10
Eur. J. Biochem. 101, 143-152 (1979) Structure and Properties of the Putrescine Carbamoyltransferase of Streptococcus faecalis Brigitte WARGNIES, Nadine LAUWERS, and Victor STALON Laboratoire de Microbiologie, Faculte des Sciences, Universite Libre de Bruxelles and Institut de Recherches du Centre d’Enseignement et de Recherches des Industries Alimentaires et Chimiques, Bruxelles (Received May 11, 3 979) Ornithine and putrescine carbamoyltransferases from Streptococcus fuecufis ATCC 11 700 have been purified and their structural properties compared. The molecular weight of native ornithine carbamoyltransferase, measured by molecular sieving, is 250 000. It is composed of six apparently identical subunits with a molecular weight of 39000, as determined by cross-linking with the bi- functional reagent glutaraldehyde followed by polyacrylamide gel electrophoresis in the presence of sodium dodecylsulfate. Using the same method, putrescine carbamoyltransferase is a trimer of 140000 consisting of three identical subunits with a molecular weight of 40000. Ornithine carbamoyltransferase displays a narrow specificity towards its substrate, ornithine. In contrast, putrescine carbamoyltransferase carbamoylates ornithine and several diamines (diamino- propane, diaminohexane, spermine, spermidine, cadaverine) in addition to its preferred substrate, putrescine, but with a considerably lower efficiency than for putrescine. The kinetic mechanism of putrescine carbamoyltransferase has been investigated. Initial velocity studies yield intersecting plots using either putrescine or ornithine as substrate, indicating a sequen- tial mechanism. The patterns of protection of the enzyme by the reactants during heat inactivation as well as the results of product and dead-end inhibition studies provide evidence for a random addition of the substrates. The putrescine inhibition that is induced by phosphate does, however, suggest that a preferred pathway exists in which carbamoylphosphate is the leading substrate. The different kinetic constants have been established. The properties of putrescine carbamoyltransferase are compared to the known properties of other carbamoyltransferases. The evolutionary implications of this comparison are discussed. The growth of Streptococcus fueculis with arginine as the energy source has been shown by Bauchop and Elsden [l]. The organism was capable of coupling the energy liberated from arginine degradation with the growth process, adenosine triphosphate being generated through the arginine deiminase pathway [2]. Agmatine is also used as an energy source [2]. Roon and Barker [3] have proposed that the catabolism of this compound follows a pathway similar to that described for the arginine catabolic pathway. It in- volves two steps catalyzed respectively by agmatine deiminase and putrescine carbamoyltransferase as shown below. Agmatine -+ carbamoyl- putrescine 3 putrescine NH3 Pi carbamoyl- phosphate ___ ~~ Enzymes. Ornithine carbamoyltransferase (EC 2.1.3.3) ; putrec- cine carbamoyltransferase (EC 2.1.3.6). Information concerning the quaternary structure and the kinetic mechanism of putrescine carbamoyl- transferase are of potential significance in the course of the study of the evolution of carbamoyltransferases in general. Indeed, the anabolic ornithine carbamoyltrdns- ferases as well as the catalytic subunit of aspartate carbamoyltransferase have been shown to share a common trimeric structure [4 - 61. These structural similarities have been taken as evidence for a common origin for ornithine and aspartate carbamoyltrans- ferases [7], a hypothesis which has received some sup- port from the comparison of the N-terminal amino acid sequences of the products of genes arg I and pyrB of Escherichiu cofi K-12 [8]. It was thus interesting to determine whether this hypothesis might also be extended to putrescine carbamoyltransferase. This work is the initial report of an attempt to define putrescine carbamoyltransferase in more detail. We describe the purification and the molecular charac- terization of putrescine carbamoyltransferase from

Structure and Properties of the Putrescine Carbamoyltransferase of Streptococcus faecalis

Embed Size (px)

Citation preview

Page 1: Structure and Properties of the Putrescine Carbamoyltransferase of Streptococcus faecalis

Eur. J . Biochem. 101, 143-152 (1979)

Structure and Properties of the Putrescine Carbamoyltransferase of Streptococcus faecalis

Brigitte WARGNIES, Nadine LAUWERS, and Victor STALON

Laboratoire de Microbiologie, Faculte des Sciences, Universite Libre de Bruxelles and Institut de Recherches du Centre d’Enseignement et de Recherches des Industries Alimentaires et Chimiques, Bruxelles

(Received May 11, 3 979)

Ornithine and putrescine carbamoyltransferases from Streptococcus fuecufis ATCC 11 700 have been purified and their structural properties compared. The molecular weight of native ornithine carbamoyltransferase, measured by molecular sieving, is 250 000. It is composed of six apparently identical subunits with a molecular weight of 39000, as determined by cross-linking with the bi- functional reagent glutaraldehyde followed by polyacrylamide gel electrophoresis in the presence of sodium dodecylsulfate. Using the same method, putrescine carbamoyltransferase is a trimer of 140000 consisting of three identical subunits with a molecular weight of 40000.

Ornithine carbamoyltransferase displays a narrow specificity towards its substrate, ornithine. In contrast, putrescine carbamoyltransferase carbamoylates ornithine and several diamines (diamino- propane, diaminohexane, spermine, spermidine, cadaverine) in addition to its preferred substrate, putrescine, but with a considerably lower efficiency than for putrescine.

The kinetic mechanism of putrescine carbamoyltransferase has been investigated. Initial velocity studies yield intersecting plots using either putrescine or ornithine as substrate, indicating a sequen- tial mechanism. The patterns of protection of the enzyme by the reactants during heat inactivation as well as the results of product and dead-end inhibition studies provide evidence for a random addition of the substrates. The putrescine inhibition that is induced by phosphate does, however, suggest that a preferred pathway exists in which carbamoylphosphate is the leading substrate. The different kinetic constants have been established.

The properties of putrescine carbamoyltransferase are compared to the known properties of other carbamoyltransferases. The evolutionary implications of this comparison are discussed.

The growth of Streptococcus fueculis with arginine as the energy source has been shown by Bauchop and Elsden [l]. The organism was capable of coupling the energy liberated from arginine degradation with the growth process, adenosine triphosphate being generated through the arginine deiminase pathway [2]. Agmatine is also used as an energy source [2]. Roon and Barker [ 3 ] have proposed that the catabolism of this compound follows a pathway similar to that described for the arginine catabolic pathway. It in- volves two steps catalyzed respectively by agmatine deiminase and putrescine carbamoyltransferase as shown below.

Agmatine -+ carbamoyl- putrescine 3 putrescine NH3 Pi carbamoyl-

phosphate

___ ~~

Enzymes. Ornithine carbamoyltransferase (EC 2.1.3.3) ; putrec- cine carbamoyltransferase (EC 2.1.3.6).

Information concerning the quaternary structure and the kinetic mechanism of putrescine carbamoyl- transferase are of potential significance in the course of the study of the evolution of carbamoyltransferases in general.

Indeed, the anabolic ornithine carbamoyltrdns- ferases as well as the catalytic subunit of aspartate carbamoyltransferase have been shown to share a common trimeric structure [4 - 61. These structural similarities have been taken as evidence for a common origin for ornithine and aspartate carbamoyltrans- ferases [7], a hypothesis which has received some sup- port from the comparison of the N-terminal amino acid sequences of the products of genes arg I and pyrB of Escherichiu cofi K-12 [8]. It was thus interesting to determine whether this hypothesis might also be extended to putrescine carbamoyltransferase.

This work is the initial report of an attempt to define putrescine carbamoyltransferase in more detail. We describe the purification and the molecular charac- terization of putrescine carbamoyltransferase from

Page 2: Structure and Properties of the Putrescine Carbamoyltransferase of Streptococcus faecalis

1 44 Str.fuecuh Putrescine Carbamoyltransferase

Str. faecalis. We report the results of a series of steady- state kinetic investigations in which the putrescine carbamoyltransferase reaction was followed in the reverse direction of its physiological function, the synthesis of carbamoylputrescine.

MATERIALS AND METHODS

Chemicals and Reagents

Amino acids and related compounds were ob- tained from Sigma Chemical Co., St Louis, Mo., U.S.A. except for propanediamine which was pur- chased from K and K Fine Chemicals, cadaverine from Mann Research Laboratories, 1,6-diamino- hexane from Fluka. L-Citrulline was purified according to the method of Rivard and Carter [9] to remove ornithine. ~-['~C]Carbamoylphosphate was obtained from New England Nuclear. The reagents for poly- acrylamide gel electrophoresis were purchased from Eastman Kodak. Sephadex G-200 Superfine and DEAE Sephadex A-50 were obtained from Pharmacia Fine Chemicals, A.B. (Uppsala, Sweden).

Enzymes used as standards were obtained from the following sources : albumin, Schwarz and Mann, glutamate dehydrogenase, aldolase and catalase for electrophoresis from Boehringer, glyceraldehyde-3- phosphate dehydrogenase, urease and hexokinase from Sigma, alkaline phosphatase and catalase for gel filtration from Worthington Biochemical Corpo- ration, myoglobin is a Seravac product and fumarase was a gift from Dr Kanarek.

Carbamoyltransferase Assays

Activity determination of putrescine carbamoyl- transferase was routinely performed by measuring the amount of carbamoylputrescine formed at 37 "C in a 2.0-ml reaction mixture containing (final concentra- tion) 50 mM Tris/HCl buffer, pH 7.0, putrescine and carbamoylphosphate 10 mM each. Lithium carba- moylphosphate solution in ice-cold water was pre- pared immediately before use and added last to initiate the reaction. After 10 min incubation the reac- tion was stopped by adding 2 ml 1 M HCI.

Ornithine carbamoyltransferase activity was deter- mined in the same conditions by measuring the citrul- line formed from ornithine and carbamoylphosphate. The buffer was 100 mM Tris/HCl, pH 8.0.

Carbamoylputrescine and citrulline were measured by the method of Archibald [lo]. When a still greater sensitivity was required, the radioactive [I4C]carba- moylphosphate assay described below has been used.

Putrescine Carbrcmoyltransjerase Assay for Kinetic Study

The assay used was essentially the ['4C]carbamoyl- phosphate assay described for Escherichia coli orni-

thine carbamoyltransferase [l 11. All incubations were performed for 5 min at 37 "C at pH 7.0. The 1 .O-ml reaction mixture contained 50 mM Tris/HCl with varying concentrations of substrate, inhibitors and putrescine carbamoyltransferase. The reaction was stopped by the addition of 1.0 ml 1 M HCl. After acidification, the test-tubes were capped and immersed in a 100 "C water bath for 10 min. The solution in the tubes was then bubbled with air for 5 min, and 1 ml of this solution was added to 10 ml scintillation cock- tail (4 parts of Triton X-100, 6 parts of a solution containing 0.5 % 2,5-diphenyloxazole and 10 % naph- thalene). Radioactivity was determined using a Beck- man LS100-C liquid scintillation spectrometer. Con- trols without enzyme were routinely run. Corrections were made for a minor radioactive component which is present with ['4C]carbamoylphosphate and is not converted to COZ after acidification and boiling. The initial velocity was linear up to about 30% substrate consumption but all experiments were performed below this range. The terminology and nomenclature for all the kinetic terms and constants are those defined by Cleland [12]. v is either expressed as pmol product formed during the time of experiment when the colorimetric assay was used or as the relative velocity when the ['4C]carbamoylphosphate assay was utilized.

All solutions of substrates, products and effectors were adjusted at the desired pH.

Protein Determination

Proteins were determined using the Folin-Ciocalteu reagent [13] and biuret [I41 methods. The Kalkar spectrophotometric method [15] was used for the last purification steps.

Polyacrylamide Gel Electrophoresis in the Presence of Sodium Dodecylsu[fate

Dissociated Enzymes. Electrophoresis was carried out in 7.5% acrylamide gels at pH 7.5 and 0.2% sodium dodecylsulfate as described by Weber and Osborn [16]. The samples were heated in the presence of 1 % sodium dodecylsulfate at 100°C for 10 min prior to electrophoresis. After staining and destaining the gels [16] the subunit molecular weights were esti- mated from a calibration curve obtained from simul- taneous electrophoresis run using as protein markers : albumin ( M , 68000), catalase ( M , 58 000), glutamate dehydrogenase ( M , 53 000), fumarase ( M , 47 500), al- dolase ( M , 40 000), glyceraldehyde-3-phosphate de- hydrogenase ( M , 36000) and myoglobin ( M , 17200).

Cross-Linking of the Subunits. The native enzyme (4- 10 mg/ml) was treated with various concentra- tions (0 - 0.05 %) cross-linking reagent, glutaralde- hyde, for 2 h at room temperature or with a single concentration (0.1 %) for different times (0 - 60 min).

Page 3: Structure and Properties of the Putrescine Carbamoyltransferase of Streptococcus faecalis

B. Wargnies, N. Lauwers, and V. Stalon 145

Table 1. Purification ofputrescine and ornithine carbamo.vltran.~~erases from Streptococcus faecalis 1 unit of transcarbamylase activity is defined as the amount of enzyme which catalyzes the formation of 1 pmol carbamoyl-product/h in the conditions described under Materials and Methods

Step Protein Putrescine carbamoyl- Ornithine carbamoyl- transferase activity transferase activity

1. Crude extract 2. Heat denaturation 3. Ammonium sulfate precipitation 4. DEAE-Sephadex: fractions 84- 114

5. Sephadex G-200 of pooled fractions 115-132

6. Second G-200 of the pooled fractions

7. Sephadex G-200 of pooled fractions 84- 114

8. Second G-200 of the pooled fractions

fractions 115-132

from DEAE-Sephadex

of ornithine transcarbamylase activity from step 5

from DEAE-Sephadex

of putrescine transcarbamylase activity from step 7

mg 9790 3020 2280

358 210

95

29

182

161

units

10 400 000 8 400 000 8150000 7 400 000

456 000

-

-

5 400 000

4950000

units/mg

1070 3 000 3 550

20600 2150

-

-

27 600

30500

~

units

1 7 400 000 13 900 000 13 265 000

800 600 9 650000

8 600 000

3350000

100000

42 500

units/mg

1780 4 600 5 800 2 250

46 000

90 000

116250

550

372

Amidination was stopped by incubation at 50 "C for 30 min in the presence of 0.2 % sodium dodecylsulfate and 10 "/, ethyleneglycol. Samples of this mixture (25 pg protein) were subject to electrophoresis in dodecylsulfate/polyacrylamide gels containing 5 % gel cross-linker as described above. Protein migration was estimated by manual measurements.

Polyacrylamide Gel Electrophoresis in Urea

The samples were heated in the presence of 8 M urea and 0.05 % 2-mercaptoethanol for 5 min at 100 "C, and then submitted to electrophoresis in 5 % acryl- amide gels at pH 8.0 in the presence of 8 M urea. The electrophoretic procedure has been described in details by Marshall and Cohen [17].

Determination of Molecular Weight by Gel Filtration

Molecular weight determinations were carried out as described previously [7] on Sephadex G-200 Super- fine columns. The protein markers for calibration, urease ( M , 480 000), catalase ( M I 230 000), hexokinase ( M , 78 000) and alkaline phosphatase (MI 48 000) were assayed following the method described ear- lier [7].

RESULTS

Purification of Ornithine and Putrescine Carbamoyltransferases

The purification of the two enzymes is shown in Table 1.

Step 1. Cell Growth and Extraction. Streptococcus faecalis ATCC 11 700 was grown in a medium con-

taining 0.5 % yeast extract, 1 % tryptone and 0.5% NaCl supplemented with 0.2 % agmatine. Previous experiments (unpublished) have demonstrated that enzymes of the agmatine deiminase pathway are in- duced after growth ceases. Cells from 32 1 culture were harvested in the stationary phase of growth after incubation at 37 "C for 22 h. The cells were collected by centrifugation in a Sharpless centrifuge and the pellet washed with a solution of 0.9 % NaCl. The cell paste was suspended in 100 mM potassium phosphate buffer (pH 6.0), supplemented with 10 mM putrescine, 1 mM dithiothreitol and disrupted for 30min in a Raytheon sonic oscillator at 10 kHz. The cell extract was centrifuged for 30 min at 20000 x g , the pellet resuspended in the extraction buffer and disrupted again. This operation was repeated three times. The supernatants of each extraction were pooled.

Step 2. Heat Denaturation. The cell extract was heated at 65 "C for 10 min, under gentle stirring in a water-bath. After cooling in an ice bath, the coagu- lated proteins were separated by centrifugation for 30 min at 20000 x g. The precipitate was washed with 50 mM potassium phosphate buffer, pH 7.5, supple- mented with 0.5 mM dithiothreitol, centrifuged, and the two supernatants were pooled. All the following steps were performed in the presence of 0.5 mM dithiothreitol.

Step 3. Ammonium Sulfate Precipitation. To the heat-denatured supernatant, solid ammonium sulfate was added slowly with constant stirring to 40 % satu- ration. After 30 min, the precipitate was removed by centrifugation at 20000 x g for 20 min. The super- natant was brought to 70% saturation by solid am- monium sulfate. The precipitate was collected by centrifugation and suspended in 50 mM potassium phosphate buffer pH 7.5 and dialyzed against the

Page 4: Structure and Properties of the Putrescine Carbamoyltransferase of Streptococcus faecalis

146

40-

3 0-

Str. fircwrlis Putrescine C.trbniiio! Itranyferase

50 100 l k 1 Fraction number

Hexokinase

E l

car barnovltransfersse ' 1 carbarnayltransferaie 2 5 0 0-

1 2 3 i i 6 7 8 9 l O zb 30 LO& Molecular weight lo1

Fig. 1 DEAi-.Sephadri A-511 ~hroniatoqrirphg of prorrin solution from srep 3. For elution conditions see Results. Fractions of 10 ml were collected at a How-rate of 40-50 ml'h Absorbance at 280 nm ( -); putrcscine carbarnoyltransferase activity (0); ornithine carhamo~ltranFferaie activity (0)

Fig 2 Gel /rlrrnrion on Sephadex G-200 Superfine of petre.rcme ~arhamo~lrr.onr/c.ru,~e /pooled actii'e frixtions from step 4 ) . For elution conditions. see Results. Absorbance at 280 nm (---), putrcscine carbamoyltrdniferaie aclivity (PTCase) (0 ) ; ornithine carhainoyitraniferaie activity (OTCase) (0)

Fig 3 Gel fdtratio,i on Sephadex G-ZOO Supwfine 41 mu51 uitive fractions of putrtwint, ~arhnmo~lrran~fe . feru~~ / r u m w p 7. For elution conditions, see Results. Ornithine carha- moyltranrferase activity (0). pntrescins carhamoyltrensferase activity (0)

same buffer at 300 mM concentration until the dis- appearance of ammonium sulfate.

Step 4 . DEAE-Sephadex A-50 Chromatography. The dialyzed protein solution was chromatographed on a column of DEAE-Sephadex A-50 (2.5 x 45 cm) which had been equilibrated with 300 mM potassium phosphate buffer, pH 7.5. The enzyme was eluted with a 1-1 linear gradient made of 500 ml 300 mM po- tassium phosphate buffer, pH 7.5, in the mixing chamber and 500 ml of the same buffer supplemented with 200 mM KC1 in the reservoir flask. Fractions of 10 ml were collected. The elution diagrams of putres- cine and ornithine carbamoyltransferase are illustrated in Fig. 1. The fractions corresponding to a particular peak were pooled and concentrated in an Amicon Dia Flow PM30 membrane to a final volume of 25 ml. These protein solutions were precipitated by addition of solid ammonium sulfate to 80% saturation. The precipitates collected by centrifugation were suspended in a minimal volume of 50 mM potassium phosphate buffer pH 7.5.

35 4a L5 3 Fraction number

I

Putrescine carbamoyltransferase Ornithine m r b a w r \

b-Glyceraldehyde -3- phosphate dehydrogenaie

-- L.--

LO 50 6 Mobility I cm 1

Fig.4 p H optimum of rhe oniirhme ~arhomo.vltran?fernse aLtzvrtg' (0 ) wpported hs the purified putrewine ~arhomogltransferasr en;yme, (0) of' the u'ell-clrorucreri=cd onirrkine furhamr?ylrrunsf~ras~. Ornithinc and carbumoylphosphate concentrattons are taken re- spectively a s 50 mM and 5 inM. The buffer used is Tns/HCI 50 rnM

Fig. 5 . Moleculur weifhz esiimafion of pvtresciiie curbam~rgltransferurse and ormrhinr ~urhamogltmn.sferase hg g d filtrution A Sephadex G-200 Superfine column ( 2 5 x 45 cm) equilibrated with 50 m M potassium phosphate, pH 7.5. w a s run with a sample of purified enzyme and the indicated 5tandards The elution volume is plotted verws the molecular weight of the protein

Fig. 6. Subunit mokcslar w i g h t determinrrtion (if putrescine curhamql.ltrans/rru~r and ornrthinr carhamoglrransferu.se. 25 pg purified enzyme and each of the indicated Ftanddrds were denatured as described in Materiels and Methods. and subjectcd to sodium dodecyl- sulfate'polyacryldmide gel electrophoresis Sor 4 h a t R mM. The molecular weight of the dcndtured protein? are plotted versus the mobility

Step 5. Molecular Sieving on Sephadex G-200 Super- fine of Ornithine Carhamoyltransjerase Pooled Frac- tions fronz Step 4 . The dissolved precipitate from step 4 (about 200 mg protein in 2.5 ml) was applied to a 2.5 x 45-cm Sephadex G-200 Superfine column prepared in 50mM potassium phosphate buffer pH 7.5. Fractions of 1.5 ml were collected. The most active fractions were pooled and concentrated by solid ammonium sulfate precipitation. They do not present any putrescine carbamoyltransferase activity.

Step 6. Second Gel Filtration on Sephadex G-200 of Ornithine Carhamoyltransjerase Fractions from Step 5. The dissolved precipitate from step 5 was chromatographed a second time on Sephadex G-200 under the same conditions. The most active fractions were collected and concentrated.

Step 7. Molecular Sieving on Sephadex G-200 Super- fine of the Putrescine Carbamoyltransjerase Pooled Fractions from Step 4. The dissolved precipitate con- taining putrescine carbamoyltransferase was sub- mitted to the same gel filtration as ornithine carba-

Page 5: Structure and Properties of the Putrescine Carbamoyltransferase of Streptococcus faecalis

B. Wargnies, N. Lauwers, and V. Stalon 147

moyltransferase. Fig. 2 shows that some ornithine carbamoyltransferase activity was present and was distributed between two peaks. One peak, correspond- ing to a higher molecular weight, was obviously due to a contamination by ornithine carbamoyltrans- ferase ; the second one was superimposed with the putrescine carbamoyltransferase activity peak. The most active fractions of putrescine carbamoyltrans- ferase were pooled and precipitated by solid am- monium sulfate.

Step 8. Evidence that Putrescine Curbamoyltrans- feruse Catalyzes the Carbamoylution of Ornithine. Attempts to eliminate the ornithine carbamoyltrans- ferase activity associated with the putrescine carba- moyltransferase activity peak by a second molecular sieving (Fig. 3) or on aminohexylsepharose 6B were unsuccessful. In consequence, ornithine carbamoyl- transferase activity appears to be an intrinsic property of putrescine carbamoyltransferase enzyme. A set of four arguments supports this view. (a) The ornithine carbamoyltransferase activity of the putrescine carba- moyltransferase enzyme exhibits a quite lower molec- ular weight during molecular sieving than the well characterized catabolic ornithine carbamoyltrans- ferase. (b) The pH optimum curves of the two ornithine carbamoyltransferase activities, supported by putrescine or by the ornithine carbamoyltransferase enzyme, are quite distinct (Fig.4). (c) Norvaline is a strong inhibitor of the catabolic ornithine carbamoyl- transferase enzyme, but moderately inhibits the putres- cine carbamoyltransferase activity or the ornithine carbamoyltransferase activity associated with it. (d) Table 2 demonstrates a broad substrate specificity of putrescine carbamoyltransferase enzyme which is able to carbamoylate a number of diamines in addition to ornithine. A few other metabolites (lysine, 2,4-di- aminobutyrate, b-alanine and ethylenediamine) are also carbamoylated by putrescine carbamoyltrans- ferase but a lower capacity than 1,3-diaminopropane. In contrast, ornithine carbamoyltransferase has ap- peared very specific for its substrates (data not shown).

Step 9. Purity of the Preparation. After polyacryl- amine gel electrophoresis under denaturing conditions (in the presence of sodium dodecylsulfate or urea) of 25 pg purified enzyme, we observed only one stained band for putrescine carbamoyltransferase and one for ornithine carbamoyltransferase.

Molecular Weight and Subunit Structure of Putrescine and Ornithine Carbamoyltransferases

Electrophoresis of the dissociated enzymes by sodium dodecylsulfate and by urea did not reveal more than one staining band. Both enzymes are thus formed by subunits with identical size and charge. The molecular weight of the native putrescine carba- moyltransferase determined by molecular sieving is

Table 2. Kinetic parameters for the alternative substrates f i r the putrescine carhamoyltran~~eruse enzyme Assays conditions were 50 mM Tris/HCI, pH 7.0, carbamoylphos- phate 1.25 x mM. Values of the apparent Michaelis constant and maximal velocity (V) were determined in a double-reciprocal plot of Lineweaver-Burk. V is expressed as a percentage of the V with putrescine as substrate

Substrate Apparent I' Michaelis constant

mM "/, Putrescine 0.029 100 Cadaverine 7.7 14.4 Ornithine 13.0 7.4 Spermidine 19.7 2.9 Spermine 0.35 0.05 1,6-Diaminohexane 6.6 0.04 1,3-Diaminopropane 2.9 0.015

140000 Jr 10000 (see Fig.5). By the same procedure ornithine carbamoyltransferase moves according to a molecular weight of 25QOOO 25000. The molecular weights of the subunits determined by sodium dodecyl- sulfate electrophoresis are 40000 k 1000 and 39000 Jr 1000 for putrescine carbamoyltransferase and or- nithine carbamoyltransferase respectively (Fig. 6). Treatment of the native enzyme with glutaraldehyde followed by electrophoresis in dodecylsulfate gel elec- trophoresis reveals three discrete bands for the putres- cine carbamoyltransferase and six for the ornithine carbamoyltransferase (Fig. 7). Assuming a molecular weight of 40000 for the subunit of putrescine carba- moyltransferase, the molecular weight of the trimer should be 120000, a value which is in good agreement with that obtained by molecular sieving. The molec- ular weight for an hexameric structure for ornithine carbamoyltransferase (234000) is also in accord with that measured by gel filtration.

General Properties of the Putrescine Curbamoyltrun.ferase Reaction

This enzyme requires only putrescine and carba- moylphosphate for its activity. The rate of carbamoyl- putrescine formation is constant for at least 30 min at 37°C. The amount of carbamoylputrescine pro- duced in 6 min increases in direct proportion to the amount of enzyme present. Roon and Barker [3] re- ported for the partially purified enzyme a fairly broad pH optimum ranging from pH 7.0-9.0. Our pure enzyme shows the same pH dependence at high sub- strate concentration (see Fig. S) but when the putres- cine concentration is lowered from 10 mM to 0.5 mM, the enzyme shows a bell-shaped curve whose pH optimum for activity is 9.0. Lowering the carbamoyl-

Page 6: Structure and Properties of the Putrescine Carbamoyltransferase of Streptococcus faecalis

148 Str.faecalis Putrescine Carbamoyltransferase

I

L I 30 LO 5n 60 I Mobility Icm I

+- 10 lCarbamoylphasphatel-l I mM I-’

8

5

50

1 IV

2:

I1

6 0 7 0 80 90 10 PH

I B I / I

Fig. 7 Soelrunt d~,decyl~ulfare~pol~ac~ylamide gel electrophoresi.~ o/ f A J ornithrne carbamoyl- wansferase and ( B ) putrescine carhamoylframferase c r o d i n k e d with glutaraldehyde. Molecular weight is plotted versus mobility

Fig. 8. p H dependence of the putrescine carhnmoyltransferose reacfron. ( 0 ) The concen- trations In carhamoylphosphate and putrescine are taken equal to 10 mM. (0) The con- centrations in carbamoylphosphate and putrescine are equal to 0.2 mM (0) Putrescine concentration 1s taken equal to 0.5 mM and carhamoylphosphate concentration to 10 mM

Fig 9. Init id velocity studlrr of the carbamoylarion reaction of purrescine by putrescine rarhamoyltrunferuse Carhamoylpbosphate IS the variable substrate Putrescine concen- trations are shown

phosphate concentration gives rise to a curve whose pH optimum is now 7.8. The diversity found in the shape of the three curves results from substrate inhibi- tion by putrescine concentration (data not shown).

Ej’eects of Reactants toward Heat Inactivation

The results of Table 3 show evidence for the binding to the enzyme of carbamoylphosphate, phos- phate (and its analog, pyrophosphate), ornithine and related compounds (2,4-diaminobutyrate, norvaline), since all these compounds provide an appreciable protection against heat denaturation.

Putrescine, a physiological reactant of the enzyme, is an extremely poor protector. This negative result does not, however, exclude its binding to the free enzyme.

The simultaneous presence of putrescine and phos- phate, or of ornithine and phosphate yielded degrees of protection which were significantly higher than those offered by each effector alone. These results

I Putrescinel 10’1mM i

200 400 600 800

I Carbamoylphosphate 1-l 1 mM I-’ 9

Fig. 10. Initial velocitv studies of the carbamoylation reaction of ornirhine by purrescine carbamoyltran~ferase. Carbamoylphosphale is the variable substrate. Ornithine concen- trations are shown

Fig. 11. Dead-end inhibition by spermrdine of the carbomoylurion reacrion of purrescine. (A) Reciprocal reverse velocity versus reciprocal putrescine concentration. Carhamoyl- phosphate concentration was chosen equal to 1.25 pM. (B) Replot of slope with respect to spermidme concentration

Fig 12. Dead-end inhibition by spsrmidine a/ the carhamoylatron reacmn of purrescine (A) Reciprocal reverse velocity versus reciprocal carhamoylphosphate concentration. Putrescine concentration was chosen equal to 0.02 mM (B) Replot of slope (0) and inter- cept (0) with respect to spermidine concentration

suggest the formation of ternary complexes enzyme . phosphate. putrescine and enzyme. phosphate. orni- thine.

That these effects were indeed specific of these agents and were not caused by the high ionic concen- trations used was established by showing that several salts, such as for instance KCl, at the concentration of 100mM did not provide any significant pro- tecting effect.

The fact that ornithine, carbamoylphosphate and phosphate can form binary complexes with the enzyme is in agreement with random binding of the reactants. But the distinction between random and ordered mechanisms does not depend only on whether en- zyme . carbamoylphosphate or enzyme . putrescine complexes form, it also depends on whether such com- plexes are productive or not.

Kinetic Studies

Kinetic studies were undertaken in order to deter- mine the order of substrate addition on putrescine

Page 7: Structure and Properties of the Putrescine Carbamoyltransferase of Streptococcus faecalis

B. Wargnies, N. Lauwers, and V. Stalon 149

Table 3, Effect of reactants against heat inacdvation of Strepto- coccus faecalis putrescine carbamoyltransferase 1 ml mixture containing 50 mM Tris/HCI, pH 7.0, reactants and enzyme at desired concentrations was incubated for 10 min at 55 “C. After 10 min the tubes were removed and chilled rapidly. Enzyme assays were initiated by adding 1 ml solution of 20 mM putrescine, 20 mM carbamoylphosphate and 20 mM Tris/HCl buffer. Each point is triplicated and compared with samples not treated

Reactants Final Activity concentration

mM % of original

None KCI Putrescine

Ornithine

2,4-Diaminobutyrate

Norvaline Phosphate

Pyrophosphate

Carbamoylphosphate Phosphate

Phosphate + putrescine

+ ornithine

100 10

100 1

10 1

10 10 10

100 1

10 10 10 10 10 1

0 < 0.1 < 0.1

3.8 6.0

88 65

100 97.5 2.6

73 87 91

100 48

33.5

carbamoyltransferase and to compare this enzyme to the other carbamoyltransferase already studied [I 1, 18 - 231. The non-physiological direction of the reac- tion has been measured. Putrescine and carbamoyl- phosphate are consequently the substrates and carba- moylputrescine and phosphate the products. These experiments were conducted with putrescine but also with ornithine as substrate.

Fig.9 and 10 show the results of studies of initial velocities for the carbamoylation of putrescine and ornithine by putrescine carbamoyltransferase.

Converging patterns were obtained when carba- moylphosphate was varied at different fixed concen- trations of putrescine or ornithine, indicating, ac- cording to Cleland analysis, that carbamoylphosphate and putrescine (or ornithine) are bound to enzyme forms that are reversibly connected. It follows that the reaction proceeds by a sequential mechanism so that both substrates must be added to the enzyme before any product is released. The kinetic constants for the interaction of substrates with the enzyme are given in Table 4.

In order to determine whether the sequence of substrate addition is ordered or random, initial velo- cities were measured in the presence of dead-end inhibitors.

Table 4. Kinetic constants for the interaction of substrates with pu- trescine carbamoyltrun,sfi.rase The values of the parameters for substrates were obtained by fitting initial velocity data such as that of Fig.9 and 10

Substrates Kinetic Values Reaction constant

mM

Carbamoyl- K,A (1.5 f 0.5) E + A E + A phosphate (A) K , A (1.54 f 0.2)

KA (2.0 0.5) lo-’” EB + A Kli 0.21 f 0.05b EB’+ A

Putrescine (B) K, B (1.5 f 0.5) lo-’ E + B KB 0.17 f 0.03 EA + B

K:B 0.58 f 0.08 E + B’ KA 6.1 f 2.0 EA + B’

Ornithme (B‘)

a Values for carbamoylphosphate obtained with putrescine as

Values for carbamoylphosphate obtained with ornithine as substrate.

substrate.

Spermidine, an analog of putrescine, is a non- competitive inhibitor with respect to carbamoylphos- phate (Fig. 11 and 12). Replots of slopes and intercepts versus inhibitor concentrations were linear. The ap- parent inhibition constants derived from these replots are given in Table 5 .

Similarly, pyrophosphate, phosphate and arsenate, competitive inhibitors of carbamoylphosphate (not shown), are non-competitive inhibitors in relation to putrescine (Fig. 13). The apparent constants for the inhibition by phosphate were also indicated in Table 5. Since no uncompetitive inhibition with respect to both substrates has been observed in the above experi- ments, the results only agree with a random addition of both putrescine and carbamoylphosphate. In addi- tion, these results suggest that the enzyme bears distinct subsites for carbamoylphosphate and putres- cine, so that enzyme . phosphate . putrescine and enzyme . carbamoylphosphate . spermidine dead-end complexes can form.

Additional evidence for the kinetic mechanism, however, results from the observation that arsenate, pyrophosphate and phosphate clearly induce putres- cine inhibition (Fig. 13). Fig. 14 shows the results of an experiment in which the concentration of carba- moylphosphate is varied in the presence of 50 mM phosphate and at different inhibitory concentrations of putrescine. Putrescine inhibition is partial and competitive with respect to carbamoylphosphate. Ac- cording to Cleland analysis, this pattern is character- istic of a random mechanism with one preferential pathway, which, for our reaction, is the one where carbamoylphosphate is the leading substrate (see Discussion).

Page 8: Structure and Properties of the Putrescine Carbamoyltransferase of Streptococcus faecalis

150 Str.faecali,r Putrescine Carbamoyltransferase

Table 5. Inhibition of putrescine carhamoyltransferase by products and substrate analogues Values for the apparent inhibition constant were obtained from slope or intercept replots corresponding to Eqns (2) and (3) according to whether the inhibition was competitive or non-competitive. Values given in parentheses are apparent inhibition constants which were calculated for comparison with the directly determined value for the apparent Ki. Calculations were performed assuming a rapid-equilibrium, random reaction mechanism using the true dissociation constant, obtained from the non-competitive inhibition with respect to the other substrate and the fixed concentration of the non-varied substrate. The true values for the product inhibition and dead-end inhibition constants were determined from the apparent values using the concentration of the non-varied Substrate and the appropriate relationships given by Morrisson [25]

Inhibitor Substrate Apparent Ki Values Reaction ~ __________ ________~ ~~ ~ ~ _ _ _ _ _ _ _ ~

varied fixed slope intercept

mM mM

Phosphate A B (0.01 mM) 6 k l (5 .4) A B' (0.5 mM) 12 k l(12.1) B A (1.25 pM) 8.5 0.5 6.8 k 0.4 K, = 4.6 k 0.8 E + I

B' A (1.25 pM) 13.5 1.5 45 * 2 Ki = 1.4 1.3 E + I

Spermidine A B (2PM) 4.5 0.6 3.8 1.0 Ki = 1.9 k 0.6 E + I

KI = 6.4 & 0.4 EB + I

Kl = 42.3 k 4.4 EB' + I

KI = 3.4 * 1.0 EB + I B A (1.25 pM) 2.5 k 0.5 (2.45)

If'hosphatel ImMl l 0 A

65 14 ICarbamoylphosphateI-' ;:b! I-'

Fig 13 Prodw I iiihihiiion by phosphate of thr carbamnjdation reaction o/ putrrxine. Reciprocal rcverse velociry versuy reciprocal of putrescine cvnccntration. Carbamoyl- phorphalr concentralion was chosen equal to 1.25 pM. Phasphatc concentrations are chown

Fig. 14. Phospholr-inducedpulrr~~mr rnhihirion 01 lhc curhamoylalron reartinn ofpulrescine. (A) Reciprocal reverse velocity with respect to reciprocal carbdmoylphoiphate concen- tmtioii. Phosphate concentration was taken equal to SO mV. Ornithine concentralions are shown (B) Replot of slopeb with respecl to putrescinc concentration

Dead-end inhibitions similar to those found with putrescine as substrate have also been observed when ornithine is used. Nevertheless, phosphate and its analogs are unable to induce ornithine inhibition in the range (up to 100 mM).

DISCUSSION

The procedure for putrescine carbamoyltransferase purification presented here results in a 30-fold purifica- tion with a satisfactory yield and a final purity of 98 % (see Table 1). All the steps of this purification scheme have been found reproducible. Ornithine carbamoyl- transferase has been obtained as a by-product of putrescine carbamoyltransferase purification. The two enzymes were separated by molecular sieving. Indeed the apparent molecular weights of the native enzymes are 140 000 for the putrescine carbamoyltransferase and 250 000 for the ornithine carbamoyltransferase. Polyacrylamide gel electrophoresis following disso- ciation is the presence of sodium dodecylsulfate or urea shows that both enzymes consist of identical subunits. The subunit molecular weights of putrescine and ornithine carbamoyltransferaser; are 40 000 and 39 000 respectively. Electrophoresis in the presence of sodium dodecylsulfate after amidination of the pro- tein with the bifunctional cross-linking reagent, glu- taraldehyde, indicates a trimeric structure for putres- cine carbamoyltransferase and a hexameric one for ornithine carbamoyltransferase. Our results concern- ing ornithine carbamoyltransferase from Streptococcus faecalis ATCC 11 700 are in agreement with those ob- tained by Marshall and Cohen [17] for the same enzyme from strain ATCC 11420.

A kinetic study of putrescine carbamoyltransferase has been performed using the conventional method of determination of the order of addition of substrate [12,23]. This method, which depends on the examina- tion of the inhibition patterns produced by the prod- ucts and analogs of the products or substrates, has

Page 9: Structure and Properties of the Putrescine Carbamoyltransferase of Streptococcus faecalis

B. Wargnies, N. Lauwers, and V. Stalon 151

been applied to both putrescine and ornithine as sub- strates.

The results from initial velocity studies with orni- thine as substrate, and those from dead-end inhibition studies, are compatible with a random addition of the substrates. This is also supported by the thermo- inactivation studies which show that both ornithine and carbamoylphosphate protect the enzyme against heat inactivation. It is of interest that significantly different maximum velocities have been observed for the reactions with putrescine or ornithine as substrate. This indicates that they cannot have a common rate- limiting step and excludes the possibility that both reactions occur by means of an ‘ordered mechanism’ in which the release of phosphate is the slowest step. If the mechanism appears random when ornithine is the substrate, it is only approximately so for putrescine. Indeed phosphate induces a putrescine inhibition which is partial and competitive with respect to carbamoylphosphate. This situation is quite analogous to that found for the ornithine carbamoyltransferases studied previously [20,21], where phosphate induced ornithine inhibition. Such a situation, according to Dannenberg and Cleland [24], could be explained by a random mechanism where the majority of the reaction flux, when both substrates are present at K, levels, goes through the path where carbamoylphos- phate adds before putrescine. The formation of the ternary complex enzyme . phosphate . putrescine would turn on an alternative reaction pathway, since this complex can break down to form the binary complex enzyme . putrescine, which can in turn react with carbamoylphosphate to generate the central ternary complex. Substrate inhibition is thus partial and competitive with respect to carbamoylphosphate, since this substrate and phosphate compete for the same enzyme . putrescine form. If consequently ap- pears that the mechanism is random but cannot be of the rapid equilibrium type. Nevertheless, we have neglected the phosphate-induced putrescine inhibition and have applied the Morrisson equations [25] for random rapid equilibrium mechanisms to the deriva- tion of the kinetic parameters of the reaction catalyzed by putrescine carbamoyltransferase. The equation has been used in its reciprocal form:

+ 5) [BI

(3)

where v is the initial velocity, V the maximal velocity, [A] and [B] represent concentrations of carbamoyl- phosphate and putrescine (or ornithine) respectively, the K terms their Michaelis constant and K, terms are dissociation constants for the reaction of the sub- strate with the free enzyme. Inhibition by an analog (I) of A, competitive versus substrate A and non- competitive with respect to B is described by Eqns (2) and (3). The same forms of Eqns (2) and (3), but with the substitution of A by B and vice-versa, describe the effect of a B analog. Ki and KI are also dissociation constants for the reaction of I with the free enzyme and the enzyme . substrate respectively.

The kinetic constants for putrescine carbamoyl- transferase were calculated using the equilibrium assumption, with either putrescine or ornithine as substrate. This mechanism does not describe the over- all process (since it neglects the phosphate-induced inhibition by putrescine) ; nevertheless, the analysis of the data of Tables 4 and 5 demonstrates that they are in quantitative agreement with this random model.

It is interesting to note that identical dissociation constants for the reaction of free enzyme with carba- moylphosphate were obtained whether putrescine or ornithine were used as substrate. The data also in- dicate that the binding of one substrate hinders the binding of the other. For instance the affinity of the enzyme towards carbamoylphosphate is considerably reduced by the presence of ornithine, and to a lesser degree of putrescine, on the enzyme. Similarly, the affinity towards phosphate is diminished by the bind- ing of ornithine to the enzyme; in contrast, this affinity is not affected by the binding of putrescine to the enzyme.

Consequently, this study allows us to conclude that the kinetic mechanism for putrescine carbamoyl- transferase of Str. fueculis is of the random type with a preferred pathway in which carbamoylphosphate is the leading substrate. Similar mechanisms have already been established for other carbamoyltransferases, namely the anabolic ornithine carbamoyltransferase of Escherichiu coli (at pH 6.8) [21] and of S. cerevisiue [20] and the catabolic ornithine carbamoyltransferase of Pseudomonus putidu [21]. Other carbamoyltrans- ferases, such as the catabolic ornithine carbamoyl- transferase of Str. fueculis [18], the anabolic ornithine carbamoyltransferase of P. putidu [19] and of E. coli (at pH 8) [ l l ] and aspartate carbamoyltransferase of E. coli [22] and of Str. fueculis [23], follow ordered addition of the substrates. Inasmuch as an ordered mechanism is an extream case of a random mechanism

Page 10: Structure and Properties of the Putrescine Carbamoyltransferase of Streptococcus faecalis

152 B. Wargnies, N

with a preferred pathway, it consequently appears that the mechanisms of these various carbamoyltransferases are related. This conclusion is supported by the ob- servation that each time an ordered mechanism is operative the leading substrate is carbamoylphosphate.

Putrescine carbamoyltransferase displays an ex- tremely broad specificity, a feature that has not been reported for the other carbamoyltransferases investi- gated so far. This enzyme, besides its ability to carbamoylate a number of substrates in addition to putrescine, is subject to inhibition by a number of compounds which bear no obvious structural analogies with its substrates or products (Wargnies and Stalon, unpublished observations). Several of these com- pounds yield parabolic inhibition curves. No clear explanation of the action of these effectors is presently available. The understanding of this behaviour was, however, not a prerequisite for the determination of putrescine carbamoyltransferase kinetic mechanism.

The findings reported in this paper concerning the structural and catalytic properties of putrescine carba- moyltransferase are interesting in relation to those of the other carbamoyltransferases. The quaternary struc- tures of a number of these enzymes have been estab- lished [4]. Trimeric structures have been observed for most of the anabolic ornithine carbamoyltransferases and catalytic subunits of aspartate carbamoyltrans- ferases so far studied. The molecular weights of these enzymes usually range between 100 000 and 150 000. On the other hand the catabolic ornithine carbamoyl- transferases have higher molecular weights and more elaborate quaternary structures. Nevertheless, there are some exceptions, both among anabolic enzymes : ornithine carbamoyltransferase from Bacillus subtilis has a molecular weight of 280000 [27], and among catabolic enzymes : ornithine carbamoyltransferase of B. lichenijormis is a trimer. Putrescine carbamoyl- transferase appears as a new exception to this rule and suggests that a clear-cut classification in catabolic and anabolic quaternary structures is not tenable.

The subunits of catabolic and anabolic carbamoyl- transferases have similar molecular weights. Yet, there has so far been no indication as to whether these sub- units have a common origin. The study of putrescine carbamoyltransferase might well be an important element in this line. The observation that putrescine carbamoyltransferase exhibits ornithine carbamoyl- transferase activity is interesting with regard to the current ideas on the evolution of carbamoyltrans- ferases. The recently observed N-terminal similarities between ornithine carbamoyltransferase and the cata- lytic subunit of aspartate carbamoyltransferase of E. coli strengthens the suggestion that these two en- zymes share a common origin, having evolved from tandem duplication of the same ancestral gene [S]. Putrescine carbamoyltransferase of Str. faecalis, which

Lauwers, and V. Stalon: Str. fuecatis Putrescine Carbamoyltransferase

harbours a certain degree of ambiguity towards its substrates, might represent a poorly differentiated form arising from a similar event.

A detailed immunological study of carbamoyl- transferases, now in progress, should yield complemen- tary information on the evolutionary development of these enzymes.

This work was supported by contract no. 2.4542/75 from the Fonds de la Recherche Fondamentale Collective. B. Wargnies is recipient of a fellowship from the Institut pour /'Encouragement de la Recherche Scientijique duns l'lndustrie et /'Agriculture and V. Stalon is chercheur qualifiP at the Fonds National de la Recherche Scientifique. We are grateful to J. P. Simon for lively discussions and we thank Professors F. Hilger and A. Pierard for help in the preparation of the manuscript.

REFERENCES 1.

2. 3. 4.

5.

6.

7.

8.

9.

10. 11. 12. 13.

14.

15. 3 6.

17.

18.

19.

20.

21.

22.

23.

24.

25. 26. 27.

Bauchop, T. & Elsden, S. R. (1960) J . 'Cen. Microbiol. 23.

Deibel, R. H. (1964) J . Bacteriol. 87, 988--992. Roon, R. J. & Barker, H. A. (1972) J . Bacteriol. 109, 44-50. Legrain, C., Stalon, V., Noullez, J. P., Mercenier, A,, Simon,

J . P., Broman, K. & Wiame, J. M. (1977) Eur. J . Biochem.

Rosenbush, J. P. & Weber, K. (1971) J . Bid. Chem. 246,

Brabson, J. S. & Swizer, R. L. (1975) J . Bid . Chem. 250, 8664 - 8669.

Legrain, C., Halleux, P., Stalon, V. & Glansdorff, N. (1972) Eur. J . Biochem. 27, 93 - 102.

Gigot, D., Glansdorff, N., Legrain, C., Pikrard, A,, Stalon, V., Konigsberg, W., Caplier, I., Strosberg, A. D. & Hervk, G. (1977) FEBS Lett. 81, 28-32.

Rivard, D. E. & Carter, H. E. (1955) J . Am. Chem. Soc. 77,

Archibald, R. M. (1944) J . B id . Chem. 156, 121-130. Legrain, C. & Stalon, V. (1976) Eur. J . Biochem. 63, 289-301. Cleland, W. W. (1963) Biochim. Biophys. Acta, 67, 104-137. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J.

Gornall, A. G., Bardawill, C. S. Sr David, M. M. (1949) J .

Kalkar, H. M. (1947) J . Bid . Chem. 167,461 -470. Weber, K. & Osborn, M. (1969) J . Biol. Clzem. 244, 4406-

Marshall, M. & Cohen, P. P. (1972) J . Biol. Chem. 247, 1641 -

Marshall, M. & Cohen, P. P. (1972) J . Biol. Chem. 247, 1654-

Stalon, V., Legrain, C. & Wiame, J. M. (1977) Eur. J . Bio-

Simon, J. P. & Stalon, V. (1977) Eur. J . Biochem. 75, 571-

Wargnies, B., Legrain, C. & Stalon, V. (1978) Eur. J . Biochem.

Porter, R. W., Mobede, M. 0. & Stark, G. R. (1969) J . Bid.

Chang, T. Y. Sr Jones, M. E. (1974) Biol,hernistry, 13, 638-

Dannenberg, K . D. Sr Cleland, W. W. (1975) Biochemistry, 14,

Morrison, J. F. &James, E. (1965) Biochem. J . 97, 37-52. Chang, T . Y . & Jones, M. E. (1974) Biochemistry, 13.629-638. Issaly, I. M. & Issaly, A. S. (1974) Eur. J . Biochem. 4Y, 485-

457 - 469.

80, 401 -409.

1644-1657.

1260 - 1261.

(1951) J . Bid. Chem. 193, 265-275.

Biol. Chem. 177, 751 - 766.

4412.

1653.

1668.

chem. 74, 319-327.

581.

89, 203-212.

Chem. 244, 1846-1859.

645.

28 - 39.

495.

B. Wargnies, N. Lauwers, and V. Stalon, Laboratoire de Microbiologie. Faculte des Sciences de I'Universite Libre de Bruxelles, Avenue Emile-Gryzon 1, B-1070 Bruxelles, Belgium