Portal de Periódicos da CAPES

The NOT proteins are part of the CCR4 transcriptional complex and affect gene expression both positively and negatively

1998; Springer Nature; Volume: 17; Issue: 4 Linguagem: Inglês

10.1093/emboj/17.4.1096

1460-2075

Haiyan Liu, Vasudeo Badarinarayana, Deborah C. Audino, Juri Rappsilber, Matthias Mann, Clyde L. Denis,

Advanced Proteomics Techniques and Applications

Article15 February 1998free access The NOT proteins are part of the CCR4 transcriptional complex and affect gene expression both positively and negatively Hai-Yan Liu Hai-Yan Liu Department of Biochemistry and Molecular Biology, Rudman Hall, University of New Hampshire, Durham, NH, 03824 USA Search for more papers by this author Vasudeo Badarinarayana Vasudeo Badarinarayana Department of Biochemistry and Molecular Biology, Rudman Hall, University of New Hampshire, Durham, NH, 03824 USA Search for more papers by this author Deborah C. Audino Deborah C. Audino Department of Biochemistry and Molecular Biology, Rudman Hall, University of New Hampshire, Durham, NH, 03824 USA Search for more papers by this author Juri Rappsilber Juri Rappsilber Peptide and Protein Group, European Molecular Biology Laboratory, Meyerhofstrasse 1, Postfach 10.2209, 69012 Heidelberg, Germany Search for more papers by this author Matthias Mann Matthias Mann Peptide and Protein Group, European Molecular Biology Laboratory, Meyerhofstrasse 1, Postfach 10.2209, 69012 Heidelberg, Germany Search for more papers by this author Clyde L. Denis Corresponding Author Clyde L. Denis Department of Biochemistry and Molecular Biology, Rudman Hall, University of New Hampshire, Durham, NH, 03824 USA Search for more papers by this author Hai-Yan Liu Hai-Yan Liu Department of Biochemistry and Molecular Biology, Rudman Hall, University of New Hampshire, Durham, NH, 03824 USA Search for more papers by this author Vasudeo Badarinarayana Vasudeo Badarinarayana Department of Biochemistry and Molecular Biology, Rudman Hall, University of New Hampshire, Durham, NH, 03824 USA Search for more papers by this author Deborah C. Audino Deborah C. Audino Department of Biochemistry and Molecular Biology, Rudman Hall, University of New Hampshire, Durham, NH, 03824 USA Search for more papers by this author Juri Rappsilber Juri Rappsilber Peptide and Protein Group, European Molecular Biology Laboratory, Meyerhofstrasse 1, Postfach 10.2209, 69012 Heidelberg, Germany Search for more papers by this author Matthias Mann Matthias Mann Peptide and Protein Group, European Molecular Biology Laboratory, Meyerhofstrasse 1, Postfach 10.2209, 69012 Heidelberg, Germany Search for more papers by this author Clyde L. Denis Corresponding Author Clyde L. Denis Department of Biochemistry and Molecular Biology, Rudman Hall, University of New Hampshire, Durham, NH, 03824 USA Search for more papers by this author Author Information Hai-Yan Liu1, Vasudeo Badarinarayana1, Deborah C. Audino1, Juri Rappsilber2, Matthias Mann2 and Clyde L. Denis 1 1Department of Biochemistry and Molecular Biology, Rudman Hall, University of New Hampshire, Durham, NH, 03824 USA 2Peptide and Protein Group, European Molecular Biology Laboratory, Meyerhofstrasse 1, Postfach 10.2209, 69012 Heidelberg, Germany *Corresponding author. E-mail: [email protected] The EMBO Journal (1998)17:1096-1106https://doi.org/10.1093/emboj/17.4.1096 PDFDownload PDF of article text and main figures. ToolsAdd to favoritesDownload CitationsTrack CitationsPermissions ShareFacebookTwitterLinked InMendeleyWechatReddit Figures & Info The CCR4 transcriptional regulatory complex consisting of CCR4, CAF1, DBF2 and other unidentified factors is one of several groups of proteins that affect gene expression. Using mass spectrometry, we have identified the 195, 185 and 116 kDa species which are part of the CCR4 complex. The 195 and 185 kDa proteins were found to be NOT1 and the 116 kDa species was identical to NOT3. NOT1, 2, 3 and 4 proteins are part of a regulatory complex that negatively affects transcription. All four NOT proteins were found to co-immunoprecipitate with CCR4 and CAF1, and NOT1 co-purified with CCR4 and CAF1 through three chromatographic steps in a complex estimated to be 1.2×106 Da in size. Mutations in the NOT genes affected many of the same genes and processes that are affected by defects in the CCR4 complex components, including reduction in ADH2 derepression, defective cell wall integrity and increased sensitivity to monoand divalent ions. Similarly, ccr4, caf1 and dbf2 alleles negatively regulated FUS1–lacZ expression, as do defects in the NOT genes. These results indicate that the NOT proteins are physically and functionally part of the CCR4 complex which forms a unique and novel complex that affects transcription both positively and negatively. Introduction There are a number of general regulatory complexes that are involved in transcriptional processes. For example, in addition to the yeast holoenzyme that contains the SRB proteins (Wilson et al., 1996), the SPT3–ADA2–GCN5 complex (Grant et al., 1997), the NOT complex (Collart and Struhl, 1994), the PAF1 holoenzyme (Wade et al., 1996) and the CCR4 complex have all been identified as playing roles in affecting gene transcription. Each of these groups of proteins appears to be unique. The interaction and functional relationship of these groups of transcriptional regulatory factors, however, remain to be clearly established. In this study, we demonstrate that the NOT protein complex is part of the CCR4 transcriptional complex and that these two groups of proteins share overlapping functions. CCR4 affects the expression of many genes and processes in yeast. It is required for the expression of ADH2 and other non-fermentative genes (Denis, 1984; Denis and Malvar, 1990) and for unidentified genes involved in cell wall integrity (Liu et al., 1997). ccr4 mutations result in a partial cell cycle block during telophase and increase the sensitivity of yeast cells to Li+ and Mg2+ (Liu et al., 1997). ccr4 is also a suppressor of spt10 mutations (Denis, 1984), defects which result in enhanced transcription at ADH2 (Denis and Malvar, 1990) and other loci (Natsoulis et al., 1991). In addition to acting as an activator, CCR4 has been implicated in negatively affecting gene expression as well (McKenzie et al., 1993; Schild, 1995). CCR4 is a component of a multi-subunit complex (Draper et al., 1994). Two of the CCR4 complex components, CAF1 (POP2) (Sakai et al., 1992; Draper et al., 1995), and DBF2, a cell cycle-regulated protein kinase (Toyn et al., 1991), function to control many of the same processes as CCR4 (Liu et al., 1997). While none of these genes by themselves are essential, the phenotypes conferred by the ccr4, caf1 and dbf2 mutations indicate that the CCR4 complex is required for optimal and proper expression of many genes. The evolutionary conservation of CAF1 across eucaryotes (Draper et al., 1995) further suggests that this complex plays an important role in eucaryotic gene control. Although the mechanism of how CCR4 functions remains unclear, the site of CCR4 action at the ADH2 locus has been shown to occur at a post-chromatin remodeling step (Verdone et al., 1997). In addition to CAF1 and DBF2, the CCR4 complex contains several unidentified proteins, 195, 185, 140 and 116 kDa in size (Draper et al., 1994). Our initial attempt at cloning the corresponding genes for these proteins by two-hybrid analysis was unsuccessful (Draper et al., 1995; Liu et al., 1997). Mass spectrometry has recently become the method of choice for rapid and unambiguous identification of gel-separated proteins. Large-scale analysis of yeast proteins is now possible (Shevchenko et al., 1996), and entire yeast protein complexes can be studied (Lamond and Mann, 1997; Neubauer et al., 1997). Here, we have used these methods to identify the 195, 185 and 116 kDa species of the CCR4 complex. The 185 and 195 kDa species were found to be NOT1 and the 116 kDa species was found to be NOT3. The NOT genes have been identified as encoding a group of factors involved in repressing the transcription of HIS3 from a non-canonical TATA (Collart and Struhl, 1994). This group of proteins contains NOT1/CDC39, NOT2/CDC36, NOT3 and NOT4/MOT2/SIG1, and genetic evidence indicates that they function as a complex in vivo (Collart and Struhl, 1993, 1994). In addition to affecting HIS3 expression, the not mutations augment the expression of many genes or reporter genes, confirming their role as a repression complex (Cade and Errede, 1994; Collart and Struhl, 1994; Irie et al., 1994; Collart, 1996). Of the four NOT genes, only NOT1 was found to be essential. We have subsequently shown that NOT2 and NOT4 also associate with the CCR4 complex. Genetic analyses reveal that NOT defects result in phenotypes similar to those observed with the deletion of CCR4 and its associated components. These results indicate that the CCR4 complex includes the NOT proteins and that this complex can affect gene transcription both positively and negatively. Results The 185/195 and 116 kDa proteins in the CCR4 complex are NOT1 and NOT3 To identify the proteins which associate with CCR4, the CCR4 complex was isolated by immunoprecipitation. Yeast extracts, containing either a LexA–CAF1 fusion protein or just LexA alone, were incubated with an antibody directed against the LexA protein, and the resulting immunoprecipitates were subjected to SDS–PAGE (Figure 1). After staining the proteins, the 116, 185 and 195 kDa species that specifically co-immunoprecipitated with CCR4 (Draper et al., 1994) were isolated and were analyzed by mass spectrometry using the strategy previously described (Shevchenko et al., 1996). A small aliquot of the peptide mixture resulting from in-gel digestion of the bands was analyzed by matrix-assisted laser desorption/ionization (MALDI). High resolution peptide mass maps were obtained of all three bands which were analyzed. Database searches with the set of measured masses resulted in the following identifications: band 116 kDa was NOT3, band 185 kDa was NOT1 and band 195 kDa was also NOT1. The identification of NOT3 was performed by MALDI peptide mapping only. The database search revealed that 26 measured peptide masses fit the sequence of NOT3 within a mass accuracy of 50 p.p.m. This corresponds to 30% of the sequence. The other two bands were subjected to both MALDI peptide mapping and mass spectrometric sequencing using nanoelectospray (Wilm et al., 1996). The peptide maps covered 29% of the protein in the band migrating at 185 kDa and 32% of the protein in the band migrating at 195 kDa. The identification of the lower band is shown in Figure 2. Sequencing of 10 of the peptides derived from the 185 kDa band and eight of the peptides derived from the 195 kDa band confirmed the identification (data not shown). No peptides of the N-terminal region of the NOT1 protein were found in the analysis of the lower band. Thus, the data are consistent with the N-terminal truncation of the NOT1 protein suggested by previous studies (Collart, 1996). Figure 1.Immunoprecipitation of the CCR4 complex for protein sequencing by mass spectrometry. The yeast whole cell extracts containing either LexA alone or full-length LexA–CAF1 were treated with the LexA antibody, and the resulting immunoprecipitates were subjected to SDS–PAGE. The resulting gel was stained with Coomassie blue. 'M' indicates the molecular weight standard. Lanes 1 and 2 are the immunoprecipitates from extracts containing LexA alone and LexA–CAF1, respectively. The 195, 185 and 116 kDa species in lane 2 were excised prior to mass spectrometric analysis. Download figure Download PowerPoint Figure 2.Identification of the yeast protein NOT1 from the 185 kDa band by MALDI mass spectrometry. The figure shows the MALDI mass spectrum obtained after in-gel digestion of the 185 kDa band. Ion signals whose measured masses match calculated masses of tryptic peptides of NOT1 within 50 p.p.m. are indicated with circles. Filled circles mark those ion signals whose corresponding peptides were sequenced additionally by nanoelectrospray mass spectrometry. In one case, nanoelectrospray sequencing revealed two peptides for one measured peptide mass (peak at 1183.634 Da, marked by two filled circles). Ion signals corresponding to trypsin autolysis products are labeled with the letter 'T'. Download figure Download PowerPoint NOT2 and NOT4 are also in the CCR4 complex The NOT1 and NOT3 proteins have been shown to be part of a complex that also includes the NOT2 and NOT4 proteins (Collart and Struhl, 1994). To examine the possibility that the NOT2 and NOT4 proteins were also part of the CCR4 complex, we carried out a series of immunoprecipitation experiments. We first examined the association of NOT1 with CCR4. A LexA–NOT1 fusion was expressed in a wild-type strain. LexA–NOT1 was immunoprecipitated with the LexA antibody while the CCR4 complex was immunoprecipitated with the CCR4 antibody. The resulting immunoprecipitates were subjected to Western blot analysis (Figure 3A). CCR4 co-immunoprecipitated with LexA–NOT1 (Figure 3A, lane 3) while LexA–NOT1 along with the NOT1 proteins (185/195 kDa) were co-immunoprecipitated with CCR4 (Figure 3A, lane 5). These results confirm the protein sequencing data. Figure 3.Co-immunoprecipitation of the NOT1 and NOT2 proteins with the CCR4 complex. (A) The yeast whole cell extracts containing either LexA–NOT1 or LexA–NOT2 were treated with LexA antibody (lanes 3 and 4) or CCR4 antibody (lanes 5 and 6). The resulting immunoprecipitates along with the crude extracts (lane 1 and 2) were subjected to immunoblot analysis and probed with NOT1, CCR4 and LexA antibodies. (B) The yeast whole extracts containing LexA–NOT2 prepared from a caf1-deleted strain (lanes 1, 3 and 5) or a ccr4-deleted strain (lanes 2, 4 and 6) were treated with LexA antibody (lanes 1 and 2), CCR4 antibody (lanes 3 and 4) or CAF1 antibody (lanes 5 and 6). The resulting immunoprecipitates were subjected to immunoblot analysis and probed with NOT1, CCR4 and LexA antibodies. The bands beneath LexA–NOT2 in lanes 1 and 2 represent degradation products of LexA–NOT2 (data not shown). Download figure Download PowerPoint To investigate the association of NOT2 with the CCR4 complex, a LexA–NOT2 fusion was expressed in a wild-type strain, a ccr4Δ strain and a caf1Δ strain. An antibody raised against the LexA protein was used to immunoprecipitate the LexA–NOT2 fusion while antibodies raised against either CCR4 or CAF1 were used to bring down CCR4 and CAF1, respectively. The resulting immunoprecipitates were subjected to SDS–PAGE, followed by Western blot analysis (Figure 3A and B). Immunoprecipitating LexA–NOT2 with the LexA antibody resulted in co-immunoprecipitation of NOT1 from the wild-type, ccr4Δ and caf1Δ extracts (Figure 3A, lane 4, and B, lanes 1 and 2, respectively). CCR4 co-immunoprecipitated along with LexA–NOT2 and NOT1 from the wild-type strain (Figure 3A, lane 4), but not from the caf1Δ strain (Figure 3B, lane 1). When the CCR4 antibody was used to repeat the immunoprecipitation experiments, the NOT1 and LexA–NOT2 proteins were found to co-immunoprecipitate with CCR4 from the wild-type strain (Figure 3A, lane 6), but not from the strains lacking either CAF1 (Figure 3B, lane 3) or CCR4 (Figure 3B, lane 4). Longer exposures of the results presented in Figure 3B, lane 3, indicated that a small amount of NOT1 and LexA–NOT2 was found to co-immunoprecipitate with CCR4 from the caf1Δ strain (data not shown). These results indicate that NOT2 physically interacts with both CCR4 and NOT1, and that the association of CCR4 with the NOT proteins is largely dependent on the presence of CAF1. The immunoprecipitation experiments were also repeated by using the CAF1 antibody. NOT1, LexA–NOT2 and CCR4 were found to co-immunoprecipitate with CAF1 from the wild-type strain (data not shown), and NOT1 and LexA–NOT2 were co-immunoprecipitated with CAF1 from the ccr4Δ strain (Figure 3B, lane 6). However, NOT1 and LexA–NOT2 failed to co-immunoprecipitate with the CAF1 antibody from the caf1Δ strain (Figure 4B, lane 7), confirming that LexA–NOT2 does not immunoprecipitate fortuitously with the CAF1 antibody. These results also suggest that the interaction between CAF1 and the NOT proteins is CCR4 independent. Figure 4.Co-immunoprecipitation of the NOT1, 2 and 4 proteins with the CCR4 complex. (A) Yeast whole cell extracts containing c-Myc–NOT4 and LexA–CAF1 were treated with LexA pre-immune serum (lane 2), LexA antibody (lane 3), CAF1 antibody (lane 4), CCR4 antibody (lane 5) or c-Myc antibody (lane 6). The resulting immunoprecipitates along with the crude extract (lane 1) were subjected to immunoblot analysis and probed with NOT1, CCR4, LexA and c-Myc antibodies. (B) Yeast whole cell extracts containing LexA–NOT2 and c-Myc–NOT4 were treated with c-Myc antibody (lane 1), LexA antibody (lane 2) or CCR4 antibody (lane 3). The resulting immunoprecipitates were subjected to immunoblot analysis and probed with NOT1, CCR4, c-Myc, NOT5 and LexA antibodies. Download figure Download PowerPoint To address the question as to whether NOT4 was in the CCR4 complex, a c-Myc-tagged NOT4 fusion was expressed along with LexA–CAF1 in a wild-type strain. Extracts treated with the LexA antibody resulted in co-immunoprecipitation of c-Myc–NOT4 with LexA–CAF1, CCR4 and NOT1 (Figure 4A, lane 3), while the LexA pre-immune serum failed to immunoprecipitate these proteins (Figure 4A, lane 2). The c-Myc–NOT4 protein also co-immunoprecipitated with CCR4 and NOT1 when the extracts were immunoprecipitated with either CAF1 antibody (Figure 4A, lane 4) or CCR4 antibody (Figure 4A, lane 5). Immunoprecipitation with the c-Myc antibody, in turn, was able to bring down LexA–CAF1, CCR4 and NOT1 along with c-Myc–NOT4 (lane 6). We also immunoprecipitated the CCR4 complex from an extract prepared from a strain expressing both LexA–NOT2 and c-Myc–NOT4 fusion proteins. The resulting immunoprecipitates were analyzed by Western blot (Figure 4B). It is clear that NOT1, NOT2 and NOT4 co-immunoprecipitated with CCR4 and CAF1. Because NOT3 is also in the CCR4 complex as determined by mass spectrometry, we conclude that the complete NOT repressive regulatory complex is part of the CCR4 complex. Two-hybrid analysis was used further to examine the interaction of the NOT proteins and the CCR4 complex components. As shown in Table I, both B42–NOT1 and B42–NOT2 interacted with LexA–CAF1, and LexA–NOT1 was found to interact with B42–CAF1. LexA–CCR4 interacted with B42–NOT1, and B42–DBF2 interacted well with LexA–NOT2. The multiplicity of these interactions confirms the above-described protein analyses. Table 1. Two-hybrid interaction assay β-Gal activity (U/mg) B42–NOT1 B42–NOT2 B42–CAF1 B42–DBF2 B42 LexA–NOT1 – – 130 – 3.6 LexA–NOT2 – 900 270 660 110 LexA–CCR4 330 – 1100 74 6.4 LexA–CAF1 1100 380 – 930 86 LexA <2 <2 <2 <2 <2 LexA–CCR4, −NOT1 and −NOT2 contain full-length CCR4, NOT1 and NOT2. LexA–CAF1 contains residue 127–444 of CAF1. All LexA fusions contain residues 1–202 of LexA. B42–NOT1, −NOT2 and −DBF2 contain full-length NOT1, NOT2 and DBF2. B42–CAF1 contains residues 148–444 of CAF1. − indicates the β-galactosidase activity is no greater than the background interaction with B42 alone. The CCR4 complex is a unique transcriptional regulatory complex Our previous studies on CCR4 indicated that the CCR4 complex is a transcriptional regulatory complex distinct from that of several other complexes such as the SNF/SWI complex, the yeast holoenzyme and the putative SPT4, 5, 6 complex (Denis et al., 1994). The size of the CCR4 complex was estimated following Superose 6 gel filtration chromatography. As shown in Figure 5A, CCR4 migrated in two separate peaks of 1.9×106 and 1.0×106 Da. In other experiments, a small portion of CCR4 migrated at 2.0×105 Da, which is close to the size of CCR4 and may represent monomeric CCR4 (Figure 5C, top panel). The two larger complexes were also unaffected by prior DNase treatment, suggesting that they do not result from non-specific binding to DNA (Figure 5A, data not shown). The 1.9×106 Da CCR4 complex is separate from that of the SRB complex which, as analyzed on a longer Superose 6 column, migrated at 1.7×106 Da (Figure 5B). Moreover, in a caf1Δ strain, most of the CCR4 protein was found at the 1.0×105 Da size, indicating that the CAF1 protein is required for CCR4 association in the 1.9×106 and 1.0×106 Da complexes (Figure 5C, top two panels). A caf1Δ had no effect, however, on the ability of the non-CCR4 complex component, SPT10, to migrate at 1.9×106 Da (Figure 5C, bottom two panels), nor on the SRB5 protein to migrate at 1.7×106 Da (data not shown). Figure 5.Analysis of the CCR4 complexes using gel filtration chromatography. (A) Yeast whole cell extracts prepared from a wild-type strain and treated with DNase were chromatographed on a Superose 6 HR10/30 column. The resulting 1 ml fractions (30 μl of each fraction) were subjected to immunoblot analysis and probed with CCR4 antibody. The two arrows indicate the size of the two peaks containing CCR4. (B) Yeast whole cell extracts were chromatographed on an extended Superose 6 HR16/50 column. The resulting 1 ml fractions (first 12 fractions) were subjected to immunoblot analysis and probed with CCR4 and SRB5 antibodies. The arrow indicates the peak containing SRB5. (C) Yeast whole cell extracts prepared from a wild-type or a caf1-deleted strain were chromatographed on the Superose 6 HR10/30 column. The resulting 1 ml fractions (every other fraction is displayed) were subjected to immunoblot analysis and probed for CCR4 (top two panels) and SPT10 (bottom two panels). 'CAF1' and 'caf1' indicate the wild-type and caf1-deleted strains, respectively. The 120 kDa band that runs above CCR4 in the CCR4-probed panel represents a non-specific protein and serves as an internal control for the experiment. Based on this control, the amount of protein loaded for the 'CAF1' experiment was about twice that of the 'caf1' experiment, resulting in the decreased level of CCR4 protein visible in the 'caf1'experiment for the SPT10 Western (bottom panel). Download figure Download PowerPoint To analyze the CCR4 complex further, we isolated the CCR4 complex from a strain in which the CAF1 gene was deleted and a CAF1 gene tagged at its C-terminus with 6×His was integrated into the genome at the TRP1 locus. This CAF1–6His gene was able to complement the defect of caf1Δ (Liu et al., 1997). The extracts prepared from this strain were first put onto a Ni2+-NTA column, and the bound proteins was eluted with 250 mM imidazole. The NOT1 protein and CAF1–6His were found to co-immunoprecipitate with CCR4 when the Ni2+ eluate was treated with CCR4 antibody (data not shown). The Ni2+ eluate subsequently was loaded onto a Mono Q column, and the bound proteins were eluted in a linear salt gradient. The Mono Q fractions were analyzed by Western blot using both CCR4 and CAF1 antibody, and CCR4 and CAF1–6His were found to co-elute (Liu et al., 1997; Figure 6). Fractions containing both CCR4 and CAF1 were pooled and the proteins were analyzed further by Superose 6 gel filtration chromatography. The fractions from these different steps in purification were subjected to Western blot analysis. The purified CCR4 complex displayed a molecular weight of 1.2×106 Da following the Superose 6 gel filtration chromatography (Figure 6), corresponding closely to the 1.0×106 Da CCR4 complex observed in crude extracts (Figure 5A). NOT1, CCR4 and CAF1 were all found to co-purify through these three purification steps. In contrast, Western blot analysis using antibodies against SRB5 and SRB6 failed to detect either of these proteins in the Mono Q and Superose 6 fractions (data not shown). These data indicate that NOT1, CCR4 and CAF1 are components of the same complex. In addition, the 1.9×106 and 1.2×106 Da CCR4 complexes appear distinct from the yeast holoenzyme containing the SRB complex. Figure 6.Co-purification of the NOT1 protein and the CCR4 complex. Yeast whole cell extracts prepared from a caf1-deleted strain containing CAF1–6His integrated at the TRP1 locus were chromatographed on a Ni2+-NTA, Mono Q and Superose 6 HR10/30 column as described in Materials and methods. Fractions of 0.5 ml from the Superose 6 chromatography were subjected to immunoblot analysis and probed for NOT1, CCR4 and CAF1. The Mono Q lane refers to the peak fraction following Mono Q chromatography that was applied to the Superose 6 column. The Ni2+-NTA eluate was not probed for NOT1. The arrow indicates the size of the peak eluted from the Superose 6 column that contains NOT1, CCR4 and CAF1–6His. Download figure Download PowerPoint Mutations in the NOT genes result in similar phenotypes to those observed with ccr4 and caf1 alleles The presence of the NOT proteins in the CCR4 complex suggest that they should function to control similar genes and processes as do CCR4 and its associated components. However, the NOT proteins have been characterized as a repression complex and CCR4 is generally considered to be an activator. To address this issue, we analyzed the effect of not mutations on several processes known to be affected by ccr4. The results from the phenotypic analyses are summarized in Table II. Mutations in the NOT genes except for NOT3 reduced ADH2 expression under non-fermentative conditions, indicating that the NOT proteins can act as activators. A not4 allele was also capable of suppressing the enhanced ADH2 expression that is caused by an spt10 defect. All of the not alleles except for not3 also displayed sensitivity to caffeine, a phenotype resulting from defects in cell wall integrity, which is shared by the ccr4, caf1 and dbf2 alleles (Liu et al., 1997). ccr4, caf1 and dbf2 mutations also result in temperature- and/or cold-sensitive phenotypes that are suppressible by 1 M sorbitol, confirming their roles in control of cell wall integrity (Liu et al., 1997). In agreement with this phenotype, it has been shown previously that a not4 allele confers a temperature-sensitive phenotype that is suppressible by 1 M sorbitol (Cade and Errede, 1994). We subsequently found that the not2 ts phenotype was also relieved by 1 M sorbitol (Table II). Also, the caffeine-sensitive phenotype of not4 was suppressed by 1 M sorbitol (data not shown). In agreement with these results, not2, not3 and not4 alleles were sensitive to staurosporine, an inhibitor of PKC1, indicative of cell wall defects. Moreover, not1, not2 and not4 alleles were sensitive to 0.04% SDS, another phenotype indicative of a defect in cell wall integrity (Igual et al., 1996) also displayed by ccr4, caf1 and dbf2 alleles (data not shown). Furthermore, not1, not2 and not4 alleles were sensitive to high concentrations of the divalent cation, Mg2+, as are ccr4-, caf1- and dbf2-containing strains (Table II). These results indicate that defects in the NOT factors result in phenotypes consistent with the NOT proteins functioning in processes similar to CCR4, CAF1 and DBF2. Table 2. Phenotypic analysis Strains ADH II spt10 ADH II Caffeine 8 mM 37°C YD 37°C YD 1 M sorbitol Mg2+ 750 mM Stauro 1 mg/ml 3 AT 20 mM wt 2400 91 + + + + + − ccr4 400 23 − + + − − − caf1 1000 7 w − + − − w/− not1 1300 78 − w w − − + not2 340 86 − − + − w + not3 2500 N.D. + + + + w + not4 1200 13 w − w w − + Growth was scored on YD plates as supplemented with 8 mM caffeine, 1 mg/ml of staurosporine (stauro), 750 mM MgCl2 or 1 M sorbitol as indicated. 3AT: growth was scored on minimal plates lacking histidine and containing 20 mM 3-aminotriazole (3AT) using strains isogenic to KY803 (wt) containing the YCp88-Sc4363 plasmid (Collart and Struhl, 1994). Strains used for monitoring caffeine, Mg2+, temperature and staurosporine sensitivity were KY803 (wt), EGY188-1 (ccr4), EGY188-c1 (caf1), MY8 (not1), MY16 (not2), MY508 (not3) and MY537 (not4). Wild-type strain EGY188 gave the same results as KY803. ADH II activities (mU/mg) represent the average of at least three determinations and were conducted following growth at 30°C on YEP medium containing 3% ethanol. No effect was observed in the not mutations on ADH II activity under glucose growth conditions (data not shown). The SEM for the ADH II activities was <20%. For ADH II assays, the following strains were used: wt, KY803-Δ3; not1, MY8-Δ1; not2, MY16-Δ1; not3, MY25-Δ1; not4, 612-1d-n4; and for spt10 ADH II assays the strains were: wt, spt10 segregants from cross 808-5c and 612-1d-n4; not1, spt10 not1-2 segregants from cross MY8Δ1 and 1366-4a; not2, spt10 not2 segregants from cross 808-5c and MY16-Δ1; not4, spt10 not4 segregants from cross 808-5c and 612-1d-n4. The isogenic parent for 612-1d-n4 is 612-1d whose ADH II activity is 3000 mU/mg. ADH II and spt10 ADH II activities for ccr4 and caf1 strains are taken from Denis (1984) and Draper et al. (1995). N.D., not done; '+', good growth; 'w', weak growth; '−', no or poor growth. The CCR4 complex has positive and negative effects on gene transcription To address whether the CCR4 complex components can act as repressors in a manner similar to that observed for the NOT proteins, we examined the effect of ccr4, caf1 and dbf2 defects on FUS1–lacZ expression. Mutations in NOT genes result in increased expression of the FUS1 gene or the FUS1–lacZ reporter gene in the absence of pheromone stimulation (Cade and Errede, 1994; Collart and Struhl, 1994; Irie et al., 1994). As shown in Figure 7A, deletion of CAF1 caused a 5-fold increase in β-galactosidase activity from the FUS1–lacZ reporter, while deletion of CCR4 and DBF2 resulted in an increase of β-galactosidase activity of ∼2- and 3-fold, respectively. These results are similar to the 2- to 5-fold effects observed for the not effects on the FUS1 promoter. The ccr4, caf1 and dbf2 effects on the FUS1–lacZ reporter were specific to the FUS1 promoter since ccr4, caf1 and dbf2 had very