CHEMISTRY

THEJOURNAL OF BIOLOGICAL
Communication Vol. 262, No. 23, Issue of August 15, pp. 10911-10913.1987
0 1987 by The American Society for Biochemistry and Molecular Biology, Inc.
Printed in U.S.A.

Hydride Transfer
Stereospecificity of Rat Liver
Aldehyde Dehydrogenases*
(Received for publication, April 14, 1987)
Kimberly Harper Jones+#, Ronald Lindahl$$,
David C. Bakergll, and Russell Timkovichgllll
A
From the $Department of Biology, §BiochemistryProgram, STRUCTURE
1
and the llDepartment of Chemistry, University of Alabama,
Tuscaloosa, Alabama35487 There has beenan earlier report onthe stereospecificity of a
liveracetaldehydedehydrogenaseby a non-NMR method
The stereospecificity of hydride transfer to NAD+by (10). However, the earlier work used a crude preparation. In
several forms of rat liveraldehyde dehydrogenase was light of the multiple forms now known to exist of aldehyde
determined by a nuclear magnetic resonance method. dehydrogenases, the earlier report is subject to some ambi-
The forms included several mitochondrial and micro- guity. The authors reported a stoichiometry that was nonin-
somal isozymes from normal liver, as well as isozymes
from xenobiotic-treated and tumor cells. The proton tegral (65% of one isomer, 20% of the other, and 15% unex-
added to NAD+ comes exclusively from the aldehyde plained), and hypothesized a variety of side reactions that
substrate and in all cases wasA (pro-R)-stereospecific. might account for their results.A reinvestigation of the ster-
eospecificity seemed warranted.
EXPERIMENTALPROCEDURES
Hepatic aldehyde dehydrogenases (aldehyde:NAD+ oxido- From normal rat liver, mitochondrial aldehyde dehydrogenase iso-
reductase (EC 1.2.1.3) and aldehyde:NAD(P)+ oxidoreductase zymes I and I1 (MTI and MTII), andmicrosomal isozymes I and I1
(EC 1.2.1.5)) are pyridine nucleotide-linked enzymes catalyz- (MCI and MCII) were purified as described previously (9). The
phenobarbital-induced enzyme (PB) and the 2,3,7,8-tetrachlorodi-
ing the oxidationof aldehydes to carboxylic acids.
benzo-p-dioxin-induced enzyme (TCDD) were also purified as de-
RCHO + NAD(P)+ + HZ0 + RCOOH + NAD(P)H + H+ scribed previously (5), as was the dehydrogenase from rat hepatoma
cells (HTC)(7)and a plasmid-encoded hepatic tumor-associated
They havebeen found in most mammals in multiple molecular aldehyde dehydrogenase (P3A1) isolated from Escherichia coli strain
forms (1-4). In rat liver, further aldehyde dehydrogenases are HBlOl (11).
induced by xenobiotics (see Ref. 5 and references therein) or The stereospecificity of NAD(P)+reduction was determined by a
during hepatocarcinogenesis (see Refs. 6 and 7 and references ‘H NMR technique (12, 13) with several conceptually important
cited therein). A rat hepatoma cell line hasbeen developed as modifications. The procedure of Arnold and co-workers (12,13) starts
an in vitro model for studying the regulation of aldehyde with deuterium-labeled NAD+ atthe pyridine 4-position. In the
dehydrogenase activity in the latter case (7). present study, normal protic NAD+ and NADP+ (Sigma) were used,
and the aldehyde substrate was prepared with deuterium in the
Pyridinenucleotide-linkedenzymes are wellknownto aldehyde proton position. In this fashion, it was possible to demon-
transfer hydride stereospecifically to (or from) the pyridine strate that the H atom transferred to the oxidized coenzyme came
4-position of the oxidized (reduced) coenzyme (for a review exclusively from the aldehyde and not from solvent water.
see Ref. 8). A-stereospecific enzymes transfer hydride to or The common aldehyde substrate benzaldehyde-1-d was used in all
from the A side (pro-R, top side of Structure l), while B- assays. Although it is not an optimum substrate for all enzyme forms,
stereospecific (pro-S) enzymes transfer at the opposite side. it is an acceptable one that undergoes reaction at reasonable rates.
Known pyridine nucleotide-linked enzymes seem to be ap- The labeled compound was synthesized by a published procedure (14)
proximately equally distributed in their stereospecificity. In and contained 94% deuterium at the aldehyde position.
rat liver the large number of basal and inducible enzymes Assays were performed at room temperature in 60 mM potassium
phosphate buffer, pH 8.5. Concentrations of benzaldehyde-1-d and
have been shown to have distinct properties (5, 7, 9). Differ- NAD+ orNADP+ were typically 1 mM each, although in some exper-
ences exist in subcellular localization (microsomal, mitochon- iments the concentrations were raised as high as 4 mM. Enzyme (3.6
drial, or cytosolic), subunit composition, stability, and, most up to 534 mIU) was added as a concentrated aliquot either in the
importantly, kinetic properties including velocity constants same assay buffer or in buffer containing in addition 1 mM EDTA, 1
and affinities for different aldehydes and NAD+ or NADP+ mM mercaptoethanol, and 0.2% (v/v) Triton X-100. Reactions were
as coenzyme. Therefore, it was possible that differences in followed either spectrophotometrically at 340 nm or by NMR spec-
stereospecificityalsoexisted.Thispresentcommunication troscopy, as will be discussed. After the reaction had progressed to
reports the stereospecificityof the multiple forms of rat liver between 30 to 99+% completion, the assay mixture was frozen,
aldehydedehydrogenasewithrespecttohydridetransfer. lyophilized, and redissolved in the NMR solvent deuterium oxide
(99.8% deuterium, Aldrich Chemical Co.) containing 0.5 mM 3-(tri-
* This work was supported by Grants GM 36264 (to R. T.) and CA methyl)-tetradeutero-sodiumpropionate as internal reference. Very
21103 (to R. L.) from the National Institutes of Health. The costs of slight differences (0.02 ppm) for chemical shifts for NADH and its
publication of this article were defrayed in part by the payment of specifically deuterated form were observed in the present study com-
page charges. This article must therefore be hereby marked “aduer- pared to the literature (12, 13) because of different temperatures and
tisement” in accordance with 18U.S.C. Section 1734 solelyto indicate concentrations employed. It was not necessary to purify products
this fact. because, in the critical region of 2-3 ppm, only the reduced pyridine
11 To whom correspondence should be addressed Dept. of Chem- nucleotides contributed an NMR detectable signal. Spectra were
istry, University of Alabama, Tuscaloosa, AL 35487. obtained on a Nicolet NTCBOO spectrometer at 200.06 MHz.

10911
10912 Stereospecificity
Dehydrogenase
of Aldehyde
RESULTS AND DISCUSSION TABLEI
Summary of kinetic parameters for rat liver aldehyde dehydrogenase
TheHTC aldehyde dehydrogenase was active in 99+%
Benzaldehyde K, values are millimolar while coenzymevalues are
deuterium oxide buffered with 60 mM phosphate to pH 8.5. micromolar. Vmaxvalues are relative to NAD+. Isozyme designations
At 1 mM in both substrates, it was 57% as active compared are: MTI and MTII, mitochondrial aldehyde dehydrogenase isozymes
to assays conducted in normal protic water. Therefore, it was I and 11; MCI and MCII, microsomal isozymes I and 11; PB, pheno-
possible to observe the appearance of NADH produced from barbital-induced enzyme; TCDD, 2,3,7,8-tetrachlorodibenzo-p-
benzaldehyde-1-d and NAD+ in NMR spectra as a function dioxin-induced enzyme; P3A1, plasma-encoded hepatic tumor-asso-
of time. Typical results are shown in Fig. 1. The resonance ciated aldehyde dehydrogenase; HTC, dehydrozenase from rat hepa-
toma cells.
that grows in at a chemical shift of2.65 ppm is the 4B
methylene proton of NADH arising from the 4 aromatic Benzaldehyde NAD+ NADP+ Stereo-
Isozyme specifi-
proton of NAD+. (The chemical shift of the latter is a t 8.95 V,. K,,, V,. K, V- Km citv
ppm.) The deuterium transferring from benzaldehyde-1-d is MTI 0.35 1.6 910
1.00
0.49 36 A
itself spectrally invisible, althoughits presence is certain MTII 0.80 50 440
1.00
0.55 22 A
because the scalar coupling pattern andchemical shift for the MCI 1.40 5.9 1.00 97 0.60 440 A
observed proton (originally on the NAD') is definitive for a MCII 0.34 3.0 1.00
270 0.27 95 A
methylene pair 'H, 'H. The observed spectrum matched that PB 0.02 1.3 1.00 2.3 100
0.01 A
TCDD 4.54 0.5 1.00
210 4.44 73 A
previously reported for Structure 1, which is [4A-'H, 4B-'H] P3A1 5.66 0.8 1.00
420 5.85 74 A
NADH (12, 13). The observed shift is clearly distinct from a HTC 5.66 0.8 1.00
420 5.85 74 A
shift of 2.77 ppm which would correspond to theother isomer.
Hence, the enzyme has A-stereospecificity.
Additional conclusions were possible in light of the follow- comes from the substrate aldehyde and not from solvent.
ing observations. When the reaction was conducted in 'HzO, When NADP+ was substituted for NAD', A-stereospecificity
the same final spectrum was obtained. By integration of was again observed.
appropriate NMR peaks in other regions of the spectrum, it Table I summarizes some kinetic parameters andthe deter-
was observed that 0.97 f 0.05 eq of [4A-'H, 4B-'HINADH mined stereospecificity for the other enzyme forms. A-stere-
were formed per equivalent of NAD+ consumed or per equiv- ospecificity was observed in all cases tested. The conformity
alent of benzaldehyde-1-d consumed. This confirms the stoi- is not as exciting as it would have been to observe a clear
chiometry of the reaction and that thehydrogen transferred difference between, for example, the normal and tumor-re-
lated forms. However,firm data now experimentally establish
the stereospecificity and eliminate speculation or assumption.
n The exact ancestral relationships between aldehyde dehy-
drogenase and other dehydrogenases have yet to be defined.
Two other dehydrogenases of interest for comparison are
alcohol and glyceraldehyde-3-phosphate dehydrogenase (15-
17). In many pathways alcohol dehydrogenase provides the
substrate for aldehyde dehydrogenase, especially in ethanol
metabolism. In this context it is interesting to note that
alcohol dehydrogenase is a dimer, requires a bound zinc atom
for activity, and has full site reactivity (15). Both alcohol and
aldehyde dehydrogenases are A (pro-R)-specific. Aldehyde
dehydrogenase is more closelyanalogous to glyceraldehyde-3-
phosphate dehydrogenase structurally. Both function as tet-
ramers of approximately equal size, both show half-site reac-
tivity with aldehyde substrates, and following substrate bind-
ing, the complex is oxidized to anintermediate which is then
hydrolyzed to an acid (16, 18).Interestingly, glyceraldehyde-
3-phosphate dehydrogenase is a B (pro-S)-specific enzyme.
FIG. 1. Production of [4A-'H, 4B-'H] NADH by HTC alde-
hyde dehydrogenase. as a function of time. The assay mixture REFERENCES
contained, in a final total volume of 0.5 ml, 1mM benzaldehyde-1-d, 1. Leicht, N., Heinz, F., and Freimuller, B. (1978) Eur. J. Biochem.
1 mM NAD+, 60 mM potassium phosphate, 0.5 mM 3-(trimethyl)- 83,189-196
tetradeutero-sodium propionate, and 0.28 mg of enzyme all in 99.8% 2. Feldman, R. I., and Weiner, H. (1972) J. Bwl. C h m . 247, 260-
2H20. Thespectral region between 2 and 3 ppm is displayed, and the 266
single peak on the right is from residual acetone (2.04 ppm) that is a
3. Crow, K. E., Kitaon, T. M., MacGibbon, A. K. H., and Batt, R.
contaminant in the stock NAD+ preparation. Spectrum A was taken
within 10 min of the addition of enzyme to the assay and is a nominal D. (1974) Biochim. Biophys. Acta360,121-128
time 0 trace. Spectra B-E were taken a t 3-h intervals. Spectrum F is 4. Greenfield, N. J., and Pietruszko, R. (1977) Biochim. Bwphys.
pure NADH with both methylene protons 'H, taken at pH8.5. Inset Acta 483,35-45
G is a computer-generated simulation of the expected line positions 5. Lindahl, R., and Evces, S. (1984) J. Bwl.Chem. 269, 11991-
for the 4A and 4B methylene protons of NADH, whereby the chemical 11996
shift of the 4A proton is 2.77 ppm and is coupled to the pyridine 6. Lindahl, R. (1982) in Enzymology of Carbonyl Metabolism: Alde-
proton 5 (J(4A,5) = 3.1 Hz), as well as to 4B (J(4A.4B) = 18 Hz), hyde Dehydrogenase and AMo/Keto Reductase, pp.121-135,
and thechemical shift of the 4B proton is 2.65 ppm and is coupled to Alan R. Liss, Inc., New York
proton 5 (J(4B,5) = 3.8 Hz) as well as 4A. The tick marks indicate 7. Lindahl, R., Baggett, D. W., and Winters, A. L. (1985) Biochim.
the chemical shifts of4A and 4B. Because of the strong scalar Biophys. Acta 843,180-185
coupling, there is no "peak" at these chemical shifts in thefully protic 8. Popjak, G. (1970) in The Enzymes (Boyer, P. D. ed) 3rd Ed., Vol.
NADH, while in [4A-'H, 4B-'HINADH the peaks form a doublet 2, p. 115, Academic Press, New York
(because of the still present 4B-5 coupling) symmetrical about the 9. Lindahl, R., and Evces, S. (1984) J. Biol.Chem. 269, 11986-
chemical shift. 11990
 StereospecificityDehydrogenase
of Aldehyde 10913
10. Levy, H. R., and Vennesland, B. (1957)J. Biol. Chem. 228,85- Vallee, B. L. (1987)in Advances in Biomedical AlcoholResearch,
96 pp. 13-23,Pergamon Press, New York
11. Jones, D. E., Brennan, M. D., Harper, K., and Lindahl, R. (1986) 16. Pietruszko, R., Ferencz-Biro, K., and MacKerrell, A. D. (1985)
Proc. Am. Assoc. Cancer Res. 27,7 in Enzymology of Carbonyl Metabolism 2: Aldehyde Dehydro-
12. Arnold, L. J., Jr., You,K., Allison, W. S., and Kaplan, N. 0. genase, A&-Keto Reductase, and Alcohol Dehydrogenase, pp.
29-41,Alan R. Liss, Inc., New York
(1976)Biochemistry 15,4844-4849 17. Hempel, J., von Bahr-Lindstrom, H., and Jornvall, H. (1984)
13. Arnold, L. J., Jr., and You, K. (1978)Methods Enzyml. LIV, Eur. J. Biochem. 141.21-35
223-232 18. Weiner, H. (1982)in Enzymology of Carbonyl Metabolism: Alde-
14. Craig, J. C.,and Ekwuribe, N. N. (1980)Synthesis 11,909 hyde Dehydrogenase and AldolKeto Reductase, pp. 1-9,Alan R.
15. Jornvall, H.,Hempel, J., von Bahr-Lindstrom, H., Hoog, J., and Liss, Inc., New York