THE JOURNAL OF BIOLOGICAL CHEJ~ISTRY Vol. 240, No. 3, March 1965

Printed in U.S.A.

Crystallization and Kinetic Properties of Glutamate

Dehydrogenase f o r m Frog Liver*

LEONARD A. FAnIEN,t BARBARA 0. WIGGERT, AND PHILIP P. COHEN

From the Department of Physiological Chemistry, University of Wisconsin, Madison 6, Wisconsin

(Received for publication, July 13, 1964)

In the course of preparing carbamyl phosphate synthetase from frog liver, a crystalline protein fraction was obtained which proved to be glutamate dehydrogenase. Because of the con- siderable interest in this enzyme and the striking differences in certain properties of this enzyme from different animal sources (l-6), it seemed of importance to study the kinetic properties of crystalline frog liver glutamate dehydrogenase.

The kinetic properties of crystalline frog liver glutamate de- hydrogenase are reported in this paper and are compared with those of crystalline beef liver glutamate dehydrogenase.

EXPERIMENTAL PROCEDURE

Enzymes and Reagents-Pyruvic acid and cetyltrimethylam- monium bromide were obtained from Eastman Organic Chemi- cals. The pyruvic acid was triply distilled before use. Satu- rated ammonium sulfate was prepared by adding 770 g of ammonium sulfate to 1 liter of water and heating the mixture to the boiling point. It was then cooled to 25” and ammonium hydroxide was added until the pH of a 1: 10 dilution in water would be 7.0. Glutamic acid-HCI was obtained from the Nu- tritional Biochemical Company. ,411 other enzymes and sub- strates were obtained from Sigma Chemical Company. The oc-ketoglutaric acid used was crystallized from acetone-benzene mixtures. Stock solutions of reagents were adjusted to the pH of the assay system and prepared as their potassium salts.

Initial VeZocity MeasurementsAll enzyme assays were carried out in 0.01 M Tris-acetate’ with 1 X IOV M EDTA, at pH 8.0 and 23”. The reactions were followed spectrophotometrically at 340 or 360 rnp by means of a Brown recorder with an expanded scale (7). For experiments which required low ammonium ion concentrations, the enzyme was dialyzed overnight against 0.1 M Tris-acetate buffer, pH 8.0, with 1 x 1O-4 M EDTA at 4”. Ammonium ion concentration could be lowered to insignificant amounts in most experiments by washing the centrifuged crystals with this buffer and diluting the enzyme. The standard assay solution used to follow the purification of the frog liver enzyme consisted of 2 X 10e2 M cy-ketoglutarate; 5 X lo-* M NH&l; and 5 x low5 M DPNH.

Protein Determination-a4t early stages in the preparation, protein concentration was determined by the method of War-

* This study was supported in part by Grant C-3571 from the National Cancer Institute, National Institutes of Health, United States Public Health Service, and by Grant G-21237 from the National Science Foundation.

t Postdoctoral Fellow, United States Public Health Service. 1 The concentration of Tris buffer refers to the concentration

with respect to Tris.

burg and Christian (8). The protein concentration of the crys- talline enzyme was determined by measuring the absorbance at 280 mp. It was found that the absorbance of a solution con- taining 1 mg per ml of the crystalline enzyme, as determined by the method of Lowry, Rosebrough, Farr, and Randall (9), was 0.95 in 0.1 M Tris-acetate with 1 X 10m4 M EDTA, pH 8.0, at 23”.

Light Scattering-Light scattering measurements were carried out in the laboratory of Dr. Carl Frieden, Department of Bio- logical Chemistry, Washington University School of Medicine, and with a similar Brice-Phoenix instrument in this laboratory. The methods used were the same as those previously described (10). All light scattering measurements were made at 90” and at 436 mM. The values shown for weight average molecular weights are in terms of apparent weight average molecular weights.

UZtracentrz’,fuation-The sedimentation coefficient of frog liver glutamate dehydrogenase was determined in a single sector cell in a Spinco model E analytical ultracentrifuge at 23”.

Gel Electrophoresis-Electrophoresis on polyacrylamide gel was performed according to the method of Ornstein and Davis (II). In the case of the experiments done with urea, only the main running gel contained 8 M recrystallized urea. The spacer and sample gel were as described by Ornstein and Davis (11).

Fluorescence StudiesSpectrophotometric studies were per- formed in the laboratory of Dr. Carl Frieden with the use of a modified Farrand spectrofluorometer previously described by Frieden (10).

Preparation of Crystalline Frog Liver Glutamate Dehydrogenase- The livers from 24 bullfrogs2 (Rana catesbeiana) were freed of the gall bladders and immediately placed in cold, 0.15 M KC1 solution. All operations were then performed at 3” in the cold room. The livers (196 g) were washed with additional cold 0.15 M KC1 and then blotted with Whatman No. 2 filter paper. The livers were homogenized in a Waring Blendor at top speed for 15 seconds in a volume of cold 0.15 M KC1 equal to 3 times the weight of the tissue. The homogenates were then pooled and centrifuged at 7970 x g in the Servall centrifuge at 0” for 15 minutes. The supernatant solution was discarded and the residue resuspended by stirring in an additional volume of cold, 0.15 M KCl. This procedure was repeated twice. The washed residue was then suspended in a volume of 0.1% cetyltrimethyl- ammonium bromide3 equal to twice the original weight of the

2 Frogs were purchased from the Lemberger Company, Oshkosh, Wisconsin.

3 The stock solutions of cetyltrimethylammonium bromide and saturated ammonium sulfate were maintained at room tempera- ture.

1083

by guest on March 24, 2018

http://ww

w.jbc.org/

Dow

nloaded from

1084 Crystalline Frog Liver Glutamate Dehydrogenase Vol. 240, No. 3

TABLE I Purification of frog liver glutamate dehydrogenase

Fraction Volume

ml

Cetyltrimethylammonium bromide extract,

BOY, ammonium sulfate pre- cipitate in Tris buffer.

407, ammonium sulfate pre- cipitate

Supcrnatant fraction from 247, ammonium sulfate fraction.

First crystals.. Second crystals.. Third crystals.. Fourth crystals.

740

215 2.7

62

71 26 14 10 (i.!

1

5

- Total xotein

- g

3.8

1.9

1.2 0.65 0.18 0.10 0.08

Specific activity*

0.7 x 10-s

1 X 10-a

1.4 X 10-a

0 3 x 10-a

11 X 10-a 20 x 10-a 25 X 1O-3

Total unitst

2.7

2.7

2.7

0 2.0 2.0 2.0 2.0

* Specific activity is defined as the millimoles of DPN produced per minute per mg of protein.

i Total units represent the product of the specific activity times the total amount of protein in milligrams.

[GLUTAMATE DEHYDROGENASEI hg per ml 1

FIG. 1. Plot of weight average molecular weight, determined by light scattering, as a function of enzyme concentration. The solid line was calculated by assuming that the enzyme dissociates from a dimer with a molecular weight of 500,000 to monomers with molecular weights of 250,000. The experiments were per- formed in 0.1 M Tris-acetate containing I X lo-’ M EDTA and 1 X 1OW M phosphate, pH 8.0, at 23”.

liver tissue, and homogenized in a Waring Blendor for 1 minute at full speed. The homogenate was then centrifuged at 7970 x g at 0” for 20 minutes. The resulting supernatant fraction was filtered through a plug of glass wool in a funnel and the filtrate collected. The residue was re-extracted with 0.1% cetyltri- methylammonium bromide in a similar manner and the two extracts were combined.

The extract was brought to 60% saturation with solid am- monium sulfate and centrifuged at 7970 X g for 20 minutes at 0”. The resulting precipitate was dissolved in a volume of 0.02 M Tris-chloride buffer, pH 7.5, containing 5 X 1O-3 M MgC12, equal to the original weight of the liver tissue. This solution was t,hen brought to 60% saturation with ammonium sulfate and the preceding step repeated. This precipitate was again dissolved in the Tris buffer as before and the solution was brought up to 10% saturation with solid ammonium sulfate and was centrifuged at 25,500 x g for 20 minutes at 0”. The precipitate was discarded and the supernatant fraction was brought up to 40y0 saturation with solid ammonium sulfate and was centri- fuged as before. The precipitate was then dissolved in 100 ml

of 0.02 M Tris-chloride buffer, pH 7.5, containing 5 X low3 M

MgC12. Saturated ammonium sulfate at pH 7.0 was then added until a faint turbidity was observed (at 23 to 24% saturation). Crystals of glutamate dehydrogenase usually formed instan- taneously but occasionally crystallization began in about I hour.

Carbamyl phosphate synthetase was found to be present in the supernatant fraction of the first preparation of crystals. The specific activity of carbamyl phosphate synthetase in the super- natant fraction was 1 X lop3 mmole of ADP produced per minute per mg of protein as assayed spectrophotometrically (7). The carbamyl phosphate synthetase can be further purified by col- umn chromatography (12).

The glutamate dehydrogenase crystals have been kept for 1 to 3 days in the cold before recrystallization, which involved separation of the crystals by centrifugation followed by solution in a small amount of 0.02 M Tris-chloride buffer, pH 7.5, con- taining 1 X IO-* M EDTA. The solution was then centrifuged at 3000 x g at 0” for 5 minut.es to remove any insoluble material. The supernatant solution was then brought up to 24% satura- tion with saturated ammonium sulfate and placed in the cold. Recrystallization was repeated four times, at which time con- stant specific activity was obtained.

RESULTS

Preparation-Four times recrystallized glutamate dehy- drogenase was obtained in a yield of 81 mg of protein with a specific activity of 2.5 X 10F2 mole of DPNH oxidized per min- ute per mg of protein under the conditions of the standard assay. Table I gives the specific activity of frog liver glutamate dehy- drogenase at various steps of purification.

Greater yields of glutamate dehydrogenase can be obtained if the first crystals are harvested 2 to 3 days after crystallization is started. However under these conditions there is a loss in carbamyl phosphate synthetase activity.

While a schlieren pattern is apparent in the ammonium sulfate suspension, and specific activity increases with recrystallization, the crystals are difficult to see under t.he microscope because of their small size. However at a magnification of 250 X, and with suitable contrast, the crystals appear as stubby, somewhat ir- regular needles.

Light Scattering-The results of light scattering experiments are shown in Fig. 1. From the results it is clear that the en- zyme tends to dissociate with dilution. While simple extrap- olation gives a value for the weight average molecular weight at zero protein concentration of 3 x 105, a lower value is ob- tained if all of the experiment,al points are considered. If it is assumed that the frog enzyme consists of a dimer, which disso- ciates upon dilution into two monomers with equal molecular weight, then the weight, average molecular weight M observed is equal to

M = olMB + (I - cx)MA

where a! is the weight fraction of monomer, MB is the molecular weight of the monomer, and Jla is the molecular weight of the dimer. The equilibrium constant K for this reaction is equal to

K = [Bl2 = ~~=CIMB [Al ___ (1 - 4

where C refers to the concentration of the enzyme in mg per ml, [B] refers to the concentration of the monomer, and [A] refers to the concentration of the dimer. The solid line shown in Fig. 1

by guest on March 24, 2018

http://ww

w.jbc.org/

Dow

nloaded from

March 1965 L. A. Fahien, B. 0. Wigged, and P. P. Cohen 1085

FIG. 2. Ultracentrifugal patterns of a solution of 4.7 mg per ml of crystalline frog liver glutamate dehydrogenase. Sedimentation is from Zeft to right; photographs were taken at 16-minute intervals. The first photograph was taken at a bar angle of 45”, the second at 35’, the third at 30”, the fourth at 25”, and the fifth at 20”. The experiment was performed in 0.1 M Tris-acetate with 1 X 10v4 M EDTA, pH 8.0, at 23”.

[GLUTAMATE DEHYDROGENASEI

(r-4 per ml)

FOG. 3. Plot of reciprocal of ~20.~ of frog liver glutamate de- hydrogenase as a function of enzyme concentration. Experimen- tal conditions are as described in legend to Fig. 2, except that values shown for glutamate dehydrogenase concentrations of 2.08, 5.20, 8.32, and 10.40 mg per ml were obtained at 19.6’; the other values were obtained at 23”.

has been calculated by letting K = 4 X 10B6 M, Ma = 5 x 105, and MB = 2.5 X 105. It can be seen that the calculated line agrees almost perfectly with all of the data over a IO-fold range of protein concentration. On this basis weight average molecular weights of 2.5 x lo5 for the monomer and of 5 X lo5 for the dimer, respectively, are proposed.

Ultracentrifugation-Frog liver glutamate dehydrogenase sedi- ments with a single boundary in the ultracentrifuge (Fig. 2). Multiple peaks were never observed. The values of szo,W as a function of protein concentration are given in Fig. 3. Again it is clear that the enzyme tends to dissociate upon dilution. Ex- trapolation gives a value for the s 20,~ at zero protein concentra- tion of 13.1 X 10-r3. This extrapolation, however, ignores light scattering data which show that dissociation is most marked when the enzyme concentration is less than 1 mg per ml. There- fore the true value of ~20,~ at zero protein concentration is in all probability lower than this value.

ElectrophoresisThe electrophoretic mobilities on polyacryl- amide gel of frog and beef liver glutamate dehydrogenase are shown in Fig. 4. The two enzymes behave very differently in that the frog enzyme migrates as one band whereas the’ beef enzyme remains stationary at the origin. In the presence of 8 M urea the two enzymes appear to behave in a similar way in that both consist of four bands which migrate at the same rates.

FIG. 4. Polyacrylamide gel electrophoresis of frog liver and beef liver glutamate dehydrogenases; 0.04 mg of protein was used in each gel. 1, beef enzyme; II, frog enzyme; 111, beef enzymein8Murea;~V,frogenzymein8Murea.

Kinetic Studies-Data obtained from kinetic studies of the beef liver enzyme with the use of TPN and TPNH as coenzymes can be represented by Equation 1 for the forward reaction (con- version of TPNH to TPN) and by Equation 2 for the reverse reaction (13).

Vf v= KA KB Kc KAKB - - -

l+ rA1+ iFI + i + [AI[Bl+

KBKC KAKBKC (1)

BIICI + LJl[BIK’l

Vf 9, E

(2) KD KF

l+[Dl+fi+

KDF

Dl IF1

where v is the initial velocity; Vf is the maximal velocity of the forward reaction; KA and A refer to the Michaelis constant and

by guest on March 24, 2018

http://ww

w.jbc.org/

Dow

nloaded from

1086 Crystalline Frog Liver Glutamate Dehydrogenase Vol. 240, No. 3

I I I I

0 IO 20 30 40

I / C TPNHl(mM)-’

FIG. 5. Double reciprocal plots of initial velocity versus TPNH concentration at various constant levels of cr-ketoglutarate in t.he presence of a high and constant amount of ammonium chloride (5 X 10V2 M). The concentrations of oc-ketoglutarate used were: A, 7 X 11Y4 M; B, 1.4 X lo-3 M; C, 2.8 X 1W’ M; D, 5.6 X lW3 M;

E, 1.4 X 10e2 M; and F, the extrapolated line for an infinite amount of a-ketoglutarate. Experiments were performed at 23” in 0.01 M Tris-acetate, pH 8.0, with 1 X 10d6 M EDTA.

-1.2 -1.0 -0.0 -0.6 -0.4 -0.2 0 0.2 0.4 0.6

I/[NH$(mM)-’

FIG. 6. Double reciprocal plots of initial velocity as a function of ammonium ion concentration at various constant levels of TPNH and in the presence of a constant a-ketoglutarate concen- tration of 7 X 10-a M. The concentrations of TPNH used were: A, 2 X 10m5 M; B, 2.5 X 10m5 M; c, 4 X 10M6 M; D, 6 X 10m6~; and E, the extrapolated line for an infinite concentration of TPNH. Other experimental conditions were identical with those described in the legend of Fig. 5.

concentration of TPNH, respectively; Ke and B refer to the Michaelis constant and concentration of ammonium ions, re- spectively; Kc and C refer to the Michaelis constant and con- centration of a-ketoglutarate, respectively; VR is the maximal velocity of the reverse reaction; KD and D refer to the Michaelis constant and concentration of TPN, respectively; KF and F refer to the Michaelis constant and concentration of glutamate, respectively; and KDF is a complex constant to be described later.

These rate equations are steady state equations derived for Mechanism I.

h EA + B \ EAB

h (Ib)

k5 e EABC

k7 EAB+C

h e EDF

h

ks \ ED + F (Ic)

k 10

k ED &E+D

he iId)

MECHANISM I

From Equation 1 it can be seen that when the concentrations of Lu-ketoglutarate and TPNH are varied in the presence of a very high and constant amount of ammonium ions, and double reciprocal plots are made of the velocity versus the concentra- tion of TPNH, the plots should be linear and parallel.

When the concentration of a-ketoglutarate is maintained constant and the concentrations of TPNH and ammonium ions are varied, and double reciprocal plots are made of velocity versus ammonium ion concentration, a series of lines should be obtained which intersect when the negative concentration of ammonium ions is equal to

-LB1 = KBO + KcIK’l) (la)

Each line should intersect the abscissa when the negative con- centration of ammonium ion is

KBO + KdMl) --[B1 = (1 + Kc/K1 + KdM)

(1 + KcI[CI) (lb)

Each line should intersect the ordinate when the reciprocal of the apparent maximal velocity is

1 1 Ka Kc -=_ V &PP vJ l+[Al+ICl >

ilc)

Fig. 5 shows the results obtained with the frog liver enzyme when the concentrations of oc-ketoglutarate and TPNH are varied in the presence of a high and constant concentration of ammonium chloride (5 x lo-* LV). The double reciprocal plots are parallel. By making secondary plots of the data presented in Fig. 5 one can obtain the values of KA, Kc, and V,.

Fig. 6 shows the results obtained when the concentrations of TPNH and ammonium ions are varied in the presence of con- stant amounts of cu-ketoglutarate (7 X 10e3 M).

From secondary plots of the type of data shown in Figs. 5 and 6, and determination of the intersection points, one can evaluate all of the constants of Equation 1 and demonstrate that KAB is KAKB, Ksc is KBKC, and KABC is KAKBK~.4 These values are shown in Table II and are compared with results ob- tained from studies of the beef enzyme.

Fig, 7 shows the results obtained when the concentrations of TPN and glutamate are varied. The results can be expressed by Equation 2. From the data presented in Fig. 7 one can ob- tain numerical values of the constants of Equation 2. These values are shown in Table II.

Thus it appears that the kinetic behavior of beef liver and frog liver glutamate dehydrogenases is identical.

Fluorometric Studies-According to kinetic theory, the ratio KaK,/Ks should equal the ratio i&%1, or the dissociation con- stant of TPNH (14). Experimentally this value is the concen- tration of d where the velocity is the same at any concentration of B. Similarly the ratio KDF/KF should equal the ratio kll/lclz, or the dissociation constant of TPN.

The spectrofluorometer used was adapted so that corrections could be made for the absorption of the incident and emitted light by the pyridine nucleotide. Thus measurements of protein quenching can be made at rather high concentrations of pyridine

4 Equation 1 is a simplification of the general case for Mecha- nism I. In the general case complex constants such as KABC are not necessarily equal to K~KBK~. These simplifications are valid in the case of the beef enzyme (13) and the frog enzyme.

by guest on March 24, 2018

http://ww

w.jbc.org/

Dow

nloaded from

March 1965 L. A. Fahien, B. 0. Wiggert, and P. P. Cohen 1087

nucleotides. In spite of this fact, in the case of TPNH, accurate measurements cannot be made when the concentration of pyri- dine nucleotide exceeds about 1 X 1OV M. Kinetic studies of the frog liver enzyme reveal t,hat the dissociation constant of TPNH is 2 times higher than this value. When the concentra- tion of TPNH is 9 X 10e5 M there is about 10% quenching of protein emission. Thus, if the maximal amount of protein quenching is 50%, this value agrees with a dissociation con- stant of 2.4 X lop4 M.

Surprisingly, unlike the beef liver enzyme (14), TPN quenches the emission of the frog liver enzyme. Fig. 8 shows the results obtained when TPN is added to the frog liver enzyme. Accurate readings can be obtained with TPN when the concentration is as high as 1 X low3 M. ilt this concentration about 50% of the protein emission is quenched. A linear plot is obtained when

-2 0 2 4 6

I/CTPNI (mM)-’ FIG. 7. Double reciprocal plots of velocity versus TPN con-

centration in the presence of various constant levels of glutamate. The concentrations of glutamate used were: A, 1 X 10-a M; B, 2 X 10m3 M; C, 4 X 10m3 M; D, 5 X 10-a M; and E, the extrapolated line for an infinite concentration of glutamate. Other experi- mental conditions were identical with those described in the legend of Fig. 5.

the reciprocal of the fluorescence is plotted against the concen- tration of TPN. The solid line drawn through the points inter- sects on the abscissa at a TPN value of 1.2 x 10F3 M, which is in excellent agreement with the kinetically determined value of the dissociation constant of TPN. This quenching by TPN cannot be due to TPNH as a contaminant since there is more quenching by TPN than by TPNH.

Haldane Relationship-Comparison of the data presented. in Table II reveal some marked differences between the beef liver and frog liver enzymes. The Haldane relationship correspond- ing to Equations 1 and 2 is

K VI&AK&C

&PP = V&m

(3)

With the use of the kinetic constants obtained for the frog enzyme, the value obtained for K,,, is 9 X lop6 M. This is about a-fold higher than the value obtained kinetically for the beef enzyme by Frieden, but is in the range of the values ob- tained by direct measurement (15). In view of the fact that

0-o (TPN)(mMT’

FIG. 8. Plot of reciprocal of percentage of total fluorescence emitted at 350 rnp as a function of TPN concentration. The enzyme concentration was 0.45 mg per ml and the activating wave length was 290 rnp. The experiments were performed in 0.1 M Tris-acetate buffer, containing 10e4 M EDTA, at pH 8.0, and at 23”.

TABLE II Comparison of kinetic data for beef and frog glutamate dehydrogenases, and lower limits of various rate constants

All experiments were done in 0.01 M Tris-acetate buffer containing 1 X 10-e M EDTA, pH 8.0, at 23”. Values for constants are calcu- lated on the basis of a molecular weight of 250,000 for both beef and frog enzymes.

constant CysyAfin

Mechanism I

KA KA KB KB KC Kc KAKBKC KAKBKC

Vi/W kz = k4

KD KD

KF KF

KDF KDF

KDFIWF) KDF/KF

VR/[J%~ VR/&

V&W(K.~4 kl

V~llEol WB) k3

VFIP-GJW~) ks

VR/L%I(KF) kla

(KDF)(VR)I[EOI(KF)(KD) kll V.dL%l (KD) hz

Constant in terms of Mechanism II Beef

KA

KB KC

KAKBKC

k - cx, k - p, k - y, k - 6 Ko

KF

KDF

KDF/KF

k(a + 11, k(O + 11, W-i + 11, k@ + 1)

k - (n+ 1) k, k-p kb

klz

2.0 x 1o-5 M

3.2 X 1o-3 M

7 x 10-d M 5.8 x 10-l’ M3

250 se6 4.7 x 10-M

1.8 X 1O-3 M

4.2 X 1OF M2

2.3 X 1lP M

8 se6 1 x 107 M-’ set-1 8 X lo* M-I set-1

3.5 x 105 M-’ se6 4.5 x 103 M-l set-1

40 see-l 1.8 x 105 M-l set-1

Frog

2 x 10-4 M

5 x 1o-4 M

5 x lo- M

5 x lo-” M3

100 set-1 5 x 10-4 M

1.8 X 10-3M

2.2 x 1o-6 M2

1.2 x 10-3M

4 se6 5 x 105 M-l set-1

2 x 106 M-l set-l 2 x 104 M-’ set-1

2.2 x 103 M-l set+ 9.6 set-1

8 X lo3 M-I se6

by guest on March 24, 2018

http://ww

w.jbc.org/

Dow

nloaded from

1088 Crystalline Frog Liver Glutamate Dehydrogenase Vol. 240, No. 3

the kinetic constants obtained with the frog enzyme are high, the precision of measurement is not as great with this enzyme. For example, the Michaelis constant of TPNH in the case of the frog enzyme is 2 x 10W4 M, which approaches a range where it is technically difficult to obtain accurate data. Since the Haldane relationship is derived by multiplying many kinetic factors, slight errors in any one constant would be exaggerated and thus could account for the discrepancy.

Rate Constants-Lower limit values of 6 of the 12 rate con- stants shown in Mechanism I can be calculated from the data presented in Table I. These values are shown in Table II and are compared with the values for the beef liver enzyme.

Bloomfield, Peller, and Alberty (16) have derived the rate equation for glutamate dehydrogenase with an arbitrary number of intermediates. Thus, Mechanism I can be written in their terms as follows.

EfF~Xi”...5L. kb + 1) \

k - (01+ 1) ’ ’ . e Xb - 1 (Ha)

Xb-l+D &-Xb= . . . = k-h

UIb)

XD \ k@ + 1) \ k

LXp+C k-@+I) ..’ k_,

xpe...exy \ k(r X 1) \ k,

k - (? X 1) “’ k_, xq + B WC)

xqti ... =x \ k(s + 1) j

k - (6 + 1) (IId)

kh + 1) . . . J-n ‘A+E

k - h + 1)

The lower limits for the values of these rate constants5 based on this expanded mechanism for the frog and beef enzymes are shown in Table II.

Reaction with DPN and DPNH-Kinetic results with the use of DPN and DPNH and the frog enzyme are similar to those obtained with the beef enzyme in that the rate varies with re- spect to the concentration of DPN or DPNH in the presence of constant amounts of other substrates, and in a manner which can be expressed by Equation 4 (15).6

K, v2 + Vl z

v= K2 K&

(4)

1+fi+[A12

Equation 4 is a steady state equation derived for the mechanism which assumes that there are two coenzyme binding sites, an active site and a modifier site. K1 and Kz refer to the Michaelis constants of the coenzymes at the active and modifier sites, respectively; V1 and VZ refer to the velocities when the coenzyme is saturating at the active site and at both sites, respectively; and A refers to the concentration of coenzyme. In the pres-

s The terms used in the original paper of Bloomfield, Peller, and Alberty (16) have been converted into the terms used con- sistently in this paper.

6 The expression of Frieden (15) has been printed incorrectly and has been corrected in the present paper.

ence of 2.5 X 50-2 M glutamate, the values of the kinetic con- stants for DPN are: V1/[Eo] = 1.4 set-I; K1 = 2 X low5 M;

VTB/[EO] = 8.6 set-r; and Kz = 3 X lop4 M. In the presence of 5 X 10W2 M ammonium ion and 2 X IO-% M a.-ketoglutarate the value of the kinetic constants for DPNH are Vr/[E,] = 250 set-l; K1 = 3 X 10e5 M; V2/[E0] = 0; and K2 = 6 X 1O-5 M;

It can be seen that the velocities are faster and the Michaelis constants lower with DPN and DPNH than is the case with TPN and TPNH.

Reaction with Pyruvate-In addition to the use of a-keto- glutarate as a substrate, frog liver glutamate dehydrogenase also catalyzes a reaction with pyruvate. The enzyme catalyzes the oxidation of 2 X 1O-5 millimole of TPNH per minute per ml under the following conditions: the presence of 5 x lo-* M pyru- vate; 0.1 M NH,CI; 1 X lop4 M TPNH; 0.01 M Tris-acetate buffer, pH 8.0, containing 1 X 1O-4 M EDTA; and 0.02 mg of enzyme per ml at 23”.

Absorption Xpectrum-The absorption spectrum of the enzyme was measured with a Beckman DIV spectrophotometer. In 0.1 M Tris-acetate, pH 8.0, and containing 1 X 10e4 M EDTA, at 23” the absorption spectrum of the enzyme reaches a maximum at 280 rnp. The ratio of absorbance at 280 rnp to 260 rnp is 1.9. When the concentration of the enzyme is as high as 4 mg per ml, there is no absorption from 310 m,u through 440 mp. The addition of TPNH or DPNH to this solution did not result in a decrease in the absorbance of the pyridine nucleotide in the region of 310 through 440 mp. The addition of DPN or TPN (1 X lop3 M) to a solution containing 4 mg of the enzyme per ml did not result in an increase in absorbance in this region.

DISCUSSIOh-

The method outlined for preparing frog liver glutamate dehy- drogenase represents a rather simple procedure for obtaining crystalline enzyme. The procedure in addition permits the preparation of both glutamate dehydrogenase and carbamyl phosphate synthetase from the same liver sample. Polyacryl- amide gel electrophoresis and sedimentation experiments indi- cate that the four times crystallized enzyme is essentially free of impurities.

Light scattering data are consistent with the interpretation that the frog enzyme dissociates upon dilution from a dimer with a molecular weight of about 500,000 to a monomer with a mo- lecular weight of about 250,000. This contrasts with the beef enzyme which dissociates from a tetramer with a molecular weight of 1 x lo6 to a monomer with a molecular weight of 250,000 (15) and the chicken liver enzyme which dissociates into subunits with weight average molecular weights of about 430,000 (6).

Consistent with the light scattering data, the sedimentation coefficient of the frog enzyme decreases with dilution of the enzyme. While the sedimentation coefficient decreases, only one peak is observed. This is consistent with the theory of Gilbert (17), which predicts that when polymers greater than dimers are formed, there should be more than one peak observed even if the equilibria are adjusted rapidly. However, Frieden has postulated that consecutive dimerizations could yield only one peak if the intermediate dimer is present in a significant amount (18).

Polyacrylamide gel electrophoresis shows a rather striking difference in the mobilities of frog liver and beef liver glutamate dehydrogena.ses. This difference could be a reflection of differ-

by guest on March 24, 2018

http://ww

w.jbc.org/

Dow

nloaded from

March 1965 L. A. Fahien, B. 0. Wiggert, and P. P. Cohen 1089

ent molecular weights. The same amount (0.2 ml) and concen- tration (0.2 mg per ml) of beef and frog enzyme were put on the gel. At’ these concentrations the molecular weight of the frog enzyme is approximately the same as that of the beef enzyme (15). Thus, the difference in mobility may be related to a dif- ference in charge. It was noted by Markert and M@ller (19) that mammalian glutamate dehydrogenase appeared to be elec- trophoretically immobile in starch gel at pH 8.6. This observa- tion is similar to the result reported here with beef liver glutamate dehydrogenase on polyacrylamide gel at pH 8.4.

The electrophoretic patterns of the beef and frog enzymes on urea gel are essentially the same. Frieden (18) has shown that in 6 R;I urea beef liver glutamate dehydrogenase is rapidly inacti- vated and sediments as one peak with an spO,w of less than 5, indicating that dissociation to lower molecular weight subunits has occurred. Since both the beef and frog enzymes show the same pattern on urea gel, it would appear that urea also causes a dissociation of frog liver glutamate dehydrogenase to lower molecular weight subunits.

Kinetic experiments with TPN and TPNH show that the glutamate dehydrogenase reaction can be described by the same mechanism for both the frog and the beef enzymes. This mecha- nism consists of a sequential order of addition of substrates with the pyridine nucleotide adding first. In the case of the oxida- tion of TPNH, ammonium ions add to the enzyme before o(- ketoglutarate. The steady state derivation of the equation for this mechanism was found to fit the data. While the values of the kinetic constants obtained with the frog enzyme differ from those obtained with the beef enzyme: the Haldane relationship is essentially the same.

The upper limit for the bimolecular rate constant for the asso- ciation of an enzyme with it’s substrate is diffusion controlled and has a, value of about log M-I see-l (20). Steric or geometrical factors would tend to diminish this value (16). The association of TPNH with the frog enzyme has a value of 5 X lo5 M-I set-I, which is 20-fold lower than this value with 6he beef enzyme (16). The bimolecular rate constant of TPN with the frog enzyme is 8 x lo3 M-I see-r, which is about 22-fold lower than the constant with the beef enzyme. This is also about 22.fold lower than the value of this constant for most dehydrogenases (16). This low value is suggestive of pronounced steric factors being in- volved in the frog enzyme-TPN complex. If the ratio of the turnover number of TPNH (Vf/[Eo]) to the rate of dissociation of TPN from the TPN complex (kn) exceeds unity, then more than one type of binary complex, ETPN, must be considered; i.e. X1, Xd, X (b-l) (16). In the case of the beef enzyme this ratio is 6.2, while in the case of the frog enzyme this ratio is 10.4. Wratten and Cleland (21) have suggested that these complexes represent different orientations of the pyridine nucleotide on the enzyme.

It is possible that the quenching of protein emission which TPN produces in the case of the frog enzyme (but not with the beef enzyme) is related to the orientation of the nucleotide. It is of interest that an oxidized pyridine nucleotide (DPN) also quenches protein emission of rabbit muscle triose dehydrogenase (22). In the presence of EDTA this enzyme-DPN complex also shows an elevated absorbance in the 320 to 400 rnp region (23, 24) which disappears in the presence of reagents which at- tack thiol groups. Racker and Krimsky (24) proposed that in this complex there is a covalent bond between a thiol group of the enzyme and DPN. Kosower (25) has suggested that the

increased absorbance is due to a charge transfer interaction between a thiol group and one face of the pyridine ring. Velick has proposed that the increased absorbance in the range of 350 rnp accounts for the quenching of protein emission (22). No increase in absorbance was noted in the 320 to 400 rnp range when 1 x 1O-3 M TPN was added to a solution containing 4 mg of frog glutamate dehydrogenase per ml. It should be noted that the extinction coefficient of the triose dehydrogenase binary complex is low; the molecular weight of this enzyme is about one-half that of the frog glutamate dehydrogenase at this con- centration; and there are three DPN binding sites on t’riose de- hydrogenase (22). It also should be added that pyridine nucleo- tide binding in the case of the beef glutamate dehydrogenase, unlike most dehydrogenases, is apparently unrelated to thiol groups (26). Thus it is possible that the quenching of protein emission observed when TPN is added to frog glutamate dehy- drogenase is related to the absorption of energy at the exciting wave length.

As in most instances when there is a sequential order of addi- tion of substrates, the bimolecular rate constant for association of the second substrate is lower than that of the first substrate but is higher than that of the third substrate (16). This may represent further orientational requirements for these steps. Table II shows that this relationship holds for the frog enzyme.

If the ratio of 2 V,/Vf exceeds unity, then either more than one complex of the type ETPNH or more than one complex ETPNH-NH,+ or both must be considered (16). In the case of the beef and frog enzymes these ratios are 0.066 and 0.08, respectively.

Pyruvate can serve as a substrate for frog liver glutamate dehydrogenase as is also the case with the beef liver enzyme (26).

SUMMARY

A method for preparing crystalline frog liver glutamate dehy- drogenase is described. Light scattering and sedimentation data are consistent with the interpretation that the frog enzyme dissociates from a dimer with a molecular weight of about 500,000 to a monomer with a molecular weight of about 250,000. In 8 M urea the frog and beef enzymes appear to be identical on polyacrylamide gel electrophoresis. The kinetic mechanisms of both enzymes are identical. Triphosphopyridine nucleotide quenches frog enzyme fluorescence but does not quench that of the beef enzyme. From the kinetic data obtained it would ap- pear that the frog enzyme preferentially uses oxidized and re- duced diphosphopyridine nucleotides instead of oxidized and reduced triphosphopyridine nucleotides.

AclenowZedgmentsThe authors are indebted to Dr. Carl Frieden, Washington University, St. Louis, for his advice and assistance with the light scattering experiments and spectro- fluorometric titrations. We wish to thank Dr. R. L. Metzen- berg for running the gel electrophoresis experiments, and Miss Shirley Miekka for the ultracentrifuge analysis.

1. OLSON, J., AND ANFINSEN, C. B., J. Biol. Chem., 197, 67 (1952). 2. STRECKER, H. J., Arch. Biochem. and Biophys., 46, 128 (1953). 3. SNOKE, J. E., J. Biol. Chem., 223, 271 (1956). 4. KUBO, H., IWATSUBO, M., WATARI, H., AND SOYAMA, T., J.

REFERENCES

Biochem. (Tokyo), 46, 1171 (1959).

by guest on March 24, 2018

http://ww

w.jbc.org/

Dow

nloaded from

1090 Crystalline Frog Liver Glutamate Dehydrogenase Vol. 240, No. 3

5.

6. 7.

8.

9.

10. 11.

12.

13. 14. 15.

TOMKINS, G. M., YIELDING, K. L., AND CURRAN, J. F., J. (Editors), The enzymes, Vol. 7, Ed. 2, Academic Press, Biol. Chem., 237,1704 (1962). Inc., New York, 1963, p. 3.

FRIEDEN, C., Biochim. et Biophys. Acta, 62, 421 (1962). FAHIEN. L. A.. AND COHEN. P. P.. J. Biol. Chem.. 239. 1925

(1964). ’ I

WARBURG, O., AND CHRISTIAN, W., Biochem. Z., 310, 354 (1941).

LOWRY, 0. H., ROSEBROUGH, N. J., FARR, A. L.,AND RANDALL, R. J.; J. BibZ. Chem., 193,.265 (i951).

FRIEDEN. C.. J. Biol. Chem.. 238.3286 (1963). ORNST&, L:, AND DAVIS, g. T.,. Disc kectkphoresis, preprint

available from Eastman Chemical Company, Rochester, New York.

MARSHALL, M., METZENBERG, R. L., AND COHEN, P. P., J. Biol. Chem., 233, 102 (1958).

FRIEDEN, C., J. Biol. Chem., 234, 2891 (1959). FRIEDEN, C., J. Biol. Chem., 238, 146 (1963). FRIEDEN, C., in P. D. BOYER, H. A. LARDY, AND K. MYRB;~CB

16. BLOOMFIELD, V., PELLER, L., AND ALBERTY, R. A., J. Am. Chem. SOL, 84, 4367 (1962).

17. GILBERT, G. A., Discussions Faraday Sot., No. 2, 68 (1955). 18. FRIEDEN; C., J.‘Biol. Chem., 237, 2396 (1962). 19. MARKERT. C. L.. AND MQJLLER. F. Proc. Natl. Acad. Sci.

20.

21.

22. 23. 24. 25. 26.

u. s., 45, 735 (i959). ’ ’ ALBERTY, R. A., AND HAMMES, G. G., J. Phys. Chem., 62,154

(1958). WRATTEN, G. C., AND CLELAND, W. W., Biochemistry, 2, 935

(1963). VELICK, S. F., J. Biol. Chem., 233, 1455 (1958). VELICK, S. F., J. Biol. Chem., 203, 563 (1953). RS~CKER, E., AND KRIMSKY, I:, J. Biol. Chem., 198, 731 (1952). KOSOWER. E. M.. J. Am. Chem. Sot.. 78. 3497 (1956). FISHER, g. F., AND MCGREGOR, L. L:, J.. Biol. &hem., 236, 791

(1961).

by guest on March 24, 2018

http://ww

w.jbc.org/

Dow

nloaded from

Leonard A. Fahien, Barbara O. Wiggert and Philip P. CohenLiver

Crystallization and Kinetic Properties of Glutamate Dehydrogenase from Frog

1965, 240:1083-1090.J. Biol. Chem. 

  http://www.jbc.org/content/240/3/1083.citation

Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/240/3/1083.citation.full.html#ref-list-1

This article cites 0 references, 0 of which can be accessed free at

by guest on March 24, 2018

http://ww

w.jbc.org/

Dow

nloaded from