- Email: [email protected]
Stability of acetaldehyde in human blood samples
BIOCHEMICAL
MEDICINE
20, 167-179
(1978)
Stability of Acetaldehyde in Human Blood Samples ALLAN
R. STOWELL,
Department
ROBERT
M. GREENWAY,
of Chemistry, Biochemistry. Palmerston North. Received
January
AND RICHARD
and Biophysics. NeM* Zealand
Massey
D. BATT
University.
12, 1978
It is well known that the measurement of acetaldehyde in ethanolcontaining human blood samples is complicated by the production of acetaldehyde from ethanol during deproteinization prior to assay (l-3). To overcome this problem, analyses can be carried out on separated plasma since the deproteinization of plasma containing ethanol results in the formation of only negligible amounts of acetaldehyde as an artifact (I ,3). However, when recovery experiments with whole blood were performed in this laboratory, acetaldehyde could not be completely recovered if plasma was separated and used for analysis. It was assumed that the added acetaldehyde was either partially metabolized or bound to blood constituents during the time required for separation of the plasma from whole blood. Accordingly, it was decided to determine accurately the rate of acetaldehyde disappearance from human blood during times corresponding to those required to take blood and separate the plasma. From this study, it was found that acetaldehyde present in blood samples as a result of ethanol metabolism in vivo did not decrease at a significant rate under conditions which resulted in the rapid disappearance of acetaldehyde which had been added to blood samples. It is suggested that the acetaldehyde formed in vivo may not be present in blood in a free form. MATERIALS
AND METHODS
Reagents. [ I-14C]Ethanol with a specific activity of 5 I .4 mCi/mmol and a radiochemical purity of 98% was obtained from the Amersham Radiochemical Centre, Bucks., England. [ I-14C]Acetaldehyde having the same specific activity was produced from the [ I-14C]ethanol by enzymic oxidation using horse liver alcohol dehydrogenase (Sigma Chemical Co., 167 OOO6-2944178/0202-0167$02.OO/O Copyright @ 1978 by Academic Press. Inc All rights of reproduction in any form reserved.
168
STOWELL,
GREENWAY,
AND
BATT
St. Louis, MO.). The reaction mixture used for the conversion of ethanol to acetaldehyde was the same as that used in the alcohol assay method of Lundquist and Wolthers (4), with semicarbazide added to trap the acetaldehyde. The [l-l”Clacetaldehyde was stored at -20°C as the crystalline semicarbazone and regenerated before use by steam distillation in the presence of dilute H,SO,. Scintillation grade 1,4-di(2-(5phenyloxazolyl))benzene (POPOP), was obtained from Hopkin and Williams, Chadwell Heath, Essex, England and reagent grade 2,5-diphenyloxazole (PPO) from BDH Chemicals Ltd., Poole, England. Acetaldehyde was redistilled before use and stored at 4°C for up to 6 months. Dilute solutions were freshly prepared when required. The scintillation solvent used for radioactivity determinations was prepared as follows: I vol of Triton Xl00 was mixed with 2 vol of redistilled toluene; PPO and POPOP were dissolved in this mixture to give final concentrations of 4 and 0.2 g/liter, respectively. Determination of radioactivity. Radioactivity determinations were carried out in a Beckman LS 350 liquid scintillation counter and corrections for quenching were made using an external standard. Counting efficiencies for samples were within the range of 84 to 87%. Blood samples. Samples of fresh human venous blood treated with heparin or EDTA were used for all experiments. The blood was obtained from an antecubital vein and, unless otherwise stated, the additions of acetaldehyde to blood or plasma samples were made in such a way that the required reagent concentrations were obtained in samples without increasing their volumes by more than 2%. Human blood samples containing endogenous ethanol and acetaldehyde were obtained from healthy volunteers who had previously consumed 1 g/kg of ethanol as vodka, 1.5 to 2 hr before the blood samples were taken. Acetaldehyde disappearance from different blood or plasma samples was determined by assaying aliquots taken at different times after acetaldehyde was mixed with the samples. The incubation of acetaldehyde with blood or plasma was carried out in all cases in sealed containers, to minimize acetaldehyde losses by evaporation. Sample treatment and assay methods. For all experiments, samples were assayed using the semi- or fully automated enzymic acetaldehyde assay methods described by Stowell et al. (5). Samples to be assayed were diluted with either I vol of ice-cold 1 M perchloric acid (PCA) or 9 vol of ice-cold 0.6 M PCA and then centrifuged. The PCA stopped the acetaldehyde disappearance and stabilized the remaining acetaldehyde levels. The clear, protein-free supernatants obtained were assayed for acetaldehyde. Aqueous acetaldehyde standards were used in all experiments. Blood ethanol levels were determined by a gas chromatographic
STABILITY
OF ACETALDEHYDE
IN BLOOD
169
method involving the analysis of head-space gas equilibrated with PCA supernatants of blood samples. Gas samples were chromatographed on a column of Porapak Q (Waters Associates Inc., Framingham, Mass.) at 130°C with a carrier gas flow rate of 40 ml/min. Acetonitrile was used as an internal standard. This method has been described fully by Couchman (6). In an attempt to identify acetic acid in blood samples by gas chromatography, a column packed with Chromasorb W, AW-DMCS, 60/80 mesh. coated with FFAP stationary phase (obtained from Varian Aerograph, Walnut Creek, Calif.) was used. For all experiments, acetaldehyde levels have been corrected for the small amounts of acetaldehyde detectable in blood samples to which no acetaldehyde had been added. These blank levels were stable for at least 30 min under the experimental conditions employed. Where blood samples containing ethanol were assayed for acetaldehyde, levels of acetaldehyde caused by deproteinization were determined for each blood sample using a correction curve prepared by deproteinizing blank blood samples (obtained before alcohol was consumed) to which ethanol was added. These were assayed in the same way as the samples. This method of correction has been described fully by Stowell et al. (3). The artifact level in each sample taken from subjects metabolizing ethanol was determined from the correction curve after the ethanol levels of the samples had been determined. Acetate determinations were performed using the method of Guynn and Veech (7) modified by using yeast acetyl coenzyme A synthetase (EC 6.2.1.1, obtained from Sigma Chemical Co.) instead of beef heart mitochondrial acetyl coenzyme A synthetase. PCA supernatants were neutralized with K&O, before being assayed for acetate. Use of I-‘4C-labeled acetafdehyde. In an experiment to determine whether acetaldehyde is metabolized by whole blood to a nonvolatile compound or possibly bound to blood protein, [I-‘“Clacetaldehyde was used as follows: Approximately 10 pg of [ I-‘*C]acetaldehyde was distilled into about 10 ml of ice-cold water and to the solution obtained, sufficient carrier acetaldehyde, NaCl, and water were added to produce 50 ml of a 0.9% NaCl solution containing 46 PM acetaldehyde. This solution (20 ml) and a sample of human blood (20 ml) were brought to 37°C in separate sealed containers, mixed, and placed in a sealed flask at 37°C. Aliquots (5 ml) were taken at intervals up to 40 min after mixing and added to ice-cold lM PCA solution (5 ml). A blank sample was prepared by deproteinizing in a similar manner, a 1: 1 (v/v) mixture of blood and 0.9% saline solution which did not contain added acetaldehyde. In a control experiment, I vol of 0.9% saline solution was mixed with 1 vol of the saline/[l-‘*C]acetaldehyde solution described above, the mixture being incubated for 40 min at 37°C. Two samples were taken from the
170
STOWELL,
GREENWAY,
AND
BATT
mixture into 1 vol of ice-cold 1M PCA (as described previously), one at zero time (the time of mixing) and one 40 min later. All PCA-treated samples were centrifuged and different aliquots of the protein-free supernatants were used for: (a) acetaldehyde assays, (b) determination of total radioactivity, and (c) determination of radioactivity in nonvolatile substances. For (b), two l-ml samples of each PCA supernatant were added to 10 ml of scintillation solvent. For (c), 2-ml samples of each supernatant were added to 2 ml of 1.2 M KHCO,, to bring the pH to 7.4. Potassium chlorate was allowed to crystallize out and 2 ml of the neutralized solutions were evaporated to half their volumes at 40°C. The solutions were made up to 2 ml again and duplicate 0.9-ml aliquots were added to 10 ml of scintillation solvent with 0.1 ml of water. RESULTS
Comparison of the rates of disappearance of acetaldehyde in human whole blood and plasma samples at 4°C and 37°C. Acetaldehyde disappearance in blood samples from an initial level of 10 PM is not only dependent on temperature but is also dependent to a large extent, on the presence of blood cells (Fig. 1). Individual variations in blood acetaldehyde disappearance rates. The disappearance of acetaldehyde added to blood to an initial level of 20 PM was studied at 37°C in samples from seven individuals (two males and five females). The results given in Fig. 2 show that the disappearance rate is exponential and there is little individual variation in this rate. Some of the blood samples used for this experiment were heparinized and for others, EDTA was used as the anticoagulant. No significant difference was found between the acetaldehyde disappearance rates in blood samples containing different anticoagulants. Experiments were performed separately on each blood sample which was stored at 4°C until required. The time between the testing of the first and last sample was 5 hr and storage at 4°C for this time did not significantly affect the rate of removal of acetaldehyde. Blank levels found in the seven different blood samples were 0.5, 1.1,4.8, 1. I, 0.7, 1.1, and 1.3 JAM. In all cases the added acetaldehyde was found to decay almost to these blank levels in 24 min. Fate of acetaldehyde added to human blood. When [ l-14C]acetaldehyde was added to blood samples the radioactivity in volatile compounds declined at essentially the same rate as acetaldehyde disappeared, while the counts in nonvolatile compounds increased at the same rate (Fig. 3). It was clear that acetaldehyde was not being converted to any appreciable extent, into a volatile product such as ethanol and the results indicated that acetaldehyde was being converted to a nonvolatile compound such as acetate. If the decay of acetaldehyde in diluted blood was due to irreversible binding to protein then the total counts in PCA supernatants would
STABILITY
OF ACETALDEHYDE
L
a
12
IN BLOOD
16
171
20
Time I minutes1 FIG. I. Comparison of the rates of disappearance of acetaldehyde in human whole blood and plasma samples at 4°C and 37°C. Acetaldehyde was added to whole blood and plasma samples to produce an initial concentration of IO pM and samples were treated and assayed as described in Materials and Methods. The results illustrated were obtained using blood from one individual and the points represent single determinations with a maximum error of 2 0.5 PM. The different samples incubated were: plasma at 4°C (O-O), plasma at 37°C (C-0). whole blood at 4°C (O---O) and whole blood at 37°C (W---W).
be expected to decline at the same rate as the acetaldehyde. However, only a slight decrease in total counts was observed in the PCA supernatants. In the control experiment no significant loss of acetaldehyde or radioactivity occurred during the 40-min incubation period at 37°C. In attempts to identify the nonvolatile radioactive compound formed from acetaldehyde in blood, acidified samples prepared from blood in which radioactive acetaldehyde had reacted were subjected to gas chromatography after the addition of carrier acetic acid. The only peak containing counts corresponded to the position of the standard acetic acid peak but a low count recovery from the column (50-60%) precluded absolute identification of acetate as the sole labeled product. The results given in Fig. 4 show that levels of acetate above the blank blood acetate level are definitely produced when acetaldehyde disappeared in blood samples from an initial level of 1.0 mM. In this experiment, no detectable ethanol appeared in the blood sample to which acetaldehyde was added. The experiment could not be repeated to give significant results using low concentrations of acetaldehyde because the acetate formed represented only a small proportion of the acetate already present in the blood samples.
172
STOWELL,
GREENWAY,
AND BATT
*Oi
Time
( minutes
I
FIG. 2. Individual variations in acetaldehyde disappearance in human whole blood samples at 37°C. The points in the figure represent the means of seven determinations, using blood samples obtained from seven individuals. The bar lines represent the range of values obtained and the figures by each point represent one standard deviation. Acetalddhyde was added to all blood samples to produce an initial concentration of 20 pM and samples were treated and assayed as described in Materials and Methods.
Disappearance of endogenous acetaldehyde in blood samples taken from subjects metabolizing ethanol. Although it was clear that acetaldehyde rapidly disappears when added in vitro to blood containing little or
no endogenous acetaldehyde or ethanol, it was considered necessary to determine whether the same rate of disappearance occurred in blood samples containing acetaldehyde which had been formed from ethanol, in vivo .
STABILITY
OF ACETALDEHYDE
IN BLOOD
173
30
25 1
\ \ \
20”
15-
10-
5-
Time
(minutes
I
FIG. 3. The metabolism of [l-*4C]acetaldehyde by human venous blood in vitro. The conversion of [ I-i4C]acetaldehyde to a nonvolatile product in blood diluted with one volume of 0.9?& NaCl is shown. The results shown are: total radioactivity (ho), volatile radioactivity (U--•),non-volatikradioactivity (C-U), and acetaldehyde (e-0). Acetaldehyde was added to the blood/saline mixture to produce an initial concentration of 23 FM. See Materials and Methods for other experimental details.
Blood samples were taken from subjects metabolizing ethanol and aliquots were deproteinized at intervals prior to estimating acetaldehyde levels. The results of such experiments are shown in Fig. 5. A I:9 dilution of blood aliquots with PCA was used to avoid excessive formation of artifact acetaldehyde (3). The results obtained were in striking contrast to those obtained in the previous experiments. Although the level of acetaldehyde measured above the artifact level appeared to fluctuate throughout the incubation period, no significant disappearance was observed. Since these blood samples contained ethanol and presumably, acetate formed from ethanol, it was considered possible that the presence of these two compounds
174
STOWELL,
GREENWAY,
0
20
10
30
AND BATT
LO
50
60
:
Time I minutes) FIG. 4. Production of acetate during the disappearance of acetaldehyde added to human venous blood. Two aliquots of a single blood sample were incubated at 37°C. To one aliquot, acetaldehyde was added to produce an initial concentration of I .O mM. The disappearance of acetaldehyde (LO) and the appearance of acetate (C-0) in this aliquot were measured. No additions were made to the second blood aliquot and this was used as a control to follow changes in the blank acetate level (S-----m). The error associated with the acetate levels is ~0.03 mM and that associated with the acetaldehyde levels is kO.04 mM.
might be responsible, in some way, for the absence of a rapid disappearance rate in endogenous acetaldehyde levels. The following experiment was therefore carried out to test such a possibility. Disappearance of added acetaldehyde from blood samples taken from subjects metabolizing ethanol. Acetaldehyde was added to blood samples containing acetaldehyde formed form ethanol in vivo and its disappear-
ance was measured. At the same time, the levels of endogenous acetaldehyde in the same blood samples to which no acetaldehyde had been added were also determined. This experiment was performed several times at two different temperatures (37°C and room temperature) using blood samples taken from different individuals. In each case a similar result was obtained. A typical set of results (Fig. 6) shows that in the same blood sample, acetaldehyde formed from ethanol in vivo does not disappear significantly over a period of 20 min after the blood sample has been taken
STABILITY
0
2
OF
L
ACETALDEHYDE
6 Incubation
8
10 Time
175
IN BLOOD
12
1L
( minutes
16
18
20
1
FIG. 5. Stability of endogenous acetaldehyde in venous blood samples taken from subjects metabolizing ethanol. The five blood samples were taken from different individuals and incubated at 37°C. The time refers to the interval from the moment the samples were drawn. The maximum error in each point is approximately ? 2.5 WM.
from a subject metabolizing ethanol, while added acetaldehyde disappears rapidly. The disappearance of the added acetaldehyde is temperature dependent while the incubation temperature has no effect on the level of acetaldehyde formed from ethanol in viva. DISCUSSION Although the decay mechanisms for acetaldehyde added to whole blood are unknown, it has been suggested by several authors that the removal could be due either to metabolism or the binding of acetaldehyde to blood constitutents. Stotz (8) observed that acetaldehyde decayed in whole blood and not in plasma. He found that this decay could be eliminated by tungstic acid deproteinization or inhibited completely for at least 20 min if blood was kept in an ice bath. It was suggested that cellular metabolism was responsible but no study of the mechanisms involved was undertaken. Acetaldehyde has been found to be produced as a product of deoxynucleoside metabolism in human erythrocyte ghosts (9) in vitro. If this production occurs in viva, it would seem likely that blood cells may have some capacity to metabolize the acetaldehyde, preventing the buildup of this toxic compound. Since the distribution of acetaldehyde metabolizing systems is known to be widespread in other mammalian tissues (10) it is possible that such a system could also be present in blood cells. Duritz and Truitt (1 l), using initial blood acetaldehyde levels of 4.5 mM found, as in this study, that the disappearance of acetaldehyde was
176
STOWELL,
0
L
GREENWAY.
8
AND BATT
12
16
20
Time (minutes I FIG. 6. The disappearance of acetaldehyde added to human venous blood samples taken from a subject metabolizing ethanol. Where acetaldehyde was added to blood samples this was carried out immediately after the blood samples were taken. The results are given for: blood incubated at room temperature (&--¤), blood incubated at 37°C (Cl-U), blood incubated at room temperature + 50 pM acetaldehyde (O----O) and blood incubated at 37°C + 50 PM acetaldehyde (C---O). The points represent single determinations with a maximum error of approximately k2.5 pM.
temperature dependent. They assumed that the disappearance was enzyme catalyzed and attempted, unsuccessfully, to inhibit it with HgCl,. However, deproteinization of blood with ZnSO,-Ba(OH), was found to completely inhibit the decay. No further study of the reaction mechanism has been reported by these workers. Malorny et al. (12) studied the disappearance of formaldehyde in blood obtained from humans and dogs and showed that formaldehyde in concentrations ranging from about 0.2 to 0.8 mM, disappeared very rapidly with a half-life of about 10 min. It was also shown that formic acid was formed almost quantitatively from the formaldehyde. Because this reaction was found to be only partially NAD+ dependent, it was suggested that oxidation of formaldehyde may be occurring partly in an aldehyde dehydrogenase catalyzed reaction and partly through a peroxidative reaction involving catalase. The significance of this study in relation to acetaldehyde decay in blood is undefined since formaldehyde was the only aldehyde studied. The results of the experiment illustrated in Fig. 3 showed that
STABILITY
OF
ACETALDEHYDE
IN
BLOOD
177
acetaldehyde was converted, quantitatively, to a soluble nonvolatile product which seemed likely, by analogy, to be acetate. Although the conversion of acetaldehyde to acetate was incomplete when acetaldehyde was added to blood at an initial level of I .O mM, it is probable that some irreversible binding of acetaldehyde to blood proteins occurred at this high acetaldehyde level. Such binding at acetaldehyde levels between 1 and IO IIIM has been described by Gaines et al. (13). Allowing for such binding, the remaining acetaldehyde would seem to have been quantitatively converted to acetate. The observed quantitative conversion of acetaldehyde, at low concentrations, to a soluble nonvolatile product in blood (Fig. 3) makes unlikely the possibility that acetaldehyde binds irreversibly to protein when it is added to blood at concentrations up to 23 PM. The results do not rule out the possibility that acetaldehyde binds irreversibly to a soluble compound which is not precipitated by PCA. However, if such a reaction occurs during the disappearance of acetaldehyde, the unidentified compound would have to exist almost exclusively in the cellular fraction of blood since acetaldehyde disappearance has been shown not to occur in any appreciable extent in plasma. Evidence for the reaction of acetaldehyde in r+vo with soluble compounds containing sulfhydryl groups has been reported ( 14- 17) and the in vitro acetaldehyde disappearance could be due, at least in part, to such a reaction. The rapid disappearance of added acetaldehyde in blood samples containing stable acetaldehyde levels produced from ethanol metabolism in ri~c~, and the differential effects of temperature on the two “types” of acetaldehyde suggest that the acetaldehyde formed in viw may be reversibly bound in such a way that it is not available to react with the acetaldehyde-metabolizing system but can be released by a PCA treatment of blood. Irreversible binding of acetaldehyde to proteins is known to occur at high acetaldehyde concentrations (13, 18, 19) but there is little evidence for a reversible binding taking place. Freundt (20) has suggested that acetaldehyde may bind to blood cell protein by the formation of a Schiff s base. However, no direct evidence for this reaction was presented. Eriksson ut al. (21) have found evidence for the reversible binding of acetaldehyde to rat red blood cells. Acetaldehyde in rat blood was found to be unevenly distributed between the cells and plasma with the acetaldehyde concentration in the plasma decreasing with time when blood samples were stored at 4°C. The decrease was matched by a simultaneous increase in the erythrocyte fraction. Although these findings could not be demonstrated with human blood, there is a possibility that a fixed amount of acetaldehyde might be bound in human blood before samples are withdrawn.
178
STOWELL,
GREENWAY,
AND BATT
If endogenous acetaldehyde is bound in some way it would seem likely that acetaldehyde added to blood would be similarly bound. This is obviously not the case, at least at low acetaldehyde concentrations, and it seems necessary to postulate that the reaction of acetaldehyde with blood in vivo is different from reactions in vitro. It may be concluded that although added acetaldehyde appears to be rapidly and quantitatively metabolized to a nonvolatile compound in human blood, acetaldehyde produced from ethanol in viva, as it appears in peripheral venous blood, is reversibly bound, most likely to proteins. SUMMARY
Acetaldehyde was found to disappear rapidly in human venous blood samples when added in concentrations ranging from 10 to 50 PM. The disappearance was dependent on temperature and the presence of blood cells. Evidence is presented for the conversion of acetaldehyde to acetate. By contrast, acetaldehyde which accumulates in human venous blood samples from the metabolism of ethanol in viva did not disappear under conditions where added acetaldehyde disappeared rapidly. It is suggested that acetaldehyde formed from ethanol in viva may be bound reversibly to blood constituents. REFERENCES 1. Truitt, E. B., Quarf. J. Stud. A/c. 31, 1. (1970). 2. Eriksson, C. J. P.. Sippel, H. W., and Forsander, 0. A., In “The Role of Acetaldehyde in the Actions of Ethanol” (K. 0. Lindros and C. J. P. Eriksson, Eds.). p. 9. Kauppakirjapaino, Helsinki, 1975. 3. Stowell, A. R., Greenway. R. M., and Batt. R. D., Biochem. Med. 18, 392 (1977). 4. Lundquist, F., and Wolthers. H.. Actu Pharm. Tax. 14, 265. (1958). 5. Stowell, A. R., Greenway. R. M., Batt, R. D., and Crow, K. E., Anal. Biochrm. 84, 384. (1978). 6. Couchman. K. G.. M. SC. thesis. Massey Univ., Palmerston North, New Zealand. (1974). 7. Guynn, R. W., and Veech, R. L., Anal. Biochem. 61, 6. (1974). 8. Stotz, E., J. Biol. Chem. 148, 585. (1943). 9. Lionetti, F. J., Fortier, N. L., and Jedziniak. J. A., Proc. Sot. Erp. Biol. Med. 116, 1080. (1964). IO. Deitrich, R. A., Biochem. Pharmucol. 15, 1911. (1966). Il. Duritz. G., and Truitt, E. B., Quurt. J. Stud. A/c. 25, 498. (1%4). 12. Malorny. G., Reitbrock. N.. and Schneider, M.. N.-S. Arch. Pharm. 250, 419. (1965). 13. Gaines, K. C., Salhany, J. M., Tuma. D. J., and Sorrell, M. F.. FEBS Left. 75, 115. (1977). 14. Ammon, H. P. T., Estler, C. J., and Heim, F., In “Biological Aspects of Alcohol, Advances in Mental Science III” (M. K. Roach, W. M. McIsaac, and P. J. Creaven, Eds.) p. 185. Univ. of Texas Press, Austin, (1971). 15. Sprince, H., Parker, C. M., Smith, G. G., and Gonzales, L. J., Agents and Actions 5, 164. (1975).
STABILITY
OF ACETALDEHYDE
IN BLOOD
16. Nagasawa. H. T.. Goon, D. J. W., Constantino. N. V.. and Alexander, C. S., Life 17, 707. (1975). 17. Nagasawa, H. T., Goon, D. J. W., Demaster, E. G.. and Alexander, C. S., Life Sci. 187. (1977). 18. Mohammed, A.. Olcott, H. S., and Fraenkel-Conrat. H.. Arch. Biochem. 24, (1949). 19. Schormuller, J.. Grampp, E., and Belitz, H. D.. Z. Lehensmittel. Unters. Forsch.
179 Sci. 20, 270. 136.
271. (1%8).
20. Freundt. K. J.. Blutakohol. 12, 389. (1975). 21. Eriksson, C. J. P., Sippel. H. W., and Forsander, 0. A., FEBS L&t. 75, 205. ( 1977).
MEDICINE
20, 167-179
(1978)
Stability of Acetaldehyde in Human Blood Samples ALLAN
R. STOWELL,
Department
ROBERT
M. GREENWAY,
of Chemistry, Biochemistry. Palmerston North. Received
January
AND RICHARD
and Biophysics. NeM* Zealand
Massey
D. BATT
University.
12, 1978
It is well known that the measurement of acetaldehyde in ethanolcontaining human blood samples is complicated by the production of acetaldehyde from ethanol during deproteinization prior to assay (l-3). To overcome this problem, analyses can be carried out on separated plasma since the deproteinization of plasma containing ethanol results in the formation of only negligible amounts of acetaldehyde as an artifact (I ,3). However, when recovery experiments with whole blood were performed in this laboratory, acetaldehyde could not be completely recovered if plasma was separated and used for analysis. It was assumed that the added acetaldehyde was either partially metabolized or bound to blood constituents during the time required for separation of the plasma from whole blood. Accordingly, it was decided to determine accurately the rate of acetaldehyde disappearance from human blood during times corresponding to those required to take blood and separate the plasma. From this study, it was found that acetaldehyde present in blood samples as a result of ethanol metabolism in vivo did not decrease at a significant rate under conditions which resulted in the rapid disappearance of acetaldehyde which had been added to blood samples. It is suggested that the acetaldehyde formed in vivo may not be present in blood in a free form. MATERIALS
AND METHODS
Reagents. [ I-14C]Ethanol with a specific activity of 5 I .4 mCi/mmol and a radiochemical purity of 98% was obtained from the Amersham Radiochemical Centre, Bucks., England. [ I-14C]Acetaldehyde having the same specific activity was produced from the [ I-14C]ethanol by enzymic oxidation using horse liver alcohol dehydrogenase (Sigma Chemical Co., 167 OOO6-2944178/0202-0167$02.OO/O Copyright @ 1978 by Academic Press. Inc All rights of reproduction in any form reserved.
168
STOWELL,
GREENWAY,
AND
BATT
St. Louis, MO.). The reaction mixture used for the conversion of ethanol to acetaldehyde was the same as that used in the alcohol assay method of Lundquist and Wolthers (4), with semicarbazide added to trap the acetaldehyde. The [l-l”Clacetaldehyde was stored at -20°C as the crystalline semicarbazone and regenerated before use by steam distillation in the presence of dilute H,SO,. Scintillation grade 1,4-di(2-(5phenyloxazolyl))benzene (POPOP), was obtained from Hopkin and Williams, Chadwell Heath, Essex, England and reagent grade 2,5-diphenyloxazole (PPO) from BDH Chemicals Ltd., Poole, England. Acetaldehyde was redistilled before use and stored at 4°C for up to 6 months. Dilute solutions were freshly prepared when required. The scintillation solvent used for radioactivity determinations was prepared as follows: I vol of Triton Xl00 was mixed with 2 vol of redistilled toluene; PPO and POPOP were dissolved in this mixture to give final concentrations of 4 and 0.2 g/liter, respectively. Determination of radioactivity. Radioactivity determinations were carried out in a Beckman LS 350 liquid scintillation counter and corrections for quenching were made using an external standard. Counting efficiencies for samples were within the range of 84 to 87%. Blood samples. Samples of fresh human venous blood treated with heparin or EDTA were used for all experiments. The blood was obtained from an antecubital vein and, unless otherwise stated, the additions of acetaldehyde to blood or plasma samples were made in such a way that the required reagent concentrations were obtained in samples without increasing their volumes by more than 2%. Human blood samples containing endogenous ethanol and acetaldehyde were obtained from healthy volunteers who had previously consumed 1 g/kg of ethanol as vodka, 1.5 to 2 hr before the blood samples were taken. Acetaldehyde disappearance from different blood or plasma samples was determined by assaying aliquots taken at different times after acetaldehyde was mixed with the samples. The incubation of acetaldehyde with blood or plasma was carried out in all cases in sealed containers, to minimize acetaldehyde losses by evaporation. Sample treatment and assay methods. For all experiments, samples were assayed using the semi- or fully automated enzymic acetaldehyde assay methods described by Stowell et al. (5). Samples to be assayed were diluted with either I vol of ice-cold 1 M perchloric acid (PCA) or 9 vol of ice-cold 0.6 M PCA and then centrifuged. The PCA stopped the acetaldehyde disappearance and stabilized the remaining acetaldehyde levels. The clear, protein-free supernatants obtained were assayed for acetaldehyde. Aqueous acetaldehyde standards were used in all experiments. Blood ethanol levels were determined by a gas chromatographic
STABILITY
OF ACETALDEHYDE
IN BLOOD
169
method involving the analysis of head-space gas equilibrated with PCA supernatants of blood samples. Gas samples were chromatographed on a column of Porapak Q (Waters Associates Inc., Framingham, Mass.) at 130°C with a carrier gas flow rate of 40 ml/min. Acetonitrile was used as an internal standard. This method has been described fully by Couchman (6). In an attempt to identify acetic acid in blood samples by gas chromatography, a column packed with Chromasorb W, AW-DMCS, 60/80 mesh. coated with FFAP stationary phase (obtained from Varian Aerograph, Walnut Creek, Calif.) was used. For all experiments, acetaldehyde levels have been corrected for the small amounts of acetaldehyde detectable in blood samples to which no acetaldehyde had been added. These blank levels were stable for at least 30 min under the experimental conditions employed. Where blood samples containing ethanol were assayed for acetaldehyde, levels of acetaldehyde caused by deproteinization were determined for each blood sample using a correction curve prepared by deproteinizing blank blood samples (obtained before alcohol was consumed) to which ethanol was added. These were assayed in the same way as the samples. This method of correction has been described fully by Stowell et al. (3). The artifact level in each sample taken from subjects metabolizing ethanol was determined from the correction curve after the ethanol levels of the samples had been determined. Acetate determinations were performed using the method of Guynn and Veech (7) modified by using yeast acetyl coenzyme A synthetase (EC 6.2.1.1, obtained from Sigma Chemical Co.) instead of beef heart mitochondrial acetyl coenzyme A synthetase. PCA supernatants were neutralized with K&O, before being assayed for acetate. Use of I-‘4C-labeled acetafdehyde. In an experiment to determine whether acetaldehyde is metabolized by whole blood to a nonvolatile compound or possibly bound to blood protein, [I-‘“Clacetaldehyde was used as follows: Approximately 10 pg of [ I-‘*C]acetaldehyde was distilled into about 10 ml of ice-cold water and to the solution obtained, sufficient carrier acetaldehyde, NaCl, and water were added to produce 50 ml of a 0.9% NaCl solution containing 46 PM acetaldehyde. This solution (20 ml) and a sample of human blood (20 ml) were brought to 37°C in separate sealed containers, mixed, and placed in a sealed flask at 37°C. Aliquots (5 ml) were taken at intervals up to 40 min after mixing and added to ice-cold lM PCA solution (5 ml). A blank sample was prepared by deproteinizing in a similar manner, a 1: 1 (v/v) mixture of blood and 0.9% saline solution which did not contain added acetaldehyde. In a control experiment, I vol of 0.9% saline solution was mixed with 1 vol of the saline/[l-‘*C]acetaldehyde solution described above, the mixture being incubated for 40 min at 37°C. Two samples were taken from the
170
STOWELL,
GREENWAY,
AND
BATT
mixture into 1 vol of ice-cold 1M PCA (as described previously), one at zero time (the time of mixing) and one 40 min later. All PCA-treated samples were centrifuged and different aliquots of the protein-free supernatants were used for: (a) acetaldehyde assays, (b) determination of total radioactivity, and (c) determination of radioactivity in nonvolatile substances. For (b), two l-ml samples of each PCA supernatant were added to 10 ml of scintillation solvent. For (c), 2-ml samples of each supernatant were added to 2 ml of 1.2 M KHCO,, to bring the pH to 7.4. Potassium chlorate was allowed to crystallize out and 2 ml of the neutralized solutions were evaporated to half their volumes at 40°C. The solutions were made up to 2 ml again and duplicate 0.9-ml aliquots were added to 10 ml of scintillation solvent with 0.1 ml of water. RESULTS
Comparison of the rates of disappearance of acetaldehyde in human whole blood and plasma samples at 4°C and 37°C. Acetaldehyde disappearance in blood samples from an initial level of 10 PM is not only dependent on temperature but is also dependent to a large extent, on the presence of blood cells (Fig. 1). Individual variations in blood acetaldehyde disappearance rates. The disappearance of acetaldehyde added to blood to an initial level of 20 PM was studied at 37°C in samples from seven individuals (two males and five females). The results given in Fig. 2 show that the disappearance rate is exponential and there is little individual variation in this rate. Some of the blood samples used for this experiment were heparinized and for others, EDTA was used as the anticoagulant. No significant difference was found between the acetaldehyde disappearance rates in blood samples containing different anticoagulants. Experiments were performed separately on each blood sample which was stored at 4°C until required. The time between the testing of the first and last sample was 5 hr and storage at 4°C for this time did not significantly affect the rate of removal of acetaldehyde. Blank levels found in the seven different blood samples were 0.5, 1.1,4.8, 1. I, 0.7, 1.1, and 1.3 JAM. In all cases the added acetaldehyde was found to decay almost to these blank levels in 24 min. Fate of acetaldehyde added to human blood. When [ l-14C]acetaldehyde was added to blood samples the radioactivity in volatile compounds declined at essentially the same rate as acetaldehyde disappeared, while the counts in nonvolatile compounds increased at the same rate (Fig. 3). It was clear that acetaldehyde was not being converted to any appreciable extent, into a volatile product such as ethanol and the results indicated that acetaldehyde was being converted to a nonvolatile compound such as acetate. If the decay of acetaldehyde in diluted blood was due to irreversible binding to protein then the total counts in PCA supernatants would
STABILITY
OF ACETALDEHYDE
L
a
12
IN BLOOD
16
171
20
Time I minutes1 FIG. I. Comparison of the rates of disappearance of acetaldehyde in human whole blood and plasma samples at 4°C and 37°C. Acetaldehyde was added to whole blood and plasma samples to produce an initial concentration of IO pM and samples were treated and assayed as described in Materials and Methods. The results illustrated were obtained using blood from one individual and the points represent single determinations with a maximum error of 2 0.5 PM. The different samples incubated were: plasma at 4°C (O-O), plasma at 37°C (C-0). whole blood at 4°C (O---O) and whole blood at 37°C (W---W).
be expected to decline at the same rate as the acetaldehyde. However, only a slight decrease in total counts was observed in the PCA supernatants. In the control experiment no significant loss of acetaldehyde or radioactivity occurred during the 40-min incubation period at 37°C. In attempts to identify the nonvolatile radioactive compound formed from acetaldehyde in blood, acidified samples prepared from blood in which radioactive acetaldehyde had reacted were subjected to gas chromatography after the addition of carrier acetic acid. The only peak containing counts corresponded to the position of the standard acetic acid peak but a low count recovery from the column (50-60%) precluded absolute identification of acetate as the sole labeled product. The results given in Fig. 4 show that levels of acetate above the blank blood acetate level are definitely produced when acetaldehyde disappeared in blood samples from an initial level of 1.0 mM. In this experiment, no detectable ethanol appeared in the blood sample to which acetaldehyde was added. The experiment could not be repeated to give significant results using low concentrations of acetaldehyde because the acetate formed represented only a small proportion of the acetate already present in the blood samples.
172
STOWELL,
GREENWAY,
AND BATT
*Oi
Time
( minutes
I
FIG. 2. Individual variations in acetaldehyde disappearance in human whole blood samples at 37°C. The points in the figure represent the means of seven determinations, using blood samples obtained from seven individuals. The bar lines represent the range of values obtained and the figures by each point represent one standard deviation. Acetalddhyde was added to all blood samples to produce an initial concentration of 20 pM and samples were treated and assayed as described in Materials and Methods.
Disappearance of endogenous acetaldehyde in blood samples taken from subjects metabolizing ethanol. Although it was clear that acetaldehyde rapidly disappears when added in vitro to blood containing little or
no endogenous acetaldehyde or ethanol, it was considered necessary to determine whether the same rate of disappearance occurred in blood samples containing acetaldehyde which had been formed from ethanol, in vivo .
STABILITY
OF ACETALDEHYDE
IN BLOOD
173
30
25 1
\ \ \
20”
15-
10-
5-
Time
(minutes
I
FIG. 3. The metabolism of [l-*4C]acetaldehyde by human venous blood in vitro. The conversion of [ I-i4C]acetaldehyde to a nonvolatile product in blood diluted with one volume of 0.9?& NaCl is shown. The results shown are: total radioactivity (ho), volatile radioactivity (U--•),non-volatikradioactivity (C-U), and acetaldehyde (e-0). Acetaldehyde was added to the blood/saline mixture to produce an initial concentration of 23 FM. See Materials and Methods for other experimental details.
Blood samples were taken from subjects metabolizing ethanol and aliquots were deproteinized at intervals prior to estimating acetaldehyde levels. The results of such experiments are shown in Fig. 5. A I:9 dilution of blood aliquots with PCA was used to avoid excessive formation of artifact acetaldehyde (3). The results obtained were in striking contrast to those obtained in the previous experiments. Although the level of acetaldehyde measured above the artifact level appeared to fluctuate throughout the incubation period, no significant disappearance was observed. Since these blood samples contained ethanol and presumably, acetate formed from ethanol, it was considered possible that the presence of these two compounds
174
STOWELL,
GREENWAY,
0
20
10
30
AND BATT
LO
50
60
:
Time I minutes) FIG. 4. Production of acetate during the disappearance of acetaldehyde added to human venous blood. Two aliquots of a single blood sample were incubated at 37°C. To one aliquot, acetaldehyde was added to produce an initial concentration of I .O mM. The disappearance of acetaldehyde (LO) and the appearance of acetate (C-0) in this aliquot were measured. No additions were made to the second blood aliquot and this was used as a control to follow changes in the blank acetate level (S-----m). The error associated with the acetate levels is ~0.03 mM and that associated with the acetaldehyde levels is kO.04 mM.
might be responsible, in some way, for the absence of a rapid disappearance rate in endogenous acetaldehyde levels. The following experiment was therefore carried out to test such a possibility. Disappearance of added acetaldehyde from blood samples taken from subjects metabolizing ethanol. Acetaldehyde was added to blood samples containing acetaldehyde formed form ethanol in vivo and its disappear-
ance was measured. At the same time, the levels of endogenous acetaldehyde in the same blood samples to which no acetaldehyde had been added were also determined. This experiment was performed several times at two different temperatures (37°C and room temperature) using blood samples taken from different individuals. In each case a similar result was obtained. A typical set of results (Fig. 6) shows that in the same blood sample, acetaldehyde formed from ethanol in vivo does not disappear significantly over a period of 20 min after the blood sample has been taken
STABILITY
0
2
OF
L
ACETALDEHYDE
6 Incubation
8
10 Time
175
IN BLOOD
12
1L
( minutes
16
18
20
1
FIG. 5. Stability of endogenous acetaldehyde in venous blood samples taken from subjects metabolizing ethanol. The five blood samples were taken from different individuals and incubated at 37°C. The time refers to the interval from the moment the samples were drawn. The maximum error in each point is approximately ? 2.5 WM.
from a subject metabolizing ethanol, while added acetaldehyde disappears rapidly. The disappearance of the added acetaldehyde is temperature dependent while the incubation temperature has no effect on the level of acetaldehyde formed from ethanol in viva. DISCUSSION Although the decay mechanisms for acetaldehyde added to whole blood are unknown, it has been suggested by several authors that the removal could be due either to metabolism or the binding of acetaldehyde to blood constitutents. Stotz (8) observed that acetaldehyde decayed in whole blood and not in plasma. He found that this decay could be eliminated by tungstic acid deproteinization or inhibited completely for at least 20 min if blood was kept in an ice bath. It was suggested that cellular metabolism was responsible but no study of the mechanisms involved was undertaken. Acetaldehyde has been found to be produced as a product of deoxynucleoside metabolism in human erythrocyte ghosts (9) in vitro. If this production occurs in viva, it would seem likely that blood cells may have some capacity to metabolize the acetaldehyde, preventing the buildup of this toxic compound. Since the distribution of acetaldehyde metabolizing systems is known to be widespread in other mammalian tissues (10) it is possible that such a system could also be present in blood cells. Duritz and Truitt (1 l), using initial blood acetaldehyde levels of 4.5 mM found, as in this study, that the disappearance of acetaldehyde was
176
STOWELL,
0
L
GREENWAY.
8
AND BATT
12
16
20
Time (minutes I FIG. 6. The disappearance of acetaldehyde added to human venous blood samples taken from a subject metabolizing ethanol. Where acetaldehyde was added to blood samples this was carried out immediately after the blood samples were taken. The results are given for: blood incubated at room temperature (&--¤), blood incubated at 37°C (Cl-U), blood incubated at room temperature + 50 pM acetaldehyde (O----O) and blood incubated at 37°C + 50 PM acetaldehyde (C---O). The points represent single determinations with a maximum error of approximately k2.5 pM.
temperature dependent. They assumed that the disappearance was enzyme catalyzed and attempted, unsuccessfully, to inhibit it with HgCl,. However, deproteinization of blood with ZnSO,-Ba(OH), was found to completely inhibit the decay. No further study of the reaction mechanism has been reported by these workers. Malorny et al. (12) studied the disappearance of formaldehyde in blood obtained from humans and dogs and showed that formaldehyde in concentrations ranging from about 0.2 to 0.8 mM, disappeared very rapidly with a half-life of about 10 min. It was also shown that formic acid was formed almost quantitatively from the formaldehyde. Because this reaction was found to be only partially NAD+ dependent, it was suggested that oxidation of formaldehyde may be occurring partly in an aldehyde dehydrogenase catalyzed reaction and partly through a peroxidative reaction involving catalase. The significance of this study in relation to acetaldehyde decay in blood is undefined since formaldehyde was the only aldehyde studied. The results of the experiment illustrated in Fig. 3 showed that
STABILITY
OF
ACETALDEHYDE
IN
BLOOD
177
acetaldehyde was converted, quantitatively, to a soluble nonvolatile product which seemed likely, by analogy, to be acetate. Although the conversion of acetaldehyde to acetate was incomplete when acetaldehyde was added to blood at an initial level of I .O mM, it is probable that some irreversible binding of acetaldehyde to blood proteins occurred at this high acetaldehyde level. Such binding at acetaldehyde levels between 1 and IO IIIM has been described by Gaines et al. (13). Allowing for such binding, the remaining acetaldehyde would seem to have been quantitatively converted to acetate. The observed quantitative conversion of acetaldehyde, at low concentrations, to a soluble nonvolatile product in blood (Fig. 3) makes unlikely the possibility that acetaldehyde binds irreversibly to protein when it is added to blood at concentrations up to 23 PM. The results do not rule out the possibility that acetaldehyde binds irreversibly to a soluble compound which is not precipitated by PCA. However, if such a reaction occurs during the disappearance of acetaldehyde, the unidentified compound would have to exist almost exclusively in the cellular fraction of blood since acetaldehyde disappearance has been shown not to occur in any appreciable extent in plasma. Evidence for the reaction of acetaldehyde in r+vo with soluble compounds containing sulfhydryl groups has been reported ( 14- 17) and the in vitro acetaldehyde disappearance could be due, at least in part, to such a reaction. The rapid disappearance of added acetaldehyde in blood samples containing stable acetaldehyde levels produced from ethanol metabolism in ri~c~, and the differential effects of temperature on the two “types” of acetaldehyde suggest that the acetaldehyde formed in viw may be reversibly bound in such a way that it is not available to react with the acetaldehyde-metabolizing system but can be released by a PCA treatment of blood. Irreversible binding of acetaldehyde to proteins is known to occur at high acetaldehyde concentrations (13, 18, 19) but there is little evidence for a reversible binding taking place. Freundt (20) has suggested that acetaldehyde may bind to blood cell protein by the formation of a Schiff s base. However, no direct evidence for this reaction was presented. Eriksson ut al. (21) have found evidence for the reversible binding of acetaldehyde to rat red blood cells. Acetaldehyde in rat blood was found to be unevenly distributed between the cells and plasma with the acetaldehyde concentration in the plasma decreasing with time when blood samples were stored at 4°C. The decrease was matched by a simultaneous increase in the erythrocyte fraction. Although these findings could not be demonstrated with human blood, there is a possibility that a fixed amount of acetaldehyde might be bound in human blood before samples are withdrawn.
178
STOWELL,
GREENWAY,
AND BATT
If endogenous acetaldehyde is bound in some way it would seem likely that acetaldehyde added to blood would be similarly bound. This is obviously not the case, at least at low acetaldehyde concentrations, and it seems necessary to postulate that the reaction of acetaldehyde with blood in vivo is different from reactions in vitro. It may be concluded that although added acetaldehyde appears to be rapidly and quantitatively metabolized to a nonvolatile compound in human blood, acetaldehyde produced from ethanol in viva, as it appears in peripheral venous blood, is reversibly bound, most likely to proteins. SUMMARY
Acetaldehyde was found to disappear rapidly in human venous blood samples when added in concentrations ranging from 10 to 50 PM. The disappearance was dependent on temperature and the presence of blood cells. Evidence is presented for the conversion of acetaldehyde to acetate. By contrast, acetaldehyde which accumulates in human venous blood samples from the metabolism of ethanol in viva did not disappear under conditions where added acetaldehyde disappeared rapidly. It is suggested that acetaldehyde formed from ethanol in viva may be bound reversibly to blood constituents. REFERENCES 1. Truitt, E. B., Quarf. J. Stud. A/c. 31, 1. (1970). 2. Eriksson, C. J. P.. Sippel, H. W., and Forsander, 0. A., In “The Role of Acetaldehyde in the Actions of Ethanol” (K. 0. Lindros and C. J. P. Eriksson, Eds.). p. 9. Kauppakirjapaino, Helsinki, 1975. 3. Stowell, A. R., Greenway. R. M., and Batt. R. D., Biochem. Med. 18, 392 (1977). 4. Lundquist, F., and Wolthers. H.. Actu Pharm. Tax. 14, 265. (1958). 5. Stowell, A. R., Greenway. R. M., Batt, R. D., and Crow, K. E., Anal. Biochrm. 84, 384. (1978). 6. Couchman. K. G.. M. SC. thesis. Massey Univ., Palmerston North, New Zealand. (1974). 7. Guynn, R. W., and Veech, R. L., Anal. Biochem. 61, 6. (1974). 8. Stotz, E., J. Biol. Chem. 148, 585. (1943). 9. Lionetti, F. J., Fortier, N. L., and Jedziniak. J. A., Proc. Sot. Erp. Biol. Med. 116, 1080. (1964). IO. Deitrich, R. A., Biochem. Pharmucol. 15, 1911. (1966). Il. Duritz. G., and Truitt, E. B., Quurt. J. Stud. A/c. 25, 498. (1%4). 12. Malorny. G., Reitbrock. N.. and Schneider, M.. N.-S. Arch. Pharm. 250, 419. (1965). 13. Gaines, K. C., Salhany, J. M., Tuma. D. J., and Sorrell, M. F.. FEBS Left. 75, 115. (1977). 14. Ammon, H. P. T., Estler, C. J., and Heim, F., In “Biological Aspects of Alcohol, Advances in Mental Science III” (M. K. Roach, W. M. McIsaac, and P. J. Creaven, Eds.) p. 185. Univ. of Texas Press, Austin, (1971). 15. Sprince, H., Parker, C. M., Smith, G. G., and Gonzales, L. J., Agents and Actions 5, 164. (1975).
STABILITY
OF ACETALDEHYDE
IN BLOOD
16. Nagasawa. H. T.. Goon, D. J. W., Constantino. N. V.. and Alexander, C. S., Life 17, 707. (1975). 17. Nagasawa, H. T., Goon, D. J. W., Demaster, E. G.. and Alexander, C. S., Life Sci. 187. (1977). 18. Mohammed, A.. Olcott, H. S., and Fraenkel-Conrat. H.. Arch. Biochem. 24, (1949). 19. Schormuller, J.. Grampp, E., and Belitz, H. D.. Z. Lehensmittel. Unters. Forsch.
179 Sci. 20, 270. 136.
271. (1%8).
20. Freundt. K. J.. Blutakohol. 12, 389. (1975). 21. Eriksson, C. J. P., Sippel. H. W., and Forsander, 0. A., FEBS L&t. 75, 205. ( 1977).