Regulation of Microsomal Enzymes by Phospholipids

A bisubstrate kinetic analysis of UDP-glucuronyltransferase (EC 2.4.1.17) has been carried out in forward and reverse directions with p-nitrophenol as aglycone. Reciprocal plots of initial rates of activity indicated that the kinetics followed a sequential mechanism. Product inhibition studies, using UDP and p-nitrophenylglucuronide as inhibitors of the forward reaction, gave a pattern of two competitive and two noncompetitive inhibitions, compatible with a rapid equilibrium, random order kinetic mechanism, or an ordered mechanism of the Theorell-Chance type. Isotope exchange experiments, however, excluded an ordered mechanism. Comparison of the kinetic parameters for the forward and reverse directions showed that the rate at V,,, is 2-fold greater for the reverse than for the forward reaction. At finite substrate concentrations, however, the forward reaction is favored because of the loo-fold higher affinity of the enzyme for p-nitrophenol than for its glucuronide. It was also observed that high concentrations of p-nitrophenol and o-aminophenol have nonspecific activating effects on UDP-glucuronyltransferase. The importance of these findings for the design and interpretation of kinetic experiments is discussed.


From the Division of Molecular
Biology, Veterans Administration Hospital, San Francisco, California 94121, and the Departments of Biochemistry and Biophysics and Medicine, University of California School of Medicine, San Francisco, California 9412.2 SUMMARY A bisubstrate kinetic analysis of UDP-glucuronyltransferase (EC 2.4.1.17) has been carried out in forward and reverse directions with p-nitrophenol as aglycone.
Reciprocal plots of initial rates of activity indicated that the kinetics followed a sequential mechanism. Product inhibition studies, using UDP and p-nitrophenylglucuronide as inhibitors of the forward reaction, gave a pattern of two competitive and two noncompetitive inhibitions, compatible with a rapid equilibrium, random order kinetic mechanism, or an ordered mechanism of the Theorell-Chance type.
Isotope exchange experiments, however, excluded an ordered mechanism. Comparison of the kinetic parameters for the forward and reverse directions showed that the rate at V,,, is 2-fold greater for the reverse than for the forward reaction. At finite substrate concentrations, however, the forward reaction is favored because of the loo-fold higher affinity of the enzyme for p-nitrophenol than for its glucuronide. It was also observed that high concentrations of p-nitrophenol and o-aminophenol have nonspecific activating effects on UDP-glucuronyltransferase.
The importance of these findings for the design and interpretation of kinetic experiments is discussed.
The classical approach to determining the properties of enzymes is to first preljare purified forms. Unfortunately, this has not been possible in many cases and is an especially difficult problem in the case of tightly bound microsomal enzymes. An abundance of recent evidence indicates, in fact, that the properties of many of these enzymes are intimately related to interactions with the microsomal phospholipids (l-9). Not only are activities at 'Vmil, and stabilities altered by perturbation of the lipid phase of the microsomal membrane, but such modifications also influence substrate specificity (8). Thus, interrelations between tightly bound microsomal enzymes and their environments xithin the membrane have important implications for * This investigation was supported in part. by grants from the United States Public Health Service (HE 10027) and the National Science Foundatioll (GB 8248) to Dr. Thomas P. Singer. the manner in which these erlzpc~s A\11 methods now available for pot,rntinlly scparat kg miclosomnl enzymes from their atjtachmerlts to t,he mcmbww \vill :jltc>1 proteirl-lipid interactions, thereby modifying the properties of the enzyme under study. On the other hand, kinetic investigations with a crude microsomal system have been useful in resolving many questions concerning the possible multiplicity of hepatic glucose 6-phosphatase (EC 3.1.3.9) (10-13). Because hepatic UDP-glucuronyltransferase is a microsomnl enzyme the properties of which are altered by modification of microsomd phospholipids (9), kinetics also would seem to be the most useful exTerimenta1 method for determining the number of different species of this enzyme, their substrate specificit.ies, the biothemical basis for the apparently complex and species-specific developmental pattern of glucuronidation reactions (14), and t)he nature of the defects in Gunn rats (15) and in human beings with the Criglar-Najjar syndrome (16). However, despite numerous investigations of this enzyme, its kinetic properties have never been carefully studied.
The kinetic mechanism is undetermined, glucuronidation has been studied only in the forward direction, and a considerable number of assumpt'ions hare been made about the usefulness of standard assay condit,ions (14)(15)(16)(17)(18)(19). This paper presents evidence concerning the kinetic mechanism of hepatic, microsomal UDP-glucuronyltransferase, the kinetic parameters for the enzyme measured in forward and reverse directions, and establishes limitations for the uscful~~cs:: of currently used assay techniques.
The importance of thrx data for the proper design of experiments to resolve some of the u11answered questions concerning the properties of ~~I)~~'-gluc~~rollyltransferase are discussed. with beef liver microsomes. The reasons for the selection of the assay conditions noted below have been discussed in a previous publication (9). With p-nitrophenol as aglycone, the reaction mixture contained in a final volume of 0.5 ml, 0.2 to 0.4 mg of microsomal protein, 0.05 M phosphate buffer (pH 7.1), and UDPglucuronic acid and p-nitrophenol at the concentrations indicated in the figures. For product inhibition studies, 5 mM TTDP or 10 mn~ p-nitrophenylglucuronide was added to the assay tubes. In the latter instance, 5.0 mM saccharic acid-l ,4-lactone was also added in order to inhibit P-glucuronidase.
The course of the reaction at 37" was followed by serial removal from the assay tubes of 0.1.ml aliquots which were added to 2.0 ml of 1.8% trichloroacet,ic acid (w/v) in order to stop the reaction and precipitate the protein.
After centrifugation, the supernatants were decanted int,o a tube containing 0.04 ml of 10 N KOH, which raised the pH of the supernatant to > 10.0; the optical density was measured at 400 nm. Activities, which are initial rates for all of the data presented in this paper, are expressed as mpmoles of p-nitrophenol conjugated per min per mg of microsomal protein, calculated from the decrease in the absorbance of p-nitrophenol. The UDP-glucuronyltransferase-catalyzed synthesis of ITTIP-glucuronic acid and p-nitrophenol from UDP and p-nitrophenylglucuronide (the reverse reaction) was assayed at 37" using the same conditions as in the forward reaction except that 5 mM saccharic acid-l, 4-lactone was added to each assay (20). Activity measurements and units are the same as for the forward reaction.
The activity of UDP-glucuronyltransferase with o-aminophenol as aglycone was assayed by the method of Dutton (21) except that Pi buffer, pH 7.6, was substituted for Tris.
As with p-nitrophenol, initial rates of activity were determined. Protein was determined by the biuret method (22).
For the isotope exchange studies, microsomes were incubated ^. ^^ at 37" with 10 mM p-nitrophenylglucuronic acid, 10 mM UDP, 5 mM saccharic acid-l ,4-lactone, 0.05 M Pi buffer (pH 7.1), and either 0.1 mM p-nitro[14C]phenol or 10 mM UDP-[14C]glucuronic acid. Serial aliquots were removed and boiled to stop the reaction. After centrifugation to remove precipitated protein, the reactants were separated chromatographically on Whatman No. 1 paper with a I:2 mixture of 1.0 M ammonium acetate, pH 3.6, and ethanol (23). Spots were located via ultraviolet absorption and by spraying with dilute KOH, cut out, and counted in a Packard Tri-Carb liquid scintillation counter.

RESULTS
Kinetic Constants of UDP-glucuronyltransjerase for Forward and Reverse Reactions-Initial rates of UDP-glucuronyltransferase activity were measured as a function of the concentration of TJDP-glucuronic acid at several fixed concentrations of p-nltrophenol and as a function of varying concentrations of p-nitrophenol at several fixed concentrations of UDP-glucuronic acid. The data were plotted in double reciprocal form as in Fig. 1, and the intersections of the primary plots were used to estimate K, for the variable substrate.
Secondary plots of the intercepts on the l/v axis were used to obtain l/V,,,,.
The UDP-glucuronyltransferase reaction was studied in the reverse direction in the same manner as for the forward direction.
The primary double reciprocal plots for initial rates of activity as a function of the concentration of p-nitrophenylglucuronide at several fixed concentrations of UDP are presented in Fig. 2. The kinetic constants determined graphically for the forward and reverse reactions are tabulated in Table I. KUDPGA1 and KPmNP are the K, terms for each substrate when it is the first to be bound and the K' terms when each is the second substrate to be bound.
UI)P-glrtcrtrotlyltr:~tlsfer~~se was assayed as described in Fig. 1  and product is rellerted also in tjltcl fact that KlpeNpG is larger than K'p.Np. As with the forward renAon, hoverer, the npparent bindittg affittity for each substrxlr in the reverse reaction is incren*ed (lower K' as cotnlxaretl to K) rvhen the substrate is the second to bhitl to ttic cirzytne.
Frutlw, the data in Table I ittdicntc t,hat thr acstivity at' T:,,.., is grcatc~r in the reverse directiotl than in the forw:rrd dircvtion.
However, because of the diffprctice bc+~veetr Zip-spG atid K _ p x\Tp one would predict that the equilibriuttl of the rc:tcation l'avors the forward direction. This follon-:: :~lso from n considrt~ntiott of the Haldane relationship for this reaction (24). transferase-catalyzed synthesis of p-tlitt'ol)hciiylglttcltrott'tl(,, butt they do not allow for a distinction between an ordered or t~trtttlotn form of sequential mechanism (25). In order to distinguish bctween these two kinetic mechanisms, initial rates for thr forward reaction were measured in the presencle of the l)roduc+s, t-D I' or p-nitrophetlyl~lncurorlidr (25). The data for each product inhibition study are plotted as l,'\ versus the reciprocal of the concentration of the variable substrate at t'wo concetttratjiotts of the fiscd hubsttxtc.
'l'lrc~ data in Figs. 3 and 4 yield a 1vtttcrn of trio c*oml&itivc> :ultl tn-o t1m1coml)etitive product itth;bitiotts, wltirtt is wttsiistettt, wit11 :t t':tttdom order kinetic mechanism but, dots not crcludc :ttt ortlcrcd mechanism of the l')ic,orell-Ch:rrt(~e type (25). The activity of IJDP-Rlucr~rott~ltratteferttse \viltt p-ttitrol)ltrttol as aglycone is several-fold greater in guinea pig liver than in beef. The primary reciltrocal plots and the Ixttterns of I)rotluc? inhibitions are the same in these two species; and t,herr is IW l:DP'-&cxronic acid lt\~ro~)ltos~)hatase activity it1 auittea pig microsomes. isotope were nearly identical (Table 11). These data, therefore, exclude a11 ordered mechanism of the Theorell-Chance type. The scheme for the over-all resction mechanism is presented in Fig. 5 With beef liver microsomes, measurements of the sctivit,y of UDP-glucuronyltransferase ill the presence of concentratiolls of p-nitrophenol greater than 0.6 to 0.8 mM, the exact concentration varying with the microsomal preparation, led to two dZiculties in the interpretation of the data. Thus, at pH 7.1, incubation of microsomes wit,h concentrations of p-nitrophenol in excess of 0.8 mM solubilized a compound which absorbs at 400 nm and which was not precipitated by trichloroacctic acid. This solubilization of all interfering chromophore \vas unaffected by TTl)l'-glucurollic acid. -1s a result, assays of TTI)P-glucllroiiyltrallsfer:1se at relatively high concentrations of p-r~itrol~liei~ol will be fnlscly low. ;\lthough it is theoretically possible to derive and al)l)ly apl)ropriate correction factors, a more serious I)roblern observed at high caollcelltrations of p-nitrophenol was anomalous kinetics.
The primary plot's of 1 /v vers76s 1 /[UDl'-glucuronic acid] at fixed concentrations of p-nitrophcnol greater than 0.6 to 0.8 IIIM did Ilot intersect in the same quadrant as those derived at lower collccntrations of p-nitrophellol (Fig.  Sil).
The exact reason for this rffect is unclear as yet; but the plots suggest that relatively high concentrations of p-nitrophenol activate IJl)l'-glucurorlSltrallsferase and decrease affinity for UDP-glucuronic acid. Decreasing the 1)H of the assay medium from 7.1 to 6.1 augmented this effect and the solubilization of the interfering chromophore, in that both were observed at concentrations of p-nitrophenol lower than those required at an assay pH of 7.1 (Fig. 6B). Since the pK of p-nitrophenol is 7.1 and the effects noted above are augmented at pH values below 7.1, it is Lkely that the phenol, as romparcd to the pht>nolatt> form of this substrate, was responsible for the difficulties ritrtl above. We should stress that this effect of high concentrations of p-llitrophenol does not represent substrate inhibition by binding at the UDP-glucuronic acid site since there was an increase in activity at saturating concentrations of UDP-glucwonic acid. It seems more likely that activation by high concentrations of p-nitrophenol results from a conformational change in ITDP-ghcuronyltransferase which in addition to increasing V,,,,x, increases K UDPGA . Activation by p-nitrophenol was Ilot limited to beef livrr microsomes, since a similar effect was observed with ITl)l'-gh~curon~ltransferase ;II guinea pig liver microsomes.
IIowcvcr, it was found that, if t,he concentratiou of o-aminophenol was increased sufficiently, inhibition of conjugation resulted.
The concentration of o-aminophenol at nhich inhibition occurred was dependent on t.he ratio of microsomal l)rotein to phenol.
One of t.he import,ant unsolved problems concerning the funct.ion of lX)P-glucuronyltrandera~se is the number of such enzymes. The data present,ed above have important implications for t,he limitations which must be placed 011 the interpretations of previous investigations of this problem and also define experimental approaches useful for its resolution.
It is often assumed that the activity of TTDP-glucuronyltransferase is measured at saturating concentrations of either aglycone or UDP-glucuronic acid (17)(18)(19).
It is clear, however, from comparison of K values in Table I and commonly used assay conditions as reported in the literature (14-19, 21, 26, 27) that so-called "saturating" conditions were not attained.
Alore import,antly, at least with o-amino-and p-nitrol)henol, t,he a&cones most often studied, the enzyme cannot be assayed at sat.uratillg concentrations because of their as yet undefined effects 011 microsomal structure and activation of ITJ)T'-glucuronyltrai~sfer:rse.
The only valid parameter which earl be utilized to indicate the amount of ellzyme present in microsomes is :&iv&y at V,,,, as determined above in Fig. 1 and l'able I.
Thus, although activit,y dctcrmined at any fixed set of xubstrate concentrations is not a useful measurement of the total amount of Ul) l'~glucurot~~ltrar~sferasc, it. is a.n implicit. assumpt.ion of rtudies of the dcvelopment,al pat.tern of UDP-glucuronyltransferasea tha.t an indes of t,he amount of each TTDP-glucuronyltranrferase present, was determined by assaying at a single set of substrat,e concentrations.
Such studies cannot distinguish, however, bet.\vcen changes in the amount of enzyme and changes in KsglyoonP. In view of the already demonstrated alteration of Kani l i ne :uld Keth?;ltl~orphine for hepatic microsomal aniline hydrosylase and cthylmorphmc demethylase (28), respectively, such developmental clw~ges in the kinetic paramrtcrs of I71 )l'-glucurorlyltraIisferase would not be unique. ,$s pointed out above, o-amino-and p-nitrophenol alter the properties of LX)P-glucuronyltransferase, which presents still another problem for the design of csperiment,s conccrnillg the problem of the multiplicity of this t'ype of enzyme. Thus, it may not be possible to obtain meaningful data using o-amillophenol as an inhibitor of the synthesis of p-~~itrol)hnlylglucuronide because of the apparent nonspecific activating effrcts of o-aminophenol on UDP-glucuronyltral1sfernse. Also, 011 the basis of these data, it is important to exclude carefully the l'oqsibility that any pot.ential substrat.e of TTT)I'-plucuro~i~ltrall~fernse has nonspecific effects 011 microsomal structure and function before embarking on kinetic studies. This type of rffrct inight account, in part, for the somewhat confusing data :~lr(xdy 1~1). lished 011 the kinetics of 1Tl)P-gl~~ruro~~~lt.rat~sfcr~t~st~ (27). 'I'hrts, in some experiments, St,orey obsorvt~d t,hat. o-nrllitiol)lic~floI :ICtually stimul:~ted the ~lllcurorl;dntioll of 2'-iiitrol)llcllol (a?'). In addition to point.& out I)ot,ciitinl ~~roblcnis in tlltl :IMS:IJ of 1: I )I'-glucnt~otlyltratisf~r:~s(~s, the dat.a itldic;~.tc ~(~vc~r:~l ~~sc$rll t,echniquc>s for studyitlg their I)ropcrtics :u~d mrdt~il)lic~ity. 'I'll:lt the revcrsc reaction is easily mrasurcd is iml)ort:mt,, sillcac ~)rob-1e1ns prcsrnt)ed by the insolubility of substratc>s ohs their IIOIIspecific effects on the microsomes may be :Lvoidcd 1)~ :iss;lyiilg UI)P-glucuronyltransferase in the reverse dir&ion. 'I'his I'?versibility also allows for expansion of the scope of I)otentinl inhibition experiments in that, for example, the efYect of p-nit,rophenol as a product inhibitor can be studied on the rate of 1-l) t-'glucuronyltransferase-catalyzed hydrolysis of o-aminophenylglucuronide.
Knowledge of the kinetic mechanism is useful in another way. Since the mechanism is random order, ral)id equilibrium, the graphically determined K nDPGri is the dissociation constant for the UDP-glucuronic acid-enzyme complex; if two aglycones are glucuronidated by a single enzyme, KunPGa must be identical when determined with either aglycone.