274

BIN1

AMPH2|AMPHL|SH3P9;bridging integrator 1;BIN1;bridging integrator 1

amphiphysin II|amphiphysin-like protein|box dependant MYC interacting protein 1|box-dependent myc-interacting protein 1|myc box-dependent-interacting protein 1

2q14

protein-coding

PURPOSE: Amplification of the MYCN proto-oncogene is strongly correlated with poor outcome in neuroblastoma (NB), although deregulated MYCN is a potent inducer of apoptosis. BIN1 (2q14) encodes multiple isoforms of a Myc-interacting adaptorprotein that has features of a tumor suppressor, including the ability to inhibit Myc-mediated cell transformation and to promote apoptosis. We hypothesized that BIN1 may function as a suppressor gene in NB, because Bin1 is highly expressed in neural tissues and binds the Myc Box motifs that are conserved in MycN. EXPERMENTAL DESIGN: expression of MYCN, total BIN1, and BIN1 isoforms were determined in 56 primary NBs using the real-time PCR. expression was correlated with biological and genetic features. To determine the functional significance of BIN1 expression we ectopically expressed BIN1 isoforms in NB cell lines with andwithout MYCN amplification, and assessed clonogenic growth. RESULTS: Four predominant BIN1 isoforms resulting from alternative splicing of exon 12A (a neural tissue-specific exon) and exon 13 (a Myc-binding domain encoding exon) were variably expressed in the 56 primary NBs. expression of BIN1 was lower in: NBs with MYCN amplification (n = 10) compared with those without, P < 0.03; in International Neuroblastoma Risk Group high-risk NB (n = 19) compared with low- or intermediate-risk NB, P < 0.01; and in metastatic NB (n = 21) compared with localized NB, P < 0.06. BIN1 inactivation by deletion or genomic rearrangement was identified infrequently. Forced expression of BIN1 isoforms containing the Myc-binding domain (with or without exon 12A) inhibited colony formation in NB cell lines with MYCN amplification (P < 0.01) but not in those without. Forced expression of BIN1 isoforms with a MBD deletion did not inhibit colony formationin any cell line assessed. CONCLUSIONS: These data support that reduced BIN1 expression contributes to the malignant phenotype of childhood NB. As we reported previously, BIN1 may function to circumvent MycN-mediated apoptosis in NBs with MYCN amplification.

Emerging evidence demonstrates that subunits of the SWI/SNF chromatin remodelingcomplex are specifically mutated at high frequency in a variety of human cancer types. SNF5 (SMARCB1/INI1/BAF47), a core subunit of the SWI/SNF complex, is inactivated in the vast majority of rhabdoid tumors (RT), an aggressive type of pediatric cancer. SNF5-deficient cancers are diploid and genomically stable, suggesting that epigenetically based changes in transcription are key drivers oftumor formation caused by SNF5 loss. However, there is limited understanding of the target genes that drive cancer formation following SNF5 loss. Here we performed comparative expression analyses upon three independent SNF5-deficient cancer data sets from both human and mouse and identify downregulation of the BIN1 tumor suppressor as a conserved event in primary SNF5-deficient cancers. Weshow that SNF5 recruits the SWI/SNF complex to the BIN1 promoter, and that the marked reduction of BIN1 expression in RT correlates with decreased SWI/SNF occupancy. Functionally, we demonstrate that re-expression of BIN1 specifically compromises the proliferation of SNF5-deficient RT cell lines. Identification ofBIN1 as a SNF5 target gene reveals a novel tumor suppressive regulatory mechanism whose disruption can drive cancer formation.