- Email: [email protected]
Rat liver alcohol dehydrogenase
ANALYTICAL
BIOCHEMISTRY
133, 350-36 1 (1983)
Rat Liver Alcohol Dehydrogenase I. Purification and Characterization
FWSHKARAJJ.LADANDHYAML.LEFFERT Department of Medicine, Division of Pharmacology, M-013 H, University of California, San Diego, L.a Jolla, California 92093 Received February 10, 1983 Alcohol dehydrogenase was purified in 14 h from male Fischer-344 rat livers by differential centrifugation, (NH&SOI precipitation, and chromatography over DEAE-Affi-Gel Blue, AffiGel Blue, and AMP-agarose. Following HPLC more than 240-fold purification was obtained. Under denaturing conditions, the enzyme migrated as a single protein band (Mr ~=40,000) on 10% sodium dodecyl sulfate-polyacrylamide gels. Under nondenaturing conditions, the protein eluted from an HPLC I-125 column as a symmetrical peak with a constant enzyme specific activity. When examined by analytical isoelectric focusing, two protein and two enzyme activity bands comigrated closely together (broad band) between pH 8.8 and 8.9. The pure enzyme showed pH optima for activity between 8.3 and 8.8 in buffers of 0.5 M Tris-HCl, 50 mM 2-(Ncyclohexylamino)ethanesulfonic acid (CHES), and 50 mM 3-(cyclohexylamino)- 1-propanesulfonic acid (CAPS), and above pH 9.0 in 50 mM glycyl-glycine. Kinetic studies with the pure enzyme, in 0.5 M Tris-HCl under varying pH conditions, revealed three characteristic ionization constants for activity: 7.4 (pK,); 8.0-8.1 (pKr), and 9.1 (pKs). The latter two probably represent functional groups in the free enzyme; pK, may represent a functional group in the enzyme-NAD+ complex. Pure enzyme also was used to determine kinetic constants at 37°C in 0.5 M Tris-HCl buffer, pH 7.4 (I = 0.2). The values obtained were V,., = 2.21 pM/min/mg enzyme, K,,, for ethanol = 0.156 mM, K,,, for NAD+ = 0.176 mM, and a dissociation constant for NAD+ = 0.306 mM. These values were used to extrapolate the forward rate of ethanol oxidation by alcohol dehydrogenam in vivo. At pH 7.4 and 10 mM ethanol, the rate was calculated to be 2.4 PM/ min/g liver. KEY WORDS: alcohol; alcohol dehydrogenase; purification; characterization; enzyme.
Alcohol is metabolized in mammals primarily by the liver. The major biochemical pathway involves NAD+-dependent alcohol dehydrogenase (alcohol:NAD+ oxidoreductase, EC 1.1.1.1). When the extracellular concentration of ethanol is ~20 mM, alcohol dehydrogenase activity appears to be the ratelimiting step for ethanol metabolism in vivo (1,2). The nature of the rate-limiting step(s) has continued to be a critical and controversial issue in the assessment of ethanol dependence and liver damage (3)-especially in the rat, an experimental animal routinely used for alcohol toxicity and metabolism studies. Rat, horse, and human liver alcohol dehydrogenase have been purified and studied 0003-2697/83 $3.00 Copyright 0 1983 by Academic Press. Inc. All rights of reproduction in any form reserved.
to different degrees (4-8). The horse enzyme is commercially available in a homogeneous form. It has been used for most biochemical and biophysical studies. Much less is known about the rat enzyme. It shows about 80% sequence homology with horse and human enzyme subunits, but its complete sequence has not been determined (9). The current view is that it lacks isozymes, unlike the horse and human liver forms which contain six or more electrophoretically separable components (10,ll). Additional problems encountered during its purification have further limited knowledge of its enzymatic mechanism. For example, it is relatively unstable compared to the horse and human enzymes, owing to sev350
PURIFICATION
OF RAT LIVER ALCOHOL
eral labile sullhydryl groups which in air form S-S bridges that generate multiple molecular forms upon electrophoresis (7,12). In order to study and clarify the molecular basis of regulation of rat liver alcohol dehydrogenase in vivo and in cultured hepatocytes, a rapid procedure to purify rat liver alcohol dehydrogenase is essential. Lange and Valee (6) have used an immobilized 4-substituted derivative of pyrazole, a competitive inhibitor of alcohol dehydrogenase, to obtain highly purified alcohol dehydrogenase from horse, humans, rats, and monkeys. Commercially available affinity resins such as N6-(6-aminohexyl)-AMP-substituted Sepharose and Blue-Sepharose also have been used to purify the horse enzyme (13,14). Here, we report a rapid purification procedure for rat liver alcohol dehydrogenase using sequential chromatography over DEAE-Affi-Gel Blue, AffiGel Blue, and AMP-agarose. Pure enzyme obtained in 14 h was used to determine its kinetic and ionization constants and to estimate the forward rate of ethanol oxidation by alcohol dehydrogenase in vivo. MATERIALS
AND METHODS
Adult male Fischer-344 rats (200 g) were obtained from Charles River Breeding laboratories (Wilmington, Mass.). DEAE-A&Gel Blue, Affi-Gel Blue, and Coomasie brilliant blue R-250 were from Bio-Rad (Richmond, Calif.). Enzyme-grade ammonium sulfate was supplied by Schwarz/Mann (Orangeburg, N. Y.). Acrylamide (enzyme grade), N,N’-methylenebisacrylamide, N,N,iV’,N’-tetramethylethylene diamine, and ammonium persulfate were from Eastman-Kodak (Rochester, N. Y.). SDS’ was from BDH Chemicals Ltd. (Poole, England). Sephadex G- 100, AMP-agarose (5’-AMP attached to crosslinked agarose through N6-amino group with 8 carbon spacer), NAD+ (Grade III), and other ’ Abbreviations used: SDS, sodium dodecyl sulfate; CHES, 2-(Akyclohexylamino)ane sulfonic acid; CAPS, 3-(cyclohexylamino)-l-propane sulfonic acid; BSA, bovine serum albumin.
DEHYDROGENASE
351
routinely used chemicals were from Sigma (St. Louis, MO.). Enzyme purification. Livers from Fischer344 rats were perfused in situ with buffer A (10 I’I’IM Tris-HCI, pH 8.0), removed, blotted onto tissue paper, weighed, and stored at -20°C. Three to five unfrozen or frozen and thawed livers (45-70 g wet wt) were suspended in 2 vol of buffer A (1:2, w:v) and homogenized at full speed for 10 s in a 1Cspeed Waring blender at 4°C. The homogenates were centrifuged at 10,OOOg for 20 min at 4°C in a Sorvall RC2B centrifuge. The supernatants were recovered and centrifuged at 100,OOOg for 60 min at 4°C in a Beckman L2-65B ultracentrifuge. The resulting 100,OOOgsupematants were 33% saturated with ammonium sulfate ( 196 g/liter). These were maintained at pH 8.0 (7) by dropwise addition of 1.0 M Tris-base, stirred for 30 min at 4°C and centrifuged at 20,OOOg for 30 min at 4°C in a Sorvall RC-2B centrifuge. The resulting supematants were pooled and adjusted to 65% saturation with ammonium sulfate (-214 g/liter) and stirred for 30 min at 4°C (pH 8.0). The suspension was centrifuged at 20,OOOg for 20 min at 4°C and the resulting pellets were pooled and suspended in buffer A. This solution was chromatographed over a Sephadex G-100 column (4 X 36 cm) equilibrated at 2 1“C in buffer B ( 10 IIIM Tris-HCl, pH 6.5). Fractions containing enzyme activity were pooled and chromatographed over a DEAE-Affi-Gel Blue column (2.4 X 30 cm) equilibrated at 2 1 “C with buffer B. The column was eluted first with 150 ml of buffer B and then with buffer B containing 10 mM NaCl. The enzyme activity eluted in two peaks: about 30% eluted with buffer B in the unbound fraction (peak l), whereas the remaining activity was eluted with buffer B containing 10 mM NaCl (peak 2). Both active fractions were pooled and chromatographed over an Affi-Gel Blue gel column (2.4 X 30 cm) equilibrated in buffer A at 2 1 “C. This column was washed with 200 ml of buffer A and then with 300 ml of buffer A containing
352
LAD
AND
0.15 M NaCl. The enzyme activity was subsequently eluted with buffer A containing 0.15 M NaCl and 2.5 mM NAD+. Fractions containing enzyme activity were pooled and dialyzed against 4 liters of buffer A for 6-8 h at 4°C. Dialysis tubing (5/8 in.; VWR, San Diego, Calif.) was boiled in 0.1% (w/v) EDTA. Subsequently, it was found that the dialysis step could be omitted by diluting the enzyme preparation 1:3 with buffer A. Prior to affinity chromatography, the AMPagarose gel was swollen in buffer A and packed into a 0.9 X 5-cm column. The column was washed with 40 ml of buffer A before applying the enzyme preparation at 4°C. Loosely bound proteins were removed by washing the column with 50 ml of buffer A and 50 ml of buffer A containing 0.15 M NaCl. About 70% of the bound alcohol dehydrogenase activity was then eluted with buffer A containing 0.15 M NaCl and 5 mM NAD+. The remaining enzyme was eluted with buffer A containing 0.15 M NaCl, 5 mM NAD+, and 5 mM pyrazole. All active fractions were pooled and the enzyme was concentrated with a Millipore immersible CX-30 ultrahltration unit (30,000 M, cutotQ. Prior to final purification with a Waters HPLC system, the insoluble material was removed from the enzyme preparation by centrifugation in Eppendorf microfuge tubes for 5 min at 4°C. Separation of the pure enzyme from NAD+, NADH, pyrazole, and other impurities was achieved over a HPLC Waters I125 protein column (silica-base packing with neutral hydrophilic moieties) with a mobile phase of 50 mM sodium phosphate (pH 7.3) and flow rate of 2 ml/min. Pure enzyme was stored at 4°C in 10 mM Tris-HCl (pH 8.0); about 80% activity was lost progressively over a 96-h period. Enzyme assay and protein determination. Alcohol dehydrogenase activity was measured at 37°C in a Beckman DU-8 spectrophotometer with the kinetics accessory unit by monitoring the rates of NADf conversion to NADH at 340 nm. Assay mixtures (2 ml final volume in 4-ml disposable Ratiolab cuvettes) contained 0.5 M Tris-HCI (pH 7.2) 2.5 mM
LEFFERT
NAD+, and 5 mM ethanol (2,15,16). The enzyme was preincubated in the assay mixture (without ethanol) for lo- 15 min at room temperature and 3 min at 37°C before adding 40 ~1 ethanol (0.25 M) to start the reaction. All reaction times were 10 min at 37°C. The activity was expressed as pM NAD+ reduced/ min/mg protein (pM/min/mg). The activity determinations were judged to be valid only when the rates were linear during this 10-min period. Nonlinearity was observed at enzyme concentrations that generated initial rates above 0.04 hM/min per assay mixture. In addition, when pure NAD+ was used, underestimated rates were obtained ( 16,17). Therefore, only grade III Sigma NAD+ was used throughout these studies. Proteins were determined according to Lowry et al. (18) with BSA as the standard. Electrophoresis. SDS-polyacrylamide slab gel electrophoresis was performed according to the general procedure of Laemmli (19). Separation and stacking gels contained 10 and 2.7% (w/v) polyacrylamide, respectively. Immediately prior to electrophoresis, the samples were denatured by boiling for 1.5 min in 50 mM Tris-HCl buffer (pH 7.0) containing 2% (w/v) SDS and 5% (w/v) 2-mercaptoethanol, and then clarified by centrifugation in an Eppendorf microfuge for 5 min at 4°C. Proteins in the gels were stained with 0.03% (w/v) Coomassie brilliant blue R-250 in 10% (v/v) acetic acid and 25% (v/v) isopropranol. Gels were destained overnight with a solution of 10% (v/v) acetic acid and 10% (v/v) isopropanol. Gels were scanned for stained proteins at 520 nm in a Beckman DU-8 spectrophotometer with a slab gel accessory unit. Some gels also were stained for proteins by the more sensitive silver nitrate technique (20) and scanned with white light. Analytical isoelectric focusing was performed with an LKB Multiphor apparatus at 6°C in commercially prepared 5% (w/v) polyacrylamide gels (pH 3.5-9.5, PAG Plate, LKB). Samples (up to 20 ~1) were applied as a streak directly to the surface of the gel. Cathode and anode electrode buffers were 1 M
PURIFICATION
OF RAT LIVER ALCOHOL
NaOH and I M phosphoric acid, respectively. Separations were performed at a constant power of 30 W/20 cm gel for 2 h. The central lane of each gel then was sliced into 0.2- or OS-cm pieces, and eluted in 1.5 ml water. The pH of the resulting solution was measured at 6°C. Gels were stained for alcohol dehydrogenase activity with a solution containing 72 mg NAD+, 50 mg pnitroblue tetrazolium, 8 mg phenazine methosulfate, and 0.6 ml 95% ethanol per 100 ml of 0.025 M Tris-HCl, pH 8.55. In order to stain proteins [with silver nitrate (20)], the gels were freed of Ampholines at room temperature by washing in 10% (w/v) trichloroacetic acid for 90-l 20 min, fixed with 10% (v/v) glutaraldehyde for 30 mitt, and then washed at 21 “C with several changes of distilled water for 12- 16 h. RESULTS
Enzyme PuriJication Table 1 gives the yields and the relative enrichment of rat liver alcohol dehydrogenase throughout the purification procedure. The procedure was economical since the relatively inexpensive chromatographic resins could be used repeatedly after their regeneration. Two critical column steps involved DEAE-Affi-Gel Blue (pH 6.5) and Affi-Gel Blue (pH 8.0) prior to AMP-agarose affinity chromatography. Both were essential to remove high-molecular-
DEHYDROGENASE
353
weight proteins from the crude preparation and to provide more than a 20-fold increase in purification. Fractions obtained after the AMP-agarose step showed a single protein band on 10% (w/v) SDS-polyacrylamide gels. Such preparations were routinely chromatographed over an HPLC I-125 column to remove NAD+, NADH, pyrazole, and other trace impurities. Roughly a 244-fold purification was achieved. The purity of final enzyme preparations was established as follows. First, the native enzyme eluted as a symmetrical peak from the HPLC column, with a constant specific activity in each of the protein-containing fractions (data not shown). Second, fractions from this peak showed a single protein-staining band on denaturing SDS-polyacrylamide gels. Figures IA and B show a typical gel and its scanning pattern, respectively (proteins were stained with silver nitrate which is 5- 10 times more sensitive than the commonly used Coomassie brilliant blue R-250). Notably, a broad activity band was seen after polyacrylamide slab gel isoelectric focusing (Fig. 2A). When stained for protein (Fig. 2B), a broad band also appeared that comigrated with the activity band (Fig. 2A). Upon closer examination, the broad band appeared to consist of two activity and protein bands (as indicated in Figs. 2A and B by short double lines). The isoelectric point range of
TABLE 1 RATLIVERALCOHOLDEHYDROGENASE PURIFICATION" Fraction Whole homogenate 100,OOOgSupematant 33-65% (NH&SO4 Sephadex G- 100 chromatography DEAE-At&Gel Blue chromatography Affi-Gel Blue chromatography AMP-agarose chromatography HPLC
Yield (%) loo 94.2 81.2 16.9 46.4 23.8 10.6 3.8
-t 7.6 -t 5.1 k 23.5 I 2 1.3 5 10.8 4 0.6 k 0.5
Specific activity (fih4/min/mg protein) 0.009 0.020 0.036 0.039 0.08 0.82 1.3 2.2
f 0.005 k 0.006 f 0.004
f 0.01 It: 0.025 r!z 0.15 f 0.4 t 0.2
’ Alcohol dehydrogenase was purified from frozen rat livers as described under Materials the averages (*SD) of six independent experiments.
Purification (-fold) 1 2.2 4.0 4.3 8.9 91.1 144.4 244.4
and Methods. Results are
354
LAD AND LEFFERT
0.8
0.8
0.0
1
I
I
I
I
0
2
4
6
8
10
DISTANCE (cm1 FIG. 1. SDS-polyacrylamide gel electrophoresis of pure rat liver alcohol dehydrogenase. Pure enzyme (-1 pg) was denatured by boiling for 1.5 min in a 50 mM Tris-HCI (pH 7.0) solution containing 2% (w/ v) SDS and 5% (w/v) 2-mercaptwthanol. The sample was centrifuged at 4°C for 5 min and the supematant was subjected to electrophoresis at 4’C for 4 h under a constant current of 13-14 mA on a 0.1% (w/v) SDS, 10% (w/v) polyacrylamide gel (pH 8.3). The gel was stained for proteins with silver nitrate (A, lane 1) and scanned with white light (B, the relative absorbance is given on the y axis). The apparent peaks seen at 0 and 10 cm are scanning artifacts occurring at the top and bottom ends of the gel, respectively. A contiguous lane of the same gel was used to electrophorese known molecular-weight standards (A, lane 2) whose subunit molecular weights are given at the right. The standards (from top to bottom) were phosphorylase b, BSA, ovalbumin, horse liver alcohol dehydrogenase, carbonic anhydrase, soybean trypsin inhibitor, and lysozyme (bottom two bands).
such doublets was 8.8-8.9. The loss of resolution seen in Figs. 2A and B may have been due to the presence of a high concentration of salt in the enzyme preparation. The activity stain showed similar results but with better resolution when fractions obtained from earlier steps (e.g., after chromatography over AffiGel Blue) were analyzed by isoelectric focusing (Fig. 2C). By comparison, commercial horse liver alcohol dehydrogenase preparations showed five activity bands (Fig. 2D) with isoelectric points of 8.9, 8.8, 8.7, 8.5, and 8.2 (the latter two bands stained for most of the enzymatic activity). Molecular weight. The molecular weight of the denatured enzyme was determined on SDS-polyacrylamide slab gels. From the mo-
bility of known molecular-weight standards (Fig. lA, lane 2), an apparent subunit mass of approximately 40,000 was calculated (lane 1). An identical result was obtained with commercial preparations of pure horse alcohol dehydrogenase (Fig. 1A, lane 2). During purification, alcohol dehydrogenase activity eluted between the void volume and BSA (M, N 66,200) of a Sephadex G-100 column; in addition, retention times of the pure rat liver enzyme and the commercially available horse liver enzyme on an HPLC column were similar (4.5 min). The results suggested that, like the horse enzyme, the mass of the native rat enzyme is =80,000, with two subunits of A4, = 40,000 (9,21,22). E@ct of pH on enzyme activity and kinetic
PURIFICATION
OF RAT LIVER
FIG. 2. Isoelectric focusing of purified and partially purified rat liver alcohol dehydrogenase. Pure enzyme (-2 rg) was subjected to isoelectric focusing at 6°C on commercially prepared 5% (w/v) polyacrylamide slab gel plates containing pH 3-1 I Ampholines. Enzyme activity was stained with tetrazolium salt (A). A contiguous lane in the gel was stained for protein with silver nitrate (B). Proteins (-10 pg) obtained after the Affi-Gel Blue step were subjected to isoelectric focusing and stained for enzyme activity (C). Commercial horse liver alcohol dehydrogenase (-5 pg) was subjected to isoelectric focusing and the gel was stained for enzyme activity (D). The pH gradients were measured as described under Materials and Methods; the isoelectric points of the activity and proteincontaining bands are given under Results.
constants. Figure 3 shows the effect of pH on pure rat liver alcohol dehydrogenase activity in different buffers at 37°C. Optimal activity was evident around pH 8.0-8.4 (or above) in each buffer system. In 0.5 M Tris-HCl, 50 I’nM CHES, and 50 mM CAPS, peak activity occurred between pH 8.3 and 8.7. In 50 tnM Tris-HCl, enzyme activity showed a broad optimum around pH 8.4, whereas in 50 mM glycylglycine the highest activity was seen at pH 9.0. An increase in enzyme activity with increasing pH also was seen in 50 mM sodium phosphate (pH 6.2-7.8). Figure 3 also shows the effect of ionic strength on the pH profile of alcohol dehydrogenase activity. The pure enzyme showed 67% more activity in 0.5 M Tris-HCl buffer (3.7 @i/min/mg protein) than in 50 tnM Tris-
ALCOHOL
DEHYDROGENASE
355
protein). Optimal acHCI (2.25 pM/ITtin/IBg tivity levels in the latter system were equivalent to optimal levels in 50 mM CHES and in 50 mM glycylglycine (2.3-2.5 pM/min/Ing), whereas the lowest activity levels occurred in 50 mM CAPS buffer (0.95 pM/min/mg protein) . To determine enzyme stability at 37°C in different buffers of varying pH, enzyme preparations were incubated for 10 min in each buffer prior to activity measurements in 0.5 M Tris-HCl, pH 7.2. Irreversible activity losses were observed on the alkaline side of the optima (e.g., after incubation at pH levels >8.8, 9.5, and 10 with 50 mM CHES, CAPS, and glycylglycine buffers, respectively). In the case of 50 tnM sodium phosphate, irreversible activity losses of 50% occurred below pH 6.8. Kinetic constants (K, for ethanol and apparent V,,) at different pH values were determined from Lineweaver-Burk double-reciprocal plots at 37°C in 0.5 M Tris-HCl buffer (I = 0.2). Ethanol concentrations were varied from 0.055 to 8.0 mM. Tris-HCl was used since it is an aldehyde-trapping agent (2). To avoid NAD+ concentration effects on the kinetic constants, the reaction mixture contained excess NAD+ (2.5 mM). The pH values were measured initially and at the end of each reaction to ensure that the pH had remained invariant during the assay. The lowest K, (-0.12 f 0.2 mM) and highest V,, (3.4 f 0.3 pM/min/mg protein) values were observed in the pH range 8.1-8.7 (data not shown). Since the K,,, is known to be affected by ionization of both the enzyme and the enzyme-substrate complex (23), the pK, (-log Km) was plotted against pH in order to obtain ionization constants for the enzyme. This is shown in Fig. 4A. Three pK values were seen: 7.4 (pK,), 8.1 (pK& and 9.1 (pK3). According to the rules published (23), the pK1 value estimated from the upward bend should represent the ionization of the enzyme-substrate complex (enzyme * NAD+ * ethanol), whereas the pK2 and pKX values (estimated from the downward bends) should correspond to the ionization constants of the free enzyme and/
356
LAD AND LEFFERT
PH PIG. 3. Effects of pH and of different buffer systems on pure rat liver alcohol dehydrogenase activity. Pure enzyme (4 pg) was pmincubated for 3 min at 37°C in different buffers at the indicated pH values: 0.5 M Tris-HCl (O), 50 mM Tris-HCl (W), 50 mM sodium phosphate (A), 50 mM CHES (A), 50 mM glycylglycine (0), and 50 mhi CAPS (0). The reaction was started by adding 64 ~1 of 0.25 M ethanol solution in a final volume of 2 ml of a solution containing 2.5 mM NAD+. The reaction time was 10 min at 31°C.
or substrate. However, since none of these constants represents an ionization constant of ethanol, the pK values 8.1 and 9.1 appear to reflect the ionization constants of the enzyme in its free state. This interpretation was further supported by plotting the log of velocity against pH at low (Fig. 4B) and at high (Fig. 4C) ethanol concentrations. Studies at low substrate concentrations yield pK(s) of the free enzyme, whereas rate analyses at high substrate concentrations yield pK values for the enzyme-substrate complex (23). It can be seen that, at the low ethanol concentration of 0.055 mM, pK values of 8.0 and 9.1 again were obtained (Fig. 4B), in agreement with results from Fig. 4A. Similarly, at the high ethanol concentration of 8.0 tIIM, a pK value of 7.5 was obtained (Fig. 4C), also in agreement with experimental results shown in Fig. 4A. Steady-State Kinetic Constants Horse liver alcohol dehydrogenase is generally thought to catalyze ethanol oxidation
reactions according to an ordered Bi-Bi mechanism of Theorell and Chance (24,25). On the other hand, Hanes and co-workers have shown that the mechanism may consist of the random addition of NAD+ and ethanol, followed by an ordered sequence for NADH and acetaldehyde formation (26,27). Since little is known about the mechanism of rat liver alcohol dehydrogenase, we used the freshly purified rat enzyme, isolated within 14 h after sacrificing the animals, to determine kinetic constants of the forward reaction (ethanol acetaldehyde). Enzyme activity was measured over an ethanol concentration range of 0.06610 mM, at 0.15-2.5 IIIM NAD+, in 0.5 M TrisHCl (pH 7.4, I = 0.2). Initial velocity data were analyzed with Lineweaver-Burk double-reciprocal plots (Fig. 5). An increase in velocity at higher ethanol concentrations occurred at all NAD+ concentrations, resulting in curvilinear plots of data in the region close to the l/v axis. Similar results were found at all pH values studied
PURIFICATION
OF RAT LIVER
ALCOHOL
DEHYDROGENASE
357
difficult. In one attempt to obviate this difficulty, the data in Fig. 5 were fitted to two component curves by least-squares analyses, resulting in two K,,, values that differed by ~50% (at all NAD+ concentrations). Another attempt to arrive at these constants was based upon the fact that a substrate range of 0.332.0 times the K,,, is theoretically ideal for K,,, determinations by such double-reciprocal plots (28). Because the Km for ethanol is 0. l0.4 mM (see above), more reliable kinetic constants could then be calculated by using rates determined only at low ethanol concentrations (0.066-2.0 mM). These data were analyzed by replotting the slopes (apparent Km/apparent Vmax) and intercepts (l/apparent I’,,) from Fig. 5 against l/[NAD+] (not shown) to obtain the four kinetic constants for the forward reaction. These are given in Table 2. DISCUSSION
In this report, Fischer-344 rat dehydrogenase has been purified using At&Gel Blue (Cibacron coupled resin), DEAE-Affi-Gel FIG. 4. Determination of ionization constants of pure rat liver alcohol dehydrogenase. Pure enzyme (4.5 pg) was assayedat varying ethanol concentrations (0.055-8.0 mM) in 0.5 M Tris-HCl buffers (pH 6-9.4) at 37°C. K,,, for ethanol at different pH values was then calculated by plotting Lineweaver-Burk double-reciprocal plots (data not shown; see Results for further details). The pKRI (-log K,,,) was plotted against pH (A). The log of the velocity of the enzyme reaction determined at a low ethanol concentration (0.055 mM) was plotted as a function of pH (B). The log of the velocity of the enzyme reaction, determined at a saturating ethanol concentration (8.0 mM), was plotted as a function of pH (C). The dashed lines and numbers indicate the assigned pK values according to Dixon (23).
liver alcohol in ~24 h by blue F3GABlue, AMP-
:1. -8
-6
-4
-2
0
2
4
6
8
IO
;2
14
16
I
[ETHANOL]
(6%9.4), although at higher pH values the curvature significantly decreased.’ These results made calculations of kinetic constants 2 Such curves often are seen when multiple enzymes or isozymes catalyze the same reaction or with enzymes that display negative cooperativity, but the reasons underlying present observations are unknown.
FIG. 5. Lineweaver-Burk double-reciprocal plots of pure alcohol dehydrogenase. Pure enzyme (4 pg) was assayed at 37°C in 0.5 M Tris-HC1 buffer, pH 7.4. Enzyme activities were determined at varying ethanol concentrations (0.066-5.0 mM) in the presence of various NAD+ concentrations: 0.15 (O), 0.25 (m), 0.5 (A), 1.25 (Cl), and 2.5 mM (0). Curves were drawn by least-squares analysis (>0.99) of rates determined at ethanol concentrations less than 1.0 mrvt.
358
LAD AND LEFFERT TABLE 2
KINJSTICCONSTANTSOFRJRERATLIVERALCOHOL DEHYDROGENASE~ Constant
Value
Vmar KNAD’ m
2.21 0.176 0.156
,yethhanal m
KNAD+ s
0.306
Units aM/min/mg
protein
mM mM mM
’ The constants were calculated from results shown in Fig. 5. A secondary plot of l/[NAD+] against the paxis intercepts of lines shown in Fig. 5 was used to determine V,,,- and K,,, for NAD+ (KEAm). Km for ethanol (Pmb”‘) and the dissociation constant of NAD+ (KrAw) were obtained by plotting l/[NAD+] against the slopes of curves shown in Fig. 5 (28). To calculate these constants, it was assumed that rat alcohol dehydrogenase follows an ordeted Bi-Bi mechanism like the horse enzyme, in which NAD+ binds to the enzyme before ethanol.
agarose, and HPLC. Rapid isolation was essential for kinetic studies because rat liver alcohol dehydrogenase is more labile than its counterparts from other mammals. Only a 4% yield was obtained, whereas in most previous studies the yield was lo-14% (6,13,14,29). However, if the enzyme preparations were scaled down to 30-35 g wet tissue, then complete purifications were achieved in 12- I4 h at the AMP-agarose chromatography step with 15-20% yields. Although affinity columns of 4- [ 3 -(N- 6 -aminocaproyl)aminopropyl]pyrazole, N6-(6-aminohexyl)-AMP, and Cibacron blue dye coupled to Sepharose were used previously to obtain different degrees of purified enzyme rapidly (6,13,14,29), the 244fold purification we obtained here is the highest purification of rat liver alcohol dehydrogenase reported to date (6,8,30,31). Similarities and differences exist between horse and rat liver alcohol dehydrogenase. We found, as previously reported (7,9,10), that both enzymes have subunit masses of 40,000 Da which, in the native state, appear to be dimers since the native protein eluted between the void volume and BSA (Mr N 66,200) from Bio-Gel P- 150 and Sephadex G- 100 columns (data not shown). The native enzymes also
showed similar retention times on an HPLC I-125 protein column. The isoelectric point of pure rat alcohol dehydrogenase was basic (8.8-8.9, Fig. 2), as are the isoelectric points of isozymes from horse (lo)-all migrate toward the cathode in starch gels or on cellulose acetate plates at pH 8.5-8.6. As seen in Fig. 2, the horse enzyme exists in at least five isoelectric forms, whereas the rat enzyme migrated as a broad band which appeared to exist as a doublet with pi’s between 8.8-8.9. Further work is needed to substantiate that this reflects two distinct isozymes, as opposed to a separation or staining artifact. Although horse and rat alcohol dehydrogenase subunits show about 80% homology in their primary sequences (9), the rat enzyme contains three unique cysteine residues (32). Their oxidation is thought to cause the lability of the rat enzyme, which we were able to minimize in these studies. The pH optima of 8.3-8.7 for purified rat alcohol dehydrogenase enzyme-observed in 0.5 M Tris-HCl, 0.05 M CHES, and 0.05 M CAPS buffers-confirms previous reports (8,30). The rat enzyme pH profile differs from those of monkey (33,34) and human enzymes ( 3 5,36); only atypical human enzymes exhibit pH optima around 8.1 (37) and 8.5-9.0 (36). The ‘mouse enzyme also shows a broad optimum between pH 8.6 and 9.2 (38) like the rat enzyme in glycylglycine buffer (Fig. 4). The pH profiles may explain discrepancies in alcohol dehydrogenase activities reported in various studies with rat liver, because buffers at different pH values and concentrations have been employed in the assays (2,7,8,15). Differences in the maximal levels of alcohol dehydrogenase activity attained at various pH values using pure enzyme were also noted (Fig. 3). Differential stability of enzyme activity under these different pH conditions did not account for these differences. However, further studies are needed to see if the rat enzyme is more active in the presence of higher ionic strength buffers containing CHES, CAPS, and glycylglycine. In addition, other factors, such as temperature, substrate concentration, and
PURIFICATION
OF
RAT
LIVER
the affinity of substrates for the enzyme, modify the effects of pH or enzyme activity. Since high ethanol (8.0 mM) and NAD+ (2.5 mM) concentrations were used in these experiments, changes in the affinity of the enzyme for either substrate probably were not involved. Pure rat liver alcohol dehydrogenase exhibited three specific ionization constants at 37°C in 0.5 M Tris-HCl buffer (Fig. 4). Even though these values were determined indirectly, similarity between the horse and rat enzyme is evident. The functional group of the free enzyme had pK’s of 8.1 (which may or may not be unique to rat enzyme) and 9.1, whereas the pK of the enzyme-substrate complex was 7.4. Comparable constants were reported for the horse liver enzyme, usually from studies performed in phosphate buffer at 25°C (39-46). From kinetic and tritrimetric studies with the horse enzyme, it seems that a functional group exists in the free enzyme that exhibits a pK of 9.2 which then decreases to 7.6 in the enzyme. NAD+ complex (39-41). Similar changes were suggested from fluorescent quenching (42) and calorimetric studies (43). It has been suggested that zinc-bound water, hydrogen bonded to a serine-48/histidine-5 1 “proton-relay system,” represents the actual ionizing structure which gives rise to the observed pK’s of 7.6 and 9.2 (41,46). Since the Ser-48/His-5 1 portion of the rat enzyme sequence may be identical with that of the horse (47), such a structure might be involved in the functioning of rat enzyme. Kinetic constants shown in Table 2 were obtained in 0.5 M Tris-HCl buffer (I = 0.2) pH 7.4, at 37°C. Comparable K,,, values for ethanol of 0.5 and 0.26 mM were obtained by Reynier (30) and Feytmans and Leighton (31) respectively, in phosphate buffer (I = 0.1) at pH 7.0 and 23.5% By contrast, Markovic et al. (7) observed a high K,,, of 2.13 mM for ethanol in 0.1 M glycine buffer (pH 10.0) at 23.5”C. This discrepancy can now be explained since the pure enzyme exhibited a lower K,,, for ethanol in the presence of excess NAD+ except at low pH, when the value was in the range of 2 mM (see Fig. 4A).
ALCOHOL
DEHYDROGENASE
359
The data presented here are insufficient to delineate the specific mechanism of rat liver enzyme. However, the forward rate of the alcohol dehydrogenase reaction (in the absence of products) can be estimated by using the constants in Table 2, because the equation is the same for all casesof ordered Bi-Bi enzymes (1,28). Assuming (as shown for the horse enzyme) that NAD+ binds first, then the forward rate equation can be written as Y,,,[ethanolJ[NAD+] ’ = [ethanol][NAD+] + K,NAD'KzToH + KsToHINAD+] + KzAD’[ethanol]
'
Using the constants from Table 2 and a tissue NAD+ concentration of 0.5 mM (48), the rates of ethanol oxidation at various ethanol concentrations were calculated. They are given in Table 3. The calculations probably overestimated the rates slightly because only the forward reaction was considered. To obtain ethanol oxidation rates from a complete rate equation, kinetic studies in the presence of products are needed. Ethanol oxidation rates from Table 3 can be used to estimate ethanol elimination rates in vim For example, at 10 mM ethanol, oxidation rates of N 1.6 pM/min/mg protein are predicted. In 1 X lo6 freshly isolated hepatocytes from Fischer-344 rats, about 15 PLgof enzyme are present (49). Therefore, 1 X lo6 hepatocytes should oxidize ethanol at a rate of 24 nM/ min. This theoretical metabolic rate is comparable to that obtained experimentally by Sjoblom and Morland (50) using hepatocytes from Wistar rats (8- 15 nM/min/ lo6 cells). Since there are Eli1 X lo* hepatocytes/ g liver, ethanol oxidation rates of 2.4 pM/min/ g liver would be predicted. These values are also in agreement with those of 3.32 f 0.14 fiM/min/g wet wt suggested by Crow et al. (5 1). However, the values in Table 3 were obtained from constants determined at pH 7.4, whereas the pH optimum of the rat enzyme is -8.3-8.9 (Fig. 3). Hence, potentially higher rates of ethanol metabolism might be possible in liver cells if an alkaline microenvironment were available to the enzyme. At this point
LAD AND LEFFERT
360 TABLE 3
F~TWATEDRATESOFETHANOLOXIDATIONBYRAT LIVER ALCOHOL DEHYDROGENASE’ Ethanol bM)
Ethanol oxidation rate (pM/min/mg enzyme)
0.7 1.0
1.000 1.292 1.380
2.0
1.496
10.0 16.0 163.0
1.605 1.616 1.633
0.3
Jiimvall, H. (1973) B&hem. Biophys. Res. Commun. 53, 1096-1101. 13. Andersson, L., Jiimvall, H., Akeson, A., and Mosbach, K. (1974) B&him. Biophys. Acta 364, l-8. A. H. (1979) Biotech. 14. Roy, S. K., and Niikawa, Bioeng. 21,775-785. 15. Lumeng, L., Bosron, W. F., and Li, T.-K. (1979) B&hem. Pharmacol. 28, 1547-l 55 1. 16. Braggins, T. J., and Crow, K. E. (1981) Eur. J. Biochem. 119,633-640. 17. Z&ten, R. N., Jacobsen, C. J., and Nejtek, M. E. (1980) Biochem. Pharmacol. 29, 1973-1976. 18. Lowry, 0. H., Rosebrough, N. H., Fatr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 26512.
LIEthanol oxidation rates were estimated from kinetic constants determined in Table 2 and from the forward reaction rate equation as given in the text [see Refs. ( 1, 28)]. The NAD+ concentration was 0.5 M (52).
kinetic constants or metabolic contributions of individual isozymes are not known. These factors may affect the above extrapolations of ethanol metabolism by the rat liver enzyme. ACKNOWLEDGMENTS
275.
19. Laemmli, U. K. (1970) Nature (London) 227, 680685. 20.
21. 22. 23. 24. 25. 26.
This work was supported by USPHS Grants AM 28392, AA 03504, and AM 28215. 21.
REFERENCES 1. Plapp, B. V. (1975) Advan. Exp. Med. Biol. 56, I?‘109. 2. Crow, K. E., Cornell, N. W., and Veech, R. L. (1977) Alcoholism Clin. Exp. Rex 1, 43-47. 3. Hawkins, R. W., and Kalant, H. (1972) Pharm. Rev. 24,67-157. 4. Bonnichsen, R. K., and Wassen, A. M. (1948) Arch. Biochem. Biophys. 18, 361-363. 5. Mourad, N., and Woronick, C. L. (1967) Arch. Biochem. Biophys. 121,431-439. 6. Lange, L. G., and Vallee, B. L. (1976) Biochemistry 15,468 l-4686. 7. Markovic, O., Theorell, H., and Rao, S. (197 1) Actn Chem. &and. 25, 195-205. 8. Arslanian, M. J., Pascoe, E., and Reinhold, J. G. ( 197 1) B&hem. J. 125, 1039-1047. 9. Jomvall, H., and Markovic, 0. (1972) Eur. J. B&hem. 29, 167-174.
10. Pietruszko, R., and Theorell, H. ( 1979) Arch. B&hem. Biophys. 131, 288-298. 11. Smith, M., Hopkinson, D. A., and Harris, H. (1971) Ann. Human Genet. 34.25 l-27 1.
28. 29. 30. 31. 32. 33.
34. 35. 36. 37. 38. 39.
Morrissey, J. H. (1981) Anal. B&hem. 117, 307310. Ehrenberg, A., and Dalxiel, K. (1958) Acta Chem. &and. 12,405-409. Jiimvall, H. (1970) Eur. J. Biochem. 16,254. Dixon, M. (1953) B&hem. J. 55, 161-170. Theorell, H., and Chance, T. (1951) Ada Chem. Stand. 5, 1127-l 144. Theorell, H., Nygaard, A. P., and Bonnichsen, R. (1955) Acta Chem. Stand. 9, 1148-1165. Wong, J. T.-F., Gurr, P. A., Bronskih, P. M., and Hanes, C. S. (1972) in Analysis and Simulation of Biochemical Systems (Hemker, H. C., and Hess, B., eds.), FEBS Symp. 25, 327-344, North-Holland, Utrecht. Hanes, C. S., Bronskill, P. M., Burr, P. A., and Wong, J. T.-F. (1973) Canad. J. Biochem. 50,1385-1413. Segel, I. H. (1975) Enzyme Kinetics, Wiley, New York. Duncan, R. J. S., Kline, J. E., and Sokoloff, L. (1976) B&hem. J. 153,561-566. Reynier, M. (1979) Acta Chem. Stand. 23, 11191129. Feytmans, E., and Leighton F. (1973) Biochem. Pharmacoi. 22, 349-360. Jiimvall, H. (1972) FEBS L&t. 28, 32-35. Dafeldecker, W. P., Pares, X., Vallee, B. L., Bosron, W. F., and Li, T.-K. (198 1) Biochemistry 20,856861. Dafeldecker, W. P., Meadow, P. E., Pares, X., and Vallee, B. L. (1981) Biochemistry 20,6729-6734. Lange, L. G., Sytkowski, A. J., and Vallee, B. L. (1976) Biochemistry 15, 4687-4693. Bosron, W. F., Li, T.-K., and Vallee, B. L. ( 1980) Proc. Natl. Acad. Sci. USA 77, 5784-5788. von Wartburg, J.-P., and Schurch, P. M. (1968) Ann. N. Y. Acad. Sci. 151, 936-947. Talbot, B. G., and Thirion, J. P. (1979) B&hem. Genet. 17, 807-824. Branden, C.-I., JomvaU, H., Ekland, H., and Furugren, B. (1975) in The Enzymes (Boyer, P., ed.), pp 103190, Academic Press, New York.
PURIFICATION
OF RAT LIVER ALCOHOL
40. Shore, J. D., Gutfreund, H., Brooks, R. L., Santiago, D., and Santiago, P. (1974) Biochemistry 13,41854190. 4 1. Kvassman J., and Pettersson, G. ( 1979) Eur. J. B&hem. 100, 115- 123. 42. Wolfe, J. K., Weidig, C. F., Halvorson, H. R., Shore, J. D., Parker, D. M., and Holbrook, J. J. (1977) J. Biol. Chem. 252, 433-436. 43. Subramanian, S., and Ross, P. D. (1979) J. Biol. Chem. 254,7827-7830. 44. Andersson, P., Kvassman, J., Lindstrom, A., Olden, B., and Pettersson, G. (1981) Eur. J. Biochem. 114, 549-554.
45. Kvassman J., Larsson, A., and Pettersson, G. (1981) Eur. J. Biochem. 114, 555-563. 46. Eklund, H., Nordstrom, B., Zeppezauer, E., Soder-
47.
48. 49. 50. 51.
DEHYDROGENASE
361
lund, G., Ohlsson, I., Bolwe, T., and Branden, C.-I. (1974) FEBS Lett. 44,200--204. JBmvall, H. (1974) in Alcohol and Aldehyde Metabolizing Systems (Thurman, R. G., Yonetani, T., Williamson, J. R., and Chanie, B., eds.), pp. 2332, Academic Press, New York. Bticher, T., Brauser, B., Come, A., Klein, F., Langguth, O., and Sies, H. (1972) Eur. J. Biochem. 27, 301-3117. Lad, P. J., and Leffert, H. L. (1983) Anal. Biochem. 133, 362-372. Sjoblom, M., and Morland, J. (1979) B&hem. Pharmacol. 28, 34 17-3423. Crow, K. E., Cornell, N. W., and Veech, R. L. (1977) in Alcohol and Aldehyde Metabolizing Systems (Thurman, R. G., Williamson, J. R., Drott, H. R., and Chance, B., eds.), Vol. 3 pp. 335-342, Academic Press, New York.
BIOCHEMISTRY
133, 350-36 1 (1983)
Rat Liver Alcohol Dehydrogenase I. Purification and Characterization
FWSHKARAJJ.LADANDHYAML.LEFFERT Department of Medicine, Division of Pharmacology, M-013 H, University of California, San Diego, L.a Jolla, California 92093 Received February 10, 1983 Alcohol dehydrogenase was purified in 14 h from male Fischer-344 rat livers by differential centrifugation, (NH&SOI precipitation, and chromatography over DEAE-Affi-Gel Blue, AffiGel Blue, and AMP-agarose. Following HPLC more than 240-fold purification was obtained. Under denaturing conditions, the enzyme migrated as a single protein band (Mr ~=40,000) on 10% sodium dodecyl sulfate-polyacrylamide gels. Under nondenaturing conditions, the protein eluted from an HPLC I-125 column as a symmetrical peak with a constant enzyme specific activity. When examined by analytical isoelectric focusing, two protein and two enzyme activity bands comigrated closely together (broad band) between pH 8.8 and 8.9. The pure enzyme showed pH optima for activity between 8.3 and 8.8 in buffers of 0.5 M Tris-HCl, 50 mM 2-(Ncyclohexylamino)ethanesulfonic acid (CHES), and 50 mM 3-(cyclohexylamino)- 1-propanesulfonic acid (CAPS), and above pH 9.0 in 50 mM glycyl-glycine. Kinetic studies with the pure enzyme, in 0.5 M Tris-HCl under varying pH conditions, revealed three characteristic ionization constants for activity: 7.4 (pK,); 8.0-8.1 (pKr), and 9.1 (pKs). The latter two probably represent functional groups in the free enzyme; pK, may represent a functional group in the enzyme-NAD+ complex. Pure enzyme also was used to determine kinetic constants at 37°C in 0.5 M Tris-HCl buffer, pH 7.4 (I = 0.2). The values obtained were V,., = 2.21 pM/min/mg enzyme, K,,, for ethanol = 0.156 mM, K,,, for NAD+ = 0.176 mM, and a dissociation constant for NAD+ = 0.306 mM. These values were used to extrapolate the forward rate of ethanol oxidation by alcohol dehydrogenam in vivo. At pH 7.4 and 10 mM ethanol, the rate was calculated to be 2.4 PM/ min/g liver. KEY WORDS: alcohol; alcohol dehydrogenase; purification; characterization; enzyme.
Alcohol is metabolized in mammals primarily by the liver. The major biochemical pathway involves NAD+-dependent alcohol dehydrogenase (alcohol:NAD+ oxidoreductase, EC 1.1.1.1). When the extracellular concentration of ethanol is ~20 mM, alcohol dehydrogenase activity appears to be the ratelimiting step for ethanol metabolism in vivo (1,2). The nature of the rate-limiting step(s) has continued to be a critical and controversial issue in the assessment of ethanol dependence and liver damage (3)-especially in the rat, an experimental animal routinely used for alcohol toxicity and metabolism studies. Rat, horse, and human liver alcohol dehydrogenase have been purified and studied 0003-2697/83 $3.00 Copyright 0 1983 by Academic Press. Inc. All rights of reproduction in any form reserved.
to different degrees (4-8). The horse enzyme is commercially available in a homogeneous form. It has been used for most biochemical and biophysical studies. Much less is known about the rat enzyme. It shows about 80% sequence homology with horse and human enzyme subunits, but its complete sequence has not been determined (9). The current view is that it lacks isozymes, unlike the horse and human liver forms which contain six or more electrophoretically separable components (10,ll). Additional problems encountered during its purification have further limited knowledge of its enzymatic mechanism. For example, it is relatively unstable compared to the horse and human enzymes, owing to sev350
PURIFICATION
OF RAT LIVER ALCOHOL
eral labile sullhydryl groups which in air form S-S bridges that generate multiple molecular forms upon electrophoresis (7,12). In order to study and clarify the molecular basis of regulation of rat liver alcohol dehydrogenase in vivo and in cultured hepatocytes, a rapid procedure to purify rat liver alcohol dehydrogenase is essential. Lange and Valee (6) have used an immobilized 4-substituted derivative of pyrazole, a competitive inhibitor of alcohol dehydrogenase, to obtain highly purified alcohol dehydrogenase from horse, humans, rats, and monkeys. Commercially available affinity resins such as N6-(6-aminohexyl)-AMP-substituted Sepharose and Blue-Sepharose also have been used to purify the horse enzyme (13,14). Here, we report a rapid purification procedure for rat liver alcohol dehydrogenase using sequential chromatography over DEAE-Affi-Gel Blue, AffiGel Blue, and AMP-agarose. Pure enzyme obtained in 14 h was used to determine its kinetic and ionization constants and to estimate the forward rate of ethanol oxidation by alcohol dehydrogenase in vivo. MATERIALS
AND METHODS
Adult male Fischer-344 rats (200 g) were obtained from Charles River Breeding laboratories (Wilmington, Mass.). DEAE-A&Gel Blue, Affi-Gel Blue, and Coomasie brilliant blue R-250 were from Bio-Rad (Richmond, Calif.). Enzyme-grade ammonium sulfate was supplied by Schwarz/Mann (Orangeburg, N. Y.). Acrylamide (enzyme grade), N,N’-methylenebisacrylamide, N,N,iV’,N’-tetramethylethylene diamine, and ammonium persulfate were from Eastman-Kodak (Rochester, N. Y.). SDS’ was from BDH Chemicals Ltd. (Poole, England). Sephadex G- 100, AMP-agarose (5’-AMP attached to crosslinked agarose through N6-amino group with 8 carbon spacer), NAD+ (Grade III), and other ’ Abbreviations used: SDS, sodium dodecyl sulfate; CHES, 2-(Akyclohexylamino)ane sulfonic acid; CAPS, 3-(cyclohexylamino)-l-propane sulfonic acid; BSA, bovine serum albumin.
DEHYDROGENASE
351
routinely used chemicals were from Sigma (St. Louis, MO.). Enzyme purification. Livers from Fischer344 rats were perfused in situ with buffer A (10 I’I’IM Tris-HCI, pH 8.0), removed, blotted onto tissue paper, weighed, and stored at -20°C. Three to five unfrozen or frozen and thawed livers (45-70 g wet wt) were suspended in 2 vol of buffer A (1:2, w:v) and homogenized at full speed for 10 s in a 1Cspeed Waring blender at 4°C. The homogenates were centrifuged at 10,OOOg for 20 min at 4°C in a Sorvall RC2B centrifuge. The supernatants were recovered and centrifuged at 100,OOOg for 60 min at 4°C in a Beckman L2-65B ultracentrifuge. The resulting 100,OOOgsupematants were 33% saturated with ammonium sulfate ( 196 g/liter). These were maintained at pH 8.0 (7) by dropwise addition of 1.0 M Tris-base, stirred for 30 min at 4°C and centrifuged at 20,OOOg for 30 min at 4°C in a Sorvall RC-2B centrifuge. The resulting supematants were pooled and adjusted to 65% saturation with ammonium sulfate (-214 g/liter) and stirred for 30 min at 4°C (pH 8.0). The suspension was centrifuged at 20,OOOg for 20 min at 4°C and the resulting pellets were pooled and suspended in buffer A. This solution was chromatographed over a Sephadex G-100 column (4 X 36 cm) equilibrated at 2 1“C in buffer B ( 10 IIIM Tris-HCl, pH 6.5). Fractions containing enzyme activity were pooled and chromatographed over a DEAE-Affi-Gel Blue column (2.4 X 30 cm) equilibrated at 2 1 “C with buffer B. The column was eluted first with 150 ml of buffer B and then with buffer B containing 10 mM NaCl. The enzyme activity eluted in two peaks: about 30% eluted with buffer B in the unbound fraction (peak l), whereas the remaining activity was eluted with buffer B containing 10 mM NaCl (peak 2). Both active fractions were pooled and chromatographed over an Affi-Gel Blue gel column (2.4 X 30 cm) equilibrated in buffer A at 2 1 “C. This column was washed with 200 ml of buffer A and then with 300 ml of buffer A containing
352
LAD
AND
0.15 M NaCl. The enzyme activity was subsequently eluted with buffer A containing 0.15 M NaCl and 2.5 mM NAD+. Fractions containing enzyme activity were pooled and dialyzed against 4 liters of buffer A for 6-8 h at 4°C. Dialysis tubing (5/8 in.; VWR, San Diego, Calif.) was boiled in 0.1% (w/v) EDTA. Subsequently, it was found that the dialysis step could be omitted by diluting the enzyme preparation 1:3 with buffer A. Prior to affinity chromatography, the AMPagarose gel was swollen in buffer A and packed into a 0.9 X 5-cm column. The column was washed with 40 ml of buffer A before applying the enzyme preparation at 4°C. Loosely bound proteins were removed by washing the column with 50 ml of buffer A and 50 ml of buffer A containing 0.15 M NaCl. About 70% of the bound alcohol dehydrogenase activity was then eluted with buffer A containing 0.15 M NaCl and 5 mM NAD+. The remaining enzyme was eluted with buffer A containing 0.15 M NaCl, 5 mM NAD+, and 5 mM pyrazole. All active fractions were pooled and the enzyme was concentrated with a Millipore immersible CX-30 ultrahltration unit (30,000 M, cutotQ. Prior to final purification with a Waters HPLC system, the insoluble material was removed from the enzyme preparation by centrifugation in Eppendorf microfuge tubes for 5 min at 4°C. Separation of the pure enzyme from NAD+, NADH, pyrazole, and other impurities was achieved over a HPLC Waters I125 protein column (silica-base packing with neutral hydrophilic moieties) with a mobile phase of 50 mM sodium phosphate (pH 7.3) and flow rate of 2 ml/min. Pure enzyme was stored at 4°C in 10 mM Tris-HCl (pH 8.0); about 80% activity was lost progressively over a 96-h period. Enzyme assay and protein determination. Alcohol dehydrogenase activity was measured at 37°C in a Beckman DU-8 spectrophotometer with the kinetics accessory unit by monitoring the rates of NADf conversion to NADH at 340 nm. Assay mixtures (2 ml final volume in 4-ml disposable Ratiolab cuvettes) contained 0.5 M Tris-HCI (pH 7.2) 2.5 mM
LEFFERT
NAD+, and 5 mM ethanol (2,15,16). The enzyme was preincubated in the assay mixture (without ethanol) for lo- 15 min at room temperature and 3 min at 37°C before adding 40 ~1 ethanol (0.25 M) to start the reaction. All reaction times were 10 min at 37°C. The activity was expressed as pM NAD+ reduced/ min/mg protein (pM/min/mg). The activity determinations were judged to be valid only when the rates were linear during this 10-min period. Nonlinearity was observed at enzyme concentrations that generated initial rates above 0.04 hM/min per assay mixture. In addition, when pure NAD+ was used, underestimated rates were obtained ( 16,17). Therefore, only grade III Sigma NAD+ was used throughout these studies. Proteins were determined according to Lowry et al. (18) with BSA as the standard. Electrophoresis. SDS-polyacrylamide slab gel electrophoresis was performed according to the general procedure of Laemmli (19). Separation and stacking gels contained 10 and 2.7% (w/v) polyacrylamide, respectively. Immediately prior to electrophoresis, the samples were denatured by boiling for 1.5 min in 50 mM Tris-HCl buffer (pH 7.0) containing 2% (w/v) SDS and 5% (w/v) 2-mercaptoethanol, and then clarified by centrifugation in an Eppendorf microfuge for 5 min at 4°C. Proteins in the gels were stained with 0.03% (w/v) Coomassie brilliant blue R-250 in 10% (v/v) acetic acid and 25% (v/v) isopropranol. Gels were destained overnight with a solution of 10% (v/v) acetic acid and 10% (v/v) isopropanol. Gels were scanned for stained proteins at 520 nm in a Beckman DU-8 spectrophotometer with a slab gel accessory unit. Some gels also were stained for proteins by the more sensitive silver nitrate technique (20) and scanned with white light. Analytical isoelectric focusing was performed with an LKB Multiphor apparatus at 6°C in commercially prepared 5% (w/v) polyacrylamide gels (pH 3.5-9.5, PAG Plate, LKB). Samples (up to 20 ~1) were applied as a streak directly to the surface of the gel. Cathode and anode electrode buffers were 1 M
PURIFICATION
OF RAT LIVER ALCOHOL
NaOH and I M phosphoric acid, respectively. Separations were performed at a constant power of 30 W/20 cm gel for 2 h. The central lane of each gel then was sliced into 0.2- or OS-cm pieces, and eluted in 1.5 ml water. The pH of the resulting solution was measured at 6°C. Gels were stained for alcohol dehydrogenase activity with a solution containing 72 mg NAD+, 50 mg pnitroblue tetrazolium, 8 mg phenazine methosulfate, and 0.6 ml 95% ethanol per 100 ml of 0.025 M Tris-HCl, pH 8.55. In order to stain proteins [with silver nitrate (20)], the gels were freed of Ampholines at room temperature by washing in 10% (w/v) trichloroacetic acid for 90-l 20 min, fixed with 10% (v/v) glutaraldehyde for 30 mitt, and then washed at 21 “C with several changes of distilled water for 12- 16 h. RESULTS
Enzyme PuriJication Table 1 gives the yields and the relative enrichment of rat liver alcohol dehydrogenase throughout the purification procedure. The procedure was economical since the relatively inexpensive chromatographic resins could be used repeatedly after their regeneration. Two critical column steps involved DEAE-Affi-Gel Blue (pH 6.5) and Affi-Gel Blue (pH 8.0) prior to AMP-agarose affinity chromatography. Both were essential to remove high-molecular-
DEHYDROGENASE
353
weight proteins from the crude preparation and to provide more than a 20-fold increase in purification. Fractions obtained after the AMP-agarose step showed a single protein band on 10% (w/v) SDS-polyacrylamide gels. Such preparations were routinely chromatographed over an HPLC I-125 column to remove NAD+, NADH, pyrazole, and other trace impurities. Roughly a 244-fold purification was achieved. The purity of final enzyme preparations was established as follows. First, the native enzyme eluted as a symmetrical peak from the HPLC column, with a constant specific activity in each of the protein-containing fractions (data not shown). Second, fractions from this peak showed a single protein-staining band on denaturing SDS-polyacrylamide gels. Figures IA and B show a typical gel and its scanning pattern, respectively (proteins were stained with silver nitrate which is 5- 10 times more sensitive than the commonly used Coomassie brilliant blue R-250). Notably, a broad activity band was seen after polyacrylamide slab gel isoelectric focusing (Fig. 2A). When stained for protein (Fig. 2B), a broad band also appeared that comigrated with the activity band (Fig. 2A). Upon closer examination, the broad band appeared to consist of two activity and protein bands (as indicated in Figs. 2A and B by short double lines). The isoelectric point range of
TABLE 1 RATLIVERALCOHOLDEHYDROGENASE PURIFICATION" Fraction Whole homogenate 100,OOOgSupematant 33-65% (NH&SO4 Sephadex G- 100 chromatography DEAE-At&Gel Blue chromatography Affi-Gel Blue chromatography AMP-agarose chromatography HPLC
Yield (%) loo 94.2 81.2 16.9 46.4 23.8 10.6 3.8
-t 7.6 -t 5.1 k 23.5 I 2 1.3 5 10.8 4 0.6 k 0.5
Specific activity (fih4/min/mg protein) 0.009 0.020 0.036 0.039 0.08 0.82 1.3 2.2
f 0.005 k 0.006 f 0.004
f 0.01 It: 0.025 r!z 0.15 f 0.4 t 0.2
’ Alcohol dehydrogenase was purified from frozen rat livers as described under Materials the averages (*SD) of six independent experiments.
Purification (-fold) 1 2.2 4.0 4.3 8.9 91.1 144.4 244.4
and Methods. Results are
354
LAD AND LEFFERT
0.8
0.8
0.0
1
I
I
I
I
0
2
4
6
8
10
DISTANCE (cm1 FIG. 1. SDS-polyacrylamide gel electrophoresis of pure rat liver alcohol dehydrogenase. Pure enzyme (-1 pg) was denatured by boiling for 1.5 min in a 50 mM Tris-HCI (pH 7.0) solution containing 2% (w/ v) SDS and 5% (w/v) 2-mercaptwthanol. The sample was centrifuged at 4°C for 5 min and the supematant was subjected to electrophoresis at 4’C for 4 h under a constant current of 13-14 mA on a 0.1% (w/v) SDS, 10% (w/v) polyacrylamide gel (pH 8.3). The gel was stained for proteins with silver nitrate (A, lane 1) and scanned with white light (B, the relative absorbance is given on the y axis). The apparent peaks seen at 0 and 10 cm are scanning artifacts occurring at the top and bottom ends of the gel, respectively. A contiguous lane of the same gel was used to electrophorese known molecular-weight standards (A, lane 2) whose subunit molecular weights are given at the right. The standards (from top to bottom) were phosphorylase b, BSA, ovalbumin, horse liver alcohol dehydrogenase, carbonic anhydrase, soybean trypsin inhibitor, and lysozyme (bottom two bands).
such doublets was 8.8-8.9. The loss of resolution seen in Figs. 2A and B may have been due to the presence of a high concentration of salt in the enzyme preparation. The activity stain showed similar results but with better resolution when fractions obtained from earlier steps (e.g., after chromatography over AffiGel Blue) were analyzed by isoelectric focusing (Fig. 2C). By comparison, commercial horse liver alcohol dehydrogenase preparations showed five activity bands (Fig. 2D) with isoelectric points of 8.9, 8.8, 8.7, 8.5, and 8.2 (the latter two bands stained for most of the enzymatic activity). Molecular weight. The molecular weight of the denatured enzyme was determined on SDS-polyacrylamide slab gels. From the mo-
bility of known molecular-weight standards (Fig. lA, lane 2), an apparent subunit mass of approximately 40,000 was calculated (lane 1). An identical result was obtained with commercial preparations of pure horse alcohol dehydrogenase (Fig. 1A, lane 2). During purification, alcohol dehydrogenase activity eluted between the void volume and BSA (M, N 66,200) of a Sephadex G-100 column; in addition, retention times of the pure rat liver enzyme and the commercially available horse liver enzyme on an HPLC column were similar (4.5 min). The results suggested that, like the horse enzyme, the mass of the native rat enzyme is =80,000, with two subunits of A4, = 40,000 (9,21,22). E@ct of pH on enzyme activity and kinetic
PURIFICATION
OF RAT LIVER
FIG. 2. Isoelectric focusing of purified and partially purified rat liver alcohol dehydrogenase. Pure enzyme (-2 rg) was subjected to isoelectric focusing at 6°C on commercially prepared 5% (w/v) polyacrylamide slab gel plates containing pH 3-1 I Ampholines. Enzyme activity was stained with tetrazolium salt (A). A contiguous lane in the gel was stained for protein with silver nitrate (B). Proteins (-10 pg) obtained after the Affi-Gel Blue step were subjected to isoelectric focusing and stained for enzyme activity (C). Commercial horse liver alcohol dehydrogenase (-5 pg) was subjected to isoelectric focusing and the gel was stained for enzyme activity (D). The pH gradients were measured as described under Materials and Methods; the isoelectric points of the activity and proteincontaining bands are given under Results.
constants. Figure 3 shows the effect of pH on pure rat liver alcohol dehydrogenase activity in different buffers at 37°C. Optimal activity was evident around pH 8.0-8.4 (or above) in each buffer system. In 0.5 M Tris-HCl, 50 I’nM CHES, and 50 mM CAPS, peak activity occurred between pH 8.3 and 8.7. In 50 tnM Tris-HCl, enzyme activity showed a broad optimum around pH 8.4, whereas in 50 mM glycylglycine the highest activity was seen at pH 9.0. An increase in enzyme activity with increasing pH also was seen in 50 mM sodium phosphate (pH 6.2-7.8). Figure 3 also shows the effect of ionic strength on the pH profile of alcohol dehydrogenase activity. The pure enzyme showed 67% more activity in 0.5 M Tris-HCl buffer (3.7 @i/min/mg protein) than in 50 tnM Tris-
ALCOHOL
DEHYDROGENASE
355
protein). Optimal acHCI (2.25 pM/ITtin/IBg tivity levels in the latter system were equivalent to optimal levels in 50 mM CHES and in 50 mM glycylglycine (2.3-2.5 pM/min/Ing), whereas the lowest activity levels occurred in 50 mM CAPS buffer (0.95 pM/min/mg protein) . To determine enzyme stability at 37°C in different buffers of varying pH, enzyme preparations were incubated for 10 min in each buffer prior to activity measurements in 0.5 M Tris-HCl, pH 7.2. Irreversible activity losses were observed on the alkaline side of the optima (e.g., after incubation at pH levels >8.8, 9.5, and 10 with 50 mM CHES, CAPS, and glycylglycine buffers, respectively). In the case of 50 tnM sodium phosphate, irreversible activity losses of 50% occurred below pH 6.8. Kinetic constants (K, for ethanol and apparent V,,) at different pH values were determined from Lineweaver-Burk double-reciprocal plots at 37°C in 0.5 M Tris-HCl buffer (I = 0.2). Ethanol concentrations were varied from 0.055 to 8.0 mM. Tris-HCl was used since it is an aldehyde-trapping agent (2). To avoid NAD+ concentration effects on the kinetic constants, the reaction mixture contained excess NAD+ (2.5 mM). The pH values were measured initially and at the end of each reaction to ensure that the pH had remained invariant during the assay. The lowest K, (-0.12 f 0.2 mM) and highest V,, (3.4 f 0.3 pM/min/mg protein) values were observed in the pH range 8.1-8.7 (data not shown). Since the K,,, is known to be affected by ionization of both the enzyme and the enzyme-substrate complex (23), the pK, (-log Km) was plotted against pH in order to obtain ionization constants for the enzyme. This is shown in Fig. 4A. Three pK values were seen: 7.4 (pK,), 8.1 (pK& and 9.1 (pK3). According to the rules published (23), the pK1 value estimated from the upward bend should represent the ionization of the enzyme-substrate complex (enzyme * NAD+ * ethanol), whereas the pK2 and pKX values (estimated from the downward bends) should correspond to the ionization constants of the free enzyme and/
356
LAD AND LEFFERT
PH PIG. 3. Effects of pH and of different buffer systems on pure rat liver alcohol dehydrogenase activity. Pure enzyme (4 pg) was pmincubated for 3 min at 37°C in different buffers at the indicated pH values: 0.5 M Tris-HCl (O), 50 mM Tris-HCl (W), 50 mM sodium phosphate (A), 50 mM CHES (A), 50 mM glycylglycine (0), and 50 mhi CAPS (0). The reaction was started by adding 64 ~1 of 0.25 M ethanol solution in a final volume of 2 ml of a solution containing 2.5 mM NAD+. The reaction time was 10 min at 31°C.
or substrate. However, since none of these constants represents an ionization constant of ethanol, the pK values 8.1 and 9.1 appear to reflect the ionization constants of the enzyme in its free state. This interpretation was further supported by plotting the log of velocity against pH at low (Fig. 4B) and at high (Fig. 4C) ethanol concentrations. Studies at low substrate concentrations yield pK(s) of the free enzyme, whereas rate analyses at high substrate concentrations yield pK values for the enzyme-substrate complex (23). It can be seen that, at the low ethanol concentration of 0.055 mM, pK values of 8.0 and 9.1 again were obtained (Fig. 4B), in agreement with results from Fig. 4A. Similarly, at the high ethanol concentration of 8.0 tIIM, a pK value of 7.5 was obtained (Fig. 4C), also in agreement with experimental results shown in Fig. 4A. Steady-State Kinetic Constants Horse liver alcohol dehydrogenase is generally thought to catalyze ethanol oxidation
reactions according to an ordered Bi-Bi mechanism of Theorell and Chance (24,25). On the other hand, Hanes and co-workers have shown that the mechanism may consist of the random addition of NAD+ and ethanol, followed by an ordered sequence for NADH and acetaldehyde formation (26,27). Since little is known about the mechanism of rat liver alcohol dehydrogenase, we used the freshly purified rat enzyme, isolated within 14 h after sacrificing the animals, to determine kinetic constants of the forward reaction (ethanol acetaldehyde). Enzyme activity was measured over an ethanol concentration range of 0.06610 mM, at 0.15-2.5 IIIM NAD+, in 0.5 M TrisHCl (pH 7.4, I = 0.2). Initial velocity data were analyzed with Lineweaver-Burk double-reciprocal plots (Fig. 5). An increase in velocity at higher ethanol concentrations occurred at all NAD+ concentrations, resulting in curvilinear plots of data in the region close to the l/v axis. Similar results were found at all pH values studied
PURIFICATION
OF RAT LIVER
ALCOHOL
DEHYDROGENASE
357
difficult. In one attempt to obviate this difficulty, the data in Fig. 5 were fitted to two component curves by least-squares analyses, resulting in two K,,, values that differed by ~50% (at all NAD+ concentrations). Another attempt to arrive at these constants was based upon the fact that a substrate range of 0.332.0 times the K,,, is theoretically ideal for K,,, determinations by such double-reciprocal plots (28). Because the Km for ethanol is 0. l0.4 mM (see above), more reliable kinetic constants could then be calculated by using rates determined only at low ethanol concentrations (0.066-2.0 mM). These data were analyzed by replotting the slopes (apparent Km/apparent Vmax) and intercepts (l/apparent I’,,) from Fig. 5 against l/[NAD+] (not shown) to obtain the four kinetic constants for the forward reaction. These are given in Table 2. DISCUSSION
In this report, Fischer-344 rat dehydrogenase has been purified using At&Gel Blue (Cibacron coupled resin), DEAE-Affi-Gel FIG. 4. Determination of ionization constants of pure rat liver alcohol dehydrogenase. Pure enzyme (4.5 pg) was assayedat varying ethanol concentrations (0.055-8.0 mM) in 0.5 M Tris-HCl buffers (pH 6-9.4) at 37°C. K,,, for ethanol at different pH values was then calculated by plotting Lineweaver-Burk double-reciprocal plots (data not shown; see Results for further details). The pKRI (-log K,,,) was plotted against pH (A). The log of the velocity of the enzyme reaction determined at a low ethanol concentration (0.055 mM) was plotted as a function of pH (B). The log of the velocity of the enzyme reaction, determined at a saturating ethanol concentration (8.0 mM), was plotted as a function of pH (C). The dashed lines and numbers indicate the assigned pK values according to Dixon (23).
liver alcohol in ~24 h by blue F3GABlue, AMP-
:1. -8
-6
-4
-2
0
2
4
6
8
IO
;2
14
16
I
[ETHANOL]
(6%9.4), although at higher pH values the curvature significantly decreased.’ These results made calculations of kinetic constants 2 Such curves often are seen when multiple enzymes or isozymes catalyze the same reaction or with enzymes that display negative cooperativity, but the reasons underlying present observations are unknown.
FIG. 5. Lineweaver-Burk double-reciprocal plots of pure alcohol dehydrogenase. Pure enzyme (4 pg) was assayed at 37°C in 0.5 M Tris-HC1 buffer, pH 7.4. Enzyme activities were determined at varying ethanol concentrations (0.066-5.0 mM) in the presence of various NAD+ concentrations: 0.15 (O), 0.25 (m), 0.5 (A), 1.25 (Cl), and 2.5 mM (0). Curves were drawn by least-squares analysis (>0.99) of rates determined at ethanol concentrations less than 1.0 mrvt.
358
LAD AND LEFFERT TABLE 2
KINJSTICCONSTANTSOFRJRERATLIVERALCOHOL DEHYDROGENASE~ Constant
Value
Vmar KNAD’ m
2.21 0.176 0.156
,yethhanal m
KNAD+ s
0.306
Units aM/min/mg
protein
mM mM mM
’ The constants were calculated from results shown in Fig. 5. A secondary plot of l/[NAD+] against the paxis intercepts of lines shown in Fig. 5 was used to determine V,,,- and K,,, for NAD+ (KEAm). Km for ethanol (Pmb”‘) and the dissociation constant of NAD+ (KrAw) were obtained by plotting l/[NAD+] against the slopes of curves shown in Fig. 5 (28). To calculate these constants, it was assumed that rat alcohol dehydrogenase follows an ordeted Bi-Bi mechanism like the horse enzyme, in which NAD+ binds to the enzyme before ethanol.
agarose, and HPLC. Rapid isolation was essential for kinetic studies because rat liver alcohol dehydrogenase is more labile than its counterparts from other mammals. Only a 4% yield was obtained, whereas in most previous studies the yield was lo-14% (6,13,14,29). However, if the enzyme preparations were scaled down to 30-35 g wet tissue, then complete purifications were achieved in 12- I4 h at the AMP-agarose chromatography step with 15-20% yields. Although affinity columns of 4- [ 3 -(N- 6 -aminocaproyl)aminopropyl]pyrazole, N6-(6-aminohexyl)-AMP, and Cibacron blue dye coupled to Sepharose were used previously to obtain different degrees of purified enzyme rapidly (6,13,14,29), the 244fold purification we obtained here is the highest purification of rat liver alcohol dehydrogenase reported to date (6,8,30,31). Similarities and differences exist between horse and rat liver alcohol dehydrogenase. We found, as previously reported (7,9,10), that both enzymes have subunit masses of 40,000 Da which, in the native state, appear to be dimers since the native protein eluted between the void volume and BSA (Mr N 66,200) from Bio-Gel P- 150 and Sephadex G- 100 columns (data not shown). The native enzymes also
showed similar retention times on an HPLC I-125 protein column. The isoelectric point of pure rat alcohol dehydrogenase was basic (8.8-8.9, Fig. 2), as are the isoelectric points of isozymes from horse (lo)-all migrate toward the cathode in starch gels or on cellulose acetate plates at pH 8.5-8.6. As seen in Fig. 2, the horse enzyme exists in at least five isoelectric forms, whereas the rat enzyme migrated as a broad band which appeared to exist as a doublet with pi’s between 8.8-8.9. Further work is needed to substantiate that this reflects two distinct isozymes, as opposed to a separation or staining artifact. Although horse and rat alcohol dehydrogenase subunits show about 80% homology in their primary sequences (9), the rat enzyme contains three unique cysteine residues (32). Their oxidation is thought to cause the lability of the rat enzyme, which we were able to minimize in these studies. The pH optima of 8.3-8.7 for purified rat alcohol dehydrogenase enzyme-observed in 0.5 M Tris-HCl, 0.05 M CHES, and 0.05 M CAPS buffers-confirms previous reports (8,30). The rat enzyme pH profile differs from those of monkey (33,34) and human enzymes ( 3 5,36); only atypical human enzymes exhibit pH optima around 8.1 (37) and 8.5-9.0 (36). The ‘mouse enzyme also shows a broad optimum between pH 8.6 and 9.2 (38) like the rat enzyme in glycylglycine buffer (Fig. 4). The pH profiles may explain discrepancies in alcohol dehydrogenase activities reported in various studies with rat liver, because buffers at different pH values and concentrations have been employed in the assays (2,7,8,15). Differences in the maximal levels of alcohol dehydrogenase activity attained at various pH values using pure enzyme were also noted (Fig. 3). Differential stability of enzyme activity under these different pH conditions did not account for these differences. However, further studies are needed to see if the rat enzyme is more active in the presence of higher ionic strength buffers containing CHES, CAPS, and glycylglycine. In addition, other factors, such as temperature, substrate concentration, and
PURIFICATION
OF
RAT
LIVER
the affinity of substrates for the enzyme, modify the effects of pH or enzyme activity. Since high ethanol (8.0 mM) and NAD+ (2.5 mM) concentrations were used in these experiments, changes in the affinity of the enzyme for either substrate probably were not involved. Pure rat liver alcohol dehydrogenase exhibited three specific ionization constants at 37°C in 0.5 M Tris-HCl buffer (Fig. 4). Even though these values were determined indirectly, similarity between the horse and rat enzyme is evident. The functional group of the free enzyme had pK’s of 8.1 (which may or may not be unique to rat enzyme) and 9.1, whereas the pK of the enzyme-substrate complex was 7.4. Comparable constants were reported for the horse liver enzyme, usually from studies performed in phosphate buffer at 25°C (39-46). From kinetic and tritrimetric studies with the horse enzyme, it seems that a functional group exists in the free enzyme that exhibits a pK of 9.2 which then decreases to 7.6 in the enzyme. NAD+ complex (39-41). Similar changes were suggested from fluorescent quenching (42) and calorimetric studies (43). It has been suggested that zinc-bound water, hydrogen bonded to a serine-48/histidine-5 1 “proton-relay system,” represents the actual ionizing structure which gives rise to the observed pK’s of 7.6 and 9.2 (41,46). Since the Ser-48/His-5 1 portion of the rat enzyme sequence may be identical with that of the horse (47), such a structure might be involved in the functioning of rat enzyme. Kinetic constants shown in Table 2 were obtained in 0.5 M Tris-HCl buffer (I = 0.2) pH 7.4, at 37°C. Comparable K,,, values for ethanol of 0.5 and 0.26 mM were obtained by Reynier (30) and Feytmans and Leighton (31) respectively, in phosphate buffer (I = 0.1) at pH 7.0 and 23.5% By contrast, Markovic et al. (7) observed a high K,,, of 2.13 mM for ethanol in 0.1 M glycine buffer (pH 10.0) at 23.5”C. This discrepancy can now be explained since the pure enzyme exhibited a lower K,,, for ethanol in the presence of excess NAD+ except at low pH, when the value was in the range of 2 mM (see Fig. 4A).
ALCOHOL
DEHYDROGENASE
359
The data presented here are insufficient to delineate the specific mechanism of rat liver enzyme. However, the forward rate of the alcohol dehydrogenase reaction (in the absence of products) can be estimated by using the constants in Table 2, because the equation is the same for all casesof ordered Bi-Bi enzymes (1,28). Assuming (as shown for the horse enzyme) that NAD+ binds first, then the forward rate equation can be written as Y,,,[ethanolJ[NAD+] ’ = [ethanol][NAD+] + K,NAD'KzToH + KsToHINAD+] + KzAD’[ethanol]
'
Using the constants from Table 2 and a tissue NAD+ concentration of 0.5 mM (48), the rates of ethanol oxidation at various ethanol concentrations were calculated. They are given in Table 3. The calculations probably overestimated the rates slightly because only the forward reaction was considered. To obtain ethanol oxidation rates from a complete rate equation, kinetic studies in the presence of products are needed. Ethanol oxidation rates from Table 3 can be used to estimate ethanol elimination rates in vim For example, at 10 mM ethanol, oxidation rates of N 1.6 pM/min/mg protein are predicted. In 1 X lo6 freshly isolated hepatocytes from Fischer-344 rats, about 15 PLgof enzyme are present (49). Therefore, 1 X lo6 hepatocytes should oxidize ethanol at a rate of 24 nM/ min. This theoretical metabolic rate is comparable to that obtained experimentally by Sjoblom and Morland (50) using hepatocytes from Wistar rats (8- 15 nM/min/ lo6 cells). Since there are Eli1 X lo* hepatocytes/ g liver, ethanol oxidation rates of 2.4 pM/min/ g liver would be predicted. These values are also in agreement with those of 3.32 f 0.14 fiM/min/g wet wt suggested by Crow et al. (5 1). However, the values in Table 3 were obtained from constants determined at pH 7.4, whereas the pH optimum of the rat enzyme is -8.3-8.9 (Fig. 3). Hence, potentially higher rates of ethanol metabolism might be possible in liver cells if an alkaline microenvironment were available to the enzyme. At this point
LAD AND LEFFERT
360 TABLE 3
F~TWATEDRATESOFETHANOLOXIDATIONBYRAT LIVER ALCOHOL DEHYDROGENASE’ Ethanol bM)
Ethanol oxidation rate (pM/min/mg enzyme)
0.7 1.0
1.000 1.292 1.380
2.0
1.496
10.0 16.0 163.0
1.605 1.616 1.633
0.3
Jiimvall, H. (1973) B&hem. Biophys. Res. Commun. 53, 1096-1101. 13. Andersson, L., Jiimvall, H., Akeson, A., and Mosbach, K. (1974) B&him. Biophys. Acta 364, l-8. A. H. (1979) Biotech. 14. Roy, S. K., and Niikawa, Bioeng. 21,775-785. 15. Lumeng, L., Bosron, W. F., and Li, T.-K. (1979) B&hem. Pharmacol. 28, 1547-l 55 1. 16. Braggins, T. J., and Crow, K. E. (1981) Eur. J. Biochem. 119,633-640. 17. Z&ten, R. N., Jacobsen, C. J., and Nejtek, M. E. (1980) Biochem. Pharmacol. 29, 1973-1976. 18. Lowry, 0. H., Rosebrough, N. H., Fatr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 26512.
LIEthanol oxidation rates were estimated from kinetic constants determined in Table 2 and from the forward reaction rate equation as given in the text [see Refs. ( 1, 28)]. The NAD+ concentration was 0.5 M (52).
kinetic constants or metabolic contributions of individual isozymes are not known. These factors may affect the above extrapolations of ethanol metabolism by the rat liver enzyme. ACKNOWLEDGMENTS
275.
19. Laemmli, U. K. (1970) Nature (London) 227, 680685. 20.
21. 22. 23. 24. 25. 26.
This work was supported by USPHS Grants AM 28392, AA 03504, and AM 28215. 21.
REFERENCES 1. Plapp, B. V. (1975) Advan. Exp. Med. Biol. 56, I?‘109. 2. Crow, K. E., Cornell, N. W., and Veech, R. L. (1977) Alcoholism Clin. Exp. Rex 1, 43-47. 3. Hawkins, R. W., and Kalant, H. (1972) Pharm. Rev. 24,67-157. 4. Bonnichsen, R. K., and Wassen, A. M. (1948) Arch. Biochem. Biophys. 18, 361-363. 5. Mourad, N., and Woronick, C. L. (1967) Arch. Biochem. Biophys. 121,431-439. 6. Lange, L. G., and Vallee, B. L. (1976) Biochemistry 15,468 l-4686. 7. Markovic, O., Theorell, H., and Rao, S. (197 1) Actn Chem. &and. 25, 195-205. 8. Arslanian, M. J., Pascoe, E., and Reinhold, J. G. ( 197 1) B&hem. J. 125, 1039-1047. 9. Jomvall, H., and Markovic, 0. (1972) Eur. J. B&hem. 29, 167-174.
10. Pietruszko, R., and Theorell, H. ( 1979) Arch. B&hem. Biophys. 131, 288-298. 11. Smith, M., Hopkinson, D. A., and Harris, H. (1971) Ann. Human Genet. 34.25 l-27 1.
28. 29. 30. 31. 32. 33.
34. 35. 36. 37. 38. 39.
Morrissey, J. H. (1981) Anal. B&hem. 117, 307310. Ehrenberg, A., and Dalxiel, K. (1958) Acta Chem. &and. 12,405-409. Jiimvall, H. (1970) Eur. J. Biochem. 16,254. Dixon, M. (1953) B&hem. J. 55, 161-170. Theorell, H., and Chance, T. (1951) Ada Chem. Stand. 5, 1127-l 144. Theorell, H., Nygaard, A. P., and Bonnichsen, R. (1955) Acta Chem. Stand. 9, 1148-1165. Wong, J. T.-F., Gurr, P. A., Bronskih, P. M., and Hanes, C. S. (1972) in Analysis and Simulation of Biochemical Systems (Hemker, H. C., and Hess, B., eds.), FEBS Symp. 25, 327-344, North-Holland, Utrecht. Hanes, C. S., Bronskill, P. M., Burr, P. A., and Wong, J. T.-F. (1973) Canad. J. Biochem. 50,1385-1413. Segel, I. H. (1975) Enzyme Kinetics, Wiley, New York. Duncan, R. J. S., Kline, J. E., and Sokoloff, L. (1976) B&hem. J. 153,561-566. Reynier, M. (1979) Acta Chem. Stand. 23, 11191129. Feytmans, E., and Leighton F. (1973) Biochem. Pharmacoi. 22, 349-360. Jiimvall, H. (1972) FEBS L&t. 28, 32-35. Dafeldecker, W. P., Pares, X., Vallee, B. L., Bosron, W. F., and Li, T.-K. (198 1) Biochemistry 20,856861. Dafeldecker, W. P., Meadow, P. E., Pares, X., and Vallee, B. L. (1981) Biochemistry 20,6729-6734. Lange, L. G., Sytkowski, A. J., and Vallee, B. L. (1976) Biochemistry 15, 4687-4693. Bosron, W. F., Li, T.-K., and Vallee, B. L. ( 1980) Proc. Natl. Acad. Sci. USA 77, 5784-5788. von Wartburg, J.-P., and Schurch, P. M. (1968) Ann. N. Y. Acad. Sci. 151, 936-947. Talbot, B. G., and Thirion, J. P. (1979) B&hem. Genet. 17, 807-824. Branden, C.-I., JomvaU, H., Ekland, H., and Furugren, B. (1975) in The Enzymes (Boyer, P., ed.), pp 103190, Academic Press, New York.
PURIFICATION
OF RAT LIVER ALCOHOL
40. Shore, J. D., Gutfreund, H., Brooks, R. L., Santiago, D., and Santiago, P. (1974) Biochemistry 13,41854190. 4 1. Kvassman J., and Pettersson, G. ( 1979) Eur. J. B&hem. 100, 115- 123. 42. Wolfe, J. K., Weidig, C. F., Halvorson, H. R., Shore, J. D., Parker, D. M., and Holbrook, J. J. (1977) J. Biol. Chem. 252, 433-436. 43. Subramanian, S., and Ross, P. D. (1979) J. Biol. Chem. 254,7827-7830. 44. Andersson, P., Kvassman, J., Lindstrom, A., Olden, B., and Pettersson, G. (1981) Eur. J. Biochem. 114, 549-554.
45. Kvassman J., Larsson, A., and Pettersson, G. (1981) Eur. J. Biochem. 114, 555-563. 46. Eklund, H., Nordstrom, B., Zeppezauer, E., Soder-
47.
48. 49. 50. 51.
DEHYDROGENASE
361
lund, G., Ohlsson, I., Bolwe, T., and Branden, C.-I. (1974) FEBS Lett. 44,200--204. JBmvall, H. (1974) in Alcohol and Aldehyde Metabolizing Systems (Thurman, R. G., Yonetani, T., Williamson, J. R., and Chanie, B., eds.), pp. 2332, Academic Press, New York. Bticher, T., Brauser, B., Come, A., Klein, F., Langguth, O., and Sies, H. (1972) Eur. J. Biochem. 27, 301-3117. Lad, P. J., and Leffert, H. L. (1983) Anal. Biochem. 133, 362-372. Sjoblom, M., and Morland, J. (1979) B&hem. Pharmacol. 28, 34 17-3423. Crow, K. E., Cornell, N. W., and Veech, R. L. (1977) in Alcohol and Aldehyde Metabolizing Systems (Thurman, R. G., Williamson, J. R., Drott, H. R., and Chance, B., eds.), Vol. 3 pp. 335-342, Academic Press, New York.