Localization and regulation of the cdk-activating kinase (Cak1p) from budding yeast.

Eukaryotic cell cycles are controlled by the activities of cyclin-dependent kinases (cdks). The major cdk in budding yeast, Saccharomyces cerevisiae, is Cdc28p. Activation of Cdc28p requires phosphorylation on threonine 169 and binding to a cyclin. Thr-169 is phosphorylated by the cdk-activating kinase (CAK), Cak1p, which was recently identified as the physiological CAK in budding yeast. Here we present our further characterization of yeast Cak1p. We have found that Cak1p is dispersed throughout the cell as shown by immunofluorescence; biochemical subcellular fractionation confirmed that most of the Cak1p is found in the cytoplasm. Cak1p is a monomeric enzyme in crude yeast lysates. Mutagenesis of potential sites of activating phosphorylation had little effect on the activity of Cak1p in vitro or in vivo. Furthermore, Cak1p contains no posttranslational modifications detectable by two-dimensional isoelectric focusing. We found that Cak1p is a stable protein during exponential growth but that its expression decreases considerably when cells enter stationary phase. In contrast, Cak1p levels oscillate dramatically during meiosis, reflecting regulation at both the transcriptional and post-translational level. The localization and regulation of Cak1p are in contrast to those of the known vertebrate CAK, p40(MO15).

Novel CDC34 (UBC3) ubiquitin-conjugating enzyme mutants obtained by charge-to-alanine scanning mutagenesis.

CDC34 (UBC3) encodes a ubiquitin-conjugating (E2) enzyme required for transition from the G1 phase to the S phase of the budding yeast cell cycle. CDC34 consists of a 170-residue catalytic N-terminal domain onto which is appended an acidic C-terminal domain. A portable determinant of cell cycle function resides in the C-terminal domain, but determinants for specific function must reside in the N-terminal domain as well. We have explored the utility of "charge-to-alanine" scanning mutagenesis to identify novel N-terminal domain mutants of CDC34 that are enzymatically competent with respect to unfacilitated (E3-independent) ubiquitination but that nevertheless are defective with respect to its cell cycle function. Such mutants may reveal determinants of specific in vivo function, such as those required for interaction with substrates or trans-acting regulators of activity and substrate selectivity. Three of 18 "single-scan" mutants (in which small clusters of charged residues were mutated to alanine) were compromised with respect to in vivo function. One mutant (cdc34-109, 111, 113A) targeted a 12-residue segment of the Cdc34 protein not found in most other E2s and was unable to complement a cdc34 null mutant at low copy numbers but could complement a null mutant when overexpressed from an induced GAL1 promoter. Combining adjacent pairs of single-scan mutants to produce "double-scan" mutants yielded four additional mutants, two of which showed heat and cold sensitivity conditional defects. Most of the mutant proteins expressed in Escheria coli displayed unfacilitated (E3-independent) ubiquitin-conjugating activity, but two mutants differed from wild-type and other mutant Cdc34 proteins in the extent of multiubiquitination they catalyzed during an autoubiquitination reation-conjugating enzyme function and have identified additional mutant alleles of CDC34 that will be valuable in further genetic and biochemical studies of Cdc34-dependent ubiquitination.

The GC box and TATA transcription control elements in the P38 promoter of the minute virus of mice are necessary and sufficient for transactivation by the nonstructural protein NS1.

To further define the transcriptional regulation of the P38 promoter in the minute virus of mice (MVM) genome, we constructed a series of internal deletion and linker scanning mutations. The mutant P38 constructs were assayed for transcriptional activity in vitro by primer extension analysis with nuclear extracts from murine A92L fibroblasts. Mutations which disrupted the GC box and TATA box severely reduced transcription in vitro. DNase I footprinting analysis confirmed that the murine transcription factor Sp1 bound to the GC box; however, no factors were observed interacting with a putative transcriptional activation regulatory element, termed the TAR element. The linker scanning mutations were analyzed in vivo by using a chloramphenicol acetyltransferase expression assay system, in both the presence and absence of constructs expressing the viral nonstructural protein, NS1. The ability of NS1 to transactivate the P38 promoter (up to 1,000-fold) depended entirely on the presence of intact GC and TATA box sequences. Disruption of the TAR element by either linker insertion mutations or an internal deletion did not inhibit transactivation of the P38 promoter. These results suggest that NS1 transactivates the P38 promoter indirectly by interacting with one or more components of the P38 core-transcription complex.

Unusual Sp1-GC box interaction in a parvovirus promoter.

The P4 promoter of the parvovirus minute virus of mice contains a single degenerate GC box sequence which binds the transcription factor Sp1 with high affinity. The two promoters of murine Sp1 were affinity purified, and their interactions with the P4 promoter were examined. Several unusual features were observed. Methylation interference experiments demonstrated that Sp1 makes contacts with both DNA strands, including the central guanine as well as an adenine residue on the cytidine-rich strand of the GC box. UV photocross-linking revealed that the 95- and the 105-kDa promoters cross-link exclusively to opposite strands of the GC box. These results suggest that the phosphorylation of the 95-kDa Sp1 promoter results in a change in the way Sp1 is positioned on the P4 GC box and identifies a high-affinity GC box motif.

Kinetics of dithionite reduction of the heme nonapeptide of cytochrome c.

The kinetics of dithionite reduction of the oxidized heme nonapeptide fragment of horse heart cytochrome c have been measured as a function of ionic strength at pH 7 and pH 9 by the stopped-flow technique. Dithionite concentration dependences indicate that the radical anion monomer, SO2-., is the active reductant. The pH 7 ionic strength dependence suggests that the heme peptide is reacting as a negatively charged molecule (its overall charge is calculated to be -1). Comparison of these results with the known rate of dithionite reduction of cytochrome c indicates that the heme nonapeptide has substantially greater inherent reactivity than cytochrome c, perhaps due to the greater accessibility of the heme.