JOURNAL OF BACTERIOLOGY, May 1972, p. 494-503Copyright 0 1972 American Society for Microbiology

Vol. 110, No. 2Printed in U.S.A.

Glutamate Dehydrogenase from Mycoplasmalaidlawii

GERALD YARRISON, DIANE W. YOUNG,' AND G. L. CHOULESDepartment of Biology, University of Utah, Salt Lake City, Utah 84112

Received for publication 27 September 1971

Mycoplasma laidlawii possesses a single glutamate dehydrogenase (GDH)with dual coenzyme specificity [specificity for nicotinamide adenine dinucleo-tide (H) and nicotinamide adenine dinucleotide phosphate (H) ]. A purificationprocedure is reported which results in an enzyme preparation with a specificactivity of 79.5 units/mg and which displays only one significant protein bandafter gel electrophoresis. This one band was determined, by activity staining,to have all of the GDH nucleotide specificities. The molecular weight of theenzyme is 250,000 + 10%, and it has a subunit size of about 48,000. The en-zyme exhibits measurable activity with aspartate and pyruvate but is inactivewith eight other possible substrates. Purine nucleotides do not affect the activ-ity. The Km for reduced nicotinamide adenine dinucleotide was 1.8 x 10-4 M.The optimal substrate concentrations and pH optimum for each of the respec-

tive GDH activities are also reported.

Mycoplasma laidlawii is interesting fromboth a physiological and phylogenetic stand-point because of the small size of its genome(1). We have chosen to study glutamate dehy-drogenase activity in M. laidlawii because ithas been characterized in many other orga-nisms and because, in vivo, it serves as a linkbetween carbohydrate metabolism and ni-trogen metabolism and is commonly a controlpoint.

Glutamate dehydrogenases catalyze the fol-lowing reversible reaction:ot-ketoglutarate + NADH + NH4 +

(NADPH)glutamate + NAD+ + H,O

(NADP+)

A given glutamate dehydrogenase (GDH)may show preference for either nicotinamideadenine dinucleotide (NAD), reduced form(NADH), or nicotinamide adenine dinucleotidephosphate (NADP), reduced form (NADPH),or may show specificity for both (dual speci-ficity). To provide clarity in the following text,we have used a shortened notation to indicateboth the form of the cofactor and direction ofreaction. For example, the NADH-specificGDH activity that catalyzes the forward re-action is abbreviated NADH-GDH (2).

In this paper, we describe methods for the1 Present address: Department of Zoology, University of

Texas, Austin 78712.

purification of GDH and give evidence thatthere exists only one species of this enzyme inM. laidlawii with dual nucleotide specificity(E.C. 1.4.1.3). Several other properties of thedual enzyme are described, followed by a dis-cussion of their relationship to properties ofGDH enzymes from other organisms.

MATERIALS AND METHODSChemicals. a-Ketoglutaric acid, NAD, NADPH,

ribonuclease (all A grade) and deoxyribonuclease (Bgrade) were obtained from Calbiochem. Phenol re-agent (Folin-Ciocalteau) was from Fisher ScientificCo. Chemicals for acrylamide gel electrophoresiswere all products of Matheson, Coleman, and Bell.Phenazine methosulfate, MTT tetrazolium [3(4,5dimethyl thiazolyl-2)-2, 5-diphenyl tetrazoliumbromide], tris(hydroxymethyl) aminomethane (Tris),NADP (Sigma grade), and NADH (grade 111) wereobtained from Sigma Chemical Co. Sepharose 6Band diethylaminoethyl (DEAE) Sephadex A-50 wereobtained from Pharmacia Fine Chemicals, Inc.Ammonium sulfate, A.R. was from MallinckrodtChemical Works. DEAE cellulose DE.11 was ob-tained from Reeve Angel, Inc.Crude extract preparation. M. laidlawii, strain

A, culture and osmotic lysis procedure were as de-scribed previously (4). The organisms were extractedonce with 10-3 M ethylenediaminetetraacetic acid(EDTA), pH 7.5, followed by two more extractionswith demineralized water. The extracts were thencombined and subjected to the enzyme fractionationprocedure described later.

Protein determination. To eliminate interference494

on April 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

MYCOPLASMA GLUTAMATE DEHYDROGENASE

from substances such as Tris, the protein in eachsample was precipitated in 5% trichloroacetic acid.The precipitate was recovered by centrifugation andsuspended in water, and protein was determined bythe Lowry method (17).

Assay procedures. The optimal substrate concen-trations (defined in Fig. 5) were used in each of thefollowing assay mixtures. Activity was measured bythe increase or decrease of 340-nm absorbancy with aprecision spectrophotometer (Bausch & Lomb, Inc.)equipped with a Sargent SRLG recorder. During theNADH-GDH assay, a-ketoglutarate was withheldfrom the assay mixture until the NADH oxidase ac-tivity could be determined. This activity was presentin the crude, but not in the pure, preparations. Aunit of enzyme activity is defined as the amount ofenzyme which catalyzes formation of 1 umole of nu-cleotide product per min. Specific activity is definedas units per milligram of protein.The NADH-GDH assay mixture contained 133

mM ammonium ion, 16.7 mM a-ketoglutarate, and0.35 mM NADH, final pH 8.9. The procedure was asfollows: to 2.6 ml of 0.1 M arginine-hydrochloridebuffer (pH 8.9) was added 0.1 ml of 4.0 M NH4C1(pH 9.1), 0.1 ml of 10.5 mm NADH, and 0.1 ml of en-zyme. The reaction was then initiated with 0.1 ml of0.5 M sodium a-ketoglutarate, pH approximately 9.0.

The NADPH-GDH assay mixture contained 80.0mM ammonium ion, 20.0 mM a-ketoglutarate, and0.45 mM NADPH, final pH 8.8. The procedure wasas follows: to 2.6 ml of 0.1 M arginine hydrochloridebuffer (pH 8.8) was added 0.1 ml of 2.4 M NH4Cl(pH 9.0), 0.1 ml of 13.5 mM NADPH, and 0.1 ml ofenzyme. Finally, 0.1 ml of 0.6 M sodium a-ketoglu-tarate, pH 9.0, was added.The NAD-GDH assay mixture contained 167 mM

L-glutamate and 2.41 mM NAD, final pH 9.6. Theprocedure was as follows: to 2.5 ml of 0.1 M glycine-hydrochloride buffer (pH 9.6) was added 0.2 ml of36.2 mM NAD, 0.1 ml of enzyme, and finally, 0.2 mlof 2.5 M sodium glutamate, pH 9.8.The NADP-GDH assay mixture contained 250 mM

L-glutamate and 1.63 mm NADP, final pH 8.9. Theprocedure was as follows: to 2.4 ml of 0.1 M arginine-hydrochloride buffer (pH 8.9) was added 0.2 ml of24.5 mm NADP, 0.1 ml enzyme, and finally, 0.3 mlof 2.5 M sodium glutamate, pH 9.1.

Gel electrophoresis. The apparatus and proce-dure for gel electrophoresis was essentially that ofChoules and Zimm (5). The electrophoresis proce-dure resembles that of Davis (7) except that in ear-lier experiments, where a variety of pH values wereneeded, a continuous buffer system was employed(buffer bath, sample, and gel all at the same pH).The sample was mixed with 10% sucrose to layer iton the column instead of using a sample gel. Therunning gels contained 7% acrylamide. The voltagewas adjusted to keep the current below 3 ma pertube. Gels were 10 cm long.

Sodium dodecyl sulfate (SDS) gel electrophoresiswas done by the method of Davis, except that 0.1%SDS was added to the gels and buffers as suggestedby Shapiro, et al. (18). The running gels contained15% acrylamide. Gels were protein-stained with a

solution of 1% aniline blue, 40% ethanol, and 7%acetic acid. Samples were dissolved in 1% SDS, withor without 0.5% mercaptoethanol, and heat-treatedfor 5 min in a boiling water bath. Reduced proteinstandards were run concurrently with the sample (8).The logarithm of molecular weight versus mobilitywas plotted for the standards, and the resulting plotwas used to find the molecular weight of the sample(18).

Activity staining. The procedure was similar tothat originally developed by Laycock, et al. (16) foruse with isoenzymes of lactic dehydrogenase. Thepositive staining solutions for detecting glutamatedehydrogenase activity consisted of 25 mg of L-glu-tamate, 0.5 mg of phenazine methosulfate, 5 mg ofMTT tetrazolium, 5 mg of NAD or NADP, and 5 mlof 0.1 M arginine buffer, pH 8.9. Gels were incubatedin the staining solution for a few minutes until bluebands appeared, marking the zones containing theglutamate dehydrogenase, and then were stored in7% acetic acid.

Since it was also desirable to detect the reverseactivities, we developed a method of negativestaining. First, gels were placed in a solution con-sisting of 25 mg of a-ketoglutarate, 10 mg of NH4Cl,5 mg of NADH or NADPH, and 5 ml of 0.1 M argi-nine buffer, pH 8.9. The gels were incubated for 5 to10 min and then rinsed with buffer. The rinsed gelswere kept dry for 5 min and then introduced into thefollowing solution: 5 mg of MTT tetrazolium, 0.5 mgof phenazine- methosulfate, and 5 ml of 0.1 M argi-nine buffer, pH 8.9. In a short time, the entire gelwas stained except where the glutamate dehydro-genase was present. The staining was specific forglutamate dehydrogenase because if glutamate (posi-tive activity staining) or a-ketoglutarate (negativeactivity staining) was deleted, no staining (or nega-tive staining) occurred.

Substrate specificity tests. To test substratespecificity, activity staining solutions were preparedas above, except that the indicated amino acids (Fig.9) were used in place of glutamate. Substitutes for a-ketoglutarate were used for negative stains in thesame figure.

After continuous gel electrophoresis at pH 9.5 for1.5 hr at 100 v, the gels were removed and placed inthe activity staining solutions. The gels were placedin a refrigerator and allowed to stain overnight.These gels were 5 cm long.

Heat stability. Heat treatments were performedin a Bausch & Lomb, Inc., thermoelectric heater-cooler. Tris buffer (0.05 M, pH 8.0) was incubated for5 min to bring it to the desired temperature andthen enzyme was added. Samples were withdrawn at5-min intervals during the treatment and were as-sayed at the conclusion of each heat experiment. Thetemperature was constant within 0.2 C during thecourse of the experiments.pH studies. The effect ofpH was studied with all

of the reactants present in optimal concentrations.The pH was determined after the completion of eachassay and was -found to vary from the initial value byless than 0.05 pH per unit. To achieve this low varia-tion, we found it necessary to prepare series of

495VOL. 110, 1972

on April 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

YARRISON, YOUNG, AND CHOULES

ammonium chloride and sodium glutamate stocksolutions, 0.2 pH units above the corresponding argi-nine or glycine buffers. Either ammonium or gluta-mate ions at these relatively high concentrationswould lower the pH.

RESULTS

Enzyme purification. All operations were

performed at 0 to 4 C unless otherwise stated.(i) Ammonium sulfate fractionation. One to

two liters of crude extract (see Materials andMethods) were adjusted to pH 7.5 and nucleicacid was broken down by digestion for 30 minat room temperature after the addition of 1 ,gof ribonuclease per ml, 1 Mg of deoxyribonu-clease per ml, and 2 x 10-3 M magnesium ion.The extract then was dialyzed against 40%saturated ammonium sulfate, pH 8.2, in a

tightly closed vessel for about 16 hr. Thesample was centrifuged for 10 min at 23,000 xg and the pellet was discarded. Solid ammo-nium sulfate then was added to the superna-tant fluid to bring it to 55% saturation. ThepH was adjusted to 8.0 and the mixture wasallowed to stir for 1 hr. The sample was centri-fuged for 45 min at 23,000 x g, and the pelletwas dissolved in about 15 ml of buffer con-

taining 0.05 M Tris and 10-3 M EDTA adjustedto pH 8.0 with HCl. Enzyme recovery rangedfrom 65 to 81% of the enzyme activity in thecrude extract.

(ii) Heat denaturation. The buffered extractwas immersed in a 62 C bath for 15 min, andthen centrifuged for 30 min at 40,000 x g. Thepellet was washed with 5 ml of fresh buffer.The combined supematant fluids retained 70to 99% of the enzyme activity before heattreatment.

(iii) DEAE cellulose absorption and gel fil-tration. The heat-treated material was dia-lyzed against 0.01 M Tris-hydrochloride buffer,pH 8.0, and 10-3 M EDTA (Tris-EDTA buffer).DEAE cellulose was hydrated and equilibratedwith the same buffer used in the dialysis.About 10 ml of the equilibrated DEAE cellu-lose was added to the dialysate. Dry NaCl was

added to 0.2 M, and the mixture was stirred for30 min and then centrifuged. The pelletedresin was placed in a 15 by 50-mm column inseries with an upward-flow Sepharose 6Bcolumn (25 by 400 mm). The two columns thenwere eluted with 0.5 M NaCl in Tris-EDTAbuffer. The enzyme usually appeared in frac-tions 21 to 29 when collecting 4.5-ml fractions.About 60% of the enzyme present in the heat-treated fraction was recovered.

(iv) DEAE Sephadex chromatography. ADEAE Sephadex A-50 column (12 by 127 mm)

was equilibrated with Tris-EDTA buffer. Thesample was dialyzed against 60% saturatedammonium sulfate containing 10-3 M EDTAfor 12 to 16 hr and then centrifuged at 38,000x g. The precipitate was dissolved in 1 to 2 mlof buffer and then was dialyzed against Tris-EDTA buffer. The dialysate was brought to10% with dry sucrose, applied to the top of thecolumn, and then eluted with a linear sodiumchloride gradient (0 to 1 M) in Tris-EDTAbuffer. The total volume of the gradient was250 ml. Enzyme was eluted at about 0.4 MNaCl. Fractions were pooled and dialyzedagainst 60% saturated ammonium sulfate con-taining 10-3 M EDTA for several hours. Theresulting precipitate was recovered by centrif-ugation at 40,000 x g and then dissolved in asmall amount of buffer. The yield was 0.1 to0.2 mg of protein. Recovery was usually about25% of the enzyme applied to the column.Table 1 summarizes a representative enzymepreparation.Gel electrophoresis and activity staining.

This purification scheme resulted in a prepara-tion that displayed only one band (with signifi-cant intensity) after gel electrophoresis. Thisband, in tum, displayed all four nucleotidespecificities (Fig. 1). For establishing nucleo-tide specificities, the gel electrophoresis was,in effect, the last step in the purification pro-cedure. Earlier activity staining studies alsoshowed that the crude extract had only oneactive band and, moreover, the respective ac-tivities were inseparable even though electro-phoresis was performed at pH 7.0, 8.0, and 9.5(the maximum range in which the enzyme isstable).The final specific activity (Table 1) was 79.5

units/mg for NADH-GDH. This is almosttwice the specific activity reported for crystal-line bovine liver NADH-GDH (9) and overthree times that for crystalline rat liver

TABLE 1. Enzyme purification

Recov-ery of NADH- Total

Stage of purification NADH- GDH proteinGDH specific proteiactivity activity (mg)

(%)Crude extract ........... 100.0 0.14 828.440% Saturated (NH4)2SO4

supernatant .107.0 0.25 492.055% Saturated (NH4)2SO4

pellet ................ 91.5 0.80 136.3Heat-treated ............ 64.2 5.63 13.6DEAE cellulose absorbed-

Sepharose 6B filtered . . 40.9 56.8 0.86DEAE Sephadex ........ 9.7 79.5 0.14

496 J. BACTERIOL.

on April 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

MYCOPLASMA GLUTAMATE DEHYDROGENASE

Copurification of GDH activities. Each ofthe activities peaked in the same fraction aftergel filtration or DEAE Sephadex chromatog-raphy (Fig. 2 and 3). However, it was of moreinterest to find that all four activities copuri-fied at essentially the same rate. As is shownin Table 2, the ratios of the GDH activitiesdiffered very little from step to step in thepurification. The NADH-oxidase activity, incontrast, diminished with each step and finallywas eliminated. The data in Table 2 were de-rived by using an earlier purification schemewhich differed from the present one in that theammonium sulfate fractionation involved abroader cut and the DEAE cellulose absorp-tion step was not used. Instead, the prepara-

l ~~~~~~~~~~~~~~~~~~........... ........

..

.. ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~.......

FIG. 1. Acrylamide gel electrophoresis of purifiedglutamate dehydrogenase at pH 8.3. Gels were splitlengthwise in half and then stained differentially.The split halves were arranged in pairs. A, The posi-tion of the NADPH-GDH band was compared toNAD-GDH. by negative and positive activity stains,respectively. B, NADH-GDH is. compared withNAD-GDH. C, NADP-GDH is compared with NAD-GDH. D, Protein stain by aniline blue is comparedwith NAD-GDH. In each case, 10 tol15, g of GDHwas applied to the gels. The NADH gel was stainedbefore -the others, and the broadened clear area wasdeveloping a positive center, apparently due to reac-tion reversal after standing.

NADH-GDH (14). However, in view of thetrace bands visible after electrophoresis (Fig.7), the enzyme may not have been more than70% pure. This conclusion is tempered by thefact that one of the trace bands (2 cm from thebottom of the gel) is in a position where wecommonly see an artifact band. The specificactivity measured for the Mycoplasma enzymecorresponds to a turnover number of 19,800 permin per mole of 250,000 MW enzyme, as-suming 100% purity. At the theoretical Vmax,and assuming 70% purity, the tumover numberwould be 47,500 per min per mole.These electrophoresis studies provided our

best evidence for a single GDH enzyme havingthe dual specificity mentioned earlier. Otherlines of evidence are described below.

FIG. 2. Sepharose 6R filtration experiment. Theenzyme preparation applied to the column waspartly purified by ammonium sulfate fractionationand heat treatment. The 25 by 400-mm column waseluted with Tris-EDTA buffer, pH 8.0. Blue dextranpeaked at fraction 15 and methylene blue peaked atfraction 49. Fraction size was 4.5 ml. Sample to bedvolume ratio was 1:50; flow rate was 25 ml/hr.

FRACTION (ml)FIG. 3. DEAE Sephadex column chromatography

of heat-treated enzyme prepared as in Fig. 3. The 25by 120-mm column was eluted with a 150-ml lineargradient of 0 to 0.8 M sodium chloride buffered atpH 7.0. The NADP-GDH activities were not meas-ured. Flow rate was 5 ml/hr.

497VOL. 110, 1972

on April 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

YARRISON, YOUNG, AND CHOULES

tion was filtered two times with Sepharose 6B.The present procedure yielded about fivetimes better purity. The last column in Table2 shows the relative percentages of the respec-tive activities in our best purity preparation.Because of its low level of activity, the

NADP-GDH activity was not detected in thecrude extract, but there was no reason to be-lieve it was absent. It was necessary to dialyzethe 35% saturated ammonium sulfate superna-tant fraction against dilute buffer before theoxidized nucleotide activities could be assayed.Ammonium ion was very inhibitory whenpresent in high concentration, presumably dueto the fact that it was a product of the reversereaction.

Effect of coenzymes on heat stability. TheGDH activities showed closely similar thermaldenaturation properties. The resulting denat-uration curves followed first-order kinetics,and accordingly, rate constants were calcu-lated from the relation: ln at = kt + ln ao, inwhich a, k, and t, respectively, are the enzymeactivity, rate constant, and time. At 66 C, thefirst-order rate constants for thermal inactiva-tion of ammonium sulfate fractionated enzyme

were 0.44, 0.49, and 0.40 at pH 8.0 for NADH-GDH, NADPH-GDH, and NAD-GDH activi-ties, respectively. At 67 C, the respective rateconstants were 0.73, 0.81, and 0.88.

Since these rate constants gave us no basisfor believing that there was more than oneenzyme, we studied next the effect of the re-spective coenzymes on the heat stability of200-fold purified enzyme to provide a morecritical test of homogeneity. We found thatNADH and NADPH at 10-4 M were equallyeffective in protecting either NADH-GDH orNADPH-GDH activities at 66 C (Fig. 4).Moreover, each cofactor provided substantialprotection at that temperature.

Similar tests were performed in the presenceof the same coenzymes at 5 x 10-6 M. No ef-fect was observed at these low cofactor levels.The special significance of these cross-protec-tion experiments is discussed later.Properties of the enzyme: substrate op-

tima. It was interesting to note that reducednucleotide GDH activities exhibited definiteoptima in their response to substrate concen-trations (Fig. 5), whereas the oxidized nucleo-tide activities did not. The substrate inhibition

TABLE 2. Relative purification of all activities

Percentage of the NADH-GDH activity at each stage in purificationGDH activity Crude 35% Cut DEAE Purified

extract treatment Sephadex GDH

NADH 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0NADPH 23.3 27.5 32.4 31.8 34.2 29.8 30.5 33.3NAD 34.7 33.0 36.6 34.5 32.4 36.2 35.0 36.2NADP 6.0 7.5 7.2 6.9 8.5 8.5 6.3NADH oxidase 30.5 [ 22.9 [ 23.5 J 11.1 3.9 2.1 0.0 0.0

15MINUTES MINUTES

FIG. 4. A, Protection of NADH-GDH activity by 10-4 M reduced coenzymes during heat treatment at 66C, pH 8.0. B, Protection of NADPH-GDH activity under the same conditions. Lower curve in each case isthe unprotected control. Natural logarithm of activity, in milliunits, is plotted against time. This enzymepreparation had a specific activity of 28 units/mg.

498 J. BACTERIOL.

on April 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

MYCOPLASMA GLUTAMATE DEHYDROGENASE

DNAD

NADP

II

0.6 1.8 3.0mM PYRIDINE NUCLEOTIDE

FIG. 5. Substrate optima of the various glutamate cdehydrogenase activities.possessed by the Mycoplasmaenzyme. In the experiments represented by each graph, all but one of the appropriate substrates were heldconstant at near optimal concentrations. The substrate that was varied is listed below each graph. The curves

are each labeled with the nucleotide specificity tested.

shown at the higher substrate concentrationsmay have resulted from the binding of more

than one substrate molecule by individual ac-tive sites. (Refer to the assay mixtures re-

ported in Materials and Methods for thevalues of the optima.)pH optima. Three of the four activities had

similar pH optima (pH 8.8-8.9, Fig. 6), but theNAD-GDH optimum was significantly dif-ferent (pH 9.6). This is discussed later. Theenzyme material used in this experiment waspurified through. two filtrations on Sepharose6B but was not DEAE cellulose-absorbed.To test whether the arginine or glycine

might be inhibitory, NADH-GDH activity wasmeasured in both arginine and Tris buffers atpH 8.8 and in both arginine and glycine buff-ers at pH 9.4. No difference was observed ineither case.

Molecular weight and subunit size. Theapproximate molecular weight of M. laidlawiiglutamate dehydrogenase was estimated by a

modification of the activity sedimentationmethod of Cohen (6). Details of our procedurewill be published elsewhere (22). We found thatthe molecular weight was 250,000 i 10%.A single subunit was observed following

SDS-gel electrophoresis (Fig. 7). This subunithad a molecular weight of approximately48,000 and did not dissociate further aftermercaptoethanol treatment.Michaelis-Menten constants. The Line-

weaver-Burk plot (Fig. 8) yielded a Km of 1.8 x10-4 M for NADH. Relatively crude ammo-

nium sulfate fractionated enzyme was used inthis study. Later, a 200-fold purified prepara-tion gave nearly the same Km, but the datawere not as good. In addition, approximate Km

I-

I-4

C.)

499VOL. 110, 1972

on April 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

YARRISON, YOUNG, AND CHOULES

41.0 2.0,

)05 7I0=

80 90 100pH

FIG. 6. Determination of pH optima. Substrateswere held at their optimal concentrations while pHwas -varied. Each- enzyme cofactor is indicated.Every enzyme preparation used was diluted to thesame level of NADH-GDH activity. Note that theNADH-GDH is plotted on a reduced scale, as indi-cated on the right vertical axis.

AgIdb1sr C c

FIG. 7. Subunit analysis by SDS-gel electropho-resis at pH 8.3. Loading was 10 to 15 jAg of proteinper gel Gels were stained with aniline blue followingelectrophoresis. A, Purified enzyme heat-treated 5min at boiling temperature (96 C) in the presence of1% SDS. B, Enzyme heat-treated in the presence of1% .SDS plus 0.5% mercaptoethanol. C, Untreatedenzyme run without SDS.

ESY1jimole-1FIG. 8. Lineweaver-Burk plot from which the

Michaelis constant for NADH was calculated. Recip-rocal velocity is plotted against reciprocal substrate(NADH) concentration.

values for the other pyridine nucleotides andsubstrates (Table 3) were calculated from re-ciprocal plots of the data in Fig. 5.Enzyme stability studies. The enzyme was

stabilized by EDTA (Table 4). Without EDTA,activity dropped off rapidly for a short timeafter the enzyme was introduced into newmedia and then remained constant. Similarstabilization was obtained with bovine serumalbumin (BSA) added, and it is likely thatBSA simply acted as a chelating agent to re-move heavy metal inhibitors. On the otherhand, the disulfide bond reducing agents, glu-tathione and mercaptoethanol, rapidly dena-tured the enzyme.Preliminary results also indicated that

ammonium sulfate had little effect in the 0 to50% saturation range and that the enzyme wasstable in the range of pH 7.0 to 8.5. Thesestudies were performed in the presence of 10-3M EDTA.

Effect of purine nucleotides. The effect ofthe purine nucleotides was investigated at anucleotide concentration of 100 jsM (Table 5).Neither inhibition nor activation was observedwith these nucleotides at the stated concentra-tion.

Substrate specificity. Substrate specificitywas investigated by means of positive andnegative staining (Fig. 9), following gel electro-phoresis, to provide high sensitivity and elimi-nate the necessity of using a highly purifiedsample. There were clearly stained bands cor-responding to aspartate and pyruvate and,possibly, a very faint band for alanine. In allother cases, however, no activity could be de-tected.

DISCUSSIONEvidence that only one GDH enzyme ex-

ists. The best evidence that M. laidlawii con-

500 J. BACTERIOL.

I.-

P.- i on April 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

MYCOPLASMA GLUTAMATE DEHYDROGENASE

TABLE 3. Approximate Km values

NADH-GDH NADPH-GDH NAD-GDH NADP-GDH

Substrate Km (mM) Substrate Km (mM) Substrate Km (mM) Substrate Km (mM)

NADH 0.18 NADPH 0.20 NAD 0.32 NADP 0.25a-kga 7.1 a-kg" 5.0 glutamate 20.0 glutamate 32.5NH4+ 30.0 NH4+ 5.5

a a-Ketoglutarate.

tains only one glutamate dehydrogenase withdual specificity comes from the finding thatthe purified preparation shows only one signif-icant band after gel electrophoresis and thatthis band, in turn, has all four nucleotide ac-

tivities. The argument is considerablystrengthened, however, by the fact that all fouractivities are copurified at essentially the same

rate through all steps in the purification proce-

dure. If, for example, there were more than one

GDH enzyme initially present in the crudeextract, we would expect to see a definitechange in the ratios during purification.

Additional evidence is provided by the re-

sult that all activities peak in the same frac-tion after Sepharose 6B filtration and DEAESephadex chromatography. Further evidence isprovided by the fact that both reduced nucleo-tides were equally effective in protecting theenzyme against thermal denaturation. Here,there is a strong implication that the activitiesare not only on one enzyme, but share thesame active site(s).We cannot explain definitely why the pH

optimum of the NAD-GDH activity is signifi-cantly different from the others, but the an-

swer may lie in a difference in attachmentpoints within the active site.Comparative properties. Evidence that M.

laidlawii possesses only one GDH enzyme con-

trasts with the finding that Escherichia coli,yeast, and Neurospora each contain two suchenzymes (11) but is similar to animal enzymeswhich have dual specificity. However, Chlo-rella pyrenoidosa also appears to possess a

single GDH enzyme (19), and it appears thatM. laidlawii may be unusual for a microorga-nism but not unique in this respect. On theother hand, the M. kaidlawii enzyme resemblesthat of other microorganisms tested (16) be-cause it is not affected by purine nucleotides.Such nucleotides have been found to affectanimal GDH enzymes when present at lessthan 100 gM. The molecular weight reportedhere (250,000 i 10%) is typical for GDH en-

zymes found in other microorganisms. Corre-sponding vertebrate enzymes tend to be muchlarger, at least in the associated form.

TABLE 4. NADH-GDH stability study

Activityremaining

Stabilizing agent Concn after 21days at4 C (%)-

None 86BSA 1 mg/ml 98BSA 10 mg/ml 98EDTA 10-4 M 96Glutathione 10-4 M 7Mercaptoethanol 10-4 M 28

a Samples were buffered with 0.05 M Tris-hydro-chloride, pH 8.0. Initially, the NADH-GDH activitywas 0.46 units/ml and the specific activity was 2.3units/mg.

TABLE 5. Lack of effect of purine nucleotides onglutamate dehydrogenase

Per cent activity with

Enzyme added nucleotides

activity' AMPb ADPb ATPb

NADH-GDH 104 105 99NADPH-GDH 104 106 100NAD-GDH 100 105 98

a The enzyme preparation used had a NADH-GDH activity of 8.2 units/ml and specific activity of28 units/mg.

bConcentration of adenosine monophosphate(AMP), adenosine diphosphate (ADP), and adeno-sine triphosphate (ATP) was 100 ,M.

Beef liver glutamate dehydrogenase wasused as a standard for our molecular weightdetermination. This enzyme has a monomermolecular weight of 320,000 (3) which is com-posed of six polypeptide chains having a mo-lecular weight of 57,000. Similarly, the Myco-plasma enzyme has a single-sized peptide sub-unit of approximately 48,000 molecular weight.This would suggest a subunit number of fivebut, with the uncertainties involved here, thenumber could be six, or even four. It is alsointeresting to note that both enzymes lack di-sulfide bonds between subunits.The substrate specificity results indicate a

VOL. 110, 1972 501

on April 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

YARRISON, YOUNG, AND CHOULES

_ _. M

* .-B4biI

_. ... _ . .....~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

,5

FIG. 9. Substrate specificity study by activity staining. Gels represent specificity for (from left to right):(A) alanine, aspartate, -glutamate, and isoleucine; (B) glutamate (again), leucine, methionine, proline, andvaline. Negative activity stains (C) represent: a-ketoglutarate, pyruvate, a-ketobutyrate, and a-ketovalerate.

relatively narrow specificity range becauseonly two of the 10 possibly alternative sub-strates tested (aspartate and pyruvate) are def-initely active. One mammalian glutamatedehydrogenase (21) and Neurospora crassa (20)both have been shown to be active with severalmore of these substrates.

Finally, we noted that the thermal stabilityof the M. Iaidlawii glutamate dehydrogenase isunusual. Such differences add to the beliefthat mycoplasmas are a separate and distinctgroup of organisms not, for example, closelyrelated to bacteria (12). Thus, it appears im-portant to do partial sequence analysis on thisand other GDH enzymes to further charac-terize this protein and better define its rela-tionships to other glutamate dehydrogenases.Work of this kind has already begun with Neu-rospora NADP-GDH enzyme (13). Consider-able progress has also been reported on thebeef liver GDH sequence (15).

ACKNOWLEDGMENTSWe thank Kent McDonald for his fine technical assist-

ance. We also thank S. F. Velick for reviewing our paperand for his many valuable suggestions. Thanks are dueGordon Lark for suggesting the effective cross-protectionexperiment.

This work was supported by,the following grants: Amer-ican Cancer Society no. P-604 and no. 235-515103, NationalInstitutes of Health no. 5505 FR07092, and the University ofUtah Research Fund.

LITERATURE CITED1. Bak, A. L., F. T. Black, C. Christianson, and E. A.

Freundt. 1969. Genome size of Mycoplasma DNA.Nature (London) 224:1209-1210.

2. Barratt, R. W., and W. N. Strickland. 1963. Purificationand characterization of a TPN-specific GDH fromNeurospora crassa. Arch. Biochem. Biophys. 102:66-67.

3. Cassman, M., and H. K. Schachman. 1971. Sedimenta-

tion equilibrium studies on glutamic dehydrogenase.Biochemistry 10:1015-1024.

4. Choules, G. L., and R. F. Bjorklund. 1970. Evidence of,-structure in Mycoplasma membranes. Biochemis-try 9:4759-4767.

5. Choules, G. L., and B. H. Zimm. 1965. An acrylamidegel soluble in scintillation fluid: its application toelectrophoresis at low pH. Anal. Biochem. 13:336-344.

6. Cohen, R. 1963. Sur un nouveau mode d'utilisation descentrifugeuses analytiques. C. R. Acad. Sci. 256:3513-3515.

7. Davis, B. J. 1964. Disc electrophoresis. II. Method andapplication to human serum proteins. Ann. N.Y.Acad. Sci. 121:404-427.

8. Fish, W. N., J. A. Reynolds, and C. Tanford. 1970. Gelchromatography of proteins in denaturing solvents. J.Biol. Chem. 245:5166-5168.

9. Frieden, C. 1959. Glutamate dehydrogenase. L. The ef-fect of coenzyme on the sedimentation velocity andkinetic behavior. J. Biol. Chem. 234:804-814.

10. Frieden, C. 1964. Comparison of effects of purine nu-cleotides on glutamate dehydrogenases from differentsources. Int. Cong. Biochem. 6:718.

11. Frieden, C. 1965. Glutamate dehydrogenase. VI. Surveyof purine nucleotides and other effects on the enzymefrom various sources. J. Biol. Chem. 240:2028-2035.

12. Hayflick, L. 1969. Fundamental biology of the classMollicutes, order Mycoplasmatales, p. 16. In L. Hay-flick (ed.), The Mycoplasmatales and the L-phase ofbacteria. Appleton-Century-Crofts, New York.

13. Jacobson, J. W., W. N. Strickland, and R. W. Barratt.1969. The amino acid composition of the subunit ofthe NADP-specific glutamate dehydrogenase fromNeurospora cmssa. Biochim. Biophys. Acta 188:283-286.

14. King, K. S., and C. Frieden. 1970. The purification andphysical properties of glutamate dehydrogenase fromrat liver. J. Biol. Chem. 245:4391-4396.

15. Landon, M., T. J. Langley, and Emil L. Smith. 1971.Sequence of bovine liver glutamate dehydrogenase. V.Sequence of a 44-residue peptide obtained by cyan-ogen bromide cleavage. J. Biol. Chem. 246:3802-3806.

16. Laycock, M. V., D. A. Thurman, and D. Boulter. 1964.An improved method for the detection of dehydrogen-ases using tetrazolium salts. Clin. Chim. Acta 11:98-100.

17. Lowry, 0. H., N. J. Rosebrough, A. L. Farr, and R. J.Randall. 1951. Protein measurement with the Folinphenol reagent. J. Biol. Chem. 193:265-275.

18. Shapiro, A. L., E. Vinuela, and J. V. Maisel, Jr. 1967.Molecular weight estimation of polypeptide chains by

502 J. BACTERIOL.

on April 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from

MYCOPLASMA GLUTAMATE DEHYDROGENASE

electrophoresis in SDS-polyacrylamide gels. Biochem.Biophys. Res. Commun. 28:815-820.

19. Shatilov, V. R., A. G. Evstigneeva, and V. L. Kretovich.1968. Glutamate dehydrogenase of Chlorella. Biokhi-miia 34:327-332.

20. Strickland, W. N. 1969. Induction of NAD-specific glu-

503

tamate dehydrogenase. Aust. J. Biol. Sci. 22:425-432.21. Struck, J., and I. W. Sizer. 1960. The substrate speci-

ficity of glutamate dehydrogenase. Arch. Biochem.Biophys. 86:260-266.

22. Yarrison, G., and G. L. Choules. 1972. Improved activitysedimentation method. Anal. Biochem., in press.

VOL. 110, 1972

on April 9, 2019 by guest

http://jb.asm.org/

Dow

nloaded from