Characteristics of alcohol/polyol dehydrogenases. The zinc-containing long-chain alcohol dehydrogenases

Mutation of Arg-115 of human class III alcohol dehydrogenase: a binding site required for formaldehyde dehydrogenase activity and fatty acid activation

The origin of the fatty acid activation and formaldehyde dehydrogenase activity that distinguishes human class III alcohol dehydrogenase (alcohol:NAD+ oxidoreductase, EC 1.1.1.1) from all other alcohol dehydrogenases has been examined by site-directed mutagenesis of its Arg-115 residue. The Ala- and Asp-115 mutant proteins were expressed in Escherichia coli and purified by affinity chromatography and ion-exchange HPLC. The activities of the recombinant native and mutant enzymes toward ethanol are essentially identical, but mutagenesis greatly decreases the kcat/Km values for glutathione-dependent formaldehyde oxidation. The catalytic efficiency for the Asp variant is < 0.1% that of the unmutated enzyme, due to both a higher Km and a lower kcat value. As with the native enzyme, neither mutant can oxidize methanol, be saturated by ethanol, or be inhibited by 4-methylpyrazole; i.e., they retain these class III characteristics. In contrast, however, their activation by fatty acids, another characteristic unique to class III alcohol dehydrogenase, is markedly attenuated. The Ala mutant is activated only slightly, but the Asp mutant is not activated at all. The results strongly indicate that Arg-115 in class III alcohol dehydrogenase is a component of the binding site for activating fatty acids and is critical for the binding of S-hydroxymethylglutathione in glutathione-dependent formaldehyde dehydrogenase activity.

Molecular cloning and expression of a Eucalyptus gunnii cDNA clone encoding cinnamyl alcohol dehydrogenase

Cinnamyl alcohol dehydrogenase (CAD) catalyses the reduction of hydroxycinnamyl aldehydes (sinapyl, paracoumaryl, coniferyl aldehydes) to the corresponding alcohols which are the direct monomeric precursors of lignins. Recently, we have purified from Eucalyptus gunnii two isoforms of CAD (CAD1 and CAD2), distinct in their biochemical and functional properties. In this paper, we report the cloning of a CAD cDNA (pEuCAD2) isolated by screening a lambda gt11 library generated from cell suspension culture of Eucalyptus gunnii, using a tobacco CAD cDNA as a probe. This full-length clone (1392 bp) encodes a protein of 356 amino acids which corresponds to the subunit molecular weight of the CAD2 isoform. Sequence analysis revealed that CAD2 is very well conserved among species (78% homology with CAD from tobacco, a herbaceous angiosperm, and 81% with the partial sequence from a gymnosperm, loblolly pine). The identity of this clone was unambiguously demonstrated (1) by comparison with peptide sequence data from purified CAD2 and (2) by functional expression of the recombinant enzyme in Escherichia coli. Recombinant CAD showed the same properties as the natural isoform CAD2, in terms of electrophoretic mobility, polypeptide structure, substrate specificity and antigenicity. The CAD2 transcript is equally abundant in stems and leaves and at the limit of detection in roots. At the tissue level the CAD2 gene is highly expressed in xylem and virtually undetectable in phloem.

Glucose-6-phosphate dehydrogenase. Structure-function relationships and the Pichia jadinii enzyme structure

The primary structure of glucose-6-phosphate dehydrogenase from the yeast Pichia jadinii (formerly Candida utilis) has been determined. It consists of a 495-residue, N-terminally acetylated protein chain. The structure shows extensive differences from those of the corresponding mammalian, fruit fly, and bacterial enzymes (52-68% residue non-identities), but also from that of another yeast, Saccharomyces cerevisiae (38%). A eubacterial type and a yeast type of glucose-6-phosphate dehydrogenase are discerned, in addition to the known mammalian type. They are distinguished from each other, from the mammalian type, and the insect enzyme, on the basis of both specific residues and pattern differences. The distribution of residues conserved in all forms locates short segments in which identities are closely grouped. Approximately 50% of these segments correspond to predicted turns and appear to mark the principal folds characteristic of the enzyme's tertiary structure. A region in the N-terminal part of the protein chain has characteristics suggestive of a coenzyme-binding site, while, in the middle third, another functionally important segment may be related to glucose-6-phosphate binding and catalysis.

Affinity labelling of sorbitol dehydrogenase from sheep liver with alpha-bromo-beta-(5-imidazolyl)propionic acid

The metal-directed alkylating agent DL-alpha-bromo-beta-(5- imidazolyl)propionic acid (BrImPpOH) is shown to be an affinity-labelling reagent for sheep liver sorbitol dehydrogenase (SDH). As previously found for horse liver alcohol dehydrogenase (ADH), it modifies a cysteine ligand to the active-site zinc. In this case it is selectively incorporated (over 90%) at Cys43 in each of the four polypeptide chains/protomers of sheep liver SDH. Incorporated reagent and residual activity correlated. The first order inactivation constant, K2, and KEI, the dissociation constant for SDH and BrImPpOH, have been determined at different pH. The reactivity of BrImPpOH for SDH is higher than that for horse liver and yeast ADH. The protection of SDH against BrImPpOH inactivation by buffers and other molecules shows some similarities to that with horse liver ADH. However, sheep liver SDH bound BrImPpOH, imidazole and phosphate ions much weaker than liver ADH. The pKa values from the plot of log (k2/KEI) against pH are approximately 7.0 and 8.8-8.9. The former pKa value probably represents ionization of an imidazole group and the latter the zinc/water ionization in SDH. These pKa values are similar to those found for horse liver ADH. They are apparently not noticeably influenced by a second cysteine ligand in liver ADH being replaced by a proposed glutamic acid residue as a ligand to the catalytic zinc in SDH. The plot of logk2 against pH shows pKa values around 7.0 and 9.2 for the SDH-BrImPpOH-complex. The pKa of 7.0 is the same as for log(k2/KEI), and indicates no significant perturbation due to the binding of BrImPpOH to SDH. The pKa around 9.2 indicates perturbation of the zinc/water ionization or the ionization of Cys43.

Cloning and analysis of a Candida maltosa gene which confers resistance to formaldehyde in Saccharomyces cerevisiae

A gene (FDH1) of Candida maltosa which confers resistance to formaldehyde in Saccharomyces cerevisiae was cloned and its nucleotide sequence determined. The gene has a single intron which possesses the highly conserved splicing signals found in S. cerevisiae introns. We demonstrated that processing of the pre-mRNA of the cloned gene occurred identically in both S. cerevisiae and C. maltosa. The predicted amino acid sequence from the cloned gene showed 65.5% identity to human alcohol dehydrogenase (ADH) class III and 23.9% identity to S. cerevisiae ADH1. The most probable mechanism of resistance to formaldehyde is thought to be the glutathione-dependent oxidation of formaldehyde which is characteristic for ADH class III. The cloned FDH1 gene was successfully employed as a dominant selectable marker in the transformation of S. cerevisiae.

Cloning and sequence analysis of the gene encoding an NADP-dependent alcohol dehydrogenase in Mycobacterium bovis BCG

The nucleotide sequence of a 1619-bp fragment of Mycobacterium bovis BCG containing the gene that encodes an alcohol dehydrogenase (ADH) has been determined. The M(r) calculated from the deduced amino acid (aa) sequence, as well as the N terminus, are in good accordance with those determined for the ADH purified from M. bovis BCG extracts. The M. bovis BCG cloned adh gene was expressed in Escherichia coli by its own promoter and the synthesized product shows ADH activity in the butane-1-ol-NADP system. Based on comparison of the aa sequence, this enzyme belongs to the zinc-containing, long-chain alcohol/polyol dehydrogenase family, which has been primarily described in eukaryotes. Of the 22 strictly conserved residues in this group, 19 are also conserved in M. bovis BCG ADH (BCGADH).

Molecular characterization of two Clostridium acetobutylicum ATCC 824 butanol dehydrogenase isozyme genes

A 4-kb segment of DNA containing two previously cloned butanol dehydrogenase (BDH) isozyme genes (D. Petersen, R. Welch, F. Rudolph, and G. Bennett, J. Bacteriol. 173:1831-1834, 1991) was sequenced. Two complete open reading frames (ORFs) were identified (bdhA and bdhB), along with a third truncated ORF (ORF1). The translation products of bdhA and bdhB corresponded to the N-terminal sequences of the purified BDH I and BDH II proteins, respectively. The two isozymes had a high amino acid identity (73%) and showed homology to a newly described class of alcohol dehydrogenases. Northern blots revealed that bdhA and bdhB did not form an operon. Primer extension experiments located single transcriptional start sites 37 and 58 bp upstream of the start codons of bdhA and bdhB, respectively. The -10 and -35 promoter regions for these genes were almost identical. bdhA and bdhB were found to be induced or derepressed immediately prior to significant butanol production in controlled pH 5.0 batch fermentations.

Cloning and expression of an NADP(+)-dependent alcohol dehydrogenase gene of Entamoeba histolytica

Ethanol is the major metabolic product of glucose fermentation by the protozoan parasite Entamoeba histolytica under the anaerobic conditions found in the lumen of the colon. Here an internal peptide sequence determined from a major 39-kDa amoeba protein isolated by isoelectric focusing followed by SDS/PAGE was used to clone the gene for the E. histolytica NADP(+)-dependent alcohol dehydrogenase (EhADH1; EC 1.1.1.2). The EhADH1 clone had an open reading frame that was 360 amino acids long and encoded a protein of approximately 39 kDa (calculated size). EhADH1 showed a 62% amino acid identity with the tetrameric NADP(+)-dependent alcohol dehydrogenase of Thermoanaerobium brockii. In contrast, EhADH1 showed a 15% amino acid identity with the closest human alcohol dehydrogenase. EhADH1 contained 18 of the 22 amino acids conserved in other alcohol dehydrogenases, including glycines involved in binding NAD(P)+ as well as histidine and cysteine residues involved in binding the catalytic zinc ion. Like the T. brockii alcohol dehydrogenase, EhADH1 lacked a 23-amino acid stretch present in other alcohol dehydrogenases that includes four cysteines that bind a second noncatalytic zinc ion. An EhADH1-glutathione-S-transferase fusion protein showed the expected NADP(+)-dependent alcohol dehydrogenase and NADPH-dependent acetaldehyde reductase activities. The enzymatic activities of the EhADH1 fusion protein were inhibited by pyrazole and 4-methylpyrazole.

Identification and characterisation of cDNA clones encoding cinnamyl alcohol dehydrogenase from tobacco

Cinnamyl alcohol dehydrogenase (CAD, EC 1.1.1.195) is an enzyme involved in lignin biosynthesis. We have previously isolated pure CAD enzyme as two closely related polypeptides of 44 and 42.5 kDa from tobacco stems. In this paper, we report partial amino acid sequences of these two polypeptides. Based on the peptide sequences mixed oligonucleotides were used to screen a tobacco stem cDNA library and CAD cDNA clones encoding the two polypeptides were identified. DNA sequence comparisons indicate very high sequence identity between these clones both in the coding and in the 5' and 3' untranslated sequences. The close similarity between the two CAD genes leads us to suggest that they do not represent different isoforms but are the same gene from each of the two parental lines of Nicotiana tabacum cv. Samsun. Sequence comparisons with alcohol dehydrogenase 1 (ADH1) from yeast shows sequence similarities of ca. 30%, while comparisons with maize, barley and potato ADH1 sequences show similarities of not more than 23%.

Cloning, expression, and sequence analysis of the Bacillus methanolicus C1 methanol dehydrogenase gene

The gene (mdh) coding for methanol dehydrogenase (MDH) of thermotolerant, methylotroph Bacillus methanolicus C1 has been cloned and sequenced. The deduced amino acid sequence of the mdh gene exhibited similarity to those of five other alcohol dehydrogenase (type III) enzymes, which are distinct from the long-chain zinc-containing (type I) or short-chain zinc-lacking (type II) enzymes. Highly efficient expression of the mdh gene in Escherichia coli was probably driven from its own promoter sequence. After purification of MDH from E. coli, the kinetic and biochemical properties of the enzyme were investigated. The physiological effect of MDH synthesis in E. coli and the role of conserved sequence patterns in type III alcohol dehydrogenases have been analyzed and are discussed.

Progressive sequence alignment and molecular evolution of the Zn-containing alcohol dehydrogenase family

Sequences of 47 members of the Zn-containing alcohol dehydrogenase (ADH) family were aligned progressively, and an evolutionary tree with detailed branch order and branch lengths was produced. The alignment shows that only 9 amino acid residues (of 374 in the horse liver ADH sequence) are conserved in this family; these include eight Gly and one Val with structural roles. Three residues that bind the catalytic Zn and modulate its electrostatic environment are conserved in 45 members. Asp 223, which determines specificity for NAD, is found in all but the two NADP-dependent enzymes, which have Gly or Ala. Ser or Thr 48, which makes a hydrogen bond to the substrate, is present in 46 members. The four Cys ligands for the structural zinc are conserved except in zeta-crystallin, the sorbitol dehydrogenases, and two bacterial enzymes. Analysis of the evolutionary tree gives estimates of the times of divergence for different animal ADHs. The human class II (pi) and class III (chi) ADHs probably diverged about 630 million years ago, and the newly identified human ADH6 appeared about 520 million years ago, implying that these classes of enzymes may exist or have existed in all vertebrates. The human class I ADH isoenzymes (alpha, beta, and gamma) diverged about 80 million years ago, suggesting that these isoenzymes may exist or have existed in all primates. Analysis of branch lengths shows that these plant ADHs are more conserved than the animal ones and that class III ADHs are more conserved than class I ADHs. The rate of acceptance of point mutations (PAM units) shows that selection pressure has existed for ADHs, implying that these enzymes play definite metabolic roles.

NAD-linked, factor-dependent formaldehyde dehydrogenase or trimeric, zinc-containing, long-chain alcohol dehydrogenase from Amycolatopsis methanolica

NAD-linked, factor-dependent formaldehyde dehydrogenase (FD-FA1DH) of the Gram-positive methylotrophic bacterium, Amycolatopsis methanolica, was purified to homogeneity. It is a trimeric enzyme with identical subunits (molecular mass 40 kDa) containing 6 atoms Zn/enzyme molecule. The factor is a heat-stable, low-molecular-mass compound, which showed retention on an Aminex HPX-87H column. Inactivation of the factor occurred during manipulation, but activity could be restored by incubation with dithiothreitol. The identity of the factor is still unknown. It could not be replaced by thiol compounds or cofactors known to be involved in metabolism of C1 compounds. Of the aldehydes tested, only formaldehyde was a substrate. However, the enzyme showed also activity with higher aliphatic alcohols and the presence of the factor was not required for this reaction. Methanol was not a substrate, but high concentrations of it could replace the factor in the conversion of formaldehyde. Presumably, a hemiacetal of formaldehyde is the genuine substrate, which, in the case of methanol, acts as a factor leading to methylformate as the product. This view is supported by the fact that formate could only be detected in the reaction mixture after acidification. Inhibition studies revealed that the enzyme contains a reactive thiol group, being protected by the binding of NAD against attack by heavy-metal ions and aldehydes. Studies on the effect of the order of addition of coenzyme and substrate suggested that optimal catalysis required NAD as the first binding component. Substrate specificity and the induction pattern clearly indicate a role of the enzyme in formaldehyde oxidation. However, since FD-FA1DH was also found in A. methanolica grown on n-butanol, but not on ethanol, it may have a role in the oxidation of higher aliphatic alcohols as well. FD-FA1DH and the factor from A. methanolica are very similar to a combination already described for Rhodococcus erythropolis [Eggeling, L. & Sahm, H. (1985) Eur. J. Biochem. 150, 129-134]. NAD-linked, glutathione-dependent formaldehyde dehydrogenase (GD-FA1DH) resembles FD-FA1DH in many respects. Since glutathione has so far not been detected in Gram-positive bacteria, FD-FA1DH could be the counterpart of this enzyme in Gram-positive bacteria. Alignment of the N-terminal sequence (31 residues) of FD-FA1DH with that of GD-FA1DH from rat liver indeed showed similarity (30% identical positions). However, comparable similarity was found with class I alcohol dehydrogenase from this organism and with cytosolic alcohol dehydrogenase from Saccharomyces cerevisiae, isozyme 1.(ABSTRACT TRUNCATED AT 400 WORDS)

A single-residue exchange gives human recombinant beta beta alcohol dehydrogenase gamma gamma isozyme properties

Amino acid residue 48 in human alcohol dehydrogenase constitutes one of 21 residue differences between the common, adult-type isozyme subunits beta and gamma. It is at the inner part of the substrate pocket and has been ascribed a role in hydrogen-bond formation with both the substrate and coenzyme. In order to allow direct evaluation of its importance, Thr48 of a recombinant non-acetylated beta subunit was mutated to Ser (as in the gamma subunit) or Ala (as in no native form, and not allowing side-chain hydrogen bonds), and the proteins were expressed in Escherichia coli. The two non-acetylated recombinant proteins, the beta 48T form and the mutant beta 48S, gave enzymatically active enzymes with indistinguishable specific activities towards ethanol, whereas the mutant beta 48A showed no enzymatic activity. The most striking differences between dimers with the beta subunit and the beta 48S subunit (both non-acetylated) were observed with cyclohexanol, hydroxysteroids, methanol and ethanol. With cyclohexanol, the Km was lowered from 11 mM to 280 microM, and the kcat/Km ratio, although still less than that for the gamma gamma isozyme, was increased 80-fold. Similarly, beta 48S could use 3 beta-hydroxy-5 beta-androstan-17-one as substrate, like gamma gamma, although again with a catalytic efficiency much less than that for the gamma gamma isozyme. Furthermore, testosterone inhibited beta 48S to 50% at a concentration of 100 microM, whereas the beta beta form was not inhibited. All these results show that residue 48 is responsible for a large part of the differences between the two isozymes beta beta and gamma gamma of human class-I alcohol dehydrogenase. The form with the inactive beta 48A subunit was possible to purify by AMP-Sepharose chromatography, suggesting the presence of a functional NAD-binding site. The enzymatic measurements, demonstrating a transition from one isozyme activity to that characteristic of another, confirmed that a side-chain hydroxyl in residue 48 is required for activity, and interpretation by computer modelling showed marked differences at the active site.

Evolution of the alcohol dehydrogenase (ADH) genes in yeast: characterization of a fourth ADH in Kluyveromyces lactis

Three alcohol dehydrogenase (ADH) genes have recently been characterized in the yeast Kluyveromyces lactis. We report on a fourth ADH in K. lactis (KADH II: KADH2* gene) which is highly similar to other ADHs in K. lactis and Saccharomyces cerevisiae. KADH II appears to be a cytoplasmic enzyme, and after expression of KADH2 in S. cerevisiae enzyme activity comigrated with a K. lactis ADH present in cells grown in glucose or in ethanol. KADH I was also expressed in S. cerevisiae and it comigrated with a major ADH species expressed under glucose growth conditions in K. lactis. The substrate specificities for KADH I and KADH II were shown to be more similar to that of SADH II than to SADH I. SADH I cannot efficiently utilize long chain alcohols, in contrast to other cytoplasmic yeast ADHs, presumably because of the presence of a methionine (residue 271) in its substrate binding cleft. A comparison of the DNA sequences of ADHs among K. lactis, S. cerevisiae and Schizosaccharomyces pombe suggests that the ancestral yeast species contained one cytoplasmic ADH. After divergence from S. pombe, the ADH in the ancestor to K. lactis and S. cerevisiae was duplicated, and one ADH became localized to the mitochondrion, presumably for the oxidative use of ethanol. Following the speciation of S. cerevisiae and K. lactis, the gene encoding the cytoplasmic ADH in S. cerevisiae duplicated, which resulted in the development of the SADH II protein as the primary oxidative enzyme in place of SADH III. In contrast, the K. lactis mitochondrial ADH duplicated to give rise to the highly expressed KADH3 and KADH4 genes, both of which may still play primary roles in oxidative metabolism. These data suggest that K. lactis and S. cerevisiae use different compartments for their metabolism of ethanol. Our results also indicate that the complex regulatory circuits controlling the glucose-repressible SADH2 in S. cerevisiae are a recent acquisition from regulatory networks used for the control of genes other than SADH2.

Environmental regulation of alcohol metabolism in thermotolerant methylotrophic Bacillus strains

The thermotolerant methylotroph Bacillus sp. C1 possesses a novel NAD-dependent methanol dehydrogenase (MDH), with distinct structural and mechanistic properties. During growth on methanol and ethanol, MDH was responsible for the oxidation of both these substrates. MDH activity in cells grown on methanol or glucose was inversely related to the growth rate. Highest activity levels were observed in cells grown on the C1-substrates methanol and formaldehyde. The affinity of MDH for alcohol substrates and NAD, as well as Vmax, are strongly increased in the presence of a Mr 50,000 activator protein plus Mg(2+)-ions [Arfman et al. (1991) J Biol Chem 266: 3955-3960]. Under all growth conditions tested the cells contained an approximately 18-fold molar excess of (decameric) MDH over (dimeric) activator protein. Expression of hexulose-6-phosphate synthase (HPS), the key enzyme of the RuMP cycle, was probably induced by the substrate formaldehyde. Cells with high MDH and low HPS activity levels immediately accumulated (toxic) formaldehyde when exposed to a transient increase in methanol concentration. Similarly, cells with high MDH and low CoA-linked NAD-dependent acetaldehyde dehydrogenase activity levels produced acetaldehyde when subjected to a rise in ethanol concentration. Problems frequently observed in establishing cultures of methylotrophic bacilli on methanol- or ethanol-containing media are (in part) assigned to these phenomena.

The 40-kilodalton allergen of Candida albicans is an alcohol dehydrogenase: molecular cloning and immunological analysis using monoclonal antibodies

To characterize the 40-kilodalton (kD) major allergen of Candida albicans (C. albicans), six monoclonal antibodies (MoAbs) against this allergen were generated. In SDS-polyacrylamide gel electrophoresis and immunoblot analysis, these MoAbs showed four different reaction patterns to antigens of six different Candida species. With the exception of one MoAb, other MoAbs were resistant to periodate treatment indicating non-carbohydrate epitopes were probably being recognized by these MoAbs. These MoAbs were used in the molecular cloning and immunological analysis of the gene coding for the 40-kD allergen. Nucleotide sequence determination of the two lambda gt11 cDNA clones obtained showed that the 40-kD allergen is an alcohol dehydrogenase (ADH) which shares a 70% amino acid sequence homology with the ADH isozyme I of Saccharomyces cerevisiae. This finding was confirmed by positive immunological response of the lysates of the clones obtained and a preparation of ADH of Saccharomyces cerevisiae to various MoAbs and to IgE antibodies in sera of allergic patients.

Characteristics of short-chain alcohol dehydrogenases and related enzymes

Different short-chain dehydrogenases are distantly related, constituting a protein family now known from at least 20 separate enzymes characterized, but with extensive differences, especially in the C-terminal third of their sequences. Many of the first known members were prokaryotic, but recent additions include mammalian enzymes from placenta, liver and other tissues, including 15-hydroxyprostaglandin, 17 beta-hydroxysteroid and 11 beta-hydroxysteroid dehydrogenases. In addition, species variants, isozyme-like multiplicities and mutants have been reported for several of the structures. Alignments of the different enzymes reveal large homologous parts, with clustered similarities indicating regions of special functional/structural importance. Several of these derive from relationships within a common type of coenzyme-binding domain, but central-chain patterns of similarity go beyond this domain. Total residue identities between enzyme pairs are typically around 25%, but single forms deviate more or less (14-58%). Only six of the 250-odd residues are strictly conserved and seven more are conserved in all but single cases. Over one third of the conserved residues are glycine, showing the importance of conformational and spatial restrictions. Secondary structure predictions, residue distributions and hydrophilicity profiles outline a common, N-terminal coenzyme-binding domain similar to that of other dehydrogenases, and a C-terminal domain with unique segments and presumably individual functions in each case. Strictly conserved residues of possible functional interest are limited, essentially only three polar residues. Asp64, Tyr152 and Lys156 (in the numbering of Drosophila alcohol dehydrogenase), but no histidine or cysteine residue like in the completely different, classical medium-chain alcohol dehydrogenase family. Asp64 is in the suggested coenzyme-binding domain, whereas Tyr152 and Lys156 are close to the center of the protein chain, at a putative inter-domain, active-site segment. Consequently, the overall comparisons suggest the possibility of related mechanisms and domain properties for different members of the short-chain family.

Isolation of Caenorhabditis elegans mutants lacking alcohol dehydrogenase activity

Alcohol dehydrogenase (ADH) and the genes encoding this enzyme have been studied intensively in a broad range of organisms. Little, however, has been reported on ADH in the free-living nematode Caenorhabditis elegans. Extracts of wild-type C. elegans contain ADH activity and display a single band of activity on a native polyacrylamide gel. Reaction rate for alcohol oxidation is more rapid with higher molecular weight alcohols as substrate than with ethanol. Primary alcohols are preferred to secondary alcohols. C. elegans is sensitive to allyl alcohol, a compound that has been used to select for ADH-null mutants of several organisms. Allyl alcohol-resistant mutant strains were selected from ethylmethanesulfonate (EMS)-mutagenized nematode populations. ADH activity was measured in extracts from eight of these strains and was found to be low or nondetectable. These results form a basis for molecular and genetic characterization of ADH expression in C. elegans.

Sorbitol dehydrogenase: cDNA coding for the rat enzyme. Variations within the alcohol dehydrogenase family independent of quaternary structure and metal content

Two separate cDNA-clones, together coding for rat sorbitol dehydrogenase, have been isolated from a liver cDNA library in lambda gt11 by screening with oligonucleotide probes. One clone contained a 1020-bp fragment starting at the codon for amino acid residue 104 and ending with a 261-bp 3' non-coding region, the second encompassed the entire 5' region and ended with a 3' truncation corresponding to amino acid residue 315. The coding region consists of 356 amino acid residues, one more than in the human and sheep enzymes. The presence of the extra residue at position 3, a proline, can be explained by a shifted splice point in the mRNA. The primary structure of rat sorbitol dehydrogenase allows triplet comparisons of three distinct rat-ungulate-human enzymes differing in quaternary structure and metal content within the zinc-containing alcohol dehydrogenase family. The variability of sorbitol dehydrogenase (tetramer with one zinc atom/subunit; no activity towards ethanol) is large (18%), exactly like that for the class I alcohol dehydrogenase (dimer with two zinc atoms/subunit; no activity towards sorbitol), differing threefold from that of the class III alcohol dehydrogenase/glutathione-dependent formaldehyde dehydrogenase (dimer with two zinc atoms/subunit; 6% variability) suggesting that the distinct extents of variability within this protein family are independent of substrate specificity, metal content and quaternary structure.

Different segment similarities in long-chain dehydrogenases

Long-chain dehydrogenases were scrutinized for common patterns. Overall molecular similarities are not discerned, in contrast to the situation for several short-chain and medium-chain dehydrogenases, but coenzyme-binding segments are discernible. Species variants of glucose-6-phosphate dehydrogenase reveal about 20% strictly conserved residues, grouped into three segments and supporting assignments of sites for coenzyme-binding and catalysis. Glycine is overrepresented among the residues conserved, typical of distantly related proteins. Two of the enzymes within the pentose phosphate pathway reveal a distant similarity of interest for further evaluation, between a C-terminal 178-residue segment of glucose-6-phosphate dehydrogenase and the N-terminal part of 6-phosphogluconate dehydrogenase.

ADH and phylogenetic relationships of Drosophila lebanonesis (Scaptodrosophila)

Increasing data on Drosophila alcohol dehydrogenase (ADH) sequences have made it possible to calculate the rate of amino acid replacement per year, which is 1.7 x 10(-9). This value makes this protein suitable for reconstructing phylogenetic relationships within the genus for those species for which no molecular data are available such as Scaptodrosophila. The amino acid sequence of Drosophila lebanonensis is compared to all of the already known Drosophila ADHs, stressing the unique characteristic features of this protein such as the conservation of an initiating methionine at the N-terminus, the unique replacement of a glycine by an alanine at a very conserved position in the NAD domain of all dehydrogenases, the lack of a slow-migrating peptide, and the total conservation of the maximally hydrophilic peptide. The functional significance of these features is discussed. Although the percent amino acid identity of the ADH molecule in Drosophila decreases as the number of sequences compared increases, the conservation of residue type in terms of size and hydrophobocity for the ADH molecule is shown to be very high throughout the genus Drosophila. The distance matrix and parsimony methods used to establish the phylogenetic relationships of D. lebanonensis show that the three subgenera, Scaptodrosophila, Drosophila, and Sophophora separated at approximately the same time.

Pseudomonas 3 beta-hydroxysteroid dehydrogenase. Primary structure and relationships to other steroid dehydrogenases

The 3 beta-hydroxysteroid dehydrogenase of Pseudomonas testosteroni commercially available was purified by an FPLC step and submitted to sequence determination by peptide analysis. The structure obtained reveals a 253-residue polypeptide chain, with an N-terminal, free alpha-amino group, and a low cysteine content. Comparisons with other hydroxysteroid dehydrogenases recently characterized reveal distant similarities with prokaryotic and, to some extent, also eukaryotic forms of separate specificities. Residue identities with a Streptomyces 20 beta-hydroxysteroid dehydrogenase are 35% and distributed over the entire molecule, whereas residue identities with the mammalian 17 beta-hydroxysteroid dehydrogenase only constitute 20%, and are essentially limited to the N-terminal and central parts, Nevertheless, all these enzymes exhibit a conserved tyrosine residue (position 151 in the present enzyme) noted as possibly having a functional role in some members of this protein family. Combined, the results establish the prokaryotic 3 beta-hydroxysteroid dehydrogenase as belonging to the family of short-chain alcohol dehydrogenases, reveal that the hydroxysteroid dehydrogenases are no more closely related than dehydrogenases with other enzyme activities within the family (e.g. glucose, ribitol, hydroxyprostaglandin dehydrogenases), show several of the mammalian hydroxysteroid dehydrogenases to have subunits of longer size with different patterns of similarity than those of the prokaryotic family members characterized, and define important segments of the coenzyme-binding region for this enzyme group.

Amphibian alcohol dehydrogenase, the major frog liver enzyme. Relationships to other forms and assessment of an early gene duplication separating vertebrate class I and class III alcohol dehydrogenases

Submammalian alcohol dehydrogenase structures can be used to evaluate the origins and functions of the different types of the mammalian enzyme. Two avian forms were recently reported, and we now define the major amphibian alcohol dehydrogenase. The enzyme from the liver of the Green frog Rana perezi was purified, carboxymethylated, and submitted to amino acid sequence determination by peptide analysis of six different digests. The protein has a 375-residue subunit and is a class I alcohol dehydrogenase, bridging the gap toward the original separation of the classes that are observable in the human alcohol dehydrogenase system. In relation to the human class I enzyme, the amphibian protein has residue identities exactly halfway (68%) between those for the corresponding avian enzyme (74%) and the human class III enzyme (62%), suggesting an origin of the alcohol dehydrogenase classes very early in or close to the evolution of the vertebrate line. This conclusion suggests that these enzyme classes are more universal among animals than previously realized and constitutes the first real assessment of the origin of the duplications leading to the alcohol dehydrogenase classes. Functionally, the amphibian enzyme exhibits properties typical for class I but has an unusually low Km for ethanol (0.09 mM) and Ki for pyrazole (0.15 microM) at pH 10.0. This correlates with a strictly hydrophobic substrate pocket and one amino acid difference toward the human class I enzyme at the inner part of the pocket. Coenzyme binding is highly similar, while subunit-interacting residues, as in other alcohol dehydrogenases, exhibit several differences.(ABSTRACT TRUNCATED AT 250 WORDS)

Eukaryotic glucose-6-phosphate dehydrogenases: structural screening of related proteins

Rapid assessment of structural relationships between yeast glucose-6-phosphate dehydrogenases and other eukaryotic types of this enzyme is described. Separation and size estimation of large fragments by sodium dodecylsulfate/polyacrylamide gel electrophoresis, electroblotting onto disks, and sequencer analysis provide data that permit alignment of the segments thus characterized with the related proteins, and utilize existing structural knowledge to assess new enzyme structures. Affinity labeling allows further correlations. The results establish the overall structural arrangements of the new proteins, including the location of the active-site lysine residue, even though the yeast enzyme structures are found to differ markedly from the few previously characterized glucose-6-phosphate dehydrogenases.

Comparison of benzyl alcohol dehydrogenases and benzaldehyde dehydrogenases from the benzyl alcohol and mandelate pathways in Acinetobacter calcoaceticus and from the TOL-plasmid-encoded toluene pathway in Pseudomonas putida. N-terminal amino acid sequences, amino acid compositions and immunological cross-reactions

1. N-Terminal sequences were determined for benzyl alcohol dehydrogenase, benzaldehyde dehydrogenase I and benzaldehyde dehydrogenase II from Acinetobacter calcoaceticus N.C.I.B. 8250, benzyl alcohol dehydrogenase and benzaldehyde dehydrogenase encoded by the TOL plasmid pWW53 in Pseudomonas putida MT53 and yeast K(+)-activated aldehyde dehydrogenase. Comprehensive details of the sequence determinations have been deposited as Supplementary Publication SUP 50161 (5 pages) at the British Library Document Supply Centre, Boston Spa. Wetherby. West Yorkshire LS23 7BQ, U.K., from whom copies can be obtained on the terms indicated in Biochem. J. (1991) 273. 5. The extent of sequence similarity suggests that the benzyl alcohol dehydrogenases are related to each other and also to established members of the family of long-chain Zn2(+)-dependent alcohol dehydrogenases. Benzaldehyde dehydrogenase II from Acinetobacter appears to be related to the Pseudomonas TOL-plasmid-encoded benzaldehyde dehydrogenase. The yeast K(+)-activated aldehyde dehydrogenase has similarity of sequence with the mammalian liver cytoplasmic class of aldehyde dehydrogenases but not with any of the Acinetobacter or Pseudomonas enzymes. 2. Antisera were raised in rabbits against the three Acinetobacter enzymes and both of the Pseudomonas enzymes, and the extents of the cross-reactions were determined by immunoprecipitation assays with native antigens and by immunoblotting with SDS-denatured antigens. Cross-reactions were detected between the alcohol dehydrogenases and also among the aldehyde dehydrogenases. This confirms the interpretation of the N-terminal sequence comparisons and also indicates that benzaldehyde dehydrogenase I from Acinetobacter may be related to the other two benzaldehyde dehydrogenases. 3. The amino acid compositions of the Acinetobacter and the Pseudomonas enzymes were determined and the numbers of amino acid residues per subunit were calculated to be: benzyl alcohol dehydrogenase and TOL-plasmid-encoded benzyl alcohol dehydrogenase, 381; benzaldehyde dehydrogenase I and benzaldehyde dehydrogenase II, 525; TOL-plasmid-encoded benzaldehyde dehydrogenase, 538.

Class IV mammalian alcohol dehydrogenase. Structural data of the rat stomach enzyme reveal a new class well separated from those already characterized

The stomach form of alcohol dehydrogenase has been structurally evaluated by peptide analysis covering six separate regions of the rat enzyme. Overall, this new structure differs widely (32-40% residue differences) from the structures of three classes of alcohol dehydrogenase characterized before from the same species. Consequently, this novel enzyme constitutes a true fourth class of mammalian alcohol dehydrogenase. In particular, differences are extensive also towards class II, although enzymatic and physicochemical properties initially suggested overall similarities with class II. The new structure establishes the presence of one further alcohol dehydrogenase mammalian gene, extends the enzyme family derived from repeated gene duplications, and confirms tissue-specific expressions.

Avian alcohol dehydrogenase: the chicken liver enzyme. Primary structure, cDNA-cloning, and relationships to other alcohol dehydrogenases

The major ethanol-active form of chicken liver alcohol dehydrogenase was characterized. The primary structure was determined by peptide analysis and, to a large part, was also deduced by cDNA analysis of a near full-length cDNA clone. The latter was detected by screening of a chicken liver cDNA library with antibodies raised against the purified dehydrogenase. The structure shows that the avian enzyme exhibits characteristics of the complex mammalian alcohol dehydrogenase system, tracing its origin and divergence, and allowing functional correlations. The chicken protein analyzed proves to be a class I alcohol dehydrogenase, with 74% residue identity to gamma chains of the human enzyme, a Km for ethanol of 0.5 mM and a Ki for 4-methyl pyrazole of 2.5 microM. Relationships to the other two classes are non-identical; residue exchanges towards the human classes increase in the order I less than III less than II, and human/chicken differences are less than inter-class differences. Consequently, the origins of the classes are more distant than the avian/mammalian separation. They reflect duplicatory events separated in time, and the lines that lead to present-day classes I and II deviate early. Integrated with the data for the quail enzyme, the structure of the chicken protein shows that within the avian enzymes the degree of variation is comparable to that within the mammalian class I enzymes, which are more variable than the class III forms. The coenzyme-binding and substrate-binding residues of this chicken alcohol dehydrogenase are largely identical to those in the mammalian class I counterparts. However, the subunit-interacting areas are more variable and suggest some relationships of the avian enzyme with both class I and III mammalian forms. One of the residues, Gly260 (mammalian class I numbering system), previously considered characteristic of all alcohol dehydrogenases, is replaced by Gln.

Avian alcohol dehydrogenase. Characterization of the quail enzyme, functional interpretations, and relationships to the different classes of mammalian alcohol dehydrogenase

The primary structure of the major quail liver alcohol dehydrogenase was determined. It is a long-chain, zinc-containing alcohol dehydrogenase of the type occurring also in mammals and hence allows judgement of the gene duplications giving rise to the classes of the human alcohol dehydrogenase system. The avian form is most closely related to the class I mammalian enzyme (72-75% residue identity), least related to class II (60% identity), and intermediately related to class III (64-65% identity). This pattern distinguishes the mammalian enzyme classes and separates classes I and II in particular. In addition to the generally larger similarities with class I, the avian enzyme exhibits certain residue patterns otherwise typical of the other classes, including an extra Trp residue, present in both class II and III but not in class I, with a corresponding increase in the UV absorbance. The avian enzyme further shows that a Gly residue at position 260 previously considered strictly conserved in alcohol dehydrogenases can be exchanged with Lys. However, zinc-binding residues, coenzyme-binding residues, and to a large extent substrate-binding residues are unchanged in the avian enzyme, suggesting its functional properties to be related to those of the class I mammalian alcohol dehydrogenases. In contrast, the areas of subunit interactions in the dimers differ substantially. These results show that (a) the vertebrate enzyme classes are of distant origin, (b) the submammalian enzyme exhibits partly mixed properties in relation to the classes, and (c) the three mammalian enzyme classes are not as equidistantly related as initially apparent but suggest origins from two sublevels.

Purification and characterisation of TOL plasmid-encoded benzyl alcohol dehydrogenase and benzaldehyde dehydrogenase of Pseudomonas putida

Benzyl alcohol dehydrogenase and benzaldehyde dehydrogenase, two enzymes of the xylene degradative pathway encoded by the plasmid TOL of a Gram-negative bacterium Pseudomonas putida, were purified and characterized. Benzyl alcohol dehydrogenase catalyses the oxidation of benzyl alcohol to benzaldehyde with the concomitant reduction of NAD+; the reaction is reversible. Benzaldehyde dehydrogenase catalyses the oxidation of benzaldehyde to benzoic acid with the concomitant reduction of NAD+; the reaction is irreversible. Benzyl alcohol dehydrogenase and benzaldehyde dehydrogenase also catalyse the oxidation of many substituted benzyl alcohols and benzaldehydes, respectively, though they were not capable of oxidizing aliphatic alcohols and aldehydes. The apparent Km value of benzyl alcohol dehydrogenase for benzyl alcohol was 220 microM, while that of benzaldehyde dehydrogenase for benzaldehyde was 460 microM. Neither enzyme contained a prosthetic group such as FAD or FMN, and both enzymes were inactivated by SH-blocking agents such as N-ethylmaleimide. Both enzymes were dimers of identical subunits; the monomer of benzyl alcohol dehydrogenase has a mass of 42 kDa whereas that of the monomer of benzaldehyde dehydrogenase was 57 kDa. Both enzymes transfer hydride to the pro-R side of the prochiral C4 of the pyridine ring of NAD+.

sti35, a stress-responsive gene in Fusarium spp

A stress-induced mRNA was identified in the phytopathogenic fungus Fusarium oxysporum f. sp. cucumerinum. Treatment of the fungus with ethanol resulted in the induction of a major mRNA species encoding a protein of approximate Mr 37,000. A full-length cDNA clone of the induced message was obtained. RNA blot analysis indicated that the mRNA was induced by various other stresses, including treatment with copper(II) chloride and heat (37 degrees C). However, it was not greatly induced by treatment with phaseollinisoflavan, an antifungal isoflavonoid produced by Phaseolus vulgaris (French bean). In contrast, phaseollinisoflavan induced the homologous mRNA in the related bean pathogen Fusarium solani f. sp. phaseoli. A genomic clone of the F. solani f. sp. phaseoli gene was obtained, and both this and the cDNA clone from F. oxysporum f. sp. cucumerinum were sequenced. The latter indicated an open reading frame of 320 codons encoding a 34,556-dalton polypeptide. The corresponding reading frame in F. solani f. sp. phaseoli was 324 codons, 89% identical to the F. oxysporum f. sp. cucumerium sequence, and was interrupted by a short intron. The gene was designated sti35 (stress-inducible mRNA). Although computer homology searches were negative, the cloned gene was observed to cross-hybridize to DNAs of other filamentous fungi, Saccharomyces cerevisiae, and soybean. Thus, sti35 appears to be a common gene among a variety of eucaryotes.

The transcripts of zeta-crystallin, a lens protein related to the alcohol dehydrogenase family, are altered in a guinea-pig hereditary cataract

Zeta-Crystallin, a major component of the guinea-pig lens proteins, is distantly related to the enzymes of the zinc-containing alcohol dehydrogenase family (ADH). Analysis of the structural similarities between zeta-crystallin and ADH reveals that while characteristics important in maintaining the tertiary structure of the molecule appear conserved, the amino acids binding the catalytic zinc atom are absent in zeta-crystallin. Significantly, zeta-crystallin does not have ADH activity. Previous studies showed that the zeta-crystallin protein is modified in the lens of guinea-pigs affected with an autosomal dominant hereditary cataract. We have further investigated the molecular origin of the lens defect by examining the steady-state levels of zeta-crystallin transcripts in normal and mutant eyes. Our data indicate that no normal zeta-crystallin mRNA is present in the lens of the homozygous animals; instead, a cross-hybridizing lower molecular weight mRNA is detected at significantly reduced concentrations. Heterozygous lenses exhibit both mRNA species.

Cloning of the Zymomonas mobilis structural gene encoding alcohol dehydrogenase I (adhA): sequence comparison and expression in Escherichia coli

Zymomonas mobilis ferments sugars to produce ethanol with two biochemically distinct isoenzymes of alcohol dehydrogenase. The adhA gene encoding alcohol dehydrogenase I has now been sequenced and compared with the adhB gene, which encodes the second isoenzyme. The deduced amino acid sequences for these gene products exhibited no apparent homology. Alcohol dehydrogenase I contained 337 amino acids, with a subunit molecular weight of 36,096. Based on comparisons of primary amino acid sequences, this enzyme belongs to the family of zinc alcohol dehydrogenases which have been described primarily in eucaryotes. Nearly all of the 22 strictly conserved amino acids in this group were also conserved in Z. mobilis alcohol dehydrogenase I. Alcohol dehydrogenase I is an abundant protein, although adhA lacked many of the features previously reported in four other highly expressed genes from Z. mobilis. Codon usage in adhA is not highly biased and includes many codons which were unused by pdc, adhB, gap, and pgk. The ribosomal binding region of adhA lacked the canonical Shine-Dalgarno sequence found in the other highly expressed genes from Z. mobilis. Although these features may facilitate the expression of high enzyme levels, they do not appear to be essential for the expression of Z. mobilis adhA.

Association of hereditary cataracts in strain 13/N guinea-pigs with mutation of the gene for zeta-crystallin

Congenital nuclear cataracts transmitted by an autosomal dominant gene are present in a line of strain 13/N guinea-pigs. Studies on the lens proteins from these animals demonstrate changes in both the composition and structure of the crystallins relative to normal controls. The most prominent difference is in the zeta-crystallin, a taxon-specific crystallin which has been shown to be related to the alcohol dehydrogenases. In animals homozygous for the cataract phenotype the normal zeta-crystallin polypeptide is absent from the lens. Quantitation is difficult in the cataractous lenses from heterozygotes because of protein changes secondary to opacification: however in liver and kidney which have catalytic levels of the protein, the concentrations are approximately half that present in tissue from normal control animals. These findings suggest that in the cataractous animals a mutation has occurred in the gene for zeta-crystallin. In addition, a novel protein which is very similar to zeta-crystallin is synthesized only in the lenses of animals with cataract. This protein appears to be the product of the mutant gene for zeta-crystallin. These data support the hypothesis that this hereditary congenital cataract results from a specific mutation in the zeta-crystallin gene.

Characterization of the murine corticosteroid binding globulin: variations between mammalian forms

Corticosteroid binding globulin (CGB) from term-pregnant mouse serum was isolated and characterized by peptide analysis after treatment with CNBr and Lys-specific protease, respectively. Amino acid sequence analysis of six segments, covering 189 of 383 positions in different regions of the protein, showed unexpectedly low overall homology (60%) to the indirectly deduced human amino acid sequence previously reported. However, some segments displayed a greater resemblance to their human counterparts. Differences were observed in at least two of six potential glycosylation sites. The nature of electrophoretic CBG variants and their immunological properties are described.

Cobalt(II)-substituted class III alcohol and sorbitol dehydrogenases from human liver

The catalytic zinc atoms in class III (chi) alcohol dehydrogenase (ADH) and sorbitol dehydrogenase (SDH) from human liver have been specifically removed and replaced by cobalt(II) with a new ultrafiltration technique. The electronic absorption spectrum of class III cobalt ADH (epsiolon 638 = 870 M-1 cm-1) is nearly identical with those of active site substituted horse EE and human class I (beta 1 beta 1) cobalt ADH. Thus, the coordination environment of the catalytic metal is strictly conserved in these enzymes. However, significant differences are noted when the spectra of class III ADH-coenzyme complexes are compared to the corresponding spectra of the horse enzyme. The spectrum of class III ADH.NADH is split into three bands, centered at 680, 638, and 562 nm. The class III ADH.NAD+ species resembles the alkaline form of the corresponding horse enzyme complex but without exhibiting the pH dependence of the latter. These spectral changes underscore the role of the coenzymes in differentially fine tuning the catalytic metal for its particular function in each ADH. The noncatalytic zinc of class III ADH exchanges with cobalt at pH 7.0. While 9 residues out of 15 in the loop surrounding the noncatalytic zinc of class III ADH differ from those of the class I ADH, the electronic absorption spectra of cobalt in the noncatalytic metal site of class III ADH establish that the coordination environment of this site is conserved as well. The spectrum of cobalt SDH differs significantly from those of cobalt ADHs.(ABSTRACT TRUNCATED AT 250 WORDS)

Cloning and sequence analysis of the fermentative alcohol-dehydrogenase-encoding gene of Escherichia coli

A 6-kb fragment of DNA, which complemented defects in the alcohol dehydrogenase (ADH)-encoding gene (adhE) of Escherichia coli, was cloned into a multicopy vector. Both ADH and coenzyme-A-linked acetaldehyde dehydrogenase (ACDH) activities were encoded by the plasmid, pHIL8. The adhE gene was identified as an open reading frame of 891 codons encoding an Mr 96,008 protein (minus the initiating methionine). Codon usage analysis indicates that adhE should be highly expressed. This gene shows no significant homology to any previously sequenced ADH-encoding gene.

Homology between hydroxybutyryl and hydroxyacyl coenzyme A dehydrogenase enzymes from Clostridium acetobutylicum fermentation and vertebrate fatty acid beta-oxidation pathways

The enzymes NAD-dependent beta-hydroxybutyryl coenzyme A dehydrogenase (BHBD) and 3-hydroxyacetyl coenzyme A (3-hydroxyacyl-CoA) dehydrogenase are part of the central fermentation pathways for butyrate and butanol production in the gram-positive anaerobic bacterium Clostridium acetobutylicum and for the beta oxidation of fatty acids in eucaryotes, respectively. The C. acetobutylicum hbd gene encoding a bacterial BHBD was cloned, expressed, and sequenced in Escherichia coli. The deduced primary amino acid sequence of the C. acetobutylicum BHBD showed 45.9% similarity with the equivalent mitochondrial fatty acid beta-oxidation enzyme and 38.4% similarity with the 3-hydroxyacyl-CoA dehydrogenase part of the bifunctional enoyl-CoA hydratase:3-hydroxyacyl-CoA dehydrogenase from rat peroxisomes. The pig mitochondrial 3-hydroxyacyl-CoA dehydrogenase showed 31.7% similarity with the 3-hydroxyacyl-CoA dehydrogenase part of the bifunctional enzyme from rat peroxisomes. The phylogenetic relationship between these enzymes supports a common evolutionary origin for the fatty acid beta-oxidation pathways of vertebrate mitochondria and peroxisomes and the bacterial fermentation pathway.

Characteristics of mammalian class III alcohol dehydrogenases, an enzyme less variable than the traditional liver enzyme of class I

Class III alcohol dehydrogenase, whose activity toward ethanol is negligible, has defined, specific properties and is not just a "variant" of the class I protein, the traditional liver enzyme. The primary structure of the horse class III protein has now been determined, and this allows the comparison of alcohol dehydrogenases from human, horse, and rat for both classes III and I, providing identical triads for both these enzyme types. Many consistent differences between the classes separate the two forms as distinct enzymes with characteristic properties. The mammalian class III enzymes are much less variable in structure than the corresponding typical liver enzymes of class I: there are 35 versus 84 positional differences in these identical three-species sets. The class III and class I subunits contain four versus two tryptophan residues, respectively. This makes the differences in absorbance at 280 nm a characteristic property. There are also 4-6 fewer positive charges in the class III enzymes accounting for their electrophoretic differences. The substrate binding site of class III differs from that of class I by replacements at positions that form the hydrophobic barrel typical for this site. In class III, two to four of these positions contain residues with polar or even charged side chains (positions 57 and 93 in all species, plus positions 116 in the horse and 140 in the human and the horse), while corresponding intraclass variation is small. All these structural features correlate with functional characteristics and suggest that the enzyme classes serve different roles. In addition, the replacements between these triad sets illustrate further general properties of the two mammalian alcohol dehydrogenase classes.(ABSTRACT TRUNCATED AT 250 WORDS)