CDC64 Encodes Cytoplasmic Alanyl-tRNA Synthetase, Ala1p, of Saccharomyces cerevisiae

Abstract

The cdc64-1 mutation causes G₁ arrest in Saccharomyces cerevisiae corresponding to a type II Start phenotype. We report that CDC64 encodes Ala1p, an alanyl-tRNA synthetase. Thus, cdc64-1 might affect charging of tRNA^Ala and thereby initiation of cell division.

In the proliferating cells of all organisms, cell growth is tightly coordinated with cell division (12, 15). Cell growth is continuous and a certain level of biosynthetic capacity is required for the completion of cell division. Saccharomyces cerevisiae cells usually arrest in the first gap phase (G₁) of the cell cycle, when cell growth is limited by the withdrawal of nutrients or the inhibition of protein synthesis. Cells become committed to divide as they pass through a late G₁ phase transition called Start. Start is responsive to the overall biosynthetic capacity of the cell, and its completion ensures the successful conclusion of cell division.

Conditional cell division cycle (cdc) mutations affect particular stage-specific events of the cell cycle and cause synchronized arrest under restrictive conditions (15). Two classes of cdc mutants that display a G₁ arrest have been identified (15). Class I mutants arrest at Start and retain mating competency. Class II mutants arrest in the early G₁ phase and are mating incompetent, suggesting a global defect in the control of metabolism and growth. Strains carrying the temperature-sensitive mutation cdc64-1 exhibit the class II phenotype at the nonpermissive temperature (1). In this report, we identified the gene that bears the cdc64-1 mutation to better understand the processes involved in the coordination of cell growth with cell division during G₁.

We used strain TC4-23-1 (MATa cdc64-1 ade1 leu2 ura3), obtained from the Yeast Genetic Stock Center, in all experiments. CDC64 was previously mapped to a locus between MET7 and PRT1 on chromosome XV (5). We identified CDC64 by complementation of the cdc64 temperature-sensitive phenotype with S. cerevisiae genomic DNA fragments from this region (Fig. 1). Integrative transformations were performed according to standard methods (9) and scored after 5 days of growth at 37°C on solid YEPD (1% yeast extract, 2% peptone, 2% dextrose) medium.

FIG. 1.

Physical map of S. cerevisiae genomic DNA fragments used to complement the cdc64-1 mutation, compared to the length (in kilobases) of chromosome XV, shown on top.

Cosmid clones (70994 and 71019) were obtained from the American Type Culture Collection. Cosmid DNA was digested with restriction endonucleases (NheI and SmaI for clones 70994 and 71019, respectively) and then transformed into the cdc64 strain. Only cosmid 70994 rescued the temperature sensitivity of strain TC4-23-1, locating cdc64-1 within clone 70994 and outside the region of overlap with clone 71019 (Fig. 1). Smaller portions of this region are contained in bacteriophage λ clones 70003 and 70120 (Fig. 1), which were obtained from the American Type Culture Collection. Phage DNA was prepared (17) and digested with SmaI, which cut in the phage arms but not within the inserts. DNA from each digest was again used to transform strain TC4-23-1. Only clone 70120 enabled cdc64 cells to grow at 37°C. Thus, cdc64-1 lies within a region of clone 70120 not shared by clone 70003. This region includes two partial open reading frames, the 3′ end of MRS2 and the 5′ end of KRE5; in addition, it contains the entire ALA1 open reading frame (Fig. 1). Open reading frames YOR334W (MRS2), YOR335C (ALA1), and YOR336W (KRE5) were obtained from Research Genetics and used directly in integrative transformations. Only YOR335C complemented the cdc64 phenotype.

Furthermore, the URA3⁺ low-copy-number centromeric plasmid YCp50 carrying ALA1 (a gift from Paul Schimmel) (16) but not an unrelated gene, the G₁ cyclin CLN3 (a gift from Fred Cross) (2), fully rescued the temperature sensitivity of the cdc64 strain (Fig. 2). Independent transformants from these plasmids were grown in liquid cultures at 30°C to a cell density of 5 × 10⁶ cells/ml. Cell proliferation was assayed by spotting 10-fold serial dilutions of the cultures on synthetic solid medium lacking uracil. The plates were incubated at the indicated temperature for 3 days. The complementation of the CEN and ALA1 transformants was plasmid linked because these transformants were not able to grow on medium containing 5-fluoroorotic acid, which selects against the URA⁺ transformants (data not shown).

FIG. 2.

ALA1 rescues the temperature-sensitive phenotype of cdc64-1 cells. Centromeric plasmids carrying either ALA1 (CEN-ALA1) or CLN3 (CEN-control) were transformed into the cdc64-1 strain and tested for complementation. Two independent transformants in each case are shown.

We then sequenced the ALA1 locus from strain TC4-23-1 to compare it to the wild-type sequence. Using a high-fidelity Taq DNA polymerase and genomic DNA as a template, the entire ALA1 open reading frame (in addition to 200 bp on either end) was PCR amplified in four 1-kb fragments. Each successive fragment overlapped the previous fragment by approximately 200 bp. The Genomics Core Facility of the Massachusetts General Hospital directly sequenced the PCR products on both strands with nested oligonucleotide primers. Comparison to the wild-type ALA1 sequence revealed a single G-to-C change at nucleotide position 818, which directs the substitution of Arg at amino acid position 273 with Pro. This mutation was verified in a separate PCR amplification with a different set of oligonucleotide primers for the PCR and sequencing reactions.

Therefore, cdc64-1 is a mutation in ALA1, which encodes an essential cytoplasmic alanyl-tRNA synthetase in S. cerevisiae (16). The mutation we identified resides in a region of the protein that could be involved in the recognition and charging of tRNA (3). Gcn4p is a transcriptional activator of genes involved in amino acid biosynthesis (6). Uncharged tRNAs accumulate during amino acid starvation, which then bind and activate the Gcn2p kinase, resulting in translational derepression of GCN4 (7). Thus, we evaluated whether expression of GCN4 is derepressed in the cdc64 strain under nonpermissive conditions.

Reporter constructs carrying the Escherichia coli gene lacZ under the control of promoter and mRNA 5′ leader sequences from GCN4 (plasmid p180, a gift from Alan Hinnebusch) (10), CYC1 (plasmid pLG669-Z) (4), and CLN3 (13) were transformed into strain TC4-23-1. β-Galactosidase expression was assayed as described previously (4) from liquid cultures under permissive and nonpermissive conditions (Table 1) in synthetic complete medium without uracil and supplemented with 2% dextrose. Only β-galactosidase expression from the pGCN4-lacZ construct was significantly derepressed at 37°C (Table 1), consistent with the notion that cdc64 cells accumulate uncharged tRNA under nonpermissive conditions.

TABLE 1.

cdc64 derepresses GCN4 expression

Reporter plasmid	β-Galactosidase activity (U)a at:
	23°C	37°C
pGCN4-lacZ	16.0 ± 1.0	92.2 ± 14.7
pCYC1-lacZ	29.9 ± 4.7	31.3 ± 6.3
pCLN3-lacZ	11.3 ± 0.6	13.4 ± 4.0

^aβ-Galactosidase activity (means ± standard deviations) from multiple independent transformants of each reporter construct is shown. 

We have previously shown that synthesis of the G₁ cyclin Cln3p is very sensitive to the availability of translation preinitiation complexes, linking protein synthesis and cell division (13). However, the processes that affect GCN4 expression, consistent with the data shown in Table 1, do not affect Cln3p synthesis (13). Certain mutations that affect formation of the 43S translation preinitiation complex cause accumulation of monosomes and repress Cln3p synthesis (13). Deletion of CLN3 exacerbates the effects of these mutations (14). However, based on polysomal fractionation experiments with cdc64 cells under nonpermissive conditions, it was concluded that translation initiation of CLN3 mRNA is not repressed (data not shown). In addition, deletion or overexpression of CLN3 did not significantly modify the cdc64 G₁ arrest phenotype (data not shown). Therefore, cdc64's effects on cell cycle progression probably do not involve Cln3p.

Our results suggest that cdc64-1 affects cell division indirectly. However, there must be some degree of specificity in cdc64's effects because mutations that severely and indiscriminately inhibit cell growth cause a random arrest at multiple stages of the cell cycle (13). For example, the translation initiation rates of ACT1 and CLN3 were not affected by the cdc64 mutation, since their polysomal profiles did not vary between the permissive and restrictive temperatures (data not shown). This is only the second mutation in an aminoacyl-tRNA synthetase shown to cause G₁ arrest in S. cerevisiae. CDC60 encodes the cytoplasmic leucyl-tRNA synthetase (8). It has been reported that cells carrying the cdc60-1 mutation arrest at Start with a type II phenotype at nonpermissive conditions (8), identical to the early G₁ arrest of cells carrying the cdc64-1 mutation (1). Interestingly, mutations in glutamine-tRNA (cdc65) also arrest cells in G₁ and affect developmental pathways as well (11). It remains to be determined how improper tRNA recognition causes specific defects in cell cycle progression.