OUP user menu

Identification and disruption of the gene encoding the K+-activated acetaldehyde dehydrogenase of Saccharomyces cerevisiae

Wayne D Tessier, Philip G Meaden, Francis M Dickinson, Melvin Midgley
DOI: http://dx.doi.org/10.1111/j.1574-6968.1998.tb13063.x 29-34 First published online: 1 July 1998


The identity of the gene encoding the mitochondrial K+-activated acetaldehyde dehydrogenase (K+-ACDH) of Saccharomyces cerevisiae has been confirmed. The gene is situated on the right arm of chromosome XV, bears the systematic name YOR374w and the deduced product shows significant homology to other members of the S. cerevisiae aldehyde dehydrogenase (ALDH) family. YOR374w has now been assigned the gene name ALD7. The N-terminal amino acid sequences of K+-ACDHs purified from several diverse strains of S. cerevisiae were determined, and found to have 81–100% identity in alignments with the product of ALD7. Haploid mutants containing a deletion of ALD7 were constructed and, in these strains, the K+-ACDH was not detectable under any growth conditions examined. The activity of the Mg2+-activated acetaldehyde dehydrogenase (Mg2+-ACDH), encoded by ALD6, remained at wild-type levels in the mutants. Growth on glucose was not affected in the mutants lacking ALD7 (in contrast to the behaviour of ald6 mutants), whereas growth on ethanol was severely impaired. This observation, together with previous work by our group, shows that both the Mg2+- and K+-ACDHs are required for growth on ethanol, whilst only the former plays a role during growth on glucose.

Key words
  • Saccharomyces cerevisiae
  • Ethanol metabolism
  • Acetaldehyde dehydrogenase
  • Aldehyde dehydrogenase family
  • Protein purification

1 Introduction

Two acetaldehyde dehydrogenases (ACDHs) have been characterised in Saccharomyces cerevisiae[1]. One is the Mg2+-activated, NADP+-linked dehydrogenase (Mg2+-ACDH) present in the cytosol [2, 3] whereas the second is a K+-activated, NAD(P)+-linked enzyme (K+-ACDH) found in mitochondria [4]. Both are members of an aldehyde dehydrogenase (ALDH) family that exists in S. cerevisiae (Table 1) [5]. We have recently identified the gene (ALD6) encoding the Mg2+-ACDH [5] and from the properties of mutants deleted for this gene, confirmed the proposal that a cytosolic pathway for the formation of acetyl-CoA exists in S. cerevisiae[6].

View this table:
Table 1

Members of the S. cerevisiae aldehyde dehydrogenase family (modified and updated from [5])

ORF and/or gene nameaGenBank accession numberGene productReference
ALD1M57887Mitochondrial (predicted)[7]
YMR170c (ALD2)X85987Induced by osmotic stress (predicted cytosolic)[18]
YMR169c (ALD3)bZ49705
YPL061w (ALD6)U39205Cytosolic, Mg2+-activated acetaldehyde dehydrogenase[5]
YOR374w (ALD7)Z75282Mitochondrial, K+-activated acetaldehyde dehydrogenase[8, 9]
YER073wcU18814Mitochondrial (predicted)
  • aThe gene names ALD1, ALD2, ALD3, ALD6 and ALD7 are those currently assigned by the Saccharomyces Genome Database [10].

  • bThe product of this gene is 92% identical to that of ALD2.

  • cThe product of this ORF is 63% identical to that of ALD7.

Several studies have served to identify the gene encoding the K+-ACDH in S. cerevisiae. ALD1 was proposed on the basis that it could restore growth on ethanol to a mutant lacking K+-ACDH activity [7]. However, ALD1 cannot be found either in the strain used in our previous work [5], or in the published yeast genome sequence. Two other candidates also exist (Table 1), based on strong homology of the deduced products to the Mg2+-ACDH [5]; these are YER073w and YOR374w. The latter gene (GenBank accession number Z75282) has been implicated in two separate studies [8, 9] and has been assigned the name ALD7 by the Saccharomyces Genome Database [10]. Nevertheless, at present both the genetic identity and role of the K+-ACDH are unclear.

In an attempt to resolve any ambiguity about the identity of the gene and to remove any doubts that may have arisen during previous studies of the K+-dependent ACDH from using different yeast strains, we have undertaken the present work. The study has two objectives, which are (i) to identify the gene encoding the protein from several diverse strains and (ii) to investigate the role of the enzyme in ethanol catabolism.

2 Materials and methods

2.1 Yeast strains and media

The diploid yeast strain YPH501 (MATa/MATαura3-52/ura3-52 lys2-801/lys2-801 ade2-101/ade2-101 trp163/trp163 his3200/his3200 leu21/leu21) [11] was the host for deletion of ALD7. YPH499 is an isogenic haploid (MATa) and was used as a reference strain for the determination of Mg2+- and K+-ACDH activities. RY270 to RY275 (ura3-52 lys2-801 ade2-101 trp163 his3200 leu21 ald7::HIS3) were obtained by sporulation of YPH501 following deletion of one copy of ALD7 from this strain. Baker's yeast was provided by Mauri Products Ltd (Hull, UK) and the brewing yeast strains BB1 and BB11 were provided by Bass Brewers (Burton-on-Trent, UK).

Basal minimal medium consisted of 0.67% (w/v) yeast nitrogen base with appropriate nutritional supplements provided at a concentration of 100 mg l−1. To this minimal medium was added the particular carbon source required (glucose or ethanol).

Complex medium consisted of 1% (w/v) yeast extract, 2% (w/v) bacteriological peptone, 0.5% (w/v) glucose and adenine (100 mg l−1).

Maximum specific growth rates of yeast strains were determined in shake-flask cultures from the increase in culture absorbance at 540 nm during exponential growth at 30°C.

2.2 Gene deletion

The method of Baudin et al. [12] was used for the deletion of ALD7, using a PCR product amplified from the HIS3 gene and flanked by sequences 40 nucleotides in length to direct recombination with ALD7 in vivo. The forward primer (ALD7-F1) was: 5′-GGATTAGAAGTATCTGGAAAACCAACCAAGAAAACTACAACTCTTGGCCTCCTCTAG-3′, in which the underlined portion corresponds to nucleotides −59 to −20 of the ALD7 sequence (relative to the first nucleotide of the translational start codon). The reverse primer (ALD7-R1) was: 5′-TGTAAGCATCGATTGGACACCAGGCTTATTGATGACCTTATCGTTCAGAATGACACG-3′, with the underlined portion in this primer being complementary to nucleotides +1597 to +1558 in the ALD7 sequence.

2.3 Southern blotting and hybridisation

Procedures for the preparation of total DNA, and digoxigenin-labelled DNA probes, and for performing Southern blotting and hybridisation, were as described previously [5, 13]. A probe for HIS3 was produced from the HIS3 gene using the primers ALD7-F1 and ALD7-R1 in the PCR. For preparation of the ALD7 probe, the forward primer was 5′-CCTCCATTGGGAGACTTCAA-3′ (corresponding to nucleotides +41 to +60 of the ALD7 sequence) and the reverse primer 5′-GGATAATGTTTTGCACGGCT-3′ (complementary to nucleotides +940 to +921).

2.4 Preparation of cell-free extracts

Cell-free extracts were prepared from actively growing cultures as follows. Cells were collected, washed and resuspended in 10 mM potassium phosphate, pH 7.5 containing 0.5 mM dithiothreitol (DTT) and 5% (v/v) glycerol. This and all subsequent steps were performed at 4°C. Cell disruption was achieved by two passages through a French press (35 MPa). The crude extract was then clarified by centrifugation at 8000×g for 15 min and used directly for enzyme assay.

2.5 Separation of ACDHs

Extracts were dialysed against extraction buffer and centrifuged for 1 h at 45 000×g. FPLC of the resulting extract was carried out on an LKB-Pharmacia system using a Bio-Rad 5-ml (Econo-Pac) hydroxyapatite mini-column. Proteins were eluted using a two-stage gradient of potassium phosphate buffer (10–400 mM) pH 7.5, containing 0.5 mM DTT and 5% (v/v) glycerol. The first 30 ml of the gradient was passed through at a flow rate of 1 ml min−1 (10–300 mM). In the second phase of the gradient (300–400 mM) 4.5 ml was collected at the same flow rate.

2.6 Acetaldehyde dehydrogenase assays

The ACDHs were assayed fluorometrically at 25°C using a filter fluorimeter [17], set to give full-scale deflection with 2 μM NAD(P)H. The Mg2+-activated ACDH was assayed under the following conditions: 50 mM Na-HEPES buffer pH 7.5, 16.25 μM NADP+, 3.75 mM MgCl2 and 125 μM acetaldehyde. The K+-activated ACDH was assayed with 62.5 mM Tris-HCl buffer pH 8.0, 375 μM NAD+, 0.75 mM DTT, 62.5 mM KCl, 1 mM pyrazole, 5 mM NaN3 and 1.5 mM acetaldehyde. The protein content of cell extracts was measured by the biuret method, using bovine serum albumin as standard.

2.7 Purification and N-terminal amino acid sequencing of a mitochondrial acetaldehyde dehydrogenase

The overall purification procedure was as follows: preparation of crude cell-free extract, heat treatment of crude extract, stepwise elution of protein from a DEAE column, and finally gradient elution of protein from an hydroxyapatite column by FPLC. The enzyme was purified from several strains of S. cerevisiae by the method of Bostian and Betts [14] with the following modifications. Cells were grown in shake flask at 30°C for 36 h in complex medium (1.5 l) and cell-free extracts were prepared as already described but with 10 mM PMSF added to the extraction buffer. The procedure used here did not include the second heat step, the acid precipitation step or the ammonium sulfate step and the DEAE treatment employed here differed in that a 200-ml column was used instead of a batch adsorption process. Protein was eluted from the column in a stepwise manner by the addition of 10 mM potassium phosphate, pH 7.5 containing 0.5 mM DTT, 5% (v/v) glycerol and increasing concentrations of KCl. With 50 mM KCl added inactive protein was eluted. Addition of 350 mM KCl elute the active enzyme. The affinity chromatography step was replaced by purification using hydroxyapatite (a 5-ml mini-column attached to an LKB Pharmacia FPLC system). The protein was eluted with a linear gradient of 10–400 mM potassium phosphate, pH 7.5 containing 0.5 mM DTT and 5% (v/v) glycerol. In total, 30 ml was collected at a flow rate of 1 ml min−1.

This purification procedure was suitable for small volumes of extract and was considerably faster than any previously published procedure, normally requiring 10 h for completion. The speed of the process and the use of PMSF were intended to minimise any proteolysis of the enzyme since this is a well-known problem with K+-ACDH [15]. SDS-PAGE was performed as previously described [16]. The proteins were typically 96% pure as determined by densitometry of an SDS-PAGE gel. The SDS-PAGE analysis also demonstrated that no detectable proteolysis of the enzymes had occurred during the purification procedure.

The purified K+-ACDH from four strains of S. cerevisiae was partially sequenced from the N-terminus by the Microchemical Facility at the Babraham Institute (Cambridge, UK).

3 Results

3.1 Deletion of ALD7 encoding the mitochondrial ACDH

The N-terminal amino acid sequence of the purified K+-ACDH from S. cerevisiae YPH499 was found to be FILPMTVPIKLPNGLEQYQP. A TFASTA search of GenBank found the best match for this sequence to be the product of the gene ALD7 (Table 1). This gene is situated on the right arm of chromosome XV of S. cerevisiae and the product shows significant similarity to other members of the S. cerevisiae ALDH family.

Fig. 1 shows the N-terminal amino acid sequence of the K+-ACDH purified from YPH499 and from three other strains of S. cerevisiae, aligned with the same sequence deduced from a translation of ALD7. The same purification method was used for the protein from each of these four strains. All four preparations behaved identically on SDS-PAGE and had similar specific activities (30 μmol min−1 (mg protein)−1). This was comparable to the activity measured by Bostian and Betts [14].

Figure 1

Alignment of N-terminal amino acid sequences of K+-activated ACDH purified from strains of S. cerevisiae. All sequences (except that of ALD7) are derived from enzyme purified in this laboratory by the procedure described in Section 2. The first 23 amino acids deduced from the translation of ALD7 (GenBank accession Z75282), representing the mitochondrial targeting sequence, have not been included.

The diploid yeast strain YPH501 was transformed to His+ with the 1.1-kb PCR product obtained by amplification of HIS3 with the primers ALD7-F1 and ALD7-R1. Total DNA isolated from three independent transformants was digested with XbaI and subjected to Southern blotting and hybridisation analysis using the HIS3 probe. A 5.4-kb fragment was detected in all three transformants (results not shown), consistent with the deletion of ALD7 as predicted from the yeast genome sequence. His+ haploids were obtained from each of the three diploid transformants, and two from each diploid were selected for analysis by Southern blotting and hybridisation. Total DNA from all six haploid strains contained the expected 5.4-kb XbaI fragment detected with the HIS3 probe, and failed to show hybridisation with the ALD7 probe (results not shown).

3.2 Deletion of ALD7 results in a loss of the mitochondrial, K+- activated ACDH

The activities of the K+- and Mg2+-ACDHs were measured in cell-free extracts prepared from both the wild-type and ald7 mutant strains (Table 2). In extracts prepared from mutants, there was no measurable activity for the K+-ACDH whereas the cytosolic Mg2+-ACDH was readily detected. Although this result was expected, it was nevertheless possible that inhibitors of the K+-ACDH were present, or that the activity of the Mg2+-ACDH was masking that of the K+-activated enzyme. The cell-free extracts from the wild-type strain and one of the ald7 mutants (RY270) were clarified and then applied to an hydroxyapatite column in order to separate the two enzyme activities. Two activity peaks were clearly observed for the wild-type strain YPH499 (ALD7) whereas only one peak was observed for the ald7 disruption mutant (Fig. 2). No residual activity for the K+-ACDH was detected in any of the fractions obtained from mutant cell-free extract.

View this table:
Table 2

K+- and Mg2+-ACDH activities in cell-free extracts of ALD7 and ald7 strains of S. cerevisiae

StrainSpecific activity (nmol min−1 (mg protein)−1)
YPH499 (ALD7)45±157.5±1
RY270 (ald7)ND58
RY271 (ald7)ND52
RY272 (ald7)ND60
RY273 (ald7)ND58
RY274 (ald7)ND59
RY275 (ald7)ND55
  • ND: not detected.

Figure 2

Separation of mitochondrial (K+-activated) and cytosolic (Mg2+-activated) acetaldehyde dehydrogenases (ACDHs) from S. cerevisiae by adsorption to an hydroxyapatite column. • K+-activated, mitochondrial ACDH from YPH499 (ALD7); ▪ Mg2+-activated, cytosolic ACDH from YPH499 (ALD7); ◯ Mg2+-activated, cytosolic ACDH from RY270 (ald7); dashed line, concentration profile (mM) of sodium phosphate in the elution buffer. No residual activity for the K+-ACDH was detected in any of the fractions obtained from mutant cell-free extract.

3.3 Deletion of ALD7 has no effect on growth on glucose but is detrimental to growth on ethanol

An analysis of all six ald7 mutants (RY270–RY275) showed no significant difference in maximum specific growth rate, compared to the parental strain, during shake flask growth on glucose minimal media. The maximum specific growth rate (μmax) for the wild-type was 0.33 h−1 and 0.30 h−1 for 0.5% (w/v) glucose and 5% (w/v) glucose respectively. The μmax values obtained for the mutants fell into the range 0.29–0.32 h−1 and 0.27–0.31 h−1 for 0.5% (w/v) glucose and 5% (w/v) glucose respectively.

Growth of the ald7 mutants on ethanol was poor. In minimal medium containing 1% (v/v) ethanol, the parental strain showed a doubling time of 9.7 h and exponential growth was observed. Under identical conditions the disruption mutants took on average 56 h to double in cell number and did not show exponential growth.

Viability following exposure to 1% (v/v) ethanol for 20 h was also measured. Cells of the ald7 mutants remained viable throughout exposure with the number of colony forming units increasing by on average 10%. This rules out the possibility of direct ethanol toxicity or, more indirectly, acetaldehyde toxicity as a cause for the lack of growth of the ald7 mutants on ethanol.

4 Discussion

The work presented here has confirmed the identity of the gene encoding the K+-ACDH in a range of S. cerevisiae strains as ALD7. This finding is consistent with previous work [8, 9] but not with that of Saigal et al. [7] who proposed ALD1. The origin and role of ALD1 remain obscure.

Our study has examined the role of the K+-ACDH by analysing the phenotype of mutants deleted for ALD7. The ald7 mutants show no measurable activity for the enzyme. Deletion of the gene has no detectable effect on the yeast during growth on glucose, in contrast to the behaviour observed in ald6 mutants [5].

It is, however, clear that the K+-ACDH does play a role during growth on ethanol since deletion of ALD7 renders the yeast incapable of growth on this substrate. This role has until now remained speculative [4], especially since the Mg2+-ACDH has also been proposed to function in ethanol catabolism [5]. Indeed, it appears that both ACDHs are required for growth on ethanol, with neither able to suppress the loss of the other. This contrasts with the findings made with alcohol dehydrogenases [19]. We have ruled out a trivial explanation for lack of growth on ethanol, namely that of accumulation of acetaldehyde to toxic levels, since viability is maintained in ald7 mutants exposed to ethanol. This is also true for ALD6 mutants (unpublished observations). From these results and from our previous work [5], it is apparent that each of the ACDHs has a distinct physiological role.


W.T. wishes to thank the Institute of Brewing for receipt of the Henry Mitchell Memorial Scholarship. The authors are grateful to Pat Barker at The Babraham Institute for his help and the protein sequencing data. We also wish to acknowledge Bass plc for funding the sequencing work.


View Abstract