A Gcn4p homolog is essential for the induction of a ribosomal protein L41 variant responsible for cycloheximide resistance in the yeast Candida maltosa.

Cycloheximide (CYH) resistance in the yeast Candida maltosa is based on the inducible expression of genes encoding a variant of ribosomal protein L41-Q, with glutamine at position 56 instead of the proline found in normal L41. The promoter of L41-Q2a, one of the L41-Q gene alleles encoding L41-Q, has an element similar to the Gcn4p-responsive element of Saccharomyces cerevisiae. In a previous study, this element was shown to be essential for the induction of L41-Q by CYH. In the present study, a C. maltosa GCN4 homolog, C-GCN4, was cloned. It had a long 5'-leader region with three upstream open reading frames. Enhanced expression of the C-GCN4 reporter fusion gene upon the addition of 3-aminotriazole or by mutations in start codons of all three upstream open reading frames indicates that C-GCN4 expression is under translation repression as was seen with GCN4. The C-GCN4-depleted mutant was unable to grow in a nutrient medium containing CYH and did not express L41-Q genes. Recombinant C-Gcn4p bound to the consensus DNA element for Gcn4p, 5'-(G/A)TGACTCAT-3', located upstream of L41-Q2a. Thus, C-Gcn4p, which likely functions in the general control of amino acid biosynthesis, is essential for the expression of L41-Q genes.

Cycloheximide (CYH) 1 is an antibiotic that inhibits the peptide elongation reaction on eukaryotic ribosomes by binding specifically to the 60 S subunit. Numerous yeast species, including Saccharomyces cerevisiae, are sensitive to low concentrations of CYH. However, some species are by nature resistant to CYH, and their resistance is based on the production of a variant ribosomal protein L41 (1). CYH-sensitive yeasts have a conventional P-type L41 protein, with proline as amino acid residue 56, whereas CYH-resistant yeasts have a Q-type L41 protein, with glutamine as residue 56 (1). Although the latter yeast species are constitutively resistant to CYH, one species of asexual yeast, Candida maltosa, develops resistance only after the addition of CYH to the culture medium (2). It has been shown previously that C. maltosa has multiple genes that encode the two types of L41 ribosomal proteins; L41-P genes encode P-type L41 proteins, and L41-Q genes encode Q-type L41 proteins. There are two L41-P genes (L41-P1 and L41-P2) on the partially aneuploid genome of C. maltosa; the former is comprised of two alleles, L41-P1a and L41-P1b. The expression of L41-P genes is constitutive. There are also three L41-Q genes, L41-Q1, L41-Q2, and L41-Q3. L41-Q1 and L41-Q2 are each comprised of two alleles, L41-Q1a and L41-Q1b, and L41-Q2a and L41-Q2b, respectively. The two L41-Q2 alleles and probably L41-Q3 are induced in the presence of CYH and are responsible for CYH resistance in this yeast, whereas the L41-Q1 alleles are not active and are not induced (3,4). Mutant strains have been constructed in which either the expressible L41-Q genes or all L41-P genes were eliminated. It has been shown that mutants lacking all functional L41-Q alleles are sensitive to CYH, whereas mutants lacking all L41-P genes are constitutively resistant to CYH. Furthermore, ribosomes with only P-type L41 have been shown to be sensitive to CYH in in vitro translation, whereas those with only Q-type L41 were resistant to CYH (3).
The promoter of L41-Q2a, one of the two alleles of CYHinducible L41-Q2, has a Gcn4p-responsive element (GCRE)like element that contains an AP-1 consensus sequence (5Ј-(G/ A)TGACTCAT-3Ј) and an adjacent GT-rich region (5). Deletion of the GCRE-like element or mutations of its consensus nucleotides abolish CYH-responsive transcription of the L41-Q2a promoter (5), indicating the essential role of this element in the induction of Q-type L41 protein.
In S. cerevisiae, starvation of single amino acid species results in increased transcription of genes encoding biosynthetic enzymes for amino acids, purine, and aminoacyl tRNAs. This regulatory network is known as the general control of amino acid biosynthesis (6 -8). Recently, a regulatory pathway that was homologous to the general control response was identified in mammalian cells, highlighting evolutionary conservation in the organization of the pathway (9). Amino acid starvation results in increased synthesis of the transcription activator, Gcn4p, which binds to specific DNA sequences called GCREs (10 -12). These elements are seen in the promoter regions of the above genes and are responsible for their increased expression. GCRE is sometimes accompanied by another element located in its upstream region. Gcn4p belongs to the so-called basic zipper (bZIP)-type regulatory protein family (13,14) and is composed of the DNA-binding and leucine zipper domains (15,16). Genes encoding Gcn4p homologs have been isolated from various fungi and yeast: cpc-1 in Neurospora crassa (17), cpcA in Aspergillus niger (18), cpcA in Aspergillus nidulans (19), cpCPC1 in Cryphonectria parasitica (20), and CaGCN4 in Candida albicans (21).
Under non-starvation conditions, Gcn4p is produced at low levels due to the negative effects of four upstream open reading frames (uORFs) in the leader sequence of GCN4 mRNA (10,11,22). Under amino acid starvation conditions, the production of Gcn4p is induced by overcoming the inhibitory effect of uORFs through the function of Gcn2p kinase (22)(23)(24)(25). In addition to this translational regulation, Gcn4p is controlled through various mechanisms that are not yet entirely clear. First, Gcn4p is degraded rapidly in a SCF cdc4 ubiquitin ligase-dependent manner in vivo. This degradation process requires the phosphorylation of specific residues in the Gcn4p activation domain and the cyclin-dependent protein kinases Pho85p and Srb10p. Amino acid starvation or CYH administration somehow represses phosphorylation of Gcn4p and extends its half-life (25,(27)(28)(29). Second, Cpc2p regulates the transcription activation function of Gcn4p in a negative manner via an unknown regulatory mechanism. Cpc2p does not affect Gcn2p-mediated GCN4 expression, the stability of Gcn4p, or the binding ability of Gcn4p to GCRE (25,30). Third, glucose addition or UV irradiation transiently activates Gcn4p through the Ras/cAMP pathway (25,31). This activation depends on Gcn2p function but apparently not on its phosphorylating activity for eIF2␣. Fourth, GCN4 mRNA levels rise to some extent under amino acid starvation or nitrogen starvation conditions (26,32), although transcriptional control seems to be of only minor importance in the regulation of Gcn4p function (32). Finally, decay of GCN4 mRNA is suppressed by a stabilizer element located between uORF4 and the GCN4 protein-coding region. This stabilization is mediated by a specific binding factor, Pub1, which is one of the major polyadenylated mRNA-binding proteins (33).
In C. maltosa, a similar general control system seems to be at work because amino acid starvation induces the expression of a histidine synthetic gene, C-HIS5, which has a GCRE-like element in its promoter. The same starvation conditions also induce the expression of L41-Q2a. Conversely, CYH also induces C-HIS5 (5). Deletion of the GCRE-like element or mutations in its consensus nucleotide sequence abolish CYH-induced transcription of the L41-Q2a promoter (5). These observations lead to the assumption that the same Gcn4p-dependent system is involved in both the general control of amino acid biosynthesis and induction of L41-Q2a by CYH.
This paper describes the cloning and characterization of C-GCN4, which encodes a Gcn4p homolog in C. maltosa. C-GCN4 mRNA was increased by the addition of CYH or 3-aminotriazole (3-AT), which blocks histidine biosynthesis, indicating the presence of control at the level of transcription or mRNA decay. The expression of C-GCN4 was tightly repressed through the function of the three uORFs in the 5Ј-leader region of C-GCN4 mRNA, and mutations that inactivated their start codons dramatically suppressed their repressive effect. C-GCN4 was essential for the induction of L41-Q genes by CYH, and recombinant C-Gcn4p bound specifically to the GCRE-like element of the L41-Q2a promoter. These findings indicate the central role of a Candida Gcn4p homolog in the induction of CYH resistance.
The ⌬C-GCN4 strain was constructed by sequential two-step gene replacements of the chromosomal C-GCN4 genes using the PvuII fragment of the C-GCN4 deletion cassette (34,35). Then the Ura Ϫ phenotype was rescued by transformation with a liner XhoI fragment containing the intact C-URA3 gene from vector pURAJ (36). Isolation of total DNA from yeast cells was described in a previous study (37). Transformations were carried out using the lithium-acetate yeast transformation method (38). Correct replacement was confirmed by Southern blot analysis.
Yeast cells were cultivated in minimal or YPD medium at 30°C. Appropriate supplements were added at the recommended amounts. Routine precultures were grown overnight, diluted with fresh media, and cultivated to an optical density of ϳ0.8 at 660 nm. Then, 50 g/ml CYH or 10 mM 3-AT for histidine starvation was added to the culture. After an appropriate growth period, cultures were harvested for RNA extraction and measurement of ␤-galactosidase activity.

Plasmids
Construction of C-GCN4 Deletion Cassettes-A 2.7-kb EcoRV fragment containing C-GCN4 was inserted into the SmaI site of vector pUC18 to obtain pUC-CGCN4. As a result of this, the entire C-GCN4 open reading frame was removed by digestion with AflII and Bst1107I. The residual fragment was blunt-ended and was combined with a blunt-ended SalI fragment containing C-ADE1 (39) to obtain pUC-CGCN41 or with a blunt-ended SalI fragment containing C-HIS5 (40) to obtain pUC-CGCN45.
Construction of pPL-CGCN4 with a C-GCN4-LAC4 Fusion Gene-A DNA fragment containing C-CGN4 promoter and its 5Ј-coding region was amplified using primers 5Ј-ACGCGTCGACAGGTCCTTATG-TATAGA-3Ј (primer 1) (SalI site is underlined) and 5Ј-AGTATACAG-CAGGAGTAGTAGC-3Ј (primer 2). The amplified fragment was digested with SalI and inserted between the SalI and SmaI sites of pPL1 (41). The resultant plasmid pPL-CGCN4 carries C-ADE1, CEN, ARS, and the C-GCN4-LAC4 fusion in which LAC4 is fused in-frame to the ninth codon of C-GCN4 (42). The correct nucleotide sequence was confirmed by sequencing.
Construction of pGEX-CGCN4 with a GST::C-GCN4 Fusion Gene-The wild-type C-GCN4 coding region was amplified from pUC-CGCN4 using the following primers: 5Ј-CGGGATCCATGTCTGCTACTACTC-CTGC-3Ј and 5Ј-GGAATTCTTAAAAGCTTATACCATGACTG-3Ј (BamHI and EcoRI sites, respectively, are underlined). The resultant PCR product was digested with BamHI and EcoRI and then inserted between the BamHI and EcoRI sites of pGEX-4T-3 (Amersham Biosciences). The correct nucleotide sequence was confirmed by sequencing.
Deletion Mutants of the L41-Q2a Promoter Region-All the deletion mutants of the L41-Q2a promoter region have been described previously (5).

Nucleotide Sequence Determination
The nucleotide sequences were determined by the dideoxy chain termination method using the Applied Biosystems sequencing system model 310 DNA sequencer (Applied Biosystems, Foster City, CA).

Northern Blot Analysis
Total RNA of C. maltosa was isolated according to the protocol described by Schmitt et al. (44). 10 g of total RNA from each sample was separated on a formaldehyde-agarose gel, transferred onto Hybond-Nϩ (Amersham Biosciences), and hybridized with a 32 P-labeled probe prepared from an 800-bp PCR-generated C-GCN4 fragment, a 500-bp PCR-generated LAC4 fragment, or a 1.1-kb HindIII fragment of L41-Q2a. The L41-Q2a probe was shown to cross-hybridize to L41-Q2b and L41-Q3 transcripts under the study conditions (data not shown). The 32 P-labeled 500-bp SalI-EcoRI fragment of ACT1 was used as an internal standard. Signals were quantified using a FLA-3000 (Fuji Film, Tokyo, Japan).

␤-Galactosidase Assay
␤-Galactosidase activities of permeabilized yeast cells were determined using ortho-nitrophenyl-␤-D-galactopyranoside as a substrate. Yeast cells were cultivated in minimal medium, and ␤-galactosidase activities after the addition of CYH or 3-AT were assayed as described previously (41).

Electrophoretic Mobility Shift Assay of DNA Binding
Complex formation between the purified recombinant C-Gcn4p and 32 P-labeled DNA fragments was performed in 15 l of 10 mM Tris-Cl (pH 7.6), 100 mM NaCl, 0.1 mM EDTA, 0.8 mM dithiothreitol, 2% glycerol, and 1 g of poly(dI-dC). The reaction mixture was incubated at room temperature for 10 min and immediately loaded onto a 5% (w/v) polyacrylamide gel with an acrylamide:bisacrylamide ratio of 50:1. It was then subjected to electrophoresis at 120 V in 1ϫ Tris/borate/EDTA as a running buffer. Three synthetic double-stranded oligonucleotides were end-labeled with 32 P and used in the binding assays. The first sequence, 5Ј-GGAGATACTTTTGGGGGGAAATTTATGAGTCATACG-TTT-3Ј, contained the consensus AP-1/GCRE and the adjacent GT-rich regions; the second, 5Ј-AAATTTATGAGTCATACGTTT-3Ј, contained the consensus AP-1/GCRE region; and the third, 5Ј-GGAGATACTTT-TGGGGGGAAATTT-3Ј, contained the GT-rich region.

RESULTS
Isolation of the C-GCN4 Gene-Based on the conserved region of S. cerevisiae Gcn4p and its homologs in various eukaryotic microorganisms, PCR primers were designed and used to amplify a part of the GCN4 homolog gene from the genomic DNA of C. maltosa. The resultant amplified DNA fragment (130 bp) was used to obtain an extended fragment of 250 bp by 5Ј-RACE-PCR. Using this fragment as a probe, a GCN4 homolog gene was isolated from the C. maltosa genomic library and was named C-GCN4.
To analyze regulation of C-GCN4 expression, we constructed a C-GCN4-LAC4 fusion gene in which the LAC4 open reading frame that codes for the ␤-galactosidase of Kluyveromyces lactis was fused to the ninth codon of C-GCN4 (41). Cells with the wild-type fusion gene showed more than a 20-fold increase in ␤-galactosidase activity at 7 h after addition of 3-AT (Fig. 1C). This marked activation could be mainly at the level of translation because the increase in mRNA levels of C-GCN4 and the fusion gene was less than 5-fold during the same period (Fig. 1,  C and D). The strain that carried the mutant C-GCN4-LAC4, in which all uORF start codons were base-substituted, showed very high levels of ␤-galactosidase activity without the addition of 3-AT (Fig. 1D). Potent repressive function of the second and third uORFs was observed (data not shown). These results suggest that the expression of C-GCN4 is under translational repression, which could be released by amino acid limitation. C-GCN4 might be also regulated at the level of mRNA since C-GCN4-LAC4 mRNA increased by 4ϫ the initial level at 1 h after the addition of 3-AT and was maintained at a high level (Fig. 1, C and D).
Disruption of C-GCN4 Simultaneously Abolishes the Inducible Resistance to CYH and the General Control Response in C. maltosa-To determine the function of C-GCN4p in inducible resistance to CYH, two C-GCN4 alleles of the strain CHAU1 (ade1, his5, ura3) were disrupted successively using two disruption cassettes, one with C-HIS5 and the other with C-ADE1 as selective markers, and the results were confirmed by Southern analysis. Because direct disruption of the second C-GCN4 allele was unsuccessful, the second disruption was done in the presence of a centromere plasmid carrying wild-type C-GCN4, which was later removed. The resultant ⌬c-gcn4 disruptor was then transformed with URA3 to avoid the influence of ura3 mutation and named ⌬C-GCN4.
The ⌬C-GCN4 strain grew as well as the wild-type strain in YPD medium; however, it grew slowly and formed pseudohyphae in minimal medium (45). This phenotype is in contrast to that of a C. albicans CaGCN4-disrupted mutant that did not show pseudohyphal growth under amino acid starvation (21). These results indicate that C-GCN4 is not essential for C. maltosa but is required for better growth of this yeast in minimal medium. In the presence of 50 g/ml CYH, the ⌬C-GCN4 strain failed to grow in YPD ( Fig. 2A) or in liquid minimal medium (Fig. 2B) and was no longer inducible in resistance to CYH. These results indicate that C-Gcn4p is essential for the induction of CYH resistance in C. maltosa.
In C. maltosa, expression of L41-Q genes is required for resistance to CYH (3) . Fig. 3, A and B, shows that neither CYH treatment nor histidine starvation induced the expression of L41-Q in the mutant ⌬C-GCN4 in contrast to the case in the wild-type strain CMT100. The same conditions induced expression of C-HIS5 in ⌬C-GCN4 but at a far lower level than in CMT100. These results indicate that C. maltosa C-GCN4 has a critical role in the transcriptional induction of both L41-Q and C-HIS5 genes under either CYH treatment or histidine starvation conditions.
Specific Binding of C-Gcn4p to the GCRE-like Element of L41-Q2a-To investigate the DNA region bound by C-Gcn4p, C-Gcn4p combined with glutathione S-transferase at its N terminus was produced in E. coli and affinity-purified using a glutathione-Sepharose 4B (Amersham Biosciences) column. The purified fusion protein showed the expected mobility when subjected to SDS-PAGE (Fig. 4A). In the electrophoretic mobility shift assay in which the purified protein was employed, the intensity of shift bands of either a 39-bp synthetic DNA fragment containing both the GCRE-like element and the GT-rich region (Fig. 4G) or a 21-bp DNA fragment containing only the GCRElike element increased in proportion to the amount of purified recombinant fusion protein (Fig. 4, B and C). However, shift bands of a 24-bp DNA of only the GT-rich region were not observed (Fig. 4D). The shift band intensity of the 39-bp fragment was reduced along with the amount of unlabeled DNA fragments containing both the GCRE-like element and the GT-rich region or containing only the GCRE-like element (Fig. 4, E and F). Scintillation counting confirmed that the specific radioactivity of the 32 P-labeled DNA probe containing both the GCRE-like element and the GT-rich region was almost equal to that containing only the GCRE-like element. Therefore, the binding of C-Gcn4p to the fragment containing both the GCRE-like element and the GTrich region was ϳ2-fold higher than that to the fragment containing only the GCRE-like element (Fig. 4, B and C). These results indicate that C-Gcn4p binds specifically to the GCRE-like element of the L41-Q2a promoter and binds more favorably to the FIG. 1. Structural features of 5-untranslated region of C-GCN4 and induction of C-GCN4-LAC4 by the addition of 3-AT. A, determination of C-GCN4 transcription start sites. The two transcriptional start sites are indicated with arrows. Lane P is the primer extension reaction product against C-GCN4 mRNA. Lanes G, A, T, and C are products of the dideoxy sequencing reaction that were carried out with the same primer. B, location of three uORFs in the 5Ј-untranslated region of C-GCN4 mRNA. Nucleotides are numbered from the start codon of the main open reading frame. C and D, the strain CMT101 containing wild-type C-GCN4-LAC4 (C) or mutant C-GCN4-LAC4 with inactivated uORFs (D) was grown in minimal medium with appropriate supplements. 10 mM 3-AT was added at time 0 and was assayed for ␤-galactosidase activity at the indicated time points. Values represent the average of at least three independent measurements (filled bars). RNA samples were also prepared and analyzed with Northern hybridization. Signals were quantified using an image analyzer FLA-3000 (Fuji Film).
Open bars indicate C-GCN4-LAC4 mRNA level relative to ACT1 mRNA. The value at time 0 is set to 1. one accompanied by the GT-rich region. The observed C-Gcn4p binding properties to various GCRE-like regions are consistent with the promoter activities as shown in Fig. 4G. In a previous paper (5), the promoter containing the GCRE-like element without the GT-rich region was reported to be not active in the induction of L41-Q2a. In this study, however, we found that it was able to induce L41-Q2a expression to some extent after the addition of CYH (Fig. 4G, see the case of pPLGCRE). These results indicate that binding of C-Gcn4p to the GCRE-like element is essential for the induction of L41-Q2a expression and that the GT-rich region is necessary for full induction of L41-Q2a by helping C-Gcn4p to bind to the GCRE-like element. DISCUSSION This paper presents several lines of evidence in favor of transcription activation by a Gcn4p homolog through binding to a GCRE-like element of the L41-Q2a promoter as a mechanism of inducible resistance to CYH in C. maltosa (a schematic overview is given in Fig. 5). The experimental data supporting this notion are as follows: 1) disruption of C-GCN4 severely inhibited CYH-induced expression of L41-Q genes (Fig. 3A); 2) over-expression of C-GCN4 led to increased expression of L41-Q genes and earlier induction of CYH resistance (data not shown); 3) C-Gcn4p bound specifically to the GCRE-like element in vitro in the electrophoretic mobility shift assay (Figs. 4B-4F); and 4) deletion of the GCRE-like element from the L41-Q2a promoter abolished the CYH-induced expression of a reporter gene (Fig. 4G) (5). Genes encoding ordinary ribosomal proteins are repressed in response to various starvation or stress conditions (46 -49). Recent microarray analysis of Gcn4pdependent gene regulation in S. cerevisiae showed that 90 ribosomal protein genes were repressed by 3-AT (50). This paper therefore describes an interesting case in which a gene encoding a ribosomal protein gene was induced by CYH or 3-AT through regulation of a GCN4 homolog.
Genetic and molecular studies of the general amino acid control transcription activator Gcn4p of S. cerevisiae have revealed specific functional domains that include a transcription activation domain, a basic DNA binding domain, and a leucine zipper-based dimerization domain (15,16). C-Gcn4p has similar structural features and should have similar functions to those of Gcn4p. In fact, recombinant C-Gcn4p bound to the AP-1 consensus sequence (5Ј-(G/A)TGACTCAT-3Ј), as observed with Gcn4p and other homologs. Furthermore, C-Gcn4p was required for the induction of C-HIS5 mRNA under histidine starvation conditions (Fig. 3B), demonstrating the operation of a C-Gcn4p-mediated general amino acid control pathway in C. maltosa. Thus, it is now obvious that C-Gcn4p functions as a transcription activator both in the general control of amino acid synthetic genes and in the induction of L41-Q genes, which are responsible for CYH-resistance in C. maltosa. In S. cerevisiae, cellular Gcn4p is maintained at a low level unless it is needed and Gcn4p is phosphorylated by Pho85p and rapidly degraded in a SCF cdc4 ubiquitin ligase-dependent manner (28). Therefore, it is interesting to see how C-Gcn4p is maintained or activated after the addition of CYH.
S. cerevisiae Gcn4p homologs have a sequence similarity to c-Jun transcription factor, and like c-Jun, they bind to the AP-1 consensus response element. Gcn4p has been shown to have functional similarity to c-Jun (51). In mammalian cells, treatment with protein synthesis inhibitors such as CYH or anisomycin accentuates and prolongs the induction of immediate early genes, including c-jun, in response to polypeptide growth factors and phorbol esters (52)(53)(54)(55)(56). Although the mechanism of this response is not clear, this process is thought to result from a combination of mRNA stabilization, activation of intracellular signaling cascades, and interference with transcriptional down-regulation (52)(53)(54)(55)(56). If the effect of CYH treatment on C. maltosa C-GCN4 is similar to that on c-jun in mammalian cells, a similar activation mechanism might be working. CYH binds to the 60 S ribosomal subunit, its only known target, producing translation arrest and polysome stabilization, which might activate an as yet unknown intracellular signaling system and result in stabilization of mRNA or even stabilization of  C-Gcn4p. This signaling system, if any, might not be the same one as that resulting from derepression of C-Gcn4p synthesis upon amino acid limitation because disruption of two alleles of a gene encoding a Gcn2p homolog in C. maltosa made this yeast sensitive to 3-AT but not to CYH. 2 Although the CYH treatment of 50 g/ml severely inhibits protein synthesis of C. maltosa during an early stage of the CYH resistance induction, there remains some protein synthesis. In an experiment where 35 S-labeled methionine and cysteine were taken up for 60 min after 3 h from the addition of CYH, the radioactivity in the trichloroacetic acid-insoluble ribosome fraction of the CYH-treated cells was not less than 2.5% of that of untreated cells (data not shown). This level of protein  1-7 were 20, 0, 1, 2, 5, 10, and 20 nM, respectively. E and F, competition in binding of GST-C-Gcn4p to the GCRE-like element and the GT-rich region by cold DNA fragments with the same sequence as the probe (E) or with only the GCRE-like element (F). 20 nM GST-C-Gcn4p was added to the respective reactions. The reactions of lanes 3-5 contained unlabeled DNA at concentrations of 1ϫ, 10ϫ, and 100ϫ that of the labeled probe, respectively. G, various L41-Q2a 5Ј-promoter constructs were placed upstream of the ␤-galactosidase gene. Each number at the end of the left-hand bars indicates the position of the deletion end point from the start codon of the wild-type sequence (A of the ATG is designated as ϩ1). Cells with respective constructs were grown for 6 h after the addition of 50 g/ml CYH, and their ␤-galactosidase activities were measured. CYH was added when OD 660 reached 0.8. Values represent the average of at least three independent measurements. synthesis will explain why L41-Q is induced under the high concentration of CYH. This leakiness of inhibition by CYH is probably brought about by another CYH resistance mechanism because a C. maltosa mutant that lacks all the functional L41-Q genes, L41-Q2a, L41-Q2b, and L41-Q3 (L41-Q3 has no allele), was able to slowly grow on 25 g/ml CYH. 3 This resistance is at least in part due to a homolog to C. albicans CaMDR1, which gave high CYH resistance to S. cerevisiae (57). 4 The stabilization of C-GCN4 mRNA by CYH could have also helped the induction of L41-Q. We observed a transient increase of C-GCN4 mRNA level after the addition of CYH. Furthermore, in the presence of thiolutin, a potent RNA polymerase inhibitor (58), the C-GCN4 mRNA in the wild-type cells rapidly decreased to 28% of the initial level within 40 min, whereas addition of 50 g/ml CYH to the culture well suppressed this decrease. 5 Using the electrophoretic mobility shift assay, recombinant C-Gcn4p was shown to bind better to DNA fragments containing both the GCRE-like element and GT-rich region than to those containing only the GCRE-like element. Promoter activity analysis of L41-Q2a demonstrated that a promoter containing only the GCRE-like element was able to induce reporter gene expression to some extent, but the GT-rich region helped full expression of L41-Q2a. There was good correlation between the results of the electrophoretic mobility shift assay and those of promoter activity analysis by the reporter gene. It is known that a homopolymer-like sequence adjacent to GCRE stimulates Gcn4p-dependent activation (59). Such a short nucleotide sequence increases accessibility to the adjacent Gcn4p binding site, thereby allowing Gcn4p to bind efficiently to the site and to stimulate transcription activation from there. The GT-rich region of the L41-Q2a gene is not highly similar to the known homopolymer-like nucleotide sequences but may have a similar function in C-GCN4-dependent induction of L41-Q2 genes.
This study focused on the identification and functional analysis of a transcription factor for the expression of a variant L41 ribosomal gene and successfully identified C-GCN4 and postulated its function in the induction of CYH resistance. Further analysis of the upstream regulatory pathway in response to CYH will give new insight into how eukaryotic cells adapt to severe inhibition of protein synthesis.