Production of antibodies that recognize the heterogeneity of immunoreactive sites in human hemoglobin chemically modified by acetaldehyde

Analysis of N(ε) -ethyllysine in human plasma proteins by gas chromatography-negative ion chemical ionization/mass spectrometry as a biomarker for exposure to acetaldehyde and alcohol

BACKGROUND

N(ε) -ethyllysine (NEL) is a major stable adduct formed by the reaction of acetaldehyde (AA) with lysine residues in proteins. However, its occurrence and levels in biological specimens and its relationship with AA/alcohol exposure-associated disorders have not been fully elucidated. In this study, we have developed a sensitive and specific method to quantitate NEL levels in human plasma proteins.

METHODS

The method consists of (1) purification of the protein fraction of interest by Sephadex G-15 to remove low molecular substances, (2) hydrolysis of proteins with Pronase E in the presence of stable isotope-labeled internal standards, (3) derivatization of amino acids with pentafluorobenzyl (PFB) bromide, and (4) quantification of the PFB derivatives of NEL and l-lysine using gas chromatography-negative ion chemical ionization/mass spectrometry in a selected ion monitoring mode.

RESULTS

Using the above method, the NEL levels in human plasma proteins obtained from 10 each of control subjects and alcoholic patients were measured. NEL was detected in all samples analyzed, the average level of NEL in the plasma proteins of alcoholic patients (1.17 ± 0.36 NEL/1,000 l-lysine) being significantly higher than that of control subjects (0.26 ± 0.07 NEL/1,000 l-lysine).

CONCLUSIONS

The method could be applied to molecular epidemiological studies to investigate possible associations between the NEL levels in human tissue proteins and human diseases associated with exposure to AA and alcohol.

Modification of carbonic anhydrase II with acetaldehyde, the first metabolite of ethanol, leads to decreased enzyme activity

BACKGROUND

Acetaldehyde, the first metabolite of ethanol, can generate covalent modifications of proteins and cellular constituents. However, functional consequences of such modification remain poorly defined. In the present study, we examined acetaldehyde reaction with human carbonic anhydrase (CA) isozyme II, which has several features that make it a suitable target protein: It is widely expressed, its enzymatic activity can be monitored, its structural and catalytic properties are known, and it contains 24 lysine residues, which are accessible sites for aldehyde reaction.

RESULTS

Acetaldehyde treatment in the absence and presence of a reducing agent (NaBH3(CN)) caused shifts in the pI values of CA II. SDS-PAGE indicated a shift toward a slightly higher molecular mass. High-resolution mass spectra of CA II, measured with and without NaBH3(CN), indicated the presence of an unmodified protein, as expected. Mass spectra of CA II treated with acetaldehyde revealed a modified protein form (+26 Da), consistent with a "Schiff base" formation between acetaldehyde and one of the primary NH2 groups (e.g., in lysine side chain) in the protein structure. This reaction was highly specific, given the relative abundance of over 90% of the modified protein. In reducing conditions, each CA II molecule had reacted with 9-19 (14 on average) acetaldehyde molecules (+28 Da), consistent with further reduction of the "Schiff bases" to substituted amines (N-ethyllysine residues). The acetaldehyde-modified protein showed decreased CA enzymatic activity.

CONCLUSION

The acetaldehyde-derived modifications in CA II molecule may have physiological consequences in alcoholic patients.

Alcoholic macrocytosis--is there a role for acetaldehyde and adducts?

Although alcohol abuse is known to cause a wide array of adverse effects on blood cell formation, the molecular mechanisms by which alcohol exerts its toxic actions have remained poorly defined. Elevated mean corpuscular volume (MCV), macrocytosis, is the most typical morphological abnormality induced by excessive ethanol consumption. This paper reviews recent data indicating that acetaldehyde, the first metabolite of ethanol, may play a role in the haematological derangements in peripheral blood cells and in bone marrow of alcoholic patients. Studies in experimental animals and in human alcoholics have shown that acetaldehyde can bind to proteins and cellular constituents forming stable adducts. Elevated adduct levels have been found from the erythrocytes of alcohol abusers, which may also be associated with ethanol-induced effects in haematopoiesis and adverse consequences in cellular functions.

Autoimmune responses against oxidant stress and acetaldehyde-derived epitopes in human alcohol consumers

BACKGROUND

Studies in experimental animals have indicated that chronic ethanol ingestion triggers the formation of antibodies directed against proteins modified with reactive metabolites of ethanol and products of lipid peroxidation. However, the nature and prevalence of such antibodies have not been compared previously in alcoholic patients.

METHODS

Autoantibodies against adducts with acetaldehyde- (AA), malondialdehyde- (MDA), and oxidized epitopes (Ox) were examined from sera of 54 alcohol consumers with (n = 28) or without (n = 26) liver disease, and from 20 nondrinking controls.

RESULTS

Anti-AA-adduct IgA and IgG antibodies were elevated in 64% and 31% of patients with biopsy-proven alcoholic liver disease (ALD, n = 28), respectively. The IgA titers were significantly higher than those from nondrinking controls (p < 0.001), or heavy drinkers without significant liver disease (p < 0.001). Anti-MDA adduct titers (IgG) were elevated in 70% of the ALD patients. These titers were significantly higher (p < 0.001) than those from nondrinking controls, or heavy drinkers without liver disease. Antibodies (IgG) against Ox epitopes occurred in 43% of ALD patients, and the titers also were significantly higher (p < 0.05) than those from nondrinking controls. The anti-AA and anti-MDA adduct titers in ALD patients correlated significantly with the combined clinical and laboratory index (CCLI) of liver disease severity (r(s) = 0.449, p < 0.05; r(s) = 0.566, p < 0.01, respectively), the highest prevalences of anti-AA-adducts (73%) and anti-MDA-adducts (76%) occurring in ALD patients with cirrhosis.

CONCLUSIONS

The present results indicated that autoantibodies against several distinct types of protein modifications are generated in ALD patients showing an association with the severity of liver disease.

Soluble proteins modified with acetaldehyde and malondialdehyde are immunogenic in the absence of adjuvant

Recent studies have shown that the alcohol metabolites malondialdehyde and acetaldehyde can combine to form a stable adduct (MAA) on proteins. This adduct has been detected in the livers of rats chronically consuming ethanol, and serum antibodies to MAA have been observed at significantly higher concentrations in ethanol-fed when compared with pair-fed or chow-fed control rats. More recently, preliminary studies have strongly suggested that the MAA adduct is capable of stimulating antibody responses to soluble proteins in the absence of adjuvants. The antibodies produced recognize either the MAA epitope or the carrier protein itself. Therefore, it was the purpose of this study to examine the potential immunogenicity of MAA-modified exogenous proteins in the absence of adjuvants. Balb/c mice were immunized in the presence or absence of adjuvant with different concentrations of unmodified or MAA-modified proteins. The antibody response to both the MAA epitope and unmodified protein epitopes were determined by ELISA. In the absence of adjuvant, significant antibody responses were induced to both the MAA epitope and nonmodified protein epitopes. Smaller immunizing doses of MAA-protein conjugate favored the production of antibodies to nonmodified proteins, whereas larger doses induced a strong anti-MAA response. In studies to begin determining a mechanism for the specificity of the response in the absence of adjuvants, peritoneal macrophages were found to bind and degrade MAA-adducted proteins through the use of a scavenger receptor. This indicated that MAA-adducted proteins may be specifically taken up and epitopes presented to the humoral immune system in the absence of adjuvants. Importantly, these are the first data showing that an alcohol-related metabolite can induce an antibody response in the absence of adjuvant and suggesting a mechanism by which antibody to the MAA adduct or its carrier (exogenous or endogenous) proteins may be generated in vivo.

Generation of acetate and production of ethyl-lysine in the reaction of acetaldehyde plus serum albumin

We report that incubation of acetaldehyde with bovine serum albumin results in the generation of acetate in a reaction that is directly proportional to the levels of albumin and exponentially dependent on the concentration of acetaldehyde. Both reactants need to be present for acetate to be formed. The oxidation of acetaldehyde into acetate requires that a reduced product also be generated in the reaction. It was hypothesized that, at high concentrations, acetaldehyde itself may reduce the Schiff bases formed in the reaction of a second molecule of acetaldehyde with amino groups in the protein, resulting in the generation of ethyl-lysine moieties. Incubation of acetaldehyde (240 mM) with bovine serum albumin was found to generate ethyl-lysine moieties as determined by a specific monoclonal antibody. Immunization of rabbits with products of the reaction of bovine serum albumin with acetaldehyde led to the generation of antibodies that reacted to reduced adducts formed in the reaction of acetaldehyde and proteins in the presence of sodium cyanoborohydride. However, the generation of acetate from acetaldehyde plus albumin was 60-fold greater than could be explained by the reduction of Schiff bases, as determined by the maximal incorporation of [14C]-acetaldehyde into an acid-precipitable protein fraction. Thus, other mechanisms to generate acetate also occur. The present findings provide an explanation for earlier reports that acetaldehyde adducts formed under "nonreducing" conditions generate antibodies that recognize reduced acetaldehyde protein adducts. However, the mechanism by which the bulk of acetate is generated in the reaction of acetaldehyde and bovine serum albumin remains to be elucidated.

Chromatographically identified alcohol-induced haemoglobin adducts as markers of alcohol abuse among women

BACKGROUND

Alcohol-induced changes in haemoglobin have been suggested as potential biochemical markers of alcohol abuse. In this study, we investigated the diagnostic value of alcohol-induced haemoglobin adducts among women.

METHODS

Whole (Hb fractions) and affinity-purified (AHb fractions) haemolysates from 87 women in three groups (a) social drinkers (n=31), (b) heavy drinkers (n=27) and (c) alcoholic subjects (n=29) - were analysed by HPLC-CEC.

RESULTS

Three fractions (HbA 1a2, HbA1dl and AHbA1d1) showed significant differences (P<0.05) between the groups and a significant positive correlation (P<0.05) with self-reported alcohol consumption (r=0.58-0.76) as determined by the Malmö modified Michigan Alcoholism Screening Test (MmMAST) and structured CAGE questionnaire (r=0.58-0.76). HbA1a2, HbA1d1 and AHbA1d1 had specificities of 97%, 97% and 100% respectively and detected 41%, 33% and 78% of heavy drinkers with overall accuracies (OAs) of 71%, 67% and 90%. Sensitivities in the detection of alcoholic subjects were 86% (OA=92%), 76% (OA=87%) and 81 % (OA=91%) respectively. The fractions had higher OAs than traditional markers of alcohol abuse.

CONCLUSION

This study indicates that at least three alcohol-induced haemoglobin adducts occurring in vivo can be measured with promising diagnostic efficiency among women.

Structural characterisation of acetaldehyde adducts formed by a synthetic peptide mimicking the N-terminus of the hemoglobin beta-chain under reducing and nonreducing conditions

This work was carried out to elucidate the structures resulting from acetaldehyde-induced modifications at the haemoglobin beta-chain N-terminal residues under different experimental conditions. A synthetic peptide (Val-His-Leu-Thr-Pro-Glu-Cys) of m/z 798, which represents the six N-terminal residues of the haemoglobin beta-chain N-terminus with an additional C-terminal cysteine, was used as a model compound. Peptide-acetaldehyde adducts were separated by reverse-phase HPLC. Fast-atom-bombardment MS and linked-scan (B/E) spectra were used to study adduct structures. Under nonreducing conditions at pH 7.4 (1:10 peptide/acetaldehyde molar ratio), two types of adducts of m/z 824 were formed, both with modifications at the N-terminal valine. These two adducts were shown to be a Schiff base, and a cyclic 2-methyl-imidazolidine-4-one. The 2-methyl-imidazolidine-4-one adduct was demonstrated to be formed via the Schiff base and to undergo dissociation gradually after 24 h. Reducing conditions at pH 7.4 (peptide /acetaldehyde molar ratio of 1:1 with 20 mM NaCNBH3) resulted in the formation of an adduct of m/z 826, which was shown to be an N-ethylated adduct of valine. A small amount of nonreduced adducts were also formed under these conditions. Reducing conditions at pH 9.0 (1:10 peptide/acetaldehyde molar ratio with 20 mM NaCNBH3) yielded two secondary, i.e. diethylated (m/z 854), products very rapidly. The cysteine residue of the peptide was not found to form an adduct with acetaldehyde under physiological pH.

Relationship between alcohol intake and immunoglobulin a immunoreactivity with acetaldehyde-modified bovine serum albumin

Acetaldehyde, the main metabolite of ethanol, is a highly reactive species that reacts with macromolecules to produce unstable and stable adducts. Acetaldehyde-modified proteins are immunogenic and have been detected in the liver and blood of alcoholics. Furthermore, antibodies reactive with acetaldehyde-modified proteins have been detected in the plasma of social drinkers and alcoholics. However, the class distribution of immunoglobulins reactive with modified proteins was different in the two groups, being predominantly immunoglobulin (Ig)M in social drinkers, but IgM and IgA in alcoholics. In this study, we demonstrate that heavy drinkers (alcohol intake > 130 g/week for females and 150 g/week for males) also exhibit IgA reactivity with acetaldehyde-modified proteins. The IgA adduct-specific reactivity (IgA reactivity with acetaldehyde-modified bovine serum albumin-reactivity with native bovine serum albumin) showed a moderate correlation with self-reported alcohol intake, but did not correlate with markers such as plasma transaminase, gamma-glutamyltransferase activity, or mean corpuscular volume. IgA adduct-specific reactivity had similar specificity to the conventional tests of alcohol abuse, but had higher sensitivity than the other tests, especially with heavy drinkers. Data presented herein demonstrate that elevated IgA reactivity with acetaldehyde-modified epitopes is associated with heavy drinking and is a potential marker for high alcohol intake.

Serum low density lipoprotein of alcoholic patients is chemically modified in vivo and induces apolipoprotein E synthesis by macrophages

This work was carried out to investigate the effect of alcohol drinking on serum LDL. Agarose gel electrophoresis showed that LDL samples from alcoholic patients without serious liver disease were more negatively charged and moved faster toward the cathode than LDL from nondrinking control subjects. Rabbit antibodies raised by using keyhole limpet hemocyanin modified in vitro by 4-hydroxynonenal or by acetaldehyde as immunogens reacted more strongly with patients' LDL than with control LDL, indicating the presence of oxidatively modified epitopes and acetaldehyde adducts in alcoholic patients' LDL. LDL of alcoholic patients has decreased vitamin E contents. The electromobility of LDL decreased after abstinence from alcohol and returned to normal in 2 wk, but this was not accompanied by a significant increase in its vitamin E contents. When incubated with mouse peritoneal macrophages, patients' LDL induced apolipoprotein E secretion by threefold over control LDL with a concomitant increase in cellular cholesterol. Our results thus demonstrate that LDL of alcoholic patients has lower vitamin E content, is chemically modified in vivo, and exhibits altered biological function. These changes in heavy alcoholic drinkers may render LDL more atherogenic and thereby may counter the antiatherosclerosis effects of moderate alcohol consumption.

Structural analysis of peptide-acetaldehyde adducts by mass spectrometry and production of antibodies directed against nonreduced protein-acetaldehyde adducts

Acetaldehyde can form protein-acetaldehyde adducts (AAs) in vivo and may play a role in the genesis of alcoholic liver disease. The nature of the chemical modification of proteins by acetaldehyde in vivo has not been elucidated. In vitro, acetaldehyde can form reversible adducts including a Schiff's base with lysine (K) and imidazolidinone with terminal amino groups of proteins such as human hemoglobin (Hb). In this study, we used FAB/MS to analyze the products of peptide-AAs (pep-AAs) formed by incubating acetaldehyde with Hb peptides. We then used an octabranched multiple antigen peptide (MAP) system containing Hb peptide-AAs to raise antibodies. Three Hb peptides [i.e., 8-pep consisting of 8 residues (V1HLTPVEK8) at the N-terminus of beta-chain of human sickle-cell Hb, 11-pep-gly consisting of 11 residues (G56NPKVKAHGKK66) in a segment of beta-chain rich in lysine, and 11-pep-pro that consists of the same sequence as 11-pep-gly, except G56 was replaced by proline (P)] were incubated with 1 mM acetaldehyde at 4 degrees C for 7d without NaCNBH3 (nonreduced conditions). Analysis by FAB/MS showed that 8-pep formed an imidazolidinone at the N-terminal valine, 11-pep-gly formed a Schiff's base and imidazolidinone at the N-terminus, whereas 11-pep-pro that lacks a free alpha-amino group formed only a Schiff's base at K59. By contrast, incubation of these Hb peptides with 250 mM acetaldehyde and NaCNBH3 at 37 degrees C for 1 hr (reduced conditions) produced mono- and diethylated modifications of all available K residues, as well as the N-terminal amino group.(ABSTRACT TRUNCATED AT 250 WORDS)

Collagen-acetaldehyde adducts in alcoholic and nonalcoholic liver diseases

Alcoholic and, to a lesser extent, nonalcoholic patients with liver disease have serum antibodies to acetaldehyde-protein adducts produced in vitro. These antibodies presumably reflect the presence of adducts in the liver, but the protein that triggers this immune response has not been identified. To study this, we measured the reactivity of cytosolic proteins to rabbit IgG developed against a P-450 2E1-acetaldehyde adduct, isolated from alcohol-fed rats, that recognizes acetaldehyde-modified epitopes in proteins. Adducts were determined on Western blots by scanning densitometry of antibody-linked alkaline phosphatase activity in 4 normal livers and in needle biopsy specimens from subjects with liver disease, 17 alcoholic and 14 nonalcoholic. In all livers, except for a normal one, we found a reactive protein of at least 200 kD, similar to the collagen-acetaldehyde adduct we reported to be markedly increased in rats with experimentally induced cirrhosis. The immunostaining intensity in the alcoholic patients with liver disease was eightfold (p < 0.01) and that in nonalcoholic patients with liver disease was fourfold, greater (p < 0.02) than the weak staining in normal livers; it correlated with the degree of inflammation and serum AST or gamma-glutamyl transpeptidase activities. The adduct was reproduced on incubation of normal cytosolic proteins with 2.5 mmol/L acetaldehyde, whereas higher concentrations yielded many additional adducts; the adduct also reacted with IgG antibody to rat collagen type I and disappeared after digestion with collagenase, suggesting that the target protein is a form of collagen.(ABSTRACT TRUNCATED AT 250 WORDS)