Eur. J. Biochem. 74, 319-327 (1977)

Anabolic Ornithine Carbamoyltransferase of Pseudomonas The Bases of Its Functional Specialization

Victor STALON, Christiane LEGRAIN, and Jean-Marie WlAME

Institut de Recherches d u Centre d’Enseignernent et de Recherches des Industries Alimentaires et Chirniques, Bruxelles and Laboratoire de Microbiologie, Universite Libre de Bruxelles

(Received September 28, 1976)

The anabolic ornithine carbamoyltransferase of Pseudomonas appears to be extremely specialized. Unlike the other carbamoyltransferases studied, this enzyme catalyzes the phosphorolytic cleavage of citrulline with a very poor efficiency.

The main goal of this paper is to understand what, in the catalytic process, causes this directed functional specialization. On the basis of kinetic data and thermodynamic properties of the reaction, it appears that the reaction mechanism is the same as for ornithine carbamoyltransferases from other sources, that is, of the sequential ordered type, where carbamoylphosphate is the first substrate to be bound and phosphate the last product to be released. In addition to this, and here lies the difference with other ornithine carbamoyltransferases, the anabolic transferase of Pseudomonas forms a binary dead-end complex with citrulline, leading to inefficient binding of phosphate and citrul- line to the enzyme. Therefore the phosphorolytic cleavage of citrulline is equally inefficient. It should be mentioned that the affinity of the enzyme for citrulline at its catalytic site is low as compared to other transferases.

Reaction (I) is catalyzed by ornithine carbamoyl- transferase.

Ornithine + carbamoylphosphate G Citrulline + phosphate. (1)

Thermodynamically the reaction is much in favour of the synthesis of citrulline [l]. The forward reaction is a universal step in the biosynthesis of arginine. The backward reaction is used in a number of micro- organisms which degrade arginine by the deiminase pathway. The efficient use of the backward reaction results from the thermodynamically favoured trans- formation of carbamoylphosphate into ATP catalyzed by carbamate kinase. Indeed coupling with carbamate kinase in vitro or the use of arsenate instead of phosphate enables the decarbamoylation of citrulline into ornithine with a number of ornithine carbamoyl- transferases, including ornithine carbamoyltrans- ferases with an obvious anabolic function. So it is well established that ornithine carbamoyltransferase may catalyze both directions of reaction I.

So far, in the few cases in which the same wild-type organism is able to use both directions of reaction I, two distinct ornithine carbamo yltransferases have been detected and their anabolic or catabolic function has

Enzyme. Ornithine carbamoyltransferase (EC 2.1.3.3)

been established at least on the basis of the regulation of their synthesis. This is the case for Pseudomonas [ 2 ] and as shown more recently for Bacillus Iicheni- formis [ 3 ] .

Also, it was observed that the two ornithine carbamoyltransferases of Pseudomonas offer regula- tory peculiarities. The catabolic ornithine carbamoyl- transferase which could catalyze the synthesis of citrulline in vitro was unable to do this in vivo despite the presence of the enzyme when grown on minimal medium (without inducer). This was explained by a very strong cooperativity for carbamoylphosphate [4]. The enzyme remains in its inactive form when carbamoylphosphate is present at cellular concentra- tion. Indeed mutations which reduce the cooperativ- ity of the catabolic enzyme allow the enzyme to fulfil an anabolic function. In other words unidirectional functioning of the enzyme results from the enzyme being in an inactive allosteric state when the condi- tions of growth require biosynthesis. When catabolic conditions appear effectors promote the transition into the active state [4].

The anabolic ornithine carbamoyltransferase is peculiar because, in contrast to other ornithine carba- moyltransferases with anabolic or catabolic functions, it shows a much reduced capacity for decarbamoyla- tion of citrulline and so it is a striking example of a

320 Pseudomonas Ornithine Carbamoyltransferase

directed, functionally specialized catalyst [5]. This paper analyses how such a kinetic property can be obtained despite a constant thermodynamic situation and in agreement with the principle of microscopic reversibility .

The need for a kinetic analysis implies the require- ment for information about the kinetic mechanism which is involved. However, the analysis of the mechanism is made difficult precisely because of the very weak backward reaction. The detailed kinetic analysis of the Escherichia coli ornithine carbamoyl- transferase where the microscopic events have been determined [7] will help to understand the peculiarities of the anabolic Pseudomonas enzyme.

EXPERIMENTAL PROCEDURE

Chemicals

Amino acids and related compounds were obtained from Sigma Chemical Co, except for homoarginine, guanidobutyrate and guanidopropionate which were purchased from Calbiochem and agmatine from ICN-K and K Laboratories. 1,-Citrulline was purified according to the method of Rivard and Carter [S] to eliminate ornithine from citrulline. ~ - [ c a r b a m o y l - ~ ~ C ] - Citrulline was obtained from the Radiochemical Center Amersham and ['4C]carbamoylphosphate from New England Nuclear.

All other chemicals were of reagent grade. Phos- phonoacetate was gently supplied by M. Penninckx.

Bacteriql Strain and Culture Medium

The "strain IRC204 has been described before [6]. Growth was carried out at 30 "C under conditions described previously 191 except that the cells were harvested in the exponential phase of growth.

Enzyme Preparation

Partial purification of the anabolic ornithine carbamoyltransferase was carried out as described previously [9] except that 10 14 ethylene glycol was added to all buffer solutions and that the enzyme preparation obtained after gel filtration was submitted to a second chromatography on DEAE-Sephadex A-50. This chromatography was achieved under the conditions described for the first chromatography except that the enzyme was eluted with a 1-1 linear gradient of 100-200 mM potassium phosphate buffer pH 7.5.

The specific activity of this enzyme preparation was 3500 units/mg protein. (One unit of ornithine carba- moyltransferase activity is defined as the amount of enzyme which catalyzes the formation of 1 pmol citrulline per h.) Such a preparation is devoid of any

interfering catabolic ornithine carbamoyltransferase activity.

Determination of Initial Velocity

Activity determination for the kinetic studies was carried out as described before [7] except that the incubation temperature was 30 "C instead of 37 "C.

Data Analysis

The nomenclature used in this study is that of Cleland [lo]. A, B, P, Q respectively represent carbamoylphosphate, ornithine, citrulline and phos- phate. I is an inhibitor other than substrates and products. Ki, and Kiq are the dissociation constants for carbamoylphosphate and phosphate. Kib and Kip are kinetic constants respectively associated with ornithine and citrulline. K,, Kb, Kp, Kq are the Michaelis constants for carbamoylphosphate, orni- thine, citrulline and phosphate respectively. K I ~ re- presents the dissociation constant for the citrulline- enzyme abortive complex and Ki terms, the inhibition constant for inhibitors other than substrates and products.

Determination of Equilibrium Constant

The equilibrium constant of the reaction catalyzed by ornithine carbamoyltransferase was determined by measuring the amount of ornithine formed from citrulline and phosphate in the presence of E. coli W ornithine carbamoyltransferase as catalyst. The ex- perimental conditions are as follows: a reaction mix- ture containing 100 mM potassium phosphate buffer and 100 mM citrulline adjusted to the desired pH was preincubated for 5 min at 30 "C or 37 "C. The reac- tion was started by adding the enzyme (about 1 mg). 1 .O-ml aliquots were removed after different incubation times and immediately transferred into test tubes containing 1 .O ml of a solution of 6 trichloroacetic acid and centrifuged to eliminate the precipitated proteins. The amount of ornithine formed after different times was determined with a Beckman 1200 amino acid analyzer. In order to calculate the equilib- rium constant using these results it was necessary to take into account the disappearance by non-enzymic hydrolysis, of part of the carbamoylphosphate formed. The effect of this hydrolysis is to displace the equilibrium in favour of the formation of orni- thine. The half-lifetime of carbamoylphosphate was determined for the conditions of pH and of tempera- ture used in our experiment. The calculation of the equilibrium constant at different pH values was done on an IBM computer with a program for numerical simulation.

V. Stalon, C. Legrain, and J.-M. Wiame 32 1

RESULTS AND DISCUSSION

Determination of the Equilibrium Constant

As already shown by Reichard [ l l ] and Marshall and Cohen [l] the equilibrium of the reaction strongly favours citrulline synthesis. The apparent equilibrium constant is defined by the following relation

[citrulline] [phosphate] [ornithine] [carbamoylphosphate] Kapp =

From the studies on the Streptococcusfaecalis enzyme [l] and E. coli enzyme [7] it is obvious that what should appear in the expression of the equilibrium constant are the ionized species of ornithine and the second ionization species of phosphate, since these are the forms involved in the catalytic mechanism. The true equilibrium constant is thus defined by the relation

[citrulline] [HPOi-] [H'] [ornithine' ] [carbamoylphosphate2 -1 K =

Using the same pK values as Marshall and Cohen, namely 6.97 for the pK2 of H3P04, 8.8 for the pK2 of the &amino group of ornithine, 9.41 for citrulline and 4.9 for carbamoylphosphate, the apparent and true K values are shown in Table 1. In the range of the temperature used (30 and 37 "C) the apparent K is not greatly modified.

The data reported here for the values of K are of the same order of magnitude as those of Marshall and Cohen [l]. The discrepancy of a factor 2.5 between their results and ours holds in the procedure followed for the determination of the product formed. By our procedure, detection of ornithine rather than car- bamoylphosphate avoids the bad effects of carbamoyl- phosphate hydrolysis in the course of manipulations after the reaction is stopped, and those resulting from incomplete transformation of carbamoylphosphate to glucose 6-phosphate through the use of carbamate kinase and hexokinase.

Table 1 . Equilibrium constant of the reaction

Effect o f p H on the Reaction Velocity

The activity curve as a function of pH is shown in Fig. 1. Unlike most ornithine carbamoyltransferases from bacterial sources (unpublished results), that of Pseudomonas does not show substrate inhibition when ornithine concentration is varied from 3- 50 mM. However, when phosphate was present orni- thine inhibition appeared as with all the transferases tested (Fig. 2).

Protection against Heat Inactivation of Ornithine Carbamoyltransferase

Binding studies are a useful tool for the determina- tion of mechanisms, but when pure material is not available, heat inactivation experiments give valuable indications as to the effectors which can bind to the enzyme [7]. As shown in Table 2, carbamoylphosphate and phosphate are potent effectors against heat in- activation. Ornithine and citrulline are without effect except if phosphate is present simultaneously.

I

PH

Fig. 1. Effect of p H on the activity of ornithine carbamo~ltrunsfercr.pe of Pseudomonas. The reaction mixture contained in 2.0 ml volume 5 mM carbamoylphosphate, 60 mM EDTA buffer at the pH in- dicated and the concentrations of ornithine were 3 mM (0) or 50 mM (0)

pH value Temperature Rate constant for hydrolysis Value of K of carbamoylphosphate ____. ~~

apparent true

"C min-' mM

5.8 30 5.23 x 10-3 26 150 3.0 6.8 30 5.65 x 10-3 48 500 3.2 6.8 37 I .64 x lo-' 48 800 3.2 7.8 30 5.60 x 10-3 145 600 2.2 8.0 37 1.20x lo-' 205 500 2.1 8.5 30 6.20 x 10-3 540 000 2.1

322 Pseudomonas Ornithine Carbamoyltransferase

7 8 9 0

6 PH

Fig.2. Effect of p H on the activity of ornithine carbamoyltrans- ferase in the presence of phosphate. Details as in Fig.l except that 25.0 mM phosphate was present in the reaction mixture

Table 2. Effect of' reactant against heat inactivation (60 "C) of anabolic ornithine curbamoyltransferase of Pseudomonas 1.9 ml of 300 mM Tris-HC1 pH 8.0 and the reactant were heated at 60 "C. 0.1 ml of a solution of enzyme at 1000 units/ml was introduced into the test tube and rapidly mixed. At various time intervals 50-p1 samples of the solution were removed and chilled rapidly in a test tube containing the reaction mixture (minus carbamoylphosphate) used to determine the enzyme activity. The remaining activity was measured as described earlier. The half-life of the enzyme was determined by plotting the log of the residual activity versus time of heating

Reactants Final concn Half-life time

mM min

None 5 KC1 500 12.5 Carbamoylphosphate 10 25 Phosphate 10 8 Phosphate 100 43 Ornithine 10 10 Ornithine I00 10 Citrulline 100 15 Arginine 100 5 Phosphate 100

+ > 200 Ornithine 100 Phosphate 100

+ 62 Citrulline 100

The effect of carbamoylphosphate against the heat inactivation is only indicative because of its lability at this temperature (half-life time of 2 min at 60 "C). At lower temperature (45 "C) the protection by carba- moylphosphate is very important.

15

10 s

5

0

I

.

1 2 3 [Carbamoylphosphate]-' (rnM-')

Fig. 3. Recbrocal plot of initial velocity pattern with carbamoyl- phosphate as variable substrate. Ornithine was held constant to (A) 0.5 mM, (A) 1 mM, (0) 2 mM, (0) 5 mM. u was measured as pmol citrulline formed in 10 min

Table 3. Kinetic constants of the anabolic ornithine carbamayl- transjerase reaction from initial velocity studies The values of the kinetic constants were calculated by fitting the data of the initial velocity studies to Eqn (1) where [Q] is equal to zero

Substrate Constant

mM

Carbamoylphosphate (A) Kj, = 0.19 K, = 1.0

Ornithine (B) Kib = 1.07 Kb = 5.0

Initial Velocity Studies

Like the ornithine carbamoyltransferases from other sources [7] (and unpublished results), the mecha- nism of the anabolic transferase of Pseudomonas is sequential. Fig. 3 shows the double-reciprocal plot where carbamoylphosphate is the variable substrate and ornithine held constant at various concentra- tions. Secondary plots of slopes and intercepts against reciprocal non-varied substrate concentration were linear. Values for the kinetic constants obtained from these plots are listed in Table 3. Corresponding ini- tial velocity data for ornithine as the variable sub- strate and carbamoylphosphate the fixed substrate also gave a converging pattern. Thus, both substrates must bind to the enzyme before any product is released.

Dead-End Inhibition

The initial velocity data do not allow one to distinguish between an ordered or a random mech-

V. Stalon, C. Legrain, and J.-M. Wiame 323

I 0.5 1.0 1.5 2.0

[Grnithinel-’ (rnM”)

Fig. 4. Non-competitive inhibition ofyhosplzonoacerate wiih ornithine as the varied substrate. Carbamoylphosphate was held constant to 2.5 mM. Phosphonoacetate concentrations were (0) 0 mM, (W) 5 mM, (0) 10 mM, (0) 20 mM. Units of v as in Fig.3

15

10

5

5 10 15 0 0

[Carbamoylphosphate].’ (mM-’)

Fig. 5. Uncompetitive inhibition of norvaline with carbamoylphos- phate as the variedsubstrate. Ornithine was held constant to 0.5 mM. Norvaline concentrations were (0) 0 mM, (0) 0.4 mM and (W) 0.8 mM. Units of v as in Fig. 3

anism, but dead-end inhibitors can be useful in further delineating the mode of substrate addition. In an ordered bi-bi mechanism an inhibitor interacting at the binding site for the first substrate (A) will give competitive inhibition with respect to that substrate and non-competitive with respect to the second. But an inhibitor interacting at the binding site for the second substrate (B) forms a dead-end complex with the binary complex enzyme-A. Inhibition will then be

Table 4. Kinetic constants for inhibition of the reaction by ornithinr analogues K i values were obtained from the replot of the graphs l j l ; against l/[ornithine] at various concentrations of the inhibitor

Inhibitor Ki

mM

~-Norvaline L-Leucine L-Isoleucine L- Arginine Glycine a,y-Diaminobutyrate L-Valine L - A 1 an i n e L-Asparagine L-Serine L-Lysine L-Methionine Putrescine

0.23 1.3 3.4 6.5 7.5 9.2

10 12 42 46 41 56

135

competitive with respect to the second substrate B and uncompetitive with respect to the first. In a random bi-bi mechanism, substrates A and B are kinetically equivalent; both can form binary complexes prior to the ternary enzyme-AB complex. An inhibitor binding at one substrate site will give competitive inhibition when the concentration of the substrate normally interacting there is varied, and non-competitive in- hibition when that of the other substrate is varied.

Phosphonoacetate was employed as dead-end in- hibitor specific for the carbamoylphosphate site, giving linear competitive inhibition with respect to carbamoylphosphate (not shown) and linear non- competitive inhibition with respect to ornithine as variable substrate (Fig. 4). These inhibition patterns are as predicted for either a random addition mech- anism or an ordered mechanism with carbamoyl- phosphate binding first, but rule out a compulsory ordered mechanism in which ornithine is the first substrate.

A number of metabolites act as ornithine anal- ogues. They give linear competitive inhibition with ornithine as variable substrate and with carbamoyl- phosphate as fixed substrate (Table 4). The inhibi- tions are uncompetitive with respect to carbamoyl- phosphate when ornithine was used as fixed substrate (Fig. 5). Thus, these observations rule out a random mechanism, which thus appears to be ordered with carbamoylphosphate as first substrate and ornithine as second. In addition to the inhibition listed in Table 2, agmatine, homoarginine, guanidomethyl, guanidobutyrate and guanidopropionate have in- hibitory action on the anabolic carbamoyltransferase of Pseudomonas. The binding of one ornithine anal- ogue kinetically excludes the binding of another.

324 Pseudomonas Ornithine Carbamoyltransferase

10

1 . -

C I

1 2 /Ornithine]-’ (mM-’)

Fig. 6. Product inhihition by phosphute with ornithine as the varied substrate. Carbamoylphosphate 1.25 mM. The concentrations of phosphate were (0) 0, (0) 10 mM, (3 20 mM, (m) 40 mM. Units of 11 as in Fig. 3

1.0 .

1 . -

I I 0.5 1.0 1.5

[Carbamoylphosphate]~’ (rnM-’)

Fig. I. Ejject of’ citrulline on the initial velocity pattern ivith carba- moylphosphate as the varied subslrate. Ornithine 2 mM. Citrulline concentrations were (0) 0, (0) 50 mh4, (0) 100 mM, (m) 150 mM, (A) 200 mM. Units of o as in Fig.3

Product In hibit ion

Phosphate (and its analogue arsenate) was found to be a competitive inhibitor with respect to carba- inoylphosphate and a non-competitive inhibitor with respect to ornithine (Fig. 6) . These patterns are thus consistent with a compulsory mechanism with carba- moylphosphate as first substrate to be bound and phosphate as last product to be released, but not with ornithine as first substrate.

Citrulline, one of the products of the carbamoyla- tion reaction, has the same inhibition pattern as phosphonoacetate or phosphate. The competitive

3

2 1 . -

1

C 1 2 3 (Ornithine]-’ (rnM’)

Fig. 8. Ejyect of’citrulline on the initial velocity pattern with ornirhine as the varied substrate. Carbamoylphosphate 1 mM. Citrulline con- centrations were (0) 0, (.) 50 mM, (A) 100 mM, (A) 200 mM. Units of t’ as in Fig. 3

action towards carbamoylphosphate is shown in Fig. 7 whereas the non-competitive pattern towards ornithine is represented in Fig. 8.

Mechanism of the Anabolic Ornithine Carbamoyltransjerase Reaction

It would appear from the cumulative results that the kinetic mechanism of the anabolic transferase of Pseudomonas presents no fundamental differences with the transferase of E. coli. Indeed, with the ex- ception of citrulline action and the high sensitivity of the transferase of Pseudomonas to a greater number of ornithine analogues the results are the same as those obtained with the E. Cali enzyme. A Theorell and Chance mechanism has been proposed for the trans- ferase of E. coli [7]. The same conclusion can be reached for the anabolic transferase of Pseudomonas, and this is based upon the following observations.

a) That the mechanism is largely ordered is clearly shown by the uncompetitive inhibition of the orni- thine analogues towards carbamoylphosphate. This pattern should appear only if the mechanism is a sequential ordered one where carbamoylphosphate is the first substrate to be bound to the enzyme and ornithine second.

b) A Theorell and Chance mechanism implies that KiaKb = K,Kib where A and B represent carbamoyl- phosphate and ornithine respectively. The K terms represent their Michaelis constants whereas the Ki terms represent their inhibition constants. The ex- perimental values determined (see Table 3 ) agree well with this condition which is necessary but insufficient to prove such a mechanism.

V. Stalon, C. Legrain, and J.-M. Wiame 325

30

20

2

10

[Carbamoylphosphak-’ (rnM-’)

Fig. 9. Competitive substrate inhibition by ornithine induced by the presence of phosphate ( 4 mM). The concentrations of ornithine were (0) 10 mM, (m) 25 mM, (0) 50 mM, (0) 75 mM, (0) 100 mM. Units of t‘ as in Fig. 3

c) Since phosphate was able to form a binary com- plex with the enzyme (see below) and was competitive towards carbamoylphosphate, this indicates that phos- phate was most probably the last product to leave the enzyme. Thus citrulline acts as a dead-end inhibitor and must interact with the free enzyme.

d) Carbamoylphosphate and phosphate are potent protectors against heat inactivation, thus suggesting binary complexes of the enzyme with those effectors. The poor protection offered by ornithine is an indica- tion that a binary enzyme-ornithine complex is less readily formed. Since ornithine increases the half-life of the enzyme only if phosphate is present, the possibility exists that a ternary enzyme-phosphate- ornithine complex is formed.

e) The hypothesis of the existence of such a ternary complex is reinforced by the observation that phos- phate induces ornithine inhibition which is hyperbolic and competitive towards carbamoylphosphate (Fig. 9). In an ordered mechanism, if A is the first substrate, substrate inhibition by B results from combination with EQ in dead-end fashion and is total and linear (that is the rate becomes zero at infinite inhibitor concentration). However, in a random mechanism, substrate inhibition by B is caused by the formation of the EBQ complex at high B concentration, and results in alternate reaction pathways since EBQ can break down to form EB which can in turn react with A to regenerate the central complex. Such substrate inhibition is always hyperbolic and competitive to- wards the first substrate (A). Thus, the observation of ornithine inhibition only when phosphate is present must be explained by a random mechanism. We must

keep in mind that when a mechanism is a close approximation of a Theorell and Chance mechanism, it is generally a more complicated scheme of partially random mechanism where one pathway carries most of the reaction flux. The fact that the anabolic transferase is sensitive to citrulline as well as to ornithine anal- ogues such as arginine or lysine, in contrast with E. coli enzyme, argues in favour of some randomness for the kinetic mechanism. However, since most of the results are more easily explained by a Theorell and Chance mechanism, the rate equation for phosphate inhibition would be represented in reciprocal form by

- K, -- 1 -

where Ki, is the dissociation constant of phosphate from the binary enzyme-phosphate complex. The true Ki, is easily obtained from slope replot of the competitive inhibition of phosphate towards carba- moylphosphate whereas the apparent Ki, obtained from the non-competitive inhibition towards orni- thine is related to the true Ki, by the following rela- tion :

K??” = Ki, (1 + g) .

(3)

(4)

The values determined and those calculated from intercept replot of the non-competitive inhibition agree well (Table 5), but some discrepancy appears for the calculated value of the slope replot of Fig.6.

Since phosphonoacetate and citrulline give the same pattern as phosphate, the same type of equation can be used to calculate the inhibition constant for these inhibitors with the following substitution of [Q] by [I] and Ki, by Ki (see Table 5) . In both cases the values agree well.

Competitive inhibition towards ornithine and un- competitive inhibition towards carbamoylphosphate are described by the following equations

326 Pseudomoizas Ornithine Carbamoyltransferase

Table 5. Kinetic constunts for inhibition of rhe reaction by products and dead-end mhibitoous The inhibition constants were calculated using the relationships given in the text for a Theorell and Chance mechanism. C = competitive, NC = non-competitive, UC = uncoinpetitive

Inhibitor Varied substrate Fixed substrate Pattern Apparent Ki True Ki

m M mM mM Q (phosphate) A (carbamoylphosphate)

13 (ornithine)

Ornithine (B) in presence of Q A (carbamoylphosphate) Phosphonoacetate A (carbamoylphosphate)

H (ornithine)

Citrulline

Norvaline

A (carbamoylphosphate) B (ornithine)

A (carbamoylphosphate) B (ornithine)

5.0 1 .o

5.0 2.5

2.0 1 .0

0.5 2.5

C NC

C C

C NC

uc C

Ki, = 5.4 Ki, 5.4 Ki, = 12.7 K,s = 80

Kil = 11 K , , = 5.5 hyperbolic inhibitmn K,s = 4 K , = 4 K,s = 54 K , = 3.9 Kii = 14 Ki = 4.0 Xis = 40 Kip = 40

Kip = 40 K,s = 250 K,i = 90 Kip = 45 K,c = 0.22 Ki = 0.25 K;s = 0.22 Ki = 0.22

Thus the apparent Ki are related to the true Ki as follows :

KPPP = Ki 1 + __ from Eqn (3) (7) ( &)

If a Theorell and Chance mechanism prevails, the kinetic constants are related to the thermodynamic equilibrium constant by the following relationships

Consequently, the Michaelis and inhibition constants for citrulline can be reached, if the value of K is known. Under our conditions, the value established was 2.05 x lo5. Values of 4 M and 38 M were obtained for the Michaelis and inhibition constants for citrulline respectively. Although these high values explain why the anabolic transferase of Pseudomonas appeared functionally irreversible, they are inadequate to ex- plain the low maximum rate measured for the phosphorolytic (or arsenolytic) cleavage of citrulline. Indeed, by supposing that the maximum rate for the carbamoylation of ornithine 6 is equal to 1, relative values for individual rate constants could be calculated, as well as the theoretical value of the maximum rate Vz for the reverse reaction the phos- phorolysis of citrulline. The value of 5 obtained for V’ shows that this is not fundanientally different from that obtained with the transferases of E. coli [7] or that of the Streptococcus enzyme [I].

Thus, how to explain that the ratio of the maximum rate in both directions appeared so high? First of all, the poor affinity of the enzyme for citrulline ex- plains why the determination of Y in the phospho-

rolytic cleavage is not accurate. Secondly, the af- finity of the enzyme for citrulline acting as an abortive inhibitor is much stronger than its affinity for citrulline acting as substrate of phosphorolysis.

In other words, even if the calculated value for the Michaelis constant for citrulline is overestimated, substrate inhibition by citrulline must occur in the reverse reaction. Maximum rate determination as pro- posed by Cleland [I21 does not take into account such abortive complexes, and the extrapolated value thus appears relatively low compared to the situation where substrate inhibition does not exist.

CONCLUSIONS

The anabolic ornithine carbamoyltransferase of Pseudomonas differs from the transferases of other sources in that it .exhibits a considerably lower ability to catalyze the reverse reaction in spite of a com- parable catalytic capacity in the anabolic direction. Whatever the mechanism of the reaction, the same K applies to all the transcarbamylases and the Haldane relationships between V, the Michaelis constants, the inhibition constants and K should hold.

Thus, if the anabolic transferase shows an unusual V ratio, then it must also show an unusual and compensatory K,,, ratio, and vice versa. If a Theorell and Chance mechanism applies to the case of the anabolic transferase of Pseudomonas the maximum rate of the reverse reaction (the phosporolytic cleavage of citrulline) is of the same order of magnitude as the maximum rate of the carbamoylation of ornithine. The affinity of the enzyme for citrulline appeared relatively low. However, the phosphorolytic cleavage of citrulline is prevented mainly by the fact that citrulline inhibits the reaction. This inhibition be- comes more pronounced as the ratio of the affinity

V. Stalon, C. Legrain, and J.-M. Wiame 327

0.3

0.2 $

0.1

1

O k 4 0 10 20 30 40

[PI (mMi

Fig. 10. P h i oj the computed V/v versus concentration of substrate P for a Theorell and Chance mechanism. The Michaelis and inhibition constants were assumed to be 1 and 5 mM respectively. Substrate Q was held at 0.025 mM, the value of its Michaelis constant. The values of KIP are for (1) cc, for (2) 10, for (3) 1 and for (4) equal to 0.1

constant Kp to the inhibition constant increases (Fig. 101.

Thus the anabolic transferase appears irreversible, at least in vitro, because of (a) the non-saturating level of citrulline and (b) the existence of a dead-end complex between the enzyme and citrulline which strongly reduces the apparent maximum velocity.

It is not possible to predict with any certainty where citrulline inhibition might play a metabolically important role in vivo, but it is interesting to note that inhibition by citrulline and arginine for the anabolic transferase of Pseudomonus is present in the physio-

logical context where two transferases (anabolic and catabolic) must compete for the same substrate. Thus at elevated levels of that metabolite, the anabolic ornithine carbamoyltransferase might advantageously be inhibited, thereby allowing metabolic access to the arginine catabolic pathway only. Indeed, these two transferases of Pseudomonus are designed for specific behaviour. Each of these catalyses with efficiency the direction of the reaction corresponding to its function.

We are indebted to M. Penninckx for suggesting the use of phosphonoacetic acid as carbamoylphosphate analogue and for performing its synthesis. We are grateful to D. Gigot and J.-P. ten Have for their help in some experiments and to K . Broinan for reading the manuscript. This work has been supported by Grant 2.4542175 from the Fonds de la Recherche Fondamentale Collective. C . Legrain was recipient of fellowship from the Fonds National de la Recherche Scientifique.

REFERENCES

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

2. Stalon, V., Ramos, F., Pierard, A. & Wiame, J. M. (1967)

3. Broman, K., Stalon, V. & Wiame, J. M. (1975) Biochem. Bio-

4. Stalon, V. (1972) Eur. J . Biochem. 29, 36-42. 5. Ramos, F., Stalon, V., Pierard, A. & Wiame, J. M. (1967)

Biochim. Biophys. Acta 139, 98 - 106. 6. Reference deleted. 7. Legrain, C. & Stalon, V. (1976) Eur. J . Biochem. 63, 289-301. 8. Rivard, D. E. & Carter, M. E. (1955) J . Am. Chem. Soc. 77,

9. Stalon, V., Ramos, F., Pierard, A. & Wiame, J . M. (1972) Eur.

1668.

Biochim. Biophys. Acta, 139, 91 - 97.

phys. Res. Commun. 66, 821 - 827.

1260-1261.

J . Biochem. 29,25- 35. 10. Cleland, W. W. (1963) Biochim. Biophys. Acta 67, 104- 137. 11. Reichard, P. (1957) Acta Chim. Scand. 11, 523. 12. Cleland, W. W. (1967) Annu. Rev. Biochem. 36, 77- 112.

V. Stalon, C. Legrain, and J.-M. Wiame, Institut de Recherches du C.E.R.I.A., Avenue Emile-Gryson 1, B-1070 Bruxelles, Belgium