Overexpressed maltose transporters in laboratory and lager yeasts: Localization and competition with endogenous transporters.

Plain and fluorescently tagged versions of Agt1, Mtt1 and Malx1 maltose transporters were overexpressed in two laboratory yeasts and one lager yeast. The plain and tagged versions of each transporter supported similar transport activities, indicating that they are similarly trafficked and have similar catalytic activities. When they were expressed under the control of the strong constitutive PGK1 promoter only minor proportions of the fluorescent transporters were associated with the plasma membrane, the rest being found in intracellular structures. Transport activity of each tagged transporter in each host was roughly proportional to the plasma membrane-associated fluorescence. All three transporters were subject to glucose-triggered inactivation when the medium glucose concentration was abruptly raised. Results also suggest competition between endogenous and overexpressed transporters for access to the plasma membrane.

Effect of C-terminal protein tags on pentitol and L-arabinose transport by Ambrosiozyma monospora Lat1 and Lat2 transporters in Saccharomyces cerevisiae.

Functional expression in heterologous hosts is often less successful for integral membrane proteins than for soluble proteins. Here, two Ambrosiozyma monospora transporters were successfully expressed in Saccharomyces cerevisiae as tagged proteins. Growth of A. monospora on l-arabinose instead of glucose caused transport activities of l-arabinose, l-arabitol, and ribitol, measured using l-[1-(3)H]arabinose, l-[(14)C]arabitol, and [(14)C]ribitol of demonstrated purity. A. monospora LAT1 and LAT2 genes were cloned earlier by using their ability to improve the growth of genetically engineered Saccharomyces cerevisiae on l-arabinose. However, the l-arabinose and pentitol transport activities of S. cerevisiae carrying LAT1 or LAT2 are only slightly greater than those of control strains. S. cerevisiae carrying the LAT1 or LAT2 gene fused in frame to the genes for green fluorescent protein (GFP) or red fluorescent protein (mCherry) or adenylate kinase (AK) exhibited large (>3-fold for LAT1; >20-fold for LAT2) increases in transport activities. Lat1-mCherry transported l-arabinose with high affinity (Km ≈ 0.03 mM) and l-arabitol and ribitol with very low affinity (Km ≥ 75 mM). The Lat2-GFP, Lat2-mCherry, and Lat2-AK fusion proteins could not transport l-arabinose but were high-affinity pentitol transporters (Kms ≈ 0.2 mM). The l-arabinose and pentitol transport activities of A. monospora could not be completely explained by any combination of the observed properties of tagged Lat1 and Lat2, suggesting either that tagging and expression in a foreign membrane alters the transport kinetics of Lat1 and/or Lat2 or that A. monospora contains at least one more l-arabinose transporter.

Three Agt1 transporters from brewer's yeasts exhibit different temperature dependencies for maltose transport over the range of brewery temperatures (0­20 °C).

Zero-trans rates of maltose transport by brewer's yeasts exert strong control over fermentation rates and are strongly temperature-dependent over the temperature range (20­0 °C) of brewery fermentations. Three α-glucoside transporters, ScAgt1(A60) (a Saccharomyces cerevisiae version of Agt1 from an ale strain), ScAgt1-A548V (a variant of ScAgt1(A60) with a single amino acid change in a transmembrane domain), and SbAgt1 (a Saccharomyces (eu)bayanus version from a lager strain), were compared. When expressed in the same laboratory yeast, grown at 24 °C and assayed at 0, 10, and 20 °C, SbAgt1 had the lowest absolute maltose uptake activity at 20 °C but smallest temperature dependence, ScAgt1-A548V had the highest activity but greatest temperature dependence, and ScAgt1(A60) had intermediate properties. ScAgt1(A60) exhibited higher absolute rates and smaller temperature dependencies when expressed in laboratory rather than brewer's strains. Absolute rates closely reflected the amounts of GFP-tagged ScAgt1(A60) transporter in each host's plasma membrane. Growth at 15 °C instead of 24 °C decreased the absolute activities of strains expressing ScAgt1(A60) by two- to threefold. Evidently, the kinetic characteristics of at least ScAgt1(A60) depended on the nature of the host plasma membrane. However, no consistent correlation was observed between transport activities and fatty acid or ergosterol compositions.

Adaptive evolution of the lager brewing yeast Saccharomyces pastorianus for improved growth under hyperosmotic conditions and its influence on fermentation performance.

An adaptive evolution method to obtain stable Saccharomyces pastorianus brewing yeast variants with improved fermentation capacity is described. The procedure involved selection for rapid growth resumption at high osmotic strength. It was applied to a lager strain and to a previously isolated ethanol-tolerant strain. Fermentation performance of strains was compared at 15 °P wort strength. A selected osmotolerant variant of the ethanol-tolerant strain showed significantly shorter fermentation time than the parent strain, producing 6.45% alcohol by volume beer in 4-5 days with mostly similar organoleptic properties to the original strain. Diacetyl and pentanedione contents were 50-75% and 3-methylbutyl acetate and 2-phenylethyl acetate 50% higher than with the original strain, leading to a small flavour change. The variant contained significantly less intracellular trehalose and glycogen than the parent. Transcriptional analysis of selected genes at 24 h revealed reduced transcription of hexose transport genes and increased transcription of the MALx1 and MALx2 genes, responsible for α-glucoside uptake and metabolism. It is suggested that an attenuated stress response contributes to the improved fermentation performance. Results show that sequential selection for both ethanol tolerance and rapid growth at high osmotic strength can provide strains with enhanced fermentation speed with acceptable product quality.

Identification of regulatory elements in the AGT1 promoter of ale and lager strains of brewer's yeast.

Agt1 is an interesting α-glucoside transporter for the brewing industry, as it efficiently transports maltotriose, a sugar often remaining partly unused during beer fermentation. It has been shown that on maltose the expression level of AGT1 is much higher in ale strains than in lager strains, and that glucose represses the expression, particularly in the ale strains. In the present study the regulatory elements of the AGT1 promoter of one ale and two lager strains were identified by computational methods. Promoter regions up to 1.9 kbp upstream of the AGT1 gene were sequenced from the three brewer's yeast strains and the laboratory yeast strain CEN.PK-1D. The promoter sequence of the laboratory strain was identical to the AGT1 promoter of strain S288c of the Saccharomyces Genome Database, whereas the promoter sequences of the industrial strains diverged markedly from the S288c strain. The AGT1 promoter regions of the ale and lager strains were for the most part identical to each other, except for one 22 bp deletion and two 94 and 95 bp insertions in the ale strain. Computational analyses of promoter elements revealed that the promoter sequences contained several Mig1- and MAL-activator binding sites, as was expected. However, some of the Mig1 and MAL-activator binding sites were located on the two insertions of the ale strain, and thus offered a plausible explanation for the different expression pattern of the AGT1 gene in the ale strains. Accordingly, functional analysis of A60 ale and A15 lager strain AGT1 promoters fused to GFP (encoding the green fluorescent protein) showed a significant difference in the ability of these two promoters to drive GFP expression. Under the control of the AGT1 promoter of the ale strain the emergence of GFP was strongly induced by maltose, whereas only a low level of GFP was detected with the construct carrying the AGT1 promoter of the lager strain. Thus, the extra MAL-activator binding element, present in the AGT1 promoter of the ale strain, appears to be necessary to reach a high level of induction by maltose. Both AGT1 promoters were repressed by glucose but their derepression was different, possibly due to a distinct distribution of Mig1 elements in these two promoters.

Selection from industrial lager yeast strains of variants with improved fermentation performance in very-high-gravity worts.

There are economic and other advantages if the fermentable sugar concentration in industrial brewery fermentations can be increased from that of currently used high-gravity (ca. 14 to 17 degrees P [degrees Plato]) worts into the very-high-gravity (VHG; 18 to 25 degrees P) range. Many industrial strains of brewer's yeast perform poorly in VHG worts, exhibiting decreased growth, slow and incomplete fermentations, and low viability of the yeast cropped for recycling into subsequent fermentations. A new and efficient method for selecting variant cells with improved performance in VHG worts is described. In this new method, mutagenized industrial yeast was put through a VHG wort fermentation and then incubated anaerobically in the resulting beer while maintaining the alpha-glucoside concentration at about 10 to 20 g.liter(-1) by slowly feeding the yeast maltose or maltotriose until most of the cells had died. When survival rates fell to 1 to 10 cells per 10(6) original cells, a high proportion (up to 30%) of survivors fermented VHG worts 10 to 30% faster and more completely (residual sugars lower by 2 to 8 g.liter(-1)) than the parent strains, but the sedimentation behavior and profiles of yeast-derived flavor compounds of the survivors were similar to those of the parent strains.

The temperature dependence of maltose transport in ale and lager strains of brewer's yeast.

Lager beers are traditionally made at lower temperatures (6-14 degrees C) than ales (15-25 degrees C). At low temperatures, lager strains (Saccharomyces pastorianus) ferment faster than ale strains (Saccharomyces cerevisiae). Two lager and two ale strains had similar maltose transport activities at 20 degrees C, but at 0 degrees C the lager strains had fivefold greater activity. AGT1, MTT1 and MALx1 are major maltose transporter genes. In nine tested lager strains, the AGT1 genes contained premature stop codons. None of five tested ale strains had this defect. All tested lager strains, but no ale strain, contained MTT1 genes. When functional AGT1 from an ale strain was expressed in a lager strain, the resultant maltose transport activity had the high temperature dependence characteristic of ale yeasts. Lager yeast MTT1 and MALx1 genes were expressed in a maltose-negative laboratory strain of S. cerevisiae. The resultant Mtt1 transport activity had low temperature dependence and the Malx1 activity had high temperature dependence. Faster fermentation at low temperature by lager strains than ale strains may result from their different maltose transporters. The loss of Agt1 transporters during the evolution of lager strains may have provided plasma membrane space for the Mtt1 transporters that perform better at a low temperature.

Improved fermentation performance of a lager yeast after repair of its AGT1 maltose and maltotriose transporter genes.

The use of more concentrated, so-called high-gravity and very-high-gravity (VHG) brewer's worts for the manufacture of beer has economic and environmental advantages. However, many current strains of brewer's yeasts ferment VHG worts slowly and incompletely, leaving undesirably large amounts of maltose and especially maltotriose in the final beers. alpha-Glucosides are transported into Saccharomyces yeasts by several transporters, including Agt1, which is a good carrier of both maltose and maltotriose. The AGT1 genes of brewer's ale yeast strains encode functional transporters, but the AGT1 genes of the lager strains studied contain a premature stop codon and do not encode functional transporters. In the present work, one or more copies of the AGT1 gene of a lager strain were repaired with DNA sequence from an ale strain and put under the control of a constitutive promoter. Compared to the untransformed strain, the transformants with repaired AGT1 had higher maltose transport activity, especially after growth on glucose (which represses endogenous alpha-glucoside transporter genes) and higher ratios of maltotriose transport activity to maltose transport activity. They fermented VHG (24 degrees Plato) wort faster and more completely, producing beers containing more ethanol and less residual maltose and maltotriose. The growth and sedimentation behaviors of the transformants were similar to those of the untransformed strain, as were the profiles of yeast-derived volatile aroma compounds in the beers.

Stimulation of zero-trans rates of lactose and maltose uptake into yeasts by preincubation with hexose to increase the adenylate energy charge.

Initial rates of sugar uptake (zero-trans rates) are often measured by incubating yeast cells with radiolabeled sugars for 5 to 30 s and determining the radioactivity entering the cells. The yeast cells used are usually harvested from growth medium, washed, suspended in nutrient-free buffer, and stored on ice before they are assayed. With this method, the specific rates of zero-trans lactose uptake by Kluyveromyces lactis or recombinant Saccharomyces cerevisiae strains harvested from lactose fermentations were three- to eightfold lower than the specific rates of lactose consumption during fermentation. No significant extracellular beta-galactosidase activity was detected. The ATP content and adenylate energy charge (EC) of the yeasts were relatively low before the [(14)C]lactose uptake reactions were started. A short (1- to 7-min) preincubation of the yeasts with 10 to 30 mM glucose caused 1.5- to 5-fold increases in the specific rates of lactose uptake. These increases correlated with increases in EC (from 0.6 to 0.9) and ATP (from 4 to 8 micromol x g dry yeast(-1)). Stimulation by glucose affected the transport V(max) values, with smaller increases in K(m) values. Similar observations were made for maltose transport, using a brewer's yeast. These findings suggest that the electrochemical proton potential that drives transport through sugar/H(+) symports is significantly lower in the starved yeast suspensions used for zero-trans assays than in actively metabolizing cells. Zero-trans assays with such starved yeast preparations can produce results that seriously underestimate the capacity of sugar/H(+) symports. A short exposure to glucose allows a closer approach to the sugar/H(+) symport capacity of actively metabolizing cells.

Analysis of beers from an 1840s' shipwreck.

Two bottles of beer from an about 170-year-old shipwreck (M1 Fö 403.3) near the Åland Islands in the Baltic Sea were analyzed. Hop components and their degradation compounds showed that the bottles contained two different beers, one more strongly hopped than the other. The hops used contained higher levels of ß-acids than modern varieties and were added before the worts were boiled, converting α-acids to iso-α-acids and ß-acids to hulupones. High levels of organic acids, carbonyl compounds, and glucose indicated extensive bacterial and enzyme activity during aging. However, concentrations of yeast-derived flavor compounds were similar to those of modern beers, except that 3-methylbutyl acetate was unusually low in both beers and 2-phenylethanol and possibly 2-phenylethyl acetate were unusually high in one beer. Concentrations of phenolic compounds were similar to those in modern lagers and ales.

The adenylate energy charge and specific fermentation rate of brewer's yeasts fermenting high- and very high-gravity worts.

Intracellular and extracellular ATP, ADP and AMP (i.e. 5'-AMP) were measured during fermentations of high- (15 degrees P) and very high-gravity (VHG, 25 degrees P) worts by two lager yeasts. Little extracellular ATP and ADP but substantial amounts of extracellular AMP were found. Extracellular AMP increased during fermentation and reached higher values (3 microM) in 25 degrees P than 15 degrees P worts (1 microM). More AMP (13 microM at 25 degrees P) was released during fermentation with industrially cropped yeast than with the same strain grown in the laboratory. ATP was the dominant intracellular adenine nucleotide and the adenylate energy charge (EC = ([ATP] + 0.5*[ADP])/([ATP] + [ADP] + [AMP])) remained high (>0.8) until residual sugar concentrations were low and specific rates of ethanol production were < 5% of the maximum values in early fermentation. The high ethanol concentrations (>85 g/l) reached in VHG fermentations did not decrease the EC below values that permit synthesis of new proteins. The results suggest that, during wort fermentations, the ethanol tolerance of brewer's strains is high so long as fermentation continues. Under these conditions, maintenance of the EC seems to depend upon active transport of alpha-glucosides, which in turn depends upon maintenance of the EC. Therefore, the collapse of the EC and cell viability when residual alpha-glucoside concentrations no longer support adequate rates of fermentation can be very abrupt. This emphasizes the importance of early cropping of yeast for recycling.

Monitoring yeast physiology during very high gravity wort fermentations by frequent analysis of gene expression.

Brewer's yeast experiences constantly changing environmental conditions during wort fermentation. Cells can rapidly adapt to changing surroundings by transcriptional regulation. Changes in genomic expression can indicate the physiological condition of yeast in the brewing process. We monitored, using the transcript analysis with aid of affinity capture (TRAC) method, the expression of some 70 selected genes relevant to wort fermentation at high frequency through 9-10 day fermentations of very high gravity wort (25 degrees P) by an industrial lager strain. Rapid changes in expression occurred during the first hours of fermentations for several genes, e.g. genes involved in maltose metabolism, glycolysis and ergosterol synthesis were strongly upregulated 2-6 h after pitching. By the time yeast growth had stopped (72 h) and total sugars had dropped by about 50%, most selected genes had passed their highest expression levels and total mRNA was less than half the levels during growth. There was an unexpected upregulation of some genes of oxygen-requiring pathways during the final fermentation stages. For five genes, expression of both the Saccharomyces cerevisiae and S. bayanus components of the hybrid lager strain were determined. Expression profiles were either markedly different (ADH1, ERG3) or very similar (MALx1, ILV5, ATF1) between these two components. By frequent analysis of a chosen set of genes, TRAC provided a detailed and dynamic picture of the physiological state of the fermenting yeast. This approach offers a possible way to monitor and optimize the performance of yeast in a complex process environment.

Characterization and functional analysis of the MAL and MPH Loci for maltose utilization in some ale and lager yeast strains.

Maltose and maltotriose are the major sugars in brewer's wort. Brewer's yeasts contain multiple genes for maltose transporters. It is not known which of these express functional transporters. We correlated maltose transport kinetics with the genotypes of some ale and lager yeasts. Maltose transport by two ale strains was strongly inhibited by other alpha-glucosides, suggesting the use of broad substrate specificity transporters, such as Agt1p. Maltose transport by three lager strains was weakly inhibited by other alpha-glucosides, suggesting the use of narrow substrate specificity transporters. Hybridization studies showed that all five strains contained complete MAL1, MAL2, MAL3, and MAL4 loci, except for one ale strain, which lacked a MAL2 locus. All five strains also contained both AGT1 (coding a broad specificity alpha-glucoside transporter) and MAL11 alleles. MPH genes (maltose permease homologues) were present in the lager but not in the ale strains. During growth on maltose, the lager strains expressed AGT1 at low levels and MALx1 genes at high levels, whereas the ale strains expressed AGT1 at high levels and MALx1 genes at low levels. MPHx expression was negligible in all strains. The AGT1 sequences from the ale strains encoded full-length (616 amino acid) polypeptides, but those from both sequenced lager strains encoded truncated (394 amino acid) polypeptides that are unlikely to be functional transporters. Thus, despite the apparently similar genotypes of these ale and lager strains revealed by hybridization, maltose is predominantly carried by AGT1-encoded transporters in the ale strains and by MALx1-encoded transporters in the lager strains.

Identification in the mold Hypocrea jecorina of the first fungal D-galacturonic acid reductase.

A d-galacturonic acid reductase and the corresponding gene were identified from the mold Hypocrea jecorina (Trichoderma reesei). We hypothesize that the enzyme is part of a fungal d-galacturonic acid catabolic pathway which has not been described previously and which is distinctly different from the bacterial pathway. H. jecorina grown on d-galacturonic acid exhibits an NADPH-dependent d-galacturonic acid reductase activity. This activity is absent when the mold is grown on other carbon sources. The d-galacturonic acid reductase was purified, and tryptic digests of the purified protein were sequenced. The open reading frame of the corresponding gene was then cloned from a cDNA library. The open reading frame was functionally expressed in the yeast Saccharomyces cerevisiae. A histidine-tagged protein was purified, and the enzyme kinetics were characterized. The enzyme converts in a reversible reaction from d-galacturonic acid and NADPH to l-galactonic acid and NADP. The enzyme also exhibits activity with d-glucuronic acid and dl-glyceraldehyde.

Engineering redox cofactor regeneration for improved pentose fermentation in Saccharomyces cerevisiae.

Pentose fermentation to ethanol with recombinant Saccharomyces cerevisiae is slow and has a low yield. A likely reason for this is that the catabolism of the pentoses D-xylose and L-arabinose through the corresponding fungal pathways creates an imbalance of redox cofactors. The process, although redox neutral, requires NADPH and NAD+, which have to be regenerated in separate processes. NADPH is normally generated through the oxidative part of the pentose phosphate pathway by the action of glucose-6-phosphate dehydrogenase (ZWF1). To facilitate NADPH regeneration, we expressed the recently discovered gene GDP1, which codes for a fungal NADP+-dependent D-glyceraldehyde-3-phosphate dehydrogenase (NADP-GAPDH) (EC 1.2.1.13), in an S. cerevisiae strain with the D-xylose pathway. NADPH regeneration through an NADP-GAPDH is not linked to CO2 production. The resulting strain fermented D-xylose to ethanol with a higher rate and yield than the corresponding strain without GDP1; i.e., the levels of the unwanted side products xylitol and CO2 were lowered. The oxidative part of the pentose phosphate pathway is the main natural path for NADPH regeneration. However, use of this pathway causes wasteful CO2 production and creates a redox imbalance on the path of anaerobic pentose fermentation to ethanol because it does not regenerate NAD+. The deletion of the gene ZWF1 (which codes for glucose-6-phosphate dehydrogenase), in combination with overexpression of GDP1 further stimulated D-xylose fermentation with respect to rate and yield. Through genetic engineering of the redox reactions, the yeast strain was converted from a strain that produced mainly xylitol and CO2 from D-xylose to a strain that produced mainly ethanol under anaerobic conditions.

The missing link in the fungal L-arabinose catabolic pathway, identification of the L-xylulose reductase gene.

The fungal L-arabinose pathway consists of five enzymes, aldose reductase, L-arabinitol 4-dehydrogenase, L-xylulose reductase, xylitol dehydrogenase, and xylulokinase. All the genes encoding the enzymes of this pathway are known except for that of L-xylulose reductase (EC 1.1.1.10). We identified a gene encoding this enzyme from the filamentous fungus Trichoderma reesei (Hypocrea jecorina). The gene was named lxr1. It was overexpressed in the yeast Saccharomyces cerevisiae, and the enzyme activity was confirmed in a yeast cell extract. Overexpression of all enzymes of the L-arabinose pathway in S. cerevisiae led to growth of S. cerevisiae on L-arabinose; i.e., we could show that the pathway is active in a heterologous host. The lxr1 gene encoded a protein with 266 amino acids and a calculated molecular mass of 28 428 Da. The LXRI protein is an NADPH-specific reductase. It has activity with L-xylulose, D-xylulose, D-fructose, and L-sorbose. The highest affinity is toward L-xylulose (K(m) = 16 mM). In the reverse direction, we found activity with xylitol, D-arabinitol, D-mannitol, and D-sorbitol. It requires a bivalent cation for activity. It belongs to the protein family of short chain dehydrogenases. The enzyme is catalytically similar and homologous in sequence to a D-mannitol:NADP 2-dehydrogenase (EC 1.1.1.138).

A novel NADH-linked l-xylulose reductase in the l-arabinose catabolic pathway of yeast.

An NADH-dependent l-xylulose reductase and the corresponding gene were identified from the yeast Ambrosiozyma monospora. The enzyme is part of the yeast pathway for l-arabinose catabolism. A fungal pathway for l-arabinose utilization has been described previously for molds. In this pathway l-arabinose is sequentially converted to l-arabinitol, l-xylulose, xylitol, and d-xylulose and enters the pentose phosphate pathway as d-xylulose 5-phosphate. In molds the reductions are NADPH-linked, and the oxidations are NAD(+)-linked. Here we show that in A. monospora the pathway is similar, i.e. it has the same two reduction and two oxidation reactions, but the reduction by l-xylulose reductase is not performed by a strictly NADPH-dependent enzyme as in molds but by a strictly NADH-dependent enzyme. The ALX1 gene encoding the NADH-dependent l-xylulose reductase is strongly expressed during growth on l-arabinose as shown by Northern analysis. The gene was functionally overexpressed in Saccharomyces cerevisiae and the purified His-tagged protein characterized. The reversible enzyme converts l-xylulose to xylitol. It also converts d-ribulose to d-arabinitol but has no activity with l-arabinitol or adonitol, i.e. it is specific for sugar alcohols where, in a Fischer projection, the hydroxyl group of the C-2 is in the l-configuration and the hydroxyl group of C-3 is in the d-configuration. It also has no activity with C-6 sugars or sugar alcohols. The K(m) values for l-xylulose and d-ribulose are 9.6 and 4.7 mm, respectively. To our knowledge this is the first report of an NADH-linked l-xylulose reductase.

Production of ethanol from L-arabinose by Saccharomyces cerevisiae containing a fungal L-arabinose pathway.

The fungal pathway for L-arabinose catabolism converts L-arabinose to D-xylulose 5-phosphate in five steps. The intermediates are, in this order: L-arabinitol, L-xylulose, xylitol and D-xylulose. Only some of the genes for the corresponding enzymes were known. We have recently identified the two missing genes for L-arabinitol 4-dehydrogenase and L-xylulose reductase and shown that overexpression of all the genes of the pathway in Saccharomyces cerevisiae enables growth on L-arabinose. Under anaerobic conditions ethanol is produced from L-arabinose, but at a very low rate. The reasons for the low rate of L-arabinose fermentation are discussed.

Identification of the first fungal NADP-GAPDH from Kluyveromyces lactis.

Deletion of the phosphoglucose isomerase gene, PGI1, in Saccharomyces cerevisiae leads to a phenotype for which glucose is toxic. This is related to overproduction of NADPH through the oxidative part of the pentose phosphate pathway and the incompetence of S. cerevisiae to deal with this overproduction. A similar deletion (rag2) in Kluyveromyces lactis does not lead to such a phenotype. We transformed a genomic library of K. lactis in a yeast vector to a S. cerevisiae strain with a pgi1 deletion and screened for growth on glucose. We found a gene (GDP1) which encodes a phosphorylating glyceraldehyde-3-phosphate dehydrogenase, NADP-GAPDH (EC 1.2.1.13), that accepts both NADP and NAD. This is the first report of a eukaryotic, nonplant, NADP-linked GAPDH. Presumably, operation of this enzyme in the reverse direction enabled the transformed S. cerevisiae pgi1 deletion mutant to reoxidize the excess NADPH produced when glucose catabolism was forced through the pentose pathway. On the other hand, transcription of the gene in K. lactis was upregulated during growth on D-xylose, which suggests that in K. lactis the enzyme is involved in regeneration of NADPH needed for xylose assimilation, but transcription was not detected in a rag2 mutant grown on glucose. The presence of an asparagine (Asn46 in NADP-GAPDH) instead of the conserved aspartate found in related but NAD-specific enzymes may explain the ability of NADP-GAPDH to work with NADP as well as NAD.