Camp. Biochem. Physiol.
0305-0491/9153.00+ 0.00 PergamonPressplc
Vol. IOOB, No. 2, pp. 321-327,1991
Printedin GreatBritain
ALDEHYDE DEHYDROGENASE (EC 1.2.1.3): COMPARISON OF SUBCELLULAR LOCALIZATION OF THE THIRD ISOZYME THAT DEHYDROGENATES y -AMINOBUTYRALDEHYDE IN RAT, GUINEA PIG AND HUMAN LIVER WOJCIECHAMBROZIAK,GLORIAKURYS and REGINAPETRUSZKO* Center of Alcohol Studies, Rutgers University, Piscataway, NJ 0885-69, (Tel: 908 932 3595); (Fax: 908 932 5944)
USA
(Received 25 March 1991) Subcellular fractionation of rat, guinea pig and human livers showed that aldehyde dehydrogenase metabolizing y-aminobutyraldehyde was exclusively localized in the cytoplasmic fraction
Ahatract-1.
in all three mammalian species. 2. Total y-aminobutyraldehyde activity of aldehyde dehydrogenase was found to be ca 0.41, 0.3 and 0.24 pmol NADH min-’ g-l tissue, respectively in rat, guinea pig and human liver, with more than 95%
of activity in the cytoplasm. 3. Partially purified cytoplasmic isozyme from rat liver showed similar chromatographic behavior and kinetic properties to the E3 isozyme isolated from human liver. 4. The rat isozyme was insensitive to disulfiram (40pM) and to magnesium (16OpM) and had &, values of 5 PM (PH 7.4) for y-aminobutyraldehyde, 7.5 PM (pH 9.0) for propionaldehyde and 4 PM @H 7.4) for NAD.
INTRODUCTION
Aldehyde dehydrogenase is universally distributed and occurs within every organ of a mammalian organism. It comprises a large variety of enzymes, some of which have relatively narrow, others wide, substrate specificities. Their coenzyme requirements also differ: some are NAD specific, others NADP specific and some can utilize both coenzymes. Some occur as tetramers, others as dimers with subunit mol. wts ranging from 40,000 to 70,000 (see review by Pietruszko, 1989). The physiological role of some aldehyde dehydrogenases is well established while that of others remains unknown. Within the liver cell the enzyme is distributed in all compartments, with different forms occurring in mitochondria, cytoplasm and microsomes. It has been reported that rat liver aldehyde dehydrogenase activity is mainly associated with the mitochondrial fraction (Lindahl and Evces, 1984; Marjanen, 1972; Tottmar et al., 1973) and that the contribution of cytoplasmic activity to total liver aldehyde dehydrogenase activity is negligible (Lindahl and Evces, 1984). However, there were also reports which suggest the presence of small but not insignificant aldehyde dehydrogenase activity (Deitrich et al., 1972; Shum and Blair, 1972; Koivula and Koivusalo, 1975) and multiple isozyme patterns in rat cytoplasm (Berger et al., 1977; Tank et al., 1981; Cao et al., 1989). In the mammalian organisms y-aminobutyraldehyde can be formed by oxidative deamination
*Author to whom correspondence should be addressed. CBPB 100,2--H
of putrescine via the cytoplasmic enzyme, diamine oxidase. y-Aminobutyraldehyde is then converted to y -aminobutyric acid by an aldehyde dehydrogenase whose identity remained unknown until recently (Kurys et al., 1989). Involvement of aldehyde dehydrogenase in dehydrogenation of y -aminobutyraldehyde could be physiologically important because it connects metabolism of putrescine with that of y -aminobutyric acid, a well known inhibitory neurotransmitter. Putrescine is also a precursor of polyamines, spermidine and spermine, which are involved in numerous cellular regulatory processes (see review by Tabor and Tabor, 1984). The involvement of rat and mouse liver aldehyde dehydrogenase in the formation of amino acids from spermidine and spermine was suggested by Seiler et al. (1981). Because of the interest in alcohol metabolism, recent research on aldehyde dehydrogenase concentrated on the EC 1.2.1.3 class of enzymes with wide substrate specificity, having low PM Km values for short chain aliphatic aldehydes and utilizing NAD as coenzyme. Within the last two decades, two aldehyde dehydrogenase (EC 1.2.1.3) isozymes, one cytoplasmic and one mitochondrial, have been purified to homogeneity from several mammalian species and characterized (see review by Pietruszko, 1989). The human isozymes are named cytoplasmic El and mitochondrial E2. The third isozyme, E3, has been purified from human liver only recently (Kurys et al., 1989). Its low & for y-aminobutyraldehyde and high affinity for aminoaldehydes (Ambroziak and Pietruszko, 1991) suggest that this is the enzyme involved in metabolism of putrescine and other amino aldehydes formed via diamine oxidase. 321
WOJCIECH AMBROZIAKet al. )f the E3 isozyme The h u m a n livers, , are o b t a i n e d from a n d as such are onation; it is only h u m a n liver fresh an. F o r this reason, ,~d as the o t h e r two )re reliable study of made. C o m p a r i s o n minobutyraldehyde hree species is preate& In addition, the enzyme d e h y d r o g e n a t i n g yf i n o b u t y r a l d e h y d e from rat liver has been partially dried a n d characterized.
MATERIALS AND METHODS
Activity assay Aldehyde dehyq photometrically, 1 at 340nm, at 25' path. The assay EDTA in 0.I M s 1 mM propional buffer, pH 7.4 wi strate. When assa mixture also cont all assay comp( propionaldehyde. Bernt, 1974), glu 1980) and alcohc activities activit were als NADH NADI or N A t nc 6.22 mM-1 cm-1 activity. One Inte en2 of enzyme activit under the above
dty was determined spectromges in NADH absorbance ~tal volume and 1 cm light ned 0.5mM NAD, 1 mM osphate buffer, pH 9.0 with 0.1 M sodium phosphate minobutyraldehyde as subasmic fraction, the reaction yrazole. Controls contained y-aminobutyraldehyde or drogenase (Bergmeyer and "ogenase (McCarthy et al., se (Bergrneyer et al., 1974) pectrophotometrically with extinction coefi~cient of le calculation of enzymatic (IU) represents the amount form I #mol NADH/min
~terials
~hemicals were reagent grade. Pyrazole, 2-mercaptoLanol, sodium deoxycholate, propionaldehyde, v-aminotyraldehyde diethyl acetal were from Aldrich Chemical ., and NAD (grade 1) from Boehringer Mannheim. 9pionaldehyde was redistilled before use. y-AminobutyrLehyde was prepared from diethyl acetal as described by abroziak and Pietruszko (1987). Sephadex G-50, DEAEahadex G-50, 5'AMP-Sepharose 4B, NAD-Agarose re from Pharmacia Fine Chemicals. All other chemicals re from Sigma Chemical Co. Spragne-Dawley female rats ~arles River) and random bred guinea pigs of Hartley (Charles ori gin of either sex were maintained on standard laboratory )w diet and water ad lib. Post-mortem human liver from chow .6 yr old female was obtained 6 hr after death to autopsy a 46, C) and was subjected immediately to subcellular fraction(4°C) an. All buffers used during tissue fractionation and enzyme ation. dfication were exhaustively evacuated and nitrogenated. purification Tissue fractionation
All M1 proc_edures were perfo_rmed. at. 0-4°C. . . Livers . . were removed from the animals afterr decapitation, washed and placed in ice-cold isolation buffer:: 10 mM sodium phosphate LM EDTA, 0.1% (v/v) 2buffer, pH 7.0 containing 1 mM mercaptoethanol and 0.25 M sucros acrose. Liver samples (5 g) were homogenized in ca 20 ml off isolatior tion buffer in Potterisolation on Elvehjem homogenizer with Teflon pestle. The homogenates ng to a modified procedure (20%) were fractionated accordir ring centrifugation at 700 g of De Duve et al. (1955). Followin for l0 min, the pellet was washed:1twice twice with isolation buffer :arded. The resulting superand the nuclear fraction was discarded. natant fraction was centrifuged at 20,000g for I0 min, the esuspended in 25 ml of isopellet washed twice and finally resus[ oudrial fraction. The comlation buffer, giving the mitochoudrial bined supernatant was centrifug;ed at 105,000g for 60 min yielding, after pellet washing, the microsomal fraction, isolation which was resuspended in 25 ml of isolati alation buffer and the cytoplasmic fraction. Prior to enzzyme activity assay, sodium deoxycholate (0.25% w/v) was added dded to the microsomal and mitochondrial fractions; the fractions actions were subsequently homogenized, incubated on ice under nitrogen for 25 rain and centrifuged at 105,000g forr 30 min to remove debris. Human mitochondrial and cytol~flasmic subcellular fractions were isolated by an analogous procedure except that the mitochondrial pellet was resuspended reded in 0.5 ml of isolation buffer followed by sonication att a pulse cycle of 2 sec and 40% duty cycle for 40 sec with th W-380 sonicator (Heat Systems-Ultrasonics Inc.); debriss was removed by centrifugation prior to assay. Human mmrosome ucrosomes were pelleted by centrifugation at 70,000g for 60 min; th natant was the cytoplasmic fraction. A assayed for all enzyme activities on the
Kineti, etic measurem Pric Prior to kinetic s, partially purified enzyme fractions Sephadex G-50 column (1.7 fractic were pat x 15 15ccm) equilibt ercaptoethanol-free 0.03 M sodiur sodium phosphat 0 (deaerated and nitrogenated). The Michi for y-aminobutyraldehyde determined J al phosphate buffer, pH 7.4 were d containing 1I mM ~UllLilllllllg 1111¥1 F . , U 1 / 5 . i l l l U I O I r " propionaldehyde proplo] in 0.1 M contai sodiur sodium pyrophosphate buffer, pH 9.0 containing I mM EDTA. The single reaction progress cur method of Yun ress curve EDT,~ Suelter (1977) was used. Disulfiram (40 # M) was used and $1 during bot propionaldehyde durin~ the progress of reaction with both as substrates. substra The effect of and 7-aminobutyraldehyde 7 magnesium (MgC12, 160#M) on aldehyde aldeh.~ dehydrogenase magne activity with both substrates was tested in 25 mM PIPES (Sigma Chemical Co.) buffer, pH 7.0. Protein assay and isoelectric focusing Protein content was determined by 280 7 nm absorbance Pro~ nr hv or by the ultra microbiuret procedure of o Goa (1953) using bovine serum albl albumin as standard. Isoel rd. Isoelectric focusing was carried out on agarose plates (114 x 225 mm) composed of 1% agarose, 12% (w/v) sorbitol, 0.06% 0.06 °, Pharmalyte, pH 3-10 (Pharmacia/LKB), for 16 hrr at at 125 125 V. ~ The enzyme was stained for activity using 100 mM Tris-H(21, pH 8.5, containuetetrazolium (3 mg/30 ml), ing NAD (10 mg/30 ml), nitrobluetetrazc phenazine methosulfate (1 mg/30 ml), 1 mM m] propionaldehyde and/or 100 # M ),-aminobutyraldehyde. Isoelectric points were determined using the standards p~rovided in an Isoelectric Focusing Calibration Kit, Broad Broa~ Range pH 3-10, from Pharmacia/LKB. The standards were w~ applied accordffacturer and were stained ing to instructions of the manufacturer with Coomassie Brilliant Blue R following followi electrophoresis. Purification procedure The cytoplasmic fraction was usually prepared from 1522 g of rat liver using the same procedu 9rocedure as in subcellular fractionation. All buffers used in purifical ratification were deaerated and nitrogenated and contained 1 mM EDTA and 0.1% 2-mercaptoethanol:
buffer 1, 0.03 M sodium phosphate, I:~H 6.8; buffer 2, 0.03 M sodium phosphate, lC~H 6.0; buffer 3, 0.I M sodium phosphate, pt~H 8.0. Purification was performed at 0 4°C. 4°C The cytoplasmic tE-Sephadex column equiwashing with buffer 1 the yraldehyde were eluted with
Aldehyde
323
dehydrogenase
Table 1. Subcellular distribution of aldehyde dehydrogenase and marker enzyme activity in rat liver
Subcellular fraction M P S Total activity
Activity @mol mm-’ g-’ liver) Glutamate dehydrogenase Aldehyde dehydrogenase propionaldehyde y-aminobutyraldehyde Activity Activity % Activity % % mean ( f SD) Total mean ( f SD) Total Total mean(fSD) 0.02 (~0.005) 0.0
4.9
0.39 ( k 0.06) 0.41 ( k 0.06)
95.1
0
100
2.35 (f0.17) 0.28 (kO.03) 0.22 (kO.03) 2.85 (* 0.20)
82.5 9.8 1.1 100
110 (*IO) 1.8 (kO.3) 1.5 (kO.6) 113 (+ 10.4)
Lactate dehydrogenase Activity mean (k SD)
% Total
91.1 1.6 1.3 100
(*:::) 3.1 (?O.l) 178 (k39) 189 (k39)
4.1 1.6 94.3 100
Subcellular fractions are: M, mitochondrial; P, microsomal; S, cytoplasmic. Aldehyde dehydrogenase activity with 1 mM propionaldehyde was determined in 0.1 M sodium pyropbosphate buffer, pH 9.0 and with 100 pM y-aminobutyraldehyde in 0.1 M sodium phosphate buffer, pH 7.4. All assays contained 0.5 mM NAD and 1mM EDTA. Activity with y-aminobutyraldehyde and cytoplasmic fraction was determined in the presence of 1 mM pyrazole. Total activity is expressed as the sum of M + P + S fractions. The results represent the means + SD from three experiments.
an NaCl linear gradient (O-O.5 M) in buffer 1. Active fractions were pooled and after adjustment to pH 6.0 were applied on a S’AMP Sepharose 4B column equilibrated with buffer 2. The majority of y-aminobutyraldehyde-active enzyme did not bind to S’AMP-Sepharose 4B and came through in buffer 2. The pooled enzyme fractions, after dilution with buffer 2 to cu 0.02 M NaCl and concentration by an ultrafiltration method (Amicon 8400 Ultrafiltration Cell with PM 30 membrane), were applied on an NAIL Agarose column equilibrated with buffer 2. The column was washed with buffer 2 and 1 vol of buffer 3; y-aminobutyraldehyde-active enzyme was eluted with buffer 3 containing 1 mg/ml of NAD. All fractions with y -aminobutyraldehyde activity were pooled and concentrated by ultrafiltration against argon saturated buffer 1 and, after addition of 30% (w/v) glycerol, were stored at - 10°C.
RESULTS
Subcellular localization activity in rat liver
of aldehyde
akhydrogenase
When rat livers were subjected to subcellular fractionation (in three separate experiments), NADdependent aldehyde dehydrogenase activity with 1 mM propionaldehyde as substrate was found in all subcellular fractions: mitochondria, microsomes and cytoplasm (Table 1). The mitochondrial fraction possessed the majority of propionaldehyde activity (ca 83% of total activity); however, cytoplasmic and microsomal fractions also contributed to total activity, 7.7% and 9.8% respectively. In contrast,
anode +
cathode123
4
5678
Fig. 1. Isoelectric focusing gels of rat liver aldehyde dehydrogenase isozymes. (A) Total homogenate developed with propionaldehyde (track 1) and y-aminobutyraldehyde (track 2). p1 Marker proteins (track 3). (B) Cytoplasmic fraction and partially purified enzyme. Cytoplasmic fraction developed with propionaldehyde (track 4) and y-aminobutyraldehyde (track 5). Partially purified enzyme developed with y-arninobutyraldehyde (track 6) and Coomassie Brilliant Blue (track 7). p1 Marker proteins (track 8). pI Markers are, from the anode: amyloglucosidase, p1 3.5; methyl red (dye) pI 3.75; soybean trypsin inhibitor, p1 4.55; /I-lactoglobulin A, p1 5.2, bovine carbonic anhydrase B, p1 5.85; human carbonic anhydrase B, pI 6.55; horse myoglobin, p1 6.85; horse myoglobin, p1 7.35; lentil lectins, p1 8.15, 8.45, 8.65 and trypsinogen, ~19.3. Samples in tracks 2,4 and 5 were overloaded to better visualize low activity bands.
WOJCIECH AMBROZIAKet al.
~hyde was used as enase activity was ytoplasmic fraction, mzyme activities of m) and glutamate icated that the mitoons were relatively kpproximately 97% activity was local)4% of total lactate the cytoplasm. No ~lal fraction.
~electric focusing of rat cytoplasmic aldehyde hydrogenase isozymes A minimum of five activity bands with pls in the nge ca 5.3-6.8 were seen when isoelectric focusing Is of the cytoplasmic fraction were developed with nM propionaldehyde (Fig. 1, track 4). When total t liver homogenate, instead of the cytoplasmic teflon was used, an additional band of mitoondrial origin was visualized with a pI value of 5.2 (Fig. 1, track 1). Isoelectric focusing gels of cytoplasmic fraction, developed with 100/~M aminobutyraldehyde, showed 2-3 activity bands th pls in the range of 5.3-5.6; the same activity ttern was observed in total homogenate (Fig. 1, tcks 5 and 2, respectively). The most intense band seen m on isoelectric focusing gels developed with ?aminobl finobutyraldehyde had a pI value in the range of 5.3-5.4.
Comparison of subcellular distribution of the enzyme(s) metabolizing -aminobutyraldehyde in human, guinea pig and rat livers Subcellular fractionation of human (one experiment ent) and guinea pig (five experiments) livers showed that ~t 7-aminobutyraldehyde activity was exclusively localized in the cytoplasmic fraction (Fig. 2). Based on subcellular distribution of the marker enzymes, mitochondrial and cytoplasmic mic fractions were free from cross-contamination. Approximately 93% of total lactate dehydrogenase activity was in the cytoplasmic fraction of guinea pig liver and negligible activity (4.5%) was observed ,~d in the mitochondrial fraction. Almost 100% of total ~tal alcohol dehydrogenase activity was located in thee ccytoplasmic fraction of the human liver and only traces Lces of activity were seen >n. More than 95% of in the mitochondrial fraction. total glutamate dehydrogenase ~e activity was associated with the mitochondrial fraction :ion of both human and guinea pig liver. Traces of g lutamate dehydrogenase the cytoplasmic fraction activity were associated with the of guinea pig liver; about 4.5% k5% of total glutamate dehydrogenase activity of human man liv, liver was associated with the cytoplasmic fraction.t. Thus in both cases, the results with marker enzymes indicated that the cytoplasmic fraction was essentiall ~lly free from contamination by mitochondria. No ?-aminobutyraldehyde activity was found in the microsomal fraction of an microsomal fraction guinea pig liver. The human nobutyraldehyde due to was not assayed with 7-aminobut the extremely low yield of thee fraction.,. .. However,, the. majority of activity was found in fraction based on the comparison activity in total homogenate. Compari
butyraldehyde human, guinea three species, cytoplasmic fr~ butyraldehyde~ tissue of humar~
subcellular fractions of livers showed that in all is associated with the 2). The total ~,-aminoa 0.24, 0.3 and 0.41 IU/g md rat liver, respectively.
Partialpurificm metabolizing 7"
r aldehyde dehydrogenase dehyde
In order to aldehyde dehyd aldehyde, a sm from the humal et aL., 1989) wa
acterize rat cytoplasmic :abolizing v-aminobutyrfication method adapted oyme purification (Kurys zytoplasmic fraction was liver
80 '~ 60 :[ ~"~ 4o
mP
ms 20 o GL[
ALDtt B. H u m a n liver
10080-
"~ 60.
[]M ms
40 20 -
~
GLDH
100 ] 80 ,~ :.~ ~
W
ADH
ALDH
C. G u i n e a pig liver / A m
60 40
mP ms
20 o
GLDH
LDH
ALDH
distribution of aldehyde Fig. 2. Comparison of subcellular distri] dehydrogenase metabolizing ?-aminob -aminobutyraldehyde and marker enzymes in rat, human and guinea gui pig liver. Subcellular fractions are: M, mitochondri," chondrial; P, microsomal; S, cytoplasmic. Total activity is expres ~ressed as the sum of M + P + S fractions. Marker enzymes are: GLDH, glutamate dehydrogenase; LDH, lactate deh,.:ydrogenase; ADH, dehydrogenase (ALDH) alcohol dehydrogenase. Aldehyde dehyd activity was determined in 0.1 M sodium phosphate buffer, pH 7.4 containing 0.5 mM NAD, 1 mM EDTA, 1% ?-aminobutyraldehyde as a 2-mercaptoethanol and 100 #M ?-amino substrate. Aldehyde dehydrogenase activity activ assay contained 1 mM pyrazole. The results represent resent th the means from five ~e experiments with rat and experiments were done in =SD are below 5%).
Aldehyde dehydrogenase f rat liver cytoplasmic aldehyde dehydrogenase Total activity 0~mol rain- l) minobutyraldehyde Prot y Yield (%) Activity 100 34 30 20
n'ninobutyraldehyde
(%)
Activity ratio -Aminobutyraldehyde: propionaldehyde 1.4 2.2 2.8 3.3
2.80 0.60 0.44 0.24
the liver,following enzyme activitywith 7-amine e to propionaldehyde. Activity assay was as des
nd propionaldehyde and highest als and Methods.
ver using a similar
tabolizing 7-aminobutyrthe range 5.3-5.4. The es for ~,-aminobutyraldeNAD and is insensitive to In all these properties it hyde dehydrogenase E3 Lohomogeneity (Table 3).
ocedure as in subcellular fractionation. The ratio
' aldehyde dehydrogenase activity with 100/~M aminobutyraldehyde (pH 7.4) as substrate to 1 mM 'opionaldehyde (pH 9.0) in the cytoplasmic fraction ~s remarkably constant (1.4-1.7) in four preparions. The results of purification following the ajority of y-aminobutyraldehyde activity and the ghest activity ratio of y-aminobutyraldehyde to opionaldehyde are shown in Table 2. Activity with ~0/~M ),-aminobutyraldehyde was bound to EAE-Sephadex G-50 and was eluted in an NaC1 lear gradient (0.0-0.5 M) with an activity ratio of 1.8-2.2. The enzyme metabolizing 7-aminobutyrdehyde, like the E3 isozyme from human liver, did )t bind to 5'AMP-Sepharose 4B, and following tssage through that column, showed an increased L C Lrio t l V of ~-aminobutyraldehyde to propionaldehyde activit, tivity of ca 2.6-2.8. Final purified enzyme, which was is eluted from NAD-A NAD-Agarose with NAD, contained 20% ~% of starting ~,-aminobutyraldehyde activity and c a 2.2-2.5 mg protein based on a purification from 17.2 '.2 g of rat liver. It had a y-aminobutyraldehyde to propionaldeh3 opionaldehyde activity ratio of ca 2.8-3.2. On isoelectric )electric focusing gels with 100/~M ~-aminobutyraldeh :lehyde as substrate, only a single activity band was visualized realized with a pI value in the range of 5.3-5.4 (see track 3 in Fig. 1). The same ame band was visualized with 1 mM propionaldehyde. le. However, between 2 and 3 bands were visualized by staining with Coomassie Brilliant Blue R" (see track 4 in Fig. 1) demonstrating that the enzyyme was not completely homogeneous but only partially :ially purified, Comparison of properties with vith those of human E3 isozyme Properties of the partially lly purified enzyme are summarized in Table 3. Rat liver cytoplasmic
aldehyde dehy¢ aldehyde has aldell enzy~~ne has low hyde, propiona disull disulfiram and resembles the resen isozy .Tree, previo
;ION
Th present The bcellular distribution of enzyme(s) deh ~-aminobutyraldehyde rat a in human, hu ; livers has demonstrated their exclusivt n in the cytoplasmic fraction (Table fracti 9n the basis of the subcellular cellul distribr arker enzymes glutamate dehyq Lydrogenase (alcohol ydrogenase and lactate dehydr dehy~ dehydrogenase for human liver), mitochondrial and cytof)lasmic fractions were essentiall ;sentially free from crossconta contamination. Thus, v-aminobutyr~ :yraldehyde dehydrogenas mase activity is totally cytoplasmic while the activity with propionaldehyde as substrate is distributed in su all subcellular compartments, y-Ami -Aminobutyraldehyde dehy, ydrogenase activity in rat liver could be part of aldeh aldehyde dehydrogenase (EC 1.2.1.3) 1.2.1.2 or could be due to a specific enzyme. Isoelectric focusing foe gels of rat liver cytoplasmic fraction, develol~~ed with 0.1 mM ~,-aminobutyraldehyde as substrate, showed 2-3 activity bands in the pI range 5.3-5.6 5.3-' with the most intense band in the range 5.3-5.4 (Fig. (1 1). The same bands also appeared with propion )ropionaldehyde as subsuggesting that strate y-aminobutyraldehyde dehydrogenase activity may be a part of aldehyde dehydrogenase. For this reason, p~ mrification of the y-aminobutyraldehyde-metabolizin~Lg enzyme from rat liver cytoplasm was undertakem. Although the enzyme was only partially purl fie and contained mrified three protein components, only one component
Table 3. Kinetic c properties of partially purified cytoplasmic rat liver aldehyde dehydrogena ~nasc metabolizing ~,-aminobul aminobutyraldehyde and comparison with homogeneous E3 isozyme from human hum." liver
Property Kr, ~-aminobutyraldeh raid©hyde (0.5 mM NAD) Kmpropionaldehyde lyde (I (0.5 mM NAD) KmNAD (0.3 mM M "~-aminobutyraldehyde) Magnesium chloride ,ride (160/aM) Disulfiram (40/~M) IEF gel, pI value Activity ratio (7-aminobutyraldehyde/prc dehyde/propionaldehyde) ~l {(I n O ~ ' t *Data from Kurys et* al. All KS values are/~M thro for y-aminobutyraldeh propionaldehyde. The
Cytoplasmic rat liver isozyme
nic human Cytoplasmic humal liver isozyme* (major componentt)
5.0 + 1.1 (n ffi 4) 7.5 __.1.3 (n ffi 3) 4.2 + 0.8 (n = 3) no effect no effect 5.3-5.4
13.8 8.0 13.6 no effect no effect 5.3
2.8-3.2
3.6
. . . .
tier,p H 7.4 pH 9.0 for
W O J C I E C H A M B R O Z I A K el
'de dehydrogenase was associated with activity; the other tctive with aldehyde at in the rat liver ]e activity also bee (EC 1.2.1.3). The rat liver aldehyde r y-aminobutyraideliver E3 isozyme in ved during enzyme )le 3). Several kinetic 'operties of the partially purified cytoplasmic rat ,er aldehyde dehydrogenase resemble those of the 3 isozyme isolated from human liver (Table 3): these elude low K m values for y-aminobutyraldehyde pM), propionaldehyde (7.5pM), N A D (4/~M), tivity ratio of y-aminobutyraldehyde to propion:lehyde (ca 2.8-3.2), pI value (close to 5.4) and sensitivity to disulfiram and magnesium• Deitrich et al. (1972), Shum and Blair (1972) and oivula and Koivusalo (1975) have reported a signifint contribution of cytoplasmic aldehyde dehydronase to total liver activity. These results were iticized by Lindahl and Evces (1984) who showed tremely low cytoplasmic aldehyde dehydrogenase tivity (below 1%), non-detectable with acetalde'de, in different rat strains and suggested minimal contrll ntribution of the cytoplasm to rat liver aldehyde dehhydrogenase. However, Berger and Weiner (1977) an~d Tank et al. (1981) had reported different major isoz, ~zyme patterns (three to five isozymes) in the cytop toplasm from a closed colony of Purdue/Wistar rats ts and recently Cao et al. (1989) confirmed these results mlts with a new colony of Wistar rats. They also calculat, iculated K~ values of five cytoplasmic isozymes with th acetaldehyde. Unfortunately, they did not report the value of cytoplasmic activity relative to total activity in the liver. Our results alts on subcellular localization of propionaldehyde activity ( mM) indicated tctivity (1 y(1 that ca 7.5% of total rat.t liver activity in the ssociated with the cytoSprague--Dawley strain is assoclat~ ;ity of cytoplasmic aldeplasmic fraction. The intensit' hyde dehydrogenase activity bandss seen se on isoelectric seen focusing gels, the multiband isoenzyme pattern similar to that reported by Weiner ~er and his colleagues in the Wistar strain and the localizatic ation of significant )calization y-aminobutyraldehyde activit ity in the cytoplasm all suggest that the contribution of cytoplasmic aldehyde tal rat liver activity candehydrogenase activity to total not be neglected. Assuming that y-aminoing in the cytoplasm is butyraldehyde activity existing represented only by our partially trtially purified isozyme with an activity ratio of y-aminobl obut' -aminobutyraldehyde to propionaldehyde of 3 : 1, a rmmmum finimum contribution of 4.3% can be expected (0.3577 IU/i IU/g of ~,-aminobutyraldehyde activity + 3 = O.1199 IU/g propionaldehyde L3% of total propionactivity, which represents 4.3! iver; see Table 1). Slight aldehyde activity in the rat liver: modifications of the tissue fractionation procedure and assay system betweena investigators cannot account for such differencess in reported values of cytoplasmic aldehyde dehydrog4 ~rogenase activity...Our experience with human liver and isozymes showed that cytoplasmic aid genase isozymes and the main isozyr
al.
y-aminobutyra tery labile and extremely sensitive to at~ ygen. Therefore, all our buffers used in 1 were extensively degassed, nitrogenated ed 0.1% v/v 2-mercaptoethanol and r fractions were immediately assayed f ~tivity. Such precautions may not neces ~een followed by earlier investigators ware of these problems. The results this paper demonstrate that enzyme(s y-aminobutyraldehyde dehydrogenatic ,~dexclusively in the cytoplasm of huma and rat livers. Isoelectric focus focusing result data from the partial purifi ~urification of from rat liver strongly sugg~ ;est that aldehyde and propionaldeh aldehyde activi 'the same enzyme, which appe~ ~ears to be ydrogenase (EC 1.2.1.3). The ]partially p ae from rat liver metabolizir olizing y-amine le resembles in properties the hhomogenec cde dehydrogenase from hum~ human liver. cies the Km values for y-am ,-aminobutyra] low, suggesting that the enzyr :yme might its metabolism. In the living g organist ,~ is metabolized to yamin,obutyrald( ~mine oxidase which is also a cytopla~ • Thus, the low Km for y-am: ',-aminobutyra] "t:l.llllllLI U t t t y I i:tlUti;ll)" O ~ U U I I ILbined L l l l l t ; O with subcellular locali localization further suggest that th~ the enzyme may be impo ~ortant in metabolism of aldeh~¢des arising from diamines. Ackn~ Acknowledgements--This work was supl: upported by US Public
Healtl Health Service National Institute on Alcohol ? Abuse and Alcoholism Grant AA00186 and Research Resear Scientist Award Alcoh K05 AA00046. REFERENCES
Ambr, broziak W. and Pietruszko R. (1987) (1987 Human aldehyde dehydrogenase: metabolism of putres( autrescine and histamine. Alcohol. din. exp. Res. I1, 528-532. Ambroziak W. and Pietruszko R. (1991) (1991 Human aldehyde dehydrogenase: activity with aldehycde metabolites of amines. J. biol. Chem. monoamines, diamines and polyami; 266, 13,011-13,018. Berger D. and Weiner H. (1977) Rel Relationship between alcohol preference and biogenic alde aldehyde metabolizing enzymes in rats. Biochem. Pharmacol armacol. 26, 841-846. Bergmeyer H. U. and Bernt E. (1974) Lactate dehydrogenase. In Methods of Enzymatic A Analysis (Edited by Bergmeyer H. U.), Vol. 2, pp. 574-579. 574-5 Verlag Chemic GmbH, Wienheim. Bergmeyer H. U., Gawehn K. and Grassl M. (1974) Alcohol ymatic Analysis (Edited dehydrogenase. In Methods of Enzyma by Bergmeyer H. U.), Vol. 1, pp. 428-~ Verlag Chemie ~. 428-429. GmbH, Weinheim. Cao Q.-N., Tu G.-C. and Weiner H. (1989) Presence of cytosolic aldehyde dehydrogenase lsozymes iso~ in adult and fetal rat liver. Biochem. Pharmacol. 38, 77-83. Deitrich R. A., Collins A. C. and Erwin Er V. G. (1972) Genetic influence upon phenobarbiU tobarbital-induced increase ¢de dehydrogenase activity. in rat liver supernatant aldehyde dehy J. biol. Chem. 247, 7232-7236. De Duve C., Pressman B. C., Gianetto Gianett( R., Wattiaux R. and Aplemans F. (1955) Tissue fracti fractionation studies. 6. Intracellular distribution patterns of enzymes e in rat liver tissues. Biochem. J. 60, 604-617. .~thod for protein determin1 protein in cerebrospinal Ft. 5, 218-222.
Aldehyde dehydrogenase 5) Different forms of and their subcellular ta 397, 9-23. szko R. (1989) Human 3n and characterization ),-aminobutyraldehyde. t Tipton K. F. (1980) )genase from ox brain 11. omparative subcellular mase in rat, mouse and 33, 3383-3389. larjanen L. A. (1972) Intracellular localization of aldehyde dehydrogenase in rat liver. Biochem. 3". 127, 633-639. ietruszko R. (1989) Aldehyde dehydrogenase (EC 1.2.1.3). In Biochemistry and Physiology o f Substance Abuse
(Edited b y W ; Inc., Boca Ra Seller N., KncH formation of .' spermine. Bio~ Shum G. T. and in rat liver. C Tabor C. W. al Biochem. 53, Tank A. W., Wt ology and sub in rat liver. B~ Tottmar S. O. C The Th, subcellul~ dehydrogenasc de! Yun S. L. and calculating K~ cal, progress curve prc
1, pp. 89-127. CRC Press, Gittos W. (1981) On the ;riving from spermidine and 23-132. 72) Aldehyde dehydrogenase • 50, 741-748. (1984) Polyamines. A. Rev. 'urman J. A. (1981) Enzymation of aldehyde oxidation acol. 30, 3265-3275. and Kiessling K. H. (1973) and properties of aldehyde ~iochem. J. 135, 577-586. 11977) A simple method for ~m single enzyme reaction ~hys. Acta 480, 1-13.
0305-0491/9153.00+ 0.00 PergamonPressplc
Vol. IOOB, No. 2, pp. 321-327,1991
Printedin GreatBritain
ALDEHYDE DEHYDROGENASE (EC 1.2.1.3): COMPARISON OF SUBCELLULAR LOCALIZATION OF THE THIRD ISOZYME THAT DEHYDROGENATES y -AMINOBUTYRALDEHYDE IN RAT, GUINEA PIG AND HUMAN LIVER WOJCIECHAMBROZIAK,GLORIAKURYS and REGINAPETRUSZKO* Center of Alcohol Studies, Rutgers University, Piscataway, NJ 0885-69, (Tel: 908 932 3595); (Fax: 908 932 5944)
USA
(Received 25 March 1991) Subcellular fractionation of rat, guinea pig and human livers showed that aldehyde dehydrogenase metabolizing y-aminobutyraldehyde was exclusively localized in the cytoplasmic fraction
Ahatract-1.
in all three mammalian species. 2. Total y-aminobutyraldehyde activity of aldehyde dehydrogenase was found to be ca 0.41, 0.3 and 0.24 pmol NADH min-’ g-l tissue, respectively in rat, guinea pig and human liver, with more than 95%
of activity in the cytoplasm. 3. Partially purified cytoplasmic isozyme from rat liver showed similar chromatographic behavior and kinetic properties to the E3 isozyme isolated from human liver. 4. The rat isozyme was insensitive to disulfiram (40pM) and to magnesium (16OpM) and had &, values of 5 PM (PH 7.4) for y-aminobutyraldehyde, 7.5 PM (pH 9.0) for propionaldehyde and 4 PM @H 7.4) for NAD.
INTRODUCTION
Aldehyde dehydrogenase is universally distributed and occurs within every organ of a mammalian organism. It comprises a large variety of enzymes, some of which have relatively narrow, others wide, substrate specificities. Their coenzyme requirements also differ: some are NAD specific, others NADP specific and some can utilize both coenzymes. Some occur as tetramers, others as dimers with subunit mol. wts ranging from 40,000 to 70,000 (see review by Pietruszko, 1989). The physiological role of some aldehyde dehydrogenases is well established while that of others remains unknown. Within the liver cell the enzyme is distributed in all compartments, with different forms occurring in mitochondria, cytoplasm and microsomes. It has been reported that rat liver aldehyde dehydrogenase activity is mainly associated with the mitochondrial fraction (Lindahl and Evces, 1984; Marjanen, 1972; Tottmar et al., 1973) and that the contribution of cytoplasmic activity to total liver aldehyde dehydrogenase activity is negligible (Lindahl and Evces, 1984). However, there were also reports which suggest the presence of small but not insignificant aldehyde dehydrogenase activity (Deitrich et al., 1972; Shum and Blair, 1972; Koivula and Koivusalo, 1975) and multiple isozyme patterns in rat cytoplasm (Berger et al., 1977; Tank et al., 1981; Cao et al., 1989). In the mammalian organisms y-aminobutyraldehyde can be formed by oxidative deamination
*Author to whom correspondence should be addressed. CBPB 100,2--H
of putrescine via the cytoplasmic enzyme, diamine oxidase. y-Aminobutyraldehyde is then converted to y -aminobutyric acid by an aldehyde dehydrogenase whose identity remained unknown until recently (Kurys et al., 1989). Involvement of aldehyde dehydrogenase in dehydrogenation of y -aminobutyraldehyde could be physiologically important because it connects metabolism of putrescine with that of y -aminobutyric acid, a well known inhibitory neurotransmitter. Putrescine is also a precursor of polyamines, spermidine and spermine, which are involved in numerous cellular regulatory processes (see review by Tabor and Tabor, 1984). The involvement of rat and mouse liver aldehyde dehydrogenase in the formation of amino acids from spermidine and spermine was suggested by Seiler et al. (1981). Because of the interest in alcohol metabolism, recent research on aldehyde dehydrogenase concentrated on the EC 1.2.1.3 class of enzymes with wide substrate specificity, having low PM Km values for short chain aliphatic aldehydes and utilizing NAD as coenzyme. Within the last two decades, two aldehyde dehydrogenase (EC 1.2.1.3) isozymes, one cytoplasmic and one mitochondrial, have been purified to homogeneity from several mammalian species and characterized (see review by Pietruszko, 1989). The human isozymes are named cytoplasmic El and mitochondrial E2. The third isozyme, E3, has been purified from human liver only recently (Kurys et al., 1989). Its low & for y-aminobutyraldehyde and high affinity for aminoaldehydes (Ambroziak and Pietruszko, 1991) suggest that this is the enzyme involved in metabolism of putrescine and other amino aldehydes formed via diamine oxidase. 321
WOJCIECH AMBROZIAKet al. )f the E3 isozyme The h u m a n livers, , are o b t a i n e d from a n d as such are onation; it is only h u m a n liver fresh an. F o r this reason, ,~d as the o t h e r two )re reliable study of made. C o m p a r i s o n minobutyraldehyde hree species is preate& In addition, the enzyme d e h y d r o g e n a t i n g yf i n o b u t y r a l d e h y d e from rat liver has been partially dried a n d characterized.
MATERIALS AND METHODS
Activity assay Aldehyde dehyq photometrically, 1 at 340nm, at 25' path. The assay EDTA in 0.I M s 1 mM propional buffer, pH 7.4 wi strate. When assa mixture also cont all assay comp( propionaldehyde. Bernt, 1974), glu 1980) and alcohc activities activit were als NADH NADI or N A t nc 6.22 mM-1 cm-1 activity. One Inte en2 of enzyme activit under the above
dty was determined spectromges in NADH absorbance ~tal volume and 1 cm light ned 0.5mM NAD, 1 mM osphate buffer, pH 9.0 with 0.1 M sodium phosphate minobutyraldehyde as subasmic fraction, the reaction yrazole. Controls contained y-aminobutyraldehyde or drogenase (Bergmeyer and "ogenase (McCarthy et al., se (Bergrneyer et al., 1974) pectrophotometrically with extinction coefi~cient of le calculation of enzymatic (IU) represents the amount form I #mol NADH/min
~terials
~hemicals were reagent grade. Pyrazole, 2-mercaptoLanol, sodium deoxycholate, propionaldehyde, v-aminotyraldehyde diethyl acetal were from Aldrich Chemical ., and NAD (grade 1) from Boehringer Mannheim. 9pionaldehyde was redistilled before use. y-AminobutyrLehyde was prepared from diethyl acetal as described by abroziak and Pietruszko (1987). Sephadex G-50, DEAEahadex G-50, 5'AMP-Sepharose 4B, NAD-Agarose re from Pharmacia Fine Chemicals. All other chemicals re from Sigma Chemical Co. Spragne-Dawley female rats ~arles River) and random bred guinea pigs of Hartley (Charles ori gin of either sex were maintained on standard laboratory )w diet and water ad lib. Post-mortem human liver from chow .6 yr old female was obtained 6 hr after death to autopsy a 46, C) and was subjected immediately to subcellular fraction(4°C) an. All buffers used during tissue fractionation and enzyme ation. dfication were exhaustively evacuated and nitrogenated. purification Tissue fractionation
All M1 proc_edures were perfo_rmed. at. 0-4°C. . . Livers . . were removed from the animals afterr decapitation, washed and placed in ice-cold isolation buffer:: 10 mM sodium phosphate LM EDTA, 0.1% (v/v) 2buffer, pH 7.0 containing 1 mM mercaptoethanol and 0.25 M sucros acrose. Liver samples (5 g) were homogenized in ca 20 ml off isolatior tion buffer in Potterisolation on Elvehjem homogenizer with Teflon pestle. The homogenates ng to a modified procedure (20%) were fractionated accordir ring centrifugation at 700 g of De Duve et al. (1955). Followin for l0 min, the pellet was washed:1twice twice with isolation buffer :arded. The resulting superand the nuclear fraction was discarded. natant fraction was centrifuged at 20,000g for I0 min, the esuspended in 25 ml of isopellet washed twice and finally resus[ oudrial fraction. The comlation buffer, giving the mitochoudrial bined supernatant was centrifug;ed at 105,000g for 60 min yielding, after pellet washing, the microsomal fraction, isolation which was resuspended in 25 ml of isolati alation buffer and the cytoplasmic fraction. Prior to enzzyme activity assay, sodium deoxycholate (0.25% w/v) was added dded to the microsomal and mitochondrial fractions; the fractions actions were subsequently homogenized, incubated on ice under nitrogen for 25 rain and centrifuged at 105,000g forr 30 min to remove debris. Human mitochondrial and cytol~flasmic subcellular fractions were isolated by an analogous procedure except that the mitochondrial pellet was resuspended reded in 0.5 ml of isolation buffer followed by sonication att a pulse cycle of 2 sec and 40% duty cycle for 40 sec with th W-380 sonicator (Heat Systems-Ultrasonics Inc.); debriss was removed by centrifugation prior to assay. Human mmrosome ucrosomes were pelleted by centrifugation at 70,000g for 60 min; th natant was the cytoplasmic fraction. A assayed for all enzyme activities on the
Kineti, etic measurem Pric Prior to kinetic s, partially purified enzyme fractions Sephadex G-50 column (1.7 fractic were pat x 15 15ccm) equilibt ercaptoethanol-free 0.03 M sodiur sodium phosphat 0 (deaerated and nitrogenated). The Michi for y-aminobutyraldehyde determined J al phosphate buffer, pH 7.4 were d containing 1I mM ~UllLilllllllg 1111¥1 F . , U 1 / 5 . i l l l U I O I r " propionaldehyde proplo] in 0.1 M contai sodiur sodium pyrophosphate buffer, pH 9.0 containing I mM EDTA. The single reaction progress cur method of Yun ress curve EDT,~ Suelter (1977) was used. Disulfiram (40 # M) was used and $1 during bot propionaldehyde durin~ the progress of reaction with both as substrates. substra The effect of and 7-aminobutyraldehyde 7 magnesium (MgC12, 160#M) on aldehyde aldeh.~ dehydrogenase magne activity with both substrates was tested in 25 mM PIPES (Sigma Chemical Co.) buffer, pH 7.0. Protein assay and isoelectric focusing Protein content was determined by 280 7 nm absorbance Pro~ nr hv or by the ultra microbiuret procedure of o Goa (1953) using bovine serum albl albumin as standard. Isoel rd. Isoelectric focusing was carried out on agarose plates (114 x 225 mm) composed of 1% agarose, 12% (w/v) sorbitol, 0.06% 0.06 °, Pharmalyte, pH 3-10 (Pharmacia/LKB), for 16 hrr at at 125 125 V. ~ The enzyme was stained for activity using 100 mM Tris-H(21, pH 8.5, containuetetrazolium (3 mg/30 ml), ing NAD (10 mg/30 ml), nitrobluetetrazc phenazine methosulfate (1 mg/30 ml), 1 mM m] propionaldehyde and/or 100 # M ),-aminobutyraldehyde. Isoelectric points were determined using the standards p~rovided in an Isoelectric Focusing Calibration Kit, Broad Broa~ Range pH 3-10, from Pharmacia/LKB. The standards were w~ applied accordffacturer and were stained ing to instructions of the manufacturer with Coomassie Brilliant Blue R following followi electrophoresis. Purification procedure The cytoplasmic fraction was usually prepared from 1522 g of rat liver using the same procedu 9rocedure as in subcellular fractionation. All buffers used in purifical ratification were deaerated and nitrogenated and contained 1 mM EDTA and 0.1% 2-mercaptoethanol:
buffer 1, 0.03 M sodium phosphate, I:~H 6.8; buffer 2, 0.03 M sodium phosphate, lC~H 6.0; buffer 3, 0.I M sodium phosphate, pt~H 8.0. Purification was performed at 0 4°C. 4°C The cytoplasmic tE-Sephadex column equiwashing with buffer 1 the yraldehyde were eluted with
Aldehyde
323
dehydrogenase
Table 1. Subcellular distribution of aldehyde dehydrogenase and marker enzyme activity in rat liver
Subcellular fraction M P S Total activity
Activity @mol mm-’ g-’ liver) Glutamate dehydrogenase Aldehyde dehydrogenase propionaldehyde y-aminobutyraldehyde Activity Activity % Activity % % mean ( f SD) Total mean ( f SD) Total Total mean(fSD) 0.02 (~0.005) 0.0
4.9
0.39 ( k 0.06) 0.41 ( k 0.06)
95.1
0
100
2.35 (f0.17) 0.28 (kO.03) 0.22 (kO.03) 2.85 (* 0.20)
82.5 9.8 1.1 100
110 (*IO) 1.8 (kO.3) 1.5 (kO.6) 113 (+ 10.4)
Lactate dehydrogenase Activity mean (k SD)
% Total
91.1 1.6 1.3 100
(*:::) 3.1 (?O.l) 178 (k39) 189 (k39)
4.1 1.6 94.3 100
Subcellular fractions are: M, mitochondrial; P, microsomal; S, cytoplasmic. Aldehyde dehydrogenase activity with 1 mM propionaldehyde was determined in 0.1 M sodium pyropbosphate buffer, pH 9.0 and with 100 pM y-aminobutyraldehyde in 0.1 M sodium phosphate buffer, pH 7.4. All assays contained 0.5 mM NAD and 1mM EDTA. Activity with y-aminobutyraldehyde and cytoplasmic fraction was determined in the presence of 1 mM pyrazole. Total activity is expressed as the sum of M + P + S fractions. The results represent the means + SD from three experiments.
an NaCl linear gradient (O-O.5 M) in buffer 1. Active fractions were pooled and after adjustment to pH 6.0 were applied on a S’AMP Sepharose 4B column equilibrated with buffer 2. The majority of y-aminobutyraldehyde-active enzyme did not bind to S’AMP-Sepharose 4B and came through in buffer 2. The pooled enzyme fractions, after dilution with buffer 2 to cu 0.02 M NaCl and concentration by an ultrafiltration method (Amicon 8400 Ultrafiltration Cell with PM 30 membrane), were applied on an NAIL Agarose column equilibrated with buffer 2. The column was washed with buffer 2 and 1 vol of buffer 3; y-aminobutyraldehyde-active enzyme was eluted with buffer 3 containing 1 mg/ml of NAD. All fractions with y -aminobutyraldehyde activity were pooled and concentrated by ultrafiltration against argon saturated buffer 1 and, after addition of 30% (w/v) glycerol, were stored at - 10°C.
RESULTS
Subcellular localization activity in rat liver
of aldehyde
akhydrogenase
When rat livers were subjected to subcellular fractionation (in three separate experiments), NADdependent aldehyde dehydrogenase activity with 1 mM propionaldehyde as substrate was found in all subcellular fractions: mitochondria, microsomes and cytoplasm (Table 1). The mitochondrial fraction possessed the majority of propionaldehyde activity (ca 83% of total activity); however, cytoplasmic and microsomal fractions also contributed to total activity, 7.7% and 9.8% respectively. In contrast,
anode +
cathode123
4
5678
Fig. 1. Isoelectric focusing gels of rat liver aldehyde dehydrogenase isozymes. (A) Total homogenate developed with propionaldehyde (track 1) and y-aminobutyraldehyde (track 2). p1 Marker proteins (track 3). (B) Cytoplasmic fraction and partially purified enzyme. Cytoplasmic fraction developed with propionaldehyde (track 4) and y-aminobutyraldehyde (track 5). Partially purified enzyme developed with y-arninobutyraldehyde (track 6) and Coomassie Brilliant Blue (track 7). p1 Marker proteins (track 8). pI Markers are, from the anode: amyloglucosidase, p1 3.5; methyl red (dye) pI 3.75; soybean trypsin inhibitor, p1 4.55; /I-lactoglobulin A, p1 5.2, bovine carbonic anhydrase B, p1 5.85; human carbonic anhydrase B, pI 6.55; horse myoglobin, p1 6.85; horse myoglobin, p1 7.35; lentil lectins, p1 8.15, 8.45, 8.65 and trypsinogen, ~19.3. Samples in tracks 2,4 and 5 were overloaded to better visualize low activity bands.
WOJCIECH AMBROZIAKet al.
~hyde was used as enase activity was ytoplasmic fraction, mzyme activities of m) and glutamate icated that the mitoons were relatively kpproximately 97% activity was local)4% of total lactate the cytoplasm. No ~lal fraction.
~electric focusing of rat cytoplasmic aldehyde hydrogenase isozymes A minimum of five activity bands with pls in the nge ca 5.3-6.8 were seen when isoelectric focusing Is of the cytoplasmic fraction were developed with nM propionaldehyde (Fig. 1, track 4). When total t liver homogenate, instead of the cytoplasmic teflon was used, an additional band of mitoondrial origin was visualized with a pI value of 5.2 (Fig. 1, track 1). Isoelectric focusing gels of cytoplasmic fraction, developed with 100/~M aminobutyraldehyde, showed 2-3 activity bands th pls in the range of 5.3-5.6; the same activity ttern was observed in total homogenate (Fig. 1, tcks 5 and 2, respectively). The most intense band seen m on isoelectric focusing gels developed with ?aminobl finobutyraldehyde had a pI value in the range of 5.3-5.4.
Comparison of subcellular distribution of the enzyme(s) metabolizing -aminobutyraldehyde in human, guinea pig and rat livers Subcellular fractionation of human (one experiment ent) and guinea pig (five experiments) livers showed that ~t 7-aminobutyraldehyde activity was exclusively localized in the cytoplasmic fraction (Fig. 2). Based on subcellular distribution of the marker enzymes, mitochondrial and cytoplasmic mic fractions were free from cross-contamination. Approximately 93% of total lactate dehydrogenase activity was in the cytoplasmic fraction of guinea pig liver and negligible activity (4.5%) was observed ,~d in the mitochondrial fraction. Almost 100% of total ~tal alcohol dehydrogenase activity was located in thee ccytoplasmic fraction of the human liver and only traces Lces of activity were seen >n. More than 95% of in the mitochondrial fraction. total glutamate dehydrogenase ~e activity was associated with the mitochondrial fraction :ion of both human and guinea pig liver. Traces of g lutamate dehydrogenase the cytoplasmic fraction activity were associated with the of guinea pig liver; about 4.5% k5% of total glutamate dehydrogenase activity of human man liv, liver was associated with the cytoplasmic fraction.t. Thus in both cases, the results with marker enzymes indicated that the cytoplasmic fraction was essentiall ~lly free from contamination by mitochondria. No ?-aminobutyraldehyde activity was found in the microsomal fraction of an microsomal fraction guinea pig liver. The human nobutyraldehyde due to was not assayed with 7-aminobut the extremely low yield of thee fraction.,. .. However,, the. majority of activity was found in fraction based on the comparison activity in total homogenate. Compari
butyraldehyde human, guinea three species, cytoplasmic fr~ butyraldehyde~ tissue of humar~
subcellular fractions of livers showed that in all is associated with the 2). The total ~,-aminoa 0.24, 0.3 and 0.41 IU/g md rat liver, respectively.
Partialpurificm metabolizing 7"
r aldehyde dehydrogenase dehyde
In order to aldehyde dehyd aldehyde, a sm from the humal et aL., 1989) wa
acterize rat cytoplasmic :abolizing v-aminobutyrfication method adapted oyme purification (Kurys zytoplasmic fraction was liver
80 '~ 60 :[ ~"~ 4o
mP
ms 20 o GL[
ALDtt B. H u m a n liver
10080-
"~ 60.
[]M ms
40 20 -
~
GLDH
100 ] 80 ,~ :.~ ~
W
ADH
ALDH
C. G u i n e a pig liver / A m
60 40
mP ms
20 o
GLDH
LDH
ALDH
distribution of aldehyde Fig. 2. Comparison of subcellular distri] dehydrogenase metabolizing ?-aminob -aminobutyraldehyde and marker enzymes in rat, human and guinea gui pig liver. Subcellular fractions are: M, mitochondri," chondrial; P, microsomal; S, cytoplasmic. Total activity is expres ~ressed as the sum of M + P + S fractions. Marker enzymes are: GLDH, glutamate dehydrogenase; LDH, lactate deh,.:ydrogenase; ADH, dehydrogenase (ALDH) alcohol dehydrogenase. Aldehyde dehyd activity was determined in 0.1 M sodium phosphate buffer, pH 7.4 containing 0.5 mM NAD, 1 mM EDTA, 1% ?-aminobutyraldehyde as a 2-mercaptoethanol and 100 #M ?-amino substrate. Aldehyde dehydrogenase activity activ assay contained 1 mM pyrazole. The results represent resent th the means from five ~e experiments with rat and experiments were done in =SD are below 5%).
Aldehyde dehydrogenase f rat liver cytoplasmic aldehyde dehydrogenase Total activity 0~mol rain- l) minobutyraldehyde Prot y Yield (%) Activity 100 34 30 20
n'ninobutyraldehyde
(%)
Activity ratio -Aminobutyraldehyde: propionaldehyde 1.4 2.2 2.8 3.3
2.80 0.60 0.44 0.24
the liver,following enzyme activitywith 7-amine e to propionaldehyde. Activity assay was as des
nd propionaldehyde and highest als and Methods.
ver using a similar
tabolizing 7-aminobutyrthe range 5.3-5.4. The es for ~,-aminobutyraldeNAD and is insensitive to In all these properties it hyde dehydrogenase E3 Lohomogeneity (Table 3).
ocedure as in subcellular fractionation. The ratio
' aldehyde dehydrogenase activity with 100/~M aminobutyraldehyde (pH 7.4) as substrate to 1 mM 'opionaldehyde (pH 9.0) in the cytoplasmic fraction ~s remarkably constant (1.4-1.7) in four preparions. The results of purification following the ajority of y-aminobutyraldehyde activity and the ghest activity ratio of y-aminobutyraldehyde to opionaldehyde are shown in Table 2. Activity with ~0/~M ),-aminobutyraldehyde was bound to EAE-Sephadex G-50 and was eluted in an NaC1 lear gradient (0.0-0.5 M) with an activity ratio of 1.8-2.2. The enzyme metabolizing 7-aminobutyrdehyde, like the E3 isozyme from human liver, did )t bind to 5'AMP-Sepharose 4B, and following tssage through that column, showed an increased L C Lrio t l V of ~-aminobutyraldehyde to propionaldehyde activit, tivity of ca 2.6-2.8. Final purified enzyme, which was is eluted from NAD-A NAD-Agarose with NAD, contained 20% ~% of starting ~,-aminobutyraldehyde activity and c a 2.2-2.5 mg protein based on a purification from 17.2 '.2 g of rat liver. It had a y-aminobutyraldehyde to propionaldeh3 opionaldehyde activity ratio of ca 2.8-3.2. On isoelectric )electric focusing gels with 100/~M ~-aminobutyraldeh :lehyde as substrate, only a single activity band was visualized realized with a pI value in the range of 5.3-5.4 (see track 3 in Fig. 1). The same ame band was visualized with 1 mM propionaldehyde. le. However, between 2 and 3 bands were visualized by staining with Coomassie Brilliant Blue R" (see track 4 in Fig. 1) demonstrating that the enzyyme was not completely homogeneous but only partially :ially purified, Comparison of properties with vith those of human E3 isozyme Properties of the partially lly purified enzyme are summarized in Table 3. Rat liver cytoplasmic
aldehyde dehy¢ aldehyde has aldell enzy~~ne has low hyde, propiona disull disulfiram and resembles the resen isozy .Tree, previo
;ION
Th present The bcellular distribution of enzyme(s) deh ~-aminobutyraldehyde rat a in human, hu ; livers has demonstrated their exclusivt n in the cytoplasmic fraction (Table fracti 9n the basis of the subcellular cellul distribr arker enzymes glutamate dehyq Lydrogenase (alcohol ydrogenase and lactate dehydr dehy~ dehydrogenase for human liver), mitochondrial and cytof)lasmic fractions were essentiall ;sentially free from crossconta contamination. Thus, v-aminobutyr~ :yraldehyde dehydrogenas mase activity is totally cytoplasmic while the activity with propionaldehyde as substrate is distributed in su all subcellular compartments, y-Ami -Aminobutyraldehyde dehy, ydrogenase activity in rat liver could be part of aldeh aldehyde dehydrogenase (EC 1.2.1.3) 1.2.1.2 or could be due to a specific enzyme. Isoelectric focusing foe gels of rat liver cytoplasmic fraction, develol~~ed with 0.1 mM ~,-aminobutyraldehyde as substrate, showed 2-3 activity bands in the pI range 5.3-5.6 5.3-' with the most intense band in the range 5.3-5.4 (Fig. (1 1). The same bands also appeared with propion )ropionaldehyde as subsuggesting that strate y-aminobutyraldehyde dehydrogenase activity may be a part of aldehyde dehydrogenase. For this reason, p~ mrification of the y-aminobutyraldehyde-metabolizin~Lg enzyme from rat liver cytoplasm was undertakem. Although the enzyme was only partially purl fie and contained mrified three protein components, only one component
Table 3. Kinetic c properties of partially purified cytoplasmic rat liver aldehyde dehydrogena ~nasc metabolizing ~,-aminobul aminobutyraldehyde and comparison with homogeneous E3 isozyme from human hum." liver
Property Kr, ~-aminobutyraldeh raid©hyde (0.5 mM NAD) Kmpropionaldehyde lyde (I (0.5 mM NAD) KmNAD (0.3 mM M "~-aminobutyraldehyde) Magnesium chloride ,ride (160/aM) Disulfiram (40/~M) IEF gel, pI value Activity ratio (7-aminobutyraldehyde/prc dehyde/propionaldehyde) ~l {(I n O ~ ' t *Data from Kurys et* al. All KS values are/~M thro for y-aminobutyraldeh propionaldehyde. The
Cytoplasmic rat liver isozyme
nic human Cytoplasmic humal liver isozyme* (major componentt)
5.0 + 1.1 (n ffi 4) 7.5 __.1.3 (n ffi 3) 4.2 + 0.8 (n = 3) no effect no effect 5.3-5.4
13.8 8.0 13.6 no effect no effect 5.3
2.8-3.2
3.6
. . . .
tier,p H 7.4 pH 9.0 for
W O J C I E C H A M B R O Z I A K el
'de dehydrogenase was associated with activity; the other tctive with aldehyde at in the rat liver ]e activity also bee (EC 1.2.1.3). The rat liver aldehyde r y-aminobutyraideliver E3 isozyme in ved during enzyme )le 3). Several kinetic 'operties of the partially purified cytoplasmic rat ,er aldehyde dehydrogenase resemble those of the 3 isozyme isolated from human liver (Table 3): these elude low K m values for y-aminobutyraldehyde pM), propionaldehyde (7.5pM), N A D (4/~M), tivity ratio of y-aminobutyraldehyde to propion:lehyde (ca 2.8-3.2), pI value (close to 5.4) and sensitivity to disulfiram and magnesium• Deitrich et al. (1972), Shum and Blair (1972) and oivula and Koivusalo (1975) have reported a signifint contribution of cytoplasmic aldehyde dehydronase to total liver activity. These results were iticized by Lindahl and Evces (1984) who showed tremely low cytoplasmic aldehyde dehydrogenase tivity (below 1%), non-detectable with acetalde'de, in different rat strains and suggested minimal contrll ntribution of the cytoplasm to rat liver aldehyde dehhydrogenase. However, Berger and Weiner (1977) an~d Tank et al. (1981) had reported different major isoz, ~zyme patterns (three to five isozymes) in the cytop toplasm from a closed colony of Purdue/Wistar rats ts and recently Cao et al. (1989) confirmed these results mlts with a new colony of Wistar rats. They also calculat, iculated K~ values of five cytoplasmic isozymes with th acetaldehyde. Unfortunately, they did not report the value of cytoplasmic activity relative to total activity in the liver. Our results alts on subcellular localization of propionaldehyde activity ( mM) indicated tctivity (1 y(1 that ca 7.5% of total rat.t liver activity in the ssociated with the cytoSprague--Dawley strain is assoclat~ ;ity of cytoplasmic aldeplasmic fraction. The intensit' hyde dehydrogenase activity bandss seen se on isoelectric seen focusing gels, the multiband isoenzyme pattern similar to that reported by Weiner ~er and his colleagues in the Wistar strain and the localizatic ation of significant )calization y-aminobutyraldehyde activit ity in the cytoplasm all suggest that the contribution of cytoplasmic aldehyde tal rat liver activity candehydrogenase activity to total not be neglected. Assuming that y-aminoing in the cytoplasm is butyraldehyde activity existing represented only by our partially trtially purified isozyme with an activity ratio of y-aminobl obut' -aminobutyraldehyde to propionaldehyde of 3 : 1, a rmmmum finimum contribution of 4.3% can be expected (0.3577 IU/i IU/g of ~,-aminobutyraldehyde activity + 3 = O.1199 IU/g propionaldehyde L3% of total propionactivity, which represents 4.3! iver; see Table 1). Slight aldehyde activity in the rat liver: modifications of the tissue fractionation procedure and assay system betweena investigators cannot account for such differencess in reported values of cytoplasmic aldehyde dehydrog4 ~rogenase activity...Our experience with human liver and isozymes showed that cytoplasmic aid genase isozymes and the main isozyr
al.
y-aminobutyra tery labile and extremely sensitive to at~ ygen. Therefore, all our buffers used in 1 were extensively degassed, nitrogenated ed 0.1% v/v 2-mercaptoethanol and r fractions were immediately assayed f ~tivity. Such precautions may not neces ~een followed by earlier investigators ware of these problems. The results this paper demonstrate that enzyme(s y-aminobutyraldehyde dehydrogenatic ,~dexclusively in the cytoplasm of huma and rat livers. Isoelectric focus focusing result data from the partial purifi ~urification of from rat liver strongly sugg~ ;est that aldehyde and propionaldeh aldehyde activi 'the same enzyme, which appe~ ~ears to be ydrogenase (EC 1.2.1.3). The ]partially p ae from rat liver metabolizir olizing y-amine le resembles in properties the hhomogenec cde dehydrogenase from hum~ human liver. cies the Km values for y-am ,-aminobutyra] low, suggesting that the enzyr :yme might its metabolism. In the living g organist ,~ is metabolized to yamin,obutyrald( ~mine oxidase which is also a cytopla~ • Thus, the low Km for y-am: ',-aminobutyra] "t:l.llllllLI U t t t y I i:tlUti;ll)" O ~ U U I I ILbined L l l l l t ; O with subcellular locali localization further suggest that th~ the enzyme may be impo ~ortant in metabolism of aldeh~¢des arising from diamines. Ackn~ Acknowledgements--This work was supl: upported by US Public
Healtl Health Service National Institute on Alcohol ? Abuse and Alcoholism Grant AA00186 and Research Resear Scientist Award Alcoh K05 AA00046. REFERENCES
Ambr, broziak W. and Pietruszko R. (1987) (1987 Human aldehyde dehydrogenase: metabolism of putres( autrescine and histamine. Alcohol. din. exp. Res. I1, 528-532. Ambroziak W. and Pietruszko R. (1991) (1991 Human aldehyde dehydrogenase: activity with aldehycde metabolites of amines. J. biol. Chem. monoamines, diamines and polyami; 266, 13,011-13,018. Berger D. and Weiner H. (1977) Rel Relationship between alcohol preference and biogenic alde aldehyde metabolizing enzymes in rats. Biochem. Pharmacol armacol. 26, 841-846. Bergmeyer H. U. and Bernt E. (1974) Lactate dehydrogenase. In Methods of Enzymatic A Analysis (Edited by Bergmeyer H. U.), Vol. 2, pp. 574-579. 574-5 Verlag Chemic GmbH, Wienheim. Bergmeyer H. U., Gawehn K. and Grassl M. (1974) Alcohol ymatic Analysis (Edited dehydrogenase. In Methods of Enzyma by Bergmeyer H. U.), Vol. 1, pp. 428-~ Verlag Chemie ~. 428-429. GmbH, Weinheim. Cao Q.-N., Tu G.-C. and Weiner H. (1989) Presence of cytosolic aldehyde dehydrogenase lsozymes iso~ in adult and fetal rat liver. Biochem. Pharmacol. 38, 77-83. Deitrich R. A., Collins A. C. and Erwin Er V. G. (1972) Genetic influence upon phenobarbiU tobarbital-induced increase ¢de dehydrogenase activity. in rat liver supernatant aldehyde dehy J. biol. Chem. 247, 7232-7236. De Duve C., Pressman B. C., Gianetto Gianett( R., Wattiaux R. and Aplemans F. (1955) Tissue fracti fractionation studies. 6. Intracellular distribution patterns of enzymes e in rat liver tissues. Biochem. J. 60, 604-617. .~thod for protein determin1 protein in cerebrospinal Ft. 5, 218-222.
Aldehyde dehydrogenase 5) Different forms of and their subcellular ta 397, 9-23. szko R. (1989) Human 3n and characterization ),-aminobutyraldehyde. t Tipton K. F. (1980) )genase from ox brain 11. omparative subcellular mase in rat, mouse and 33, 3383-3389. larjanen L. A. (1972) Intracellular localization of aldehyde dehydrogenase in rat liver. Biochem. 3". 127, 633-639. ietruszko R. (1989) Aldehyde dehydrogenase (EC 1.2.1.3). In Biochemistry and Physiology o f Substance Abuse
(Edited b y W ; Inc., Boca Ra Seller N., KncH formation of .' spermine. Bio~ Shum G. T. and in rat liver. C Tabor C. W. al Biochem. 53, Tank A. W., Wt ology and sub in rat liver. B~ Tottmar S. O. C The Th, subcellul~ dehydrogenasc de! Yun S. L. and calculating K~ cal, progress curve prc
1, pp. 89-127. CRC Press, Gittos W. (1981) On the ;riving from spermidine and 23-132. 72) Aldehyde dehydrogenase • 50, 741-748. (1984) Polyamines. A. Rev. 'urman J. A. (1981) Enzymation of aldehyde oxidation acol. 30, 3265-3275. and Kiessling K. H. (1973) and properties of aldehyde ~iochem. J. 135, 577-586. 11977) A simple method for ~m single enzyme reaction ~hys. Acta 480, 1-13.
