The Expression Profile of FRAT1 in Human Gliomas

BRAIN RESEARCH 1320 (2010) 152– 158

available at www.sciencedirect.com

www.elsevier.com/locate/brainres

Research Report

The expression profile of FRAT1 in human gliomas

Geng Guoa,1, Xinggang Maoa,1, Peng Wanga,1, Bolin Liua, Xiang Zhanga,⁎, Xiaofan Jianga, Chengliang Zhongb, Junli Huoa, Ji Jinc, Yuzhen Zhuod aDepartment of Neurosurgery, Xijing Hospital, Fourth Military Medical University, 15# West Chang Le Road, Xi'an, Shaanxi Province 710032, People's Republic of China bDepartment of Health Statistics, Faculty of Preventative Medicine, Fourth Military Medical University, 17# West Chang Le Road, Xi'an, Shaanxi Province 710032, People's Republic of China cShanxi Medical University, 56# South Xin Jian Road, Taiyuan, Shanxi Province 030001, People's Republic of China dTianjin Institute of Acute Abdomen, Tianjin Nankai Hospital, 122# San Wei Road, Tianjin 300100, People's Republic of China

ARTICLE INFO ABSTRACT

Article history: FRAT1 was originally characterized as a protein frequently rearranged in advanced T cell Accepted 13 January 2010 lymphoma, which inhibits GSK-3-mediated phosphorylation of β-catenin and positively Available online 21 January 2010 regulates the Wnt signaling pathway. FRAT1 has been identified as a proto-oncogene involved in tumorigenesis. Previous studies have shown that FRAT1 is strikingly Keywords: overexpressed in some human cancers. However, the relationship between FRAT1 and FRAT1 human gliomas is unclear. In this study, we detected the expression of FRAT1 in human β-catenin gliomas by immunohistochemistry, Western blot and RT-PCR. FRAT1 was found to be Glioma specifically expressed in the majority of glioma samples, and their expression levels Immunohistochemistry increased markedly with the increase of WHO grades. In addition, there was a positive Western blot correlation between FRAT1 immunoreactivity score (IRS) and β-catenin IRS. Our results RT-PCR suggest that FRAT1 may be an important factor in the tumorigenesis and progression of gliomas, and could be used as a potential molecular marker for pathological diagnosis and a target for biological therapy. © 2010 Elsevier B.V. All rights reserved.

1. Introduction patients diagnosed with high-grade gliomas die within the first year, even after receiving multidisciplinary treatments, in- Gliomas are the most common primary neoplasms arising cluding surgical resection, radiotherapy and chemotherapy from the brain or spinal cord tissue, composing over 40% of all (Hess et al., 2004). New therapies based on a better under- such tumors and 78% of central nervous system malignancies standing of the molecular mechanisms of gliomagenesis are in adults (Buckner et al., 2007). This type of tumor is necessary to improve the outcome of treatment. Therefore, it is characterized by progressive overgrowth of glial tissue, diffuse essential for us to define specific tumor markers in tumori- and relentless invasion. Despite recent advances in treatment genesis and progression of gliomas, and to design more strategies, gliomas remain a poor prognosis in decades due to a effective therapeutic strategies (van den Bent et al., 2006). high rate of recurrence, which also decreased the effectiveness FRAT1 (frequently rearranged in advanced T cell lymphomas- of surgical therapy (Kaba and Kyritsis, 1997). Till now, half of 1) gene, which is a human homologue of mouse proto-oncogene

⁎ Corresponding author. Fax: +86 29 84775567. E-mail address: [email protected] (X. Zhang). 1 These authors contributed equally to this work.

Frat1 (Jonkers et al., 1997; Saitoh and Katoh, 2001), is located on expression of β-catenin increased gradually with the increase of human chromosome 10q24.1, encoding a 29-kDa protein pathological grade of glioma. There was a positive correlation comprising 279 amino acids. FRAT1 was initially cloned in a between FRAT1 IRS and β-catenin IRS (r=0.782, P<0.001). screen to identify genes that accelerate lymphomagenesis in Representative images of β-catenin immunostaining are oncogene-bearing transgenic mice (Freemantle et al., 2002). shown in Fig. 1 and the related results are shown in Tables 2 Its biological function, however, remained elusive until its and 3. Xenopus homolog GBP (GSK3-binding protein) was isolated (Yost et al., 1998). All FRAT/GBP homologs have an established role in Xenopus embryonic development through inhibition of 3. Discussion GSK-3 phosphorylation of β-catenin (Jonkers et al., 1999a; Saitoh et al., 2001; Yost et al., 1998), and they were thus shown to be a Our study focused on the relationship between tumorigenesis core component of the canonical Wnt pathway in Xenopus of gliomas and FRAT1 expression. Gliomas are one of the most (Dominguez and Green, 2000; Farr et al., 2000; Yost et al., 1998). aggressive human malignancies. Patients with the most Moreover, there are some supportive reports showing that malignant histopathologic subtype, glioblastoma, carry the FRAT1 plays a role in tumor progression (Jonkers et al., 1997, worst clinical prognosis, despite advanced surgery and 1999b; Saitoh et al., 2002; Wang et al., 2006). However, little is adjuvant radiotherapy and chemotherapy (van den Bent known about the relationship between FRAT1 and glioma. In the et al., 2006). Although the molecular and genetic basis present study we found that FRAT1 expression correlated underlying the pathogenesis and treatment resistance of positively with the increasing WHO grade of glioma and these tumors is becoming better understood in recent years, expression level of β-catenin, which may have important much still remains unclear at present (Sanson et al., 2004). It implications in both predicting the clinical prognosis and has been elucidated that many critical signal pathways understanding the biology of these tumors. important for the development of nervous system and neural stem cells also have great impact on the tumorigenesis of gliomas, such as Notch, BMP, EGFR, BMI1, Hedgehog, PTEN, 2. Results Wnt/β-catenin, etc., which might be aberrantly regulated in gliomas (Mao et al., 2009). Wnt/β-catenin signaling is a 2.1. Immunohistochemical study of FRAT1 conserved molecular mechanism in metazoan animals. This pathway broadly influence changes in gene expression that In our study, FRAT1 protein was overexpressed in brain govern embryogenesis and postnatal responses, such as cell gliomas. Immunopositive tumor cells showed primarily cyto- proliferation, cell-fate determination, cell survival, cell behav- plasmic labeling under light microscope. However, normal ior and migration during morphogenesis (Logan and Nusse, brain tissues had exceedingly weak or absent immunoreactiv- 2004). Abnormal Wnt/β-catenin signaling is associated with ity for this protein (Fig. 1I). The positive expression rate of many human diseases, including cancer, osteoporosis, aging FRAT1 was 66.7% (56/84), and its IRS was 4.07±3.60 for 84 cases and degenerative disorders (Clevers, 2006; Moon et al., 2004). of tumor specimen. The FRAT1 IRS (immunoreactivity score) As one essential regulator for the development of nervous were positively and markedly correlated with increasing WHO system, Wnt/β-catenin signaling participates in the process of grades (P<0.001). There were significant FRAT1 IRS differences almost all aspects of neural development, including stem between Grade I and Grade III (P<0.001), Grade I and Grade IV cell proliferation, maintenance and differentiation (Ille and (P<0.001), Grade II and Grade III (P=0.033), Grade II and Grade IV Sommer, 2005). Recently, Wnt/β-catenin signaling was (P<0.001), and Grade III and Grade IV (P=0.035) gliomas. reported to contribute to the formation of gliomas, and some Representative images of FRAT1 immunostaining are shown proteins involved in the Wnt/β-catenin pathway were abnor- in Fig. 1 and the related results are shown in Tables 1, 2, and 3. mally expressed in gliomas (Sareddy et al., 2009; Yu et al., 2007). In addition, it has been demonstrated that Wnt/β- 2.2. The mRNA and protein expressions of FRAT1 catenin signaling played important roles in the genesis of other tumors such as leukemia, colon cancers, etc. (Van der In this study, the mRNA and protein expression levels of Flier et al., 2007; Zhao et al., 2007). However, the exact ways FRAT1 were represented by the ratio of gray value of FRAT1 to that Wnt/β-catenin signaling acts on the tumorigenesis of that of internal control (β-actin or GAPDH) in Western blot and gliomas are not clarified. Therefore, this study focused on RT-PCR. The results of Western blot and RT-PCR were investigating the expression of FRAT1 on gliomas, in order to coincident with that of immunohistochemistry. The results provide more knowledge about the roles that Wnt/β-catenin showed that mRNA and protein expression levels of FRAT1 pathway plays during the tumorigenesis of gliomas. increased markedly with the increase in pathologic grade of FRAT1 has been identified as a positive regulator of the brain gliomas (P<0.001, Tables 2 and 3; Figs. 2 and 3). Wnt/β-catenin pathway, which can inhibit GSK3 activity and ultimately down-regulate β-catenin (Ferkey and Kimelman, 2.3. Expression of β-catenin and its correlation with 2002; Fraser et al., 2002; Giles et al., 2003). It has been proposed pathologic grade and FRAT1 expression in gliomas that activation of the Wnt signaling cascade can cause Dishevelled (Dvl) family protein to recruit FRAT/GBP into the The expression of β-catenin was detected in cytoplasm and/or β-catenin degradation complex, leading to dissociation of nucleus of glioma cells by immunohistochemistry. There GSK-3 from Axin and consequently stabilization of β-catenin was no expression of β-catenin in normal brain tissues. The (Li et al., 1999; Salic et al., 2000). This may result in the 154 BRAIN RESEARCH 1320 (2010) 152– 158

Fig. 1 – Representative sections for FRAT1 and β-catenin Immunoreactivity in normal control brain and glioma tissue. FRAT1 immunoreactivity shows homogeneously brown-yellow staining in cytoplasm; β-catenin immunoreactivity shows brown-yellow staining in cytoplasm and/or nucleus (LSAB method, hematoxylin counterstain; original magnifications 400× for A–I). (A and E) Glioma of Grade I; (B and F) glioma of Grade II; (C and G) glioma of Grade III; (D and H) glioma of Grade IV; (I) normal brain tissue showing no specific immunopositive staining. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

cytoplasmic accumulation of β-catenin, which then translo- responsive genes are then induced (Tolwinski and Wieschaus, cates into the nucleus, where complexes with members of the 2004). Several transcriptional targets of Wnt/β-catenin signal- TCF/LEF (T cell factor/lymphoid enhancer factor) family of ing are known oncogenes in the pathogenesis of cancers transcription factors are formed and the transcription of Wnt (Polakis, 2000). For example, two of the TCF/LEF target genes, c- myc and cyclin D1 are supposed to be involved in the oncogenic function of inappropriate Wnt signal activation (Behrens et al., 1996; Pelengaris et al., 2002). This mechanism is Table 1 – FRAT1-IRS in normal brain tissues and human gliomas of different pathologic grades. believed to contribute to tumor progression. Previous studies have shown that FRAT1 is strikingly overexpressed in several Pathologic n FRAT1-IRS The human cancers including esophageal cancer, cervical cancer, grade positive − + ++ +++ expression breast cancer, ovarian cancer, and so on (Saitoh et al., 2002; Wang et al., 2006, 2008). However, recent study by van (0) (1–3) (4–6) (8–12) rate Amerongen and colleagues demonstrated that FRAT is not NBT 6 6 0 0 0 0.000(0/6) essential components of Wnt/β-catenin signaling in mam- Grade I 16 9 6 1 0 0.438(7/16) mals (van Amerongen and Berns, 2005; van Amerongen et al., Grade II 22 10 4 4 4 0.545(12/22) 2005). The paradoxical phenomenon deserves further study to Grade III 19 4 2 7 6 0.789(15/19) investigate the gene function of FRAT in gliomagenesis. Grade IV 27 5 0 5 17 0.815(22/27) Total 84 28 12 17 27 0.667(56/84) To the best of our knowledge, this study is the first detection of FRAT1 expression in human gliomas. We found that Note: NBT represents normal brain tissue. The normal brain tissues FRAT1 was overexpressed in most subtypes of gliomas by were not included in the total number listed in the table. immunohistochemistry, RT-PCR and Western blot. Our results BRAIN RESEARCH 1320 (2010) 152– 158 155

Table 2 – Expression level of FRAT1 in normal brain tissues and human gliomas of different pathologic grades (mean±SD). Histological n FRAT1 IRS β-catenin IRS mRNA relative Protein relative classification density of FRAT1 density of FRAT1

NBT 6 0 0 0 0 Grade I 16 1.25±1.77 1.06±1.18 0.22 ±0.12 0.18±0.06 Grade II 22 2.64±3.13 2.86±2.75 0.42 ±0.08 0.38±0.12 Grade III 19 4.63±3.17 4.74±3.07 0.65 ±0.14 0.64±0.13 Grade IV 27 6.52±3.42 7.15±1.90 0.86 ±0.13 0.84±0.14 Total 84 4.07±3.60 4.32±3.25 0.57 ±0.27 0.55±0.28 P value a <0.001 <0.001 <0.001 <0.001 P value b <0.001 <0.001 <0.001 <0.001

Note: The mRNA relative density of FRAT1 was represented by the gray value of FRAT1/gray value of own GAPDH in RT-PCR. The protein relative density of FRAT1 was represented by the gray value of FRAT1/gray value of own β-actin in Western blot. NBT represents normal brain tissue. The normal brain tissues were not included in the total number listed in the table. There was a positive correlation between FRAT1 IRS and β-catenin IRS by the Spearman rank test (r=0.782, P<0.001). a Difference of FRAT1-IRS, β-catenin IRS, mRNA relative density and protein relative density of FRAT1 in different pathologic grade groups was compared using analysis of variance (ANOVA). b The Spearman rank test was used to establish correlation of histological grades with FRAT1 IRS, β-catenin IRS, mRNA relative density and protein relative density of FRAT1.

indicate that mRNA and protein expression levels of FRAT1 However, it should be acknowledged that this study was were significantly positively correlated with increasing WHO retrospective, and therefore has some limitations. First, our grades. Furthermore, FRAT1 only showed weak immunoreac- investigation contains only 90 samples, and larger retrospec- tivity in some glial cells of normal brain tissue by immuno- tive studies are necessary to determine the expression of histochemistry. These results indicated that FRAT1 might not FRAT1. Second, the biases of tumor histologic subtypes may only be necessary in the normal physiological state to main- influence the results. In addition, it must be kept in mind that tain normal functions of glial cells but also be involved in tumorigenesis and malignancy progression of glioma under certain pathological conditions. In addition, we used immunohistochemistry to investigate β-catenin expression in glioma samples. There was a positive correlation between FRAT1 IRS and β-catenin IRS. In most cases, β-catenin cytoplasm and/or nucleus accumulation was correlated with high levels of FRAT1 expression in tumor specimens. Taken together, since key genetic, epigenetic and environmental factors associated with gliomagenesis remain incompletely defined, our findings shed some light on the unresolved molecular mechanism of gliomagenesis, which may lead to a better understanding of the roles of dysregulated Wnt signaling in tumorigenesis of human gliomas.

Table 3 – ANOVA analysis of comparison between pathological grades. Variables P-value

FRAT1 β-catenin mRNA Protein IRS IRS relative relative density density of FRAT1 of FRAT1

NBT vs. Grade I 0.378 0.334 <0.001 0.002 NBT vs. Grade II 0.056 0.008 <0.001 <0.001 Fig. 2 – mRNA expression of FRAT1 in normal brain tissues NBT vs. Grade III <0.001 <0.001 <0.001 <0.001 NBT vs. Grade IV <0.001 <0.001 <0.001 <0.001 and human gliomas of different grades by RT-PCR. The Grade I vs. II 0.156 0.019 <0.001 <0.001 FRAT1 fragment was amplified by 30 cycles using the paired Grade I vs. III <0.001 <0.001 <0.001 <0.001 primers as mentioned in Experimental procedures. GAPDH Grade I vs. IV <0.001 <0.001 <0.001 <0.001 was used as an internal standard. M, marker (125 bp DNA Grade II vs. III 0.033 0.010 <0.001 <0.001 Ladder); N, normal brain tissue; lane 1, glioma of Grade I; lane Grade II vs. IV <0.001 <0.001 <0.001 <0.001 2, glioma of Grade II; lane 3, glioma of Grade III; lane 4, glioma Grade III vs. IV 0.035 <0.001 <0.001 <0.001 of Grade IV. 156 BRAIN RESEARCH 1320 (2010) 152– 158

Table 4 – Histologic subtypes and pathologic grades of human glioma samples. Histopathologic diagnosis WHO grade Sample size

Pilocytic astrocytoma I 16 Diffuse astrocytoma II 17 Oligodendroglioma II 5 Anaplastic astrocytoma III 15 Anaplastic oligodendroglioma III 4 Glioblastoma IV 27 Total 84

trauma who consented to the procedure were used as control. All tumor tissues were obtained from the initial surgery, and none of the patients had been subjected to chemotherapy or radiation therapy before tumor excision. For the experimental use of the surgical specimens, informed consent was obtained prior to surgery from all enrolled patients according to the hospital ethical guidelines. All experiments and control experiments were performed three times on different days using the same protocol and time exposures. All experiments were approved by the local ethics committee. Fig. 3 – Protein expression of FRAT1 in normal brain tissues and human gliomas of different grades by Western blot. Lane 4.2. Immunohistochemistry 1, normal brain tissue; lane 2 and lane 3, glioma of Grade I; lane 4 and lane 5: glioma of Grade II; lane 6 and lane 7: glioma Formalin-fixed, paraffin-embedded specimens were cut into of Grade III; lane 8 and lane 9: glioma of Grade IV. 4 μm-thick serial sections, then mounted on poly-L-lysine- coated slides and baked overnight at 60 °C. The sections were dewaxed in xylene, followed by rehydration in a decreasing FRAT1 are not expressed in all glioma samples. The functions concentration of ethanol solutions. For routine pathological of FRAT1 need further investigation in the future. examination, deparaffinized sections from all blocks were In summary, the high expression of FRAT1 in both levels of stained with hematoxylin and eosin. No histological abnormal- mRNA and protein, as well as its expression pattern which ities were detected in the sections from any of the six normal positively correlated with increasing WHO grades in glioma control brain tissues. Antigen retrieval was carried out by tissues, strongly suggests that FRAT1 may play a crucial role in pressure-cooking in 0.01 M citrate buffer (pH 6.0) for 10 min. tumorigenesis and progression of gliomas. We also propose Endogenous peroxidase activity and non-specific binding site that FRAT1 protein may be useful as a valuable biomarker for were blocked with 3% H2O2 and non-immune serum, respec- the molecular diagnosis of glioma and a potential target for tively. Sections were then incubated with primary antibodies gene therapy of glioma. overnight at 4 °C in a humidified chamber. The rabbit anti-frat1 (U-25) polyclonal antibody and the rabbit anti-β-catenin (H-102) polyclonal antibody were used at 1:50 and 1:100 dilution (Santa 4. Experimental procedures Cruz Biotechnology, Santa Cruz, CA, USA), respectively. Then the primary antibodies were detected using the appropriate labeled 4.1. Patients Streptavidin-Biotin (LSAB) kit (Maixin Biotechnology, Fuzhou, China) according to the manufacturer's instructions. Immuno- Tumor specimens from 84 patients with glioma who underwent labeled sections were visualized with 3′,3′-diaminobenzidine surgical resection in the Xijing Hospital of the Fourth Military tetrahydrochloride (DAB; Sigma, St. Louis, MO, USA), and Medical University between 2007 and 2009 were randomly counterstained with hematoxylin. As a specificity control, selected from the archival files of the specimen bank of the phosphate buffered saline (PBS) was used instead of the primary Institute of Neurosurgery of People's Liberation Army (Xi'an, antibody to exclude non-specific binding of the secondary China). Of the patients, 47 were male and 37 were female. antibody. All immunostaining experiments were assessed by Patients' ages at the time of surgery ranged from 12 to 75 (41.2 an experienced pathologist blinded to all clinical data. ±18.3, mean±standard deviation [SD]). According to the revised World Health Organization criteria for the central nervous 4.3. Evaluation of immunostaining system (Buckner et al., 2007), patients were categorized into Grade I (n=16), Grade II (n=22), Grade III (n=19) and Grade IV Microscopic analysis was performed by two independent (n=27). The histological subtypes and pathologic grades of all neuropathologists. Digital microscope images were captured by glioma samples which were confirmed by two pathologists Olympus CH 30 microscope (Olympus, Tokyo, Japan). For each independently are given in Table 4. Six samples of slightly slide, 10 high power (400×) fields (about 1000 cells) were impaired brain tissue fragments of volunteers with cerebral randomly picked for quantification. The FRAT1 staining BRAIN RESEARCH 1320 (2010) 152– 158 157 intensity (FRAT1-SI), percentage of FRAT1-positive tumor cells anti-goat IgG for FRAT1, Santa Cruz Biotechnology, CA, USA; (FRAT1-PP), and a resulting FRAT1 immunoreactivity score donkey anti-mouse IgG for β-actin, Santa Cruz Biotechnology, (FRAT1-IRS) were assessed as a modification of the technique CA, USA) at room temperature for 1 h. Finally, the antibody described previously (Remmele et al., 1986). Briefly, the immu- antigen complex was detected by using an enhanced chemilu- noreactivity score (FRAT1-IRS: negative 0; weak 1–3; moderate 4– minescence's (ECL) detection solution (Pierce, Rockford, IL, USA). 6; strong 8–12) was determined by multiplication of the values for Densitometric analysis was performed using Scion Image FRAT1 staining intensity (FRAT1-SI: 0, no staining; 1, weak software (Scion Corporation, Frederick, MD, USA). All bands staining; 2, moderate staining; 3, strong staining) and the values were normalized as percentages of the control values. for percentage of FRAT1-positive tumor cells (FRAT1-PP: 0, <1%; 1, 1–25%; 2, 26–50%; 3, 51–75%; 4, >75%). Because of the 4.6. Statistical analysis heterogenous staining intensity of tumor cells, SI was determined according to the staining intensity of most cells. All statistical analysis was performed using the SPSS statistical software package, version 16.0 (SPSS Inc, Chicago, IL, USA). Data 4.4. RT-PCR were expressed as mean±SD. Comparisons between groups of different pathologic grades were performed using one-way Total RNA was isolated from human normal brain and brain analysis of variance (ANOVA), then the Student–Newman– tumor tissues using TRIZOL reagent (Invitrogen, Carlsbad, CA, Keuls test (SNK test, i.e., q test) was used for comparison of USA). The reverse transcriptase polymerase chain reaction (RT- differences between the two groups by ANOVA. The Spearman PCR) was done with TaKaRa RNA PCR Kit (AMV) version 3.0 rank test was used to establish correlation not only between (TaKaRa, Dalian, China). All procedures were performed accord- FRAT1 expression value and histological grades but also FRAT1 ing to the manufacturer's protocol. The primer pair used for IRS and β-catenin IRS. Values of P<0.05 were considered amplification of the human FRAT1 was: sense primer, 5′- statistically significant. GCCCTGTCTAAAGTGTATTTTCAG-3′, and anti-sense primer, 5′-CGCTTGAGTAGGACTGCAGAG-3′. As an internal control to normalize variances, GAPDH was co-amplified by PCR using the Acknowledgments following primers: sense primer: 5′-GAAGTGAAGGTCG- GAGTCA-3′; and anti-sense primer: 5′-TTCACACCCATGACGAA- We are grateful to Dr. Zhiqiang Yang (Department of CAT-3′. All PCR reactions were performed using standard PCR Pathology, College of Basic Medical Sciences, China Medical conditions: 95 °C for 5 min, 95 °C for 1 min, annealing at 56 °C for University) and Dr. Fucheng Ma (Department of Pathology, 1 min and extension at 72 °C for 1 min for 30 cycles and a final Xijing Hospital, Fourth Military Medical University) for con- extension at 72 °C for 10 min. The lengths of the FRAT1, and firming the histopathological diagnosis of all the samples used GAPDH amplicons were 325 bp and 402 bp, respectively. The PCR in this research and helping to perform microscopic analysis products were electrophoresed in a 1.5% agarose gel containing for FRAT1 immunoreactivity. We appreciate Prof. Guirong 0.1 μg/μl ethidium bromide and photographed using Kodak gel Ding for her constructive suggestions and generous help in documentation system. Densitometric analysis was performed preparation of the manuscript. We also thank Mrs. Wanjuan using Scion Image software (Scion Corporation, Frederick, MD, Yang, Mrs. Xiaoyan Chen, Ms. Juan Li and Mrs. Yufen Shi for USA). A grayscale intensity value was determined for each target preparing for this research. band.

4.5. Western blot analysis REFERENCES

Frozen tissues were homogenized in RIPA lysis buffer on ice. Behrens, J., von Kries, J.P., Kuhl, M., Bruhn, L., Wedlich, D., After centrifugation, the supernatant was collected, and total Grosschedl, R., Birchmeier, W., 1996. Functional interaction of protein concentrations were determined using the BioRad beta-catenin with the transcription factor LEF-1. Nature 382, protein assay kit (Hercules, CA, USA) and measured with the 638–642. micro-BCA protein assay (Pierce, Rockford, IL, USA). The protein Buckner, J.C., Brown, P.D., O'Neill, B.P., Meyer, F.B., Wetmore, C.J., samples (40 μg/lane) were separated by 12% sodium dodecyl Uhm, J.H., 2007. Central nervous system tumors. Mayo Clin. sulfate-polyacrylamide gel electrophoresis (SDS-PAGE), and Proc. 82, 1271–1286. then transferred to nitrocellulose membranes (Sigma-Aldrich Clevers, H., 2006. Wnt/beta-catenin signaling in development and disease. Cell 127, 469–480. Co., St. Louis, MO, USA). All blots were blocked by incubating Dominguez, I., Green, J.B., 2000. Dorsal downregulation of with 5% nonfat dry milk in Tris-buffered saline (TBS; pH 7.4) GSK3beta by a non-Wnt-like mechanism is an early molecular containing 0.1% Tween 20 (TBS-T) at room temperature for consequence of cortical rotation in early Xenopus embryos. 1 h. Next, the membrane was incubated with primary Development 127, 861–868. antibodies diluted in blocking buffer at 4 °C overnight. The Farr 3rd, G.H., Ferkey, D.M., Yost, C., Pierce, S.B., Weaver, C., respective dilution rates of the primary antibodies were 1:200 Kimelman, D., 2000. Interaction among GSK-3, GBP, axin, and APC in Xenopus axis specification. J. Cell Biol. 148, 691–702. and 1:1000 for FRAT1 (goat anti-human IgG for FRAT1, Santa Ferkey, D.M., Kimelman, D., 2002. Glycogen synthase kinase-3 beta Cruz Biotechnology, CA, USA) and β-actin (mouse anti-human mutagenesis identifies a common binding domain for GBP and β IgG for -actin, Santa Cruz Biotechnology, CA, USA). After that, Axin. J. Biol. Chem. 277, 16147–16152. the membrane washed with PBS-T and incubated with horse- Fraser, E., Young, N., Dajani, R., Franca-Koh, J., Ryves, J., Williams, radish peroxidase-conjugated secondary antibodies (donkey R.S., Yeo, M., Webster, M.T., Richardson, C., Smalley, M.J., Pearl, 158 BRAIN RESEARCH 1320 (2010) 152– 158

L.H., Harwood, A., Dale, T.C., 2002. Identification of the Axin Saitoh, T., Katoh, M., 2001. FRAT1 and FRAT2, clustered in human and Frat binding region of glycogen synthase kinase-3. J. Biol. chromosome 10q24.1 region, are up-regulated in gastric Chem. 277, 2176–2185. cancer. Int. J. Oncol. 19, 311–315. Freemantle, S.J., Portland, H.B., Ewings, K., Dmitrovsky, F., Saitoh, T., Moriwaki, J., Koike, J., Takagi, A., Miwa, T., Shiokawa, K., DiPetrillo, K., Spinella, M.J., Dmitrovsky, E., 2002. Katoh, M., 2001. Molecular cloning and characterization of Characterization and tissue-specific expression of human FRAT2, encoding a positive regulator of the WNT signaling GSK-3-binding proteins FRAT1 and FRAT2. Gene 291, 17–27. pathway. Biochem. Biophys. Res. Commun. 281, 815–820. Giles, R.H., van Es, J.H., Clevers, H., 2003. Caught up in a Wnt Saitoh, T., Mine, T., Katoh, M., 2002. Molecular cloning and storm: Wnt signaling in cancer. Biochim. Biophys. Acta 1653, expression of proto-oncogene FRAT1 in human cancer. Int. J. 1–24. Oncol. 20, 785–789. Hess, K.R., Broglio, K.R., Bondy, M.L., 2004. Adult glioma incidence Salic, A., Lee, E., Mayer, L., Kirschner, M.W., 2000. Control of trends in the United States, 1977–2000. Cancer 101, 2293–2299. beta-catenin stability: reconstitution of the cytoplasmic steps Ille, F., Sommer, L., 2005. Wnt signaling: multiple functions in of the wnt pathway in Xenopus egg extracts. Mol. Cell 5, 523–532. neural development. Cell. Mol. Life Sci. 62, 1100–1108. Sanson, M., Thillet, J., Hoang-Xuan, K., 2004. Molecular changes in Jonkers, J., Korswagen, H.C., Acton, D., Breuer, M., Berns, A., 1997. gliomas. Curr. Opin. Oncol. 16, 607–613. Activation of a novel proto-oncogene, Frat1, contributes to Sareddy, G.R., Panigrahi, M., Challa, S., Mahadevan, A., Babu, P.P., progression of mouse T-cell lymphomas. EMBO J. 16, 441–450. 2009. Activation of Wnt/beta-catenin/Tcf signaling pathway in Jonkers, J., van Amerongen, R., van der Valk, M., human astrocytomas. Neurochem. Int. 55, 307–317. Robanus-Maandag, E., Molenaar, M., Destree, O., Berns, A., 1999a. Tolwinski, N.S., Wieschaus, E., 2004. Rethinking WNT signaling. In vivo analysis of Frat1 deficiency suggests compensatory Trends Genet. 20, 177–181. activity of Frat3. Mech. Dev. 88, 183–194. van Amerongen, R., Berns, A., 2005. Re-evaluating the role of Frat Jonkers, J., Weening, J.J., van der Valk, M., Bobeldijk, R., Berns, A., in Wnt-signal transduction. Cell Cycle. 4, 1065–1072. 1999b. Overexpression of Frat1 in transgenic mice leads to van Amerongen, R., Nawijn, M., Franca-Koh, J., Zevenhoven, J., van glomerulosclerosis and nephrotic syndrome, and provides der Gulden, H., Jonkers, J., Berns, A., 2005. Frat is dispensable for direct evidence for the involvement of Frat1 in lymphoma canonical Wnt signaling in mammals. Genes Dev. 19, 425–430. progression. Oncogene 18, 5982–5990. van den Bent, M.J., Hegi, M.E., Stupp, R., 2006. Recent Kaba, S.E., Kyritsis, A.P., 1997. Recognition and management of developments in the use of chemotherapy in brain tumours. gliomas. Drugs 53, 235–244. Eur. J. Cancer 42, 582–588. Li, L., Yuan, H., Weaver, C.D., Mao, J., Farr 3rd, G.H., Sussman, D.J., Van der Flier, L.G., Sabates-Bellver, J., Oving, I., Haegebarth, A., De Jonkers, J., Kimelman, D., Wu, D., 1999. Axin and Frat1 interact Palo, M., Anti, M., Van Gijn, M.E., Suijkerbuijk, S., Van de with dvl and GSK, bridging Dvl to GSK in Wnt-mediated Wetering, M., Marra, G., Clevers, H., 2007. The intestinal Wnt/ regulation of LEF-1. EMBO J. 18, 4233–4240. TCF signature. Gastroenterology 132, 628–632. Logan, C.Y., Nusse, R., 2004. The Wnt signaling pathway in Wang, Y., Hewitt, S.M., Liu, S., Zhou, X., Zhu, H., Zhou, C., Zhang, development and disease. Annu. Rev. Cell Dev. Biol. 20, G., Quan, L., Bai, J., Xu, N., 2006. Tissue microarray analysis of 781–810. human FRAT1 expression and its correlation with the Mao, X.G., Zhang, X., Zhen, H.N., 2009. Progress on potential subcellular localisation of beta-catenin in ovarian tumours. Br. strategies to target brain tumor stem cells. Cell. Mol. Neurobiol. J. Cancer 94, 686–691. 29, 141–155. Wang, Y., Liu, S., Zhu, H., Zhang, W., Zhang, G., Zhou, X., Zhou, C., Moon, R.T., Kohn, A.D., De Ferrari, G.V., Kaykas, A., 2004. WNT and Quan, L., Bai, J., Xue, L., Lu, N., Xu, N., 2008. FRAT1 beta-catenin signalling: diseases and therapies. Nat. Rev. overexpression leads to aberrant activation of beta-catenin/ Genet. 5, 691–701. TCF pathway in esophageal squamous cell carcinoma. Int. J. Pelengaris, S., Khan, M., Evan, G., 2002. c-MYC: more than just a Cancer 123, 561–568. matter of life and death. Nat. Rev. Cancer. 2, 764–776. Yost, C., Farr 3rd, G.H., Pierce, S.B., Ferkey, D.M., Chen, M.M., Polakis, P., 2000. Wnt signaling and cancer. Genes Dev. 14, Kimelman, D., 1998. GBP, an inhibitor of GSK-3, is implicated in 1837–1851. Xenopus development and oncogenesis. Cell 93, 1031–1041. Remmele, W., Hildebrand, U., Hienz, H.A., Klein, P.J., Vierbuchen, Yu, J.M., Jun, E.S., Jung, J.S., Suh, S.Y., Han, J.Y., Kim, J.Y., Kim, K.W., M., Behnken, L.J., Heicke, B., Scheidt, E., 1986. Comparative 2007. Role of Wnt5a in the proliferation of human glioblastoma histological, histochemical, immunohistochemical and cells. Cancer Lett. 257, 172–181. biochemical studies on oestrogen receptors, lectin receptors, Zhao, C., Blum, J., Chen, A., Kwon, H.Y., Jung, S.H., Cook, J.M., Lagoo, and Barr bodies in human breast cancer. Virchows Arch. A A., Reya, T., 2007. Loss of beta-catenin impairs the renewal of Pathol. Anat. Histopathol. 409, 127–147. normal and CML stem cells in vivo. Cancer Cell 12, 528–541.