Medulloblastomas represent about 25% of all pediatric intracranial neoplasms. These highly malignant tumors arise from the cerebellum affecting mainly children between ages 5 and 15. Although the etiology of medulloblastomas has not yet been elucidated, several reports suggest that insulin-like growth factor I (IGF-I) may contribute to the development of these tumors. Results of this study show that the majority of cases examined were characterized by the abundant presence of the receptor for IGF-I (IGF-IR) protein (16 of 20 cases), and its major signaling molecule, insulin receptor substrate 1 (IRS-1; 15 of 20). Protein levels for IGF-IR and IRS-1, determined by Western blot and immunohistochemistry, were significantly higher in medulloblastoma biopsies than in control cerebellar tissue. By immunohistochemistry, 10 of 17 biopsies examined were also positive for the anti-pY1316 antibody staining that specifically recognizes the phosphorylated (active) form of the IGF-IR. These findings correlate with the fact that phosphorylated forms of the downstream-signaling molecules Erk-1, Erk-2, and Akt/protein kinase B were found in medulloblastoma biopsies but not in control cerebellar tissue. Importantly, there is a strong inverse correlation between biopsies that are positive for anti-pY1316 and for anti-Trk-C immunoreactivity. These observations direct our attention to the IGF-IR system as a potential therapeutic target in medulloblastomas and suggest a possibility of using the anti-pY1316 antibody as a potential prognostic marker for medulloblastomas.

The IGF-IR3 is a multifunctional membrane-associated tyrosine kinase that sends a mitogenic signal (1, 2), protects normal and tumor cells from apoptosis (3, 4), may induce cell differentiation (5), and plays an important role in the maintenance of the transformed phenotype by different viral and cellular oncogenes (6, 7). After ligand binding (to the α-subunit), the cytoplasmic portion of the receptor (the β-subunit) becomes phosphorylated, creating specific binding sites for direct IGF-IR substrates. Signaling molecules that bind activated IGF-IR include the family of IRS-1 proteins (8), Shc proteins (8), p85 subunit of PI3-K(9), GRB-10 (10), 14-3-3 (11), SH-containing phosphotyrosine phosphatase (12), and COOH-terminal Src kinase (13). Among multiple IGF-IR substrates, IRS-1 and PI3-K have been reported to play a role in IGF-IR-mediated cell proliferation and cell protection from apoptosis by recruiting downstream phosphorylation events such as the activation of the Ras/MAPK pathway and PI-3 kinase-mediated phosphorylation of Akt/PKB (14, 15, 16).

In the nervous system IGF-IR and its ligand(s) are present during embryonal and early postnatal development of the brain, decreasing substantially during adolescence and adult life (17, 18). In cerebellar cortex the IGF-IR is found in cells of the granular layer, where IGF-I is mitogenic and protects them from low potassium induced apoptosis (19, 20). Tumors of the nervous system secrete both IGF-I and IGF-II and often over-express IGF-IR (18). The presence of an active IGF system has been established in glioblastomas (21, 22, 23), neuroblastomas (24), astrocytomas and meningiomas (25). Despite data that point to the IGF system involvement in the cerebellum and in embryonic tumors of the peripheral nervous system (neuroblastomas), IGF-I and IGF-IR have been investigated minimally in medulloblastomas, the tumors that arise from the cerebellum and that are the most common embryonic tumors of the central nervous system. These few studies have demonstrated the presence of IGF-I binding sites in one medulloblastoma specimen (26), as well as the presence of IGF-IR protein and IGF-I mRNA in human medulloblastoma cell lines (27), and have shown overexpression of IGF-2 in medulloblastomas and rhabdomyosarcomas, which develop in mice with deletions of the PTCH gene (28). Finally, activation of the IGF-IR system has been demonstrated in human medulloblastoma cell lines, and in mouse medulloblastoma cells obtained from cerebellar tumors, which develop in transgenic mice carrying JC virus T-antigen and the JC virus archetype promoter (29).

It has been previously established that besides IGF-I, there are other factors that control growth and development of medulloblastomas. For instance, neurotrophins and their cognate receptors, Trk-A, Trk-B, and Trk-C, were shown to mediate proliferation, differentiation, and apoptosis of neuronal progenitor cells (30, 31). Importantly, Trk-C has been repeatedly shown to induce apoptosis in medulloblastoma cell lines (31, 32), and its elevated expression has been associated with favorable clinical outcome in medulloblastoma patients (32, 33).

To further explore the involvement of the IGF-IR system in medulloblastomas, we have examined 17 paraffin-embedded and 3 frozen cerebellar biopsies from patients with this tumor. Several parameters of the IGF-IR system were examined in this study: (a) the protein levels for IGF-IR, IRS-1, and PI3-K; (b) the cellular localization of IGF-IR and IRS-1; and (c) the phosphorylation status of the IGF-IR and its downstream signaling molecules, including Erk-1, Erk-2, and Akt/PKB. In addition, we made an attempt to correlate activation of the IGF-IR system with Trk-C expression in multiple biopsies isolated from medulloblastoma patients.

Clinical Samples and Cells.

A total of 17 human medulloblastoma samples were obtained from the pathology archives of Medical College of Pennsylvania-Hahnemann University, Philadelphia, PA. Formalin-fixed, paraffin-embedded surgical resections were histologically classified according to WHO classification of tumors of the nervous system (34). Three frozen medulloblastoma biopsies (p138, p494, and p609) were obtained from The Cooperative Human Tissue Network (CHTN). Finally, samples of frozen human cerebellum (N906, N989, and N1061) were kindly provided by Drs. Mary Herman and Joel E. Kleinman from the Neuropathology Section-Clinical Brain Disorder Branch, National Institute of Mental Health, Bethesda, MD.

Previously characterized R600 mouse embryo fibroblasts were used in Western blot analyses as a positive control to monitor the presence and activity of different components of the IGF-IR signaling pathways (2). Briefly, R600 cells were derived from R cells (35), which originated from mouse embryos with a targeted disruption of the IGF-IR gene (36). After stable transfection with the pMRIGFIR12 plasmid, which contains the hygromycin B phosphotransferase gene of Escherichia coli, and the human IGF-IR cDNA under control of the rat (−2350 + 1640) IGF-IR promoter (2), clones with different numbers of the IGF-IR were selected (35). By Scatchard analysis, clone R600 was shown to express 3 × 104 IGF-IR molecules per cell (35). To determine the presence of the phosphorylated forms of Akt/PKB and MAPKs, quiescent R600 cells were stimulated with IGF-I (50 ng/ml), and total proteins were extracted 15 min after stimulation.

Western Blots.

To determine total IGF-IR, IRS-1, and PI3-K protein levels, monolayer cultures of R600 cells or small fragments of cerebellar tissues were lysed on ice with 400 μl of lysis buffer [50 mm HEPES (pH 7.5), 150 mm NaCl, 1.5 mm MgCl2, 1 mm EGTA, 10% glycerol, 1% Triton X-100, 1 mm phenylmethylsulfonyl fluoride, 0.2 mm sodium orthovanadate, and 10 μg/ml aprotinin]. Protein concentrations were determined by the Bio-Rad Protein Assay (Bio-Rad, Hercules, CA), and 50 μg of total proteins were separated on a 4–15% gradient SDS-PAGE (Bio-Rad) and transferred into nitrocellulose membranes. Blots were blocked with 5% nonfat dry milk in TBST [10 mm Tris-HCl (pH 7.5), 150 mm NaCl, and 0.1% Tween 20] and were probed with rabbit anti-IGF-IRβ antibody (Santa Cruz Inc., Santa Cruz, CA), rabbit anti-IRS-1 antibody (Upstate Biotechnology Inc., Lake Placid, NY), rabbit anti-PI3-K (p85) antibody (Upstate Biotechnology), and antirabbit horseradish peroxidase-conjugated secondary antibody (Calbiochem, San Diego, CA). Blots were developed with ECL detection reagents (Amersham, Arlington, IL). Phosphorylated forms of Akt/PKB were detected with the PhosphoPlus Akt(Ser473) antibody kit (New England BioLabs, Beverly, MA), and phosphorylated Erk-1 and Erk-2 were detected by PhosphoPlus Phospho-p44/42 MAPK (Thr202/Tyr204) antibody kit (New England BioLabs, Beverly, MA). To determine equal loading conditions, corresponding blots were “stripped” and reprobed with anti-Grb-2 antibody (Transduction Laboratories, Lexington, KY). Finally, to evaluate IGF-IR tyrosine phosphorylation, frozen tissues were lysed on ice with 400 μl of lysis buffer, and protein concentration was determined by a Bio-Rad Protein Assay (Bio-Rad, Hercules, CA). A 500 μg of total protein extract was used for immunoprecipitation the anti-αIGF-IR antibody (Ab-1, Calbiochem) in the presence of agarose conjugated mouse IgG (Sigma). The immunoprecipitated proteins were resolved on 4–15% gradient polyacrylamide gels containing SDS and electroblotted onto nitrocellulose filters. Membranes were blocked with 5% BSA in TBST buffer overnight at 4°C. Blots were probed with the mouse anti-phosphotyrosine horseradish peroxidase-conjugated antibody (PY20; Transduction Laboratories) and developed with ECL reagents (Amersham).

Immunohistochemistry.

Formalin-fixed, paraffin-embedded samples were sectioned at 4-μm thickness and stained with H&E for routine examination and the histological classification of the tumors. Immunohistochemistry was performed using the avidin-biotin-peroxidase complex system, according to the manufacturer’s instructions (Vectastain Elite ABC Peroxidase Kit; Vector Laboratories). Briefly, sections were deparaffinized in xylene and rehydrated through descending alcohols to water. For nonenzymatic antigen retrieval, sections were heated in 0.01 m sodium citrate buffer (pH 6.0) to 95°C for 40 min and allowed to cool for 20 min at room temperature. To quench endogenous peroxidase, slides were then rinsed with PBS and incubated in methanol/3% H2O2 for 20 min. Sections were then washed with PBS and blocked in PBS/0.1% BSA containing 5% normal horse or goat serum for 2 h at room temperature, and incubated overnight at room temperature with primary antibodies. The primary antibodies used were rabbit anti-pY1316 IGF-IR (1:500 dilution; Ref. 36); rabbit anti-IGF-IR (1:500 dilution; Santa Cruz Inc.); rabbit anti-IRS-1 (1:100 dilution; Upstate Biotechnology Inc. ); mouse anti-GFAP (1:100 dilution, clone b2F; DACO); mouse antineurofilaments antibody (1:500 dilution, clone SMI-312; Sternberger Monoclonals); and mouse anti-synaptophysin antibody (1:500 dilution, clone SY38; Roche Molecular, Indianapolis, IN). Finally, rabbit anti-Trk-C antibody (sc-117, 1:500 dilution; Santa Cruz Inc.), which specifically recognizes the COOH terminus of the Trk-C and does not cross-react with Trk-A and Trk-B, was used as well.

Secondary antibodies were biotin-conjugated horse antimouse and goat antirabbit IgGs (Vector). After secondary antibodies, avidin-biotin peroxidase complexes (Vector) were incubated for 1 h at room temperature, sections were developed with a DAB substrate (Sigma), counterstained with hematoxylin, and mounted with Permont.

Localization and Activity of the IGF-IR in Biopsies from Patients with Medulloblastoma.

A recently developed and characterized antibody against the phosphorylated (active) form of the IGF-IR was used in these immunohistochemical experiments (37). The antibody specifically recognizes phosphotyrosine 1316 (pY1316) of the β subunit of IGF-IR, and does not cross-react with the corresponding moiety of the activated insulin receptor (37). To determine the status of IGF-IR activity in human medulloblastomas, paraffin-embedded cerebellar sections taken from 17 biopsies of patients diagnosed with medulloblastoma were stained with the anti-pY1316 antibody. Ten (59%) of the total 17 cases, demonstrated neoplastic cells with tyrosine-phosphorylated IGF-IR (Table 1). In addition, 76% and 82% of these biopsies showed strong immunoreactivity for anti-IGF-IRα and anti-IRS-1 antibodies, respectively (Table 1). Importantly, all anti-pY1316-positive sections were characterized by the presence of the IGF-IR and IRS-1 proteins, and sections that lacked either IRS-1 or IGF-IR staining were negative for anti-pY1316 immunoreactivity. Other criteria considered with respect to the anti-pY1316 antibody staining included: (a) WHO histological subtype of medulloblastoma; (b) gender and age of the patient; (c) location of the tumor; (d) presence of neurofilaments and synaptophysin and absence of GFAP immunostaining; and, finally, (e) presence or absence of the Trk-C in the neoplastic cells. In general, results of these multiple comparisons did not show a clear association with the anti-pY1316 antibody staining. The only exception was found with Trk-C-positive human medulloblastoma samples, in which a strong inverse correlation with anti-pY1316 immunostaining was detected. As illustrated in Table 1, all of the strongly positive Trk-C samples (biopsies 1, 5, 10, 11, and 12) were negative for anti-pY1316 antibody staining. In agreement with this finding, all of the strongly positive pY1316 samples (biopsies 2, 3, 15, and 16) were negative for the anti-Trk-C antibody staining. There are two cases however, biopsies 6 and 8, in which a weak anti-pY1316 immunostaining was accompanied by the presence of a weak Trk-C protein detection.

Fig. 1 demonstrates immunohistochemical localization of the phosphorylated and total IGF-IR in two representative medulloblastoma biopsies (biopsy 3, A and B; and biopsy 4, C and D). Groups of tumor cells strongly positive for the phosphorylated IGF-IR, detected with anti-pY1316, are shown in Fig. 1, A and C. The immunoproduct appears to be cytoplasmic and membrane associated because DAB granules (brown) are visible not only in perinuclear regions (arrowheads) but also above the hematoxylin-counterstained nuclei (arrows). Control cerebellar tissue does not stain with this antibody (Fig. 1,E). Control cells treated with secondary antibody in the presence of an irrelevant (anti-bromodeoxyuridine) antibody instead of anti-Y1316, were negative as well (data not shown). Consecutive sections were also probed with the antibody that recognizes total IGF-IR (Fig. 1, B and D). In contrast to the phosphorylated form of the IGF-IR, which was detected only in certain regions within the tumors, the majority of cells within these sections were uniformly stained. Sections of normal human cerebellum were also positive for the total IGF-IR (Fig. 1,F), mostly detected in Purkinje cells (arrows) and were completely negative for anti-pY1316 antibody staining (Fig. 1,E). To ensure the specificity of the Y1613 antibody staining we introduced an additional control in which a strong anti-Y1316 antibody staining was detected in BsB8 medulloblastoma cell line detected within 15 min after IGF-I stimulation (Fig. 1,G). Lack of the staining in parallel cultures that were left without IGF-I stimulation confirmed again specific recognition of the phosphorylated form of the IGF-IR by this antibody (Fig. 1 H).

Fig. 2 shows representative immunohistochemical staining with anti-pY1316 (A and C) and anti-Trk-C (B and D) antibodies. Biopsy 2, which has a strong anti-pY1316 immunoreactivity, is completely negative for anti-Trk-C antibody staining (Fig. 2; compare C and D). Conversely, biopsy 1, strongly positive for the presence of Trk-C (see Table 1) is negative for anti-pY1316 antibody staining (Fig. 2; compare A and B). Although this inverse correlation between phosphorylation of the IGF-IR and the Trk-C expression is not absolute, compare, for instance, biopsies 6 and 8 (Table 1), an opposite immunostaining for these two receptor proteins was detected in the majority cases examined in this study (15 of total 17 cases).

Protein Levels for IGF-IR, IRS-1, and PI3-K in Medulloblastoma Biopsies.

To analyze protein levels for IGF-IR, PI3-K and IRS-1 in a semiquantitative fashion, Western blots were performed on frozen biopsy samples from three patients with classic medulloblastoma (p138, p609, and p494). Thick frozen sections of human normal cerebellum collected from autopsies of an 11-week-old (N906), a 26-week-old (N1061), and a 36-year-old (N989) were used as reference samples. Fig. 3,A (top) demonstrates elevated levels of the IGF-IR protein in three medulloblastoma frozen biopsies. By densitometry, the IGF-IR from biopsies showed, on average, a 4-fold increase in comparison with three samples isolated from human normal cerebellum. Additionally, IGF-IR protein levels in medulloblastoma biopsies were compared with IGF-IR protein levels detected in R600 fibroblasts, previously shown by Scatchard analysis to express 3 × 104 IGF-IR molecules/cell (35). Again densitometry allowed us to estimate the quantity of IGF-IR protein in normal and neoplastic cerebellum. These semiquantitative determinations showed that the population of cells from the biopsies p138, p609, and p494 expressed an average of 1.3 × 104; 2.6 × 104; and 2.1 × 104 IGF-IR/cell, respectively. In comparison, cells from normal human cerebellum, N906 and N1061, expressed an average of 0.6 × 104 and 0.8 × 104 IGF-IR/cell, respectively. It is noteworthy that IGF-IR protein level in cerebellar tissue from the adult (N989) is significantly lower (0.14 × 104 IGF-IR/cell) than the IGF-IR level detected in two cerebellar autopsies from the newborns (N906 and N1061). Importantly, the IGF-IR protein, abundantly present in all of the medulloblastoma biopsies, was tyrosine phosphorylated as indicated (Fig. 3,B) by immunoprecipitating the IGF-IR and developing corresponding blot with anti-phosphotyrosine antibody (pY). Because of restricted sensitivity of the Western blot in comparison with immunohistochemistry, detection of the phosphorylated band was possible when the IGF-IR protein was immunoprecipitated from 500 μg of total proteins. Control samples from normal cerebellar tissue did not show any traces of the IGF-IR phosphorylation (Fig. 3,B). These results confirmed our immunohistochemical evaluation of the IGF-IR tyrosine phosphorylation depicted in Fig. 1, by using a different antibody and different biopsy samples (Fig. 3 B).

Western blot for PI3-K is shown in the Fig. 3 A, middle panel. In comparison with IGF-IR detected in the same set of protein samples, PI3-K has a different pattern of expression. It appears to be abundant in both normal cerebellar tissue and medulloblastoma biopsies. One could even argue that in three medulloblastoma biopsies examined, PI3-K protein levels were slightly lower than in control cerebellar tissues. Interestingly, an additional slightly lower band was detected in all three of the samples from control cerebellum. Although we did not attempt to investigate the nature of this lower band in this study, it may reflect the presence of hypophosphorylated p85 in normal cerebellar tissue, and its disappearance in actively growing medulloblastomas.

Fig. 3,A, bottom panel, illustrates IRS-1 protein levels determined in the samples previously analyzed for the IGF-IR and PI3-K. The IRS-1 protein was detected in all of themedulloblastoma samples within seconds after exposing ECL-treated blots to photographic material. In contrast, control samples obtained from normal cerebellar tissues did not show any detectable traces of IRS-1 signal at these short exposure times. Overexposure of the film for 15 min resulted in visualization of weak bands at a Mr range of 160,000 in all of the control samples (Fig. 3,C). This may reflect a low level of IRS-1 expression in normal human cerebellum. Because reactivation of the IRS-1 expression in combination with an apparent IGF-IR activation in medulloblastomas may potentially affect aggressiveness of the tumor cells, it was critical to confirm these findings by using different medulloblastoma samples and a different methodological approach. In this respect, Fig. 4 illustrates IRS-1 immunostaining of paraffin-embedded biopsies obtained from two additional patients with a diagnosed classic medulloblastoma. As expected, the IRS-1 staining appears to be cytoplasmic because DAB-positive granules (light brown) are visible in perinuclear regions of the cells. Control sections, treated with secondary antibody in the presence of an irrelevant antibody, were negative (data not shown). In comparison, a faint anti-IRS-1 immunostaining of the granular layer (arrows) and Purkinje cells (arrowheads) was observed in normal cerebellum (Fig. 4,A), whereas the medulloblastoma samples were characterized by much stronger cytoplasmic staining (Fig. 4, B and C). This result supports our previous date from the Western blot (Fig. 3 A), confirming apparent up-regulation of the IRS-1 protein in medulloblastomas.

Activity of the Downstream Signaling Pathways.

Activation of two downstream signaling pathways, Ras/MAPK and the PI3-K-mediated phosphorylation of Akt/PKB, is thought to facilitate IGF-IR-mediated cell proliferation and cell protection from apoptotic death (5, 27, 38, 39). We first compared levels of Erk1 and Erk2 phosphorylation between medulloblastoma biopsies and normal cerebellar tissue (Fig. 5, top two panels). IGF-I-stimulated R600 fibroblasts were used as a positive control for Erk1/Erk2 phosphorylation (2). Two medulloblastoma biopsies (p138 and p494) were characterized by Erk1/Erk2 phosphorylation on threonine 202 and tyrosine 204, an event that triggers MAPK activity (40). In contrast, the third biopsy (p609), and all of the control samples (N906, N989, and N1061) were negative for the anti-phospho-Erk-1/Erk-2 immunostaining.

The Western blot depicted in Fig. 5, third panel, is the same one shown in the top two panels after reprobing it with the anti-phospho-Akt antibody. This was done to monitor Akt phosphorylation in the same experimental samples as described for MAPKs. Two samples that demonstrated activated MAPKs (p138 and p494) were also characterized by Akt/PKB phosphorylation. However, an apparent difference in the phosphorylation pattern was observed in biopsy p609. In this particular sample Akt phosphorylation was apparently stronger than in the other two samples (p138 and p494) and appeared to be comparable with the level of Akt phosphorylation in control R600 fibroblasts stimulated in vitro with IGF-I. Interestingly, the medulloblastoma sample strongly positive for the Akt phosphorylation (p609) was negative for the Erk-1/Erk-2 phosphorylation. Observed changes in Erks and Akt phosphorylation do not reflect alterations on the level of gene expression, because total protein levels for Erk-1, Erk-2, and Akt were very similar (Fig. 5, second and fourth panels). In addition, to monitoring protein levels independently from MAPKs and Akt, the same blot was additionally reprobed with the anti-Grb-2 antibody (41). This confirms that, despite of the presence of total Erks, p85, and Akt in all of the samples examined, only actively growing medulloblastoma cells are characterized by the presence of phosphorylated forms of the signaling molecules, which are likely activated by the IGF-IR signaling system in these tumors.

Despite the data that point to the involvement of the IGF system in the cerebellum (19, 20, 27) and in embryonic tumors in the peripheral nervous system, neuroblastomas (24), the IGF-I, and the IGF-IR have been investigated minimally in medulloblastomas, which arise from the cerebellum and are the most common embryonic tumors of the central nervous system. These studies include detection of the IGF-I-binding sites in one medulloblastoma specimen (26), and the demonstration of the presence of IGF-IR protein by Western blot and the presence of IGF-I transcripts by RT-PCR (27). We have recently demonstrated the up-regulation of the IGF-IR system in human and mouse medulloblastoma cell lines, as well as the phosphorylation of the IGF-IR in several medulloblastoma biopsies (29). These preliminary studies, together with recent reports by Patti et al.(27) emphasize the potential involvement of the IGF-IR system in the development and progression of medulloblastomas. To further analyze the IGF-IR system in medulloblastomas, we have examined 17 paraffin-embedded and 3 frozen biopsies from patients with medulloblastomas.

In agreement with previous results obtained from medulloblastoma cell lines (27, 29), both IGF-IR and its major substrate IRS-1 were significantly up-regulated in medulloblastoma biopsies. Importantly, the phosphorylated (active) form of the IGF-IR was detected in 59% of all medulloblastoma cases examined. Because not all medulloblastoma samples demonstrated positive staining with the anti-pY1316 antibody, it was relevant to ask whether the presence of the phosphorylated form of the IGF-IR correlates with some of the parameters commonly used for the clinical characterization of medulloblastomas. We were initially able to demonstrate that all of the pY1316-positive samples were also positive for the IGF-IR and IRS-1 immunostaining. Importantly, all IGF-IR-negative samples were also negative for anti-pY1316 antibody staining, further confirming the specificity of the anti-pY1316 antibody. In the next set of comparisons, we expected to find phosphorylated IGF-IR in more aggressive classic and/or neuroblastic medulloblastomas rather than in the less aggressive desmoplastic medulloblastomas. However, the data shown in Table 1 indicate the absence of such a correlation. This was somewhat surprising because activation of the IGF-IR mediates both cell proliferation (1, 2) and cell motility (42, 43, 44), parameters that could potentially increase invasiveness of the tumors. On the other hand, elevated IRS-1 expression, although favorable for the IGF-I-dependent primary tumor growth (45), has been conversely related to metastatic potential of breast cancer (46), and prostate cancer (43, 44) cell lines. In this respect, overexpression of IRS-1 in the metastatic prostate cancer cell line, LNCaP, markedly attenuated cell motility and increased cell aggregation and adhesion to extracellular matrix proteins (43, 44). Therefore, it is possible that, in medulloblastomas, activation of the IGF-IR, together with strong up-regulation of the IRS-1, is associated with primary tumor growth rather than with tumor invasiveness (ability to escape from the primary site into the subarachnoid space and infiltration of medulloblastoma cells into healthy cerebellar tissues). Other comparisons made with respect to anti-pY1316 antibody staining included: gender and age of the patient; location of the tumor; presence of cellular markers such as neurofilaments and synaptophysin, and absence of the GFAP. These multiple comparisons did not show any significant association with the IGF-IR activity in medulloblastoma samples. In contrast, a strong inverse correlation was noticed between the presence of the phosphorylated IGF-IR and Trk-C expression (last two columns in Table 1). Elevated Trk-C levels were already proposed as a molecular marker associated with good prognosis in medulloblastomas (32, 33). Results in this study (Table 1 and Fig. 2) suggest that the presence of the activated form of the IGF-IR may also carry some diagnostic values; however, more studies are required to correlate low levels of the Trk-C expression, activation of the IGF-IR phosphorylation, and the clinical outcome in medulloblastoma patients.

Fig. 1, A and B, show two different examples of medulloblastoma biopsies. In these samples, there are groups of neoplastic cells strongly positive for the anti-pY1316 antibody staining. Cytoplasmic location of the staining may suggest that the activated (Y1316-positive) receptor was internalized at the time the biopsy was collected. Sections from the same samples were also probed with the antibody that recognizes total IGF-IR (Fig. 1, D and E). Cells within these sections were uniformly stained, suggesting the presence of the IGF-IR protein in the majority of neoplastic cells in the affected region of the cerebellum. On the other hand, the phosphorylated form of the IGF-IR was detected only in certain areas within the tumors and only some of the cells showed strong anti-pY1316 reactivity (Fig. 1, A and B). This irregular pattern of IGF-IR activation may suggest that, only in certain discrete regions within the cerebellar tumor IGF-IR, ligand(s) are secreted and activate IGF-IR-dependent cellular processes that may possibly contribute to the malignant growth within the neoplasm.

The results presented in Fig. 3 show apparent increases in the IGF-IR and IRS-1 protein levels in medulloblastoma samples. These up-regulations were evaluated in biopsy samples obtained at surgery, in relation to control samples isolated from frozen autopsies of normal cerebellar tissue. The biopsy samples were frozen within minutes after their surgical excision, whereas the normal cerebellar tissues were obtained between 13.5 to 36 h postmortem. Therefore, it is possible that the proteins isolated from the control sections are not well preserved. Microscopic examination of the sections, and the abundant presence of the PI3-K band (Mr 85,000), Grb-2, Erk1, and Erk2, detected in all of the protein samples, including normal cerebellar tissues (Figs. 3 and 5), partially excludes this possibility. Additionally, we have shown, in our previous work, blots containing proteins from the postmortem cerebellum in which both cytoplasmic and membrane-associated proteins were well preserved (29).

Ligand-activated IGF-IR leads to the recruitment and the activation of the Ras/MAPK pathway via the IRS-1 and/or Shc proteins (16, 47). In parallel, PI3-K can also be activated via a direct interaction with the β-subunit of the IGF-IR or by an interaction with tyrosine-phosphorylated IRS-1 (16). Once PI3-K is activated, it mediates phosphorylation, and activates Akt/PKB, an event repeatedly shown in many different systems to protect cells from apoptosis (39). Fig. 5, top, illustrates the phosphorylation status of Erk-1 and Erk-2. Although both IGF-IR and IRS-1 were strongly up-regulated in all three of the medulloblastoma samples examined (Fig. 3,A), biopsy p609 did not exhibit a detectable level of phosphorylated Erk-1 and Erk-2 (Fig. 5, top panel). Interestingly, when the same blot was reprobed with the antibody that recognizes the phosphorylated Akt/PKB, biopsy p609 was characterized by the strongest Akt/PKB phosphorylation (Fig. 5, middle panel). The explanation for this peculiar phosphorylation pattern may come from several recent reports showing the inhibition of the Ras/MAPK pathway in cells with constitutively active Akt/PKB (48). Although the question as to whether medulloblastoma cells are able to proliferate without activating Ras/MAPK when Akt/PKB is constitutively active remains to be addressed, our present results indicate activation of different signal transduction pathways that may originate from the auto and/or paracrine IGF-IR system in medulloblastomas.

In summary, we have detected the IGF-IR and its major signal transduction substrates, IRS-1 and PI3-K, in the majority of paraffin-embedded and frozen sections from patients with medulloblastoma. Protein levels and the intensity of immunostaining for IGF-IR and IRS-1 show strong up-regulation in medulloblastoma biopsies when compared with control, age-appropriate, normal cerebellar tissue. Importantly, 10 of 17 paraffin-embedded biopsies from patients diagnosed with medulloblastoma demonstrated the presence of the phosphorylated form of IGF-IR (pY1316) and subsequent phosphorylation of downstream IGF-IR signaling molecules, Erk-1, Erk-2, and Akt/PKB. Finally, an inverse pattern for anti-pY1316 and anti-Trk-C immunostaining was demonstrated in 15 of 17 medulloblastoma cases examined. These results strongly suggest the involvement of the IGF-IR system in the development of medulloblastomas and direct our attention to the IGF-IR molecule as a potential therapeutic target in these cerebellar tumors.

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

Supported by NIH Grant PO1 NS 36466.

3

The abbreviations used are: IGF-I, insulin-like growth factor I; IGF-IR, receptor for IGF-I; IRS-1, insulin receptor substrate 1; Akt/PKB, Akt/protein kinase B; MAP, mitogen-activated protein; PI3-K, phosphatidylinositol 3′-kinase; GRB-10, growth factor receptor-bound protein-10; MAPK, MAP kinase; GFAP, glial fibrillary acidic protein kinase; DAB, 3,3′-diaminobenzidine; Erk, extracellular signal-regulated protein kinase; Trk, tyrosine receptor kinase.

Fig. 1.

Immunohistochemical analysis of IGF-IR in medulloblastomas. A and C, representative paraffin-embedded sections of two medulloblastomas treated with the anti-pY1346 antibody that specifically recognizes the phosphorylated (active) form of the IGF-IR. B and D, consecutive sections incubated with anti-IGF-IR antibody that recognizes both phosphorylated and nonphosphorylated IGF-IR. E, a section from normal human cerebellum treated with anti-pY1316 antibody. Cells that are positive for anti-pY1316 antibody staining are exclusively present in discrete areas within the tumor, and are not detectable in sections from the normal human cerebellum. F, normal human cerebellum stained with anti-IGF-IR antibody. Almost all cells in medulloblastoma biopsies are stained with the antibody against total IGF-IR, whereas sections of normal human cerebellum are weakly positive at best, most often in Purkinje cells (arrows). G and H, immunocytochemical staining with anti-Y1316 antibody of BsB8 mouse medulloblastoma cell line in monolayer culture. Cells were made quiescent by culturing them in serum-free medium for 48 h, and were subsequently stimulated with IGF-I (50 ng/ml) for 15 min (G) or were left without stimulation (H). There is an apparent membrane association and cytoplasmic immunoreactivity in cells stimulated with IGF-I. A–D, G, and H, ×40; E and F, ×20.

Fig. 1.

Immunohistochemical analysis of IGF-IR in medulloblastomas. A and C, representative paraffin-embedded sections of two medulloblastomas treated with the anti-pY1346 antibody that specifically recognizes the phosphorylated (active) form of the IGF-IR. B and D, consecutive sections incubated with anti-IGF-IR antibody that recognizes both phosphorylated and nonphosphorylated IGF-IR. E, a section from normal human cerebellum treated with anti-pY1316 antibody. Cells that are positive for anti-pY1316 antibody staining are exclusively present in discrete areas within the tumor, and are not detectable in sections from the normal human cerebellum. F, normal human cerebellum stained with anti-IGF-IR antibody. Almost all cells in medulloblastoma biopsies are stained with the antibody against total IGF-IR, whereas sections of normal human cerebellum are weakly positive at best, most often in Purkinje cells (arrows). G and H, immunocytochemical staining with anti-Y1316 antibody of BsB8 mouse medulloblastoma cell line in monolayer culture. Cells were made quiescent by culturing them in serum-free medium for 48 h, and were subsequently stimulated with IGF-I (50 ng/ml) for 15 min (G) or were left without stimulation (H). There is an apparent membrane association and cytoplasmic immunoreactivity in cells stimulated with IGF-I. A–D, G, and H, ×40; E and F, ×20.

Close modal
Fig. 2.

Immunohistochemical detection of phosphorylated IGF-IR and Trk-C protein in medulloblastoma biopsies. Representative paraffin-embedded sections of two medulloblastoma cases (biopsy 1, A and B; and biopsy 2, C and D; see Table 1 for details) were stained with anti-pY1316 antibody that specifically recognizes the phosphorylated (active) form of the IGF-IR (A and C). Consecutive sections were also stained with anti-Trk-C antibody (B and D). Cells from biopsy 2 are strongly positive for anti-pY1316 antibody staining and are negative for anti-Trk-C antibody staining (C and D, respectively). Conversely, cells from biopsy 1 are characterized by a strong Trk-C immunoreactivity in the absence of anti-pY1316 antibody staining (A and B). ×40.

Fig. 2.

Immunohistochemical detection of phosphorylated IGF-IR and Trk-C protein in medulloblastoma biopsies. Representative paraffin-embedded sections of two medulloblastoma cases (biopsy 1, A and B; and biopsy 2, C and D; see Table 1 for details) were stained with anti-pY1316 antibody that specifically recognizes the phosphorylated (active) form of the IGF-IR (A and C). Consecutive sections were also stained with anti-Trk-C antibody (B and D). Cells from biopsy 2 are strongly positive for anti-pY1316 antibody staining and are negative for anti-Trk-C antibody staining (C and D, respectively). Conversely, cells from biopsy 1 are characterized by a strong Trk-C immunoreactivity in the absence of anti-pY1316 antibody staining (A and B). ×40.

Close modal
Fig. 3.

IGF-IR, PI3-K (p85), and IRS-1 protein levels and IGF-IR tyrosine phosphorylation in medulloblastomas. A, protein levels detected by Western blots developed with anti-IGF-IR (β subunit), anti-IRS-1, and ant-PI3-K (p85) antibodies. For a regular Western blot, 50 μg of total proteins isolated from the same samples were separated by PAGE, transferred onto nitrocellulose filters, and probed with corresponding antibodies (see “Materials and Methods”). There is an apparent up-regulation of the IGF-IR and IRS-1 proteins in medulloblastoma samples in comparison with normal human cerebellar tissue. B, IGF-IR tyrosine phosphorylation was evaluated by immunoprecipitating IGF-IR (IP) and developing the blot with antiphosphotyrosine antibody (pY). For IP/Western (W/pY), 500 μg protein samples were collected from R600 mouse fibroblasts (positive control), from medulloblastoma biopsies (p138, p494, and p609), and from normal human cerebellar tissues (N906, N989, and N1061). C, part of the blot containing proteins from normal human cerebellum (A) was purposely overexposed for 15 min to visualize low levels of IRS-1 protein in these samples. kD, Mr in thousands.

Fig. 3.

IGF-IR, PI3-K (p85), and IRS-1 protein levels and IGF-IR tyrosine phosphorylation in medulloblastomas. A, protein levels detected by Western blots developed with anti-IGF-IR (β subunit), anti-IRS-1, and ant-PI3-K (p85) antibodies. For a regular Western blot, 50 μg of total proteins isolated from the same samples were separated by PAGE, transferred onto nitrocellulose filters, and probed with corresponding antibodies (see “Materials and Methods”). There is an apparent up-regulation of the IGF-IR and IRS-1 proteins in medulloblastoma samples in comparison with normal human cerebellar tissue. B, IGF-IR tyrosine phosphorylation was evaluated by immunoprecipitating IGF-IR (IP) and developing the blot with antiphosphotyrosine antibody (pY). For IP/Western (W/pY), 500 μg protein samples were collected from R600 mouse fibroblasts (positive control), from medulloblastoma biopsies (p138, p494, and p609), and from normal human cerebellar tissues (N906, N989, and N1061). C, part of the blot containing proteins from normal human cerebellum (A) was purposely overexposed for 15 min to visualize low levels of IRS-1 protein in these samples. kD, Mr in thousands.

Close modal
Fig. 4.

Immunohistochemical detection and localization of IRS-1. Biopsies obtained from two representative patients with classic medulloblastoma diagnoses were stained with anti-IRS-1 antibody (B and C) and compared with IRS-1 staining in control cerebellar tissue (A). Present are a strong cytoplasmic immunoreactivity in the two medulloblastomas examined, significantly weaker immunoreactivity of the granular layer (arrows), and Purkinje cells (arrowheads) in normal human cerebellum. A, ×20; B and C, ×40.

Fig. 4.

Immunohistochemical detection and localization of IRS-1. Biopsies obtained from two representative patients with classic medulloblastoma diagnoses were stained with anti-IRS-1 antibody (B and C) and compared with IRS-1 staining in control cerebellar tissue (A). Present are a strong cytoplasmic immunoreactivity in the two medulloblastomas examined, significantly weaker immunoreactivity of the granular layer (arrows), and Purkinje cells (arrowheads) in normal human cerebellum. A, ×20; B and C, ×40.

Close modal
Fig. 5.

Detection of phosphorylated forms of MAPKs and Akt/PKB in medulloblastomas. Protein samples described in Fig. 3 were probed with antibodies that recognize the phosphorylated (active) forms of Erk-1, Erk-2, and Akt/PKB. The blot probed with anti-phospho-Akt antibody was “stripped” and reprobed with anti-phospho-Erk-1/Erk-2, anti-Akt, anti-Erk-1, and anti-Grb-2 antibodies. Grb-2 protein, which is usually expressed at a constant level, was used to monitor equal loading conditions. All of the primary and secondary antibodies are described in “Materials and Methods.” kD, Mr in thousands.

Fig. 5.

Detection of phosphorylated forms of MAPKs and Akt/PKB in medulloblastomas. Protein samples described in Fig. 3 were probed with antibodies that recognize the phosphorylated (active) forms of Erk-1, Erk-2, and Akt/PKB. The blot probed with anti-phospho-Akt antibody was “stripped” and reprobed with anti-phospho-Erk-1/Erk-2, anti-Akt, anti-Erk-1, and anti-Grb-2 antibodies. Grb-2 protein, which is usually expressed at a constant level, was used to monitor equal loading conditions. All of the primary and secondary antibodies are described in “Materials and Methods.” kD, Mr in thousands.

Close modal
Table 1

Clinical and immunohistochemical analysis of human medulloblastomas

Diagnosis of the tumors is based on the WHO classification of brain tumors, using the criteria for medulloblastomas as detailed in “Materials and Methods.” The anatomical location of the tumors, the age at the time of resection, and the gender of each patient are shown. Presence (+), or absence (−) of immunohistochemical reaction for GFAP, synaptophysin (SY-38), neurofilaments (NF), IRS-1, IGF-IR (IGF-IR Total), IGF-IR phosphorilated at tyrosine 1316 (IGF-IR Phosp), and Trk-C are indicated. Ten of 17 medulloblastomas were positive when tested with the anti-Y1316 antibody, indicating the presence of the phosphorilated (active) form of the IGF-IR in these samples.

No.DiagnosisGender/AgeLocationImmunohistochemistry
GFAPSY-38bNFIRS-1IGF-IR TotalIGF-IR PhospTrk-C
Desmoplastic medulloblastoma Male/6 yr Left hemisphere − − − − − +++ 
Classic medulloblastoma Male/2 yr Vermis ++ ++ − 
Desmoplastic medulloblastoma Male/7 yr Right hemisphere − ++ − ++ ++ − 
Desmoplastic medulloblastoma Female/3 yr Vermis − − ++ ++ − 
Medulloblastoma (neuroblastic) Female/17 yr Right hemisphere − − − − − ++ 
Desmoplastic medulloblastoma Female/16 mo Posterior fossa − +++ +a 
Desmoplastic medulloblastoma Male/11 yr Posterior fossa − − − 
Desmoplastic medulloblastoma Male/15 yr Left hemisphere − − 
Medulloblastoma (neuroblastic) Male/11 yr Posterior fossa − − ++ ++ − 
10 Desmoplastic medulloblastoma Female/16 mo Posterior fossa − +a − +++ 
11 Desmoplastic medulloblastoma Male/5 yr Posterior fossa − +a − − − ++ 
12 Classic medulloblastoma Male/7 yr Vermis − +++ − − 
13 Medulloblastoma (neuroblastic) Male/9 yr Vermis − ++ − − 
14 Desmoplastic medulloblastoma Female/8 yr Posterior fossa − − − − 
15 Medulloblastoma (neuroblastic) Male/5 yr Posterior fossa − +++ − ++ ++ − 
16 Medulloblastoma (neuroblastic) Male/12 yr Left hemisphere − − ++ ++ − 
17 Medulloblastoma (neuroblastic) Male/5 yr Posterior fossa − ++ − − − − +++ 
No.DiagnosisGender/AgeLocationImmunohistochemistry
GFAPSY-38bNFIRS-1IGF-IR TotalIGF-IR PhospTrk-C
Desmoplastic medulloblastoma Male/6 yr Left hemisphere − − − − − +++ 
Classic medulloblastoma Male/2 yr Vermis ++ ++ − 
Desmoplastic medulloblastoma Male/7 yr Right hemisphere − ++ − ++ ++ − 
Desmoplastic medulloblastoma Female/3 yr Vermis − − ++ ++ − 
Medulloblastoma (neuroblastic) Female/17 yr Right hemisphere − − − − − ++ 
Desmoplastic medulloblastoma Female/16 mo Posterior fossa − +++ +a 
Desmoplastic medulloblastoma Male/11 yr Posterior fossa − − − 
Desmoplastic medulloblastoma Male/15 yr Left hemisphere − − 
Medulloblastoma (neuroblastic) Male/11 yr Posterior fossa − − ++ ++ − 
10 Desmoplastic medulloblastoma Female/16 mo Posterior fossa − +a − +++ 
11 Desmoplastic medulloblastoma Male/5 yr Posterior fossa − +a − − − ++ 
12 Classic medulloblastoma Male/7 yr Vermis − +++ − − 
13 Medulloblastoma (neuroblastic) Male/9 yr Vermis − ++ − − 
14 Desmoplastic medulloblastoma Female/8 yr Posterior fossa − − − − 
15 Medulloblastoma (neuroblastic) Male/5 yr Posterior fossa − +++ − ++ ++ − 
16 Medulloblastoma (neuroblastic) Male/12 yr Left hemisphere − − ++ ++ − 
17 Medulloblastoma (neuroblastic) Male/5 yr Posterior fossa − ++ − − − − +++ 
a

Focal positivity.

b

Sy-38, Synaptophysin.

Samples of frozen human cerebellum were kindly provided by Drs. Mary Herman and Joel E. Kleinmen from Neuropathology Section–Clinical Brain Disorder Branch, National Institute of Mental Health, Bethesda, MD. We would like to express our gratitude to Dr. Christos D. Katsetos for his scientific input and critical discussion of the results.

1
Baserga R. Oncogenes and the strategy of growth factors.
Cell
,
79
:
927
-930,  
1994
.
2
Reiss K., Valentinis B., Tu X., Xu S. Q., Baserga R. Molecular markers of IGF-I-mediated mitogenesis.
Exp. Cell Res.
,
242
:
361
-372,  
1998
.
3
Sell C., Baserga R., Rubin R. Insulin-like growth factor I (IGF-I) and the IGF-I receptor prevent etoposide-induced apoptosis.
Cancer Res.
,
55
:
303
-306,  
1995
.
4
Gluckman P., Klempt N., Guan J., Mallard C., Sirimanne E., Dragunow M., Klempt M., Singh K., Williams C., Nikolics K. A role for IGF-I in the rescue of CNS neurons following hypoxic-ischemic injury.
Biochem. Biophys. Res. Commun.
,
182
:
593
-599,  
1992
.
5
Morrione A., Romano G., Navarro M., Reiss K., Valentinis B., Dews M., Eves E., Rosner M. R., Baserga R. Insulin-like growth factor I receptor signaling in differentiation of neuronal H19-7 cells.
Cancer Res.
,
60
:
2263
-2272,  
2000
.
6
Baserga R. The insulin-like growth factor I receptor: a key to tumor growth?.
Cancer Res.
,
55
:
249
-252,  
1995
.
7
Baserga R., Resnicoff M., Dews M. The IGF-I receptor and cancer.
Endocrine
,
7
:
99
-102,  
1997
.
8
White M. F. The IRS-1 signaling system.
Curr. Opin. Genet. Dev.
,
4
:
47
-54,  
1994
.
9
Yamamoto K., Altschuler D., Wood E., Horlick K., Jacobs S., Lapetina E. G. Association of phosphorylated insulin-like growth factor I receptor with the SH2 domains of phosphatidylinositol 3-kinase p85.
J. Biol. Chem.
,
267
:
11337
-11343,  
1992
.
10
Morrione A., Plant P., Valentinis B., Staub O., Kumar S., Rotin D., Baserga R. mGrb10 interacts with Nedd4.
J. Biol. Chem.
,
274
:
24094
-24099,  
1999
.
11
Furlanetto R. W., Dey B. R., Lopaczynski W., Nissley S. P. 14-3-3 proteins interact with the insulin-like growth factor receptor but not the insulin receptor.
Biochem. J.
,
327
:
765
-771,  
1997
.
12
Seely B. L., Reichart D. R., Staubs P. A., Jhun B. H., Hsu D., Maegawa H., Milarski K. L., Saltiel A. R., Olefsky J. M. Localization of the insulin-like growth factor I receptor binding sites for the SH2 domain proteins p85, Syp, and GTPase activating protein.
J. Biol. Chem.
,
270
:
19151
-19157,  
1995
.
13
Arbet-Engels C., Tartare-Deckert S., Eckhart W. C-terminal Src kinase associates with ligand-stimulated insulin-like growth factor I receptor.
J. Biol. Chem.
,
274
:
5422
-5428,  
1999
.
14
Kauffmann-Zeh A., Rodriguez-Viciana P., Ulrich E., Gilbert C., Coffer P., Downward J., Evan G. Suppression of c-Myc-induced apoptosis by Ras signalling through PI(3)K and PKB.
Nature (Lond.)
,
385
:
544
-548,  
1997
.
15
Kulik G., Klippel A., Weber M. J. Antiapoptotic signalling by the insulin-like growth factor I receptor, phosphatidylinositol 3-kinase, and Akt.
Mol. Cell. Biol.
,
17
:
1595
-1606,  
1997
.
16
Myers M. G., Jr., Sun X. J., Cheatham B., Jachna B. R., Glasheen E. M., Backer J. M., White M. F. IRS-1 is a common element in insulin and insulin-like growth factor I signaling to the phosphatidylinositol 3′-kinase.
Endocrinology
,
132
:
1421
-1430,  
1993
.
17
LeRoith D., Werner H., Faria T. N., Kato H., Adamo M., Roberts C. T., Jr. Insulin-like growth factor receptors. Implications for nervous system function.
Ann. N.Y. Acad. Sci.
,
692
:
22
-32,  
1993
.
18
Glick R. P., Unterman T. G., Blaydes L., Hollis R. Insulin-like growth factors in central nervous system tumors.
Ann. N.Y. Acad. Sci.
,
692
:
223
-229,  
1993
.
19
D’Mello S. R., Galli C., Ciotti T., Calissano P. Induction of apoptosis in cerebellar granule neurons by low potassium: inhibition of death by insulin-like growth factor I and cAMP.
Proc. Natl. Acad. Sci. USA
,
90
:
10989
-10993,  
1993
.
20
Cui H., Meng Y., Bulleit R. F. Inhibition of glycogen synthase kinase 3β activity regulates proliferation of cultured cerebellar granule cells.
Brain. Res. Dev. Brain Res.
,
111
:
177
-188,  
1998
.
21
Trojan J., Blossey B. K., Johnson T. R., Rudin S. D., Tykocinski M., Ilan J. Loss of tumorigenicity of rat glioblastoma directed by episome-based antisense cDNA transcription of insulin-like growth factor I.
Proc. Natl. Acad. Sci. USA
,
89
:
4874
-4878,  
1992
.
22
Trojan J., Johnson T. R., Rudin S. D., Ilan J., Tykocinski M. L. Treatment and prevention of rat glioblastoma by immunogenic C6 cells expressing antisense insulin-like growth factor I RNA.
Science (Wash. DC)
,
259
:
94
-97,  
1993
.
23
Trojan J., Duc H. T., Upegui-Gonzalez L. C., Hor F., Guo Y., Anthony D., Ilan J. Presence of MHC-I and B-7 molecules in rat and human glioma cells expressing antisense IGF-I mRNA.
Neurosci. Lett.
,
212
:
9
-12,  
1996
.
24
Martin D. M., Yee D., Carlson R. O., Feldman E. L. Gene expression of the insulin-like growth factors and their receptors in human neuroblastoma cell lines.
Brain Res. Mol. Brain Res.
,
15
:
241
-246,  
1992
.
25
Antoniades H. N., Galanopoulos T., Neville-Golden J., Maxwell M. Expression of insulin-like growth factors I and II and their receptor mRNAs in primary human astrocytomas and meningiomas; in vivo studies using in situ hybridization and immunocytochemistry.
Int. J. Cancer
,
50
:
215
-222,  
1992
.
26
Kurihara M., Tokunaga Y., Ochi A., Kawaguchi T., Tsutsumi K., Shigematsu K., Niwa M., Mori K. Expression of insulin-like growth factor I receptors in human brain tumors: comparison with epidermal growth factor receptor by using quantitative autoradiography (in Japanese).
No To Shinkei
,
41
:
719
-725,  
1989
.
27
Patti R., Reddy C. D., Geoerger B., Grotzer M. A., Raghunath M., Sutton L. N., Phillips P. C. Autocrine secreted insulin-like growth factor I stimulates MAP kinase-dependent mitogenic effects in human primitive neuroectodermal tumor/medulloblastoma.
Int. J. Oncol.
,
16
:
577
-584,  
2000
.
28
Hahn H., Wojnowski L., Specht K., Kappler R., Calzada-Wack J., Potter D., Zimmer A., Muller U., Samson E., Quintanilla-Martinez L. Patched target IGF2 is indispensable for the formation of medulloblastoma and rhabdomyosarcoma.
J. Biol. Chem.
,
275
:
28341
-28344,  
2000
.
29
Wang J. Y., Del Valle L., Gordon J., Rubini M., Romano G., Croul S., Peruzzi F., Khalili K., Reiss K. Activation of the IGF-IR system contributes to malignant growth of human and mouse medulloblastomas.
Oncogene
,
20
:
3857
-3868,  
2001
.
30
Chou T. T., Trojanowski J. Q., Lee V. M. A novel apoptotic pathway induced by nerve growth factor-mediated TrkA activation in medulloblastoma.
J. Biol. Chem.
,
275
:
565
-570,  
2000
.
31
Kim J. Y., Sutton M. E., Lu D. J., Cho T. A., Goumnerova L. C., Goritchenko L., Kaufman J. R., Lam K. K., Billet A. L., Tarbell N. J., Wu J., Allen J. C., Stiles C. D., Segal R. A., Pomeroy S. L. Activation of neurotrophin-3 receptor TrkC induces apoptosis in medulloblastomas.
Cancer Res.
,
59
:
711
-719,  
1999
.
32
Evans A. E., Kisselbach K. D., Yamashiro D. J., Ikegaki N., Camoratto A. M., Dionne C. A., Brodeur G. M. Antitumor activity of CEP-751 (KT-6587) on human neuroblastoma and medulloblastoma xenografts.
Clin. Cancer Res.
,
5
:
3594
-3602,  
1999
.
33
Grotzer M. A., Janss A. J., Fung K., Biegel J. A., Sutton L. N., Rorke L. B., Zhao H., Cnaan A., Phillips P. C., Lee V. M., Trojanowski J. Q. TrkC expression predicts good clinical outcome in primitive neuroectodermal brain tumors.
J. Clin. Oncol.
,
18
:
1027
-1035,  
2000
.
34
Kleihues, P., and Cavanee, W. K. (eds.), Pathology and Genetics of Tumours of the Nervous System, pp. 56–61. London, France: IARC Press, 2000.
35
Rubini M., Hongo A., D’Ambrosio C., Baserga R. The IGF-I receptor in mitogenesis and transformation of mouse embryo cells: role of receptor number.
Exp. Cell Res.
,
230
:
284
-292,  
1997
.
36
Baker J., Liu J. P., Robertson E. J., Efstratiadis A. Role of insulin-like growth factors in embryonic and postnatal growth.
Cell
,
75
:
73
-82,  
1993
.
37
Rubini M., D’Ambrosio C., Carturan S., Yumet G., Catalano E., Shan S., Huang Z., Criscuolo M., Pifferi M., Baserga R. Characterization of an antibody that can detect an activated IGF-I receptor in human cancers.
Exp. Cell Res.
,
251
:
22
-32,  
1999
.
38
Valentinis B., Morrione A., Peruzzi F., Prisco M., Reiss K., Baserga R. Antiapoptotic signaling of the IGF-I receptor in fibroblasts following loss of matrix adhesion.
Oncogene
,
18
:
1827
-1836,  
1999
.
39
Majewski M., Nieborowska-Skorska M., Salomoni P., Slupianek A., Reiss K., Trotta R., Calabretta B., Skorski T. Activation of mitochondrial Raf-1 is involved in the antiapoptotic effects of Akt.
Cancer Res.
,
59
:
2815
-2819,  
1999
.
40
Payne D. M., Rossomando A. J., Martino P., Erickson A. K., Her J. H., Shabanowitz J., Hunt D. F., Weber M. J., Sturgill T. W. Identification of the regulatory phosphorylation sites in pp42/mitogen-activated protein kinase (MAP kinase).
EMBO J.
,
10
:
885
-892,  
1991
.
41
Peruzzi F., Prisco M., Dews M., Salomoni P., Grassilli E., Romano G., Calabretta B., Baserga R. Multiple signaling pathways of the insulin-like growth factor 1 receptor in protection from apoptosis.
Mol. Cell. Biol.
,
19
:
7203
-7215,  
1999
.
42
Zheng B., Clemmons D. R. Blocking ligand occupancy of the αVβ3 integrin inhibits insulin-like growth factor I signaling in vascular smooth muscle cells.
Proc. Natl. Acad. Sci. USA
,
95
:
11217
-11222,  
1998
.
43
Reiss K., Wang J. Y., Romano G., Tu X., Peruzzi F., Baserga R. Mechanisms of regulation of cell adhesion and motility by insulin receptor substrate 1 in prostate cancer cells.
Oncogene
,
20
:
490
-500,  
2001
.
44
Reiss K., Wang J. Y., Romano G., Furnari F. B., Cavenee W. K., Morrione A., Tu X., Baserga R. IGF-I receptor signaling in a prostatic cancer cell line with a PTEN mutation.
Oncogene
,
19
:
2687
-2694,  
2000
.
45
D’Ambrosio C., Keller S. R., Morrione A., Lienhard G. E., Baserga R., Surmacz E. Transforming potential of the insulin receptor substrate 1.
Cell Growth Differ.
,
6
:
557
-562,  
1995
.
46
Jackson J. G., White M. F., Yee D. Insulin receptor substrate 1 is the predominant signaling molecule activated by insulin-like growth factor-I, insulin, and interleukin-4 in estrogen receptor-positive human breast cancer cells.
J. Biol. Chem.
,
273
:
9994
-10003,  
1998
.
47
Waters S. B., Yamauchi K., Pessin J. E. Functional expression of insulin receptor substrate-1 is required for insulin-stimulated mitogenic signaling.
J. Biol. Chem.
,
268
:
22231
-22234,  
1993
.
48
Zimmermann S., Moelling K. Phosphorylation and regulation of Raf by Akt (protein kinase B).
Science (Wash. DC)
,
286
:
1741
-1744,  
1999
.