A 38 kDa allylic alcohol dehydrogenase from the cultured cells of Nicotiana tabacum

Article abstract of DOI:10.1016/S0031-9422(00)00326-5

An NADP⁺-dependent alcohol dehydrogenase (allyl-ADH) was isolated from the cultured cells of Nicotiana tabacum. The allyl-ADH was found to be efficient for the dehydrogenation of secondary allylic alcohols rather than saturated secondary alcohols and it was specific for the S-stereoisomer of the alcohols. The enzyme catalyzed the reversible reaction whereby the carbonyl group of enones is reduced to the corresponding allylic alcohol or vice versa. Two possible primary structures of the allyl-ADH were deduced by the sequence analyses of full-length cDNAs (ally-ADH1 and ally-ADH2), which were cloned by the PCR method. These analyses indicated that the allyl-ADHs are composed of 343 amino acids-having the molecular weights 38 083 and 37 994, respectively, and they showed approximately 70% homology to the NADP⁺-dependent oxidoreductases belonging to a plant ζ-crystallin family. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0031-9422(00)00326-5

Phytochemistry 55 (2000) 297±303
www.elsevier.com/locate/phytochem
A 38 kDa allylic alcohol dehydrogenase from the
cultured cells of Nicotiana tabacum
Toshifumi Hirata *, Yoshitaka Tamura, Naoyuki Yokobatake, Kei Shimoda,
Yoshiyuki Ashida
Department of Mathematical and Life Sciences, Graduate School of Science, Hiroshima University, 1-3-1 Kagamiyama,
Higashi-Hiroshima 739-8526, Japan
Received 17 April 2000; received in revised form 18 July 2000
Abstract
An NADP+-dependent alcohol dehydrogenase (allyl-ADH) was isolated from the cultured cells of Nicotiana tabacum. The allyl-
ADH was found to be ecient for the dehydrogenation of secondary allylic alcohols rather than saturated secondary alcohols and
it was speci®c for the S-stereoisomer of the alcohols. The enzyme catalyzed the reversible reaction whereby the carbonyl group of
enones is reduced to the corresponding allylic alcohol or vice versa. Two possible primary structures of the allyl-ADH were deduced
by the sequence analyses of full-length cDNAs (ally-ADH1 and ally-ADH2), which were cloned by the PCR method. These analyses
indicated that the allyl-ADHs are composed of 343 amino acids having the molecular weights 38 083 and 37 994, respectively, and
they showed approximately 70% homology to the NADP+-dependent oxidoreductases belonging to a plant z-crystallin family.
# 2000 Elsevier Science Ltd. All rights reserved.
Keywords: Nicotiana tabacum; Solonaceae; Allylic alcohol dehydrogenase; Allylic alcohol; Enone; Carveol; Carvone; z-crystallin family
1. Introduction
The primary structure of the allyl-ADH was deduced by
analyzing the nucleotide sequence of the allyl-ADH
cDNA in the cultured cells of N. tabacum.
Alcohol dehydrogenases (ADHs) acting on primary
alcohols (Dalziel and Dickinson, 1966; Dickinson and
Dalziel, 1967; Lange et al., 1976; Tassin and Vandecas-
teele, 1972) and secondary alcohols (Coleman and
Perry, 1985; Barrett et al., 1980; Bryant et al., 1988) are
well characterized. However, little attention has been
paid to those participating in the oxidation of allylic
alcohols. It has been shown that cultured cells of
Nicotiana tabacum transform monoterpene allylic alco-
hols, such as carveol and verbenol, into the corre-
sponding enones (Hamada, 1988). In connection with
studies on the biotransformation of enones with cul-
tured plant cells as biocatalysts (Hirata et al., 1982), we
isolated from the cultured suspension cells of N. taba-
cum an allylic alcohol dehydrogenase (allyl-ADH) hav-
ing a dehydrogenase activity speci®c for the (S)-alcohol.
2. Results and discussion
2.1. Isolation and characterization of the allyl-ADH
The allylic alcohol dehydrogenase was puri®ed from a
cell free extract of the cultured cells of N. tabacum by
DEAE-Toyopearl anion exchange column chromato-
graphy, followed by Red-Toyopearl anity column
chromatography. The puri®cation steps of the enzyme
are summarized in Table 1. SDS±PAGE of the puri®ed
enzyme fraction yielded a single protein band at 37 kDa.
Gel ®ltration chromatography on a TSK G3000SW
HPLC column revealed the molecular mass of the
enzyme to be approximately 74 kDa. These results sug-
gested that the native enzyme is a dimer composed of
two identical subunits. The enzyme had a pH optima of
8.0 for the dehydrogenation of (2S, 4S)-carveol (1a) and
at 7.4 for the reduction of the carbonyl group of (S)-
* Corresponding author. Tel.: +81-824-247435; fax: +81-824-
247435.
E-mail address: thirata@sci.hiroshima-u.ac.jp (T. Hirata).
0031-9422/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0031-9422(00)00326-5

298
T. Hirata et al. / Phytochemistry 55 (2000) 297±303
2
1
Table 1
Puri®cation of the allyl-ADH from N. tabacum
5.2Â10 M for 3b, and 5.9Â10 M for 5a. The Km
values for the (S)-alcohols were one order of magnitude
lower than those for the (R)-alcohols, indicating that the
allyl-ADH has a higher anity for the S-con®guration
than for the R-con®guration. It is interesting to note
that the anity for the allylic alcohol was much higher
than that for the saturated secondary alcohol according
to the di€erences of Km values by up to two orders of
magnitude. These showed that the allyl-ADH favored
allylic alcohols having the S-con®guration.
Step
Total protein Total activity Speci®c activity Fold
(unitsa)
(mg)
(units/mg
protein)
3
Crude extract 552
DEAE-Toyopearl 61
0.29
0.28
0.01
0.53Â10
1
9
3
4.6Â10
Red-Toyopearl
0.09
0.11
210
a
One unit of enzyme activity is de®ned as the amount of enzyme
that catalyzed the oxidation of 1.0 mmol of (2S, 4S)-carveol (1a) per
min.
The possibility that allyl-ADH catalyzes a reversible
reaction for the dehydrogenation of alcohols was
examined. The yields in the enzymatic reactions of
ketones, 2, 4, 6, 7, and 11±13, as substrates are sum-
marized in Table 3. Carvones, 2 and 4, having a con-
jugated C C double bond were transformed into the
corresponding allylic alcohols, whereas the saturated
ketones, 6, 7, and 11±13 were not, suggesting that the
allyl-ADH is speci®c for the reduction of a ketone
adjacent to the C C double bond, in agreement with
the results described above. It is of interest that the
allyl-ADH from N. tabacum hardly reduced saturated
ketones. In the reduction of carvones, (S)-allyl alcohols
were preferentially obtained rather than (R)-stereo-
isomers, showing that the stereospeci®city of the reduc-
tion by the allyl-ADH corresponds to the Prelog rule
(Prelog, 1964) for reduction the of ketone, resulting in
the predominant formation of a S-alcohol by re-face
carvone (2). The utilization of NADP+ by the enzyme
was four times greater than for NAD+.
The substrate speci®cities in the dehydrogenation of
each enantiomer of the alcohols, 1, 3, 5, and 8±10, with
the allyl-ADH were examined, and are shown in Table
2. Carveols, 1 and 3, were converted into carvones, 2
and 4, respectively, whereas saturated alcohols, 5 and 8±
10, were not oxidized. This indicated that the allyl-ADH
catalyzed speci®cally the dehydrogenation of the sec-
ondary alcohols adjacent to the C C double bond,
rather than the saturated secondary alcohols. In order
to complete the characterization of the substrate speci-
®city of the enzyme, the Michaelis parameter (Km) for
the enzymatic dehydrogenation was measured: 2.5Â
3
2
3
10 M for 1a, 3.8Â10 M for 1b, 1.9Â10 M for 3a,

T. Hirata et al. / Phytochemistry 55 (2000) 297±303
299
Table 2
Dehydrogenation of alcohols with the allyl-ADH from N. tabacum
1 a
Substrate
Product
Kcat(min
)
(2S,4S)-Carveol (1a)
(2R,4R)-Carveol (1b)
(S)-Carvone (2)
(R)-Carvone (4)
4.6
1.2
5.8
1.5
0.2
0
(2S,4R)-Carveol (3a)
(2R,4S)-Carveol (3b)
(R)-Carvone (4)
(S)-Carvone (2)
(1R,2S, 4S)-Neoisodihydrocarveol (5a)
(1S,2R,4R)-Neoisodihydrocarveol (5b)
(1R,3R,4S)-Menthol (8a)
(1R,4S)-Isodihydrocarvone (6)
(1S,4R)-Isodihydrocarvone (7)
±
±
±
±
±
±
0
(1S,3S,4R)-Menthol (8b)
(1R,2S,4R)-Borneol (9a)
0
0
(1S,2R,4S)-Borneol (9b)
(1S,2S,3S,5R)-Isopinocampheol (10a)
(1R,2R,3R,5S)-Isopinocampheol (10b)
0
0
0
a
Kcat are expressed as turnover number in the enzymatic formation of the product.
Table 3
Hydrogenation of ketones with the allyl-ADH from N. tabacum
1 a
Substrate
Product
Kcat(min
)
(S)-Carvone (2)
(2S, 4S)-Carveol (1a)
(2R,4S)-Carveol (3b)
1.5
0.6
0.7
1.9
0.2
0
(R)-Carvone (4)
(2R,4R)-Carveol (1b)
(2S,4R)-Carveol (3a)
(1R,2S, 4S)-Neoisodihydrocarveol (5a)
(1R, 4S)-Isodihydrocarvone (6)
(1S, 4R)-Isodihydrocarvone (7)
(1R, 4S)-Menthone (11)
±
±
±
±
0
(1R, 4R)-Camphor (12)
(1S, 4S)-Camphor (13)
0
0
a
Kcat are expressed as turnover number in the enzymatic formation of the product.
attack of hydrogen. These results show that the enzy-
matic reaction with the allyl-ADH is a reversible oxi-
doreduction and the reaction is enantioselective.
ADH2 were di€erent in two bp at the 50-untranslated
regions and four bp at the ORF. The allyl-ADH1 and
allyl-ADH2 encoded polypeptides of 343 amino acids with
calculated molecular weights of 38 083 and 37 994,
respectively, and the amino acid sequences of the deduced
allyl-ADHs were di€erent in four amino acid residues.
The sequence variation of allyl-ADH cDNA and deduced
polypeptide from the sequence in Fig. 1 is thus; nucleotide
a20g, g27a, a80g, t120c, t765c, and g864a, polypeptide
T17A, V30A, L345P, and R278H. The estimated mole-
cular weight almost coincided with the values for the allyl-
ADH observed by SDS±PAGE, and the partial amino
acid sequences completely agreed with the sequence
deduced from the cDNA clones (Fig. 1; underlined).
The deduced amino acid sequence from two allyl-ADH
cDNA clones showed about 70% sequence homology to
NADPH-dependent oxidoreductase/z-crystallin family
from Arabidopsis thaliana (GenBank accession no.
Z49768 and Z49268; Babiychuk et al., 1995). The frag-
ment 160±178 of the amino acid sequence had sig-
ni®cant homology to NADP+-dependent leukotriene
B4 12-hydroxydehydrogenase (67±86% homology) and
also to the ꢀ-crystallin family/quinone reductase (58±
72% homology) containing the putative NAD+/
NADP+ binding domain of the other short chain
2.2. Cloning and sequencing of a cDNA encoding the
allyl-ADH
The N-terminal amino acid of the allyl-ADH was
blocked and consequently the partial amino acid
sequences were determined for two peptide fragments
prepared by digestion of the allyl-ADH with lysyl
endopeptidase. The sequences obtained were KGETV
FVSAASGAVGQLVGQFAKMLG and VLESGDPX
FQKGDLVXG.
The complete amino acid sequence of the allyl-ADH
was determined from the cDNA clone, which was
obtained by reverse transcriptase±polymerase chain reac-
tion (RT-PCR), 30- and 50- rapid ampli®cation of cDNA
ends (RACE) method. The ®rst RT-PCR was performed
by using primers determined from the partial amino
acid sequence given above. These PCR experiments
produced two full-length cDNA clones (allyl-ADH1 and
allyl-ADH2) with 1243-base pairs (bp). These nucleotide
sequences contained a 1029-base open reading frame
(ORF), as shown in Fig. 1. The allyl-ADH1 and allyl-

300
T. Hirata et al. / Phytochemistry 55 (2000) 297±303
Fig. 1. Nucleotide sequences and deduced amino acid sequence of the allyl-ADH1 clone. Two peptide sequences of the puri®ed allyl-ADH are
underlined. The bold letters show the putative NADP+/NAD+ binding motif.
alcohol dehydrogenases (Scrutton et al., 1990; Baker et
al., 1992; Persson et al., 1994), as shown in Fig. 2. It was
reported that the nucleotide-binding domain with the
consensus amino acid sequences, AXXGXXG or
GXXGXXG, preferentially bind, NADP+ as a co-fac-
tor (Persson et al., 1994; Thorn et al., 1995). Since the
allyl-ADH had the sequence AXXGXXG, its co-factor
speci®city was predicted to be a NADP+ rather than
NAD+. It has been suggested that three glycines at 152,
155, and 166 of leukotriene B4 12-hydroxydehydro-
genase from human and pig have a crucial role for the
enzyme activity by producing an NADP+ binding

T. Hirata et al. / Phytochemistry 55 (2000) 297±303
301
Fig. 2. Amino acid alignment of NAD+/NADP+-binding domains of allyl-ADH and other homologous proteins. Amino acid sequence alignments
were generated using the CLUSTAL W multiple sequence alignment program (Thompson et al., 1994). Identical amino acid residues with allyl-
ADH are indicated by a black background. Upper-column indicates the NADPH-dependent oxidoreductases: Arabidopsis thaliana P1_ARATH
(GenBank accession no. Q39172), and Arabidopsis thaliana P2_ARATH (Q39173). Center-column indicates the NADP-dependent leukotriene B4
12-hydroxidehydrogenase: Botryotinia fuckeliana BFU68722 (U68722), rabbit LB4D_RABIT (Q28719), pig LB4D_PIG (Q29073), and human
LB4D_HUMAN (Q14914). Bottom-column indicates the quinone oxidoreductases: Bacillus subtilis A69813 (PIR), Cavia porcellus QOR_CAVPO
(P11415), Escherichia coli D64897 (PIR), Saccharomyces cerevisiae QOR_YEAST (P38230), cowpea T11672 (PIR), Pseudomonas aeruginosa
QOR_PSEAE (P43903), Leishmania amazonensis QOR_LEIAM (P42865), Rhodobacter capsulatus 2420322B (PRF), Deinococcus radiodurans
AE001957 (AAF10634), mouse QOR_MOUSE (P47199), and human QOR_HUMAN (Q08257).
pocket (Yokomizo et al., 1996). A similar pattern for
the positioning of three glycines was found in all dehy-
drogenases cited in Fig. 2.
Thus, the primary structure of the allyl-ADH from N.
tabacum was demonstrated to be similar to the
NADPH-dependent oxidoreductase belonging to the
plant z-crystallin family. Further biochemical studies
and molecular analyses will be required to determine
why the pocket of allyl-ADH from N. tabacum is spe-
ci®c for allylic alcohols.
Chromosorb W (AW-DMCS; 80±100 mesh) at 120ꢀC,
or a capillary column (0.25 mmÂ25 m) coated with
0.25 mm CP-cyclodextrin b-236 M-19 (WCOT), using
N2 as carrier gas (column temp: 110ꢀC, split ratio: 50,
make up: 50 ml min 1). GC±MS was carried out with a
capillary column (0.25 mmÂ25 m) coated with 0.25 mm
of OV-101.
3.2. Substrate
(S)-Carvone (2) {[ꢁ]2D5 +57.1 (neat)}, (R)-Carvone (4)
{[ꢁ]2D5 60.1 (neat)}, (1R,3R,4S)-menthol (8a) {[ꢁ]2D5 49.1
(c 1.5, EtOH)}, (1S,3S,4R)-menthol (8b) {[ꢁ]2D5 +48.9 (c
1.5, EtOH)}, (1R,2S,4R)-borneol (9a) {[ꢁ]2D5 +37.5
(neat)}, (1S,2R,4S)-borneol (9b) {[ꢁ]2D5 37.3 (neat)},
(1S,2S,3S,5R)-isopinocampheol (10a) {[ꢁ]2D5 +30.7 (c
2.0, EtOH)}, (1R,2R,3R,5S)-isopinocampheol (10b)
{[ꢁ]2D5 31.7 (c 2.0, EtOH)}, (1R,4S)-menthone (11)
3. Experimental
3.1. General experimental procedures
GLC analyses were carried out with FID and a glass
column (3 mmÂ2 m) packed with 15% DEGS on

302
T. Hirata et al. / Phytochemistry 55 (2000) 297±303
{[ꢁ]2D5 27.3 (c 1.0, EtOH)}, (1R,4R)-camphor (12)
{[ꢁ]2D5 +42.9 (c 1.5, EtOH)} and (1S,4S)-camphor (13)
{[ꢁ]2D5 42.6 (c 1.5, EtOH)} were purchased from
Aldrich. (2S, 4S)- and (2R,4S)-carveols (1a and 3b) and
(2R,4S)- and (2S,4R)-carveols (1b and 3a) were pre-
pared from (S)- and (R)-carvones (2 and 4), respec-
tively, by reduction with NaBH4. (1R,2S, 4S)- and (1 S,
2R,4R)-Neoisodihydrocarveols (5a and 5b) were given
by the NaBH4 reduction of (1R,4S)- and (1S,4R)-iso-
dihydrocarvones (6 and 7), which were prepared from
(S)- and (R)-carvones (2 and 4), respectively (Hirata et
al., 1982).
100 mM 3-(N-morpholino)propane sulfonic acid with
pH adjusted from 6.0 to 8.5.
3.4. Enzyme assay
The standard assay mixture was composed of 2 ml
enzyme preparation (in 25 mM Tris±HCl bu€er at
pH 8.0), 10 mmol NADP+, and 5 mmol (2S, 4S)-carveol
(1a) solubilized by Triton X-100 (0.3% ®nal concentra-
tion). The mixture was incubated for 24 h at 35ꢀC and
then extracted with diethyl ether. The organic layer was
subjected to GLC and GC±MS analyses. The enzyme
activity was determined as the amount of (S)-carvone
(2) produced. One unit of enzyme activity is de®ned as
the amount of enzyme that catalyzed the oxidation of
1.0 mmol of (2S, 4S)-carveol (1a) per min.
3.3. Puri®cation of the enzyme
Cultured cells of N. tabacum (Hirata et al., 1982)
grown for 4 weeks in Murashige and Skoog's medium
were used in the present work. The fresh cells (150 g)
were frozen with liquid N2, ground in a chilled mortar,
and homogenized with 500 ml of 50 mM Na±Pi bu€er
(pH 6.8). The resulting suspension was ®ltered through
four layers of cheesecloth. The crude extract was cen-
trifuged at 10 000 g for 30 min and then the supernatant
was treated with 40±80% sat. of ammonium sulfate to
give a crude enzyme preparation. The preparation was
suspended in 25 mM Tris±HCl bu€er at pH 8.0 (basic
bu€er) and dialyzed overnight against the basic bu€er.
The dialyzate was applied to a Sephadex G-25 column
equilibrated with the basic bu€er. The desalted protein
material was subsequently loaded onto a 6Â35 cm col-
umn packed with DEAE-Toyopearl (Tosoh Co. Ltd.)
equilibrated with the basic bu€er. After non-adsorbed
proteins were eluted, adsorbed proteins were eluted with
the basic bu€er containing a 0±1.0 M linear gradient of
NaCl. The active enzyme fraction for the dehydrogena-
tion of (2S, 4S)-carveol (1a) was collected. The enzyme
fraction was further puri®ed on a Red-Toyopearl a-
nity column (Tosoh Co. Ltd.) (2Â20 cm) equilibrated
with the basic bu€er. After washing with the bu€er
solution, the proteins were eluted with a 0±1.0 M linear
gradient of NaCl in the basic bu€er. The active enzyme
fractions were pooled and used for subsequent experi-
ments.
In the activity assay for the oxidation of alcohols, the
assay mixture was composed of puri®ed enzyme (10 mg)
in 2 ml of 25 mM Tris±HCl bu€er at pH 8.0, 1.0 mmol
NADP+, and 1.0 mmol substrate solubilized by Triton
X-100 (0.3% ®nal concentration). The mixture was
incubated for 3 h at 35ꢀC and the amount of product
was estimated by GLC analyses. In the activity assay for
the reduction of ketones, the reactions were carried out
under the same conditions containing 2 ml enzyme
solution (10 mg) in 25 mM Na±Pi bu€er at pH 7.4,
1.0 mmol NADPH, and 1.0 mmol substrate solubilized
with 0.3% Triton X-100.
3.5. Partial amino acid sequence of allyl-ADH
Puri®ed protein on a 12.5% SDS±PAGE gel was
degraded partially by lysyl endopeptidase and the pep-
tide fragment on the SDS±PAGE gel was blotted onto a
polyvinylidene di¯uoride (PVDF) membrane (Immobi-
lon-PSQ, Millipore). The Coomassie stained peptide
band was excised. Sequence analysis was performed
with an online phenylthiohydantoin amino acids analy-
zer (Applied Biosystems Model 473A pulsed liquid
sequencer).
3.6. Cloning a cDNA encoding allyl-ADH
The puri®ed enzyme was subjected to sodium dodecyl
sulfate-polyacrylamide gel electrophoresis (SDS±
PAGE) on a 12.5% gel according to the standard pro-
tocol of Laemmli (Laemmli, 1970). The molecular mass
of the denatured enzyme was determined by SDS±
PAGE with the LMW electrophoresis calibration kit
(Pharmacia). The molecular mass of the native enzyme
Total cellular RNA was isolated from cultured cells
(9.3 g) of N. tabacum with guanidium thiocyanate±phe-
nol±chloroform extraction method (Chomczynski and
Sacchi, 1987). Using dT20-M4 primer and M-MLV
reverse transcriptase (Sawady Technology Co. Ltd.),
single-stranded cDNA was synthesized from the total
RNA. The single-stranded cDNA was used as template
for 1st PCR ampli®cation with EX Taq DNA poly-
merase (Takara Shuzo Co., Ltd.) and the primers
ADH1 (50-aaa gtr yta gaa tct ggr gat cca-30) and ADH2
(50-ctt gtg gga caa ttt gct aag atg-30). The ampli®cation
was carried out at 94ꢀC for 1 min for denaturation,
56ꢀC for 1 min for annealing, and 72ꢀC for 1 min for
was estimated by gel ®ltration through
a TSK
G3000SW HPLC column using aldolase, bovine serum
albumin, ovalbumin and ribonuclease A, respectively as
marker proteins.
The pH optimum of the enzyme reaction was esti-
mated by reaction of substrate 1a in a bu€er solution of

T. Hirata et al. / Phytochemistry 55 (2000) 297±303
303
synthesis, followed by 30 cycles. The ®rst PCR solution
was used as template for second PCR using the primers
ADH3 (50-gag aga cag ttt ttg tgt cag ctg-30) and ADH4
(50-ttc cag gta ggw gay tta gtc-30). 30-RACE-PCR was
performed with the following primers; nucleotides 448±
473, which are derived from the RT-PCR clone, for
upstream primer, and 30-RACE1 (50-tgg aag aat tcg cgg-
30) and M13M4 (50-gtt ttc cca gtc acg ac-30) for down-
stream primer. 50-RACE-PCR was performed with the
following primers; nucleotides 336±382, which are
derived from the RT-PCR clone, for downstream pri-
mer, and 50-RACE1 (50-tgg aag aat tcg cgg ccg ctt aag
ggg ggg ggg ggg-30) and 50-RACE2 (50-cgc ggc cgc tta-30)
for upstream primer. The ampli®ed cDNA fragment
was excised from an agarose gel, puri®ed, and ligated to
the pPCR-Script Amp vector (Stratagene). DNA
sequence analysis was performed with an ABI
PRISM310 Genetic Analyzer with BigDye terminator
cycle sequencing kit (PE Biosystems). Two full-length
cDNA (ally-ADH1 and ally-ADH2) of a 1243-bp were
sequenced (DDBJ/EMBL/GenBank accession no.
AB036735). The nucleotide sequence of the cDNA con-
tained a 1029-base ORF (Fig. 1).
terrigena that exhibit stereospeci®city towards secondary alcohols.
Biochemistry Journal 187, 703±709.
Bryant, F.O., Wiegel, J., Ljungdahl, L.G., 1988. Puri®cation and
properties of primary and secondary alcohol dehydrogenases from
Thermoanaerobacter ethanolicus. Applied Environmental Micro-
biology 54, 460±465.
Chomczynski, P., Sacchi, N., 1987. Single-step method of RNA isola-
tion by acid guanidinium thiocyanate±phenol±chloroform extrac-
tion. Analytical Biochemistry 162, 156±159.
Coleman, J.P., Perry, J.J., 1985. Puri®cation and characterization of
the secondary alcohol dehydrogenase from propane-utilizing
Mycobacterium vaccae strain JOB-5. Journal of General Micro-
biology 131, 2901±2907.
Dalziel, K., Dickinson, F.M., 1966. The kinetics and mechanism of
liver alcohol dehydrogenase with primary and secondary alcohols as
substrates. Biochemistry Journal 100, 34±46.
Dickinson, F.M., Dalziel, K., 1967. The speci®cities and con®gura-
tions of ternary complexes of yeast and liver alcohol dehy-
drogenases. Biochemistry Journal 104, 165±172.
Hamada, H., 1988. Enantioselectivity in the biotransformation of
mono- and bicyclic monoterpenoids with the cultured cells of
Nicotiana tabacum. Bulletin of the Chemical Society of Japan 61,
869±878.
Hirata, T., Hamada, T., Aoki, T., Suga, T., 1982. Stereoselectivity of
the reduction of carvone and dihydrocarvone by suspension cells of
Nicotiana tabacum. Phytochemistry 21, 2209±2212.
Laemmli, U.K., 1970. Cleavage of structural propeins during the
assembly of the head of bacteriophage T4. Nature 277, 680±685.
Lange, L.G., Sytkowski, A.J., Vallee, B.L., 1976. Human liver alcohol
dehydrogenase: puri®cation, composition, and catalytic features.
Biochemistry 15, 4687±4693.
Acknowledgements
Persson, B., Zigler Jr., J.S., Jornvall, H., 1994. A super-family of
medium-chain dehydrogenases/reductases (MDR). Sub-lines
including z-crystallin, alcohol and polyol dehydrogenases, quinone
oxidoreductase enoyl reductases, VAT-1 and other proteins. Eur-
opean Journal of Biochemistry 226, 15±22.
This work was supported in part by a Grant-in-Aid
for Scienti®c Research (No. 11694086) from the Minis-
try of Education, Science, Sports and Culture, Japan.
The authors thank Dr L. John Goad, the University of
Liverpool, UK, for critical reading of the manuscript.
The authors also thank the Instrument Center for Che-
mical Analysis of Hiroshima University for the mea-
Prelog, V., 1964. Speci®cation of the stereospeci®city of some oxidor-
eductases by diamond lattice sections. Pure and Applied Chemistry
9, 119±130.
Scrutton, N.S., Berry, A., Perham, R.N., 1990. Redesign of the coen-
zyme speci®city of a dehydrogenase by protein engineering. Nature
343, 38±43.
1
surements of H NMR and GC±MS spectra.
Tassin, J.P., Vandecasteele, J.P., 1972. Separation and characteriza-
tion of long-chain alcohol dehydrogenase isoenzymes from Psudo-
monas aeruginosa. Biochimica Biophysica Acta 276, 31±42.
Thompson, J.D., Higgins, D.G., Gibson, T.J., 1994. CLUSTAL W:
improving the sensitivity of progressive multiple sequence alignment
through sequence weighting, positions-speci®c gap penalties and
weight matrix choice. Nucleic Acids Research 22, 4673±4680.
Thorn, J.M., Barton, J.D., Dixon, N.E., Ollis, D.L., Edwards, K.J.,
1995. Crystal structure of Escherichia coli QOR quinone oxidor-
eductase complexed with NADPH. Journal of Molecular Biology
249, 785±799.
References
Babiychuk, E., Kushnir, S., Belles-Boix, E., Van Montagu, M., Inze,
D., 1995. Arabidopsis thaliana NADPH oxidoreductase homologs
confer tolerance of yeasts toward the thiol-oxidizing drug diamide.
Journal of Biological Chemistry 270, 26224±26231.
Baker, P.J., Britton, K.L., Rice, D.W., Rob, A., Stillman, T.J., 1992.
Structural consequences of sequence patterns in the ®ngerprint
region of the nucleotide binding fold. Implications for nucleotide
speci®city. Published erratum appears in Journal of Molecular
Biology, 228, 662±671.
Yokomizo, T., Ogawa, Y., Uozumi, N., Kume, K., Izumi, T., Shi-
mizu, T., 1996. cDNA cloning, expression, and mutagenesis study of
leukotriene B4 12-hydroxydehydrogenase. Journal of Biological
Chemistry 271, 2844±2850.
Barrett, C.H., Dodgson, K.S., White, G.F., Payne, W.J., 1980. Pre-
liminary observations on alcohol dehydrogenases in Comamonas