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.     

Polyol metabolism by Rhizobium trifolii.

In Rhizobium trifolii 7000, the polyols myo-inositol, xylitol, ribitol, D-arabitol, D-mannitol, D-sorbital, and dulcitol are metabolized by inducible nicotinamide adenine dinucleotide-dependent polyol dehydrogenases. Five different polyol dehydrogenases were recognized: inositol dehydrogenase, specific for inositil; ribitol dehydrogenase, specific for ribitol; D-arabitol dehydrogenase, which oxidized D-arabitol, D-mannitol, and D-sorbitol; xylitol dehydrogenase, which oxidized xylitol and D-sorbitol; and dulcitol dehydrogenase, which oxidized dulcitol, ribitol, xylitol, and sorbitol. Apart from inositil and xylitol, all of the polyols induced more than one polyol dehydrogenase and polyol transport system, but the heterologous polyol dehydrogenases and polyol transport systems were not coordinately induced by a particular polyol. With the exception of xylitol, all of the polyols tested served as growth substrates. A mutant of trifolii 7000, which was constitutive for dulcitol dehydrogenase, could also grow on xylitol.     

Extensive variations and basic features in the alcohol dehydrogenase-sorbitol dehydrogenase family

Structural comparisons of sorbitol dehydrogenase with zinc-containing 'long' alcohol dehydrogenases reveal distant but clear relationships. An alignment suggests 93 positional identities with horse liver alcohol dehydrogenase (25% of 374 positions) and 73 identities with yeast alcohol dehydrogenase (20%). Sorbitol dehydrogenase forms a link between these distantly related alcohol dehydrogenases and is in some regions more similar to one of them that they are to each other. 43 residues (11%) are common to all three enzymes and include a heavy over-representation of glycine (half of all glycine residues in sorbitol dehydrogenase), showing the importance of space restrictions in protein structures. Four regions are well conserved, two in each domain of horse liver alcohol dehydrogenase. They are two segments close to the active-site zinc atom of the catalytic domain, and two in the central beta-pleated sheet strands of the coenzyme-binding domain. These similarities demonstrate the general importance of internal and central building units in proteins. Large variations affect a region adjacent to the third protein ligand to the active-site zinc atom in horse liver alcohol dehydrogenase. Such changes at active sites of related enzymes are unusual. Other large differences concern the segment around the non-catalytic zinc atom of horse liver alcohol dehydrogenase; three of its four cysteine ligands are absent from sorbitol dehydrogenase. Three segments with several exchanges correspond to a continuous region with superficial areas, inter-domain contacts and inter-subunit interactions in the catalytic domain of alcohol dehydrogenase. They may correlate with the altered quaternary structure of sorbitol dehydrogenase. Regions corresponding to top and bottom beta-strands in the coenzyme-binding domain of the alcohol dehydrogenase are also little conserved. Within sorbitol dehydrogenase, a large segment shows an internal similarity. The two distantly related alcohol dehydrogenases and sorbitol dehydrogenase form a triplet of enzymes illustrating basic protein relationships. They are ancestrally close enough to establish similarities, yet sufficiently divergent to illustrate changes in all but fundamental properties.     

The primary structure of alcohol dehydrogenase from Drosophila lebanonensis. Extensive variation within insect 'short-chain' alcohol dehydrogenase lacking zinc

Insect alcohol dehydrogenase is highly different from the well-known yeast and mammalian alcohol dehydrogenases. The enzyme from Drosophila lebanonensis has now been characterized by protein analysis and was found to have a 254-residue protein chain with an acetyl-blocked N-terminal Met. Comparisons with the structures of the enzyme from other species allows judgement of the extent of variability within the insect alcohol dehydrogenases. They have diverged to a considerable extent; two forms analyzed at the protein level differ at 18% of all residues, and all the known Drosophila alcohol dehydrogenase structures reveal differences at 72 positions. Some deviations, against a background similarity, in the extent of changes are noted among the parts corresponding to different exons. The structural variation within Drosophila is about as large as the one for the mammalian zinc-containing alcohol dehydrogenase. Consequently, the results illustrate Drosophila relationships and establish great variations also for group of alcohol dehydrogenases lacking zinc.     

Two malate dehydrogenases in Methanobacterium thermoautotrophicum

Methanobacterium thermoautotrophicum (strain Marburg) was found to contain two malate dehydrogenases, which were partially purified and characterized. One was specific for NAD⁺ and catalyzed the dehydrogenation of malate at approximately one-third of the rate of oxalacetate reduction, and the other could equally well use NAD⁺ and NADP⁺ as coenzyme and catalyzed essentially only the reduction of oxalacetate. Via the N-terminal amino acid sequences, the encoding genes were identified in the genome of M. thermoautotrophicum (strain ΔH). Comparison of the deduced amino acid sequences revealed that the two malate dehydrogenases are phylogenetically only distantly related. The NAD⁺-specific malate dehydrogenase showed high sequence similarity to l-malate dehydrogenase from Methanothermus fervidus, and the NAD(P)⁺-using malate dehyrogenase showed high sequence similarity to l-lactate dehydrogenase from Thermotoga maritima and l-malate dehydrogenase from Bacillus subtilis. A function of the two malate dehydrogenases in NADPH:NAD⁺ transhydrogenation is discussed. Received: 29 December 1997 / Accepted: 4 March 1998     

The possible occurrence of zinc in ribitol and sorbitol dehydrogenases from Mycobacterium sp. 279

The activity of polyhydric alcohol dehydrogenases in Mycobacterium sp. 279 was studied under limitation of zinc in the growth medium. It was found that the activity of ribitol and sorbitol dehydrogenases were markedly lowered and that of D-arabinitol dehydrogenase remained unchanged in the Zn2+-deficient cells. Other ions tested i.e., Co2+, Cu2+, Ni2+ and Mn2+ failed to substitute Zn2+ ions in their effect on the enzyme activities. The Zn2+-responsive enzymes were sensitive to the chelating agents (1,10-phenanthroline, 2,2'-dipyridyl), whereas D-arabinitol dehydrogenase was insensitive. The results indicate possible existence of a zinc component in the ribitol and sorbitol dehydrogenases from Mycobacterium sp. 279.     

Calorimetric investigation of NAD binding to some dehydrogenases.

The thermodynamic parameters for the binding of NAD to some dehydrogenases have been determined calorimetrically at 25° and pH 7.6. Except for liver alcohol dehydrogenase (LADH) the ΔG^o, ΔH^o and ΔS^o values for NAD binding to the dehydrogenases are very similar pointing out a possible structure - thermodynamics correlation. The large deviation observed in the case of LADH would be consistent with the occurrence of a conformational change in this enzyme upon binding NAD.     

Crystal structure of formaldehyde dehydrogenase from Pseudomonas putida: the structural origin of the tightly bound cofactor in nicotinoprotein dehydrogenases

Formaldehyde dehydrogenase from Pseudomonas putida (PFDH) is a member of the zinc-containing medium-chain alcohol dehydrogenase family. The pyridine nucleotide NAD(H) in PFDH, which is distinct from the coenzyme (as cosubstrate) in typical alcohol dehydrogenases (ADHs), is tightly but not covalently bound to the protein and acts as a cofactor. PFDH can catalyze aldehyde dismutations without an external addition of NAD(H). The structural basis of the tightly bound cofactor of PFDH is unknown. The crystal structure of PFDH has been solved by the multiwavelength anomalous diffraction method using intrinsic zinc ions and has been refined at a 1.65 A resolution. The 170-kDa homotetrameric PFDH molecule shows 222 point group symmetry. Although the secondary structure arrangement and the binding mode of catalytic and structural zinc ions in PFDH are similar to those of typical ADHs, a number of loop structures that differ between PFDH and ADHs in their lengths and conformations are observed. A comparison of the present structure of PFDH with that of horse liver ADH, a typical example of an ADH, reveals that a long insertion loop of PFDH shields the adenine part of the bound NAD(+) molecule from the solvent, and a tight hydrogen bond network exists between the insertion loop and the adenine part of the cofactor, which is unique to PFDH. This insertion loop is conserved completely among the aldehyde-dismutating formaldehyde dehydrogenases, whereas it is replaced by a short turn among typical ADHs. Thus, the insertion loop specifically found among the aldehyde-dismutating formaldehyde dehydrogenases is responsible for the tight cofactor binding of these enzymes and explains why PFDH can effectively catalyze alternate oxidation and reduction of aldehydes without the release of cofactor molecule from the enzyme.     

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.     

Plagiorchis elegans: Histochemical localization of dehydrogenases in the cercarial stage

Cercariae of Plagiorchis elegans Rudolphi 1802 collected from experimentally infected snails, Lymnaea palustris, were subjected to various histochemical tests for dehydrogenase systems. A high degree of activity was demonstrated for succinic dehydrogenase (EC 1.3.99.1), malic dehydrogenase (EC 1.1.1.37), isocitric dehydrogenase (EC 1.1.1.41), α-glycerophosphate dehydrogenase (EC 1.1.1.8), and glucose 6-phosphate dehydrogenase (EC 1.1.1.49). These enzymes were present in the tegument, tail, caudal pocket, excretory bladder, acetabulum, and oral sucker, particularly in the muscles around the stylet. Only moderate activity was obtained for lactic dehydrogenase (EC 1.1.1.27) and 6-phosphogluconate dehydrogenase (EC 1.1.1.44) at these sites, glutamic dehydrogenase (EC 1.4.1.2) was localized only in the tails of the cercariae and tests for alcohol dehydrogenase (EC 1.1.1.1) were completely negative. The cerebral ganglia and its commissures stained intensely in the tests for succinic, isocitric, α-glycerophosphate, and glucose 6-phosphate dehydrogenase systems. The results indicate the possibility that several energy-producing sequences may be available to these cercariae.     

UDP-glucose dehydrogenase from bovine liver: primary structure and relationship to other dehydrogenases.

The primary structure of bovine liver UDP-glucose dehydrogenase (UDPGDH), a hexameric, NAD(+)-linked enzyme, has been determined at the protein level. The 52-kDa subunits are composed of 468 amino acid residues, with a free N-terminus and a Ser/Asn microhetergeneity at one position. The sequence shares 29.6% positional identity with GDP-mannose dehydrogenase from Pseudomonas, confirming a similarity earlier noted between active site peptides. This degree of similarity is comparable to the 31.1% identity vs. the UDPGDH from type A Streptococcus. Database searching also revealed similarities to a hypothetical sequence from Salmonella typhimurium and to "UDP-N-acetyl-mannosaminuronic acid dehydrogenase" from Escherichia coli. Pairwise identities between bovine UDPGDH and each of these sequences were all in the range of approximately 26-34%. Multiple alignment of all 5 sequences indicates common ancestry for these 4-electron-transferring enzymes. There are 27 strictly conserved residues, including a cysteine residue at position 275, earlier identified by chemical modification as the expected catalytic residue of the second half-reaction (conversion of UDP-aldehydoglucose to UDP-glucuronic acid), and 2 lysine residues, at positions 219 and 338, one of which may be the expected catalytic residue for the first half-reaction (conversion of UDP-glucose to UDP-aldehydoglucose). A GXGXXG pattern characteristic of the coenzyme-binding fold is found at positions 11-16, close to the N-terminus as with "short-chain" alcohol dehydrogenases.(ABSTRACT TRUNCATED AT 250 WORDS)     

Sorbitol dehydrogenase is a zinc enzyme.

Evidence is given that tetrameric sorbitol dehydrogenase from sheep liver contains one zinc atom per subunit, most probably located at the active site, and no other specifically bound zinc or iron atom. In alcohol dehydrogenases that are structurally related to sorbitol dehydrogenase, more than one zinc atom per subunit can complicate investigations of zinc atom function. Therefore, sorbitol dehydrogenase will be particularly valuable for defining the precise roles of zinc in alcohol and polyol dehydrogenases, and for establishing correlations of structure and function with other important zinc-containing proteins.     

Characterization of membrane-bound dehydrogenases of Gluconobacter oxydans 621H using a new system for their functional expression

Acetic acid bacteria are used in biotechnology due to their ability to incompletely oxidize a great variety of carbohydrates, alcohols, and related compounds in a regio- and stereo-selective manner. These reactions are catalyzed by membrane-bound dehydrogenases (mDHs), often with a broad substrate spectrum. In this study, the promoters of six mDHs of Gluconobacter oxydans 621H were characterized. The constitutive promoter of the alcohol dehydrogenase and the glucose-repressed promoter of the inositol dehydrogenase were used to construct a shuttle vector system for the fully functional expression of mDHs in the multi-deletion strain G. oxydans BP.9 that lacks its mDHs. This system was used to express each mDH of G. oxydans 621H, in order to individually characterize the substrates, they oxidize. From 55 tested compounds, the alcohol dehydrogenase oxidized 30 substrates and the polyol dehydrogenase 25. The substrate spectrum of alcohol dehydrogenase overlapped largely with the aldehyde dehydrogenase and partially with polyol dehydrogenase. Thus, we were able to resolve the overlapping substrate spectra of the main mDHs of G. oxydans 621H. The described approach could also be used for the expression and detailed characterization of substrates used by mDHs from other acetic acid bacteria or a metagenome.

     

Expression of membrane-bound dehydrogenases from a mother of vinegar metagenome in Gluconobacter oxydans

Acetic acid bacteria are well-known for their membrane-bound dehydrogenases rapidly oxidizing a variety of substrates in the periplasm. Since many acetic acid bacteria have not been successfully cultured in the laboratory yet, studying membrane-bound dehydrogenases directly from a metagenome of vinegar microbiota seems to be a promising way to identify novel variants of these enzymes. To this end, DNA from a mother of vinegar was isolated, sequenced, and screened for membrane-bound dehydrogenases using an in silico approach. Six metagenomic dehydrogenases were successfully expressed using an expression vector with native promoters in the acetic acid bacterium strain Gluconobacter oxydans BP.9, which is devoid of its major native membrane-bound dehydrogenases. Determining the substrates converted by these enzymes, using a whole-cell DCPIP assay, revealed one glucose dehydrogenase with an enlarged substrate spectrum additionally oxidizing aldoheptoses, D-ribose and aldotetroses, one polyol dehydrogenase with an extreme diminished spectrum but distinguishing cis and trans-1,2-cyclohexandiol and a completely new secondary alcohol dehydrogenase, which oxidizes secondary alcohols with a hydroxyl group at position 2, as long as no primary hydroxyl group is present. Three further dehydrogenases were found with substrate spectra similar to known dehydrogenases of G. oxydans 621H.

     

Determination of some biochemical and structural features of alcohol dehydrogenases from Drosophila simulans and Drosophila virilis. Comparison of their properties with the Drosophila melanogaster Adhs enzyme.

The biochemical properties of the enzyme alcohol dehydrogenase of two different Drosophila species, Drosophila simulans and Drosophila virilis, were studied and compared with those of Drosophila melanogaster Adhs enzyme. All of them consist of two identical subunits of molecular weight 27800 and share significant similarities in function. The substrate specificities of these enzymes were characterized and Km(app.) and Vmax.(app.) values were calculated. All these alcohol dehydrogenases show greater affinity for secondary rather than for primary alcohols. The amino acid compositions of the three enzymes were determined, and there is a close similarity between the D. simulans and the D. melanogaster enzymes, but there are significant differences from the alcohol dehydrogenase of D. virilis. The N-terminal amino acid is blocked and the C-terminal amino acid is the same for all three alcohol dehydrogenases. The enzymes from the three species were carboxymethylated and digested with trypsin. The peptide 'maps' reveal, as expected, more homologies between the enzymes of D. simulans and D. melanogaster than with the enzyme of D. virilis.     

The primary structure of Escherichia coli L-threonine dehydrogenase

The complete primary structures of Escherichia coli L-threonine dehydrogenase has been deduced by sequencing the cloned tdh gene. The primary structure so determined agrees with results obtained independently for the amino acid composition, the N-terminal amino acid sequence (20 residues), and a short sequence at the end of an internal peptide of the purified enzyme. The presence of a predicted Asp-Pro bond at residues 148 and 149 was confirmed by treatment of purified threonine dehydrogenase with dilute acid and subsequent analysis of the resulting cleavage products. The primary structure of L-threonine dehydrogenase from E. coli has been examined for possible homology to other NAD+-dependent dehydrogenases; indications are that this enzyme is a member of the zinc-containing long-chain alcohol/polyol dehydrogenase family.     

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.     

Geraniol and geranial dehydrogenases induced in anaerobic monoterpene degradation by Castellaniella defragrans

Castellaniella defragrans is a Betaproteobacterium capable of coupling the oxidation of monoterpenes with denitrification. Geraniol dehydrogenase (GeDH) activity was induced during growth with limonene in comparison to growth with acetate. The N-terminal sequence of the purified enzyme directed the cloning of the corresponding open reading frame (ORF), the first bacterial gene for a GeDH (geoA, for geraniol oxidation pathway). The C. defragrans geraniol dehydrogenase is a homodimeric enzyme that affiliates with the zinc-containing benzyl alcohol dehydrogenases in the superfamily of medium-chain-length dehydrogenases/reductases (MDR). The purified enzyme most efficiently catalyzes the oxidation of perillyl alcohol (k(cat)/K(m) = 2.02 × 10(6) M(-1) s(-1)), followed by geraniol (k(cat)/K(m) = 1.57 × 10(6) M(-1) s(-1)). Apparent K(m) values of <10 μM are consistent with an in vivo toxicity of geraniol above 5 μM. In the genetic vicinity of geoA is a putative aldehyde dehydrogenase that was named geoB and identified as a highly abundant protein during growth with phellandrene. Extracts of Escherichia coli expressing geoB demonstrated in vitro a geranial dehydrogenase (GaDH) activity. GaDH activity was independent of coenzyme A. The irreversible formation of geranic acid allows for a metabolic flux from β-myrcene via linalool, geraniol, and geranial to geranic acid.     

Carbohydrate metabolism in Thermoproteus tenax: in vivo utilization of the non-phosphorylative Entner-Doudoroff pathway and characterization of its first enzyme, glucose dehydrogenase

Thermoproteus tenax is a hyperthermophilic, facultative heterotrophic archaeum. In this organism the utilization of the two catabolic pathways, a variant of the Embden-Meyerhof-Parnas (EMP) pathway and the modified (nonphosphorylative) Entner-Doudoroff (ED) pathway, was investigated and the first enzyme of the ED pathway, glucose dehydrogenase, was characterized. The distribution of the ¹³C label in alanine synthesized by cells grown with [1-¹³C]glucose indicated that in vivo the EMP pathway and the modified ED pathway operate parallel, with glucose metabolization via the EMP pathway being prominent. To initiate studies on the regulatory mechanisms governing carbon flux via these pathways, the first enzyme of the ED pathway, glucose dehydrogenase, was purified to homogeneity and its phenotypic properties were characterized. The pyridine-nucleotide-dependent enzyme used both NAD⁺ and NADP⁺ as cosubstrates, showing a 100-fold higher affinity for NADP⁺. Besides glucose, xylose was used as substrate, but with significantly lower affinity. These data suggest that the physiological function of the enzyme is the oxidation of glucose by NADP⁺. A striking feature was the influence of NADP⁺ and NAD⁺ on the quaternary structure and activity state of the enzyme. Without cosubstrate, the enzyme was highly aggregated (mol. mass > 600 kDa) but inactive, whereas in the presence of the cosubstrate the aggregates dissociated into enzymatically active, homomeric dimers with a mol. mass of 84 kDa (mol. mass of subunits: 41 kDa). The N-terminal amino acid sequence showed striking similarity to the respective partial sequences of alcohol dehydrogenases and sorbitol dehydrogenases, but no resemblance to the known pyridine-nucleotide-dependent archaeal and bacterial glucose dehydrogenases. Received: 25 October 1996 / Accepted: 15 April 1997     

Characterization of the genes of the 2,3-butanediol operons from Klebsiella terrigena and Enterobacter aerogenes.

The genes involved in the 2,3-butanediol pathway coding for alpha-acetolactate decarboxylase, alpha-acetolactate synthase (alpha-ALS), and acetoin (diacetyl) reductase were isolated from Klebsiella terrigena and shown to be located in one operon. This operon was also shown to exist in Enterobacter aerogenes. The budA gene, coding for alpha-acetolactate decarboxylase, gives in both organisms a protein of 259 amino acids. The amino acid similarity between these proteins is 87%. The K. terrigena genes budB and budC, coding for alpha-ALS and acetoin reductase, respectively, were sequenced. The 559-amino-acid-long alpha-ALS enzyme shows similarities to the large subunits of the Escherichia coli anabolic alpha-ALS enzymes encoded by the genes ilvB, ilvG, and ilvI. The K. terrigena alpha-ALS is also shown to complement an anabolic alpha-ALS-deficient E. coli strain for valine synthesis. The 243-amino-acid-long acetoin reductase has the consensus amino acid sequence for the insect-type alcohol dehydrogenase/ribitol dehydrogenase family and has extensive similarities with the N-terminal and internal regions of three known dehydrogenases and one oxidoreductase.