Aug 13, 1986 - (alcohol dehydrogenase substrate specificity/aldehyde reduction). GcRAN MARDH ... methoxyphenyl)-acetaldehyde] were synthesized as their.
Proc. Natl. Acad. Sci. USA Vol. 83, pp. 8908-8912, December 1986 Biochemistry
Human class II (IT) alcohol dehydrogenase has a redox-specific function in norepinephrine metabolism (alcohol dehydrogenase substrate specificity/aldehyde reduction)
GcRAN MARDH, AMY L. DINGLEY, DAVID S. AULD, AND BERT L. VALLEE Center for Biochemical and Biophysical Sciences and Medicine, Harvard Medical School, and the Department of Pathology, Brigham and Women's Hospital, Boston, MA 02115
Contributed by Bert L. Vallee, August 13, 1986
Studies of the function of human alcohol ABSTRACT dehydrogenase (ADH) have revealed substrates that are virtually unique for class II ADH (ir ADH). It catalyzes the formation of the intermediary glycols of norepinephrine metabolism, 3,4-dihydroxyphenylglycol and 4-hydroxy-3-methoxyphenylglycol, from the corresponding aldehydes 3,4-dihydroxymandelaldehyde and 4-hydroxy-3-methoxymandelaldehyde with Km values of 55 and 120 MAM and kct/Km ratios of 14,000 and 17,000 mM'1 min'1; these are from 60- to 210-fold higher than those obtained with class I ADH isozymes. The catalytic preference of class II ADH also extends to benzaldehydes. The kct/Km values for the reduction of benzaldehyde, 3,4-dihydroxybenzaldehyde and 4-hydroxy-3-methoxybenzaldehyde by T ADH are from 9- to 29-fold higher than those for a class I isozyme, P1V2 ADH. Furthermore, the norepinephrine aldehydes are potent inhibitors of alcohol (ethanol) oxidation by if ADH. The high catalytic activity of X ADH-catalyzed reduction of the aldehydes in combination with a possible regulatory function of the aldehydes in the oxidative direction leads to essentially 'unidirectional" catalysis by if ADH. These features and the presence of ir ADH in human liver imply a physiological role for ir ADH in the degradation of circulating epinephrine and norepinephrine.
benzaldehyde (HMBAL) with activated Raney nickel (4). The aldehydes were purified by HPLC (Waters Associates), using a gradient of 10-80% methanol in 10 ,AM sodium phosphate buffer (pH 7.5), on a Waters Radial-PAK NOVAPAK C18 column and dried by rotary evaporation. The products were identified by ammonia CI (chemical ionization) mass spectrometry on a Finnegan MAT 312 mass spectrometer, direct inlet, yielding ions of m/z 183, 200, and 217 for HMMAL that correspond to m/z (M + H)+, (M + NH4)+, and (M + NH4 + NH3)+, respectively. A mass spectrum of DHMAL could not be obtained, presumably due to its instability. However, proton NMR obtained in 2H6dimethyl sulfoxide with an 80 MHz Varian CFT 20 spectrometer gave consistent data for the norepinephrine aldehydes. DHMAL: 9.70(s), 7.23(s), 7.27(d), 6.90(d), 3.28(s). HMMAL: 9.68(s), 7.38(s), 7.30(d), 6.90(d), 3.30(s). ADH Isozymes. Human liver alcohol dehydrogenase isozymes were isolated, purified to homogeneity, and characterized, as described (5-8). The present study was performed with class I (apl, ay,, 8iy, P1jy2, and PijP), class II (Xr), and class III (X) ADH isozymes. Measurement of Kinetic Constants. The kinetics of the synthetic aldehydes were measured within 6 hr of preparation. The aldehyde concentration in the assay solution was assessed spectrophotometrically by allowing the reaction to go to completion. Changes in absorbance at 340 nm were monitored with a Gilford 2600 spectrometer. Assays were buffered with 0.1 M sodium phosphate, pH 6.5 or 7.4, at 250C containing 0.25 mM NADH. pH 6.5 was used for DHMAL to avoid problems of instability that were noted at pH 7.4. Kinetic parameters were calculated from duplicate determinations of initial reaction rates of from 6 to 10 different substrate concentrations between 10-500 AuM. Values of Km and kcat were determined from linear regression analysis of Lineweaver-Burk plots. Since both DHMAL and HMMAL absorb visible radiation at 340 nm with e = 3180 M-1cm-1 and 2130 M-1cm'1, respectively, kcat values for their reductions were calculated with the combined extinction coefficients of the aldehydes and that of NADH, 6220 M-1 cm-l. Materials. HMBAL and DHBAL were purchased from Aldrich, whereas HMPG, DHPG, and NADH were obtained from Sigma.
Ethanol affects human norepinephrine metabolism by increasing the urinary excretion of 4-hydroxy-3-methoxyphenylglycol (HMPG) and decreasing that of the corresponding acid 4-hydroxy-3-methoxymandelic acid (VMA) (1). The ethanol-induced decrease in VMA, which is formed predominantly by oxidation of HMPG (2), could result from a competition between ethanol and HMPG for class I alcohol dehydrogenase (ADH) isozymes (3). To examine the potential involvement of ADH in norepinephrine metabolism further, we have synthesized its intermediary aldehydes and examined their capacity to serve as substrates for the three classes of the ADH isozymes. As expected, class I ADH catalyzes the formation of HMPG and 3,4-dihydroxyphenylglycol (DHPG) from their corresponding aldehydes. However, class II (ir) ADH catalyzes these reactions considerably more efficiently than afty class I isozyme and, most remarkably, ir ADH does not oxidize HMPG or DHPG at all (3).
MATERIALS AND METHODS Preparation of Aldehydes. The norepinephrine aldehydes 3,4-dihydroxymandelaldehyde [DHMAL, 2-hydroxy-2-(3,4dihydroxy-phenyl)-acetaldehyde] and 4-hydroxy-3-methoxymandelaldehyde [HMMAL, 2-hydroxy-2-(4-hydroxy-3methoxyphenyl)-acetaldehyde] were synthesized as their racemates by reduction of the 2-hydroxynitriles of 3,4-dihydroxybenzaldehyde (DHBAL), and 4-hydroxy-3-methoxy-
RESULTS The class I ADH isozymes readily catalyze the reduction of the norepinephrine intermediates DHMAL and HMMAL. Abbreviations: ADH, alcohol dehydrogenase; ALDH, aldehyde dehydrogenase; DHBAL, 3,4-dihydroxybenzaldehyde; DHMAL, 3,4-dihydroxymandelaldehyde; DHPG, 3,4-dihydroxyphenylglycol; HMBAL, 4-hydroxy-3-methoxybenzaldehyde; HMMAL, 4-hydroxy-3-methoxymandelaldehyde; HMPG, 4-hydroxy-3-methoxyphenylglycol; VMA, 4-hydroxy-3-methoxymandelic acid.
The publication costs of this article were defrayed in part by page charge payment. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. §1734 solely to indicate this fact.
8908
Biochemistry: MArdh et al.
Proc. Natl. Acad. Sci. USA 83 (1986)
Table 1. Kinetic parameters for individual class I ADH isozymes with norepinephrine aldehydes Class I isozymes
DHMAL Km
kt
kcat/Km
kcat
HMMAL* Km kcat/Km
22 190 55 160 17 260 ply, 10 62 43 280 181Y2 N.A. N.A. P3 N.A., no activity noted at 10-500 AM substrate and 1.0 /ibM enzyme concentration. Units are as follows: kca, min-'; Km, uM; kcat/Km, mM-1'min-1. Values are for pH 6.5; 250C at 10-500 IM DHMAL and HMMAL. *All conditions were as for DHMAL except the buffer was at pH 7.4.
ay,
apB
110 50 180
8.4 9.6 12
76 190 66 160
4.1 9.0 4.5 12
The kcat/Km values range from 66 to 280 mM-1 min1 (Table 1) and are one to two orders of magnitude higher than those observed for the oxidation ofDHPG and HMPG, 1-10 mM-' min' (3). This difference in reactivity is due mainly to a much lower Km value for the reduction of the aldehydes. Thus the Km values for reduction of HMMAL range from 17 to 55 AM (Table 1) whereas the corresponding values for the oxidation of HMPG are 440-1300 A.M (3). There was no detectable activity at all using the 8i31 isozyme, the least reactive isozyme in the direction of oxidation (kcat/Km, 0.14 mM-1 min-1) (3). Class III ADH did not display any reductive activity over the substrate concentration range, from 10 to 500 ,M. It has also been shown (3) that this isozyme does not have any detectable catalytic activity towards the corresponding glycols either. Class II ADH reduces DHMAL and HMMAL with extreme efficiency. The kcat/Km values, 14,000 and 17,000 mM-1min-1, respectively, are two orders of magnitude higher than those for the class I isozymes (Table 2). This increased catalytic efficiency is due to the markedly higher kcat values for the ir enzyme, 780 and 1980 min- compared to average values of 7 and 10 min- for the class I isozymes toward DHMAL and HMMAL, respectively (Table 2). The Km values for the class I and II isozymes are all in the range from 17 to 180 ,uM (Tables 1 and 2). The products of ir ADH-catalyzed reduction of DHMAL and HMMAL, the glycols DHPG and HMPG, were identified by HPLC and by thin layer chromatography on silica (isobutanol/glacial acetic acid/water; 4:1:1; vol/vol) and compared with authentic samples. The catalytic preference of ir ADH for DHMAL and HMMAL, 2-hydroxylated phenylacetaldehydes, also includes benzaldehyde (9), DHBAL, and HMBAL (vanillin) (Table 3), while aldehydes that lack a carbonyl or hydroxyl Table 2. Kinetic parameters for class I and II isozymes with norepinephrine intermediates Isozyme Substrate class klat/Km kcat Km 55 780 DHMAL* II 14,000 II 120 HMMAL 1980 17,000 100 120 I 10 DHMAL* HMMAL DHPG
I I I
7 10 10
34 2700 2000
220 4 7
HMPG II 520t DHPG 460t II HMPG Class I parameters are the average value for the individual isozymes (Table 1) (3). Units are as follows: k,,, min1; Kin, AM; kcat/Km, mM-' min'. Values are for pH 7.4; 250C. *Values are for pH 6.4. tCalculated from the Haldane relationship (see Results).
Table 3. Kinetic parameters for with benzaldehydes ir ADH
Aldehyde
kt
Km
ir
and
kcat/Km
P1y2
8909
ADH 1Y2
kat
ADH
Km
kcat/Km
0.093 37 1.3 1.3 14 DHBAL 29 0.14 4.1 4.1 29 HMBAL 570 140 4.8 BAL* 45 320 65 450 10 Units are as follows: kt, min-'; Kmi, AM; kcat/Km, ALM-'-min-'. Values are for pH 7.4; 250( at 10-500 ,uM DHBAL and HMBAL. BAL, benzaldehyde. *Data from ref. 11.
adjacent to the phenyl ring do not appear to be substrates for irADH (11). The difference in the capacities of class II (IT) ADH and class I (P1y2) ADH to reduce benzaldehyde and the substituted analogs HMBAL and DHBAL (Table 3) is somewhat less dramatic. All benzaldehydes are good substrates of ir ADH with kcat/Km values from 9- to 20-fold higher than catalysis by 131y2 ADH. The catalytic preference again resides in the higher kcat values for the XT group
isozyme.
vT ADH also oxidizes benzyl and vanillyl alcohol with Km and kcat values similar to those for the corresponding reductions (11). By comparison, the kcat/Km values are 79 and 15 versus 15 and 45 1LM-1 min1 for benzyl and vanillyl alcohol and benzaldehyde and vanillin, respectively. It is quite surprising that Ir ADH apparently oxidizes neither HMPG nor DHPG (4), and we have, therefore, examined this phenomenon further. Apparent equilibrium constants, Keq, for the class I isozyme catalyzed reactions HMPG/HMMAL and DHPG/ DHMAL can be calculated from the Haldane relationship (12), assuming that the dissociation constant for the cofactor does not vary greatly for the different isozymes (Table 4). The calculated mean values of Keq for HMPG, 0.027, and for DHPG, 0.037, are nearly the same. Using these values the kcat/Km values for ir ADH oxidation of DHPG and HMPG are calculated to be 520 and 460 mM- 'min1, respectively (Table 2). These values are two orders of magnitude higher than those observed for the class I isozymes (3) indicating that it should be readily possible to determine the reactivity of the enzyme in the direction of oxidation. However, spectrophotometrically, no catalysis at glycol substrate concentrations from 0.4 gM to 10 mM is noted either at pH 7.4 or 10.0. In contrast, incubation of HMPG with Ir ADH and NAD' in the presence of semicarbazide as an aldehyde trapping agent does yield an HPLC analysis peak corresponding to HMMAL-semicarbazone. In the absence of semicarbazide, aldehyde formation could not be detected and neither aldehyde dimers nor any adducts could be isolated. Both DHPG and HMPG inhibit the ADH-catalyzed oxidation of ethanol. Fig. 1 shows the competitive inhibition by HMPG with a Ki value of 2.2 mM. A Ki value in this range would be reasonable in view of the Km values for the class I isozymes (Table 2). It, therefore, seems that the glycols bind to the enzyme, but turnover is greatly decreased due to a IT
IT
Table 4. Equilibrium constants for the interconversion of norepinephrine glycols and aldehydes DHPG/DHMAL Isozyme HMPG/HMMAL ayl
0.036 0.011
0.032 0.0058
a.81 p3ly,
0.025
0.073
P1Y2
0.036
0.038
Data were calculated as the ratio between kcat/Km values for the oxidation (3) and reduction reactions using the Haldane relationship (12).
8910
Biochemistry: MArdh et al.
Proc. Natl. Acad. Sci. USA 83 (1986) with respect to the present findings, ir ADH has not been found in human brain or in peripheral neural tissue (16). The present data reveal the existence of metabolic substrates that are seemingly unique for ir ADH This isozyme preferentially and essentially "unidirectionally" reduces the two intermediary aldehydes of norepinephrine metabolism, DHMAL and HMMAL, to DHPG and HMPG, respectively. We first detected this substrate specificity by generating the aldehydes in situ with monoamine oxidase acting on norepinephrine or normetanephrine. As both of these crude preparations exhibited high activity toward irADH, the aldehydes were synthesized, purified, and characterized as described above. Earlier studies were limited to the use of partially purified aldehydes prepared by monoamine oxidase-catalyzed deaminations (17), but the authentic analogues do not seem to have been synthesized and studied in this manner. Class I ADH also can catalyze the reduction of DHMAL and HMMAL but at significantly lower rates (Table 1). Such reductions by class I isozymes are reversible as both HMPG and DHPG are good substrates for oxidation by these isozymes (3). In contrast with iTADH, no oxidation of HMPG or DHPG was detected by spectrophotometric determination of the concurrent formation of NADH (3). However, in the presence of an incubation mixture of X ADH, HMPG and semicarbazide, the aldehyde product, is trapped since HPLC analysis detects a peak corresponding to HMMAL-semicarbazone, indicating the ir ADH-catalyzed reactions in norepinephrine metabolism are reversible. Thus, in a situation where the aldehydes are metabolized further by an aldehyde dehydrogenase, ir ADH should be able to catalyze HMPG and DHPG oxidation. Regardless of the mechanism for the decreased catalysis by iT ADH in the oxidative direction, the results suggest an interesting control mechanism for the ADH isozymes. In a situation where the concentration of class I and II isozymes are essentially the same, the inhibition of the class II enzyme will reduce activity toward the norepinephrine intermediates by a factor of 100, i.e., to the level of the class I isozymes. The present studies indicate that this can happen in the direction of oxidation, possibly through noncompetitive in-
1/v
2.
1.0
0.5
[Ethanol],
M
FIG. 1. Inhibition of Tr ADH-catalyzed ethanol oxidation by HMPG. HMPG concentrations were 0 (e), 2.0 (o), and 4.0 (A) mM in assays containing 1.0-8.0 mM ethanol, 2.5 mM NADI, and 0.4 AM enzyme in 0.1 M sodium phosphate, pH 7.40 at 25°C. The Ki value is 2.2 mM. v is expressed as umol of NAD+/hr per mg of protein.
drastic reduction in kcat, possibly through noncompetitive inhibition of glycol oxidation by these aldehydes. In this regard, the aldehydes DHMAL and HMMAL are potent inhibitors of ethanol oxidation (33.3 mM); 23 and 60% inhibition is observed by 10 and 20 ,uM DHMAL, respectively, concentrations much lower than the Km value for this substrate, 55 uM (Table 2).
DISCUSSION The enzymatic and inhibition characteristics of Ir ADH show that they differ from those ofclass I isozymes (13, 14). ITADH does not oxidize methanol or digitoxigenin at all, and ethanol only at much higher concentrations (7, 10), but it preferentially attacks benzyl alcohol and octanol (7). It is much less sensitive to 4-methylpyrazole and testosterone inhibition than the class I enzymes (10, 15). The occurrence of ir ADH seems to be restricted to the human liver, the only organ in which it has been detected (7). Importantly and particularly OH
HO.
CH-CH2
HO J
NH2
NOREPINEPHRINE COMT
MAO
~> CH 3°
OH
I
f 1pCH-CHO
ALDEHYDE (HMMAL) -...
'W ADH and/or ALDEHYDE REDUCTASE
ALDEHYDE DEHYDROGENASE
OH
-&
CH -CH 2H
CHO
HO-t) HMPG
OH
CH-COOH
CH3 0 HO
.1,
CLASS I ADH
ALDEHYDE DEHYDROGENASE
VMA
FIG. 2. Enzymes in norepinephrine metabolism. The formation of HMPG occurs predominantly through the combined action of
catechol-O-methyltransferase (COMT), monoamine oxidase (MAO), and either ir ADH or an aldehyde reductase and that of VMA through HMPG oxidation catalyzed by class I ADH and an aldehyde dehydrogenase. (A similar scheme can be drawn for the formation of DHPG by omitting the COMT step.)
Biochemistry: Ma'rdh et al. Table 5. Kinetic parameters for DHMAL toward ig ADH and ALDH Enzyme Km kct/Km kct r ADH 780 55 14,000 8 Cytosolic isozyme of ALDH* 22t 2,800 Mitochondrial isozyme of ALDH* lot 18 600 Units are as follows: kc.,, min'; Km, AM; kcat/Km, AM-'-min'l. *Data from ref. 29. tCalculated from V, values and a M, of 200,000.
hibition by the product and/or substrate of the reaction. Other metabolites could of course control the reductive direction in a similar manner. This might provide a means of regulating the reversibility of these reactions. The mechanism of the inhibition by the norepinephrine intermediary glycols and aldehydes clearly calls for further investigation. The reduction of aldehydes in intermediary norepinephrine metabolism has been attributed to various aldehyde reductases (18) (Fig. 2), though this conclusion remains somewhat questionable owing to the uncertain specificity of the reductases (19). There is little information regarding the identity of reductases that might be involved in catecholamine metabolism of human tissues, though some studies have indicated an involvement of ADH and/or aldehyde reductases (20, 21). Since NAD+/NADH is the cofactor for ADH while NADP+/NADPH activates the reductases, reduction by ADH under physiological conditions has been considered improbable (20, 22). The present Km values for ir ADH reduction of DHMAL and HMMAL are similar to those obtained with aldehyde reductases (17, 20, 21, 23, 24). None of the earlier studies were performed with synthetic and well-characterized aldehydes; those examined were generated in situ by monoamine oxidase-catalyzed deamination of the corresponding amines. In addition the Vma,, values reported probably constitute significant overestimates since they were not corrected for the absorption of visible radiation at 340 nm due to the mandelaldehydes. Since both ADH and aldehyde reductases are cytosolic enzymes (25), they qualify for the degradation of norepinephrine. The majority of the norepinephrine metabolites are formed outside the brain (26-28), and it is, therefore, reasonable to postulate that both HMPG and DHPG are formed in the liver, the only organ in which irADH occurs (7). The high specificity of ir ADH towards the mandelaldehydes suggests an in vivo role in the formation of HMPG and DHPG from circulating epinephrine and norep-
inephrine. Once the intermediary aldehydes are formed by monoamine oxidase-catalyzed deamination they can be reduced to glycols or oxidized to acids (Fig. 2). As most VMA is formed by the oxidation by HMPG (2), the fate of the aldehydes is primarily reductive. This implies either that aldehyde reduction is more efficient than aldehyde oxidation or that there exists a compartmentalization in tissue localization of substrates and enzymes or intracellular transport mechanisms of substrates to prevent the newly formed aldehydes from being oxidized by an aldehyde dehydrogenase (ALDH). The present kinetic data show that ir ADH is more efficient in DHMAL turnover than either of the two ALDH isozymes isolated from human liver (Table 5) (29) or the class I ADH isozymes (Table 2). At equal enzyme concentrations ir ADH is about 5 and 20 times more efficient than the cytosolic and mitochondrial ALDH isozymes, respectively, and 100 times more efficient than the class I ADH isozymes. This is consistent with a preferentially reductive metabolism of the aldehydes (2). The further fate of the glycols are conjugation (sulfate and glucuronide) or oxidation to the corresponding acids (Fig. 2). The first step of the latter reaction is thermodynamically favored (Table 4) and occurs readily with class
Proc. Natl. Acad. Sci. USA 83 (1986)
8911
I but not with class II ADH (3). It is likely that this reaction and the further oxidation to acids takes place at a location different from that where the glycols are formed. Under suitable conditions the ADH isozymes may operate in the formation of alcohol intermediates of norepinephrine metabolism, even when the NAD+/NADH ratio at the site of enzyme action does not favor reduction by ADH. Thus, a change in the redox state of the liver would be expected to enhance aldehyde reduction by ir ADH when NADH is provided by concomitant oxidation of ethanol. Alternatively, ir ADH-catalyzed ethanol oxidation/aldehyde reduction (30) could occur as a coupled reaction when the cofactor does not dissociate from the enzyme. Since dissociation of NADH is rate limiting (7), a mechanism where the aldehyde binds instantaneously to the enzyme-NADH complex on formation during ethanol oxidation would suggest stimulation of HMPG and DHPG formation. During ethanol metabolism their further metabolism (Fig. 2) is inhibited by competition of ethanol for class I ADH (3). Hence, the observed changes in metabolic patterns of norepinephrine subsequent to ethanol intake (1) may be explained partially by both an activated Ir ADH catalysis and competitive inhibition of class I ADH. The authors thank Dr. Maurice Morelock who made preliminary observations relevant to this manuscript. Mass spectra were obtained at the Mass Spectrometry facility of the Harvard School of Public Health. This work was supported by a grant from the Samuel Bronfman Foundation, Inc., with funds provided by Joseph E. Seagram and Sons, Inc. A.L.D. was supported by Postdoctoral Fellowship IFJ2HL06572-01 from the National Institutes of Health. 1. Davis, V. E., Brown, H., Huff, J. A. & Cashaw, J. E. (1967) J. Lab. Clin. Med. 69, 787-799. 2. MArdh, G. & AnggArd, E. (1984) J. Neurochem. 42, 43-46. 3. MArdh, G., Luehr, C. A. & Vallee, B. L. (1985) Proc. Nad. Acad. Sci. USA 82, 4979-4982. 4. Tinapp, P. (1971) Chem. Ber. 104, 2266-2272. 5. Lange, L. G. & Vallee, B. L. (1976) Biochemistry 15, 4681-4693. 6. Wagner, F. W., Burger, A. R. & Vallee, B. L. (1983) Biochemistry 22, 1857-1863. 7. Ditlow, C. C., Holmquist, B., Morelock, M. M. & Vallee, B. L. (1984) Biochemistry 23, 6363-6368. 8. Wagner, F. W., Pares, X., Holmquist, B. & Vallee, B. L. (1984) Biochemistry 23, 2193-2199. 9. Deetz, J. S., Luehr, C. A. & Vallee, B. L. (1984) Biochemistry 23, 6822-6828. 10. Li, T.-K., Bosron, W. F., Dafeldecker, W. P., Lange, L. G. & Vallee, B. L. (1977) Proc. Nati. Acad. Sci. USA 74, 43784381. 11. M&rdh, G. & Vallee, B. L. (1986) Biochemistry 25, in press. 12. Segel, I. H. (1975) in Enzyme Kinetics (Wiley, New York), pp. 34-37. 13. Bosron, W. F., Li, T.-K., Lange, L. G., Dafeldecker, W. P. & Vallee, B. L. (1977) Biochem. Biophys. Res. Commun. 74, 85-91. 14. Bosron, W. F., Li, T.-K., Dafeldecker, W. P. & Vallee, B. L. (1979) Biochemistry 18, 1101-1105. 15. MArdh, G., Falchuk, K. H., Auld, D. S. & Vallee, B. L. (1986) Proc. Nati. Acad. Sci. USA 83, 2836-2840. 16. Beisswenger, T., Holmquist, B. & Vallee, B. L. (1985) Proc. Nati. Acad. Sci. USA 82, 8366-8373. 17. Tabakoff, B., Anderson, R. & Alivisatos, S. G. A. (1973) Mol. Pharmacol. 9, 428-437. 18. Tipton, K. F., Housley, M. D. & Turner, A. J. (1977) in Essays in Neurochemistry and Neuropharmacology, eds. Youdim, M. B. H., Lovenberg, W., Sharman, D. F. & Lagnado, J. R. (Wiley, New York), Vol. 1, pp. 104-138. 19. Flynn, T. G. (1982) Biochem. Pharmacol. 31, 2705-2712. 20. Wermuth, B. & Munch, J. D. B. (1979) Biochem. Pharmacol. 28, 1431-1433. 21. von Wartburg, J. P. & Wermuth, B. (1982) Methods Enzymol. 89, 506-513. 22. Tabakoff, B. & DeLeon-Jones, F. (1983) in MHPG: Basic Mechanisms and Psychopathology, ed. Maas, J. W. (Aca-
8912
Biochemistry: Mirdh et al.
demic, New York), pp. 1-18. 23. Duncan, R. J. S. & Sourkes, T. L. (1974) J. Neurochem. 22, 663-669. 24. Turner, A. J., Illingworth, J. A. & Tipton, K. F. (1974) Biochem. J. 144, 353-360. 25. Ryle, C. M. & Tipton, K. F. (1981) Biochem. J. 197, 715-720. 26. Mirdh, G., Sjoquist, B. & AnggArd, E. (1981) J. Neurochem. 36, 1181-1185.
Proc. Nati. Acad. Sci. USA 83 (1986) 27. Hoeldtke, R. D., Cilmi, K. M., Reichard, G. A., Bodne, G. & Owen, 0. E. (1984) J. Lab. Clin. Med. 101, 772-782. 28. Kopin, I. J., Jimmerson, D. C., Markey, S. P., Ebert, M. H. & Polinsky, R. J. (1984) Pharmacopsychiatry 17, 3-8. 29. MacKerrell, A. D., Jr., Blatter, E. E. & Pietruszko, R. (1986) Alcohol. Clin. Exp. Res. 10, 266-270. 30. Gupta, N. K. & Robinson, W. G. (1966) Biochim. Biophys. Acta 118, 431-434.
