Vol. 53 No. 4/2006, 739–745 Regular paper on-line at: www.actabp.pl

The ER24 aci (acidification) mutant of Saccharomyces cerevisiae excreting protons in the absence of glucose was transformed with a multicopy yeast DNA plasmid library. Three different DNA fragments restored the wild-type phenotype termed Aci- because it does not acidify the complete glucose medium under the tested conditions. Molecular dissection of the transforming DNA fragments identified two multicopy suppressor genes YJL185C, YJR129C and one allelic YLR376C. Disruption of either of the three genes in wild-type yeast strain resulted in acidification of the medium (Aci+ phenotype) similarly to the original ER24 mutant. These data indicate the contribution of the ER24 gene product Ylr376Cp and of the two suppressor gene products Yjl185Cp and Yjr129Cp to a complex regulation of the glyoxylate cycle in yeast.


INTRoDuCTIoN
The yeast Saccharomyces cerevisiae growing on complete Kok medium containing glucose and bromocresol purple forms gray colonies on a violet background (Kok et al., 1975;Goffeau, 2000).At least 17 non-allelic single-gene aci (acidification) mutants whose colonies are surrounded by a yellow acidic zone have been isolated.This determined the Aci + mutant phenotype (Gonchar et al., 1990;Boniewska-Bernacka et al., 1998;Grochowalska et al., 2003).The mutant cells show abrupt proton liberation when suspended in distilled water.Gas chromatography analysis indicated that the Krebs/glyoxylate cycle intermediates are the proton carriers excreted in the medium and that these intermediates preexisted in the cells at the start of the acidification test (Machnicka et al., 2004).We report here a detailed study of the ER24 mutant that belongs to the complementation group III of the aci mutants (Grochowalska et al., 2003).
Growth conditions.The strains were held on slants at 4 o C and cultivated at 28 o C in liquid medium with shaking.The acidifying mutants were screened

2006
E. Boniewska-Bernacka and others on Kok medium as forming colonies surrounded by yellow zones (Kok et al., 1975).
Proton extrusion.Proton extrusion by yeast cells was estimated according to Haworth et al. (1991) with a modification by Sigler et al. (1991) on a computer-linked pH-meter (Radomski et al., 1995).
Genetic manipulations.Strain construction, crossing and tetrad analysis were carried out by standard genetic techniques (Sherman et al., 1987).Tetrads were dissected with a Singer micromanipulator.For chromosomal mapping, the mutants were crossed with a collection of 16 cir 0 tester strains, each containing plasmid DNA integrated near the centromere of suitable chromosomes (Wakem & Sherman, 1990).
Cloning strategies.aci strains with the Δura3 disruption (introduced by recombination) were transformed with a yeast DNA library on the plasmid shuttle vector pFL44L (Bonneaud et al., 1991) using the lithium acetate procedure (Gietz & Woods, 1998).The transformants selected for uracil prototrophy were screened for the Aci -phenotype on Kok medium.Plasmid DNA was isolated from the yeast transformants and cloned in Escherichia coli.Restriction fragments from the cloned plasmids were isolated from low melting point agarose gel (Sambrook et al., 1989).Sequencing of about 100 nucleotides from the ends of yeast DNA fragments was carried out using the dideoxy chain termination procedure (Sanger et al., 1977).
Physical mapping of the plasmid DNA isolated from the transformants was performed according to the recommendations of Sherman et al. (1987).
The stability of plasmids in the transformants was tested by mitotic segregation.Yeast cells were serially cultivated for about 24 generations in complete YPD medium.After each passage a sample of about one hundred cells was plated on Kok medium and the fraction of Aci + colonies was determined (Kok et al., 1975).

Genetics of the aci1 mutant ER24
The mutant ER24 excreted acids during growth on glucose complete medium.It was unable to grow on glycerol, acetate, ethanol, or citrate and this feature co-segregated 2 : 2 with the Aci + character in 20 tetrads from a cross with the isogenic wildtype stain F87-24B (not shown).
The mutated genes responsible for acidification are recessive, as the diploid obtained from a cross to the isogenic strain F87-24B MATa his3 did not excreting acids.The inability to grow on glycerol is not caused by cytochrome defects which were present (Claisse et al., 1992).
In order to localize the aci1 gene causing the ER24 phenotype, the Sherman collection of cir 0 tester strains for each chromosomes of both MATα and MATa were crossed to the ER24 mutant in which ura3 was introduced by recombination (Wakem & Sherman, 1990).
The diploids obtained were grown in complete YPD medium for at least 24 generations and an increased frequency of Aci -colonies was observed in the diploids of the mutant ER24 and chromosome X or XII tester strains, indicating that ER24 was located on one of these two chromosome (not shown).

Proton extrusion test
When cells of the wild-type strains D273-10B/A1 or F87-24B were suspended in distilled water, the pH remained at a constant level of 5 to 6. Addition of glucose led to a decrease of pH (Fig. 1a and b).A further decrease of pH was stimulated by addition of potassium ions.
The same experiment was carried out with the ER24 mutant (Fig. 1c).In this case addition of yeast cells to distilled water led to an abrupt decrease of pH, and the following addition of glucose or potassium ions was without effect.
Acidification by the ER24 mutant without glucose addition does not occur at the cost of endogenous substrates as starvation for 16 h does not deprive the mutant of the acidification ability (not shown).
The glucose-independent acidification cosegregated with the Aci + character as shown in Fig. 1.d-h.According to the recesiveness of the aci mutations the heterozygous diploid excretes protons upon glucose addition although some abrupt decrease of pH immediately after addition of yeast cells to distilled water is also observed.In the tetrad, the two Aci + spore clones extruded protons upon addition of water and glucose had no effect.The two Aci -spores excrete acids upon glucose addition although as in the case of the diploid, the addition of water alone cause some slight abrupt pH decrease.

Isolation and cloning of genes complementing the mutant phenotype of ER24
In order to isolate genes restoring wild-type (Aci -) phenotype in the ER24 mutant we used a yeast bank gene library on multicopy plasmid pFL44L obtained thanks to courtesy of Dr. Francois Lacroute (Gif-sur-Yvette, France).
The ∆ura3 marker was incorporated by recombination into the mutant used a recipient in the transformation.The recombinant aci1 ∆ura3 cells were transformed with the DNA bank and the transformants ER24 selected for Ura prototrophy were screened for the Aci -phenotype.Three transformants of ER24 were isolated and designated tER24-1, -2 and -3.

Test of mitotic segregation
In order to check the stability of the plasmids carrying the genes restoring the Aci + phenotype, mitotic segregation of the transformed markers was determined.
After about 24 generations on complete medium (without selective pressure) approx.80% of YJL185C, YLR376C and YJR129C genes of Saccharomyces cerevisiae the cells retained the complete plasmids.This indicates that the transformants were rather stable.

sequencing and identification of yeast genome fragments
Plasmid DNA was isolated from the transformants.The plasmids were cloned in E. coli strain DH5α and for each yeast transformant three bacterial clones were isolated.The plasmids were restriction mapped with EcoRI and HindIII.For further investigation we chose three plasmids that differed in restriction maps (pER24-1/No1, pER24-3/No1, pER24-3/No2).After retransformation of the plasmids chosen into the ER24 strain, one hundred of transformants obtained (selected for Ura prototrophy) appeared Aci -.
The fragments of the yeast genome carried on the plasmids were cut out, separated by electrophoresis and their 3' and 5' ends were sequenced.The obtained nucleotide sequences were identified using the BLAST program (Altschul et al., 1990).

Identification of the gene(s) restoring wild-type phenotype in the ER24 mutant
As cloned, the yeast genomic fragment of plasmid tER24-1/No1 restoring the wild-type Aci -phenotype contained two complete genes.In order to determine which of these genes is responsible for the suppression, we successively cut out the genes YJL184W and YJL185C with the restriction enzyme SalI (Fig. 2), inserted them into empty plasmid pFL44L and transformed the ER24 mutant (Table 2).The Aci -(wild-type) phenotype was found only in transformants that obtained the YJL185C gene.

2006
E. Boniewska-Bernacka and others order to determine which of them was responsible for the suppression, we successively cut out the genes with restriction enzymes (Fig. 2).Out of the four constructs obtained only one with the YLR376C gene restored the wild-type Aci -phenotype in the ER24 mutant.Plasmid pER24-3/No2 restoring the wild-type Aci -phenotype in ER24 contained yeast DNA carrying four genes: STR2, YJR129C, SNR3 and YJR 128W (Fig. 2).The genes were separately subcloned in pFL44L.Restoration of the wild-type phenotype was obtained only with the plasmid carrying the YJR129C gene (Fig. 2).

Disruption of the YJL185C, YLR376C and YJR129C genes
Each of the genes restoring the wild-type phenotype in ER24, YJL185C, YLR376C and YJR129C, was disrupted in the wild-type strain.The disruptants (from Euroscarf) were unable to grow on glycerol and produced the acid yellow halo on Kok medium.In the liquid acidification test, each mutant extruded protons without glucose stimulation.The disruptants were recessive as they gave wildtype Aci -diploids when crossed with the wild-type strain D273-10B/A 1 .In the diploids the phenotype of abrupt proton extrusion in the absence of glucose gave monogenic 2:2 segregation.In other words, they showed the typical mutated Aci + phenotype.
The suppression appears non-specific as each of the wild type genes YJL185C, YLR376C and YJR129C on a multicopy plasmid restored the Aci - wild-type phenotype in several other non-allelic aci mutants (Table 3).This will be explained in more detail in a next publication.
In order to determine which of the three genes is allelic to ER24 (aci1) mutation, we carried out complementation experiments with three disruptants.The disruptants, named YJL185C, YLR376C and YJR129C, were crossed with the original ER24 mutant and the obtained diploids were tested for the acidification phenotype.One of the analyzed diploids, ER24 x YLR376, was Aci + indicating allelism of YLR376 with the ER24 mutant aci1 gene.To confirm this finding, the YLR376 gene was introduced into the centromeric plasmid pFL38 and the original ER24 mutant was transformed with the obtained construct.The resulting transformant was Aci -, as expected.The other two genes (YJL185C and YJR129C) when similarly tested, appeared to be multicopy, but not single-copy, suppressors of the mutant Aci + phenotype.

DIsCussIoN
The aci1 mutant ER24 was transformed with a genomic DNA library on the multicopy shuttle plas-mid pFL44L.We found that transformants carrying the YJL185C, YLR376C or YJR129C genes from chromosomes X, XII and X, respectively, restored the wild-type Aci -phenotype in the ER24 mutant.
The suppression is unspecific as each of this three genes restores the wild-type phenotype of many other non-allelic aci mutants.When disrupted in a wild-type strain, each of the genes gives the Aci + phenotype and prevents growth on glycerol.According to the test of complementation one of these genes appeared allelic with the ER24 mutation.The other two genes appeared to be multicopy suppressors of the ER24 mutant Aci + phenotype.
Their function is far from being fully understood.Out of several possibilities considered also in our previous publication (Machnicka et al., 2002) we propose the following two most likely interpretations.Either the multicopy suppressors modify the components of the cellular membrane in such a way that leakage of acid intermediates is prevented or they function in a complex regulatory mechanism that turns on the glyoxylic acid cycle when the keto acid substrates for amino acid synthesis are not required.The second possibility may be more likely, at least for the YLR376C and YJR185C genes.Indeed, while Yjl129Cp is still of unknown function, recent evidence suggests possible regulatory functions for Ylr185Cp either in chromatin action or in protein trafficking and for Yjr376Cp at the level of DNA maintenance.The protein Ylr185Cp is a putative methyltransferase.Its expression is drastically increased in Histone 4 mutants and this protein may be involved in microtubule biogenesis as a member of the folding of a prechaperone complex (Huh et al., 2003).Ylr376Cp/Psy3p has been reported to be part of a large novel protein complex involved in error-free DNA repair (Shor et al., 2005).How such proteins may regulate the expression and or activity of the glyoxilic and tricarboxilate cycles remains to be determined.The described genes seem to represent a new family of genes encoding a new family of proteins.