A regulatory shortcut between the Snf1 protein kinase and RNA polymerase II holoenzyme.

RNA polymerase II holoenzymes respond to activators and repressors that are regulated by signaling pathways. Here we present evidence for a "shortcut" mechanism in which the Snf1 protein kinase of the glucose signaling pathway directly regulates transcription by the yeast holoenzyme. In response to glucose limitation, the Snf1 kinase stimulates transcription by holoenzyme that has been artificially recruited to a reporter by a LexA fusion to a holoenzyme component. We show that Snf1 interacts physically with the Srb/mediator proteins of the holoenzyme in both two-hybrid and coimmunoprecipitation assays. We also show that a catalytically hyperactive Snf1, when bound to a promoter as a LexA fusion protein, activates transcription in a glucose-regulated manner; moreover, this activation depends on the integrity of the Srb/mediator complex. These results suggest that direct regulatory interactions between signal transduction pathways and RNA polymerase II holoenzyme provide a mechanism for transcriptional control in response to important signals.

Direct interaction between Rsc6 and Rsc8/Swh3,two proteins that are conserved in SWI/SNF-related complexes.

The RSC complex of Saccharomyces cerevisiae is closely related to the SWI/SNF complex. Both complexes are involved in remodeling chromatin structure and they share conserved components. The RSC proteins Sth1, Rsc8/Swh3, Sfh1 and Rsc6 are homologs of the SWI/SNF proteins Swi2/Snf2, Swi3, Snf5 and Swp73 respectively. To investigate the RSC complex, we isolated a temperature-sensitive swh3 allele. A screen for multicopy suppressors yielded plasmids carrying the RSC6 and MAK31 loci. RSC6 also suppressed the formamide sensitivity of a strain with a C-terminal truncation of SWH3 . We show that Swh3 and Rsc6 fusion proteins interact in the two-hybrid system and that the swh3-ts mutation impairs this interaction. Finally, bacterially produced Swh3 and Rsc6 fusion proteins interact in vitro , supporting the genetic evidence for direct interaction between Swh3 and Rsc6 in vivo . We have previously shown that Swh3 also interacts with Sth1. These findings, together with the conservation of these proteins in the SWI/SNF complex and in mammalian SWI/SNF-related complexes, strongly suggest that these proteins form a structural core for the complex.

Interaction of a Swi3 homolog with Sth1 provides evidence for a Swi/Snf-related complex with an essential function in Saccharomyces cerevisiae.

The Saccharomyces cerevisiae Swi/Snf complex has a role in remodeling chromatin structure to facilitate transcriptional activation. The complex has 11 components, including Swi1/Adr6, Swi2/Snf2, Swi3, Snf5, Snf6, Snf11, Swp73/Snf12, and Tfg3. Mammalian homologs of these proteins have been shown to form multiple Swi/Snf-related complexes. Here we characterize an S. cerevisiae Swi3 homolog (Swh3) and present evidence that it associates in a complex with a Snf2 homolog, Sthl. We identified Swh3 as a protein that interacts with the N terminus of Snf2 in the two-hybrid system. Swh3 and Swi3 are functionally distinct, and overexpression of one does not compensate for loss of the other. Swh3 is essential for viability and does not activate transcription of reporters. The Snf2 sequence that interacts with Swh3 was mapped to a region conserved in Sth1. We show that Swh3 and Sth1 fusion proteins interact in the two-hybrid system and coimmunoprecipitate from yeast cell extracts. We also map interactions between Swh3 and Sth1 and examine the role of a leucine zipper motif in self-association of Swh3. These findings, together with previous analysis of Sth1, indicate that Swh3 and Sth1 are associated in a complex that is functionally distinct from the Swi/Snf complex and essential for viability.

SSN genes that affect transcriptional repression in Saccharomyces cerevisiae encode SIN4, ROX3, and SRB proteins associated with RNA polymerase II.

The RNA polymerase II of Saccharomyces cerevisiae exists in holoenzyme forms containing a complex, known as the mediator, associated with the carboxyl-terminal domain. The mediator includes several SRB proteins and is required for transcriptional activation. Previous work showed that a cyclin-dependent kinase-cyclin pair encoded by SSN3 and SSN8, two members of the SSN suppressor family, are identical to two SRB proteins in the mediator. Here we have identified the remaining SSN genes by cloning and genetic analysis. SSN2 and SSN5 are identical to SRB9 and SRB8, respectively, which encode additional components of the mediator. Genetic evidence implicates the SSN genes in transcriptional repression. Thus, these identities provide genetic insight into mediator and carboxyl-terminal domain function, strongly suggesting a role in mediating transcriptional repression as well as activation. We also show that SSN4 and SSN7 are the same as SIN4 and ROX3, respectively, raising the possibility that these genes also encode mediator proteins.

SNF11, a new component of the yeast SNF-SWI complex that interacts with a conserved region of SNF2.

The yeast SNF-SWI complex is required for transcriptional activation of diverse genes and has been shown to alter chromatin structure. The complex has at least 10 components, including SNF2/SWI2, SNF5, SNF6, SWI1/ADR6, and SWI3, and has been widely conserved in eukaryotes. Here we report the characterization of a new component. We identified proteins that interact in the two-hybrid system with the N-terminal region of SNF2, preceding the ATPase domain. In addition to SWI3, we recovered a new 19-kDa protein, designated SNF11. Like other SNF/SWI proteins, SNF11 functions as a transcriptional activator in genetic assays. SNF11 interacts with SNF2 in vitro and copurifies with the SNF-SWI complex from yeast cells. Using a specific antibody, we showed that SNF11 coimmunoprecipitates with members of the SNF-SWI complex and that SNF11 is tightly and stoichiometrically associated with the complex. Furthermore, SNF11 was detected in purified SNF-SWI complex by staining with Coomassie blue dye; its presence previously went unrecognized because it does not stain with silver. SNF11 interacts with a 40-residue sequence of SNF2 that is highly conserved, suggesting that SNF11 homologs exist in other organisms.

The yeast SNF2/SWI2 protein has DNA-stimulated ATPase activity required for transcriptional activation.

The yeast SNF2 (SWI2) protein functions with SNF5, SNF6, SWI1, and SWI3 in the transcriptional activation of many differently regulated genes. These proteins appear to facilitate activation by gene-specific regulatory proteins. SNF2 is highly conserved among eukaryotes and defines a family of proteins with similarity to helicases and nucleic acid-dependent NTPases. Here, we present genetic and biochemical evidence that SNF2 has DNA-stimulated ATPase activity. Mutations in the nucleoside triphosphate (NTP)-binding motif and other conserved motifs impair SNF2 function. Swapping experiments with another member of this family indicate that the helicase-related domains are functionally interchangeable. Finally, bacterially expressed SNF2 protein has ATPase activity that is stimulated by double-stranded DNA, and mutation of the NTP-binding site abolishes this activity. Deletion analysis shows that the helicase-like region of SNF2 is necessary, but not sufficient, for transcriptional activation.

RPC53 encodes a subunit of Saccharomyces cerevisiae RNA polymerase C (III) whose inactivation leads to a predominantly G1 arrest.

RPC53 is shown to be an essential gene encoding the C53 subunit specifically associated with yeast RNA polymerase C (III). Temperature-sensitive rpc53 mutants were generated and showed a rapid inhibition of tRNA synthesis after transfer to the restrictive temperature. Unexpectedly, the rpc53 mutants preferentially arrested their cell division in the G1 phase as large, round, unbudded cells. The RPC53 DNA sequence is predicted to code for a hydrophilic M(r)-46,916 protein enriched in charged amino acid residues. The carboxy-terminal 136 amino acids of C53 are significantly similar (25% identical amino acid residues) to the same region of the human BN51 protein. The BN51 cDNA was originally isolated by its ability to complement a temperature-sensitive hamster cell mutant that undergoes a G1 cell division arrest, as is true for the rpc53 mutants.

Determination of lysine residues affinity labeled in the active site of yeast RNA polymerase II(B) by mutagenesis.

In a previous study, yeast RNA polymerase II(B) was affinity labeled with two nucleotide derivatives (III and VIII) (1). In both cases, the labeled site was localized to the C-terminal part of the B150 subunit. The potential target lysyl residues of derivative III were mapped to the conserved domain H, between Asn946 and Met999. In the present work, we have mutagenized to arginine the five lysines present in domain H. Three lysines can be replaced, individually or simultaneously, without affecting cell growth, and each mutated enzyme can still be affinity labeled. Hence one or both of the other two lysyl residues, Lys979 and Lys987, is the target of the affinity reagent. These two lysines were each found to be essential for cell viability. Derivative VIII labeled another domain in addition to domain H. Supported by analogous results obtained for E. coli RNA polymerase using derivative VIII (2), we hypothesized that the second domain labeled by this derivative in the B150 subunit was domain I. Mutagenesis of the unique lysine present in domain I demonstrated that Lys 1102 was the target of derivative VIII. These results indicate that in both prokaryotic and eukaryotic RNA polymerases, domains H and I are in close proximity and participate to the active site.

Effect of mutations in a zinc-binding domain of yeast RNA polymerase C (III) on enzyme function and subunit association.

The conserved amino-terminal region of the largest subunit of yeast RNA polymerase C is capable of binding zinc ions in vitro. By oligonucleotide-directed mutagenesis, we show that the putative zinc-binding motif CX2CX6-12CXGHXGX24-37CX2C, present in the largest subunit of all eukaryotic and archaebacterial RNA polymerases, is essential for the function of RNA polymerase C. All mutations in the invariant cysteine and histidine residues conferred a lethal phenotype. We also obtained two conditional thermosensitive mutants affecting this region. One of these produced a form of RNA polymerase C which was thermosensitive and unstable in vitro. This instability was correlated with the loss of three of the subunits which are specific to RNA polymerase C: C82, C34, and C31.

RPC10 encodes a new mini subunit shared by yeast nuclear RNA polymerases.

Yeast RNA polymerases A, B, and C share five small subunits, two of which, ABC10 alpha and ABC10 beta, comigrate on SDS polyacrylamide gels. The gene encoding ABC10 alpha, RPC10, was isolated based on microsequence data. RPC10 is a single copy gene localized on chromosome VIII. It codes for a very basic protein of only 70 amino acids, which contains a zinc binding domain of the form CX2CX13CX2C. Deletion of its gene indicated that, despite its very small size, the ABC10 alpha subunit is essential for yeast cell viability. ABC10 alpha and ABC10 beta have little sequence similarity.

Two additional common subunits, ABC10 alpha and ABC10 beta, are shared by yeast RNA polymerases.

Yeast nuclear RNA polymerases are multisubunit enzymes that contain in common some small subunits. We show that the smallest, a 10-kDa component of three enzymes (A10, B10, and C10), is heterogeneous. In each case, it can be resolved into two distinct polypeptides (alpha and beta) by reverse-phase chromatography. A10 alpha, B10 alpha, and C10 alpha were indistinguishable on the basis of their electrophoretic and chromatographic behaviors, characteristic silver staining, and tryptic peptide analysis. All three polypeptides are blocked at their amino termini. By the same criteria, A10 beta, B10 beta, and C10 beta were also indistinguishable. The amino-terminal sequence of A10 beta and C10 beta corresponded to that of subunit B10 recently cloned by Woychik and Young (Woychik, N. A., and Young, R. A. (1990) J. Biol. Chem. 265, 17816-17819). Thus, the three forms of RNA polymerase share two additional and distinct polypeptides, ABC10 alpha and ABC10 beta, that therefore can be considered bona fide subunits of these enzymes. Interestingly, these two subunits bind zinc.

Zinc-binding subunits of yeast RNA polymerases.

The zinc-binding subunits of yeast RNA polymerase A(I) and B(II) have been identified by a zinc-blotting technique. The two largest subunits of each enzyme (A190, A135, B220, and B150), as well as A12.2, A10, B44.5, B12.6, and B10, bind 65Zn(II). Predicted zinc-binding motifs have been noted in the NH2-terminal part of B220 and the COOH-terminal region of B150 subunits. Subdomains encompassing these motifs have been overproduced as MalE-fusion proteins and shown to retain zinc binding activity. Site-directed mutagenesis in the predicted metal-binding domain of B150 demonstrated its role in zinc binding. Mutations of cysteine residues C1163, C1166, C1182, and C1185 affected 65Zn2+ binding in vitro and caused a lethal or thermosensitive phenotype for growth. The ability to bind zinc is not sufficient for function since mutations in vicinal residues not affecting zinc binding were either lethal or thermosensitive. The role of zinc in RNA polymerase structure and function is discussed in the light of the present results.

ABF1 binding sites in yeast RNA polymerase genes.

We have used gel retardation and DNase protection assays to investigate the trans-acting factors involved in the regulation of yeast RNA polymerase genes RPC160 and RPC40. The same binding component was found to interact with the promoter of the two genes, at a short distance (100-150 base pairs) from the transcription start sites. From its size, its DNA-binding specificity and its immunological properties, this factor appears to correspond to the autonomous replication sequences and silencer-binding factor ABF1/SBF-B. The interaction of ABF1 with the polymerase upstream box sequence was characterized using gel DNA-binding assay. The factor binds with high affinity to the polymerase upstream box sequence (Kapp = 5.10(-10) M). A mutational analysis showed that nine base pairs belonging to two separated attachment sites are involved in factor binding. The consensus sequence RTCRYB(N)4ACG was derived from the present binding studies. These data provide an experimental basis for evaluating the efficiency of known or potential ABF1 sites and for comparing several factors with ABF1-like binding properties.

Isolation, sequence, and regulation by oxygen of the yeast HEM13 gene coding for coproporphyrinogen oxidase.

The HEM13 gene of Saccharomyces cerevisiae codes for coproporphyrinogen oxidase (EC 1.3.3.3) catalyzing the sixth enzymic step in the heme biosynthetic pathway. Its expression has been previously shown to be regulated negatively by heme and oxygen. We have isolated the HEM13 gene by functional complementation of a hem13 gene by functional complementation of a hem13 mutant and determined its nucleotide sequence. The open reading frame encodes a protein of 328 amino acids. Its calculated molecular weight (37,673), amino acid composition and amino-terminal sequence predicted from the DNA sequence are in agreement with those determined for the native enzyme (Camadro, J. M., Chambon, H., Jolles, J., and Labbe, P. (1986) Eur. J. Biochem. 156, 579-587). The 5' ends of the HEM13 transcripts were identified by nuclease S1 mapping; induction of HEM13 resulted in an equivalent increase of the level of all the transcripts. 5' deletion analysis revealed that DNA sequence located upstream of 409 nucleotides from the translational initiation codon was needed for depression under oxygen limitation. The loss of induction of coproporphyrinogen oxidase activity by anaerobiosis caused a considerable decrease of heme formation in anaerobic cells.

RPC40, a unique gene for a subunit shared between yeast RNA polymerases A and C.

Yeast RNA polymerases A and C share an approximately equal to 40 kd subunit. We have identified, sequenced, and mutagenized in vitro the AC40 subunit gene. The RPC40 gene is unique in the yeast genome and is required for cell viability. This gene contains an open reading frame encoding a 37.6 kd protein having no significant homology with bacterial RNA polymerase subunits. The promoter region contains a 19 bp sequence also present in the largest subunit of RNA polymerase C. It also contains a well-conserved RPG box, a sequence found in the promoter region of many genes encoding the translational apparatus. A novel, plasmid-shuffling method was developed to isolate a large number of RPC40 ts mutants. One of these, ts4, was shown to be defective in the synthesis of RNA polymerases A and C at the restrictive temperature. In contrast, RNA polymerase B was made normally.

Isolation of structural genes for yeast RNA polymerases by immunological screening.

A lambda gt11 yeast genomic library was screened with antibodies directed against yeast RNA polymerases A, B, and C. Thirty-five individual recombinant phages that expressed proteins in Escherichia coli that were antigenically related to RNA polymerases A, B, or C were isolated by using 22 distinct antisera. Thus, all 22 genes for the RNA polymerase subunits were potentially cloned. In three cases (lambda A-43, lambda A-40, and lambda A-34.5), an antigenic protein was expressed in E. coli with the same molecular weight as the corresponding subunit. When lambda A-40 DNA was used to hybrid-select yeast mRNA, the protein translated in vitro was the expected size for the A-40 subunit, further supporting our isolation of the A-40 gene. However, mRNA hybrid selected by lambda A-27 DNA did not code for a protein of the correct size. The lengths of the mRNA that hybridized to phage lambda A-190 or lambda C-160 DNA on RNA blots were in agreement with the predicted sizes of the coding regions of the corresponding genes. As predicted by our previous immunological results, yeast DNA inserts of the lambda A-190 and lambda C-160 clones cross-hybridized to the B-220 subunit gene. The cloned genes for the RNA polymerase subunits will prove to be valuable tools for the study of the function, regulation, and genetics of the yeast RNA polymerases.

Effects of methyl mercury and triethyllead on Na+K+ATPase and pyruvate dehydrogenase activities in glioma C6 cells.

The effect of methyl mercury (MeHg) and triethyllead (Et3Pb) on the membrane bound SH-enzymes Na+K+ATPase and pyruvate dehydrogenase (PDH) was studied in relation to the effect on the galactosyl ceramide sulfotransferase (CST) and to morphological changes in glioma C6 cells. Two-day-old cultures were incubated for 1 or 20 hrs with 5-30 microgram MeHgC1 and 2-8 microgram Et3PbC1/mg cell protein. The results show that both compounds induced morphological changes and a reduction of CST activity at growth inhibitory concentrations. A less marked reduction of Na+K+ATPase was induced with increasing exposure time only in MeHgC1 treated cultures, and PDH activity was not affected by either of the compounds under the experimental conditions. Thus, an interference with Na+K+ATPase and PDH activities do not appear to be a primary effect of MeHg and Et3Pb intoxication.

Different levels of changes induced by the insecticide lindane in cultured C-6 glioma cells.

The effects of lindane on alpha-aminoisobutyric acid and uridine uptake were studied in the rat line C-6 of glial origin. Different levels of modifications could be distinguished by use of 2 modes of investigation. One study, carried out within minutes of contact with lindane, has shown that the inhibition of alpha-aminoisobutyric acid and uridine uptake was rapid and immediately reversible. The other study, carried out after several hours of treatment followed by a 15-min triple washing, has allowed to attribute the non-reversible decrease in uridine uptake induced by a 20-h treatment to an inhibition of protein synthesis, and the increase in alpha-aminoisobutyric acid uptake to the derepression of synthesis of transport proteins.