Rearrangement of the transcription factor gene CHOP in myxoid liposarcomas with t(12;16)(q13;p11)

Abstract
Most myxoid liposarcomas (MLS) are characterized cytogenetically by a t(12; 16)(q13;p11). It is reasonable to assume that this translocation corresponds to the consistent rearrangement of one or two genes in 12q 13 and/or 16p11, and that the loci thus affected are important in the normal control of fat cell differentiation and proliferation. We have used Southern blot technique to test whether a gene of the CCAAT/enhancer binding protein (C/EBP) family, CHOP, which maps to 12q13 and is assumed to be involved in adipocyte differentiation, could be the 12q gene in question. Using a cDNA probe that spans the CHOP coding region, we detected one rearranged and one wild type allele in nine of nine MLS with t(l2;l6). Using PCR generated, site-specific probes corresponding to the non-coding exons 1 and 2 and intron 2 of CHOP, rearrangements in five of seven tumors mapped to the 2.4 and 1.6 kbp Pstl fragments that contain the first two exons and introns of the gene and the upstream promoter region. In contrast to the findings in MLS, no tumor without a t(12;16) exhibited aberrant CHOP restriction digest patterns. These tumors included one highly differentiated liposarcoma with abnormal karyotype but no involvement of 12q 13, seven lipomas with various cytogenetic aberrations of 12q 13–15, two uterine leiomyomas with t( 12; 14) (q 14–15;q23–24), and one hemangiopericytoma and one chondroma, both of which also had 12q 13 changes. These findings demonstrate that the 12q breakpoint of the t( 12;16) in MLS differs from those in the other tumors investigated, even in cases with no cytogenetically visible differences in breakpoint position, that CHOP rearrangement is specific for MLS, and that the breakpoints cluster to the 5′ region of the gene. We assume that the CHOP alteration is an important step in MLS tumorigenesis and speculate that this is achieved by interference with the normal function of the C/EBP family of transcription factors.