plaut_1974

7/28/2019 plaut_1974

1/9

THE JOURNAL OF BIOLOGICAL CHEMISTRYVol. 249, No. 6, hue of March 25, PP. 1848-1856, 1974Printed in U.S .A.

Action of Magnesium Ion on Diphosphopyridine Nucleotide-linked Isocitrate Dehydrogenase from Bovine HeartCHARACTERIZATION OF THE FORMS OF THE SUBSTRATE AND THE MODIFIER OF THE REACTION*

(Received for publication, November 22, 1972, and in revised form, July 6, 1973)GERHARD W. E. PLAUT, VERN L. SCHRAMM, AND TADASHI AOGAICHIFrom the Department of Biochemistry, Temple University School of Medicine, Philadelphia, Pennsylvania19160

SUMMARYThe act ivi ty of DPN-linked isocitrate dehydrogenase is

dependent on the concentration of magnesium n-isocitrate(MgI-) under conditions where the concentrations of freeMg2+ and isocitrate approach zero and saturation, respec-tively. These results can be interpreted in terms of mag-nesium isocitrate as the active substrate, however the presentdata cannot rule out the possibility of free Mg2f and iso-citrate as the substrates if the reaction mechanism is rapidequilibrium ordered with free Mg*+ being the first reactant.In the absence of ADP and in presence of 0.33 mM DPN+,&o.~ of magnesium DL-isocitrate has been found to be 0.2,0.4, 0.5, and 1.0 mM at pH 6.65, 7.2, 7.4, and 7.75, respec-tively , with values of Hill slopes (n) near 2. Excess freeMg2+ causes apparent inhibition which is competitive withrespect to magnesium isocitrate, the proposed substrateof the reaction. The inhibition by Mg*+ is much moresevere at pH 6.65 than at pH 7.2 and above, e.g. the inhibi-tion constant for Mg2+ is 40. to 80.fold larger at pH 7.2than at pH 6.65. In contrast to the results which are con-sistent with magnesium isocitrate as the substrate, freeADP3- is the modifier of the enzyme and MgADP- is inac-tive . At pH 7.2 and 7.4, increasing concentrations of ADP3-lead to decreasing values of So.s for magnesium isocitrateaccompanied by a decline of Hill slopes from n = 2 to nearunity ; these changes are less pronounced at pH 6.65. Thevalues of V,,, are the same in the absence and presenceof ADP at each pH tested; however, V,,, at pH 6.65 isabout one-half of that at pH 7.2 and above. There is aninterdependence in the interaction of the enzyme withsubstrate and modifier. Thus, at pH 7.2 and 7.4, the valuesof K, for ADP3- decline with increasing concentrations ofmagnesium isocitrate; however, the Hill slopes for ADP3-are not influenced by substrate concentration and remainconstant at m = 1 between pH 6.65 and 7.75.

A reaction model has been proposed in which initial bind-ing of substrate to the enzyme at one site leads to a con-formational change and subsequent binding of magnesiumisocitrate at an additional site which is catalytically active.

DPN-dependent isocitrate dehydrogenase from animal tissues(EC 1.1.1.41) catalyzing the reactionn-three-Isocitrate + DPN+ -

or-ketoglutarate + CO2 + DPNH + H+has been shown to require Mg2+, Mn+, or Co2+ for activi ty(l-8). Mn*+ is a more effect ive activator than Mg*+, exhibitinga lower apparent K, and a somewhat higher Vmax;however,severe inhibition has been noted at higher concentrations ofMn2f (8, 9). With the enzyme from liver the apparent K, ofMg2+ was lowered by increased isocitrate and decreased furtherby the positive modifier ADP. The question of whether or nota complex of the divalent metal ion with isocitrate is the truesubstrate for the enzyme was not resolved in the earlier work (8).

In recent systematic studies of the role of Mn2+ with a purifiedenzyme from porcine heart, Cohen and Colman (9) have pro-posed that dibasic isocitrate is the active form of the substrateof the enzyme with an apparent K, (approximately 30 PM)independent of Mn2+ concentration and pH. Inhibition bysubstrate, reported at pH 6 but not at pH 7 and 8, was attrib-uted to a complex o f manganous dibasic isocitrate. Activationby a number of chelating agents (GDP, UDP, and citrate) wasobserved (9) and the effe ct explained as due to removal of Mn2+from the isocitrate complex resulting in an increase in the activespecies of isocitrate. In accord with earlier observations (2, 4,6) these investigators found that ADP lowers the apparent K,of total isocitrate; but they also reported that ADP lowers theK, of dibasic isocitrate. They propose that activation by ADPis caused by direct modification of the enzyme, and by chelationof Mn2+ which leads to increased free dibasic isocitrate. Thus,MS+ would function as a negative regulator; controlling thesupply of the substrate (free dibasic isocitrate) and the inhibitor(isocitrate chelate). Nevertheless, Cohen and Colman (9) haveconfirmed previous observations (l-8) for the absolute require-ment for act ivi ty of divalent metal ions, although the mech-anism of interaction between Mn2+ and enzyme was not eluci-dated.

* This work was supported in part by Grant AM 15404 from theNational Institute of Arthritis and Metabolic Diseases, NationalInstitutes of Health.

yg,

y

,

jg

7/28/2019 plaut_1974

2/9

1849Certain of the observations with the enzyme from porcine

heart differ from previous work with the dehydrogenase fromother animal tissues. Thus, only inhibition of activity was ob-tained with a number of nucleoside di- and triphosphates (otherthan ADP) with enzymes from bovine and rat heart (2, 4),Ehrlich ascites tumor (6) and liver (S), and the effect has beenattributed partly to chelation of divalent metal ions. WithATP, inhibition competitive with isocitrate occurs at a lowerconcentrat ion than with other nucleotides (2) ; while it appearsto be accompanied by binding of divalent metal ions (4) thekinetic effect m ay be due to the specif ic interaction of enzymeand ATP which has been shown to occur without added diva lentmetal ions (10). Citrate is a positiv e modifie r of the enzymefrom yeast (ll), Neurosporu (12), and certain other m icrobialsources, and is reported by Cohen and Colman (9) to activatethe enzyme from hog heart; however, this was not observedwith the hog liver enzyme (8).

Furthermore, it is diff icul t to explain with the mechanism ofCohen and Colman (9) protection by the combination of mag-nesium (or Mn*) and isocitrate against inhibition by low con-centrations of mercurials (3, 8) or by 5,5-dithiobis(2-nitro-benzoate) (13), not obtained with either isocitrate or divalentmetal ion alone.

Therefore, a further clarification of the role of divalent metalions became desirable. In the studies presented here the roleof Mg*+ was investigated. This has the advantage that Mg*+ isless inhibitory than Mn2+, an inhibition which can be irreversibleunder certain conditions. Evidence will be presented that isconsistent with magnesium isocitrate acting as the true sub-strate, free ADP is active, and MgADP- is inactive as a modi-fier, and free Mg2+ is not required for activity.

EXPERIMENTAL PROCEDUREMaterials

The following chem icals were purchased from the comme rcialsources indicated. Inorganic salts were analytical reagent gradefrom Mallinckrodt Chem ical Works. 1,3-Dimercapto-2-propa nolwas from Chemical Procurement Co. ADP, DPN+, EDTA,Hepes,2 Pipes, DL-isocitric la&one, and n-isocitric acid (mono-potassium salt) were from Sigma.

DL-ISOCitrate lactone was recrystallized by dissolving 1 g in 25ml of ethyl n-butyrate under reflux followed by treatment with50 mg of Norit A and filtration of the hot solution. The colorlesscrys tals formed overnight at room temperature were collected byfiltration, washed with a small quantity of ethyl butyrate, anddried in a vacuum for 3 hours at 95. Recrystallized DL-isocitriclactone was recovered in 75 to 80% yield with m .p. 162. Beforeuse, isocitric lactone was hydrolyzed in alkali as described previ-ously (14) and then adjusted with HCl at 25 to the exact pHused in the incubation mixture.

DPN-linked isocit rate dehydrogenase from bovine heart,3which was prepared by the method of Giorgio et al. (15), showed asingle band in polyacrylamide gel disc electrophoresis and had aspecif ic activity of 28,000 nmoles of DPN+ reduced per mg ofprotein per min at 25 under the conditions of assay reportedpreviously (14).

Methods

the concentrations indicated in the text, tables, and figures. Th econcentrations of ADP and DPN+ were determined spectrophoto-metrically at 260 nm (16) and the magnesium content of stock so-lutions was assayed by atomic absorption spectrometry.4

Reaction mixtures containing all components except enzyme ina final volume of 1 ml in silica cuvettes of l-cm light path werebrought to temperature equilibrium at 25 in a thermostatedcopper block cuvette holder. Reactions were started by the addi-tion of 5 to 10 ~1 of diluted enzyme solution. Initial reactionrates were determined at 25 by following the reduction of DPN+spectrophotometrically at 340 nm in a Gilford model 240 spectro-photometer operated in the 0 to 0.2 absorbance range. Thespectrophotometer output was recorded with a Honeywell model190 recorder at a chart speed of 6 to 12 inches per min.Stock solutions of the enzyme containing 1.2 mg of protein perml were stored at -90 in a solution containing 10% glycerol, 5mM sodium phosphate (pH 7.2), and 0.1 mM dithiothreitol. Fora days experiment enzyme stock solutions were diluted 1:5 or1:lO into a solution containing 100 mM sodium Hepes at pH 7.2,0.1 mM of 1,3-dimercapto -2-propanol, and 764 mM Li.$JOa . Underthese conditions no activity was lost in the diluted sample for 21hours at 0.

Velocitie s are expressed as change in absorbance at 340 nmper min; the values are comparable between experiments sincethey were calculated in all cases on the basis of the same amountof enzyme protein (1.2 fig) per ml of reaction mixture.

Calculations-Grzyb owski et al. (17) have reported the stab ilityconstants of magnesium isocitrate complexes asK = [MgH1l = 0.027 mM-l1 [Mg*l[HI+]

and[MgI-1Kz = ~Mg~1~13-l = 0.521 mM-

where MgHI = magnesium I I-isocitrate; HI = H-isocitratez-;and 13- = isocitratea-.According to the acid dissociation constants of isocitric acid(pK,l = 3.02, pK,2 = 4.28, pK,3 = 5.75) and the acid dissocia-

tion constant for magnesium H-isocitrate (17)K

c= [H+lDkI-I

D&HI1 = 3.44 x lo-2 rnMIsocitrate should be almost fully ionized above pH 7 (i.e. [I] %[I]), and magnesium H-isocitrate formation should be negligible.Therefore, the equilibrium can be described essentially by theconstant Kf. With the terminology used in this commun ication,the equilibrium is described by

(1)where MI = (magnesium isocitrate)- complex; M = free Mga+;Z = free isocitrate.

Above pH 7, the value of KI is assumed to be 0.521 mM-1; how-ever, at pH 6.65, the concentration of H-isocitrate becomessignificant and an apparent equilibrium constant of 0.463 rnM-1,calculated from Kz and pK,3 has been used to estimate the con-centration of magnesium isocitrate.The concentrations of magnesium isocitrate can be calculatedfrom total magnesium (M!Z) and total isocitrate (IT) by sub-stitution into Equation 1

IMZI([MI] - [MZI)([ZT] - [MI]) = Kr

Assay-All incubation mixtures contained 0.33 mM DPN+ and167 mM sodium Henes at DH 7.2 to 7.75 or 167 mM sodium Pines and solution of the resulting quadratic equationbuffer at pH 6.65. * Isocitrate, ADP, and MgSO4 were added- at KAWI + [MT]) + 1 f

1 C. Fan (1971) unpublished observations. [MI] =d(KAZT] f [MT]) + 1Y - 4K,*[MTl[ZTl (3)2Kr2 The abbreviatio ns used are: Hepes, N-2-hydroxyethylpipera-

zinc-N-2-ethanesulfonic acid ; Pipes, piperazine-N,N-bis(2-ethanesulfonic acid).

The concentrations of free Mg2+ and free isocitrate can then3 These enzyme preparations were made available to us by Dr. 4 We wish to thank Dr. M. C. Scrutton for making these de-

C. Fan. terminations.

byguest,onJuly22,2012

www.jbc.org

Downloadedfrom

7/28/2019 plaut_1974

3/9

1850be calculated from the relationships

[Ml = IMTI - [MI]and

[ZI = IZTI - [MI]For calculation of concentration of the magnesium ADP com-plex (MgADP-) the stability constants of OSullivan and Perrin

(18) were used with values of KA of 2.3, 2 .4, and 4 .1 rnrv-1 at pH7.2, 7.4, and 7.75, respectively. The stability constant5 at pH6.65 was estimated to be 1.4 mM0. The concentrations of freeMg*+ were calculated from total magnesium, total isocitrate andtotal ADP (AT) according to the relationship of Equation 4,which was solved by use of the computer and the Newton-Raphsoniteration technique.

[Ml8 + & + j& + [ATI + UT1 - [MT] 1Ml*+ [ITI - [MT1 + MT1 - [MT] +KA KI K*] [Ml (41

A!ELO-- --Concentrations of MgADP- (MA) and magnesium isocitratewere then calculated from the relations

fMzl = [MIUTIK,1 + KIM (6)

and concentrations of free ADP (A) and free isocitrate werecalculated from the conservation equations6

Since the binding of Mg*+ to DPN+ is relatively weak and theconcentration of DPN+ is low it has not been considered here inthe distribution of MgZ+ among the various forms.Analysis of Data-The nature of initial velocity curves wasdetermined by graphical analysis and then analyzed by fitting tothe appropriate compute r programs developed by Clelan d (19)using a CDC 6400 digital computer. Data which gave lineardouble reciprocal plots of initial velocity against substrate con-centration were fitted to Equation 7.

(7)Data which conformed to linear competitive inhibition werefitted to Equation 8, with the exceptions discusse d underResults.

Plots which were rectangular hyperbolas, but did not pass throughthe origin in plots of ZI against ADP, were fitted to Equation 9.

I ADP\v = vo (9)

where vg is the initial velocity of the reaction in the absence ofADP. The maximal velocity will be given by setting ADP -+

6 OSu llivan and Perrin (18) reported that KA = 2.4 rnM-1 atpH 7.4. They estimated that KA = 2.0 rnM-r at pH 7.0 by con-sideration of the equilibrium HADP+ ti H+ + ADP5-, pK =6.65. The interpolated value of KA at pH 6.65 was calculatedfrom the same assumptions.li V. L. Schramm (1972) unpublished.

~0, in which case:V 00)

Th e K,,, for ADP (ADP concentration required for one-halfmaximal increase in velocity (u)) is given by the relationship:Km = KD.

Calculated values of maximal velocities obtained with theaid of computer fits of the data were used in the construction ofHill plots. Details underlying these calculatio ns are describedin the text.

All experiments were done in the presence of a constant con-centratio n of DPN+ (0.33 mM). Under the standard assay con-ditions used to determine the activity of the enzyme (i.e. pH 7.2and Mn*+ as activator (14)), Km of DPN+ is approximately 0.08mM (13), and the velocity with 0.33 mM DPN+ is about 80% ofV msx

RESULTSInitial Velocity as Function of Total Isocitrate and Mgzf-

When experiments were done with different fixed ratios of totalmagnesium to total DL-isocitrate, the results shown in Fig. 1A

FR EE O L - ISO C ITR ATE,mM

IIx) 2.0 4.0 6.0MASNESIUY CL-ISOCl fRAT2,mW

FIG. 1. Changes in substrate concentrations at fixed ratios oftotal is ocitra te (IT) and total Mg*+ (MT) at pH 7.2. Subs trateconcentrations were varied by addition of solutions with [totalMga+] to [total isocitrate] fixed at the ratios: 0.5, 0 ; 1.0, X ; 2.0,a. A, velocity X10 as a function of free nn-isocitrate concentra-tion; B, velocity as a function of magnesium nn-isocitrate; C,Hill plot of velocities corrected for Mg2+ inhibition as a function ofmagnesium isocitrate. In A and B the points are fitted by smoothcurves and in C by the method of least squares.

yg,

y

,

jg

7/28/2019 plaut_1974

4/9

1851

1 1.1 I * 1 aI0 0.04 0.08 0.12A ,M-[I TlFIG. 2. The effect of varying total isocitrate at fixed levels oftotal magnesium at pH 7.75. Concentrations of total magnesiumwere fixed at the levels shown by the numbers in the figure. Totalnn-isocitrate [IT] var ied between 10 and 100 mM. The reactionmixtures contained 0.167 M Hepes buffer at pH 7.75 and initialvelocities (AA3d0 X 10 min-I) were determined as described

under Methods. No act ivi ty was observed in a control experi-ment with 166 mM nn-isocitrate in the absence of added magnesium.The lines were fitted to the points by the method of least squares.were obtained. These results indicated, as has been prev ious lyshown (l), that magnesium is required for catalysis. In orderto determine whether the major ionic species of isocitrate whichwas present in the experiment, namely magnesium isocitrate-should be tested as a potential active substrate species, the con-centration of magnesium isocitrate- was calculated. The rela-tionship between initial velocity and magnesium isocitrate- con-centration is shown in Fig. 1B. Since there appears to be acorrelation between reaction rate and magnesium isocitrate-concentrat ion (Fig. 1B) as well as between rate and free iso-citrate concentration, the following species were considered aspossible substrate requirements for the isocitrate dehydrogenasereaction: (i) free Mg2+ and free isocitratea-; (ii) free Mg*+ andfree H-isocitrate2-; (iii) free Mgti and magnesium isocitrate-;(iv) only magnesium isocitrate- complex; (v) only magnesiumH-isocitrate complex; and (vi) both magnesium H-isocitrate andmagnesium isocitrate- can be substrates,The results of Fig. 1 do not eliminate any o f the above possi-bilities, but suggest that there is an interdependent mechanismof substrate addition since plots of initial velocity against eitherfree isocitrate or magnesium isocitrate- appeared sigmoidal.

Requirement of Free JZagnesium as Substrate-It has recentlybeen reported that H-isocitratez- and free Mgfi are the requiredsubstrates for the pig heart isocitrate dehydrogenase (9). Inorder to test the hypothesis that free magnesium is required forthe bovine heart enzyme, initial velocities were measured at pH7.75, where the ratio of isocitrate+ to H-isocitrate* would beexpected to be 100. In this experiment, illustrated in Fig. 2,the concentration of total isocitrate was increased at fixed mag-nesium concentrations so as to increase the concentration ofmagnesium isocitrate- while decreasing the concentration of freemagnesium. It will be apparent that at the ordinate interceptsin Fig. 2, the free Mg2+ concentration approaches zero. Sincethe ordinate intercept values are strongly dependent on the con-centration of total magnesium the following conclusions can bereached.

1. Any reaction mechanism which requires finite concentra-tions of free Mg2f can be eliminated since the observed velocit,ies

at the ordinate represent initial rates when all of the magnesiumis converted to magnesium isocitrate- and magnesium H-iso-citrate. This eliminates Mechanisms i, ii, and iii except for thepossibility of rapid equilibrium-ordered mechanisms with freemagnesium adding fi rst followed b y free isocitrate. The rapidequilibrium-ordered mechanism gives the same initial veloc ityequation as a mechanism in which magnesium isocitrate is thesubstrate and free magnesium is a competitive inhibitor withrespect to the complex. Thus, the two mechanisms cannot bedistinguished by the present kinetic approach. However, sincerapid equilibrium-ordered mechanisms with metal adding fi rstoccur very rarely for metal-requiring enzymes, and sincemetal-chelate substrates are commonly found in biological sys -tems, the remainder of the results will be interpreted in terms ofmagnesium isocitrate as the active substrate for the enzyme.2. It appears that free isocitratea-, which is the predominantform of isocitrate present under these experimental conditions,does not act as an inhibitor of the reaction, since no apparentsubstrate inhibition occurs at 100 mM concentrations of totalisocitrate.

3. Initial velocity is a sigmoidal function of magnesium iso-citrate- concentration. A replot of the ordinate intercepts ofFig. 2 as a function of reciprocal magnesium isocitrate- concen-tration was nonlinear, concave up, and gave a SO.~ value of 0.87mM for magnesium isocitrate-. This experiment cannot dis-tinguish between magnesium isocitrate- and magnesium H-isocitrate as substrates, but since the experiment was done atpH 7.75, the ratio of magnesium isocitrate- to magnesium H-isocitrate is approximately 1900 : 1. Therefore, if magnesiumH-isocitrate is the substrate, the L&, for magnesium H-isocitratewould be 0.46 PM. The possibil ity that magnesium H-isocitrateis the substrate for isocitrate dehydrogenase is considered lesslikely than the magnesium isocitrate- complex being the sub-strate since the predominance of the magnesium isocitrate-complex would make magnesium isocitrate- inhibition a distinctpossibility . If both magnesium isocitrate- and magnesium H-isocitrate act as substrates with similar kinetic properties, onlythe magnesium isocitrate- complex need be considered as themagnesium H-isoc itrat,e contribution to the kinetic patternswould be negligible. For the remainder of the kinetic studiesdescribed here, magnesium isocitrate- should be interpreted asbeing the sum of the magnesium isocitrate- and magnesium H-isocitrate complexes.

Inhibition by Mg2+ in Presence and Absence of ADP-In initialexperiments (Fig. 1B) lower veloci ties were observed for equiva-lent magnesium isocitrate- concentrations at elevated Mg2+ tomagnesium isocitrate- ratios. This result suggested that freeMg+ mav act as an inhibitor of bovine heart isocitrate dehy-drogenase. In order to test this hypothesis, experiments weredone with magnesium isocitrate- as the substrate and the con-centration of free Mg*+ was varied.

Evidence has been obtained in several experiments that in theabsence of ADP the Hill slope for magnesium isocitrate- ap-proaches the value of n = 2 and that plots of 1 /v against 1 /[MI12are linear over the range of magnesium isocitrate- concentrationsused in these experiments. For calculation of magnesium inhibi-tion constants it was assumed, therefore, that the velocity of thereaction in the absence o f free Mg2+ (v) can be described by

v rnSXx DfZ12 = K $ [Ml]2 (11)

Double reciprocal plots of initial velocity as a function of [MZ12at pH 6.65 and 7.2 yielded intersecting lines (Fig. 3) which gave

byguest,

onJuly22,2012

www.jbc.org

Downloadedfrom

7/28/2019 plaut_1974

5/9

1852good fit s to the equation for competitive inhibition:

(12)

where &I is the magnesium isocitrate complex, JZ is free Mg2+,Kmic is the magnesium inhibition constant, and K is an apparentconstant which should be equivalent to (X0.5)2 of magnesiumisocitrate. Correction for inhibition by free Mg2+ can then beexpressed by the relationship

The kinetic con&ants obtained from the computer analyses arerecorded in Table I. These constants could be used in Equation13 to calculate the uninhibited initial velocities of reactions con-taining inhibitory levels of free Mg2+.Between pH 7.2 and 7.75 in the presence of relatively highconcentrations of ADP the Hill coef ficients for magnesium iso-citrate- approach values of n = 1. In an experiment similar tothat described in Fig. 3 inhibition by Mg2f at pH 7.2 was testedin the presence of 0.3 mM ADP and was found to be competitivewith magnesium isocitrate. The constant,s obtained from fit s ofthe dat,a to Equat.ion 8 are reported in Table I. In the presenceof the modifier the correction of velocity can be expressed in theusual way by

4 0 4 6 12 16

(14)

FIQ. 3. Inhibition by Mg*+. A and B, double reciprocal plotsof velocity versus [magnesium isocitrate]* (MI). The numbersabove the lines refer to concentrations of free Mg2+ as millimolar.The lines were fitted to the points by a computer program (19)and Equation 8. A, pH 7.2; B, pH 6.65. The constants derivedfrom these fi ts are reported in Table I,

where K, is the apparent Michaelis constant of magnesium iso-cikate and f(i is the inhibition constant of free Mg2+.As indicat,ed in Equations 13 and 14, the methods for calcula-

t,ion of inhibition correction factors di ffer in the absence or pres-ence of ADP, and at pH 7.2 the numerical values of the con&&sobtained and the units o f K are not the same (Table I). Never-theless, at comparable concentrations of Mg2f and magnesiumisocitrate the inhibition correction factors are similar, whencalculated from the constants of Table I. This implies, at leastat pH 7.2, that the relative kinetic constants for Mg2+ and mag-nesium isocitrate are not changed substantially by the modifier.Initial Velocity as Function o f Magnesium Isoci frafe- Concen-tration-when the correction factors from Table I are appliedto experiments where several fixed levels of magnesium isocitratehad been maint.ained by varying concentrations of free isocitrateand free Mg*+ at pH 7.2 (Fig. lB), the corrected velocities aredependent only on magnesium isocitrate- concentration. Thedata of Fig. lB, corrected for Mg2+ inhibition, are shown as aHill plot in Fig. 1C. A Hill coeff icient of 1.6 and an SO.s valueof 0.75 mM was obtained from the data. The results of similarexperiments at pH values of 6.65, 7.2, 7.4, and 7.75 are sum-marized in Table II . The X0.s for magnesium isocitrate- in-creases from 0.18 to 0.95 mM as a function of pH, while the Hillcoeff icient remains constant near 2.0.

Experiments at pH 7.2 using u-isocitrate as substrate gaveapproximately the same Hill coe fficient as when nn-isocitratewas used (1.8 and 2.0, respect ively). The LS&, value for D-isocitrate was one-half of that found when nn-isocitrate was usedas substrate. This result is consistent with the previous ob-servation that n-three-isocitrate is the substrate for DPN-dependent isocitrate dehydrogenase (1).

Activating Species of ADP-In the above studies, evidence hasbeen presented that magnesium isocitrate is the substrate forthe enzyme. Since ADP is a relatively strong chelator of mag-nesium, it became of interest to know whether the free or thechelated forms of ADP are active modifiers.An experiment similar to that shown in Fig. 2 was designed inwhich free hilg2+ and magnesium ADP complex are eff ect ive lyremoved by adding free isocitrate in large excess. The eff ect ofincreasing concenkations of isocitrate at constant total mag-nesium (0.10 mM) was studied at several fixed levels of ADP,and plotted as 1 /v against reciprocals of isocitrate concentration(Fig. 4A). As free isocikate increases, the magnesium is con-verted to magnesium isocitrate and velocity increases. Forexample, without ADP, the concentration of magnesium iso-citrate rises from 0.087 to 0.096 mM as nn-isocitrate increasesfrom 12.5 to 50 mM. At 50 mM isocitrate with varying ADP,the concentration of free h/Ig2+ is reduced to between 3.4 and3.7 PM and only about 1% of total AD1 is present as MgADP-(2 to 8 PM). Extrapolation to the ordinate leads to di$erentintercept values, indicating that free ADl can activate isocitratedehydrogenase. If MgADP- were the active species all lineswould intersect at a common point on the ordinate. Further-more, as there is no tendency for the lines to curve upward asisocitrate concentration increases, free isocitrate does not actas an inhibitor with respect to magnesium isocitrate. If suchinhibition did occur it should be observed in this experiment(as well as in Fig. 2) since the ratio of free isocitrate to magnesiumisocitrate is in excess of 500.

This experiment also confirms the previous results which in-dicated that free magnesium is not required for the reaction,

The inhibition correction factors are the terms shown inbrackets in Equations 13 and 14.

yg,

y

,

jg

7/28/2019 plaut_1974

6/9

1853TABLE I

Constants from ncagnesium inhibition experimentsExamples of the conditions of incubation are given in Fig. 3. Calculations are described in the text and Equations 13 and 14.

ADP?nM000.3

6.657.27.2

K,,,ic f SE. K i SE. Vmx zt S.E. [Magnesium isocitrate-]range-

?nM he Aaro mine mu0.0210 zt 0.0091 0.0189 f 0.0087 0.0427 f 0.004 0.15~0.301.83 f 0.20b 0.239 f 0.041* 0.096 * 0.004* 0.25-1.0

rnM0.79 f 0.10 0.093 f 0.006 0.105 III 0.003 0.065-O. 39

-.- [Free Mgs+l range

?nM0.0250-0.150.026-4.0

0.04-l .5@Values expressed in terms of concentrations of magnesium DL-isocitrate. In experiments where the D isomer is used the value ofthe constant is one-fourth and one-half that reported here in the absence and presence of ADP, respectively . 4X = Sa.s as de-scribed by Equation 11.b Weighted mean from three experiments.

since extrapolation to t.hc ordinat,e gives t.he velocity at zero freemagnesium. These intercept values represent the initial veloc-ity at 0.1 mM magnesium isocitrate, zero fret magnesium, zeroMgADP-, and the free ADP3- concentrations shown in Fig. 4A.The lines in Fig. 44 show approximate linearity over the experi-mental range of magnesium isocitrate employed, since its con-centration changed only by small increments (from 0.067 to0.096 mM), however, over wider concentration ranges such plotswould not be linear. A secondary plot of the reciprocals of thedifferences of the velocities at the intercepts (l/v - vO) againstthe reciprocals of concentrat.ions of ADP yields a straight linewhich intersects the abscissa at -l/K,,, (Fig. 4B). The datawere fit ted to Equation 10 as described under Methods andan apparent K, of 0.63 mM has been calculated for free ADP.In another experiment, where the concentrations of ADP werelower (0.15 to 0.20 m M ) , the value of K, was 0.68 mM.The results above suggested that free ADP can activate theenzyme; however, they leave uncertain the act ivi ty of t,he mag-nesium ADP complex. The react,ion was studied in an experi-ment where magnesium m-isocitrate was maintained at thesame concentration as that at the intercepts of Fig. 4A (0.10mM), free Mg*+ was kept constant and total ADP was variedbetween 0 and 2 mM. When free Mgz+ was constant at 0.217or 0.434 mM the ratio of [free ADP] to [magnesium ADP] wasfixed at 2 or 1, respectively. Comparison wit,h K, values forfree ADP obtained under the experimental conditions of Fig. 4should give an indication of the act ivi ty of magnesium ADP.Thus, similar const.ants would indicate that magnesium ADP isinert; if K, of free ADP is lower or higher than that of Fig. 4magnesium ADP would be an activator or an inhibitor, respec-tively.The results o f such an experiment for total ADP arc repre-sented in Fig. 5A in the form of a rectangular hyperbola with apositive ordinate intercept. The experimental points were fittedto Equation 9 to give values of ~0, KN, and KD; the latter repre-sents the apparent K, for ADP. The values of apparent K, o ftotal ADP are 0.94 and 1.14 mM where rat ios of [free ADP] to[magnesium ADP] are 2 and 1, respect ively. However, theconstants of free ADl are essentially the same (0.63 rnhf versus0.57 mM) (Fig. 5B), and comparable to the value of K, (0.63mM) obtained under the experimental conditions shown in Fig. 4.These results suggest that free ADP is the active modifier andthat magnesium ADP is inert. The results f rom Figs. 4 and 5and of additional experiments are summarized in Table II I.Relationship between Substrate and ADP--In studies similarto those shown above, the values o f K, of free ADP were foundto vary considerably depending on the concentration of sub-

strat,e. Thus, at pH 7.2 the values of K,, of free ADP were 0.6and 0.2 mM with 0.10 and 0.2Gl mM magnesium nr,-isocitrate,respectively. It became of interest, therefore, to study theinterdependence of substrate and modifier. In these experi-ments free Xg2+ was held constant at relatively low and un-inhibit,ory levels (0.025 to 0.05 mnf) ; magnesium isocitrate andfree ADP were varied by appropriate changes in isocitrate andtotal ADP.

The results of experiments at pH values of 6.65, 7.2, 7.4, and7.75 are presented in Table II . The experiments at pH 7.2 and7.4 show the apparent K, of ADP decreasing with increasingconcentrations of magnesium isocitrate, and So.5 of magnesiumnL-isoeitrate declining wit.h increasing AD1 (Table II) . Thevalue of the Hill slope declines progressively from n = 1.9 int.he absence of ADP toward n = 1.0 as t.he concentration ofADP increases. The same trend in values of slopes is noted inexperiments at pH 7.75.

At pH 6.65 t.he slopes of the Hill plots for magnesium iso-citrate are near n = 2 and unaffected by ADP in the concentra-t.ion range 0 t.o 0.2 mM . This has been confirmed by the para-bolic shape of plots of l/v versus l/[magnesium isocitrate-] withthe fixed levels of ADP shown in Table II. It is possible, how-ever, that higher concentrations of ADP are necessary at pH6.65 in order to decrease the slope of Hill plots to 1.0. It maybe that there is no interdependence between substrate and ADP,as suggested by the lack o f response of the Hill slope (n), how-ever, in the absence of data on inhibition by Mg2f in the pres-ence of ADP, and in view of the fact that, Mg2+ appears to in-hibit more strongly at low pII values, no firm conclusions canbe reached. 13~ the same criteria, it is diff icul t to determinewhet.her or not the kinetic constar1t.s of ADP and substrate arecompletely independent at pH 6.65 (Ta.ble II),

The effect of magnesium isocitrate concentrat,ion on ,!!& ofADP has also been examined in Hill plots. Slopes (m) withvalues near unity have been obtained at. pH 6.65, 7.2, 7.4, and7.75 (Table II). The values o f m are not affected by changesin magnesium isocitrate concentration.Eject of pa-Increasing pH between pH 7.2 and 7.75 leadsto relatively small increases of K, for magnesium isocitrateand ADP; inhibition by free Mg2+ becomes less severe and V,,is either the same or shows a small increase (Tables I and II).The constants at pH 6.65 are markedly different from thosebetween pH 7.2 and 7.75, showing in the absence of ADP muchlower So.5 for magnesium isoci t,rate, lower Vmsx, and much moresevere inhibition by free Mgz+.

The inhibition by Mg2+ as a function of pH is of special in-terest in view o f the proposal of Cohen and Colman that in-

byguest,

onJuly22,2012

www.jbc.org

Downloadedfrom

7/28/2019 plaut_1974

7/9

1854TABLE II

hflect of varying ADP on So.5 of magnesium isocitrate and ofchanging magnesium isocilrate complex concentrations on K,,,of ADPAt pH 7.2, 7.4, and 7.75 the incubation mixtures contained0.167 M sodium Hepes buffe r, free Mgz+ was held constant at 0.05M, and the concentrations of magnesium isocitrate were main-

tained at the levels indicated by appropriate changes in DL-isocitrate. At pH 6.65 the incubation mixtures contained 0.167M Pipes buffer and 0.025 M free Mg*+. Concentrations of ADPand values of K,,, of ADP are reported as total ADP. Concentra-tions o f magnesium isocitrate (MI) and values of So.5 are on thebasis of nn-isocitrate. Values o f K,,, of ADP are calculated fromthe computer fi ts of the data to Equation 9 as described in thetext. The slopes (m) were determined in Hill plots. Values ofSO.5 of magnesium nn-isocitrate and of slopes (n) were calculatedfrom Hill plots at pH 7.2 and 7.4. At pH 6.65 an average value ofV msx of 0.0314A Aa40 min-1 was used for calculations of the Hillplots; i t was obtained from plots of l/v versus l/[magnesium iso-citrate-je at various fixed levels of ADP (range of V,,, ,, was from0.0337 to 0.0305 A As& min-I). Incubation conditions and thecalculation of concentrations of components of the reaction mix-tures from dissociation constants of magnesium isocitrate and ofmagnesium ADP were as described under Methods.

110

100 ADP,mM0

soI

v

x

607.E 70E

g 607- /0

30

20 At--vime0.D2 0.04VDL-ISOCITRATE (TOTAL),mM-FIG. 4. Response of velocity at constant total magnesium andvarying isocitrate at several fixed levels of ADP (pH 7.2). Theconcentration of total magnesium was 0.10 mM, total nn-isocitratewas varied from 12.5 to 50 mM, and ADP concentrations were 0,0.20, 0.25, 0.33, 0.50, and 1.00 mM. A, double reciprocal plot ofvelocity versus total nn-isocitrate. B, secondary plot of A inwhich the differences between the velocities at the intercept in theabsence (V,) and presence o f ADP (V) are plotted as the recipro-cals l/(V - V,) versus l/ADP. The intercept at the base-line isequivalent to -1/K, of free ADP. The lines to which the experi-mental points were fitted were constructed by the computer pro-

gram for a rectangular hyperbola (19) and Equation 7.

-7pH 6.69 pH 1.2 pH 1.4 PI3 1.15

WI

n&M0.100.150.200.300.400.500.600.750.801.50[ADPI7md

00.050.100.150.200.300.501.00[ADPI

WZM0

cd

A%P7m.u

0.160.230.21

nzicope1.01.01.2

n

A% mslope1.01.01.21.01.2

A%e&M

i%P

0.270.220.160.13

0.440.340.290.220.16

mSlC+C

1.10.91.01.01.1

0.260.220.49

?nslope

1.11.11.0

0.6 Ml )t 0.6 Ml n So.s MI nnzM slope ?nM SlO$%

0.41 1.9 0.54 2.00.30 1.7 0.46 1.90.25 1.6 0.38 1.7

mld0.956

0.19 1.6 0.29 1.60.16 1.5 0.23 1.50.12 1.4 0.17 1.40.09 1.3 0.10 1.30.63c0.77c0.87~

&PC2b

1C1C10

--

--s--

----

(--

hibition of pig heart isocitrate dehydrogenase occurs at lowervalues of pH due to an increase in the concentration of themanganous complex of dibasic isocitrate (9). If this were thecase for the magnesium isocitrate complex in the present studies,the apparent inhibition constant for magnesium would be ex-pected to decrease approximately 4-fold as the pH decreasesfrom 7.2 to 6.65, In fac t, the magnesium inhibition constantdecreases by a factor of 85 over this pH range (Table I) whichmakes the simple explanation of magnesium H-isocitrate forma-tion and inhibition unlikely. At present the reason for such alarge change in the inhibition constant is unknown, but it couldbe due to a pH-dependent conformational transition in the pro-tein over such a pH range.

DISCUSSIONSteady state kinetic studies on the role of magnesium in the

mechanism of bovine heart DPN-linked isocit.rate dehydrogenaseare consistent either with magnesium isocitrate being the actualsubstrate or with a rapid equilibrium-ordered mechanism withfree magnesium adding before the addition of free isocitrate.Since mechanisms of the latter type are very uncommon formetal-requiring enzymes, the former mechanism was assumedfor the purpose of discussing the results and postulating a re-action sequence for the substrate and modifier of bovine heartisocitrate dehydrogenase. The proposed substrate function ofmagnesium isocitrate is in accord with the earlier observationthat a combination of divalent metal ion activator and n-iso-

0.1 Mlmd(

0.180.120.110.100.10

iope2.02.02.02.01.9

1

V rnP.xAApo v&r-

0.1150.0849

V,,X V,.XAAsro mitt-l AAaro min-

0.0817 0.1150.0887 0.13-0.2~

AAsao min-0.0305

).044-0.051

(LResults uncorrected for inhibition by 0.025 mM Mg2+. Thevelocity correction factors for magnesium inhibition at 0.1, 0.2,and 0.3 mM of magnesium isocitrate- are 1.20, 1.37, 1.76, respec-tively.* From plot l/v versus l/[magnesium isocitrate-12.0 From plot l/v versus l/[magnesium isocitrate-1. At 0.3, 0.6,and 1.0 mM ADP the values of V,,, were 0.15, 0.18, and 0.2, re-spectively.d Calculated from Equation 9 except where indicated.

yg,

y

,

jg

7/28/2019 plaut_1974

8/9

1855, I

V,=O.O69 d I

l- - ~/FREE AknM--I K-*1.14, , 111 . 11.0 1.5 2.6T O T A L A O P , m M

FIG. 5. Activi ty as a function of ADP at fixed ratios of freeADP to magnesium ADP at pH 7.2. Magnesium nn-isocitratewas constant at 0.10 mM. Total ADP was varied and the ratiosof [free ADP] to [magnesium ADP] were fixed at 1 or 2 by main-taining free Mg*+ at 0.434 mM (Curve 1) and 0.217 mM (Curve B),respectively. A, velocity as a function of total ADP. The ex-perimental points were fit ted b y a computer program (19) toEquation 9. B, double reciprocal plot of (V - VO) versus freeADP, where V, is the velocity in the absence of ADP. Thepoints were fitted by the method of least squares. The interceptat the base-line is --1/K,,,. At 0.217 and 0.434 mM Mg2+ the in-hibition factors? are about 1.1 and 1.3, respectively. The inhibi-tion of velocity is reflected in the differences in calculated I,, ,for Curves 1 and 2 shown in Panel A and the differing slopes andordinate intercepts (with practically identical base-line inter-cepts) of the lines in Panel B.

TABLE IIIComparison of activation by total ADP and free ADP

Magnesium nn-isocitrate (0.10 mru) was constant at pH 7.2 inall experiments. Free Mg2+ and ADP varied as indicated. Thedata were obtained in experiments as described in Figs. 4 and 5.

Y-Free Free ADP Total ADPMP to MgADP- range

~___?nM ?nM

0 0.2-1.00.1 47034 0.3-1.50.217 2 0.2-1.00.434 1 0.4-2.0 -

Km of total ADP K,,, of Free ADP

mu f S.E.0.63 f 0.060.84 f 0.070.94 f 0.071.14 f 0.06

- -

vim f S.E.0.63 f 0.060.68 f 0.050.63 f 0.050.57 f 0.03

citrate is required for protection against inhibition by low con-centrations of sulfhydryl group-binding agent.s (2, 13) whileeither the metal ions or isocitrate alone were inef fect ive. Itwas reported previously that manganese is an inhibitor of theenzyme (8, 9), but such an ef fec t was not observed with mag-nesium (8). The present studies are consistent with the ideathat inhibition by free Mg2+ can occur; that it is competitivewith magnesium isocitrate-, and that the inhibition becomesmore severe with decreasing pH (Fig. 3 and Table I). In con-trast to the requirement for the combination of magnesium andisocitrate as substrate, only the free form of the modifier isactive (Figs. 4 and 5; Table III) .Previous kinetic investigations of mammalian DPN-speci ficisocitrate dehydrogenase were handicapped by lack of informa-tion on the nature of the ionic species which interact with theenzyme, and interpretation of kinetic results became somewhatambiguous. Thus, the ef fec t of total isocitrate concentrationon velocity with constant relat ively high divalent metal ion

concentration was examined in most earlier studies o f the ki-netics o f this enzyme (2, 4, 6, 8). The present results suggestthat such determinations of apparent kinetic constants, depend-ing substantially on lower concentrations of isocit.rat.e, tend tobe imprecise because of substantial apparent inhibition by freeMg2+ at the relatively high [Mg*+] to [magnesium isocitrate-]ratios. Since reported values of Hill coefficients are particularlyvariable, experiments (not shown here) have been done at pH7.2, 7.4, and 7.6 in which total magnesium was fixed at 6.6 mMand total nn-isocitrate was varied from 0.4 to 40 mM. Hillplots o f the results as a function of magnesium isocitrate con-centrations gave Hill coeff icients (n) ranging from 1.9 to 2.0;the values in the absence of ADP are thus in agreement withthose report.ed in Table II. However, plots of the same experi-mental data as a function of total isocitrate yielded Hill co-efficients considerably lower than n = 2 (1.3 t.o 1.4). Thesedifferences in part reflect the fac t that with constant total mag-nesium (6.6 mM) and changing isocitrate (total or free) the ratioof [isocitrate]/[ma.gnesium isocitrate] is not constant (Equations2 and 3) but rises with increasing isocitrate concentration.

The elucidation of active substrate and modifier forms in t.hepresent study eliminates a number of mechanisms which hadpreviously been considered. Among t.hese are separate com-binations of free Jlg2+ and free isocitrate at the substrate site,except for the rapid equilibrium ordered possibi1 it.y previouslydiscussed, and t,he combination of the Mg;ADP- complex at themodifier site.

Although the experiments presented in this paper were notdesigned for quantitative analysis of the reaction mechanism,the results do permit the formulation of a reaction mechanismwhich shows quali tative agreement with the experimental resul ts.Such a mechanism is presented in Scheme I, where E representsenzyme saturated with DPNf.

MI\ \ MI,E- E*MI - Eu (MI),w ------A

AD P \I1\r

PRODUCTS

The over-all initial velocity equation for Schemewritten in general fo rm as:V(IMZP + bL4DPINfII)

I can be

= [MZ]2 + c[MZ] + d[ADPJ + e]ADPI]MZl + j (15)where b, c, . . f represent kinetic constants. In the absence ofthe modifier, ADP3-, only the upper reaction sequence of SchemeI would operate, and Equation 15 would reduce to:

V[MZ]2v = Mifz12 + c[M Zl + j (16)

This ordered combination of 2 substrate molecules is consistentwith the results o f initial velocity studies, which gave good fit s(19) to Equation 16,* i.e. parabolic plots of l/v against l/mag-nesium isocitrate- and Hill plots for magnesium isocitrate- withslopes near 2 in the absence of ADP. These kinetic featuresmay be most simply explained by the existence of two inter-dependent sites for substrate combination, both of which mustbe occupied before products can be released. It is thought that

* T. Aogaichi, V. L. Sehramm, and G. W. E. Plaut (1972) un-published observations.

yg,

y

,

jg

7/28/2019 plaut_1974

9/9

1856initial binding of substrate induces a change in the enzyme whichtransforms the remaining site (or sites) into a cataly ticallyactive conformation. In the absence of substrate these sitesare presumed to be identical since previous observations indicatethat the enzyme is composed of eight identical subunits (15).Since steady state kinetic data cannot be used to determine thestoichiometry of substrate binding, additional experiments willbe required to determine the actual number of each type o fbinding site on the enzyme. In the absence of such quantita-tive data no meaningful consideration can be given to binding atmultiple identical sites. Such combinations would not changethe general form o f Equation 16, but would have the ef fec t ofchanging the kinetic constants by a statistical factor for multipleinteractions.In contrast to the interdependence of enzyme-substrate inter-actions in the absence of ADI-, kinetic studies in the presenceof relatively high concentrations of the modifier are consistentwith a single (or multiple independent) combination of substratewith enzyme since the data give good fits to the equation de-scribing a single such combination (Equation 7). Such resultsare consistent with the pattern predicted from Equation 15,which reduces to the same form as Equation 7 as ADP concen-tration becomes very high. This interaction is shown in SchemeI, and leads to the formation of the ADP .enzyme .magnesiumisocitrate- complex. Saturation of the modifier site (or sites)causes the enzyme to exhibit independence of substrate sites,with an apparent loss of the requirements for the catalytical lyinactive site (or sites) to be filled before catalysis can occur.This phenomenon could be explained by ADP causing such anappreciable increase in aff ini ty between substrate and the non-catalytic site (or sites) that saturation occurs at all experimentalconcentrations of magnesium isocitrate. Alternatively, in thepresence of ADP, binding of substrate at the noncatalytic sitemay no longer be required for catalysis. The latter postulateconstitutes the simplest mechanism and has been illustrated inthe sequence of reactions from E to the formation of ADI.E.magnesium isocitrate-. A third possibility is that ADP causesboth sets of sites to become catalytically active, equivalent, andindependent. The latter seems less likely t,han the previouspostulates since the formation of additional cat.a lytic si tes mightbe expected to lead to increased activity. However, the resultsindicate that Fmnlr is independent of modifier concentration(Tables I and II).Indications for binding of free ADPa+ to the free enzyme havebeen obtained in gel filtration esperiments (3) and by the dem-onstration that the modifier causes protein concentration-dependent dimerization of the enzyme (15) in the absence ofadded divalent metal ions and substrates. The kinetic datapresented here are consistent with these earlier resul ts, and alsoshow that modifier combines in an independent manner at itsrespective site (or sites), as the experimental data fit well toEquat.ion 17, which describes a single combination of ADP. Itwill be noted that Equation 17 has t.he same general form asEquation 9. r I + e[ADPl 121 V

1 Cd + eM4ZIL4I~Pl+ lMZIZ f c[MZl + f (17)Furthermore, Equation 17 predicts that the apparent K, valuesfor ADP- will be dependent up011 the concentration of substratepresent, as

K, for ADP = [MZI~ + c[MIl + fd + elMI

This prediction is confirmed by the data (Table II).Although Scheme I is capable of explaining the kinetic prop-erties of DPN-linked isocitrate dehydrogenase from heart, itshould be emphasized that this is the simplest mechanism con-sistent with the data and in no way precludes more complexreaction pathways which may well become apparent with furtherkinetic and thermodynamic studies. Indeed, isocitrate dehy-drogenases from other sources appear to exhibit more complexmechanisms than that proposed here (11, 20, 21), althoughfailure to identify the active substrate form could have addedan additional degree of complexity to the kinetic results.

The ef fec ts of a number of factors which may regulate mito-chondrial isocitrate oxidation at the stage of DPN-linked iso-citrate dehydrogenase have been discussed previously (for reviewsee Ref. 13). The present results emphasize the need to con-sider the level of intramitochondrial divalent metal ion (or ions)(e.g. Ref . 22) on the flux of metabolites at this step, since freeMg*f or M& inhibits the reaction and the rate of reaction isdependent on the concentrations of the active species of sub-strate (e.g. magnesium isocitrate) and modifier (free ADP3-).In addition, fluctuations in intramitochondrial pH, which mayoccur in various metabolic states (23, 24), could affect the rateof catalysis since some of the kinetic constants are particularlysensitive to changes in pH (Tables I and II).

1.2.3.4.5.

6.7.8.9.

10.11.12.13.14.15.16.17.18.19.20.21.22.23.

24.

REFERENCESPLAUT, G. W. E., AND SUNG, S-C. (1954) J. Biol. Chem. 207,305-314CHICN, R. F., AND PLAUT , G. W. E. (1963) Biochem istry 2,

1023-1032CHEN, R. F., BROWN, D. M., AND PLAUT, G. W. E. (1964)

Biochemistry 3, 552-559GOEUILL, II., AND KLINGENBERG, M. (1964) Biochem. 2. 340,

441-464KLINQENBERG, M., GOXBELL, H., AND WENSK E, G. (1965)

Biochem. 2. 341, 199-223STUN, A. M., KIRKMAN, S. K., AND STEIN, J. H. (1967) Bio-chemistry 6, 3197-3203STEXN, A. M., STEIN, J. H., AND KIRICMAN, S. K. (1967) Bio-

chemistry 6, 1370-1379PLAUT, G. W. E., AND AOGNCHI, T. (1968) J. Biol. Chem. 243,

5572-5583COHEN, P. F., AND COLMAN, R. F. (1972) Bioche mistry 11,

1501-1508HARVEY, R. A., HERON, J. I., AND PLAUT, G. W. E. (1972) J .

Biol. Chem. 247, 1801-1808ATKINSON, D. E., H.~THAW.LY, J. A., AND SMITH, E. C. (1965)

J. Biol. Chem. 240, 2682-2690SANWALL, B. D., AND COOK, II. (1966) Biochemistry 6, 886-894PLAUT, G. W. E. (1970) C urr. Top . C ell. Regul. 2, l-25PLAUT, G. W. E. (1969) Methods Enzymol. 13, 34-42GIORGIO, N. A., JR., YIP, A. T., FLF:MING, J., AND PLAUT,

G. W. E. (1970) J. Biol. Chem. 246, 5469-5477SIEGEL, J. M., MONTGOMERY, G. A., AND Bocq R. M. (1959)Arch. Biochem. Biophys. 82, 288-299

GHZYUOWSKI, A. K., TATI,: , S. S., AND DATTA, S. P. (1970) J.Chem. Sot., Sect. A, 241-245

OSULLIVAN, W. S., AND PNUUN, D. L). (1964) Biochemistry 3,18-26

CLELAND, W. W. (1963) Nature 198,463-465ATKINSON, 11. E. (19G6) Annu. Rev. Biochem. 36, 85-124S.~N~.UL, B. D., STXHOW, C. S., AND COOK, It. A. (1965)

Biochemistry 4, 410-421BOGUCKA, K., AND WOJTCZX