodendrocytes, type 1 astrocytes, and type 2 astrocytes. To explore associated from sporadic PAs. the possibility that PAs express markers suggestive of an O2A cell, we Hierarchical cluster analysis clearly stratified the samples in our analyzed transcript expression of specific genes associated with O2A series into two subgroups. The first group separated the PA tumors cells. We observed increased expression of PMP-22, PLP, myelin and normal fetal astrocytes from the oligodendrogliomas and white oligodendrocyte glycoprotein, and MBP in many of the PAs. On the matter samples. This result was obtained using a variety of different basis of histological review of the frozen specimens, these increases in mathematical clustering strategies. From these results, the overall expression did not reflect the presence of contaminating nonneoplastic pattern of gene expression of the PAs most closely resembles the fetal parenchyma. astrocyte cells, rather than oligodendrocyte-rich white matter or well- Third, a number of genes demonstrated a relatively unique pattern differentiated oligodendroglial tumors. The second level of stratifica- of gene expression in the PA tumors relative to the various controls tion separated out the NHA cells, the two NF1-associated astrocyto- (NHA cells, oligodendrogliomas, and normal white matter). These mas, and one of the six sporadic PAs. The one female patient with the genes included GAP43, ␣-7 integrin, connexin 43, fatty acid binding sporadic PA at age 10.5 who clustered with the NF1-associated protein-7, cyclin D2, translation EF1-␣2, and lung type-1 cell mem- tumors did not meet diagnostic criteria for NF1 at age 13. Moreover, brane-associated gp36. Each of these genes was differentially ex- this tumor was in a typical location for a sporadic PA (posterior fossa). pressed in all of PAs relative to NHA cells and oligodendrogliomas, Therefore, it is unclear why this tumor more closely resembled the except ␣-7 integrin (increased in six of the eight PA tumors) and expression profile of the two NF1-associated tumors or whether this connexin-43 (decreased in five of the eight PA tumors). The decreased association is significant. Additional studies will be required to de- 2088

Fig. 4. Comparison of gene expression profiles in each of the 16 tumor samples as represented by GeneChip analysis in this study and transcript abundance in one PA and one NHA sample as determined by electronic SAGE analysis. Left, table of absolute gene expression levels as determined by GeneChip SADV metric. Center, pictorial representation of GeneChip expression values depicted in linear grayscale (highest expression in black). Right, analysis of the same genes in an electronic comparison (SAGE Xprofiler) of a single juvenile pilocytic astrocytoma and NHA sample. The number of SAGE tags identified in each of the samples is proportional to the transcript abundance in that sample. Only genes with concordant statistically ,ء .UniGene IDs and qualitatively similar differential expression patterns in both analyses are displayed. Each gene is accompanied by its GenBank accession number significant difference in GeneChip gene expression values (adjusted P Ͻ 0.05, t test) between PAs (n ϭ 8) and normal human astrocytes (n ϭ 2). See text for details concerning the statistical methods. termine whether this pattern of gene expression is predictive of a in lipid metabolism may be associated with astrocytic tumors. The different long-term clinical behavior compared with sporadic PAs finding of decreased merlin expression relative to NHA cells is lacking this gene expression profile. intriguing, given the role of this gene in regulating nervous system cell Many of the genetic alterations typical of high-grade astrocytomas proliferation. Although individuals with NF2 develop astrocytomas at were not observed in the PA tumors by GeneChip expression profil- an increased frequency, it is unlikely that merlin plays a significant ing. Specifically, we did not detect any changes in tumor suppressor role in astrocytoma formation, based on previous studies (26–29). gene expression (PTEN/MMAC1, p16, p19, p53, and retinoblastoma) We identified one gene, EF-1␣2, that was increased in all sporadic or metastasis-associated gene expression (metalloproteinase inhibi- PAs relative to the two NF1-associated tumors. Specific antibodies tors, vascular endothelial growth factor, and Tiam-1). These results were not available to confirm this observation at present. EF-1␣ has corroborate other studies examining PAs using protein- (17) and been implicated in malignant progression of metastatic rat mammary DNA-based strategies (13–16). Whereas one report demonstrated adenocarcinoma (30) and may determine the susceptibility to trans- increased platelet-derived growth factor-R␣ expression in sporadic formation (31) or apoptosis (32). The observation that EF-1␣ expres- PAs compared with normal brain, we could not confirm these results sion is lower in NF1-associated tumors that are typically less prone to in either this GeneChip analysis or in our previous studies on the clinical progression is intriguing. We are presently examining a larger protein level (17). Whether this increased expression in PAs reflects series of NF1-associated and sporadic PAs for differences in EF-1␣2 an O2A lineage (21) will require additional study. expression. In this study, we identified transcripts that were differentially Several genes that encode cytoskeleton/adhesion-associated mole- expressed in PAs relative to cultured NHAs, normal white matter, or cules demonstrated patterns of expression that differentiated PAs from oligodendrogliomas. These gene expression profiles will require in- NHA cells and oligodendroglioma tumors. These included GAP43, dependent confirmation on the RNA (e.g., quantitative PCR or North- connexin-43, ezrin, integrin ␣-7, and the lung type I-cell membrane- ern blot) and protein (e.g., immunohistochemistry or Western blot) associated gp36. GAP43 was increased in all of the PA tumors in this level on a larger series of PA tumors to validate our initial observa- series. GAP43 is a neuronal protein kinase C substrate expressed in tions. Despite these inherent limitations, our findings provide a start- cells of the O2A lineage that has been implicated in regulating actin ing point for future studies by identifying a number of specific genes cytoskeleton dynamics (33–35). We have found recently that GAP43 that warrant further investigation. expression is lost in high-grade astrocytomas, and its replacement in Cyclin D2 expression was increased in most of the PAs in our astrocytoma cell lines results in growth arrest (36). Connexin-43 series. Amplification of CCND2 has been reported previously in expression was significantly decreased in many of the PA tumors. In high-grade astrocytomas (22), suggesting that aberrant cell cycle support of a growth regulatory role for connexin-43, overexpression regulation may contribute to the pathogenesis of both high- and of connexin-43 results in decreased C6 cell proliferation (37). Ezrin low-grade astrocytomas. ApoD was increased in PA tumors in both expression was also found to be decreased moderately in PA tumors our study and in the SAGE analysis, as reported recently (23). The compared with normal white matter and NHA cells. This lower significance of this finding is not known but might relate to specific expression has been reported in grade II fibrillary astrocytomas but features of these low-grade tumors. We also observed increased appears to increase as the tumors become more malignant (38). expression of the brain fatty acid-binding protein, in the PA tumors. Integrin ␣-7 expression was increased in most of the PA tumors. It has Brain fatty acid-binding protein expression has been reported previ- been implicated in the regulation of cell migration through Crk ously in malignant gliomas (24, 25), suggesting that specific changes adapter molecule coupling to p130CAS (39). Lastly, lung type-I cell 2089

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 2002 American Association for Cancer Research. PILOCYTIC ASTROCYTOMA GENE EXPRESSION ANALYSIS membrane-associated glycoprotein (gp36 in humans and gp40 in will necessitate additional studies on the contribution of dysregulated rodents) is expressed at the apical plasma membrane (40) and has O2A lineage cell growth to the molecular pathogenesis of grade I PAs. been hypothesized to represent a marker of cellular differentiation in Lastly, because PAs exhibit clinical behavior distinct from other alveolar cells (41). Interestingly, gp36 is expressed in brain and neural astrocytic tumors, the preliminary insights provided by this study may derivative tissues beginning after embryonic day 13.5 in the rat but is serve to identify targets for PA-specific tumor therapies. down-regulated in the adult nervous system (42). Although a precise function for gp36 has not been proven, other similar proteins function to modulate cell adhesion. The identification of several genes in- ACKNOWLEDGMENTS volved in adhesion and cytoskeleton-associated processes, whose expression pattern is different from NHA cells and oligodendroglioma We thank Dr. William Shannon and Maciej Faifer for providing access to the Multitest tool for statistical analysis; the technical assistance of Sarah tumors, raises the possibility that specific changes in astroglial cell Kunz, Victoria Holtschlag, and Kate Hamilton; and the secretarial support of interactions may contribute to the pathogenesis of PAs. In this regard, Nancy North during the preparation of this work. we have recently found significant alterations in cell adhesion and motility in mouse astrocytes lacking Nf1 gene expression (43). Other gene expression changes were observed between PA tumors REFERENCES and NHA cells. These include N-CAM, L1-CAM, and syndecan. 1. Kleihues, P., and Cavanee, W. K. World Health Organization Classification of N-CAM is also expressed in a subpopulation of astrocytes in normal Tumours: Tumours of the Nervous System. Lyon, France: IARC Press, 2000. and glaucomatous human optic nerve (44). In tumors, expression of 2. Listernick, R., Louis, D. N., Packer, P. J., and Gutmann, D. H. Optic pathway gliomas N-CAM has been correlated inversely with astrocytic malignancy in children with neurofibromatosis 1: consensus statement from the NF1 optic pathway glioma task force. Ann. Neurol., 41: 143–149, 1997. grade (45). It has been implicated in intracellular signaling pathways 3. Ilgren, E. B., and Stiller, C. A. Cerebellar astrocytomas. Part II. Pathologic features involved in the negative regulation of astrocyte proliferation (46, 47). indicative of malignancy. Clin. Neuropathol., 6: 201–214, 1987. 4. Listernick, R., Charrow, J., and Gutmann, D. H. Intracranial gliomas in NF1. Am. J. Another related CAM (L1-CAM) was also increased in the PA tumors Med. Genet., 89: 38–44, 1999. in our series. L1-CAM has been reported to be overexpressed in 5. Raff, M. C., Miller, R. H., and Noble, M. A glial progenitor cell that develops in vitro malignant gliomas (48). The increase in syndecan expression in the into an astrocyte or an oligodendrocyte depending on culture medium. Nature (Lond.), 303: 390–396, 1983. PA tumors is intriguing given a recent report describing this trans- 6. Ffrench-Constant, C., and Raff, M. C. Proliferating bipotential glial progenitor cells membrane proteoglycan as a specific NF1 gene product (neurofibro- in adult rat optic nerve. Nature (Lond.), 319: 499–502, 1986. min) interacting protein (49). Because these particular gene expres- 7. Raff, M. C., Abney, E. R., and Miller, R. H. Two glial cell lineages diverge prenatally in rat optic nerve. Dev. Biol., 106: 53–60, 1984. sion changes were observed when PAs were compared with human 8. Landry, C. F., Verity, M. A., Cherman, L., Kashima, T., Black, K., Yates, A., and fetal astrocytes grown in vitro, additional experiments will be required Campagnoni, A. T. Expression of oligodendrocytic mRNAs in glial tumors: changes associated with tumor grade and extent of neoplastic infiltration. Cancer Res., 57: to exclude the possibility that these alterations reflect the differences 4098–4104, 1997. between tumors in situ and cultured astroglial cells. 9. Figols, J., Iglesias-Rozas, J. R., and Kazner, E. Myelin basic protein (MBP) in human PAs express some proteins that suggest an O2A lineage. On his- gliomas: a study of twenty-five cases. Clin. Neuropathol., 4: 116–120, 1985. 10. Figarella-Branger, D., Daniel, L., Andre, P., Guia, S., Renaud, W., Monti, G., Vivier, topathological examination, they typically exhibit a more astrocytic E., and Rougon, G. The PEN5 epitope identifies an oligodendrocyte precursor cell phenotype with long, hair-like cytoplasmic processes, which account population and pilocytic astrocytomas. Am. J. Pathol., 155: 1261–1269, 1999. for the name “pilocytic.” However, a significant subset appears re- 11. Gutmann, D. H., Donahoe, J., Brown, T., James, C. D., and Perry, A. Loss of neurofibromatosis 1 (NF1) gene expression in NF1-associated pilocytic astrocytomas. markably oligodendrocyte-like and is occasionally misdiagnosed as J. Neuropathol. Exp. Neurol., 26: 361–367, 2000. oligodendroglioma (1). Therefore, to determine whether PA tumors 12. Platten, M., Giordano, M. J., Dirvem, C. M. F., Gutmann, D. H., and Louis, D. N. Upregulation of specific NF1 gene transcripts in sporadic pilocytic astrocytomas. express transcripts typically associated with O2A cells, we examined Am. J. Pathol., 149: 621–627, 1996. oligodendrocyte-specific genes in our sporadic and NF1-associated 13. White, F. V., Anthony, D. C., Yunis, E. J., Tarbell, N. J., Scott, R. M., and Schofield, PAs. Although the overall expression pattern most closely resembled D. E. Nonrandom chromosomal gains in pilocytic astrocytomas of childhood. Hum. Pathol., 26: 979–986, 1995. fetal astrocytes, we observed the expression of oligodendrocyte-asso- 14. Zattara-Cannoni, H., Gambarelli, D., Lena, G., Dufour, H., Choux, M., Grisoli, F., ciated transcripts, such as PLP, PMP-22, MBP, and oligodendrocyte and Vagner-Capodano, A. M. Are juvenile pilocytic astrocytomas benign tumors? A myelin glycoprotein, suggesting that these tumors have at least some cytogenetic study in 24 cases. Cancer Genet. Cytogenet., 104: 157–160, 1998. 15. Sanoudou, D., Tingby, O., Ferguson-Smith, M. A., Collins, V. P., and Coleman, N. gene expression features of O2A cells. In contrast, increased expres- Analysis of pilocytic astrocytomas by comparative genomic hybridization. Br. J. sion of PLP and MBP has not been observed in high-grade astrocy- Cancer, 82: 1218–1222, 2000. 16. Cheng, Y., Pang, J. C. S., Ng, H-K., Ding, M., Zhang, S. F., Zheng, J., Liu, D. G., and tomas (50). The observation that the PEN5 marker of an oligoden- Poon, W. S. Pilocytic astrocytomas do not show most of the genetic changes drocyte precursor cell population (10) is expressed in PAs, but not commonly seen in diffuse astrocytomas. Histopathology, 37: 437–444, 2000. high-grade astrocytomas, also supports a separate lineage for PAs 17. Li, J., Perry, A., James, C. D., and Gutmann, D. H. Cancer-related gene expression profiles in neurofibromatosis 1 (NF1)-associated pilocytic astrocytomas. Neurology, compared with fibrillary astrocytomas. The expression of genes more 56: 885–890, 2001. typically associated with oligodendrocytes or O2A progenitor cells 18. Luzzi, V., Holtschlag, V., and Watson, M. A. Expression profiling of ductal carci- raises the possibility that PA tumors arise from precursor cells of an noma in situ by laser capture microdissection and high-density oligonucleotide arrays. Am. J. Pathol., 158: 2005–2010, 2001. intermediate differentiation stage. Future work will be required to 19. Eisen, M. B., Spellman, P. T., Brown, P. O., and Botstein, D. Cluster analysis and definitively establish the cell of origin relevant to the histogenesis of display of genome-wide expression patterns. Proc. Natl. Acad. Sci. USA, 95: 14863– 14868, 1998. PA tumors. 20. Gutmann, D. H., Giordano, M. J., Fishback, A. S., and Guha, A. Loss of merlin In summary, we identified transcripts that were uniquely expressed expression in sporadic meningiomas, ependymomas and schwannomas. Neurology, in PAs relative to oligodendrogliomas, NHA cells, or normal white 49: 267–270, 1997. ␣ 21. Raff, M. C., Lillien, L. E., Richardson, W. D., Burne, J. F., and Noble, M. D. matter. In addition, we found one potential gene (EF-1 2) that may Platelet-derived growth factor from astrocytes drives the clock that times oligoden- distinguish NF1-associated from sporadic PA tumors. Future studies drocyte development in culture. Nature (Lond.), 333: 562–565, 1988. on larger cohorts of PA tumors will be required to validate these initial 22. Buschges, R., Weber, R. G., Actor, B., Lichter, P., Collins, V. P., and Reifenberger, G. Amplification and expression of cyclin D genes (CCND1, CCND2, CCND3)in gene expression profiles and enable a more detailed functional anal- human malignant gliomas. Brain Pathol., 9: 435–442, 1999. ysis of these genes to define their respective roles in astroglial cell 23. Hunter, S., Young, A., Olson, J., Van Meir, E., Brat, D., Bower, G., Medrinos, S., and Neish, A. Expression microarray analysis: pilocytic astrocytomas and oligodendro- growth regulation and tumorigenesis. Similarly, the finding that PAs gliomas compared with anaplastic astrocytomas. J. Neuropathol. Exp. Neurol., 60: express molecular markers of oligodendrocytes and O2A progenitors 534, 2001. 2090

24. Godbout, R., Bisgrove, D., Shkolny, D., and Day, R. S. Correlation of B-FABP and 38. Geiger, K. D., Stoldt, P., Schlote, W., and Derouiche, A. Ezrin immunoreactivity is GFAP expression in malignant glioma. Oncogene, 16: 1955–1962, 1998. associated with increasing malignancy of astrocytic tumor but is absent in oligoden- 25. Young, J. K., Baker, J. H., and Muller, T. Immunoreactivity for brain-fatty acid drogliomas. Am. J. Pathol., 157: 1785–1793, 2000. binding protein in gomori-positive astrocytes. Glia, 16: 218–226, 1996. 39. Mielenz, D., Hapke, S., Poschl, E., von Der Mark, H., and von Der Mark, K. The 26. Hoang-Xuan, K., Merel, P., Vega, F., Hugot, J. P., Cornu, P., Delattre, J. Y., Poisson, integrin ␣ 7 cytoplasmic domain regulates cell migration, lamellipodia formation, and M., Thomas, G., and Delattre, O. Analysis of the NF2 tumor suppressor gene and of p130CAS/Crk coupling. J. Biol. Chem., 276: 13417–13426, 2001. chromosome 22 deletions in gliomas. Int. J. Cancer, 60: 478–481, 1995. 40. Zimmer, G., Lottspeich, F., Maisner, A., Klenk, H-D., and Herrler, G. Molecular 27. Ng, H. K., Lau, K. M., Tse, J. Y. M., Lo, K. W., Wong, J. H. C., Poon, W. S., and characterization of gp40, a mucin-type glycoprotein from the apical plasma mem- Huang, D. P. Combined molecular genetic studies of chromosome 22q and the brane of Madin-Darby canine kidney cells (type I). Biochem. J., 326: 99–108, 1997. neurofibromatosis type 2 gene in central nervous system tumors. Neurosurgery, 37: 41. Zimmer, G., Oeffner, F., von Messling, V., Tschernig, T., Grones, H-J., Klenk, H-D., 764–773, 1995. and Herrler, G. Cloning and characterization of gp36, a human mucin-type glyco- 28. Rubio, M-P., Correa, K. M., Ramesh, V., MacCollin, M. M., Jacoby, L. B., von protein preferentially expressed in vascular endothelium. Biochem. J., 341: 277–284, Deimling, A., Gusella, J. F., and Louis, D. N. Analysis of the neurofibromatosis 2 gene in human ependymomas and astrocytomas. Cancer Res., 54: 45–47, 1995. 1999. 29. Merel, P., Hoang-Xuan, K., Sanson, M., Moreau-Aubry, A., Bijsma, E. K., Lazaro, 42. Rishi, A. K., Joyce-Brady, M., Fisher, J., Dobbs, L. G., Floros, J., VanderSpek, J., C., Moisan, J. P., Resche, F., Nishisho, I., Estivill, X., Delattre, J. Y., Poisson, M., Brody, J. S., and Williams, M. C. Cloning, characterization, and developmental Theillet, C., Hulsebos, T., Delattre, O., and Thomas, G. Predominant occurrence of expression of a rat lung alveolar type I cell gene in embryonic endodermal and neural somatic mutations of the NF2 gene in meningiomas and schwannomas. Genes derivatives. Dev. Biol., 167: 294–306, 1995. Chromosomes Cancer, 13: 211–216, 1995. 43. Gutmann, D. H., Wu, Y. L., Hedrick, N. M., Zhu, Y., Guha, A., and Parada, L. F. 30. Edmonds, B. T., Wyckoff, J., Yeung, Y. G., Wang, Y., Stanley, E. R., Jones, J., Heterozygosity for the neurofibromatosis 1 (NF1) tumor suppressor results in abnor- Segall, J., and Condeelis, J. Elongation factor-1 ␣ is an overexpressed actin binding malities in cell attachment, spreading and motility in astrocytes. Hum. Mol. Genet., protein in metastatic rat mammary adenocarcinoma. J. Cell Sci., 109: 2705–2714, 10: 3009–3016, 2001. 1996. 44. Ricard, C. S., Pena, J. D., and Hernandez, M. R. Differential expression of neural cell 31. Tatsuka, M., Mitsui, H., Wada, M., Nagata, A., Nojima, H., and Okayama, H. adhesion molecule isoforms in normal and glaucomatous human optic nerve heads. Elongation factor-1-␣ gene determines susceptibility to transformation. Nature Brain Res. Mol. Brain Res., 74: 69–82, 1999. (Lond.), 359: 333–336, 1992. 45. Sasaki, H., Yoshia, K., Ikeda, E., Asou, H., Inaba, M., Otani, M., and Kawase, T. 32. Duttaroy, A., Bourbeau, D., Wang, X-L., and Wang, E. Apoptosis rate can be Expression of the neural cell adhesion molecule in astrocytic tumors: an inverse accelerated or decelerated by overexpression of reduction of the level of elongation correlation with malignancy. Cancer (Phila.), 82: 1921–1931, 1998. ␣ factor-1- . Exp. Cell Res., 238: 168–176, 1998. 46. Krushel, L. A., Sporns, O., Cunningham, B. A., Crossin, K. L., and Edelman, G. M. 33. Laux, T., Fukami, K., Thelen, M., Golub, T., Frey, D., and Caroni, P. GAP43, Neural cell adhesion molecule (N-CAM) inhibits astrocyte proliferation after injury to MARCKS, and CAP23 modulate PI(4,5)P2 at plasmalemmal rafts, and regulate cell different regions of the adult rat brain. Proc. Natl. Acad. Sci. USA, 92: 4323–4327, cortex actin dynamics through a common mechanism. J. Cell Biol., 149: 1455–1471, 1995. 2000. 47. Sporns, O., Edelman, G. M., and Crossin, K. L. The neural cell adhesion molecule 34. Caroni, P. Actin cytoskeleton regulation through modulation of PI(4,5)P2 rafts. (N-CAM) inhibits proliferation in primary cultures of rat astrocytes. Proc. Natl. Acad. EMBO J., 16: 4332–4336, 2001. 35. Frey, D., Laux, T., Xu, L., Schneider, C., and Caroni, P. Shared and unique roles of Sci. USA, 92: 542–546, 1995. CAP23 and GAP43 in actin regulation, neurite outgrowth, and anatomical plasticity. 48. Tsuzuki, T., Izumoto, S., Ohnishi, T., Hiraga, S., Arita, N., and Hayakawa, T. Neural ␤ J. Cell Biol., 149: 1443–1453, 2000. cell adhesion molecule L1 in gliomas: correlations with TGF- and p53. J. Clin. 36. Gutmann, D. H., Huang, Z-Y., Hedrick, N. M., Ding, H., Guha, A., and Watson, Pathol., 51: 13–17, 1998. M. A. Mouse glioma gene expression profiling identifies novel human glioma- 49. Hsueh, Y. P., Roberts, A. M., Volta, M., Sheng, M., and Roberts, R. G. Bipartite associated genes. Ann. Neurol., in press. interaction between neurofibromatosis type 1 protein (neurofibromin) and syndecan 37. Bradshaw, S. L., Naus, C. C., Zhu, D., Kidder, G. M., D’Ercole, A. J., and Han, V. K. transmembrane heparan sulfate proteoglycans. J. Neurosci., 21: 3764–3770, 2001. Alterations in the synthesis of insulin-like growth factor binding proteins and insulin- 50. Golfinos, J. G., Norman, S. A., Coons, S. W., Norman, R. A., Callecer, C., and like growth factors in rat C6 glioma cells transfected with a gap junction connexin43 Scheck, A. Expression of the genes encoding myelin basic protein and proteolipid cDNA. Regul. Pept., 48: 99–112, 1993. protein in human malignant gliomas. Clin. Cancer Res., 3: 799–804, 1997.

2091

Downloaded from cancerres.aacrjournals.org on September 24, 2021. © 2002 American Association for Cancer Research. Comparative Gene Expression Profile Analysis of Neurofibromatosis 1-associated and Sporadic Pilocytic Astrocytomas

David H. Gutmann, Nicolé M. Hedrick, Jun Li, et al.

Cancer Res 2002;62:2085-2091.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/62/7/2085

Cited articles This article cites 46 articles, 14 of which you can access for free at: http://cancerres.aacrjournals.org/content/62/7/2085.full#ref-list-1

Citing articles This article has been cited by 9 HighWire-hosted articles. Access the articles at: http://cancerres.aacrjournals.org/content/62/7/2085.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/62/7/2085. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.