Alterations in the Rap1 Signaling Pathway Are Common in Human Gliomas
Total Page:16
File Type:pdf, Size:1020Kb
Oncogene (1997) 15, 1611 ± 1616 1997 Stockton Press All rights reserved 0950 ± 9232/97 $12.00 SHORT REPORT Alterations in the rap1 signaling pathway are common in human gliomas David H Gutmann1, Susan Saporito-Irwin1, Jerey E DeClue2, Ralf Wienecke2 and Abhijit Guha3 1Department of Neurology, Washington University School of Medicine, Box 8111, 660 South Euclid Avenue, St. Louis, Missouri 63110, USA; 2Laboratory of Cellular Oncology, Building 36, Room 1D-32, National Cancer Institute, Bethesda, Maryland 20892, USA; 3Lunenfeld Research Institute, Mount Sinai Hospital, Division of Neurosurgery, The Toronto Hospital, University of Toronto, 2-415 McLaughlin; 399 Bathurst, Toronto, Ontario M5T 2S8, Canada Several inherited predisposition to cancer syndromes are To this end, the TSC2 gene product, tuberin, functions associated with the development of nervous system as a negative regulator of rap1. Loss of tuberin tumors. Tuberous sclerosis complex (TSC) is an expression has recently been demonstrated by our autosomal dominant disorder in which aected indivi- laboratory in one third of sporadic gliomas, suggesting duals are at risk for developing astrocytomas. One of the that alterations in the rap1 signaling pathway may be genes responsible for this disorder is TSC2, located on associated with the development of human astrocyto- chromosome 16p, and encoding a 180 kDa protein mas (Wienecke et al., in press). In this study, we test (tuberin) that functions in part as a negative regulator the hypothesis that alterations in the rap1 signaling of rap1. Previous studies from our laboratory demon- pathway, either due to rap1 overexpression or reduced strated that 30% of sporadic astrocytomas have reduced levels of the rap1 negative regulator, tuberin, are or absent tuberin expression. In addition to loss of frequently associated with gliomas compared to other tuberin in sporadic astrocytomas, aberrant rap1 mediated tumors of the nervous system. signaling may also result from overexpression of rap1. In Tuberin has been demonstrated to suppress cell this study, we test the hypothesis that alterations in the growth and this activity resides in the C-terminus of rap1 signaling pathway are frequently observed in certain tuberin where the domain that functions as a GTPase subsets of gliomas compared to other tumors of the activating protein (GAP) for rap1 is located (Jin et al., nervous system. Analysis of sporadic astrocytomas and 1996). Biochemical studies demonstrated that this GAP ependymomas demonstrated either increased rap1 or activity is speci®c for rap1 and not other related small reduced/absent tuberin protein expression in 50 ± 60% of GTPase proteins, including rap2, H-ras and rho dierent cohorts of these gliomas, compared to 30 ± 33% (Wienecke et al., 1995). Loss of tuberin expression of sporadic schwannomas and meningiomas and none of would therefore be predicted to lead to elevated rap1 eight oligodendrocyte tumors. These results suggest that activity in these tumors. At present, there is no alterations in the rap1 signaling pathway are important biochemical method available to determine rap1 in the development of certain sporadic human gliomas. activity in cells to directly test this hypothesis. In addition to loss of tuberin expression, another Keywords: astrocytomas; tumor suppressor gene; mechanism for elevated rap1 signaling might result tuberous sclerosis complex 2; tuberin from increased rap1 expression. To this end, amplifica- tion of a segment of chromosome 12q containing the rap1B gene has been detected in many dierent tumor Studies on the molecular genetic events associated with types, including astrocytomas (Cheng et al., 1995; He the development of human malignant astrocytic tumors et al., 1994, 1995; Khatib et al., 1993; Reifenberger et (the most common subtype of glioma and primary al., 1996). In these tumors, there is ampli®cation of the human brain cancer) have demonstrated alterations in 12q chromosomal region variably containing the the p21-ras signaling pathway. Ampli®cation of the CDK4/SAS, MDM2, GLI, AM2R, GADD153, IFNG epidermal growth factor receptor (EGF-R) has been and RAP1B genes. detected in a high proportion of the most malignant To determine whether alterations in the rap1 astrocytoma or glioblastoma multiforme (GBM; Olson signaling pathway are observed in human astrocyto- et al., 1995), contributing to increased activation of mas, Western blot analysis was performed on twenty p21-ras. Previously, we demonstrated elevated levels of adult astrocytomas (Figure 1a). Tumor samples were p21-ras activity in astrocytoma cell lines as well as obtained from patients undergoing surgery in the fresh GBM surgical specimens (Gutmann et al., 1996; Division of Neurosurgery at the University of Toronto Guha et al., manuscript submitted). Inhibition of p21- or Washington University School of Medicine. All ras activity using either a dominant inhibitory p21-ras tumors were obtained in accordance with the respective mutant or farsenyltransferase inhibitors resulted in institutions' Institutional Review Board Human Studies reduced astrocytoma cell proliferation in vitro and Protocols and classi®ed according to the World Health tumor growth in vivo. Organization (WHO) classi®cation of tumors of the In addition to the p21-ras signaling pathway, central nervous system. For control purposes, normal activation of other small GTPase proteins may also cortex tissue was obtained from patients who underwent result in increased cell proliferation in astrocytomas. cerebral resection because of contusion. Mouse neocortical astrocyte cultures were generated as previously described (Hewett et al., 1995). The tumor Correspondence: DH Gutmann Received 14 March 1997; revised 28 May 1997; accepted samples were coded in order to conduct the experiments 28 May 1997 in a blind fashion. Brie¯y, specimens were minced and Tuberin and rap1 in nervous system tumors DH Gutmann et al 1612 lysed with the use of a Dounce homogenizer in NP40 lysis buer (50 mM Tris, pH 7.4; 150 mM NaCl; 0.5% NP-40, 1 mM DTT) and protease inhibitors while maintained on ice (Wienecke et al., in press). The a protein content was determined using a modi®ed * * * * Bradford reagent (BIORAD) and 75 micrograms of brain 317 352 379 397 410 mA 5295 52395 JPA 4295 WU47 each sample was subjected to 18% SDS polyacrylamide gel electrophoresis. The proteins were electroblotted — tuberin onto Immobilon-P membranes and the blots were incubated with the appropriate primary antibodies and developed with horseradish peroxidase-based enhanced chemiluminescence detection (ECL; Amersham). Reduced or absent tuberin expression was detected in — rap1 nine of twenty tumors relative to normal human brain or mouse astrocyte culture controls. For these studies, reduced or absent tuberin expression was de®ned as less * * than 10% of control tissue (brain or primary cultured mouse neocortical astrocytes) as determined by scan- * * * * * brain 794 805 806 808 813 820 827 834 858 859 ning densitometry. Equal loading was veri®ed using anti-actin polyclonal antibodies (data not shown; see — tuberin Figure 3b). Similar analysis of an additional 18 adult astrocytomas demonstrated reduced or absent tuberin expression in ®ve tumors. In total, reduced tuberin expression was observed in a total of 14 of 38 astrocytomas (37%). Increased rap1 expression was observed in ®ve of the 20 astrocytomas shown in Figure — rap1 1. In this study, increased rap1 expression was de®ned as greater than tenfold increased expression over control tissues (brain or primary astrocytes) as determined by scanning densitometry. The ®ve astro- * * * cytomas overexpressing rap1 had normal levels of tuberin. Similar analysis of an additional 18 adult astrocytomas demonstrated increased rap1 expression b in ®ve tumors, all of which had normal levels of tuberin brain 315 405 553 606 686 749 845 924 expression. In total, increased rap1 expression was observed in a total of 10 of 38 astrocytomas (26%). Therefore, 24 of 38 (63%) sporadic astrocytomas — tuberin demonstrated potential alterations in the rap1 signaling pathway. Consistent with the hypothesis that alterations in the rap1 signaling pathway result from either an increase in rap1 expression or a loss of tuberin expression, we observed overexpression of rap1 only in gliomas that maintained tuberin expression. There were no clinical dierences in terms of age, gender, — rap1 location of tumor and survival between this cohort of malignant astrocytomas compared to other astrocyto- mas, or between the group with decreased tuberin Figure 1 Tuberin and rap1 expression in sporadic human expression compared to those overexpressing rap1. astrocytic tumors. (a) Western blotting of protein lysates from 20 To determine if alterations in rap1 were unique to sporadic astrocytomas with the rabbit polyclonal tuberin C20 astrocytomas, we examined ependymomas and oligo- antibody (Santa Cruz Biotechnology; 1 : 1000 dilution) and mouse monoclonal rap1 antibody (Transduction Laboratories; 1 : 500 dendrogliomas (both less prevalent subtypes of human dilution) demonstrated reduced or absent tuberin expression in gliomas) as well as other non-glial nervous system nine tumors (317, 352, 5295, 4295, 813, 820, 834, 858 and 859) and tumors. In six spinal ependymomas, tuberin expression increased rap1 expression in an additional ®ve tumors (397, 410, was reduced in two tumors, while rap1 expression was 805, 806 and 827). Brain and m.A. correspond to human normal increased in an additional tumor, for a total