[CANCER RESEARCH 63, 2933–2939, June 1, 2003] The 43,000 Growth-associated Protein Functions as a Negative Growth Regulator in Glioma 1 Zhi-yong Huang, YanLi Wu, Stephen P. Burke, and David H. Gutmann 2 Department of Neurology, Washington University School of Medicine, St. Louis, Missouri 63110 ABSTRACT Previous molecular analyses of human astrocytomas have identified many genetic changes associated with astrocytoma formation and pro- gression. In an effort to identify novel gene expression changes associated with astrocytoma formation, which might reveal new potential targets for glioma therapeutic drug design, we used the B8-RAS-transgenic mouse astrocytoma model. Using multiplex gene expression profiling, we found that growth-associated protein 43 (GAP43) RNA and protein expression were lost in select human and mouse glioma cell lines. In this study, we demonstrate that re-expression of GAP43 in deficient C6 glioma cells results in growth suppression in clonogenic assays, as well as in multiple independently derived C6 glioma cell lines in vitro. GAP43-expressing C6 cells also exhibit reduced tumor growth as s.c. explants in immunocom- promised mice in vivo. In addition, GAP43-expressing C6 clones demon- strate impaired cell motility and increased homophilic aggregation. GAP43 re-expression is also associated with reduced mitogen-activated protein kinase and AKT activation in C6 cells, suggesting that GAP43 functions as a novel glioma growth suppressor by modulating mitogenic signaling pathways. INTRODUCTION Astrocytomas (or gliomas) are one of the most devastating central nervous system tumors, with a median survival of 9 –12 months after diagnosis (1). Previous studies of the molecular genetic events asso- ciated with glioma formation and progression have identified a num- ber of specific genetic alterations, including amplification or mutation of the epidermal and platelet-derived growth factor receptors and loss of cell cycle growth regulators such as p53 and p16 (2–5). In addition to these known genetic changes, it is highly likely that there exist other alterations in gene expression associated with glioma pathogen- esis, which may yield novel insights into potential new targets for therapeutic drug design. In an effort to identify some of these previ- ously unidentified glioma-associated gene expression changes, we used gene expression profiling on a recently developed transgenic mouse model of glioma. One of the most common signaling abnormalities observed in gliomas is activation of the Ras mitogenic signaling pathway (6). This increased Ras pathway activation may occur as a result of aberrant receptor tyrosine kinase signaling. In this respect, constitutive activa- tion of the epidermal growth factor receptor leads to increased Ras activation and increased Ras pathway mitogenic signaling relevant to astrocytoma growth (2). Inhibition of Ras pathway activation with molecules that inhibit Ras posttranslational modification and proper membrane localization required for efficient effector molecule signal propagation results in marked reduction in glioma growth in vitro and in vivo (7). In addition, individuals with the NF1 3 inherited tumor syndrome develop astrocytomas at an increased frequency. The NF1 tumor suppressor protein, neurofibromin, contains a central domain with Ras-GTPase activating protein activity. Loss of NF1 expression results in impaired neurofibromin-mediated Ras inactivation and leads to increased Ras pathway activation and tumorigenesis (8). Because of these observations, we and our colleagues generated a transgenic mouse in which an oncogenic activated H-Ras molecule was specif- ically expressed in astrocytes (9). These Ras transgenic (B8) mice develop high-grade gliomas by 3– 4 months of age that are histolog- ically identical to their human counterparts. The availability of the B8 glioma transgenic mouse provided a unique opportunity to compare the gene expression profiles of wild-type astrocytes, nonneoplastic B8 astrocytes, and B8 astrocytoma cells. Using this approach, we were able to identify a number of novel gene expression changes that were specifically associated with the neoplastic state (10). We were able to validate some of these changes on the protein level in both mouse and human glioma cell lines, confirming the results obtained on the RNA level. One of these novel gene expression changes was loss of GAP43 expression in human and mouse glioma cell lines. GAP43 was originally identified as a neuronal protein important in cytoskeleton-associated processes relevant to nerve growth cone ex- tension and sprouting (11–15). GAP43 belongs to a family of proteins that include CAP23 and MARCKS, which have also been implicated in neuronal outgrowth (16). GAP43 expression and function has been studied in neurons (17–21), but little is known about its role in other cell types in the brain. Several reports have described the expression of GAP43 in type 1 astrocytes (22–24), as well as type 2 astrocyte precursors and mature type 2 astrocytes (25–27). Because astrocytes are not postmitotic cells like neurons and continue to divide in the adult brain, abnormal GAP43 function in astrocytes might have dif- ferent effects than previously reported for neurons. In this respect, expression of the GAP43 family member, MARCKS, is dramatically reduced in various cell lines after oncogenic or chemical transforma- tion (28 –33). In addition, re-expression of MARCKS in melanoma cell lines results in inhibition of both anchorage-dependent and anchorage-independent cell growth (34). These observations prompted us to examine the growth regulatory properties of GAP43 in gliomas. MATERIALS AND METHODS Constructs, Antibodies, and Cell Culture. The pcDNA3.chicken GAP43 plasmid was kindly supplied by Dr. Pico Caroni. The pcDNA3.mouse GAP43 plasmid was generated by reverse transcription-PCR using mouse brain RNA and mouse-specific primers 5'-TCTAGATCTGTGCTGTATGAGAA- GAACC-3' and antisense 5'-TCAGGCATGTTCTTGGTCAGC-3'. Anti- GAP43 polyclonal antibody was purchased from Chemicon (Temecula, CA). The anti-phospho MAPK (T202/Y204) antibody (9106S), anti-phospho AKT (Ser 473 ) antibody (9271S), and anti p44/42 MAPK antibody (9102) were purchased from Cell Signaling Technology (Beverly, MA), whereas anti-actin and anti--tubulin antibodies were purchased from Sigma (St. Louis, MO). All antibodies were used according to the manufacturer’s recommendations. Received 10/15/02; accepted 4/2/03. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by funding from NIH Grant NS41097 and American Cancer Society Grant RPG-00-231-01-CNE (to D. H. G.). 2 To whom requests for reprints should be addressed, at Department of Neurology, Washington University School of Medicine, Box 8111; 660 South Euclid Avenue, St. Louis, MO 63110. Phone: (314) 362-7379; Fax: (314) 362-2388; E-mail: gutmannd@ neuro.wustl.edu. 3 The abbreviations used are: NF1, neurofibromatosis 1; GAP43, growth-associated protein 43; MAPK, mitogen-activated protein kinase. 2933 Research. on November 26, 2021. © 2003 American Association for Cancer cancerres.aacrjournals.org Downloaded from