Elucidation of reaction scheme describing malondialdehyde - Acetaldehyde - Protein adduct formation

Article abstract of DOI:10.1021/tx000222a

Malondialdehyde and acetaldehyde react together with proteins and form hybrid protein conjugates designated as MAA adducts, which have been detected in livers of ethanol-fed animals. Our previous studies have shown that MAA adducts are comprised of two distinct products. One adduct is composed of two molecules of malondialdehyde and one molecule of acetaldehyde and was identified as the 4-methyl-1,4-dihydropyridine-3,5-dicarbaldehyde derivative of an amino group (MDHDC adduct). The other adduct is a 1:1 adduct of malondialdehyde and acetaldehyde and was identified as the 2-formyl-3-(alkylamino)butanal derivative of an amino group (FAAB adduct). In this study, information on the mechanism of MAA adduct formation was obtained, focusing on whether the FAAB adduct serves as a precursor for the MDHDC adduct. Upon the basis of chemical analysis and NMR spectroscopy, two initial reaction steps appear to be a prerequisite for MDHDC formation. One step involves the reaction of one molecule of malondialdehyde and one of acetaldehyde with an amino group of a protein to form the FAAB product, while the other step involves the generation of a malondialdehyde - enamine. It appears that generation of the MDHDC adduct requires the FAAB moiety to be transferred to the nitrogen of the MDA - enamine. For efficient reaction of FAAB with the enamine to take place, additional experiments indicated that these two intermediates likely must be in positions on the protein of close proximity to each other. Further studies showed that the incubation of liver proteins from ethanol-fed rats with MDA resulted in a marked generation of MDHDC adducts, indicating the presence of a pool of FAAB adducts in the liver of ethanol-fed animals. Overall, these findings show that MDHDC - protein adduct formation occurs via the reaction of the FAAB moiety with a malondialdehyde - enamine, and further suggest that a similar mechanism may be operative in vivo in the liver during prolonged ethanol consumption.

Full text of DOI:10.1021/tx000222a

822
Chem. Res. Toxicol. 2001, 14, 822-832
Elu cid a tion of Rea ction Sch em e Descr ibin g
Ma lon d ia ld eh yd e-Aceta ld eh yd e-P r otein Ad d u ct
F or m a tion
Dean J . Tuma,*,†,‡ Mark L. Kearley,†,§ Geoffrey M. Thiele,†,‡ Simon Worrall,|
Alvin Haver,†,‡ Lynell W. Klassen,†,‡ and Michael F. Sorrell†,‡
Department of Veterans Affairs Alcohol Research Center and the Department of Internal Medicine,
University of Nebraska Medical Center, Omaha, Nebraska 68105, the Department of Chemistry,
Creighton University, Omaha, Nebraska 68178 and the Alcohol Research Unit,
Department of Biochemistry, University of Queensland, Queensland 4072, Australia
Received October 19, 2000
Malondialdehyde and acetaldehyde react together with proteins and form hybrid protein
conjugates designated as MAA adducts, which have been detected in livers of ethanol-fed
animals. Our previous studies have shown that MAA adducts are comprised of two distinct
products. One adduct is composed of two molecules of malondialdehyde and one molecule of
acetaldehyde and was identified as the 4-methyl-1,4-dihydropyridine-3,5-dicarbaldehyde
derivative of an amino group (MDHDC adduct). The other adduct is a 1:1 adduct of
malondialdehyde and acetaldehyde and was identified as the 2-formyl-3-(alkylamino)butanal
derivative of an amino group (FAAB adduct). In this study, information on the mechanism of
MAA adduct formation was obtained, focusing on whether the FAAB adduct serves as a
precursor for the MDHDC adduct. Upon the basis of chemical analysis and NMR spectroscopy,
two initial reaction steps appear to be a prerequisite for MDHDC formation. One step involves
the reaction of one molecule of malondialdehyde and one of acetaldehyde with an amino group
of a protein to form the FAAB product, while the other step involves the generation of a
malondialdehyde-enamine. It appears that generation of the MDHDC adduct requires the
FAAB moiety to be transferred to the nitrogen of the MDA-enamine. For efficient reaction of
FAAB with the enamine to take place, additional experiments indicated that these two
intermediates likely must be in positions on the protein of close proximity to each other. Further
studies showed that the incubation of liver proteins from ethanol-fed rats with MDA resulted
in a marked generation of MDHDC adducts, indicating the presence of a pool of FAAB adducts
in the liver of ethanol-fed animals. Overall, these findings show that MDHDC-protein adduct
formation occurs via the reaction of the FAAB moiety with a malondialdehyde-enamine, and
further suggest that a similar mechanism may be operative in vivo in the liver during prolonged
ethanol consumption.
protein conjugates that have been designated as MAA
adducts (5). These adducts have been detected in livers
of ethanol-fed rats (5), and circulating antibodies against
MAA adducts have been identified in ethanol-fed rats (6)
and in patients with alcohol-induced liver disease (7).
Further studies have shown that proteins with MAA
adducts can serve as immunogens and can alter the
function of various cell types in the liver. Adduction of
proteins with MDA and acetaldehyde results in a potent
immune response, resulting in circulating antibodies
against both the carrier protein epitopes, as well as to
the MAA adduct epitope (8). MAA adducts also appear
to possess both pro-inflammatory and profibrogenic
properties. These adducts have been shown, in vitro, to
increase the secretion of chemokines and to upregulate
ICAM expression in hepatic stellate cells (9) and to
increase the secretion of TNFR and chemokines, to
upregulate adhesion molecule expression, and to enhance
fibronectin secretion in liver endothelial cells (10). These
findings suggest that MAA adducts may play a signifi-
cant role in the development of alcoholic liver injury.
In tr od u ction
Ethanol metabolism in the liver results in the forma-
tion of acetaldehyde, which can covalently bind to hepatic
proteins, and this process has been implicated as playing
an important role in the development of alcoholic liver
injury (1, 2). There is also considerable evidence indicat-
ing that chronic ethanol consumption induces oxidative
stress and lipid peroxidation in the liver (3, 4) which in
turn generates another reactive aldehyde, malondialde-
hyde (MDA).1 Studies from our laboratory have demon-
strated that acetaldehyde and MDA can react together
with proteins in a synergistic manner and form hybrid
* To whom correspondence should be addressed. Phone: (402) 346-
8800 (ext. 3548). Fax: (402) 449-0604. E-mail: dean@tuma.net.
† Department of Veterans Affairs Alcohol Research Center.
‡ Department of Internal Medicine.
§ Department of Chemistry.
| Alcohol Research Unit.
1
Abbreviations: MDA, malondialdehyde; MAA, malondialdehyde-
acetaldehyde; FAAB, 2-formyl-3-(alkylamino)butanal; MDHDC, 4-meth-
yl-1,4-dihydropyridine-3,5-dicarbaldehyde; BSA, bovine serum albu-
min; PLL, poly-L-lysine; hexyl-MDHDC, 1-hexyl-MDHDC; ethyl-
MDHDC, 1-ethyl-MDHDC; hexyl-FAAB, 2-formyl-3-(hexylamino)-
butanal; ELISA, enzyme-linked immunosorbent assay.
Chemical characterization and structural analysis of
MAA-protein adducts is an essential step in clarifying
10.1021/tx000222a CCC: $20.00 © 2001 American Chemical Society
Published on Web 05/26/2001

Malondialdehyde-Acetaldehyde Adduct Formation
Chem. Res. Toxicol., Vol. 14, No. 7, 2001 823
(15) and was synthesized by the method of Nair et al. (16). This
procedure gives N-ethylaminopropenal as a mixture of isomers
(E and Z). The 1H NMR signals for each of these compounds
are reported below. 1-Ethyl-4-methyl-1,4-dihydro-3,5-pyridinedi-
carboxaldehyde (ethyl-MDHDC) is a new compound and was
synthesized by the same procedure as hexyl-MDHDC (11). It
was characterized by 1H and 13C NMR, IR spectroscopy, and
microanalysis (C, H, N) performed by Atlantic Microlab, Inc.
(Norcross, GA). The IR spectrum was recorded using a Nicolet
Impact 410 spectrophotometer and melting point was deter-
mined on a MelTemp melting point apparatus.
1
1
Na-MDA. H NMR (D2O) δ 8.6 (d, 2H), 5.3 (t, 1H). H NMR
(pD 7.4 buffer) δ 8.7 (s, 2H). In buffer, the R-hydrogen exchanges
with the solvent within two minutes.
Hexyl-FAAB. 1H NMR (pD 7.4 buffer) δ 8.6 (s, 2H, -CHO),
4.2 (q, 1H, CH-CH3), 2.9 (t, 3H, N-CH2), 1.6 (m, 2H, CH2-
C-N), 1.5 (d, 3 H, CH3-CH), 1.2-1.4 (m, (CH2)3), 0.9 (t, 3H,
CH3).
N-Ethylaminopropenal. 1H NMR (pD 7.4 buffer) δ 8.7 (d, 1H,
-CHO), 8.6 (d,1H, -CHO), 7.6 (m, 1H and 1H, dCH-N), 5.4
(m, 1H and 1H, CHdC), 3.3 (q, 2 H, N-CH2), 3.2 (q, 2H,N-
CH2), 1.3 (t, 3H, CH3), 1.2 (t, 3H, CH3). These signals are for
the mixture of E and Z isomers.
Ethyl-MDHDC. Column chromatography (Rf ) 0.56, 1:9
methanol:chloroform) followed by recrystallization from metha-
nol afforded the product as yellow needles: mp 154-155 °C; IR
(KBr) 3031, 2963, 1635, 1567, 715 cm-1;1H NMR (pD 7.4 buffer)
δ 9.2 (s, 2H, -CHO), 7.3 (s, 2H, CdCH), 3.8 (q, 1H, CH-CH3),
3.6 (q, 2H, N-CH2), 1.3 (t, 3H, CH3-CH2), 1.0 (d, 3H, CH3-
CH); 13C NMR (pD 7.4 buffer) δ 193.0, 150.7, 123.3, 50.2, 22.3,
21.6, 14.4. Anal. calcd for C10H13NO2: C, 67.02; H, 7.31; N, 7.82.
Found: C, 66.87; H, 7.21; N, 7.74.
Hexyl-MDHDC. 1H NMR (pD 7.4 buffer) δ 9.2 (s, 2H, -CHO),
7.3 (s, 2H, CdCH), 3.7 (q, 1H, CH-CH3), 3.6 (t, 2H, N-CH2),
1.2-1.8 (m, 8 H, (CH2)4), 1.0 (d, 3H,CH3-CH), 0.9 (t, 3H, CH3).
P r ep a r a tion of F AAB-BSA a n d F AAB-P LL. A solution of
BSA (10 mg/mL) was incubated with 20 mM acetaldehyde and
2 mM MDA, and a solution of PLL (10 mg/mL) was incubated
with 10 mM acetaldehyde and 1 mM MDA. Incubations were
conducted in phosphate buffer (0.1 M, pH 8.2) at 15 °C for 24 h
in polypropylene vessels that were sealed to minimize losses
due to volatility. At the end of incubation, the reaction mixtures
were exhaustively dialyzed against phosphate buffer (0.1 M, pH
7.4) for 24 h at 4 °C. These reaction conditions generated MAA
adducts of BSA and PLL where 99% or greater of the total
adducts formed was the FAAB adduct and less than 1% was
the MDHDC adduct.
P r ep a r a tion of MAA Ad d u cts (con sistin g of a m ixtu r e
of F AAB a n d MDHDC a d d u cts) w ith BSA a n d P LL.
Solutions of BSA (1 mg/mL) and PLL (1 mg/mL) were incubated
with 1 mM acetaldehyde and 1 mM MDA in phosphate buffer
(0.1 M, pH 7.4) at 37 °C for 3 days. At the end of incubation,
the reaction mixtures were exhaustively dialyzed against
phosphate buffer (0.1 M, pH 7.4) for 24 h at 4 °C. These reaction
conditions yielded MAA adducts composed of 60% FAAB adduct
and 40% MDHDC adduct for both BSA and PLL.
P r ep a r a tion of MDA Sch iff Ba ses of BSA. BSA (1 mg/
mL) was incubated with 1 mM MDA in phosphate buffer (0.1
M, pH 7.4) at 37 °C for 24 h. At the end of incubation, the reac-
tion mixture was dialyzed against phosphate buffer for 2 h at 4
°C to get rid of the free, unreacted MDA. This procedure has
been shown to yield Schiff base adducts of MDA with BSA, par-
ticularly the mono Schiff base derivative, aminopropenal (17,
18).
Qu a n tifica tion of MDHDC a n d F AAB Ad d u cts. Total
MAA adducts were quantified by a [14C]acetaldehyde binding
assay as previously described (5, 19). This assay measures the
nanomoles of stably bound acetaldehyde per milligram of protein
or polypeptide, and based on the known structures of MAA
adducts which contain 1 mol of acetaldehyde per mol of MAA
derivative, this assay can be used to quantify total MAA adducts
on a protein. MDHDC adducts were determined directly by a
F igu r e 1. Chemical structures of MAA-protein adducts.
the role of these adducts in ethanol-induced hepatotox-
icity. Our previous work showed that two major adducts
result from the reaction of MDA and acetaldehyde with
proteins, and their structures are depicted in Figure 1
(11, 12). One adduct is a highly fluorescent, cyclic product
composed of two molecules of MDA and one molecule of
acetaldehyde and has been identified to be a 4-methyl-
1,4-dihydropyridine-3,5-dicarbaldehyde derivative of an
amino group (MDHDC adduct). The other adduct is a 1:1
adduct of MDA and acetaldehyde, forming a 2-formyl-3-
(alkylamino)butanal derivative of an amino group (FAAB
adduct). The mechanism which accounts for the forma-
tion of these adducts both in vitro and in vivo is still
unclear. Our previous studies showed that when MDA
and acetaldehyde are allowed to react with proteins, the
FAAB adduct is the prominent adduct at early reaction
times, whereas, the formation of the MDHDC adduct is
delayed to more prolonged times of reaction (11, 12).
These findings suggest that the FAAB adduct may serve
as a precursor for the formation of the MDHDC adduct.
Therefore, the purpose of the present study was to gain
information concerning the mechanism of MAA adduct
formation, focusing on whether the FAAB adduct can
serve as a precursor for the generation of the MDHDC
adduct. Both in vitro and in vivo experiments were
conducted for this purpose, and some additional impor-
tant information concerning reaction sites on proteins
undergoing MAA adduction was obtained.
Exp er im en ta l P r oced u r es
Ma ter ia ls. [1,2-14C]Acetaldehyde (5 mCi/mmol) was pur-
chased from New England Nuclear (Boston, MA). Radiolabeled
acetaldehyde was received from the manufacturer frozen as an
aqueous solution (1 mCi/mL), thawed, and diluted to 250 µCi/
mL with distilled water, rapidly refrozen, and stored at -70
°C. The specific activity of the acetaldehyde was checked as
described by Miwa et al. (13). Nonradioactive acetaldehyde was
obtained from the Aldrich Chemical Co. (Milwaukee, WI).
Bovine serum albumin (BSA) (crystallized, lyophilized, and fatty
acid free) and poly-L-lysine (PLL) (mol. wt. 16 100) were from
Sigma Chemical Co. (St. Louis, MO). All other chemicals were
of analytical grade.
Or ga n ic Syn th esis. MDA was synthesized as the sodium
salt by the method of Grabowski and Autrey (14). 1-Hexyl-4-
methyl-1,4-dihydro-3,5-pyridinedicarboxaldehyde (hexyl-MDH-
DC) and 2-formyl-3-(hexylamino)butanal (hexyl-FAAB) were
previously synthesized in this laboratory (11, 12). N-Ethylami-
nopropenal has been previously characterized by Makin et al.

824 Chem. Res. Toxicol., Vol. 14, No. 7, 2001
Tuma et al.
Ta ble 1. Effects of In cu ba tion of Aceta ld eh yd e or MDA
w ith F AAB-BSA on th e F or m a tion of MDHDC-BSA
fluorescence assay as previously described (5, 11). Fluorescence
measurements (excitation max 398 nm and emission max at 460
nm) were obtained on postdialysis samples using a Perkin-
Elmer (Norwalk, CT) LS-5B spectrophotofluorometer attached
to a Perkin-Elmer GP-100 graphics printer. The data were
expressed as nanomoles of MDHDC adduct per milligram of
protein or polypeptide, using hexyl-MDHDC as a standard. The
quantity of FAAB adducts on a protein or polypeptide were
estimated by subtracting the amounts of MDHDC adducts from
total MAA adducts. The suitability of this method is based on
our previously reported finding that MAA adducts are comprised
almost entirely of the FAAB and MDHDC adducts (12), and this
finding was again confirmed by applying 13C NMR spectra
analyses of the variously generated MAA adducts used in this
study as previously described (12).
NMR Sp ectr oscop y. All 1H and 13C NMR spectra were
recorded on a Varian INOVA 300 spectrometer, and the solvent
was 0.5 M phosphate buffer in D2O (pD 7.4). For 1H NMR, the
chemical shifts (δ) are relative to the HOD signal (δ ) 4.81 ppm).
For 13C NMR, the chemical shifts are relative to 3-(trimethyl-
silyl)propionic-2,2,3,3-d4 acid, sodium salt (δ ) -2.58 ppm). Each
1H NMR reaction consisted of adding 0.030 mmol of each
reactant to 0.6 mL of D2O buffer, and acquiring an initial 1H
NMR 16-scan spectrum. The NMR tube was then placed in a
37 °C oil bath and removed at various time intervals and 16-
scan spectra acquired.
reaction conditionsa
MDHDCb
FAABb
FAAB-BSA (starting material)
FAAB-BSA + no additions (control)
FAAB-BSA + 2 mM acetaldehyde
FAAB-BSA + 10 mM acetaldehyde
FAAB-BSA + 2 mM MDA
1.7
6.5
6.2
136.4
31.1
36.4
30.9
48.2
50.4
5.1
21.2
33.9
0.6
FAAB-BSA + 10 mM MDA
unmodified BSA + 2 mM MDA
unmodified BSA + 10 mM MDA
2.4
a
FAAB-BSA (10 mg/mL), prepared as described in the Experi-
mental Procedures, or unmodified BSA (10 mg/mL) were incubated
in phosphate buffer (pH 7.4) at 37 °C for 24h in the absence or
presence of acetaldehyde or MDA. Following 24-h dialysis at 4 °C,
the levels of FAAB-BSA and MDHDC-BSA were determined. b The
data are expressed as nanomoles per milligram of BSA and
represent the averages of two determinations.
An a lysis of MDHDC-P r otein Ad d u cts in Liver s of
Eth a n ol-F ed a n d Con tr ol Ra ts. Male Wistar rats (150-160
g) were pair-fed the Lieber-DeCarli ethanol-containing diet and
the control liquid diets (20) for up to 11 weeks as previously
reported (21). Liver homogenates from these animals were
prepared in Hepes-buffered sucrose (pH 7.4). In some cases, the
livers were perfused as previously described (22) in the presence
or absence of MDA prior to the preparation of liver homogenates.
Relative levels of MDHDC-liver protein adducts were deter-
mined by a direct ELISA as previously described in detail (5,
6). An affinity-purified antibody that specifically recognizes
MDHDC epitopes on proteins (11) was used to assay for
MDHDC adducts on liver proteins. Optical density at 405 nm
was measured by a Dynatech Micro ELISA Reader MR700
(Dynatech, Chantilly, VA), and increasing optical density read-
ings were reflective of increasing levels of MDHDC adducts.
Resu lts
Rea ction of F AAB-P r otein Ad d u cts w ith MDA.
When a 10-fold molar excess of acetaldehyde relative to
MDA was incubated with BSA at pH 8.2 and 15 °C, these
conditions resulted in the formation of MAA adducts
enriched with the FAAB adduct (99%). Reaction of this
BSA-FAAB adduct at pH 7.4 and 37 °C for 24 h with
MDA generated substantial levels of BSA-MDHDC with
a corresponding decrease in the levels of the FAAB
adduct; however, when no MDA or when acetaldehyde
was added under these conditions, minimal levels of the
MDHDC adduct were formed (Table 1). Furthermore,
when MDA was incubated with unmodified BSA, only
slight MDHDC adduct formation was observed (Table 1).
When the effects of concentration and time on the
conversion of the FAAB adduct to the MDHDC adduct
were examined, addition of MDA to the reaction mixtures
at all concentrations tested resulted in substantial for-
mation of the MDHDC adduct after 1 day of incubation,
and formation of the MDHDC product continued up to 5
days of incubation, but at a decreased rate (Figure 2A).
Also, MDA at concentrations of 5 and 10 mM produced
similar yields of the MDHDC adduct which were higher
than those observed for the 1 mM concentration (Figure
2A). On the other hand, when acetaldehyde was added
to the reaction mixture containing the FAAB adduct,
F igu r e 2. Effects of incubation of acetaldehyde or MDA with
FAAB-BSA on the formation of MDHDC-BSA. FAAB-BSA
(1 mg/mL), prepared as described in the Experimental Proce-
dures, was incubated in phosphate buffer (pH 7.4) at 37 °C in
the absence (O) or presence of 1 (4), 5 (0), and 10 mM (3)
acetaldehyde or 1 (2), 5 (9), and 10 mM (1) MDA for 1, 2, and
5 days. Following 24-h dialysis at 4 °C, the levels of MDHDC-
BSA (A) and FAAB-BSA (B) were determined. The data are
expressed as nanomoles per milligram of BSA and represent
the averages of three determinations.
minimal formation of the MDHDC product was observed,
and this was neither time nor concentration dependent
(Figure 2A). Interestingly, decreases in the FAAB adduct
were less pronounced in the presence of MDA compared
to acetaldehyde or to the untreated sample (Figure 2B).
A similar time- and concentration-dependent conversion
of the FAAB adduct to the MDHDC adduct was also
observed in the presence of MDA but not acetaldehyde
when PLL-FAAB was treated with these aldehydes
(Figure 3). Overall, these data indicate that the addition

Malondialdehyde-Acetaldehyde Adduct Formation
Chem. Res. Toxicol., Vol. 14, No. 7, 2001 825
Ta ble 2. Effects of In cu ba tion of Aceta ld eh yd e or MDA
w ith MAA-BSA on th e F or m a tion of MDHDC-BSA
reactions conditionsa
MDHDCb
FAABb
MAA-BSA (starting material)
MAA-BSA + no additions (control)
MAA-BSA + 1 mM acetaldehyde
MAA-BSA + 5 mM acetaldehyde
MAA-BSA + 10 mM acetaldehyde
MAA-BSA + 1 mM MDA
53.8
42.1
43.1
42.8
40.8
40.6
41.9
40.2
81.8
60.0
56.4
57.7
59.7
58.9
60.5
62.4
MAA-BSA + 5 mM MDA
MAA-BSA + 10 mM MDA
a
MAA-BSA (1 mg/mL), consisting of a mixture of FAAB-BSA
and MDHDC-BSA, was prepared as described in the Experimental
Procedures and was then incubated in phosphate buffer (pH 7.4)
at 37 °C for 24 h in the absence or presence of various concentra-
tions of acetaldehyde or MDA. Following 24-h dialysis at 4 °C,
the levels of FAAB-BSA and MDHDC-BSA were determined. b The
data are expressed as nanomoles per milligram of BSA and
represent the averages of three determinations.
aldehyde, methine, and methyl protons which have
chemical shifts of δ 8.6 (singlet), 4.2 (quartet), and 1.5
(doublet), respectively. The proton signals that indicate
the presence of the MDHDC product are the aldehyde,
vinyl, methine, and methyl protons which have chemical
shifts of δ 9.2 (singlet), 7.3 (singlet), 3.7 (quartet), and
1.0 (doublet), respectively. After 24 h, hexyl-FAAB had
decomposed significantly (Figure 4). The most significant
new peak is at δ 3.0 (triplet) which is consistent with
the formation of hexylamine. There were small signals
at δ 9.2 (singlet) and 7.3 (singlet) which indicate that a
small amount of hexyl-MDHDC was formed. In addition,
there were many small signals between δ 8.0-9.0, which
can be attributed to MDA-acetaldehyde products.2
Since the MDHDC product requires an extra mole of
MDA, the hexyl-FAAB was allowed to react with MDA
F igu r e 3. Effects of incubation of acetaldehyde or MDA with
FAAB-PLL on the formation of MDHDC-PLL. FAAB-PLL (1 mg/
mL), prepared as described in the Experimental Procedures, was
incubated in phosphate buffer (pH 7.4) at 37 °C in the absence
(O) or presence of 1 (4), 5 (0), and 10 mM (3) acetaldehyde or
1 (2), 5 (9), and 10 mM (1) MDA for 1,2, and 5 days. Following
24-h dialysis at 4 °C, the levels of MDHDC-PLL (A) and FAAB-
PLL (B) were determined. The data are expressed as nanomoles
per milligram of PLL and represent the averages of three
determinations.
1
(Figure 5). For MDA, the representative H NMR signal
is due to the aldehydic protons with a chemical shift of δ
8.7 (singlet). Again, after 24 h hexyl-FAAB decomposed,
but the MDHDC adduct was not a significant product.
The major products were hexylamine, as evidenced by
the signal at δ 3.0 (triplet), and 2,4-dihydroxymethylene-
3-methylglutaraldehyde, a 2:1 MDA:acetaldehyde adduct
shown in Figure 5. The representative signals for this
glutaraldehyde are at δ 8.3 (singlet), 4.1 (quartet), and
1.2 (doublet) as reported by Gomez-Sanchez et al. (23).
J ust like the decomposition of hexyl-FAAB, there were
small signals at δ 9.2 (singlet) and 7.3 (singlet) to show
that a small amount of hexyl-MDHDC was formed.
Since MDA was ineffective in converting the FAAB
adduct to the MDHDC adduct, we wondered whether
MDA had to be in the form of a Schiff base in order to be
reactive with the FAAB adduct; therefore, the reaction
of hexyl-FAAB and N-ethylaminopropenal, the MDA-
enamine of ethylamine, was monitored by 1H NMR
(Figure 6). After 24 h, more than half of the starting
hexyl-FAAB had been converted to hexylamine and ethyl-
MDHDC as evidenced by the new signals at δ 9.2
(singlet), 7.4 (singlet), and 1.1 (doublet). After 48 h, there
was virtually complete conversion to MDHDC product.
In the reaction of hexyl-FAAB and N-ethylaminoprope-
nal, it is the nitrogen from the N-ethylaminopropenal
that ends up in the MDHDC product. At δ 3.6-3.8, there
are two quartets which correspond to the methine
adjacent to the methyl and the methylene adjacent to the
nitrogen. If the nitrogen from hexyl-FAAB had been in
of MDA to the preformed FAAB-adduct results in the
conversion of this adduct to the MDHDC adduct.
Rea ction of MAA Ad d u cts (con sistin g of a m ix-
t u r e of F AAB a n d MDHDC a d d u ct s) w it h MDA.
MAA adducts on BSA were generated so that 60% of the
adducts were FAAB and 40% were MDHDC adducts.
Unlike the results obtained with the BSA-FAAB en-
riched adduct, reaction of this mixed adduct with MDA
did not result in increased formation of the MDHDC
adduct. Instead, treatment with MDA or acetaldehyde
or the untreated sample for 24 h of incubation resulted
in small decreases (20-30%) in the levels of both the
MDHDC and FAAB adducts (Table 2). Extending the
times of reaction under these experimental conditions to
2 and 5 days did not appreciably change the results (data
not shown). In addition, a mixed adduct of PLL (FAAB:
MDHDC ratio of 60:40) behaved similarly to the mixed
BSA adduct, yielding 20 to 30% decreases in both the
MDA-, acetaldehyde-treated, and untreated reaction
mixtures (data not shown).
Rea ction s of Hexyl-F AAB. The conversion of the
FAAB adduct to the MDHDC adduct in the presence of
MDA, which we observed for BSA- and PLL-FAAB, was
studied in more detail using small/organic analogues of
these adducts. By using an analogue such as hexyl-
FAAB, we could conveniently monitor the conversion of
this adduct to the MDHDC adduct by NMR spectroscopy.
Hexyl-FAAB was dissolved into pD 7.4 buffer and its
decomposition was followed for 24 h by 1H NMR. For the
FAAB adduct, the representative signals are due to
2
M.L.K., and D.J .T., unpublished observations.

826 Chem. Res. Toxicol., Vol. 14, No. 7, 2001
Tuma et al.
F igu r e 4. 1H NMR spectra for the decomposition of hexyl-FAAB at pD 7.4. The signals correspond to the compounds as follows:
hexyl-FAAB (A), hexylamine (B), and hexyl-MDHDC (C).
F igu r e 5. 1H NMR spectra for the reaction of hexyl-FAAB and MDA at pD 7.4. The signals correspond to the compounds as follows:
hexyl-FAAB (A), MDA (D), hexylamine (B), 2,4-dihydroxymethylene-3-methylglutaraldehyde (E), and hexyl-MDHDC (C).
the MDHDC, there would have been a quartet and a
triplet at δ 3.6-3.8.
Overall, these data are in contrast to the data obtained
for the polypeptide-FAAB adducts where addition of
MDA readily converted this adduct to the MDHDC
adduct. Instead, it takes the Schiff base derivative of
MDA, but not MDA alone, to convert the smaller, organic
derivative of FAAB (i.e., hexyl-FAAB) to the MDHDC
product.
Rea ction of F AAB-BSA or MAA-BSA w ith MDA
or a Sch iff Ba se-Der iva tive of MDA. Since the results
obtained when hexyl-FAAB reacted with a MDA-Schiff
base indicated that the FAAB moiety is transferred to
the nitrogen of the MDA-Schiff base to form the MDHDC

Malondialdehyde-Acetaldehyde Adduct Formation
Chem. Res. Toxicol., Vol. 14, No. 7, 2001 827
F igu r e 6. 1H NMR spectra for the reaction of hexyl-FAAB and N-ethyl-aminopropenal at pD 7.4. The signals correspond to the
compounds as follows: hexyl-FAAB (A), N-ethylaminopropenal (F), hexylamine (B), and ethyl-MDHDC (G).
adduct, the applicability of this mechanism to protein
adducts was investigated. When FAAB-BSA was treated
with MDA, only minor differences in the levels of the
MDHDC adduct was observed before and after dialysis.
However, treatment of FAAB-BSA with the MDA-Schiff
base resulted in a marked reduction of the levels of the
MDHDC adduct following dialysis (Figure 7A). Similarly,
when the MAA-BSA (60% FAAB and 40% MDHDC)
adduct was treated with MDA, dialysis did not alter the
levels of the MDHDC product, but when treated with
MDA-Schiff base we again observed a marked decrease
in the MDHDC adduct in the retentate following dialysis
(Figure 7B). Overall, these results indicate for MDA to
convert the FAAB-BSA to the MDHDC adduct, the MDA
must first form a Schiff base derivative with an amino
group of the protein. Formation of this Schiff base allows
for the transfer of the FAAB moiety to nitrogen of the
Schiff base to form the MDHDC adduct which in this case
is attached to the protein and hence resistant to dialysis.
On the other hand, when a small, dialyzable Schiff base
of MDA reacts with FAAB-BSA, the FAAB moiety is
directly transferred to the small Schiff base, forming a
small, dialyzable MDHDC product.
For m a tion of MDHDC Ad d u cts on Liver P r otein s
fr om Eth a n ol-Fed Ra ts Followin g Tr ea tm en t with
MDA. Our previous studies have shown the presence of
MAA adducts on hepatic proteins from ethanol-fed rats
(5). Later studies showed that the antibody used in the
immunoassay to detect these MAA adducts specifically
recognized only MDHDC epitopes (11); therefore, whether
FAAB-protein adducts are present in the liver during
ethanol consumption is still unknown. In view of these
previous findings, the following experiments were de-
signed to test the effect of in vitro treatment of liver
homogenates obtained from ethanol-fed and control rats
with MDA on the formation of MDHDC-protein adducts,
and these results are shown in Figure 8. In confirmation
of our previous study (5), elevated levels of MDHDC-
protein adducts were detected in liver proteins from the
ethanol-fed rats, and comparably very low levels (∼10%
of ethanol-fed) were detected in the controls (p < 0.01).
In vitro incubation of liver homogenates from the ethanol-
fed animals with MDA increased MDHDC levels nearly
2-fold (p < 0.05) and also elevated the low levels of this
adduct in the controls (p < 0.02). In addition, the specific
inhibitor of antibody binding, hexyl-MDHDC, reduced the
antibody response in all cases, indicating that the major-
ity of the antibody response to liver proteins was at-
tributable to the MDHDC adduct (Figure 8).
Rea ction of Hexyl-F AAB w ith MDA-BSA Sch iff
Ba se. To further establish that the transfer of the FAAB
moiety to the nitrogen of a MDA-Schiff base is an
essential step in the formation of the MDHDC adduct,
hexyl-FAAB was incubated with BSA which had been
previously treated with MDA to generate Schiff bases on
the protein. As shown in Table 3, treatment of this
modified BSA with hexyl-FAAB resulted in significant
formation of the MDHDC-BSA adduct, indicating trans-
fer of the FAAB moiety from the small organic analogue
to the protein. It should be noted that some MDHDC-
BSA also formed when unmodified BSA was treated with
hexyl-FAAB (Table 3).
Since the FAAB adduct can dissociate during in vitro
incubation, further experiments were conducted to mini-
mize the effects of homogenization and subsequent
sample handling on the breakdown of FAAB. Therefore,
MDA was directly perfused into the livers prior to liver
homogenization and sample processing. In this case,
MDA treatment resulted in a marked increase in MDH-
DC adducts in livers from the ethanol-fed rats (p < 0.01)
and also increased the low level of the adduct in the
controls (p < 0.01)(Figure 9). Again, hexyl-MDHDC

828 Chem. Res. Toxicol., Vol. 14, No. 7, 2001
Tuma et al.
F igu r e 8. Effect of MDA on the formation of MDHDC epitopes
on liver proteins obtained from ethanol-fed rats. Liver homo-
genates from pair-fed (open bars) and ethanol-fed (closed bars)
rats were incubated in phosphate buffer (pH 7.4) at 37 °C for
24 h in the presence and absence of 1 mM MDA. An anti-MAA
antibody, specific for the recognition of MDHDC epitopes, was
used in an ELISA to detect the relative formation of MDHDC
epitopes on liver proteins. ELISAs were conducted in the
absence and presence of hexyl-MDHDC, a specific inhibitor of
anti-MAA antibody binding. Values are expressed as means (
SE for four pairs of animals.
F igu r e 7. Reaction of FAAB-BSA or MAA-BSA with a soluble
Schiff base derivative of MDA. (A) FAAB-BSA and (B) MAA-
BSA (comprised of both FAAB and MDHDC adducts) at a
concentration of 1 mg/mL BSA were allowed to react with
various concentrations of MDA or a soluble Schiff base deriva-
tive of MDA (N-ethyl aminopropenal) at pH 7.4 and 37 °C for
24 h. Fluorescent measurements (indicator of MDHDC adducts)
were conducted before (open bars) and after 24-h dialysis (closed
bars). The data are expressed as nanomoles per milligram of
BSA and represent the averages of two determinations.
Ta ble 3. Effects of In cu ba tion of Hexyl-F AAB w ith
MDA-BSA Sch iff Ba se on th e F or m a tion of MDHDC-BSA
MDHDC-BSAb
F igu r e 9. Effects of liver perfusion with a perfusate containing
MDA on the formation of MDHDC epitopes on liver proteins.
Prior to liver perfusion, two hepatic lobes were tied-off and
removed, and ELISAs were conducted on homogenates from
pair-fed (open bars) and ethanol-fed (closed bars) rats to obtain
baseline values of MDHDC-protein epitopes (represented on
figure as no additions). The remaining liver was perfused via
the portal vein for 10 min with a perfusate containing 1 mM
MDA. Liver homogenates were then prepared and incubated
in phosphate buffer (pH 7.4) at 37 °C for 3 days in the presence
of 1 mM MDA. ELISAs were performed to detect MDHDC
epitopes on liver proteins as described in Figure 7. Values are
expressed as means ( SE for 3 pairs of animals.
1-day
2-day
reactions conditionsa
incubation
incubation
BSA-MDA Schiff base
control (no additions)
+ MDA (1 mM)
0.93
1.33
1.33
1.70
+ MDA (5 mM)
2.19
3.12
+ hexyl-FAAB (1 mM)
+ hexyl-FAAB (5 mM)
unmodified BSA
26.30
59.20
50.60
76.40
+ hexyl-FAAB (1 mM)
9.35
13.85
a
MDA Schiff bases of BSA, prepared as described in the
Experimental Procedures, or unmodified BSA at 1 mg/mL were
incubated in phosphate buffer (pH 7.4) at 37 °C for 1 or 2 days in
the presence of MDA or hexyl-FAAB. Following 24-h dialysis at 4
°C, the level of MDHDC-BSA was determined. The data are
expressed as nanomoles per milligram of BSA and represent the
averages of two determinations.
Discu ssion
MAA adducts, resulting from the reaction of proteins
with acetaldehyde and MDA, have been detected in livers
of ethanol-fed animals (5), and circulating antibodies
against MAA adducts have been observed in ethanol-fed
rats (6) and in patients with alcoholic liver disease (7).
MAA adducts have been shown to be very immunogenic
(8) and to possess pro-inflammatory and profibrogenic
properties (9, 10). Our previous studies demonstrated
that MAA adducts are comprised of two major products
(11, 12), the FAAB adduct and the MDHDC adduct
b
effectively inhibited antibody binding (Figure 9). These
data show that in vitro treatment of livers from ethanol-
fed rats with MDA results in considerable formation of
MDHDC-protein adducts, indicating the likely in vivo
presence of FAAB adducts in livers of ethanol-fed ani-
mals.

Malondialdehyde-Acetaldehyde Adduct Formation
Chem. Res. Toxicol., Vol. 14, No. 7, 2001 829
F igu r e 10. Proposed reaction scheme for the formation of MDHDC-protein adducts.
(Figure 1). It should also be pointed out that in addition
to proteins MDHDC and FAAB-like adducts have also
been shown to form with DNA (24, 25). The purpose of
the present study was to clarify the chemistry of MAA
adduct formation, especially focusing on whether the
FAAB adduct serves as a precursor for the generation of
the MDHDC adduct, and to gain information concerning
the relevance of this reaction in vivo.
Based on the present findings, a reaction scheme
describing the formation of the MDHDC-protein adduct
is proposed and illustrated in Figure 10. Two initial
reaction steps appear to be a prerequisite to MDHDC
generation. One step involves the reaction of one molecule
of MDA and one of acetaldehyde with an amino group of
a protein to form the FAAB adduct. The other step
involves the reaction of MDA with another amino group
to form a MDA Schiff base (MDA-enamine). Reaction
of these two intermediates results in the formation of the
MDHDC adduct.
Several lines of experimental evidence support this
proposed reaction scheme. Previous studies showed that,
during the reaction of proteins with MDA and acetalde-
hyde, FAAB adduct formation precedes the formation of
the MDHDC adduct (11, 12), an observation consistent
with a precursor-product relationship. Furthermore,
MDA-Schiff base formation is a prominent product when
MDA reacts with proteins (17, 18). Combined together,
these studies clearly establish the presence of these two
proposed intermediates prior to the generation of the
MDHDC adduct.
The results of the present study support the occurrence
of the key step of the reaction scheme, which is the
apparent “transfer” of the FAAB moiety to the nitrogen
of the MDA-enamine. First of all, addition of MDA to a
preformed FAAB-protein adduct resulted in a significant
conversion of this adduct to the MDHDC-protein adduct.
However, in the case of the small, organic FAAB analogue
(i.e., hexyl-FAAB), the addition of the MDA-enamine
(but not free MDA) was necessary to form the MDHDC
adduct. The most likely explanation of these differences
in reactivity with free MDA is that in the case of hexyl-
FAAB, there is a lack of availability of free amino groups
to form MDA-enamines; whereas, with proteins an
ample number of amino groups are available for MDA-
enamine formation.
mechanisms require 2 mol of MDA and 1 mol each of
acetaldehyde and a primary amine, but they differ with
respect to the intermediate species. However, none of the
proposed mechanisms invoke the FAAB adduct as an
intermediate. In fact, it is surprising, at first consider-
ation, that the FAAB adduct would serve as a precursor
to the MDHDC adduct because in FAAB the acetaldehyde
and amine are connected; whereas, in MDHDC the
acetaldehyde is attached to both molecules of MDA. Thus,
either the FAAB adduct dissociates prior to the formation
of MDHDC or the FAAB and MDA-enamine react in an
unusual concerted reaction.
If the dissociation pathway is at play, there are two
probable mechanisms as shown in Figure 11. FAAB most
likely initially dissociates into free amine and intermedi-
ate I (although this reactive intermediate was not
observed in the NMR spectra), and this intermediate can
follow two paths of reaction to generate the MDHDC
adduct (Figure 11). Studies by Nair et al. (26) support
path A as a possible mechanism for MDHDC formation,
whereas several groups have indicated that path B may
be the major mechanism (23, 26, 27). Of the two mech-
anisms shown, our evidence supports path A. In the
hexyl-FAAB decomposition experiments (Figures 4 and
5), at 24 h there was very little MDHDC produced in the
absence or presence of MDA. However, when hexyl-FAAB
was allowed to react with MDA-enamine, MDHDC was
a significant product at 24 h (Figure 6). Thus the second
mole of MDA was from the MDA-enamine (path A) and
not from free MDA (path B). Further experiments with
proteins also support path A and the participation of the
MDA-enamine. When a small, organic Schiff base of
MDA (N-ethylaminopropenal) was added to BSA-FAAB,
MDHDC adduct did not form on the protein but instead
a small, dialyzable MDHDC adduct was generated.
Presumably, this MDHDC was formed on the nitrogen
of the MDA-enamine. Conversely, when BSA was al-
lowed to form Schiff bases with MDA, treatment of this
derivatized BSA with hexyl-FAAB resulted in the MDH-
DC adduct being present on the protein. Overall, these
data indicate the necessity of both FAAB and MDA-
enamine for the formation of the MDHDC adduct on
BSA. Further studies are currently in progress to further
elucidate the mechanistic details of the transfer of the
FAAB moiety to the nitrogen of the MDA-enamine.
As expected, the levels of the FAAB adduct decreased
when preformed BSA-FAAB (Figure 2B) and PLL-
FAAB (Figure 3B) were treated with MDA during forma-
tion of the MDHDC adduct. However, decreases in FAAB
exceeded the formation of the MDHDC product, indicat-
ing that the transfer of the FAAB moiety to the MDA-
enamine of proteins was not quantitative. Although the
When considering the mechanism of formation of the
MDHDC-protein adduct, it is necessary to refer to
studies of the classical Hantzsch reaction because the
MDHDC adduct is an example of an Hantzsch adduct.
The Hantzsch reaction has been studied by numerous
groups and there are a number of plausible mechanisms
that have been proposed (23, 26, 27). All of the proposed

830 Chem. Res. Toxicol., Vol. 14, No. 7, 2001
Tuma et al.
F igu r e 11. Reaction schemes describing the conversion of the FAAB adduct to the MDHDC adduct.
conversion of the FAAB adduct by treatment with a
MDA-enamine to the MDHDC product essentially went
to completion when small organic analogues were used,
the transfer of FAAB to the MDA-enamine appears to
be much less efficient for proteins. Apparently with
proteins some of the FAAB dissociates and perhaps
decomposes before it can react with the MDA-enamine.
This dissociation, coupled with the observation that
FAAB can decompose and reform the aldehydes, could
also explain why low levels of MDHDC adducts are
formed when preformed FAAB adducts are incubated
alone at 37 °C or when incubated with acetaldehyde.
Related to this is the observation that decreases in FAAB
adduct were less pronounced in the presence of MDA
compared to the acetaldehyde-treated or untreated
samples. Either MDA can stabilize the FAAB adduct or,
more likely, some FAAB adduct can reform after dis-
sociation in the presence of excess MDA.
transfer reaction and that the position of the FAAB
adduct relative to the MDA-enamine in the protein is
an important determinant for the efficient transfer of the
FAAB moiety to the MDA-enamine and subsequent
formation of the MDHDC product. This would imply that
an important factor in governing the formation of the
MDHDC adduct in a given protein would be the ability
of the protein to not only form the necessary intermedi-
ates (i.e., FAAB and MDA-enamine), but to form them
in the appropriate relative positions of close proximity
to each other to allow for the efficient transfer of FAAB
to the MDA-enamine. This consideration would explain
our previous findings which showed a wide variation in
MDHDC adduct formation among various proteins tested
and further showed that the extent of MDHDC adduct
formation for a protein was not dependent simply on the
number of amino groups (11). Therefore, proteins which
are capable of generating the FAAB and MDA-enamine
intermediates in the proper location on the peptide chain
would be selective targets of MDHDC adduct formation.
This factor could potentially be helpful in identifying
protein targets for MDHDC adduct formation in vivo
during ethanol consumption.
An interesting and important feature concerning the
transfer of the FAAB moiety to the MDA-enamine was
noted when the extent of this transfer reaction was
compared in proteins with different ratios of MAA
adducts. In the case of MAA adducts enriched in the
FAAB product (99%), addition of MDA to this adduct
caused a significant formation of the MDHDC adduct;
whereas, MAA adducts, consisting of a 60 and 40%
mixture of FAAB and MDHDC adducts, respectively, did
not generate more MDHDC adducts upon addition of
MDA. Apparently, when preformed MDHDC adducts are
present, available reaction sites are already occupied by
the MDHDC; therefore, these sites are unavailable to
form the MDA-enamine and accept transfer of the FAAB
moiety. If the number of amino groups, occupied by either
the FAAB or MDHDC adduct in the starting material
(MAA adducts composed of a 60:40 ratio of FAAB to
MDHDC), is calculated, about nine amino groups are
occupied by these adducts, leaving about 50 amino groups
still available to potentially react. Therefore, the failure
of the FAAB transfer reaction in this case cannot be
merely due to the lack of available free amino groups.
Instead, it appears that there is some selectivity in this
In the final set of experiments, information was
obtained concerning MAA adduct formation in the in vivo
situation during ethanol consumption in the rat. Our
previous studies demonstrated the presence of MDHDC-
protein adducts in livers from ethanol-fed rats (5, 11);
however, to date, we have been unable to obtain an
appropriate antibody that recognizes the FAAB adduct.
Therefore, the formation of the FAAB adduct in the liver
during ethanol consumption could not be ascertained.
Our current results showed that in vitro incubation of
liver homogenates from ethanol-fed rats with MDA
markedly increased the levels of MDHDC-protein ad-
ducts, indicating indirectly that FAAB adducts are likely
present in livers from ethanol-fed animals. Furthermore,
these data suggest that MDHDC-protein adduct forma-
tion in vivo may very well proceed by the same reaction
scheme as shown in Figure 10. Interestingly, in the pair-
fed controls, a low signal indicating the presence of

Malondialdehyde-Acetaldehyde Adduct Formation
Chem. Res. Toxicol., Vol. 14, No. 7, 2001 831
(8) Thiele, G. M., Tuma, D. J ., Willis, M. S., Miller, J . A., McDonald,
T. L., Sorrell, M. F., and Klassen, L. W. (1998) Soluble proteins
modified with acetaldehyde and malondialdehyde are immuno-
genic in the absence of adjuvant. Alcohol: Clin. Exp. Res. 22,
1731-1739.
(9) Kharbanda, K. K., Todero, S. L., Sorrell, M. F., and Tuma, D. J .
(1998) MAA adducts increase chemokine production and the
expression of cell adhesion molecule, ICAM-1, in hepatic stellate
cells. Hepatology (Suppl.) 28, 311A (Abstr.).
MDHDC adducts was observed. The ability of MDA
treatment to further elevate the levels of MDHDC
adducts was also observed, indicating the presence of
FAAB adducts in control livers as well. These data
suggest that both types of MAA adducts are present in
the livers of controls, but at very low levels compared to
the ethanol-fed animals.
In conclusion, our present studies show that MDHDC-
protein adduct formation occurs via the reaction scheme
illustrated in Figure 10 and further suggest that MDHDC
adduct formation in ethanol-fed rats may also proceed
by such a reaction pathway. Extrapolation of these
findings, that are relevant to the role of MAA adducts in
the pathogenesis of alcoholic liver injury, allows us to
propose a series of events that could take place during
ethanol consumption that would lead to liver injury. In
the early stages of ethanol consumption, FAAB adduct
formation would be favored because acetaldehyde levels
would likely exceed those of MDA. This situation would
result in a pool of FAAB adducts in the liver. With more
prolonged ethanol intake, which has been shown to
induce oxidative stress (3, 4, 28, 29) and elevate sub-
stantially the levels of MDA (30, 31), increased MDA
would be available to generate Schiff base adducts with
proteins, and subsequently these Schiff bases would be
available to accept the transfer of the FAAB moiety and
form MDHDC adducts. Elevated levels of MDHDC-
protein adducts, which have been shown to be very
immunogenic (8) and to possess pro-inflammatory3 and
profibrogenic3 properties (9, 10), could by virtue of these
potential toxic effects contribute to the pathogenesis of
alcoholic liver injury.
(10) Duryee, M. J ., Klassen, L. W., Willis, M. S., Sorrell, M. F., Tuma,
D. J ., and Thiele, G. M. (1999) Malondialdehyde-acetaldehyde
(MAA) modified proteins induce pro-inflammatory and pro-fibrotic
responses by liver endothelial cells. Hepatology (Suppl.) 30, 414A
(Abstr.).
(11) Xu, D. S., Thiele, G. M., Kearley, M. L., Haugen, M. D., Klassen,
L. W., Sorrell, M. F., and Tuma, D. J . (1997) Epitope character-
ization of malondialdehyde-acetaldehyde adducts using an enzyme-
linked immunosorbent assay. Chem. Res. Toxicol. 10, 978-986.
(12) Kearley, M. L., Patel, A., Chien, J ., and Tuma, D. J . (1999)
Observation of a new nonfluorescent malondialdehyde-acetalde-
hyde-protein adduct by 13C NMR spectroscopy. Chem. Res.
Toxicol. 12, 100-105.
(13) Miwa, G. T., Levin, W., Thomas, P. E., and Lu, A. Y. H. (1978)
The direct oxidation of ethanol by
a catalase- and alcohol
dehydrogenase-free reconstituted system containing cytochrome
P-450. Arch. Biochem. Biophys. 30, 464-475.
(14) Grabowski, E. J . J ., and Autrey, R. L. (1969) Oxygenated dienes
and the syntheses of methylenedioxybiphenyl derivatives. Tet-
rahedron 25, 4315-4330.
(15) Makin, S. M., Ishmael, A. A., Yastrebov, V. V., and Petrov, K. I.
(1971) Synthesis and investigation of â-alkyl and â-arylami-
noacrylaldehydes. J . Org. Chem. USSR (Eng. Transl.) 7, 2201-
2204.
(16) Nair, V., Vietti, D. E., and Cooper, C. S. (1981) Degenerative
chemistry of malondialdehyde. Structure, stereochemistry, and
kinetics of formation of enaminals from reaction with amino acids.
J . Am. Chem. Soc. 103, 3030-3036.
(17) Kikugawa, K., and Sugimura, Y. (1986) Properties of conjugated
Schiff bases of malondialdehyde. Chem. Pharm. Bull. 34, 1794-
1800.
(18) Beppu, M., Fukata, Y., and Kikugawa, K. (1988) Interaction of
malondialdehyde-modified bovine serum albumin and mouse
peritoneal macrophages. Chem. Pharm. Bull. 36, 4519-4526.
(19) Hoffmann, T., Meyer, R. J ., Sorrell, M. F., and Tuma, D. J . (1993)
Reaction of acetaldehyde with proteins: Formation of stable
fluorescent adducts. Alcohol: Clin. Exp. Res. 17, 69-74.
Ack n ow led gm en t. We are grateful to Tomas Hoff-
mann for his valuable technical assistance and to Mary
Barak-Bernhagen and Crystal Miller for their help in
preparation of this manuscript. This research was sup-
ported by grants AA04961 and AA11627 from the Na-
tional Institute of Alcohol Abuse and Alcoholism and by
the Department of Veterans Affairs Alcohol Research
Center.
(20) Lieber, C. S., and DeCarli, L. M. (1986) The feeding of ethanol in
liquid diets: 1986 update. Alcohol: Clin. Exp. Res. 10, 550-553.
(21) Casey, C. A., Kragskow, S. L., Sorrell, M. F., and Tuma, D. J .
(1987) Chronic ethanol administration impairs the binding and
endocytosis of asialoorosomucoid in isolated hepatocytes. J . Biol.
Chem. 262, 2704-2710.
(22) Casey, C. A., Volentine, G. D., J ankovich, C. J ., Kragskow, S. L.,
and Tuma, D. J . (1990) Effect of chronic ethanol administration
on the uptake and degradation of asialoglycoproteins by the
perfused rat liver. Biochem. Pharmacol. 40, 1117-1123.
Refer en ces
(1) Sorrell, M. F., and Tuma, D. J . (1985) Hypothesis: Alcoholic liver
injury and the covalent binding of acetaldehyde. Alcohol: Clin.
Exp. Res. 9, 306-309.
(2) Nicholls, R., de J ersey, J ., Worrall, S., and Wilce, P. (1992)
Modification of proteins and other biological molecules by acetal-
dehyde: Adduct structure and functional significance. Int. J .
Biochem. 24, 1899-1906.
(3) Cederbaum, A. I. (1989) Role of lipid peroxidation and oxidative
stress in alcohol toxicity. Free Radical Biol. Med. 7, 537-539.
(4) Ishii, H., Kurose, I., and Kato, S. (1997) Pathogenesis of alcoholic
liver disease with particular emphasis on oxidative stress. J .
Gastroenterol. Hepatol. 12 (Suppl.), S272-S282.
(5) Tuma, D. J ., Thiele, G. M., Xu, D. S., Klassen, L. W., and Sorrell,
M. F. (1996) Acetaldehyde and malondialdehyde react together
to generate distinct protein adducts in the liver during long-term
ethanol administration. Hepatology 23, 872-880.
(6) Xu, D. S., Thiele, G. M., Beckenhauer, J . L., Klassen, L. W.,
Sorrell, M. F., and Tuma, D. J . (1998) Detection of circulating
antibodies to malondialdehyde-acetaldehyde adducts in ethanol-
fed rats. Gastroenterology 115, 686-692.
(23) Gomez-Sanchez, A., Hermosin, I., Lassaletta, J .-M., and Maya,
I. (1993) Cleavage and oligomerization of malondialdehyde.
Tetrahedron 49, 1237-1250.
(24) Le Curieux, F., Pluskota, D., Munter, T., Sjoholm, R., and
Kronberg, L. (1998) Formation of a fluorescent adduct in the
reaction of 2′-deoxyadenosine with a malondialdehyde-acetalde-
hyde condensation product. Chem. Res. Toxicol. 11, 989-994.
(25) Le Curieux, F., Pluskota, D., Munter, T., Sjoholm, R., and
Kronberg, L. (2000) Identification of fluorescent 2′-deoxyadenosine
adducts formed in reactions of conjugates of malondialdehyde and
acetaldehyde and of malondialdehyde and formaldehyde. Chem.
Res. Toxicol. 13, 1228-1234.
(26) Nair, V., Offerman, R. J ., Turner, G. A., Pryor, A. N., and
Baenziger, N. C. (1988) Fluorescent 1,4-dihydropyridines: the
malondialdehyde connection. Tetrahedron 44, 2793-2803.
(27) Slatter, D. A., Murry, M., and Bailey, A. J . (1998) Formation of a
dihydropyridine derivative as a potential cross-link derived from
malondialdehyde in physiological systems. FEBS Lett. 421, 180-
184.
(28) Nordmann, R., Ribiere, C., and Rouach, H. (1992) Implications
of free radical mechanisms in ethanol-induced cellular injury. Free
Radical Biol. Med. 12, 219-240.
(7) Rolla, R., Vay, D., Mottaran, E., Parodi, M., Traverso, N., Arico,
S., Sartori, M., Bellomo, G., Klassen, L. W., Thiele, G. M., Tuma,
D. J ., and Albano, E. (2000) Detection of circulating antibodies
against malondialdehyde-acetaldehyde adducts in patients with
alcohol-induced liver disease. Hepatology 31, 878-884.
(29) Lieber, C. S., Leo, M. A., Aleynik, S. I., Aleynik, M. K., and
DeCarli, L. M. (1997) Polyenylphosphatidylcholine decreases
3
K.K.K., G.M.T., and D.J .T., unpublished observations.

832 Chem. Res. Toxicol., Vol. 14, No. 7, 2001
Tuma et al.
alcohol-induced oxidative stress in the baboon. Alcohol: Clin. Exp.
Res. 21, 375-379.
(30) Tsukamoto, H., Horne, W., Kaminura, S., Niemela, O., Parkila,
S., Yla-Herttula, S., and Brittenham, G. M. (1995) Experimental
liver cirrhosis induced by alcohol and iron. J . Clin. Invest. 96,
620-630.
(31) Chen, J ., Petersen, D. R., Schenker, S., and Henderson, G. I.
(2000) Formation of malondialdehyde adducts in livers of rats
exposed to ethanol: Role in ethanol-mediated inhibition of
cytochrome C oxidase. Alcohol: Clin. Exp. Res. 24, 544-552.
TX000222A