A mutational analysis of the insulin gene transcription control region: expression in beta cells is dependent on two related sequences within the enhancer.

Abstract

Cell-specific expression of the insulin gene is controlled by cis-acting DNA sequences located within "-350 base pairs of the 5' flanking DNA immediately upstrVam from the transcription start site. Using synthetic oligonucleotides, we have constructed a systematic series of block replacement mutants spanning this region. No single sequence appears to be absolutely required for expression. However, three of the mutants exhibit 5-10 times less activity and several others show 2-3 times less. Simultaneous mutation of twQ of the most mutationally sensitive regions leads to virtual abolition of activity. These two elements are structurally related and presumably represent key components of the machinery determining the cell-specific expression of the insulin gene.The development of complex multicellular organisms requires selective spatial and temporal regulation of the genetic repertoire. Although cell-specific gene expression in higher eukaryotes can be controlled at multiple levels (1), a primary determinant is the efficiency of initiation of new transcripts (2). Specific transcription is apparently controlled by the interaction of cellular factors with certain cis-acting DNA sequences in the vicinity of the gene.Cis-acting elements involved in controlling cell-specific expression have been identified in several cellular genes. In the insulin gene, two distinct positive control elements have been identified in the 5' flanking DNA (3): a cell-specific enhancer and a cell-specific promoter (4). Additional negatively acting cis elements have also been detected in this region (5). In immunoglobulin genes, cell specificity is also attributable to cell-specific enhancers (6-9) and cell-specific promoters (10-12). A further measure of control may be exerted in this system at the post-transcriptional level (1,12).In vitro binding studies of transcriptional control regions reveal multiple sites for binding of sequence-specific protein factors (13). Cell-specific transcription may be the consequence of the interaction of specific DNA-binding protein factors, which are present at elevated levels in cells actively expressing the specific gene (14-16).To more completely define the elements of the insulin gene that control transcription in vivo and to correlate in vitro protein binding to transcriptional activity, we have constructed a systematic series of block replacement mutants spanning the proximal 345 base pairs (bp) of flanking DNA by using an oligonucleotide-directed approach that permits mutation of a region without changing distances or the stereospecific alignment between transcriptionally important elements. Each mutant contained a block of noncomplementary transversions (i.e., A + C, G ++ T; see ref. 17) spanning -'10 bp. None of these mutations leads to complete loss of activity. However, several produce dramatic reductions (by a factor of 5-10): the "TATA" region of the promoter and two short sequences within the enhancer, which are similar to each other and are also related to the transcriptional control...

Pdx1 and BETA2/NeuroD1 Participate in a Transcriptional Complex That Mediates Short-range DNA Looping at the Insulin Gene

“…6B shows that mutation of the A3 element completely abrogated loop formation. These results are consistent with studies emphasizing the particular importance of the A3 element in Ins gene transactivation (27).…”

Mouse MafA, homologue of zebrafish somite Maf 1, contributes to the specific transcriptional activity through the insulin promoter

“…Other important elements are known as E1 and A3 [40][41][42][43][44][45]. The pancreatic islet restricted transcription factors BETA2/NeuroD and PDX1, which bind to the E1 and A3 elements, have been isolated.…”

An endocrine-specific element is an integral component of an exocrine-specific pancreatic enhancer.

“…A B-like sequence adjacent to an enhancer core (GTGGAAAA; Weiher et al 19831 is conserved among the human, two mouse, and two rat insulin genes. The B-like and core elements are contained within the 48-bp insulin gene E 1 regulatory domain, which binds one or more islet-specific nuclear factors {Ohlsson and Edlund 1986; Philippe et al 1988) and is necessary for optimal expression in cultured insulinoma cells (Karlsson et al 1987) and in transgenic mice (Fromont-Racine et al 1990). The regulatory region of the glucagon gene contains a potential B element that overlaps the distal part of the 40-bp G1 region, which is critical for specific expression of the glucagon gene (Philippe et al 1988).…”