Tupl and Ssn6 transcriptionally repress a wide variety of genes in yeast but do not appear to bind DNA. We provide genetic and biochemical evidence that the DNA-binding protein a2, a regulator of cell-type-specific genes, recruits the Tupl/Ssn6 repressor by directly interacting with Tupl. This interaction is mediated by a region of Tupl containing seven copies of the WD repeat, a 40 amino acid motif of unknown function found in many other proteins. We have found that a single WD repeat will interact with a2, indicating that the WD repeat is a protein-protein interaction domain. Furthermore, a fragment of Tupl containing primarily WD repeats provides at least partial repression in the absence of Ssn6, suggesting that the repeats also mediate interaction between Tupl and other components of the repression machinery.

No abstract

The tetratricopeptide repeat (TPR) is a 34-amino-acid degenerate sequence motif that is found in a large variety of proteins, both prokaryotic and eukaryotic. TPRs are usually found in tandem arrays of up to 16 copies. In this paper we identify a direct interaction between the TPRs of Ssn6, a general transcriptional repressor, and a2, a cell-type regulator in Saccharomyces cerevisiae. Five of the Ssn6 TPRs were tested individually, and all were found to interact specifically with a2. These results suggest a model for TPR-protein interactions and for the role that a tandem array of TPRs may have in mediating transcriptional repression.

To bring about repression of a family of genes in Saccharomyces cerevisiae called the a-specific genes, two transcriptional regulatory proteins, a2 and GRM (general regulator of mating type), bind cooperatively to an operator found upstream of each a-specffic gene. To date, GRM has been defined only biochemically. In this communication we show that the product of a single yeast gene (MCMI) is sufficient to bind cooperatively with oa2 to the operator. We also show that antiserum raised against the MCM1 gene product recognizes GRM from yeast cells. These results, in combination with previous observations, provide strong evidence that MCM1 encodes the GRM activity.The expression of many eucaryotic genes appears to be controlled by combinations of regulatory proteins (for a recent review, see reference 2). An example of this phenomenon is found in the repression of a group of cell-typespecific genes (called the a-specific genes) in the a cell type of the yeast Saccharomyces cerevisiae. This repression requires the activity of two proteins that bind cooperatively to an operator located upstream of each a-specific gene. Either protein alone can recognize the operator, but proteinprotein interactions between the two couple their affinities for the DNA and ensure tight binding by the complex. The proteins are a2, a cell-type-specific protein, and GRM, a non-cell-type-specific protein. GRM has been defined only in terms of biochemical function, as follows: GRM is an activity that is (i) capable of binding alone to the a-specific gene operator and (ii) able to bind the operator cooperatively with a2 (4).Considerable evidence suggests that GRM is encoded by the MCMJ gene. The MCMJ gene was originally identified by the isolation of a mutation, mcml-l, that rendered strains unable to maintain certain plasmids (hence its name, from minichromosome maintenance) (6; G. R. Maine, Ph.D. thesis, Cornell University, 1984). However, a cells that carried the mcml-l allele exhibited an additional phenotype; they mated poorly (9), a property expected for a cell deficient in GRM. This connection was strengthened by the observation that GRM-binding activity in extracts derived from mcml-l strains was lower than that in extracts from wild-type strains (4). Recently, the MCMJ gene product (produced in Escherichia coli) has been shown to meet the first criterion for GRM activity, namely, it recognizes the a-specific gene operator (8

Proliferation and phenotypic modulation of smooth muscle cells (SMCs) are major components of the vessel's response to injury in experimental models of restenosis. Some of the growth factors involved in restenosis have been identified, but to date little is known about the transcription factors that ultimately regulate this process. We examined the expression of the four members of the myocyte enhancer binding factor-2 (MEF2) family of transcription factors in cultured rat aortic SMCs (RASMCs) and a rat model of restenosis because of their known importance in regulating the differentiated phenotype of skeletal and cardiac muscle. In skeletal and cardiac muscle, the MEF2s are believed to be important for activating the expression of contractile protein and other muscle-specific genes. Therefore, we anticipated that the MEF2s would be expressed at high levels in medial SMCs that are producing contractile proteins and that they would be downregulated along with the contractile protein genes in neointimal SMCs. On the contrary, we observe that MEF2A, MEF2B, and MEF2D mRNAs are upregulated in the neointima, with the highest levels in the layer of cells nearest to the lumen, whereas MEF2C mRNA levels do not appreciably increase. Moreover, few cells in the media are making MEF2 proteins detectable by immunohistochemistry, whereas large numbers of neointimal cells are positive for all four MEF2s. These data suggest that the MEF2s are involved in the activated smooth muscle phenotype and not in the maintenance of contractile protein gene expression.