Search

The zinc finger protein MAZ, originally identified as a factor that binds to the c-myc P2 promoter, is associated with transcriptional termination. As shown in these studies, a termination sequence between the closely spaced human complement genes C2 and Factor B contains a protein binding site which interacts with three different proteins in vitro. Binding of one of these factors, MAZ, correlates with activity of the C2 termination sequence in vivo. Cloned MAZ was used to obtain a consensus binding site, G5AG5. This allowed identification of new sites, between the closely spaced human genes g11 and C4 and within an intron of the mouse IgM-D gene, where termination is known to occur and regulate the expression of IgD. The g11 and IgM MAZ sites lie within sequences that have activity in a termination assay and, furthermore, mutation of C2 or g11 MAZ sites severely reduces termination activity. MAZ bends DNA, and inherently bent DNA is highly active as a terminator, suggesting that MAZ-induced bending is important for C2 and g11 termination. We propose that MAZ sites exist in promoters which require protection against transcriptional interference, such as those of closely spaced genes, to cause efficient termination. The MAZ consensus sequence will facilitate the identification of further sites.

We have investigated the nature of regulatory sequences within the vicinity of the murine c-myc locus by analyzing the expression of myc-chloramphenicol acetyl transferase (CAT) vectors transfected into a human lymphoblastoid cell line (BJAB) and a monkey fibroblast line (COS). CAT enzymatic assays and S1 nuclease protection experiments reveal that a negative element resides 428-1188 bp 5' of the first c-myc promoter, P1. This 760-bp segment of 5'-flanking c-myc DNA dramatically inhibits CAT gene expression in the pSV2CAT vector when placed in either orientation approximately 1.7 kb 3' (and approximately 3.2 kb 5' on the circular plasmid) from the SV40 promoter region. By employing this strategy, we were unable to identify an analogous DNA segment that is closer to or within the first c-myc exon. We propose that this 5' c-myc region be termed a 'dehancer' since this negative element has the opposite properties of a transcriptional enhancer.

Five novel murine plasma cell (PC) tumors with chromosome translocations 350-500 bp 5' of the first c-myc exon are described. The t(12;15)s of TEPC 1194, ABPC 33 and TEPC 1165 position the intact c-myc locus 5' of the Cu, C gamma 2a and C alpha IgCH genes respectively. In ABPC 17, the IgH enhancer element and adjacent switch (Su) sequences were found 5' of the first c-myc exon while this enhancer is associated with the reciprocal products of the TEPC 1194, ABPC 33 and TEPC 1033 translocations. Quantitative S1 nuclease analyses demonstrate that the ratios of transcription from the two c-myc promoters (P1 and P2) are increased 4-to 7-fold in these five tumors. With the exception of TEPC 1165, (which contains a small deletion in exon 1), such increases in P1:P2 ratios appear to be manifested by a reduction in P2 usage in comparison to other tumors without such promoter shifts. A survey of 27 additional PC and non-PC B lymphoid tumors and cell lines revealed that myc promoter shifts of this magnitude are unique to PC tumors with 5'-proximal translocations. We propose that (i) these clustered breakpoints identify a normal c-myc regulatory element located at least 350 bp 5' of c-myc exon 1; (ii) the loss or disruption of this cis-acting upstream element and the linkage of c-myc to the IgCH locus would result in abnormal expression of this oncogene in these as well as most other PC tumors.

We have employed a retroviral vector, ZN(Smu/S gamma 2b)tk1, as a substrate for detecting the presence of immunoglobulin heavy chain constant region (CH) gene switch (S) recombination activity in murine pre-B cells. ZN(Smu/S gamma 2b)tk1 contains a neomycin (neo) resistance gene in addition to the herpes simplex virus thymidine kinase (Htk) gene which is positioned between murine Smu and S gamma 2b sequences. Stable acquisition of the ZN(Smu/S gamma 2b)tk1 vector was selected in G-418 and switch region recombination within these proviruses was selected by resistance to the drug bromodeoxyuridine (BUdR). Fluctuation analyses of ZN(Smu/S gamma 2b)tk1 infected 18-8tk- and 38B9tk- pre-B lines revealed Htk gene inactivations with apparent frequencies of 5 X 10(-5) and 1 X 10(-5) events/cell/generation, respectively, while G-418 resistant Ltk- fibroblasts lost the HTK phenotype at an apparent rate of 4 X 10(-8). Southern blot analysis demonstrated that switch recombination caused the deletion of the Htk gene in all pre-B clones examined while the loss of Htk in Ltk- clones was not mediated by S region recombination. In 21 out of 24 pre-B clones, the recombinations involved the tandemly repetitive portions of the Smu and S gamma 2b sequences. These results demonstrate that the CH gene S region segments inserted into ZN(Smu/S gamma 2b)tk1 are sufficient for B-cell-specific recombination/deletion within the S region tandem repeats.

We have identified a positive modulator within the c-myc first exon downstream of the gene's transcription initiation sites, P1 and P2. We introduced myc-CAT (chloramphenicol acetyltransferase) hybrid genes into three cell lines (BJAB, COS and HeLa) and measured their expression by either CAT enzymatic activity, S1 nuclease protection or by a nuclear 'run-on' transcription assay. Removal of 46 bp from the 3' end of the first exon results in a decrease of myc-CAT expression and P2 activity. A 438-bp exon 1 segment, lacking the normal myc promoters, efficiently drives the expression of SV40 early promoters. We find that this first exon segment efficiently functions as a positive modulator only in its sense orientation, 3' of a nearby promoter. The positive effects of the myc first exon and the SV40 enhancer are complementary.

We present a detailed analysis of strand-specific transcription in different regions of the murine c-myc locus. In normal and transformed cell lines, RNA polymerase II directed transcription occurs in the sense and anti-sense direction. Three noncontiguous regions show a high level of transcription in the anti-sense orientation: upstream of the first exon, within the first intron and in the 3' part of the gene (intron 2 and exon 3). In a cell line carrying a c-myc amplification (54c12), anti-sense transcription is not uniformly increased throughout the locus and is differentially affected by inhibition of protein synthesis. These results suggest that anti-sense transcription in various parts of the locus is independently regulated. In the sense orientation, transcriptional activity is higher in the first exon than in the rest of the gene indicating that transcription pauses near the 3' end of the first exon. The extent of this intragenic pausing varies among different cell lines and is most severe in cells with a c-myc amplification. Transcription initiation and pausing are both negatively regulated by labile proteins.

To assess possible alterations of c-myc transcriptional control in murine B-cell tumors, we have investigated the pattern of DNaseI hypersensitive sites in the gene's putative regulatory region and within the gene in a variety of genomic contexts. A number of such sites were found in several cell types, but none of these was detectable in a gene which was shown to be transcriptionally silent by the criterion of elongation of nascent transcripts in isolated nuclei. These results differ from those of a previous study, in which a DNaseI-hypersensitive site approximately 2 kb upstream of the gene was proposed to be associated with negative regulation of c-myc transcription in human cells. An analysis of DNA sequences presented here reveals that this region is highly homologous between mouse and human, suggesting that these upstream hypersensitive sites do not reflect species-specific regulatory elements. We also present data indicating that this hypersensitive site distinguishes the c-myc alleles in translocation-positive plasma cell tumors which lack c-myc rearrangement. Furthermore, we report the existence of hypersensitive sites within the gene. One of these appears to be associated with cryptic promoters that are employed only when the normal promoters are lost as a consequence of chromosome translocation. These results are discussed in the context of c-myc translocation and gene breakage and with respect to possible stage-specific regulation of the gene's transcriptional competence.