A simple approach to identify the mechanism of intermediate transfer: enzyme system related to triose phosphate metabolism

Biochimica et Biophysica Acta 915 (1987) 53-59

53

Elsevier BBA 32913

A simple approach to identify the m e c h a n i s m o f intermediate transfer. e n z y m e s y s t e m related to triose phosphate m e t a b o l i s m Ferenc Orosz and Judit Ovfidi Institute of Enzymology, Biological Research Center, Hungarian Academy of Sciences, Budapest (Hungary)

(Received 6 March 1987)

Key words: Intermediatetransfer; Kineticanalysis;Isotope dilution; Triosephosphateconvertingenzyme; Enzymekinetics

A new approach is described to identify the mechanism of transfer of intermediates of consecutive reactions catalysed by two functionally related enzymes. Interactions resulting in conformational changes of the individual enzymes and/or channelling of the intermediate can be identified by comparing the rate constants of the coupled and individual reactions. Using these kinetic parameters, the relative specific radioactivity of the end product can be calculated according to the different mechanisms. The comparison of these values with the experimentally determined relative specific radioactivity enhances the sensitivity of the determination. The interaction between aldolase (D-fructose-l,6-bisphosphate D-glyceraldehyde-3-phosphate-lyase, EC 4.1.2.13) and glyceraldehyde-3-phosphate dehydrogenase (D-glyceraldehyde-3phosphate:NAD + oxidoreductase (phosphorylating), EC 1.2.1.12) was analysed. The data agree with the model in which dmnneling of the intermediate was assumed. The results suggest that glyceraldehyde 3-phosphate is functionally compartmentalised within the reconstituted enzyme system, which may be relevant under physiological conditions.

Introduction One of the consequences of complex formation between two functionally related enzymes may be the channeling of the intermediate, i.e., the direct transfer of the product from the active centre of the first enzyme to that of the second in a coupled reaction. This phenomenon can be investigated by kinetic approaches [1-6]. Recently, direct transfer of N A D H between various pairs of dehydro-

Abbreviations: Fru-l,6-P2, fructose 1,6-bisphosphate; glyceraldehyde-3-P, glyceraldehyde 3-phosphate; 3-P-glycerate, 3phosphoglycerate.

Correspondence:J. Ovhdi, Institute of Enzymology,Biological Research Center, Hungarian Academy of Sciences, H-1113 Budapest, Karolina ut 29, Hungary.

genases has been revealed using a technique of enzyme buffering of metabolite concentrations [7]. A direct approach for showing channelling of intermediates in metabolically related enzyme systems is the isotope dilution technique [3,8,9]. The interaction of aldolase (D-fructose-l,6-bisphosphate D-glyceraldehyde-3-phosphate-lyase, EC 4.1.2.13) and glyceraldehyde-3-phosphate dehydrogenase (D-glyceraldehyde-3-phosphate : N A D ÷ oxidoreductase (phosphorylating), EC 1.2.1.12) has been examined by several authors, using different techniques [1,10-14]; however, the results are rather contradictory. In previous works [1,10], we provided kinetic and physicochemical evidence for complex formation between the two enzymes. Recently, we demonstrated [15] the alternative binding of glyceraldehyde-3-phosphate dehydrogenase and triosephosphate isomerase (D-

0167-4838/87/$03.50 © 1987 Elsevier Science Publishers B.V. (BiomedicalDivision)

54 glyceraldehyde-3-phosphate ketot-isomerase, EC 5.3.1.1) to aldolase. The kinetics of this three-enzyme system were analysed under steady-state conditions. In this work the mechanism of interaction of aldolase and glyceraldehyde-3-phosphate dehydrogenase is investigated kineticaUy and also by using an isotope dilution technique. Triosephosphate isomerase added to the enzyme system ensures the rapid equilibration of triose phosphates produced by aldolase, but no significant amount of aldolase-isomerase complex could be formed [15]. Quantitative determination of isotopic label in the end product, 3-phosphoglycerate (3-P-glycerate), was rendered possible by a method developed recently in our laboratory, using high-performance liquid chromatography [16,17]. Materials and Methods

Fru-l,6-P2, glyceraldehyde 3-phosphate (glyceraldehyde-3-P) and NAD were from Boehringer. [U-14C]Fru-l,6-P2 (specific radioactivity 8840 MBq/mmol) was obtained from Amersham. All other chemicals were reagent-grade commercial preparations. Aldolase [18] and glyceraldehyde-3-phosphate dehydrogenase [19] were purified from rabbit skeletal muscle as described previously, while triosephosphate isomerase was purchased from Boehringer. Concentration of enzymes was determined spectrophotometrically using absorption coefficients (A°~10~) of 1.0, 0.74 and 1.32 for glyceraldehyde-3-phosphate dehydrogenase [20], aldolase [21], and triosephosphate isomerase [22], respectively. The specific activities of the enzymes, determined as described previously [23,24], were 28-30 k a t / m o l for aldolase, 4500-4800 k a t / m o l for triosephosphate isomerase, measured with glyceraldehyde-3-P as substrate, and 230-250 k a t / m o l for tetrameric glyceraldehyde-3-phosphate dehydrogenase.

Enzyme assay The consecutive enzymatic reactions were carried out at 25 o C in 40 mM triethanolamine buffer (pH 8.5). The assay mixture comprised 2 mM [U-14C]Fru-1,6-P2 (radioactivity of 2.5 ml assays was 1.25.106 d p m ) / 6 mM N A D + / 4 mM arsenate/0.20 /~M aldolase/0.60 #M glyceralde-

hyde-3-phosphate dehydrogenase/0.30 /~M triosephosphate isomerase. The conditions were chosen such that aldolase reaction should be zero order and the dehydrogenase reaction first order, with respect to their substrates. The reaction was started by adding either Fru-l,6-P2, or Fru-l,6-P 2 plus glyceraldehyde-3-P, and was followed by monitoring N A D H formation at 340 nm for determination of the total concentration of 3-Pglycerate. The reaction was stopped at 22 s by addition of 0.2 ml 70% perchloric acid to 2.5 ml sample. The samples were kept at 4 o C for 20 min, then centrifuged for 15 min at 6000 × g. 20/~1 of supernatant was analysed by high-performance liquid chromatography.

Procedures and equipment for high-performance liquid chromatographic measurements The labelled 3-P-glycerate was separated from the other radioactive metabolites by high-performance liquid chromatography [16,17], utilising a Liquopump OE 312 pump (Labor MIM, Budapest, Hungary) and a SynChropack AX300 anion-exchanger column 250 × 4.1 mm i.d. (SynChrom, Linden, IN) The apparatus was operated at room temperature, at about 7 MPa (70 bar) overpressure, which gave a flow rate of 0.86 m l / min. Isocratic elution was used with a mobile phase of 150 mM KH2PO4, adjusted to pH 2.5 by addition of phosphoric acid. For determination of the amount of the radioactive 3-P-glycerate by scintillation counting, the total volume of fractions containing only 3-P-glycerate was applied. The radioactivity was measured by a LKB Wallace 1211 RackBeta liquid scintillation counter. For the determination of measured relative specific radioactivity, which is the ratio of the labelled and total concentrations of 3-P-glycerate, the concentration of isotopic 3-P-glycerate was obtained from the scintillation measurements after chromatographic separation. The total concentration of 3-P-glycerate was taken from the N A D H concentration which was formed in equimolar concentration with 3-P-glycerate. Results and Discussion

Theoretical considerations The scheme of two consecutive reactions cata-

55

lysed by enzymes E 1 and E 2 in the presence of an external intermediate is shown in Scheme I, B

C

Ax__E_~Bx ~ )C x v k'([B] + [BX]) Scheme I.

where the superscript x indicates radioactive metabolites. If the first reaction proceeds at constant velocity and the concentration of the intermediate, [B] + [BX], is low enough for the second reaction to be first order with respect to its substrate, then the progress curve of product formation can be described by the sum of Eqns. 1 and 2 [3]: [C x ] = v t - v r ( 1 - e

(1)

-t/y)

[C l = [ B 0 ] ( 1 - e -k'`)

(2)

where, v is the E~ activity measured at a given concentration. 1/~" and k' are the pseudofirst-order rate constants for conversion of B, determined in the consecutive and individual reactions, respectively, in the two-enzyme system. [B0] is the initial concentration of unlabelled B. In the case of a noninteracting enzyme system, the slope of the linear, steady-state part of the progress curve of [C x] vs. t in the absence of external intermediate equals v. The intercept, r, on the time axis equals the reciprocal value of the pseudo-first-order rate constant, 1 / k , of conversion of B [5]. Naturally, k is not influenced by the presence of E~ ( k ' = k). The relative specific radioactivity (r) of the end product at a given time can be calculated from Eqn. 3: rlc*le =

[C X] [CX]+ [C]

vt - k ( 1 - e - k t ) Vt--k(1--e-kt)+[Bo](1--e-kt

(3) )

Obviously, this value ( r ~ c) should equal that determined directly from the ratio of the labelled and total concentrations of the final product (cf. Eqn. 3). If any change in the kinetic parameters takes place due to the interaction between E 1 and E 2, it should be reflected in the value of measured rela-

tive specific radioactivity (rm~s). If this value (rm~s) is compared with the r calculated according to different models, the mechanism of intermediate transfer can be identified. Obviously, the mechanism can be also deduced from the comparison of the kinetic parameters measured in the coupled and individual reactions. However, use of isotope dilution technique combined with kinetic analysis enormously enhances the sensitivity of the determination. The following models will be considered with the relevant equations for the calculations of relative specific radioactivity. If the interaction of two enzymes induces alteration in the tertiary and quaternary structure of the enzymes without producing channelling of the intermediate, then the steady-state velocity of the coupled reaction (v') may differ from v (El activity), a n d / o r the pseudo-first-order rate constant, k, may differ from k', measured in the two enzyme system. However, k' should equal 1/r. Therefore, the relative specific radioactivity for the model 'interaction inducing conformational changes' can be calculated from Eqn. 4: r~a ¢ _

v' t - v ' ( 1 - e -t/~) v ' t - o"r (1 - e - t / r ) + [B0](1 - e - t / r )

(4)

In an interacting system, on one hand, if the intermediate produced endogenously by the E1-E 2 complex is channelled between E 1 and E2, then 1/~-> k [5]. This relationship may result from either steric hindrance which impedes the diffusion of intermediate into the bulk solution or simply from the juxtaposition of active sites of the enzymes in complexed form. The latter case may be considered special to the intermediate channelling denoted as 'leaky channel'. On the other hand, if the binding of exogenous B to the E1-E 2 complex is sterically hindered, then k > k' [25], whereas for 'leaky channel' k = k', but k < 1/~'. If this kind of mechanism of the intermediate transfer is fulfilled, the steady-state velocity of the coupled reaction should equal v. Therefore, the relative specific radioactivity for the 'channelling' model is given by Eqn. 5: r~¢ =

vt - v ' r ( 1 - e - t / ¢ )

vt - v r ( 1 - e - ' / r ) + t B o l ( 1 - e

(5)

-k'')

56

This approach does not say anything about the extent of the channel, since the kinetic parameters are the weighted sums of the corresponding kinetic parameters for interacting and noninteracting enzyme species [8], because the reaction mixture always contains both complexed and free E 1 and E 2. Nevertheless, the mechanism of the interaction of functionally related enzymes can be determined without knowing the actual concentration of the heterologous complex. This is possible because the kinetic parameters (v, % k ' ) for the calculation of relative specific radioactivities according to different models, as well as the concentrations of final product for the determination of rme,s, were determined at the same enzyme concentrations. Naturally, it is assumed in all calculations that the degree of dissociation of the E1-E 2 complex does not change during the coupled reaction until time t, when the reaction is stopped. This type of analysis also enables one to evaluate the effect of an intermediate on the heterologous interaction. If the ratio of rmeas and rcajc increases or decreases with the elevation of [B], then intermediate-induced association or dissociation of the heterologous enzyme complex occurs.

Experimental We analysed the coupled reaction catalysed by aldolase and glyceraldehyde-3-phosphate dehydrogenase in the presence of triosephosphate isomerase. Isomerase was added to the system to ensure the rapid equilibration between the two triose phosphates. Under the experimental conditions used, a significant amount of aldolase-dehydrogenase complex can be formed [15]. Obviously, the kinetic parameters, v, r and k', were determined in the presence of isomerase; therefore, the validity of the equations presented in the theoretical part is not influenced by the presence of isomerase. The effect of isomerase on the kinetics of consecutive reaction catalysed by aldolase and dehydrogenase has been presented in a previous paper [15]. Fig. 1 shows the time-course of glyceraldehyde3-P oxidation catalysed by dehydrogenase in the presence of isomerase and nonfunctioning aldolase (Fru-l,6-P2 was not added to the assay mixture) (curve 1), as well as the progress curves

20

~10.

I

I 05 Time (rnin)

I 10

Fig. 1. Time-course of the coupled aldolase-glyceraldehyde 3-phosphate dehydrogenase reaction in the presence of tnosephosphate isomerase. The irreversible oxidation of glyceraldehyde-3-P catalysed by dehydrogenase in the presence of aldolase (curve 1); Fru-l,6-P 2 conversion into 3-P-glycerate in the coupled reaction of aldolase, isomerase, and dehydrogenase (curve 2); the same as in curve 2 with added glyceraldehyde-3-P (curve 3). Enzyme concentrations were 0.20, 0.60 and 0.30 #M for aldolase, dehydrogenase, and isomerase, respectively. Initial concentrations of Fru-l,6-P 2, and glyceraldehyde-3-P, if present, were 2 and 5.9.10 -2 mM, respectively. For the conditions of the assay, see Materials and Methods.

of NADH formation from [U-14C]Fru-l,6-P2 in the three-enzyme system. The experimental conditions for the coupled reaction were chosen to meet criteria described in the theoretical part, viz., Fru1,6-P2 was high enough to saturate aldolase *, and the reaction catalysed by aldolase was the ratelimiting step of final product formation [15]. The progress curve was also measured in the presence of external unlabeUed glyceraldehyde-3-P (curve 3). The endogenously formed and exogenously added glyceraldehyde-3-P was low enough for the dehydrogenase action to be first-order. The concentration of added glyceraldehyde-3-P was commensurate with that expected to be formed endo* The reverse reaction of aldolase in the coupled reaction can be disregarded, since (i) Fru-l,6-P 2 is in high excess to shift the equilibrium towards triose phosphates formed in equimolar m o u n t s by aldolase: (ii) the reaction catalysed by dehydrogenase is irreversible in the presence of arsenate.

57 genously in the above in vitro system under steady-state conditions. The reaction shown (Fig. 1, curve 1) can be described by a single pseudo-first-order reaction, indicating that: (i) isomerase was present in sufficient excess to secure the rapid equilibrium of triose phosphates, and (ii) the glyceraldehyde-3-P concentration was below the K m for dehydrogenase ( K m = 1.3.10 -3 M), as determined in the presence of isomerase. Thus, the apparent firstorder rate constant, k, reflects the activity of dehydrogenase in the given system. In addition, we found that aldolase in the absence of Fru-l,6-P 2 did not influence the time course of glyceraldehyde-3-P conversion catalysed by the dehydrogenase. The same result was obtained if the aldolase was added in excess with respect to dehydrogenase (not shown). This finding in itself suggests a lack of interaction inducing conformational changes. Therefore, further possibilities should be considered. The phenomenon can be explained by either the lack of interaction or interaction without steric hindrance of bound aldolase, i.e., the exogenous glyceraldehyde-3-P can bind to the complexed dehydrogenase with the same probability as to the uncomplexed one. The time-course analysis of the coupled reaction catalysed by aldolase and dehydrogenase could be used with success to distinguish between these possibilities. Under the experimental conditions used, the activity of the first enzyme, aldolase, was constant (Fru-l,6-P 2 >> K ald°lase) and also low, relative to the second [15]. The linear part of the final prod-

uct formation represents the steady-state velocity (v) of the coupled reaction as demonstrated previously. The intercept on the time axis of the linear part of curve 2 equalled the transient time, ~', in the three enzyme system. Table I illustrates the mean values of kinetic parameters from five sets of experiments as well as relative specific activities calculated from the kinetic parameters, assuming different mechanisms for the intermediate transfer. F r o m these data, it is clear that the value of 1/~" determined in the coupled reaction is significantly higher than the value of k ' measured by glyceraldehyde-3-P conversion catalysed by dehydrogenase in the presence of isomerase and non-functioning aldolase. The reduction in the value of "r suggests the direct transfer of glyceraldehyde-3-P between the active sites of the two enzymes. The same conclusion can be drawn from the isotope dilution experiments. For the determination of the isotope incorporation into the end product from [U14C]Fru-l,6-P2, the coupled enzymatic reaction in the presence of added non-labelled glyceraldehyde-3-P was stopped at time t. The concentrations of the labelled and total product are presented in Table II. The measured relative specific radioactivity was obtained from the ratio of the labelled and total concentrations of 3-P-glycerate, where the latter values equal the N A D H concentration determined spectrophotometrically [17]. Time t was chosen to be shorter than that needed for the complete conversion of exogeneous (added) glyceraldehyde-3-P in the coupled system. The relative specific radioactivity

TABLE I KINETIC PARAMETERS MEASURED IN THE THREE-ENZYME SYSTEM AND THE CALCULATED RELATIVE SPECIFIC RADIOACTIVITIES Procedures were carried out as described under Materials and Methods. v is the slope of the progress curve of the coupled reaction in the absence of added glyceraldehyde-3-P; ¢ is the transient time; k' is the pseudo-first-order rate constant of glyceraldehyde-3-P oxidation. The mean values were calculated from five series of experiments. The r values were calculated from Eqns. 3-5, using the mean values of kinetic parameters at t = 22 s. Kinetic parameters measured v (~ M. s - 1)

1/¢ (s - 1)

k' (s - 1)

6.3 + 0.1

0.20 + 0.02

(4.3 + 0.1). 10- 2

Relative specificradioactivities (r) calculated for models noninteracting interaction inchannelling ducing conformational changes 0.58 0.65 0.75

58 TABLE II THE CONCENTRATIONOF THE END-PRODUCTSOF THE COUPLED REACTION AND THE MEASURED RELATIVE SPECIFIC RADIOACTIVITY Procedures were carried out as described under Materials and Methods. The specific radioactivity of 3-P-glycerate* without dilution was 2.47.10s dpm/mmol. The background count was less than 30 dpm. rmeas~rea= [3-P-glycerate*]/[3-P-glycerate]toua = [3-Pglycerate* ]/[NADH]. Added [glyceraldehyde-3-P](/~M)

AA34o

[NADH] (/xM)

dpm

[3-P-glycerate* ] (/~M)

rme..~,, ~

59

0.71 5:0.03 0.94 5:0.04

1145:5 151 5:6

563 5:24 558 5:24

114 5:5 113 5:5

1.0 0.75

should be unity with no dilution. This may occur either in the absence of exogenous glyceraldehyde3-P (cf. Table II), or when the glyceraldehyde-3-P formed endogenously is completely channelled by the enzyme complex and no free dehydrogenase is in the solution. For the complete mixing of endogenous radioactive glyceraldehyde-3-P and exogenous (added) glyceraldehyde-3-P, the value of relative specific radioactivity of the final product is 0.58, as calculated for the noninteracting system (cf. Table I). Obviously, this value depends on both the concentrations of added unlabeUed glyceraldehyde-3-P and the time, t, when the reaction was stopped. The measured relative specific radioactivity value should compare with the values of that calculated assuming different models (cf. Table I). On the basis of this comparison, the mechanism of interaction may be also identified with the 'cbannelling' model. The same conclusion can be drawn from Fig. 2, in which the measured and calculated progress curves of the coupled reactions in the presence of exogenous glyceraldehyde-3-P are presented. The theoretical progress curves were calculated assuming a noninteracting mechanism, interaction inducing conformational changes, and channelling of intermediate. It is clear that the experimental curve fits to the theoretical curve, assuming direct transfer of the intermediate. In fact, glyceraldehyde-3-P in the coupled reaction seems to be directly transferred from aldolase to dehydrogenase due to the juxtaposition of active sites of the enzymes. Since the time-course of conversion of exogenous glyceraldehyde-3-P by dehydrogenase is not influenced by aldolase, it is probable that the active site of dehydrogenase is not buried,

at least not completely, by aldolase. Recently, we demonstrated [15] that the isomerase at relatively high concentration can modulate the interaction between aldolase and dehydrogenase due to its alternative binding with dehydrogenase for aldolase. However, under physio-

I

I

2.0,

~10

I 05 Time (rain)

I 10

Fig. 2. Progress curves of NADH formation in the coupled reactions catalysed by aldolase, triosephosphate isomerase and glyceraldehyde-3-phosphate dehydrogenase in the presence of added glyceraldehyde-3-P. Different interaction mechanisms were considered. The continuous curves 1, 2 and 3 were computed according to the noninteraction, interaction with conformational changes, and 'channelling' models, respectively. For the calculation of the theoretical curves, the denominators of the right-hand side of Eqns. 3-5 were used. The parameters, v, ~- and k', for the equations were taken from Table I. Dashed line is the experimental curve measured as curve 3 in Fig. 1.

59

logical conditions where the concentration of dehydrogenase is significantly higher than that of isomerase, the functional compartmentation of glyceraldehyde-3-P may be envisaged. Considering the new approach applied in this study, we conclude that it can be easily used for investigation of the mechanism of intermediate transfer in an enzyme system containing functionally related enzymes.

Acknowledgements The authors are sincerely grateful to Professor T. Keleti for his constant interest and help in this work, and to Mrs. M. Nuridshny for skillful and conscientious technical assistance.

References 10vAdi, J. and Keleti, T. (1978) Eur. J. Biochem. 85, 157-161 2 Salerno, C., Ovhdi, J., Keleti, T. and Fasella, P. (1982) Eur. J. Biochem. 121, 511-517 3 Bryce, C.F.A., Williams, D.C., John, R.A. and Fasella, P. (1976) Biochem. J. 153, 571-577 4 Grazi, E. and Trombetta, G. (1980) Eur. J. Biochem. 107, 369-373 5 Hess, B. and Boiteux, A. (1972) in Protein-Protein Interaction, Proc. 23rd Mosbach Colloquium (Jaenicke, R. and Helmreich, E., eds.), pp. 271-297, Springer Verlag, Berlin 6 Bartha, F. and Keleti, T. (1979) Oxid. Commun. 1, 75-84 7 Srivastava, D.K. and Bernhard, S.A. (1986) Science 234, 1081-1086 8 Keleti, T. and Ovfidi, J. (1987) Curt. Topics Cell. Reg. 29, in the press

9 Manley, E.R., Webster, T.A. and Spivey, H.O. (1980) Arch. Bioehem. Biophys. 205, 380-387 10 OvAdi, J., Salerno, C., Keleti, T. and Fasella, P. (1978) Eur. J. Biochem. 90, 499-503 11 Patthy, L. and Vas, M. (1978) Nature (London) 276, 94-95 12 Masters, C.J. and Winzor, D.J. (1981) Arch. Biochem. Biophys. 209, 185-190 13 Kwon, T.W. and Olcott, H.S. (1965) Biochem. Biophys. Res. Commun. 19, 300-305 14 FSldi, J., Szabolcsi, G. and Friedrich, P. (1973) Acta Biochim. Biophys. Acad. Sci. Hung. 8, 263-265 15 Orosz, F. and OvAdi, J. (1986) Eur. J. Biochem. 160, 615-619 16 Orosz, F. and Ovhdi, J. (1986) in Chromatography 84, the 4th Annual American-Eastern European Symposium on Liquid Chromatography (Kalfisz, H. and Ettre, L.S., eds.), pp. 425-434, Akad6miai Kiad6, Budapest, Elsevier, Amsterdam 17 Orosz, F., Nuridsfiny, M. and Ov~di, J. (1986) J. Biochem. Biophys. Methods 13, 325-332 18 Taylor, J., Green, A.A. and Coil, G.T. (1948) J. Biol. Chem. 173, 591-604 19 El&ti, P. and SzSr6nyi, E. (1956) Acta Physiol. Acad. Sci. Hung. 9, 339-350 20 Fox, J.B. and Dandliker, W.B. (1956) J. Biol. Chem. 221, 1005-1017 21 Biszku, E., Boross, L. and Szabolcsi, G. (1964) Acta Physiol. Acad. Sci. Hung. 25, 161-167 22 Michel, G. and Beufler, H.O. (1974) in Methods in Enzymatic Analysis, 2nd Edn. (Bergmeyer, H.U., ed.), Vol. 3, pp. 1314-1319, Academic Press, New York 23 Salerno, C., and Ovfidi, J. (1982) FEBS Lett. 138, 270-272 24 OvAdi, J., Mohamod Osman, I.R. and Batke, J. (1982) Biochemistry 21, 6375-6382 25 V6rtessy, B. and Ov~di, J. (1987) Eur. J. Biochem. 164, 655-659