A protein involved in minichromosome maintenance in yeast binds a transcriptional enhancer conserved in eukaryotes

ACR1, a yeast ATF/CREB repressor

Members of the mammalian ATF/CREB family of transcription factors, which are associated with regulation by cyclic AMP and viral oncogenes, bind common DNA sequences (consensus TGACGTCA) via a bZIP domain. In the yeast Saccharomyces cerevisiae, ATF/CREB-like sequences confer either repression or activation of transcription, depending on the promoter context. By isolating mutations that alleviate the repression mediated by ATF/CREB sites, we define a new yeast gene, ACR1, which encodes an ATF/CREB transcriptional repressor. ACR1 contains a bZIP domain that is necessary for homodimer formation and specific DNA binding to an ATF/CREB site. Within the bZIP domain, ACR1 most strongly resembles the mammalian cyclic AMP-responsive transcriptional regulators CREB and CREM; it is less similar to GCN4 and YAP1, two previously described yeast bZIP transcriptional activators that recognize the related AP-1 sequence (consensus TGACTCA). Interestingly, deletion of the ACR1 gene causes increased transcription through ATF/CREB sites that does not depend on GCN4 or YAP1. Moreover, extracts from acr1 deletion strains contain one or more ATF/CREB-like DNA-binding activities. These genetic and biochemical observations suggest that S. cerevisiae contains a family of ATF/CREB proteins that function as transcriptional repressors or activators.

CDC46/MCM5, a yeast protein whose subcellular localization is cell cycle-regulated, is involved in DNA replication at autonomously replicating sequences

Saccharomyces cerevisiae cells containing mutations in the cell-division-cycle gene CDC46 arrest with a large bud and a single nucleus with unreplicated DNA at the non-permissive temperature. This G1/S arrest, together with the increased rates of mitotic chromosome loss and recombination phenotype, suggests that these mutants are defective in DNA replication. The subcellular localization of the CDC46 protein changes with the cell cycle; it is nuclear between the end of M phase and the G1/S transition but is cytoplasmic in other phases of the cell cycle. Here we show that CDC46 is identical to MCM5, based on complementation analysis of the mcm5-1 and cdc46-1 alleles, complementation of the minichromosome maintenance defect of mcm5-1 by CDC46, and the genetic linkage of these two genes. Like mcm5-1, cdc46-1 and cdc46-5 also show a minichromosome maintenance defect thought to be associated with DNA replication initiation at autonomously replicating sequences. Taken together, these observations suggest that CDC46/MCM5 acts during a very narrow window at the G1/S transition or the beginning of S phase by virtue of its nuclear localization to effect the initiation of DNA replication at autonomously replicating sequences.

HAP1 positive control mutants specific for one of two binding sites

The expression of the yeast CYC1 and CYC7 genes is controlled by the HAP1 activator. A GAL4-like zinc finger (residues 1-148) specifies binding to the dissimilar sites UAS1 (of CYC1) and CYC7, and an acidic domain (residues 1307-1483) is essential for activation of transcription. To analyze how HAP1 binds to UAS1 and CYC7, we performed saturation mutagenesis of the DNA-binding domain and recovered mutants with altered activity. Class 1 mutants had a reduced activity at both UAS1 and CYC7, and class 2 mutants selectively eliminated activity at CYC7. Surprisingly, several mutants of both classes exhibited wild-type DNA binding, indicating that they were specifically defective in activation. These positive control (PC) mutants alter residues that bracket the zinc finger. We explain these mutants in a model involving cofactor proteins that bind UAS1 and CYC7 along with HAP1. The existence of PC mutants that only affect activity at CYC7 raises the possibility that different cofactors may exist for UAS1 and CYC7.

Chromosome loss, hyperrecombination, and cell cycle arrest in a yeast mcm1 mutant

The original mcm1-1 mutant was identified by its inability to propagate minichromosomes in an ARS-specific manner, suggesting that it is defective in the initiation of DNA synthesis at ARSs. This mutant is also defective in expression of alpha-mating-type-specific genes. Further genetic and biochemical studies confirmed that Mcm1 is a transcription factor that mediates the transcriptional regulation of a number of genes, including genes outside of the mating type complement, by interacting with different cofactors. Although MCM1 is an essential gene, none of the previously characterized mcm1 mutants exhibits significant growth defects. To assess which of the many roles of Mcm1 is essential for growth, we constructed and characterized a temperature-sensitive conditional mutant of mcm1, mcm1-110L. This mutant exhibits a temperature-dependent cell-cycle arrest, with a large, elongated bud and a single, undivided nucleus that has a DNA content of close to 2n. In addition, it shows elevated levels of chromosome loss and recombination. In spite of the severity of the mcm1-110L mutation, this mutant still retains an ARS-specific pattern of minichromosome instability. All of these phenotypes are precisely those exhibited by mutants in three MCM genes, MCM2, MCM3, and MCM5/CDC46, that have been shown to play interacting roles in the early steps of DNA replication.

Human myocyte-specific enhancer factor 2 comprises a group of tissue-restricted MADS box transcription factors

The MEF2 site is an essential element of muscle enhancers and promoters that is bound by a nuclear activity found, so far, only in muscle and required for tissue-specific transcription. We have cloned a group of transcription factors from human muscle that are responsible for this activity: They are present in muscle-specific DNA-binding complexes, have a target sequence specificity identical to that of the endogenous activity, and are MEF2 site-dependent transcriptional activators. These MEF2 proteins comprise several alternatively spliced isoforms from one gene and a related factor encoded by a second gene. All share a conserved amino-terminal DNA-binding domain that includes the MADS homology. MEF2 transcripts are ubiquitous but accumulate preferentially in skeletal muscle, heart, and brain. Specific alternatively spliced isoforms are restricted to these tissues, correlating exactly with the presence of endogenous MEF2 activity. Furthermore, MEF2 protein is detected only in skeletal and cardiac muscle nuclei and not in myoblast and nonmuscle cells. Thus, post-transcriptional regulation is important in the generation of tissue-specific MEF2 activity. Cardiac and smooth, as well as skeletal, muscles contain functionally saturating levels of MEF2 trans-activating factors that are absent in nonmuscle cells. Moreover, MEF2 is induced in nonmuscle cells by MyoD; however, MEF2 alone is insufficient to produce the full muscle phenotype. Implications for the molecular mechanisms of myogenesis are considered.

Human and Drosophila homeodomain proteins that enhance the DNA-binding activity of serum response factor

Cells with distinct developmental histories can respond differentially to identical signals, suggesting that signals are interpreted in a fashion that reflects a cell's identity. How this might occur is suggested by the observation that proteins of the homeodomain family, including a newly identified human protein, enhance the DNA-binding activity of serum response factor, a protein required for the induction of genes by growth and differentiation factors. Interaction with proteins of the serum response factor family may allow homeodomain proteins to specify the transcriptional response to inductive signals. Moreover, because the ability to enhance the binding of serum response factor to DNA residues within the homeodomain but is independent of homeodomain DNA-binding activity, this additional activity of the homeodomain may account for some of specificity of action of homeodomain proteins in development.

The N-terminal 96 residues of MCM1, a regulator of cell type-specific genes in Saccharomyces cerevisiae, are sufficient for DNA binding, transcription activation, and interaction with alpha 1

MCM1 performs several functions necessary for its role in regulating cell type-specific gene expression in the yeast Saccharomyces cerevisiae: DNA binding, transcription activation, and interaction with coregulatory proteins such as alpha 1. We analyzed a set of MCM1 deletion derivatives using in vivo reporter gene assays and in vitro DNA-binding studies to determine which regions of MCM1 are important for its various activities. We also analyzed a set of LexA-MCM1 hybrids to examine the ability of different segments of MCM1 to activate transcription independent of MCM1's DNA-binding function. The first third of MCM1 [MCM1(1-96)], which includes an 80-residue segment homologous to the mammalian serum response factor, was sufficient for high-affinity DNA binding, for activation of reporter gene expression, and for interaction with alpha 1 in vitro and in vivo. However, the ability of MCM1(1-96) to activate transcription and to interact with alpha 1 was somewhat reduced compared with wild-type MCM1 [MCM1(1-286)]. Optimal interaction with alpha 1 required residues 99 to 117, in which 18 of 19 amino acids are acidic in character. Optimal transcription activation required a segment from residues 188 to 286, in which 50% of the amino acids are glutamine. Deletion of this segment from MCM1 reduced expression of reporter genes by about twofold. Moreover, LexA-MCM1 hybrids containing this segment were able to activate expression of reporter genes that rely on LexA binding sites as potential upstream activation sequences. Thus, glutamine-rich regions may contribute to the activation function of yeast transcription activators, as has been suggested for glutamine-rich mammalian proteins such as Sp1.

The yeast alpha 1 and MCM1 proteins bind a single strand of their duplex DNA recognition site

The yeast cell type regulator alpha 1 cooperates with a constitutive factor, MCM1 protein, to recognize the promoter and activate transcription of several alpha-specific genes. I show here that the alpha 1 and MCM1 proteins bind specifically to one of the two strands of their recognition sequence. This single-strand-binding activity shares several characteristics with the duplex-binding properties of these proteins: (i) the MCM1 protein binds alone to single-stranded and duplex sequences of both the alpha-specific (P'Q) and a-specific (P) binding sites; (ii) the alpha 1 protein requires both the MCM1 protein and the Q sequence to bind either single-stranded or duplex DNA; (iii) the alpha 1 protein stimulates binding of the MCM1 protein to both single-stranded and duplex DNAs; and (iv) the affinities of the proteins for single-stranded and duplex DNAs are comparable.

The N-terminal 96 residues of MCM1, a regulator of cell type-specific genes in Saccharomyces cerevisiae, are sufficient for DNA binding, transcription activation, and interaction with alpha 1

MCM1 performs several functions necessary for its role in regulating cell type-specific gene expression in the yeast Saccharomyces cerevisiae: DNA binding, transcription activation, and interaction with coregulatory proteins such as alpha 1. We analyzed a set of MCM1 deletion derivatives using in vivo reporter gene assays and in vitro DNA-binding studies to determine which regions of MCM1 are important for its various activities. We also analyzed a set of LexA-MCM1 hybrids to examine the ability of different segments of MCM1 to activate transcription independent of MCM1's DNA-binding function. The first third of MCM1 [MCM1(1-96)], which includes an 80-residue segment homologous to the mammalian serum response factor, was sufficient for high-affinity DNA binding, for activation of reporter gene expression, and for interaction with alpha 1 in vitro and in vivo. However, the ability of MCM1(1-96) to activate transcription and to interact with alpha 1 was somewhat reduced compared with wild-type MCM1 [MCM1(1-286)]. Optimal interaction with alpha 1 required residues 99 to 117, in which 18 of 19 amino acids are acidic in character. Optimal transcription activation required a segment from residues 188 to 286, in which 50% of the amino acids are glutamine. Deletion of this segment from MCM1 reduced expression of reporter genes by about twofold. Moreover, LexA-MCM1 hybrids containing this segment were able to activate expression of reporter genes that rely on LexA binding sites as potential upstream activation sequences. Thus, glutamine-rich regions may contribute to the activation function of yeast transcription activators, as has been suggested for glutamine-rich mammalian proteins such as Sp1.

Global regulators of chromosome function in yeast

In the yeast Saccharomyces cerevisiae, several abundant, sequence-specific DNA binding proteins are involved in multiple aspects of chromosome function. In addition to functioning as transcriptional activators of a large number of yeast genes, they are also involved in transcriptional silencing, the initiation of DNA replication, centromere function and regulation of telomere length. This review will consider each of these proteins, focusing on what is known about the mechanisms of their multiple functions.

The yeast alpha 1 and MCM1 proteins bind a single strand of their duplex DNA recognition site

The yeast cell type regulator alpha 1 cooperates with a constitutive factor, MCM1 protein, to recognize the promoter and activate transcription of several alpha-specific genes. I show here that the alpha 1 and MCM1 proteins bind specifically to one of the two strands of their recognition sequence. This single-strand-binding activity shares several characteristics with the duplex-binding properties of these proteins: (i) the MCM1 protein binds alone to single-stranded and duplex sequences of both the alpha-specific (P'Q) and a-specific (P) binding sites; (ii) the alpha 1 protein requires both the MCM1 protein and the Q sequence to bind either single-stranded or duplex DNA; (iii) the alpha 1 protein stimulates binding of the MCM1 protein to both single-stranded and duplex DNAs; and (iv) the affinities of the proteins for single-stranded and duplex DNAs are comparable.

SRF and MCM1 have related but distinct DNA binding specificities

The mammalian transcription factor SRF and the yeast regulatory protein MCM1 contain DNA binding domains that are 70% identical; moreover, both proteins can bind the serum response element in the human c-fos promoter. Here we present an analysis of MCM1 sequence specificity by selection of sites from random sequence oligonucleotides. In this assay the MCM1 DNA binding domain selects binding sites containing the consensus (NotC)CCY(A/T)(A/T)(T/A)NN(A/G)G, distinct from the SRF binding consensus CC(A/T)6GG. Carboxylethylation interference analysis of a set of selected sites suggests that MCM1 contacts DNA in its major groove throughout one helical turn. These differences in specificity are largely due to sequence differences between the N terminal basic parts of the SRF and MCM1 DNA binding domains. Comparison of the relative binding affinities of MCM1 and SRF for a panel of representative binding sites showed that many high affinity MCM1 sites have negligible affinity for SRF and vice versa. Thus MCM1 and SRF have significantly different sequence specificities.

Expression of Xenopus laevis histone H5 gene in yeast

As an approach to assess linker histone function, we engineered a cDNA encoding Xenopus laevis histone H5 (XLH5), into the yeast Saccharomyces cerevisiae, which lacks any known proteins homologous to linker histones. XLH5 cDNA when fused to the yeast GAL10 promoter and 5' untranslated region (UTR) was shown to be accurately transcribed at relatively high levels in cells harvested at mid to late log after exposure to at least 22 mM galactose. The resultant 0.95 kb XLH5 transcript reached steady state levels by approx. 2 h after galactose induction. In contrast, the product, detected by anti-XLH5 antibody, was not stably expressed until 4 h or more after induction, when no apparent growth takes place. The expression product was 27% smaller than native H5 and may have been proteolytically processed. Constitutive transcription and loss of XLH5 expression product occurred using a plasmid construct containing a 275 bp fragment of the pBR322 tetr gene inserted downstream of the GAL10 promoter. This fragment carries a putative yeast cell-type-specific upstream activation sequence.

The mcm2 mutation of yeast affects replication, rather than segregation or amplification of the two micron plasmid

We have studied the maintenance of the endogenous two micron (2 mu) plasmid in a strain of yeast carrying the nuclear mutation mcm2. This mutation, earlier shown to affect the maintenance of yeast minichromosomes in an ARS-dependent manner, also affected the copy number of the 2 mu plasmid. The effect was more pronounced at 35 degrees C leading to the elimination of the plasmid from the cells cultured at this temperature. The mutant cells could be efficiently cured of the circle by transformation with 2 mu ORI-carrying hybrid vectors, an observation consistent with the low copy number of the endogenous plasmid. A chromosomal revertant of this mutant for another ARS(ARS1) was found also to confer stability on the 2 mu ORI-carrying minichromosomes and had elevated levels of the endogenous plasmid. The mutation neither affected the segregation nor the amplification process mediated by site-specific recombination at FRT sites requiring the FLP gene-encoded protein action. ARS131C, an ARS that was unaffected in the mutant at 25 degrees C, could elevate the copy number of a 2 mu hybrid vector in the mutant cells. In view of these results, some aspects of segregation and copy number control of the endogeneous plasmid have been discussed. We propose that the mutation impairs the 2 mu ORI function, leading to its loss.

The SRF and MCM1 transcription factors

The mammalian transcription factor SRF (serum-response factor) and the related Saccharomyces cerevisiae transcription factor MCM1 are the prototypes of a new class of dimeric DNA-binding proteins. Their function is regulated in part by the interactions of their DNA-binding domains with accessory proteins. Recent work has advanced the functional characterization of the contributions of SRF and MCM1, and their accessory proteins to transcriptional activation.

Stable nucleosome positioning and complete repression by the yeast alpha 2 repressor are disrupted by amino-terminal mutations in histone H4

Nucleosomes are positioned in the presence of the yeast repressor alpha 2 in minichromosomes containing the alpha 2 operator and on the promoters of a-cell-specific genes regulated by alpha 2. To investigate the possibility that alpha 2 directs nucleosome position through an interaction with a component of the core particle, we analyzed chromatin structures adjacent to the operator in alpha cells containing mutations in the amino-terminal region of histone H4. Deletion or point mutation of specific amino acids in histone H4 altered the location and/or stability of nucleosomes adjacent to the alpha 2 operator. These changes in chromatin structure were accompanied by partial derepression of a beta-galactosidase reporter construct under alpha 2 control, even though alpha 2 remained bound to its operator sequence. Our data suggest that complete repression by alpha 2 requires stable positioning of nucleosomes in promoter regions and this positioning involves the conserved amino-terminal region of histone H4.

Ssn6-Tup1 is a general repressor of transcription in yeast

The homeodomain protein alpha 2 and the SRF-like protein Mcm1 are required to establish cell type in the yeast Saccharomyces cerevisiae. Together, these regulatory proteins recognize a specific DNA operator, marking a set of genes for transcriptional repression. In this paper, we show that occupancy of the operator by alpha 2-Mcm1 is not sufficient to bring about repression. Rather, repression is effected only when Ssn6 (a TPR protein) and Tup1 (a beta-transducin repeat protein) are also present in the cell. We show that Ssn6 represses transcription when brought to a promoter by a bacterial DNA-binding domain and that Tup1 is required for this repression. Based on these and other results, we propose that Ssn6-Tup1 is a general repressor of transcription in yeast, recruited to target promoters by a variety of sequence-specific DNA-binding proteins.

Characterization of SAP-1, a protein recruited by serum response factor to the c-fos serum response element

We used a yeast genetic screen to isolate cDNAs that encode a protein, SRF accessory protein-1 (SAP-1), that is recruited to the c-fos serum response element (SRE) as part of a ternary complex that includes serum response factor (SRF). SAP-1 requires DNA-bound SRF for ternary complex formation and makes extensive DNA contacts to the 5' side of SRF, but does not bind DNA autonomously. Ternary complex formation by SAP-1 requires only the DNA-binding domain of SRF, which can be replaced by that of the related yeast protein MCM1. We isolated cDNAs encoding two forms of SAP-1 protein, SAP-1a and SAP-1b, which differ at their C termini. Both SAP-1 proteins contain three regions of striking homology with the elk-1 protein, including an N-terminal ets domain. Ternary complex formation by SAP-1 requires both the ets domain and a second conserved region 50 amino acids to its C-terminal side. SAP-1 has similar DNA binding properties to the previously characterized HeLa cell protein p62/TCF.

RPD1 (SIN3/UME4) is required for maximal activation and repression of diverse yeast genes

We show that the extent of transcriptional regulation of many, apparently unrelated, genes in Saccharomyces cerevisiae is dependent on RPD1 (and RPD3 [M. Vidal and R. F. Gaber, Mol. Cell. Biol. 11:6317-6327, 1991]). Genes regulated by stimuli as diverse as external signals (PHO5), cell differentiation processes (SPO11 and SPO13), cell type (RME1, FUS1, HO, TY2, STE6, STE3, and BAR1), and genes whose regulatory signals remain unknown (TRK2) depend on RPD1 to achieve maximal states of transcriptional regulation. RPD1 enhances both positive and negative regulation of these genes: in rpd1 delta mutants, higher levels of expression are observed under repression conditions and lower levels are observed under activation conditions. We show that several independent genetic screens, designed to identify yeast transcriptional regulators, have detected the RPD1 locus (also known as SIN3, SD11, and UME4). The inferred RPD1 protein contains four regions predicted to take on helix-loop-helix-like secondary structures and three regions (acidic, glutamine rich, and proline rich) reminiscent of the activating domains of transcriptional activators.

Both activation and repression of a-mating-type-specific genes in yeast require transcription factor Mcm1

Mcm1 is a yeast transcription factor with homologs throughout the metazoa. MCM1 was first identified as a gene involved in maintenance of artificial minichromosomes in yeast. More recently Mcm1 has been shown to serve as a transcriptional regulator of mating-type-specific genes. Biochemical data suggest that Mcm1 coactivates alpha-specific genes and corepresses a-specific genes by binding to a 10-base-pair dyad symmetry element in their upstream regions. We reported previously that an mcm1 point mutation reduced activation of alpha-specific genes but had little effect on the expression of a-specific genes. We now show that another mcm1 allele, which depletes the Mcm1 protein, affects both activation and repression of a-specific genes. The mutant strain remains capable of high levels of pheromone induction of a-specific genes, although with retarded kinetics. Mcm1 joins an increasing number of transcription factors involved in both positive and negative regulation of gene expression.

RPD1 (SIN3/UME4) is required for maximal activation and repression of diverse yeast genes

We show that the extent of transcriptional regulation of many, apparently unrelated, genes in Saccharomyces cerevisiae is dependent on RPD1 (and RPD3 [M. Vidal and R. F. Gaber, Mol. Cell. Biol. 11:6317-6327, 1991]). Genes regulated by stimuli as diverse as external signals (PHO5), cell differentiation processes (SPO11 and SPO13), cell type (RME1, FUS1, HO, TY2, STE6, STE3, and BAR1), and genes whose regulatory signals remain unknown (TRK2) depend on RPD1 to achieve maximal states of transcriptional regulation. RPD1 enhances both positive and negative regulation of these genes: in rpd1 delta mutants, higher levels of expression are observed under repression conditions and lower levels are observed under activation conditions. We show that several independent genetic screens, designed to identify yeast transcriptional regulators, have detected the RPD1 locus (also known as SIN3, SD11, and UME4). The inferred RPD1 protein contains four regions predicted to take on helix-loop-helix-like secondary structures and three regions (acidic, glutamine rich, and proline rich) reminiscent of the activating domains of transcriptional activators.

Cell-specific regulation of oncogene-responsive sequences of the c-fos promoter

We have identified oncogene-responsive sequences in the human c-fos promoter that mediate induction of transcription by several nonnuclear oncoproteins and the tumor promoter TPA. These sequences are regulated in a cell-specific manner. (i) In NIH 3T3 cells, the CArG box of the c-fos promoter is sufficient to mediate activation by oncogenes. (ii) In contrast, in HeLa cells, additional flanking sequences are also required, including the outer arm of the serum response element and the FAP site. We also show that the serum response factor, which binds to the CArG box, activates transcription in vivo in NIH 3T3 cells but not in HeLa cells. Finally, we present evidence that the intracellular level of the c-Fos protein could be a major determinant of cell-specific regulation of these oncogene-responsive elements of the c-fos promoter.

MAT alpha 1 can mediate gene activation by a-mating factor

In the yeast Saccharomyces cerevisiae, expression of alpha-specific genes is governed by the MAT alpha 1 and MCM1 gene products. MAT alpha 1 and MCM1 bind cooperatively to PQ elements upstream of alpha-specific genes. The PQ element not only directs alpha-specific expression but can also direct gene induction in response to treatment with a-mating pheromone. We have used gene fusions to investigate whether induction conferred by the PQ box is mediated through either MAT alpha 1 or MCM1, or a combination of both. When MCM1 is fused to the DNA-binding domain of the bacterial repressor LexA, this fusion protein is capable of trans-activating a lacZ reporter gene driven by a LexA operator. However, the transcriptional activity of the MCM1-LexA fusion is not further enhanced by treatment of cells with a-factor. A MAT alpha 1-LexA fusion protein is also capable of trans-activation through a LexA operator. Moreover, the activity of the MAT alpha 1-LexA fusion protein can be further induced by treatment with a-factor. When progressive deletions are made from the amino terminus of MAT alpha 1 in the fusion protein, the basal level of trans-activation progressively decreases, but the inducibility of the fusion protein increases. MAT alpha 1-LexA fusion proteins, which have greater than or equal to 57 amino acids deleted from the amino terminus of MAT alpha 1 are not capable of trans-activation. In addition, the activity of the MAT alpha 1-LexA fusion protein is dependent on the functions of the STE7, STE11, and STE12 genes that encode components of the pheromone response pathway.

Participation of ABF-1 protein in expression of the Saccharomyces cerevisiae CAR1 gene

DNA fragments previously shown to be required for expression of the CAR1 (arginase) gene in Saccharomyces cerevisiae and to support transcriptional activation of a reporter gene in a heterologous expression vector were shown to bind purified regulatory protein ABF-1. Two ABF-1 sites were identified in the CAR1 upstream region, one to which ABF-1 protein bound with high affinity and a second to which it bound much less avidly. The higher-affinity ABF-1 binding site upstream of CAR1 was an effective competitor of the HMRE, ARS1 B domain, and COR2-GFI binding sequences for protein binding. Point mutations in the CAR1 high-affinity ABF-1 binding site resulted in a 12-fold loss of transcriptional activation of a reporter gene compared with the wild-type CAR1 DNA fragment. These data are consistent with the suggestion that ABF-1 protein is one of the transcription factors involved in expression of the CAR1 gene.

Cell-specific regulation of oncogene-responsive sequences of the c-fos promoter

We have identified oncogene-responsive sequences in the human c-fos promoter that mediate induction of transcription by several nonnuclear oncoproteins and the tumor promoter TPA. These sequences are regulated in a cell-specific manner. (i) In NIH 3T3 cells, the CArG box of the c-fos promoter is sufficient to mediate activation by oncogenes. (ii) In contrast, in HeLa cells, additional flanking sequences are also required, including the outer arm of the serum response element and the FAP site. We also show that the serum response factor, which binds to the CArG box, activates transcription in vivo in NIH 3T3 cells but not in HeLa cells. Finally, we present evidence that the intracellular level of the c-Fos protein could be a major determinant of cell-specific regulation of these oncogene-responsive elements of the c-fos promoter.

The serum response factor is extensively modified by phosphorylation following its synthesis in serum-stimulated fibroblasts

Growth factor regulation of c-fos proto-oncogene transcription is mediated by a 20-bp region of dyad symmetry, termed the serum response element. The inner core of this element binds a 67-kDa phosphoprotein, the serum response factor (SRF), that is thought to play a pivotal role in the c-fos transcriptional response. To investigate the mechanism by which SRF regulates c-fos expression, we generated polyclonal anti-SRF antibodies and used these antibodies to analyze the biochemical properties of SRF. These studies indicate that the synthesis of SRF is transient, occurring within 30 min to 4 h after serum stimulation of quiescent fibroblasts. Newly synthesized SRF is transported to the nucleus, where it is increasingly modified by phosphorylation during progression through the cell cycle. Within 2 h of serum stimulation, differentially modified forms of SRF can be distinguished on the basis of the ability to bind a synthetic serum response element. SRF protein exhibits a half-life of greater than 12 h and is predominantly nuclear, with no change occurring in its localization upon serum stimulation. We find that the induction of SRF synthesis is regulated at the transcriptional level and that cytoplasmic SRF mRNA is transiently expressed with somewhat delayed kinetics compared with c-fos mRNA expression. These features of SRF expression suggest a model whereby newly synthesized SRF functions in the shutoff of c-fos transcription.

The serum response factor is extensively modified by phosphorylation following its synthesis in serum-stimulated fibroblasts

Growth factor regulation of c-fos proto-oncogene transcription is mediated by a 20-bp region of dyad symmetry, termed the serum response element. The inner core of this element binds a 67-kDa phosphoprotein, the serum response factor (SRF), that is thought to play a pivotal role in the c-fos transcriptional response. To investigate the mechanism by which SRF regulates c-fos expression, we generated polyclonal anti-SRF antibodies and used these antibodies to analyze the biochemical properties of SRF. These studies indicate that the synthesis of SRF is transient, occurring within 30 min to 4 h after serum stimulation of quiescent fibroblasts. Newly synthesized SRF is transported to the nucleus, where it is increasingly modified by phosphorylation during progression through the cell cycle. Within 2 h of serum stimulation, differentially modified forms of SRF can be distinguished on the basis of the ability to bind a synthetic serum response element. SRF protein exhibits a half-life of greater than 12 h and is predominantly nuclear, with no change occurring in its localization upon serum stimulation. We find that the induction of SRF synthesis is regulated at the transcriptional level and that cytoplasmic SRF mRNA is transiently expressed with somewhat delayed kinetics compared with c-fos mRNA expression. These features of SRF expression suggest a model whereby newly synthesized SRF functions in the shutoff of c-fos transcription.

Molecular characterization of two stamen-specific genes, tap1 and fil1, that are expressed in the wild type, but not in the deficiens mutant of Antirrhinum majus

Deficiens, a homeotic gene involved in the genetic control of flower development, codes for a putative transcription factor. Upon mutation of the gene, petals are transformed to sepals and stamens to carpels, indicating that deficiens is essential for the activation of genes required for petal and stamen formation. In a search for putative target genes of deficiens, several stamen- and petal-specific genes were cloned that are expressed in wild type but not in the deficiensglobifera mutant. In this report the molecular characterization of two of these genes, tap1 and fil1, is presented. They are transiently expressed during flower development. In situ hybridization data demonstrate that tap1 is expressed in the tapetum of the anthers and fil1 in the filament of the stamen and at the bases of the petals. Both genes encode small proteins with N-terminal hydrophobic domains suggesting that they are secreted. We discuss possible functions of the gene products and their relationship to the deficiens gene.

Striking similarities between the regulatory mechanisms governing yeast mating-type genes and mammalian major histocompatibility complex genes

Expression of a mammalian major histocompatibility complex (MHC) class I gene is in part regulated by a silencer DNA sequence element which binds a complex of silencer factors. This negative regulatory system is shown to be strikingly similar to the yeast alpha 2 mating-type repression system. A moderate DNA sequence homology exists between the MHC class I silencer DNA element and the yeast alpha 2 operator. Mammalian silencer factors specifically bind to the yeast alpha 2 operator DNA and also specifically interact with a yeast alpha 2-binding protein. Furthermore, the alpha 2 operator functions as a silencer element in mammalian cells when placed upstream of a MHC class I promoter.

Striking similarities between the regulatory mechanisms governing yeast mating-type genes and mammalian major histocompatibility complex genes

Expression of a mammalian major histocompatibility complex (MHC) class I gene is in part regulated by a silencer DNA sequence element which binds a complex of silencer factors. This negative regulatory system is shown to be strikingly similar to the yeast alpha 2 mating-type repression system. A moderate DNA sequence homology exists between the MHC class I silencer DNA element and the yeast alpha 2 operator. Mammalian silencer factors specifically bind to the yeast alpha 2 operator DNA and also specifically interact with a yeast alpha 2-binding protein. Furthermore, the alpha 2 operator functions as a silencer element in mammalian cells when placed upstream of a MHC class I promoter.

Pheromone response elements are necessary and sufficient for basal and pheromone-induced transcription of the FUS1 gene of Saccharomyces cerevisiae

The FUS1 gene of Saccharomyces cerevisiae is transcribed in a and alpha cells, not in a/alpha diploids, and its transcription increases dramatically when haploid cells are exposed to the appropriate mating pheromone. In addition, FUS1 transcription is absolutely dependent on STE4, STE5, STE7, STE11, and STE12, genes thought to encode components of the pheromone response pathway. We now have determined that the pheromone response element (PRE), which occurs in four copies within the FUS1 upstream region, functions as the FUS1 upstream activation sequence (UAS) and is responsible for all known aspects of FUS1 regulation. In particular, deletion of 55 bp that includes the PREs abolished all transcription, and a 139-bp fragment that includes the PREs conferred FUS1-like expression to a CYC1-lacZ reporter gene. Moreover, three or four copies of a synthetic PRE closely mimicked the activity conferred by the 139-bp fragment, and even a single copy of PRE conferred a trace of activity that was haploid specific and pheromone inducible. In the FUS1 promoter context, four copies of the synthetic PRE inserted at the site of the 55-bp deletion restored full FUS1 transcription. Sequences upstream and downstream from the PRE cluster were important for maximal PRE-directed expression but, by themselves, did not have UAS activity. Other yeast genes with PREs, e.g., STE2 and BAR1, are more modestly inducible and have additional UAS elements contributing to the overall activity. In the FUS1 promoter, the PREs apparently act alone to confer activity that is highly stimulated by pheromone.

Relative contributions of MCM1 and STE12 to transcriptional activation of a- and alpha-specific genes from Saccharomyces cerevisiae

We have examined the relative contributions of MCM1 and STE12 to the transcription of the a-specific STE2 gene by using a 367 bp fragment from the STE2 5'-noncoding region to drive expression of a reporter lacZ gene. Mutation of the MCM1 binding site destroyed MCM1.alpha 2-mediated repression in alpha cells and dramatically reduced expression in a cells. The residual expression was highly stimulated by exposure of cells to pheromone. Likewise, the loss of STE12 function reduced lacZ expression driven by the wild-type STE2 fragment. In the absence of both MCM1 and STE12 functions, no residual expression was observed. Thus, the STE2 fragment appears to contain two distinct upstream activation sequences (UASs), one that is responsible for the majority of expression in cells not stimulated by pheromone, and one that is responsible for increased expression upon pheromone stimulation. In further support of this idea, a chemically synthesized version of the STE2 MCM1 binding site had UAS activity, but the activity was neither stimulated by pheromone nor reduced in ste12 mutants. Although transcription of alpha-specific genes also requires both MCM1 and STE12, these genes differ from a-specific genes in that they have a single, MCM1-dependent UAS system. The activity of the minimal 26 bp UAS from the alpha-specific STE3 gene was both stimulated by pheromone and reduced in ste12 mutants. These data suggest that at alpha-specific genes STE12 and MCM1 exert their effects through a single UAS.

Pheromone response elements are necessary and sufficient for basal and pheromone-induced transcription of the FUS1 gene of Saccharomyces cerevisiae

The FUS1 gene of Saccharomyces cerevisiae is transcribed in a and alpha cells, not in a/alpha diploids, and its transcription increases dramatically when haploid cells are exposed to the appropriate mating pheromone. In addition, FUS1 transcription is absolutely dependent on STE4, STE5, STE7, STE11, and STE12, genes thought to encode components of the pheromone response pathway. We now have determined that the pheromone response element (PRE), which occurs in four copies within the FUS1 upstream region, functions as the FUS1 upstream activation sequence (UAS) and is responsible for all known aspects of FUS1 regulation. In particular, deletion of 55 bp that includes the PREs abolished all transcription, and a 139-bp fragment that includes the PREs conferred FUS1-like expression to a CYC1-lacZ reporter gene. Moreover, three or four copies of a synthetic PRE closely mimicked the activity conferred by the 139-bp fragment, and even a single copy of PRE conferred a trace of activity that was haploid specific and pheromone inducible. In the FUS1 promoter context, four copies of the synthetic PRE inserted at the site of the 55-bp deletion restored full FUS1 transcription. Sequences upstream and downstream from the PRE cluster were important for maximal PRE-directed expression but, by themselves, did not have UAS activity. Other yeast genes with PREs, e.g., STE2 and BAR1, are more modestly inducible and have additional UAS elements contributing to the overall activity. In the FUS1 promoter, the PREs apparently act alone to confer activity that is highly stimulated by pheromone.

Mcm2 and Mcm3, two proteins important for ARS activity, are related in structure and function

MCM2 and MCM3 are essential genes believed to play important roles in the initiation of DNA replication in Saccharomyces cerevisiae. Mutants defective in Mcm2 or Mcm3 are remarkably similar in phenotype. They both show an autonomously replicating sequence (ARS)-specific minichromosome maintenance defect, although their ARS specificities are not identical. In addition, these mutants exhibit a premitotic cell cycle arrest and an increase in chromosome loss and recombination. Genetic studies suggest that the two MCM gene products play interacting or complementary roles in DNA replication. Double mutants of mcm2-1 and mcm3-1 are inviable at the permissive growth temperature (23 degrees C) for each of the single mutants. Furthermore, overproduction of Mcm3 accentuates the deleterious effect of the mcm2-1 mutation, whereas overproduction of Mcm2 partially complements the mcm3-1 mutation. MCM2 encodes a protein of 890 amino acids containing a putative zinc-finger domain that is essential for Mcm2 function. Mcm2 shows striking homology to Mcm3 and three other proteins, Cdc46 of S. cerevisiae, and Nda4 and Cdc21 of Schizosaccharomyces pombe. The phenotypes of mutants defective in these proteins suggest that they belong to a protein family involved in the early steps of DNA replication.

Differential modulation of yeast actin, tubulin, and YPT1 mRNA levels by cycloheximide

We have examined the effects of cycloheximide (Chx) on transcription of genes encoding the yeast beta-actin (ACT), beta-tubulin (TUB) and yeast protein 1 (YPT1). As in mammalian cells, Chx caused an increase in levels of ACT, but not TUB, transcripts. The YPT1 gene was also activated. Induction of Chx was further studied by placing the promoter regions of the ACT and YPT1 genes in front of a globin (GLB)-encoding reporter gene (GLB) on yeast plasmids. Induction of GLB mRNA synthesis by Chx was not seen with either promoter; the YPT1 promoter was, however, strongly inducible by Chx if glucose was present. The YPT1 coding sequence conferred Chx inducibility on both the YPT1 and ACT promoters, suggesting that it may contain a transcription regulatory element.

Functional domains of the yeast transcription/replication factor MCM1

MCM1 is an essential yeast DNA-binding protein that affects both minichromosome maintenance, in a manner suggesting that it has DNA replication initiation function, and gene expression. It activates alpha-specific genes together with MAT alpha 1, and represses a-specific genes together with MAT alpha 2. Alone, MCM1 can activate transcription. To determine whether different domains of the protein mediate these diverse functions, we constructed and analyzed several mcm1 mutants. The gene expression and minichromosome maintenance phenotypes of these mutants suggest that the role of MCM1 in DNA replication initiation may not involve transcriptional activation. However, both transcription and replication activities require only the 80-amino-acid fragment of MCM1 homologous to the DNA-binding domain of the serum response factor (SRF). This small fragment is also sufficient for cell viability and repression of a-specific genes. A polyacidic amino acid stretch immediately adjacent to the SRF homologous domain of MCM1 was found to be important for activation of alpha-specific genes in alpha cells. Mutants lacking the acidic stretch confer higher expression from an alpha-specific UAS in a cells in addition to lower expression in alpha cells, suggesting that negative regulation at this site occurs in a cells, in addition to the well-documented positive regulation in alpha cells.

RAP1 protein activates and silences transcription of mating-type genes in yeast

RAP1 is a sequence-specific DNA-binding protein essential for cell growth. The occurrence of RAP1-binding sites in many promoter regions, the mating-type gene silencer elements, and telomeres suggests that RAP1 has multiple functions in the cell. To assess its role in transcription, temperature-sensitive mutations in RAP1 were generated. Analysis of rap1ts strains provides evidence that RAP1 functions in both transcriptional activation and silencing of mating-type genes. Several observations indicate that rap1ts strains are defective in the expression of MAT alpha, whose upstream activation sequence (UAS) contains a RAP1-binding site. At nonpermissive temperatures, decreases in MAT alpha steady-state transcript levels can be detected in MAT alpha rap1ts strains. Furthermore, these strains are deficient in alpha-pheromone production and simultaneously express at least two alpha-specific genes. These phenotypes can be reversed by replacing the RAP1-binding site at MAT alpha with a binding site for the GAL4 transcriptional activator. Certain rap1ts alleles have an opposite effect on the silent mating-type locus HMR, which becomes partially derepressed at nonpermissive temperatures.

Morphogenetic and molecular correlates of teratogenesis in the amphibian embryo

In attempting to develop a system to study molecular mechanisms of teratogenesis, examination of the effects of a teratogen (dimethyl sulfoxide) on both molecular and morphological aspects of embryonic development in the amphibian Xenopus laevis has been conducted. Characteristic morphological effects, which occur during the period from 7 to 16 hours after fertilization (i.e., gastrulation) are noted. Delays in gastrulation are accompanied by changes in the regulation of transcription of several genes known to be active during gastrulation in normal development. Later morphological effects are also observed, and these probably arise as a consequence of the changes occurring during gastrulation. Thus, molecular responses to a teratogen have been detected, and a correlation between molecular and morphological responses to a teratogen is observed. These findings represent the first demonstration of the effects of a teratogen on the transcription of specific genes, and invite speculation that one or more molecular events mediate teratogenesis. They further suggest that the amphibian system may be useful for studying early molecular responses to teratogens.

The Schizosaccharomyces pombe homolog of Saccharomyces cerevisiae HAP2 reveals selective and stringent conservation of the small essential core protein domain

The fission yeast Schizosaccharomyces pombe is immensely diverged from budding yeast (Saccharomyces cerevisiae) on an evolutionary time scale. We have used a fission yeast library to clone a homolog of S. cerevisiae HAP2, which along with HAP3 and HAP4 forms a transcriptional activation complex that binds to the CCAAT box. The S. pombe homolog php2 (S. pombe HAP2) was obtained by functional complementation in an S. cerevisiae hap2 mutant and retains the ability to associate with HAP3 and HAP4. We have previously demonstrated that the HAP2 subunit of the CCAAT-binding transcriptional activation complex from S. cerevisiae contains a 65-amino-acid "essential core" structure that is divisible into subunit association and DNA recognition domains. Here we show that Php2 contains a 60-amino-acid block that is 82% identical to this core. The remainder of the 334-amino-acid protein is completely without homology to HAP2. The function of php2 in S. pombe was investigated by disrupting the gene. Strikingly, like HAP2 in S. cerevisiae, the S. pombe gene is specifically involved in mitochondrial function. This contrasts to the situation in mammals, in which the homologous CCAAT-binding complex is a global transcriptional activator.

A cDNA encoding a human CCAAT-binding protein cloned by functional complementation in yeast

We constructed a comprehensive cDNA library from HeLa cell mRNA in a vector that directs expression of the cDNA in Saccharomyces cerevisiae. We used this library to clone the human counterpart of the Sa. cerevisiae CCAAT-binding transcription factor, Hap2, by functional complementation of a hap2 mutation. The cDNA encoding the human Hap2 homolog encodes a protein of 257 amino acids that has a 62-amino acid carboxyl-terminal region 73% identical to the essential core region of Hap2. The amino terminus of the protein is highly enriched in glutamine residues, reminiscent of transcriptional activation domains of several other mammalian transcription factors. Analysis of human Hap2 expression reveals three major transcripts: a 4.1-kilobase species found in all cell types examined, a 7.0-kilobase species specific to B lymphocytes, and a 1.6-kilobase species that is expressed preferentially in HeLa cells and that likely corresponds to our cDNA clone. Thus, the human Hap2 homolog and related factors may play both a constitutive and cell type-specific role in gene expression. The general approach of cloning by complementation should allow the isolation of many human genes for which corresponding yeast mutations exist.

A chicken beta-actin gene can complement a disruption of the Saccharomyces cerevisiae ACT1 gene

Recently it was demonstrated that beta-actin can be produced in Saccharomyces cerevisiae by using the expression plasmid pY beta actin (R. Karlsson, Gene 68:249-258, 1988), and several site-specific mutants are now being produced in a protein engineering study. To establish a system with which recombinant actin mutants can be tested in vivo and thus enable a correlation to be made with functional effects observed in vitro, a yeast strain lacking endogenous yeast actin and expressing exclusively beta-actin was constructed. This strain is viable but has an altered morphology and a slow-growth phenotype and is temperature sensitive to the point of lethality at 37 degrees C.

The Schizosaccharomyces pombe homolog of Saccharomyces cerevisiae HAP2 reveals selective and stringent conservation of the small essential core protein domain

The fission yeast Schizosaccharomyces pombe is immensely diverged from budding yeast (Saccharomyces cerevisiae) on an evolutionary time scale. We have used a fission yeast library to clone a homolog of S. cerevisiae HAP2, which along with HAP3 and HAP4 forms a transcriptional activation complex that binds to the CCAAT box. The S. pombe homolog php2 (S. pombe HAP2) was obtained by functional complementation in an S. cerevisiae hap2 mutant and retains the ability to associate with HAP3 and HAP4. We have previously demonstrated that the HAP2 subunit of the CCAAT-binding transcriptional activation complex from S. cerevisiae contains a 65-amino-acid "essential core" structure that is divisible into subunit association and DNA recognition domains. Here we show that Php2 contains a 60-amino-acid block that is 82% identical to this core. The remainder of the 334-amino-acid protein is completely without homology to HAP2. The function of php2 in S. pombe was investigated by disrupting the gene. Strikingly, like HAP2 in S. cerevisiae, the S. pombe gene is specifically involved in mitochondrial function. This contrasts to the situation in mammals, in which the homologous CCAAT-binding complex is a global transcriptional activator.

Distinct protein targets for signals acting at the c-fos serum response element

The c-fos serum response element (SRE) is a primary nuclear target for intracellular signal transduction pathways triggered by growth factors. It is the target for both protein kinase C (PKC)-dependent and -independent signals. Function of the SRE requires binding of a cellular protein, termed serum response factor (SRF). A second protein, p62TCF, recognizes the SRE-SRF complex to form a ternary complex. A mutated SRE that bound SRF but failed to form the ternary complex selectively lost response to PKC activators, but retained response to PKC-independent signals. Thus, two different signaling pathways act through discrete nuclear targets at the SRE. At least one of these pathways functions by recruitment of a pathway-specific accessory factor (p62TCF). These results offer a molecular mechanism to account for the biological specificity of signals that appear to act through common DNA sequence elements.

The PRE and PQ box are functionally distinct yeast pheromone response elements

Saccharomyces cerevisiae mating pheromones function by binding to cell surface receptors and activating signal transduction processes which regulate gene expression. In this report, we have analyzed the minimum sequence requirements for conferring both a and alpha mating pheromone inducibilities onto a heterologous promoter. Here we show that the repetitive pheromone response element (PRE) which binds to STE12 protein is sufficient to confer pheromone responsiveness only when present in multiple copies. Moreover, by itself, it is preferentially responsive to alpha factor in a cells. In contrast, a single copy of the PQ box of the STE3 upstream activation sequence (UAS) is sufficient to confer a-factor responsiveness in alpha cells. The PQ box binds both MCM1 and MAT alpha 1 in a cooperative manner, and neither the P nor Q site alone is sufficient to confer a-factor responsiveness. In a cells, however, even multiple copies of the PQ box fail to confer alpha-factor responsiveness. Therefore, the PRE and the PQ box are functionally distinct pheromone-responsive elements with opposite cell type specificities. Moreover, these results indicate that the MCM1 protein functions in a signal transduction pathway in a manner analogous to that of its mammalian homolog, the serum response factor, which regulates the expression of the c-fos proto-oncogene in mammals.

A chicken beta-actin gene can complement a disruption of the Saccharomyces cerevisiae ACT1 gene

Recently it was demonstrated that beta-actin can be produced in Saccharomyces cerevisiae by using the expression plasmid pY beta actin (R. Karlsson, Gene 68:249-258, 1988), and several site-specific mutants are now being produced in a protein engineering study. To establish a system with which recombinant actin mutants can be tested in vivo and thus enable a correlation to be made with functional effects observed in vitro, a yeast strain lacking endogenous yeast actin and expressing exclusively beta-actin was constructed. This strain is viable but has an altered morphology and a slow-growth phenotype and is temperature sensitive to the point of lethality at 37 degrees C.