acetaldehyde and acrolein, is their reaction with the cysteine ligands of zinc sites in proteins and concomitant zinc release.. Since minute changes in the availability of cellular zinc
Trang 1oxidative stress ⁄lipid peroxidation to cellular functions
of zinc
Qiang Hao and Wolfgang Maret
Departments of Preventive Medicine & Community Health and Anesthesiology, The University of Texas Medical Branch, Galveston, TX, USA
The aldehyde group is the most reactive among the
functional groups of biomolecules It is involved in
Schiff base formation in the chemistry of pyridoxal
phosphate-catalyzed reactions, and in vision
photo-receptors, where retinal reacts with the e-amino group
of a specific lysine in rhodopsin There are many
sources of endogenous aldehydes For instance,
glycer-aldehyde 3-phosphate is an intermediate in glycolysis
Thiohemiacetal⁄ thioester intermediates between
glycer-aldehyde 3-phosphate and the sulfhydryl group of the
active site cysteine are formed during turnover of
glyceraldehyde 3-phosphate dehydrogenase,
demon-strating that aldehydes also react with the sulfhydryl
group of cysteine Several enzymes control the levels
of aldehydes by oxidation or reduction, thus avoiding
unspecific reactions of endogenous aldehydes and detoxifying xenobiotic aldehydes In many degenerat-ive diseases, the concentrations of aldehydes increase, and their reactivity becomes a liability In diabetes, for example, prolonged elevation of blood glucose, an aldose, leads to nonenzymatic glycations such as the addition of glucose to the a-amino groups of the b-chains of hemoglobin [1] In yet other glycation reac-tions, a-hydroxy-aldehydes or oxy-aldehydes formed from ketone bodies give rise to advanced glycation end-products [2] Concentrations of aldehydes also increase with age and in diseases that are accompan-ied by oxidative stress Oxidative stress causes lipid peroxidation and formation of aldehydes such as malon(di)aldehyde, 4-hydroxynonenal (4-HNE), and
Keywords
acetaldehyde; acrolein; metallothionein;
oxidative stress; zinc
Correspondence
W Maret, Division of Human Nutrition,
Preventive Medicine and Community
Health, The University of Texas Medical
Branch, 700 Harborside Drive, Galveston,
TX 77555, USA
Fax: +1 409 772 6287
Tel: +1 409 772 4661
E-mail: womaret@utmb.edu
(Received 2 May 2006, revised 14 July
2006, accepted 20 July 2006)
doi:10.1111/j.1742-4658.2006.05428.x
Oxidative stress, lipid peroxidation, hyperglycemia-induced glycations and environmental exposures increase the cellular concentrations of aldehydes
A novel aspect of the molecular actions of aldehydes, e.g acetaldehyde and acrolein, is their reaction with the cysteine ligands of zinc sites in proteins and concomitant zinc release Stoichiometric amounts of acrolein release zinc from zinc–thiolate coordination sites in proteins such as metallothion-ein and alcohol dehydrogenase Aldehydes also release zinc intracellularly
in cultured human hepatoma (HepG2) cells and interfere with zinc-depend-ent signaling processes such as gene expression and phosphorylation Thus both acetaldehyde and acrolein induce the expression of metallothionein and modulate protein tyrosine phosphatase activity in a zinc-dependent way Since minute changes in the availability of cellular zinc have potent effects, zinc release is a mechanism of amplification that may account for many of the biological effects of aldehydes The zinc-releasing activity of aldehydes establishes relationships among cellular zinc, the functions of endogenous and xenobiotic aldehydes, and redox stress, with implications for pathobiochemical and toxicologic mechanisms
Abbreviations
ADH, alcohol dehydrogenase; DNP, 2,4-dinitrophenyl; DTNB, 5,5¢-dithiobis-2-nitrobenzoic acid; 4-HNE, 4-hydroxynonenal; MCA,
(7-methoxycoumarin-4-yl)-acetyl; 4-MP, 4-methylpyrazole hydrochloride; MRE, metal response element; MT, metallothionein; MT2,
metallothionein isoform 2; MTF-1, metal response element-binding transcription factor-1; PAR, 4-(2-pyridylazo)-resorcinol; PTP, protein tyrosine phosphatase; TCEP, tris(2-carboxyethyl)-phosphine; TPEN, N,N,N¢,N¢-tetrakis(2-pyridylmethyl)-ethylenediamine.
Trang 2acrolein [3,4] Aldehydes from the environment can
exacerbate the burden of exposure Endogenous
alde-hydes that increase during these and other episodes of
exposure include: formaldehyde, used as a preservative
but also found in cigarette smoke and burning
veget-ation; acrolein, found in cigarette smoke, herbicides,
and acrylics, and produced during fossil fuel
combus-tion, during petrochemical processing, and when
over-heating cooking oil; and methylglyoxal, a metabolite
formed during acetone detoxification [5,6]
Endogen-ously generated or inhaled aldehydes are involved in
cardiovascular disease, atherosclerosis, vascular
com-plications of diabetes [7] and respiratory diseases [8]
Another prominent example is acetaldehyde, the
meta-bolic product of ethanol from alcoholic beverages
Excess acetaldehyde can accumulate to levels of a few
hundred micromoles per liter [9], especially in
indi-viduals with a slow-metabolizing variant of
mito-chondrial aldehyde dehydrogenase Accumulation of
acetaldehyde has been discussed in the pathology of
alcohol-induced tissue injury [10]
Cysteine is now recognized as a ligand in a large
number of zinc coordination sites The cysteine ligands
are remarkably reactive towards oxidizing agents and
nucleophiles, both of which release zinc [11] Even
minute amounts of released zinc are potent effectors of
cellular metabolism and signaling [12,13] This study
addresses the reactivity of aldehydes with cysteine
lig-ands of zinc in proteins Moreover, it demonstrates
that aldehydes release zinc from isolated proteins and
in cultured cells and that the released zinc affects
phos-phorylation signaling and gene expression
Results
Aldehydes release zinc from zinc-binding proteins
The effect of aldehydes on the zinc-binding capacity of
zinc proteins was assayed by employing
spectropho-tometry and the chromophoric indicator
4-(2-pyridyl-azo)-resorcinol (PAR) for zinc ions Acrolein at
concentrations as low as 10 lm releases zinc from
metallothionein (MT) (Fig 1) In this experiment, the
concentration of zinc MT isoform 2 (MT2) is 0.5 lm,
corresponding to 10 lm in thiols, as there are 20
cys-teines in MT Thus stoichiometric amounts of acrolein
with regard to the thiols in MT release zinc The
reac-tion continues for 20 h until all seven zinc ions from
MT2 are released (Fig 1) Zinc release is based on the
reaction of MT with 10 lm ebelsen, which releases all
seven zinc ions from MT within 20 min [14] At a
con-centration of 1 mm acrolein, all seven zinc ions are
released within 6 h
The zinc-releasing activity of other aldehydes was determined with the same assay (Fig 2) Because some aldehydes are much less reactive than acrolein, the measurements were performed at aldehyde concentra-tions of 1 mm (Fig 2) Among the aldehydes tested, acrolein is the most reactive aldehyde, followed by butyraldehyde, propionaldehyde, acetaldehyde, benzal-dehyde, and glyceraldehyde 4-HNE releases only 5%
of zinc from MT, while malondialdehyde releases only 3% At physiologic pH, malondialdehyde exists as the enolate, which is much less reactive than its enol form
at acidic pH (b-hydroxyacrolein)
The following investigations focus on the effects of acetaldehyde and acrolein because of the relevance of these aldehydes for the biological effects of ingested ethanol and lipid peroxidation, respectively
In order to explore whether or not acetaldehyde releases zinc from other zinc–sulfur coordination envi-ronments, its reaction with the zinc enzyme yeast alco-hol dehydrogenase (ADH) in the absence of coenzyme was followed with the PAR assay Acetaldehyde (1 mm) also releases zinc from this enzyme (Fig 3A, line 2) Acrolein (1 mm) releases significantly more zinc than acetaldehyde (Fig 3A, line 3) The activity of the enzyme is affected differently by the two aldehydes
Fig 1 Acrolein releases zinc from metallothionein (MT) The kinetics of zinc transfer from MT isoform 2 (MT2) (0.5 l M ) to 4-(2-pyridylazo)-resorcinol (PAR) (100 l M ) was monitored spectro-photometrically in the absence and presence of 10 l M acrolein in
20 m M Tris ⁄ HCl (pH 7.4) The reaction was recorded immediately after acrolein was added to the solution and recorded for 1200 min Zinc release is based on the reaction of MT with 10 l M ebselen, which releases all seven zinc ions from MT within 20 min [14], because evaporation of liquid during the long time period of the assay leads to a more concentrated sample, a higher absorbance reading, and hence an apparent release of more zinc than is possi-ble based on the initial concentration of 0.5 l M MT2, when calcula-ted on the basis of the extinction coefficient of PAR Line 1: control (no acrolein) Line 2: with acrolein.
Trang 3(Fig 3B) Incubation with acetaldehyde has virtually
no effect on its activity, whereas incubation with
acro-lein inhibits enzymatic activity, suggesting that
acetal-dehyde removes only the noncatalytic zinc and that
acrolein, an irreversible inhibitor [15], removes both
the noncatalytic and the catalytic zinc ions from the
enzyme
Aldehydes react with the sulfhydryl groups of
metallothionein and thionein
A thiol assay with 5,5¢-dithiobis-2-nitrobenzoic acid
(DTNB, Ellman’s reagent) was employed to explore
the reactions of MT2 with acetaldehyde (Fig 4) When
the ratio between MT2 and DTNB is 1 : 200, the
reac-tion reaches a plateau after 2 h (Fig 4, line 1), at
which point all of the 20 sulfhydryl groups in MT are
titrated with DTNB Preincubation of MT2 with
acet-aldehyde for 30 min changes the sulfhydryl reactivity
of MT2 significantly Only 67% of the thiols now
react, indicating that the remaining 33% are modified
with acetaldehyde and can no longer react with DTNB
(Fig 4, line 2) Under these conditions, 2.1 zinc ions
are released from MT The reaction of the apoprotein
thionein (1.2 lm) with DTNB (200 lm) is rapid and
complete in less than 10 min Acetaldehyde (1 mm)
quenches the reactivity of the 20 thiols in thionein, as
the absorbance does not change when DTNB is added
To determine whether or not acetaldehyde also reacts
directly with 2-nitro-5-theobenzoic acid, the product of
the reaction of DTNB with thiols, the excess of acetal-dehyde in the above reaction mixture was removed enzymatically with yeast ADH [1 unitÆmL)1 (one unit converts 1 micromole ethanol per min at pH 8.8,
25C)] and NADH (2 mm) before DTNB was added
As virtually the same absorbance reading was recor-ded, save for a small increase due to the sulfhydryls in ADH, the experiment demonstrates that acetaldehyde reacts directly with the sulfhydryl groups of MT and does not react with TNB
In order to determine whether the modification of any of the eight lysines in MT by aldehydes would
Fig 3 Aldehydes release zinc from alcohol dehydrogenase (ADH) (A) The kinetics of zinc transfer from ADH (0.5 l M , 12.8 unitsÆmL)1)
to 4-(2-pyridylazo)-resorcinol (PAR) (100 l M ) was monitored spectro-photometrically in the absence and presence of aldehydes in
20 m M Tris ⁄ HCl (pH 7.4) The ADH concentration is based on the data provided by the manufacturer The reaction was recorded for
20 min immediately after aldehydes were added to the solution Line 1: control (no aldehyde) Line 2: 1 m M acetaldehyde Line 3:
1 m M acrolein (B) Effect of aldehydes on ADH activity ADH (0.15 units) was incubated with either 1 m M acetaldehyde or 1 m M
acrolein for 20 min, the mixture was added to the buffer ⁄ substrate mix, and the reaction was followed spectrophotometrically at
340 nm d, control (no aldehyde preincubation); n, acetaldehyde;
m , acrolein.
Fig 2 Zinc-releasing activities of different aldehydes The amount
of zinc released from metallothionein isoform 2 (MT2) (0.5 l M ) by
aldehydes (1 m M ) was determined with 4-(2-pyridylazo)-resorcinol
(PAR) after 30 min Ebselen (10 l M ) was used as a positive control
because it releases all seven zinc ions from MT within 20 min Data
are presented as means ± SD of triplicate determinations.
Trang 4contribute to zinc release, the E-amino groups of
lysines in MT2 were carbamoylated with potassium
cyanate and the modified protein was assayed for zinc
release as described above Acetaldehyde releases
almost the same amount of zinc from the modified
protein (90%), clearly indicating that the reaction of
lysines in MT with aldehydes has little, if any, effect
on zinc release and that the predominant mechanism
of zinc release is the modification of the cysteine
lig-ands of zinc
Aldehydes increase the concentration of
available cellular zinc
Cultured human hepatocellular carcinoma (HepG2)
cells were used to examine whether or not aldehydes
release zinc intracellularly HepG2 cells were incubated
with acetaldehyde (1 mm) or acrolein (10 lm) for
30 min, and Zinquin ester was added to introduce a
fluorescent chelating agent into the cell for
measure-ment of intracellular zinc HepG2 cells without any
treatment have a fluorescence signal that corresponds
to 15.4% saturation of Zinquin with zinc (Fig 5A)
Treatment of cells with acrolein (10 lm) increases the
saturation to 22% Because the effect of acetaldehyde
(1 mm) on zinc saturation of Zinquin is small (17%),
albeit statistically significant, a different approach was
employed to increase cellular acetaldehyde
concentra-tions When cells were treated with 2 lm disulfiram to
inhibit aldehyde dehydrogenase and ethanol was
added, a significant release of zinc was detected, with
saturation of Zinquin reaching 22% (Fig 5B) Ethanol
alone had a small but statistically significant effect,
while disulfiram alone lowered the amount of zinc available to the probe due to its metal-chelating capa-city [16]
Fig 4 Effect of acetaldehyde on the thiol reactivity of
metallothion-ein isoform 2 (MT2) MT2 (1.2 l M ) was incubated without (line 1)
or with (line 2) acetaldehyde (1 m M ) for 30 min in 20 m M Tris ⁄ HCl
(pH 7.4), 5,5¢-dithiobis-2-nitrobenzoic acid (DTNB) was added to a
final concentration of 0.2 m M , and the absorbance at 412 nm was
recorded Line 1: control (no acetaldehyde) Line 2: 1 m M
acetalde-hyde.
Fig 5 Aldehydes increase the amount of available intracellular zinc
in HepG2 cells (A) HepG2 cells (1 · 10 6
) were treated with acetal-dehyde (1 m M ) or acrolein (10 l M ) for 30 min The cells were col-lected and labeled with Zinquin ester Fluorescence intensities were recorded with excitation and emission wavelengths of 370 and 490 nm, respectively (B) HepG2 cells (1 · 10 6 ) were treated with 2 l M disulfiram for 1 h to inhibit aldehyde dehydrogenases After addition of 5 m M ethanol to the medium and incubation for another hour, cells were collected and cellular zinc was measured
as described above Data are presented as means ± SD of triplicate determinations Fluorescence changes are insignificant when etha-nol is added to the cells Disulfiram decreases the fluorescence intensity slightly (see text) The asterisk indicates significance at
P < 0.05.
Trang 5Aldehydes induce expression of metallothionein
in HepG2 cells
A cadmium-binding assay was used to examine the
expression levels of MT in HepG2 cells after aldehyde
treatment The experiment is based on the hypothesis
that released zinc induces the expression of MT
The MT concentration in control HepG2 cells is
75.4 ± 7.6 ngÆ(g cells))1 (Fig 6) Treating the cells
with ethanol, a known inducer of MT [17], for 12 h
increases the concentration of MT to 101 ngÆ(g cells))1
To examine whether ethanol or its metabolic product
acetaldehyde induces MT, inhibitors of ADH [4-meth-ylpyrazole hydrochloride (4-MP)] and aldehyde dehy-drogenase (disulfiram) were used in conjunction with ethanol 4-MP inhibits the conversion of ethanol to acetaldehyde, lowering acetaldehyde concentrations, whereas disulfiram inhibits the conversion of acetalde-hyde to acetic acid, increasing the concentrations of acetaldehyde The concentration of MT in 4-MP⁄ etha-nol-treated cells does not change, whereas it increases
to 118 ngÆ(g cells))1 in disulfiram⁄ ethanol-treated cells (Fig 6A) Treatment of HepG2 cells with 1 mm acetal-dehyde increases the MT concentration two-fold These results clearly demonstrate that acetaldehyde and not ethanol induces MT in HepG2 cells Relatively low concentrations of acrolein (10 lm) increase MT2 expression by 35% (Fig 6B)
Aldehydes inhibit protein tyrosine phosphatase activity in HepG2 cells through modulation of intracellular zinc
To further investigate the effect of aldehydes on zinc-mediated biological processes, the effects of acetalde-hyde and acrolein on protein tyrosine phosphatase (PTP) activity were investigated The rationale for this experiment is that intracellular zinc modulates PTP activity [18] Incubation of HepG2 cells with acetalde-hyde or acrolein significantly inhibits PTP activity to 45% and 52% of the control, respectively (Fig 7) This inhibition could be caused by a reaction of
Fig 6 Aldehydes increase the expression levels of
metallothio-nein (MT) in HepG2 cells (A) Ethanol (5 m M ), 4-methylpyrazole
hydrochloride (4-MP) ⁄ ethanol (5 l M ⁄ 5 m M ), disulfiram ⁄ ethanol
(5 l M ⁄ 5 m M ) or acetaldehyde (1 m M ) were incubated with
2 · 10 6
HepG2 cells for 12 h (B) Acrolein (10 l M ) was incubated
with 2 · 10 6 HepG2 cells for 12 h Control or treated cells were
collected, washed, and homogenized MT concentrations were
determined with a cadmium-binding assay Data are presented as
means ± SD of triplicate determinations The asterisk indicates
sig-nificance at P < 0.05 No significant difference was found for
4-MP ⁄ ethanol treatment.
Fig 7 Aldehydes inhibit protein tyrosine phosphatase (PTP) activity
in HepG2 cells through a zinc-mediated mechanism Acetaldehdye (1 m M ) or acrolein (10 l M ) was incubated with 2 · 10 6 HepG2 cells for 12 h Control or treated cells were collected, washed, and homogenized PTP activity was measured with a fluorescent phos-photyrosine peptide An aliquot of the homogenized cells was incu-bated with 5 l M N,N,N¢,N¢-tetrakis(2-pyridylmethyl)-ethylenediamine (TPEN) for 30 min before measurement of PTP activity –, without TPEN; +, with TPEN Emission wavelength 395 nm, excitation wavelength 328 nm Data are presented as means ± SD of tripli-cate determinations The asterisk inditripli-cates significance at P < 0.05.
Trang 6acetaldehyde with the catalytic cysteine of PTP, zinc
inhibition of PTP, or both After addition of the
zinc-chelating agent
N,N,N¢,N¢-tetrakis(2-pyridylmethyl)-ethylenediamine (TPEN), PTP activity in both control
and aldehyde-treated cells increases, indicating that
aldehydes affect PTP activity in part through zinc
release and zinc inhibition of PTP
Discussion
Aldehydes affect zinc–sulfur (Zn–SCys)
coordination environments in proteins
Zn–SCys sites in proteins are remarkably reactive
Oxi-dation of the sulfur ligands and concomitant zinc
release establishes multiple pathways for redox control
of zinc metabolism and dynamic regulation of protein
structure and function [11] Oxidants such as
glutathi-one disulfide, nitric oxide and reducible
selenium-containing compounds release zinc from proteins with
Zn–SCyssites [19–21] Based on the above results,
alde-hydes can now be added to the growing list of agents
that affect the cellular functions of zinc A structure–
activity relationship for the limited number of
alde-hydes tested here cannot be given, as many factors
other than steric factors determine the reactivity In
aqueous solutions, aldehydes undergo side reactions
that compete with the reactivity under investigation
Examples are slow oxidation to the corresponding
acid, aldol condensation of short-chain aldehydes and
hydration of alkyl aldehydes to gem-diols [22]
There-fore, it is critical to prepare fresh stock solutions from
the anhydrous aldehyde immediately before the
experi-ment In addition, the two aldehydes discussed,
acrolein and acetaldehyde, react differently with
sulf-hydryls Acetaldehyde reacts via the aldehyde group,
whereas acrolein, an a,b-unsaturated aldehyde, forms a
Michael adduct The zinc-releasing activity of
alde-hydes has implications for toxicologic and
patho-biochemical mechanisms
Acrolein
Concentrations of cellular aldehydes increase during
environmental and nutritional exposures, as well as in
various diseases with oxidative stress that increases
lipid peroxidation Malondialdehyde, 4-HNE and
acro-lein are the major aldehyde products of lipid
peroxida-tion Acrolein is also formed from spermine and
spermidine by amine oxidases [23] In the brain of
Alz-heimer’s disease victims, the concentrations of acrolein
and 4-HNE increase 7–8-fold [24–26] For
refer-ence, basal values in hippocampus are 0.3 and
0.265 nmolÆ(mg protein))1, respectively In acute iron loading⁄ toxicosis, cytotoxic aldehydes increase through lipid peroxidation, which is initiated by Fenton chem-istry-generated free radicals [27] In diabetes, there are pathways for the increased formation of a-keto-aldehydes such as glyoxal and methylglyoxal from glyceraldehyde 3-phosphate Autoxidation of a-hydroxy-aldehydes to a-ketoa-hydroxy-aldehydes generates hydrogen per-oxide, which contributes to oxidative stress and lipid peroxidation in the disease [1]
Acrolein induces transcription of phase II genes by activating the transcription factor Nrf2 [24] Nrf2 translocates to the nucleus when released from the pro-tein Keap1, a zinc metallopropro-tein with Zn–SCys co-ordination and the sensor for electrophiles such
as aldehydes A proposed mechanism of activation involves a reaction of electrophiles with the cysteine ligands of Keap1, followed by zinc release [28] The reactions of aldehydes with MT and ADH and con-comitant zinc release provide direct experimental sup-port for such a mechanism
Acetaldehyde Under normal conditions, aldehyde dehydrogenases maintain acetaldehyde at relatively low levels, e.g below 0.2 lm for plasma acetaldehyde that is not pro-tein-bound [29] However, acetaldehyde concentrations are significantly higher when alcoholic beverages are consumed, in individuals with an inactive mitochon-drial aldehyde dehydrogenase or in alcoholic patients under treatment with disulfiram or other alcohol-sensitizing drugs In animals treated with aldehyde dehydrogenase inhibitors and ethanol, blood acetalde-hyde can reach concentrations of almost 1 mm [30,31] Acetaldehyde is discussed as a mediator of tissue injury in alcoholic liver disease and myopathies, in the etiology of cancer of the respiratory and digestive tracts, and in other diseases [10,32]
In summary, the reactivity of aldehydes with zinc proteins demonstrates that elevated levels of aldehydes affect zinc metabolism and that zinc release and ensu-ing bindensu-ing of zinc to other proteins is one aspect of the molecular actions of aldehydes that are generated during lipid peroxidation and metabolism of ethanol
Zinc signals generated by aldehydes The concentrations of ‘free’ zinc are orders of magni-tude smaller than those of total cellular zinc, which is
a few hundred micromoles per liter [33] Very small but significant changes in the availability of cellu-lar zinc have profound biological effects Thus, an
Trang 7increase from 520 to 870 pm ‘free’ zinc is characteristic
for a transition between normal and diabetic
cardio-myocytes [34] Changes from picomolar to low
nano-molar concentrations of zinc affect gene expression in
cardiomyocytes [35] Similarly, low nanomolar
concen-trations of zinc inhibit phosphorylation signaling,
metabolic enzymes, and mitochondrial respiration
[18,36,37] Because a very potent zinc signal is
gener-ated, aldehyde-induced zinc release from proteins is
significant for even relatively small increases of
alde-hyde concentrations Hence, the actions of zinc may
explain at least some of the regulatory functions of
ethanol and its metabolite acetaldehyde in cellular
signaling, where molecular mechanisms remain largely
unknown [38,39] There is a striking similarity between
the effects of acetaldehyde and those of zinc
Acetalde-hyde inhibits PTP 1B in Caco-2 cells and increases
protein tyrosine phosphorylation, much as zinc does in
other cell lines [18,40,41] Also, acetaldehyde affects
the nuclear factor-jB pathway in a way similar to zinc
or MT [42,43] Indeed, in addition to a direct
interac-tion of aldehydes with protein sulfhydryls, an indirect
action of aldehydes via binding of released zinc to
pro-tein sulfhydryls is evident from the effects of released
zinc on gene expression (Fig 6) and phosphorylation
signaling (Fig 7) Short-chain alcohols induce thionein
through an indirect mechanism [44] It is now apparent
that the induction occurs through zinc that is released
by aldehydes formed from the corresponding alcohols
during metabolism
Protective functions of zinc and MT against
ethanol toxicity
Both zinc and MT protect the liver and the heart
against the toxic effects of ethanol [45–47] The above
results suggest that a critical aspect of the protective
function of MT is the scavenging of the acetaldehyde
formed from ethanol and concomitant zinc release
Micromolar cellular concentrations of MT [48] make it
a significant source of aldehyde-released zinc Zinc
released in the cell or zinc provided by
supplementa-tion activates metal response element (MRE)-binding
transcription factor-1 (MTF-1) and transcription of
the apoprotein thionein, which also reacts with
alde-hydes Indeed, addition of a hexapeptide that contains
three of the 20 cysteines of thionein suppresses the
for-mation of protein–hydroxynonenal adducts in retinal
pigmented epithelial cells [49] Most cells have
concen-trations of thionein commensurate with those of MT
[50] Reactions of aldehydes with cellular thiols such as
thionein and glutathione will affect the cellular redox
balance and the capacity to scavenge reactive species
Thionein, with its 20 thiols, is an efficient reducing agent [20] and can serve as a cofactor for methionine sulfoxide reductase, an enzyme that protects tissue against oxidative injury [51] The reaction of acetalde-hyde with the Zn–SCys bonds in ADH and concomit-ant zinc release underscores the significance of these reactions for compromising the functions of other pro-teins with Zn–SCys sites, such as ‘zinc fingers’ 4-HNE modifies the cysteine ligands in liver ADH, leading to ubiquitinylation and proteasomal degradation [52] However, whether the released zinc is cytoprotective
or cytotoxic depends on the concentrations of released zinc, as zinc has both pro-antioxidant and pro-oxidant functions [53] If concerns for safety can be overcome [54], zinc supplementation could be an efficient way of inducing MT⁄ thionein for protection against toxic aldehydes On the other hand, nutritional or condi-tional zinc deficiency will increase cellular damage by aldehydes Zinc deficiency elicits oxidative stress [55], thus increasing lipid peroxidation and aldehyde con-centrations, releasing more zinc from proteins, and initiating a vicious cycle that will exacerbate zinc defi-ciency and increase the toxicity of aldehydes
Experimental procedures
Materials
4-HNE was obtained from Biomol (Plymouth Meeting, PA), Sephadex G-25 and G-50 from Amersham Biosciences (GE Healthcare, Piscataway, NJ), Cleland’s reagent (dithio-threitol) from Calbiochem (San Diego, CA), and Zinquin ester from Molecular Probes (Eugene, OR) All other chem-icals were from Sigma (St Louis, MO)
Reconstitution of MT2 with zinc
Commercial rabbit MT2 (Sigma) contains both cadmium and zinc To prepare zinc MT2 [56], 5 mg of MT2 was dis-solved in 1 mL of 20 mm Tris⁄ HCl (pH 7.4) containing
50 mg dithiothreitol, and incubated at 25C for 24 h After incubation, the sample was adjusted to pH 1 with HCl, and centrifuged at 10 000 g for 5 min (Eppendorf centrifuge model 5415C, Hamburg, Germany) to remove any precipi-tate The clear supernatant was then loaded onto a Sepha-dex G-25 column (1· 120 cm), which was equilibrated and eluted with 10 mm HCl Fractions containing thionein were collected and quantified based on both absorbance readings (A220¼ 48 000 m)1Æcm)1) and assay of thiols A ten-fold molar excess of zinc sulfate was added to the nitrogen gas-purged solution of thionein, and the pH value was adjusted
to 8.6 by slowly adding nitrogen gas-purged 1 m Tris base The sample was concentrated to about 2 mL by
Trang 8centrifuga-tion for 4 h at 4000 g using Centricon centrifugal filter
devices (MWCO 3000) (Millipore, Bedford, MA), loaded
onto a Sephadex G-50 column (1· 120 cm), and eluted
with 20 mm Tris⁄ HCl (pH 7.4) at a flow rate of 10 mLÆh)1
MT fractions were pooled after measuring the
concentra-tion of protein (A220¼ 159 000 m)1Æcm)1) and thiols and
determining zinc by atomic absorption spectrophotometry
(Perkin-Elmer model 5100, Wellesley, MA)
Preparation of thionein from MT2
Zinc MT2 (0.5 mg) was incubated in 1 mL of 20 mm
Tris⁄ HCl (pH 7.4) containing 0.1 m dithiothreitol overnight
at 25C The reaction mixture was adjusted to pH 2 with
HCl, and thionein was separated from excess dithiothreitol
and zinc ions by gel filtration on a Sephadex G-25 column
(1· 30 cm) equilibrated with 10 mm HCl at 25 C To
min-imize the oxidation of thionein, the elution buffer (20 mm
Tris⁄ HCl, pH 7.4) was purged with nitrogen gas Thionein
was located in the fractions by measurement of its
absorb-ance at 220 nm and by assaying its thiols with
2,2¢-dithiodi-pyridine (see below) Thionein was either used immediately
or stored at liquid nitrogen temperatures
Thiol assay
The concentration of thiols in MT was determined by
incubating the protein with 100 mgÆL)12,2¢-dithiodipyridine
[57] and taking absorbance readings (A343¼ 7600
m)1Æcm)1) with a Beckman-Coulter DU 800 UV–visible
spectrophotometer (Fullerton, CA)
PAR metal transfer assay
Metallochromic indicators provide a rapid means of
investi-gating metal–protein equilibria [58,59] PAR is such an
indicator Binding of zinc ions changes its absorbance at
500 nm Zn7-MT2 or yeast ADH (0.5 lm) and PAR
(100 lm from a 1 mm stock solution in 20 mm Tris⁄ HCl,
pH 7.4) were incubated with or without aldehydes and the
absorbance change was followed (A500¼ 65 000 m)1Æcm)1)
Aldehyde stock solutions (100 mm) were prepared
immedi-ately before use Owing to the toxicity of some aldehydes,
all stock solutions were prepared in a fume hood A stock
solution of 4-HNE was prepared from the compound
stored at ) 80 C and used immediately Malonaldehyde
tetrabutylammonium salt was used as a source of
‘malondi-aldehyde’ Evaporation of acetaldehyde during
measure-ments was minimized by sealing the cuvettes with Parafilm
A 1 mm solution of dl-glyceraldehyde (Sigma) in 20 mm
Tris⁄ HCl (pH 7.4) was found to contain 20 lm zinc Thus
the absorbance change after incubation of 1 mm
glyceralde-hyde with PAR was subtracted The experiments were
repeated at least three times Aldehydes (1 mm) were also
mixed with PAR (100 lm) in the absence of MT, and the absorbance at 500 nm was recorded With the exception of formaldehyde, none of the aldehydes affects the absorbance
of PAR The data for the reaction of MT with formalde-hyde were corrected for the absorbance changes in the absence of MT
Thiol reactivities of MT and thionein
The reactivity of thiols in MT and thionein was determined with DTNB under pseudo-first-order rate conditions The reaction between MT or thionein (1.2 lm) and DTNB (200 lm) in 20 mm Tris⁄ HCl (pH 7.4) was followed spec-trophotometrically at 412 nm (25C) The number of sulf-hydryls modified by acetaldehyde was determined by incubating MT or thionein with acetaldehyde for 30 min, removing the excess of aldehyde with 1 unitÆmL)1 of yeast ADH, 2 mm NADH and 100 mm potassium chloride, and then assaying the protein with DTNB
Modification of lysine residues in MT
Lysine residues in MT were modified according to an estab-lished protocol [60] Briefly, 1 mg of MT2 was concentrated with Centricon centrifugal microconcentrators (MWCO 3000; Millipore), and diluted with 0.5 m sodium borate buf-fer (pH 9.2) to a final concentration of 10 mgÆmL)1, and solid potassium cyanate was added to a final concentration
of 1 m The reaction mixture was incubated at 37C for
24 h Excess potassium cyanate was then removed by gel filtration on a Sephadex G-25 column (0.2· 8 cm) Protein concentrations were determined spectrophotometrically at
220 nm
Yeast ADH assay
ADH activity was determined with acetaldehyde as sub-strate The assay was performed in 0.1 m Tris⁄ HCl (pH 8.0), 0.67 mm NADH, 100 mm KCl, 10 mm 2-mercap-toethanol, 2 mm acetaldehyde and 0.0007% (w⁄ v) BSA The reaction was monitored by measuring the decrease in NADH absorbance at 340 nm after initiation of the reaction
by addition of enzyme (0.15 units) The effects of aldehydes
on ADH activity were examined by mixing ADH (0.15 units
in 5 lL) with an equal volume of either 2 mm acetaldehyde
or 2 mm acrolein and incubating for 20 min An aliquot was then added to the assay solution to initiate the reaction Aldehydes introduced into the assay in this way increase the total aldehyde concentration by less than 1%
Tissue culture
HepG2 cells (#HB-8065, American Type Culture Collec-tion, Manassas, VA) were cultured in DMEM containing
Trang 94.5 gÆL)1 glucose, supplemented with 10% (v⁄ v) FBS
(defined; Hyclone, Salt Lake City, UT), 0.12 mgÆmL)1
streptomycin sulfate, and 0.1 mgÆmL)1 gentamicin sulfate
Cells were maintained at 5% CO2 and 37C in a
humid-ified atmosphere All other cell culture products were
pur-chased from Gibco (Invitrogen, Carlsbad, CA)
Determination of available cellular zinc
HepG2 cells (1· 106
cells per well) were seeded in 12-well plates and grown for 24 h Freshly prepared acetaldehyde
and acrolein were added to the medium to final
concentra-tions of 1 mm and 10 lm, respectively, and incubated for
30 min Additionally, cells were incubated for 1 h with
tetraethylthiuram disulfide (disulfiram), an aldehyde
dehy-drogenase inhibitor, at a final concentration of 2 lm,
eth-anol was added to each well to a final concentration of
5 mm, and the cells incubated for an additional hour [18]
The fluorescence probe Zinquin ethyl ester (dissolved in
dimethyl sulfoxide) was added to the cells to a final
concen-tration of 25 lm The measurements were normalized by
measuring the total protein concentration of each sample
with a Micro-BCATM protein assay kit from Pierce
(Rock-ford, IL) The protein concentration of control cells without
disulfiram or ethanol was taken as 100% To determine the
extent of saturation of Zinquin with zinc, 1· 106
cells were incubated with the dye as described above, washed three
times with Dulbecco’s NaCl⁄ Pi, and detached in 3 mL of
NaCl⁄ Pi, and the fluorescence intensity (F) was measured
at 370 nm (excitation) and 490 nm (emission) with an
SLM-8000 spectrofluorimeter equipped with data
acquisi-tion and processing electronics from ISS (Champaign, IL)
Fluorescence intensities are the averages of three
measure-ments The working range for measurements of fluorescence
intensity was determined by adding zinc and the ionophore
pyrithione (20 lm final concentrations for both) The
meas-ured value corresponds to the maximum fluorescence
(Fmax) The minimum fluorescence (Fmin) was obtained
from a reading in the presence of the zinc-chelating agent
TPEN (100 lm) The percentage of saturation was then
calculated from [(F) Fmin)⁄ (Fmax) Fmin)]· 100 Addition
of 20 lm zinc alone increased fluorescence slightly This
fluorescence increase is quenched with cell-impermeable
EDTA, and is therefore due to zinc binding to residual,
extracellular Zinquin This fluorescence was subtracted
from Fmax
Determination of MT in HepG2 cells
The total amount of MT in HepG2 cells was determined
with a cadmium-binding assay [61] with modifications
HepG2 cells (2· 106
) were homogenized in a Potter-Elveh-jem homogenizer with at least 20 strokes
Microsco-pic inspection verified that 90% of the cells were broken
The supernatant (200 lL) obtained after centrifugation at
14 000 g (Eppendorf centrifuge model 5415C) was mixed with the same volume of a CdCl2solution (2 lgÆmL)1), and incubated at 25C for 10 min One hundred microliters of bovine hemoglobin solution (2%, w⁄ v) was added to the tubes, and the sample was mixed and heated in a boiling water bath for 2 min The samples were then placed on ice for 5 min, and centrifuged at 14 000 g for 2 min (Eppen-dorf centrifuge model 5415C); another aliquot of 100 lL of 2% hemoglobin solution was then added to the superna-tant, and heating, cooling and centrifugation were repeated Finally, a 500 lL aliquot of the supernatant was removed and diluted with 3.5 mL of 0.1 m HNO3 Cadmium concen-trations in the supernatants were determined by atomic absorption spectrophotometry (Perkin-Elmer model 5100)
MT concentrations were calculated based on an MT⁄ Cd stoichiometry of 1 : 7
PTP assay
PTP activity in HepG2 cells was determined with a tyro-sine-phosphorylated oligopeptide MCA-Gly-Asp-Ala-Glu-Tyr(PO3H2)-Ala-Ala-Lys(DNP)-Arg-NH2 (Calbiochem, La Jolla, CA) [18] In this peptide, the DNP group quenches the fluorescence of the (7-methoxycoumarin-4-yl)-acetyl (MCA) group Assays were performed at 37C in 20 mm Hepes⁄ NaOH (pH 7.5) containing 1 mm Tris-(2-carboxy-ethyl)-phosphine (Molecular Probes) and 1 lm substrate in
a total volume of 1 mL After 5 min of equilibration of substrate with buffer, the reaction was initiated by adding
an aliquot containing 10 mg of total protein from the extract of the control or aldehyde-treated cells (sample from determination of MT concentration) The reaction was quenched after 15 min by adding 10 lL of chymotryp-sin⁄ sodium orthovanadate to final concentrations of 0.05% (w⁄ v) and 0.1 mm, respectively Chymotrypsin cleaves only the peptide that is dephosphorylated by PTPs Cleavage disrupts fluorescence resonance energy transfer, thereby increasing MCA fluorescence MCA fluorescence was mon-itored at 328⁄ 395 nm, with slit widths of 1.5 nm (excita-tion) and 10 nm (emission), using an SLM-8000 spectrofluorimeter Background fluorescence was deter-mined in the absence of cell extract and was subtracted
Statistical analysis
Values are given as means ± SD and analyzed by Student’s t-test Significance was assessed at the P < 0.05 level
Acknowledgements
We thank Dr V M Sadagopa Ramanujam (The University of Texas Medical Branch, Galveston, TX) for help with metal analyses by atomic absorption spectrophotometry (supported by the Human
Trang 10Nutri-tion Research Facility) and Professor Richard Glass
(University of Arizona, Tucson, AZ) for helpful
dis-cussions This work was supported by NIH Grant
GM 065388 to WM
References
1 Robertson RP (2004) Chronic oxidative stress as a
cen-tral mechanism for glucose toxicity in pancreatic islet
beta cells in diabetes J Biol Chem 279, 42351–42354
2 Beisswenger PJ, Howell SK, Nelson RG, Mauer M &
Szwergold BS (2003) Alpha-oxoaldehyde metabolism
and diabetic complications Biochem Soc Trans 31,
1358–1363
3 Esterbauer H, Schaur RJ & Zollner H (1991) Chemistry
and biochemistry of 4-hydroxynonenal, malonaldehyde
and related aldehydes Free Radic Biol Med 11, 81–128
4 Kehrer JP & Biswal SS (2000) The molecular effects of
acrolein Toxicol Sci 57, 6–15
5 Dost FN (1991) Acute toxicology of components of
vegetation smoke Rev Environ Contam Toxicol 119, 1–46
6 Thornalley PJ (1996) Pharmacology of methylglyoxal:
formation, modification of proteins and nucleic acids,
and enzymatic detoxification) a role in pathogenesis
and antiproliferative chemotherapy Gen Pharmacol 27,
565–573
7 Uchida K (2000) Role of reactive aldehyde in
cardiovas-cular diseases Free Radic Biol Med 28, 1685–1696
8 Leikauf GD (1992) Mechanisms of aldehyde-induced
bronchial reactivity: role of airway epithelium Res Rep
Health Eff Institute 49, 1–35
9 Johnsen J, Stowell A & Morland J (1992) Clinical
responses in relation to blood acetaldehyde levels
Phar-macol Toxicol 70, 41–45
10 Eriksson CJ (2001) The role of acetaldehyde in the
actions of alcohol (update 2000) Alcohol Clin Exp Res
25, 15S–32S
11 Maret W (2004) Zinc and sulfur: a critical biological
partnership Biochemistry 43, 3301–3309
12 Beyersmann D & Haase H (2001) Functions of zinc in
signaling, proliferation and differentiation of
mamma-lian cells Biometals 14, 331–341
13 Frederickson CJ, Koh JY & Bush AI (2005) The
neuro-biology of zinc in health and disease Nat Rev Neurosci
6, 449–462
14 Jacob C, Maret W & Vallee BL (1998) Ebselen, a
sele-nium-containing redox drug, releases zinc from
metal-lothionein Biochem Biophys Res Commun 248, 569–573
15 Rando RR (1974) Allyl alcohol-induced irreversible
inhibition of yeast alcohol dehydrogenase Biochem
Pharmacol 23, 2328–2331
16 Shiah SG, Kao YR, Wu FYH & Wu CW (2003)
Inhibi-tion of invasion and angiogenesis by zinc-chelating
agent disulfiram Mol Pharmacol 64, 1076–1084
17 Ka¨gi JHR (2001) Overview of metallothionein Methods Enzymol 205, 613–626
18 Haase H & Maret W (2003) Intracellular zinc fluctua-tions modulate protein tyrosine phosphatase activity in insulin⁄ insulin-like growth factor-1 signaling Exp Cell Res 291, 289–298
19 Maret W (1994) Oxidative metal release from metallo-thionein via zinc–thiol⁄ disulfide interchange Proc Natl Acad Sci USA 91, 237–241
20 Chen Y & Maret W (2001) Catalytic selenols couple the redox cycles of metallothionein and glutathione Eur J Biochem 268, 3346–3353
21 Chen Y, Irie Y, Keung WM & Maret W (2002) S-nitro-sothiols react preferentially with zinc thiolate clusters of metallothionein III through transnitrosation Biochemis-try 41, 8360–8367
22 Deetz JS, Luehr CA & Vallee BL (1984) Human liver alcohol dehydrogenase isozymes: reaction of aldehydes and ketones Biochemistry 23, 6822–6828
23 Toninello A, Pietrangeli P, De Marchi U, Salvi M & Mondovi B (2006) Amine oxidases in apoptosis and cancer Biochim Biophys Acta 1765, 1–13
24 Tirumulai R, Rajesh Kuram T, Mai KH & Biswal S (2002) Acrolein causes transcriptional induction of phase II genes by activation of Nrf2 in human lung type
II epithelial (A549) cells Toxicol Lett 132, 27–36
25 Markesbery WR & Lovell MA (1998) Four-hydroxyno-nenal, a product of lipid peroxidation, is increased in the brain in Alzheimer’s disease Neurobiol Aging 19, 33–36
26 Lovell MA, Xie CS & Markesbery WR (2001) Acrolein
is increased in Alzheimer’s disease brain and is toxic to primary hippocampal cultures Neurobiol Aging 22, 187– 194
27 Bartfay WJ, Hou D, Lehotay DC, Luo X, Bartfay E, Backx PH & Liu PP (2000) Cytotoxic aldehyde genera-tion in heart following acute iron-loading J Trace Elem Med Biol 14, 14–20
28 Dinkova-Kostova AT, Holtzclaw WD & Wakabayashi
N (2005) Keap1, the sensor for electrophiles and oxi-dants that regulates the phase 2 response, is a zinc metalloprotein Biochemistry 44, 6889–6899
29 Helander A, Lowenmo C & Johansson M (1993) Distri-bution of acetaldehyde in human blood: effects of etha-nol and treatment with disulfiram Alcohol Alcohol 28, 461–468
30 Isse T, Oyama T, Kitagawa K, Matsuno K, Matsumoto
A, Yoshida A, Nakayama K, Nakayama K & Kawa-moto T (2002) Diminished alcohol preference in trans-genic mice lacking aldehyde dehydrogenase activity Pharmacogenetics 12, 621–626
31 Keung WM, Lazo O, Kunze L & Vallee BL (1995) Daidzin suppresses ethanol consumption by Syrian golden hamsters without blocking acetaldehyde metabo-lism Proc Natl Acad Sci USA 92, 8990–8993