Two types of alcohol dehydrogenase in separate protein families are the "medium-chain" zinc enzymes (including the classical liver and yeast forms) and the "shortchain" enzymes (including the insect form). Although the medium-chain family has been characterized in prokaryotes and many eukaryotes (fungi, plants, cephalopods, and verte-brates), insects have seemed to possess only the short-chain enzyme. We have now also characterized a medium-chain alcohol dehydrogenase in Drosophila. The developmental stages of the fly, compatible with the constitutive nature of the vertebrate enzyme. Taken together, the results bridge a previously apparent gap in the distribution of medium-chain alcohol dehydrogenases and establish a strictly conserved class m enzyme, consistent with an important role for this enzyme in cellular metabolism.The "classical" alcohol dehydrogenase is part of a widespread system of zinc-containing enzymes (1). In mammalian tissues, at least six classes of this enzyme occur. They differ considerably and represent stages between separate enzymes and ordinary isozymes. Class I is the well-known liver enzyme with ethanol dehydrogenase activity (2), class III is identical with glutathione-dependent formaldehyde dehydrogenase (3), class IV is a form preferentially expressed in stomach (4, 5), while classes II, V, and VI, although little studied, are known also to exhibit distinct properties (6, 7, 44). The class origins have been traced to gene duplications early in vertebrate evolution [the I/III duplication (8)] or during that evolution [the IV/I duplication (5)], with emerging activities toward ethanol (9); class III corresponds to an ancestral form. These properties and the different evolutionary patterns, with class III being "constant" and class I "variable" (10), result in a consistent picture of the enzyme system and place the classes of medium-chain alcohol dehydrogenases as separate enzymes in the cellular metabolism.Similarly, another protein family, short-chain dehydrogenases, has also evolved into a family comprising many different enzyme activities, including an alcohol dehydrogenase (11). This form operates by means of a completely different catalytic mechanism and is related to mammalian prostaglandin dehydrogenases/carbonyl reductase (12). Thus far, this alcohol dehydrogenase has been found in insects, the Drosophila enzyme being recognized early to differ from the zinc-containing alcohol dehydrogenases (13,14). Its properties in various Drosophila species are well established (15).These two alcohol dehydrogenase types demonstrate that ethanol dehydrogenase activity has evolved in different manners, with many organisms now employing a medium-chain enzyme, while others depend on a short-chain enzyme. The medium-chain family has not been identified in insects, although it is of ancient origin and has been characterized in other eukaryotes and in prokaryotes. We now show that the family is indeed present also in insects and that its major representative is the typical class III t...

Background: Human appetite is stimulated by alcohol but the underlying mechanism is unknown. It is possible that hunger-stimulating hormones are mediators of this effect of alcohol. Ghrelin stimulates hunger, but how alcohol affects human ghrelin secretion has never been studied before. Objective: To investigate whether alcohol ingestion exerts an acute influence on serum ghrelin concentrations in healthy subjects. Subjects and design: Eight healthy non-obese subjects participated in the study. All were investigated on two occasions (experiments A and B). Alcohol (0.55 g ethanol/kg body weight) was ingested in experiment A, and drinking-water in experiment B. Venous blood was collected before, and 30 and 60 min after consumption of the drinks. Serum concentrations of ghrelin, cortisol and ethanol were determined and neuropeptide Y (NPY) concentrations were determined in plasma. Results: Alcohol lowered the ghrelin level by 13.9^5.0% at 30 min and by 17.5^2.6% at 60 min, in contrast to drinking-water which was without significant effect. Serum levels of cortisol and insulin were similar after alcohol and water as was plasma NPY. Conclusion: Alcohol has an acute inhibitory influence on human ghrelin secretion but no measurable effect on the secretion of NPY and cortisol. Hence, none of these hormones mediate the orexigenic effect of the drug.

No abstract

Analysis of the activity and structure of lower vertebrate alcohol dehydrogenases reveals that relationships between the classical liver and yeast enzymes need not be continuous. Both the ethanol activity of class I-type alcohol dehydrogenase (alcohol:NAD+ oxidoreductase, EC 1.1.1.1) and the glutathione-dependent formaldehyde activity of the class rI-type enzyme [formaldehyde:NAD+ oxidoreductase (glutathione-formylating), EC 1.2.1.1] are present in liver down to at least the stage of bony fishes (cod liver: ethanol activity, 3.4 units/mg of protein in one enzyme; formaldehyde activity, 4.5 units/mg in the major form of another enzyme).Structural analysis of the latter protein reveals it to be a typical class HI enzyme, with limited variation from the mammalian form and therefore with stable activity and structure throughout much of the vertebrate lineage. In contrast, the classical alcohol dehydrogenase (the class I enzyme) appears to be the emerging form, first in activity and later also in structure. The class I activity is present already in the piscine line, whereas the overall structural-type enzyme is not observed until amphibians and still more recent vertebrates. Consequently, the class I/I duplicatory origin appears to have arisen from a functional class HI form, not a class I form. Therefore, ethanol dehydrogenases from organisms existing before this duplication have origins separate from those leading to the "classical" liver alcohol dehydrogenases. The latter now often occur in isozyme forms from further gene duplications and have a high rate of evolutionary change. The pattern is, however, not simple and we presently find in cod the first evidence for isozymes also within a class Im alcohol dehydrogenase. Overall, the results indicate that both of these classes of vertebrate alcohol dehydrogenase are important and suggest a protective metabolic function for the whole enzyme system. ancestral lineages (15), apparently utilized to different extents in different organisms. The Drosophila line is now known to be part of another large protein family, short-chain dehydrogenases, encompassing also human prostaglandin and steroid dehydrogenases (16)(17)(18).Second, separate classes of human alcohol dehydrogenase were discovered (19). These classes have been structurally characterized and shown to typify mammalian alcohol dehydrogenases, with separate evolutionary properties (20). The explanation is a series ofgene duplications at different stages (21), generating a system of isozymes and enzymes, which have been studied for structure-function relationships (22,23).Third, analysis of amphibian alcohol dehydrogenase recently established a first estimate of the timing of one of the duplications that explain the classes. This timing placed the class I/III separation early in vertebrate development (24), and analysis of the ethanol-active cod enzyme showed this to be a protein with mixed-class properties (25). These results seemed puzzling, since both glutathione-dependent formaldehyde-active class III [...

The structural framework of cod liver alcohol dehydrogenase is similar to that of horse and human alcohol dehydrogenases. In contrast, the substrate pocket differs significantly, and main differences are located in three loops.Nevertheless, the substrate pocket is hydrophobic like that of the mammalian class 1 enzymes and has a similar topography in spite of many main-chain and side-chain differences. The structural framework of alcohol dehydrogenase is also present in a number of related enzymes like glucose dehydrogenase and quinone oxidoreductase. These enzymes have completely different substrate specificity, but also for these enzymes, the corresponding loops of the substrate pocket have significantly different structures. The domains of the two subunits in the crystals of the cod enzyme further differ by a rotation of the catalytic domains by about 6". In one subunit, they close around the coenzyme similarly as in coenzyme complexes of the horse enzyme, but form a more open cleft in the other subunit, similar to the situation in coenzyme-free structures of the horse enzyme. The proton relay system differs from the mammalian class I alcohol dehydrogenases. His 51, which has been implicated in mammalian enzymes to be important for proton transfer from the buried active site to the surface is not present in the cod enzyme. A tyrosine in the corresponding position is turned into the substrate pocket and a water molecule occupies the same position in space as the His side chain, forming a shorter proton relay system.