Aspartate transcarbamylase from the hyperthermophilic archaeon Pyrococcus abyssi Insights into cooperative and allosteric mechanisms Sigrid Van Boxstael 1 , Dominique Maes 2,3 and Raymond Cunin 1 1 Erfelijkheidsleer en Microbiologie, Vrije Universiteit Brussel, Belgium 2 Ultrastructuur, Vrije Universiteit Brussel, Belgium 3 Vlaams Interuniversitair Instituut voor Biotechnologie, Belgium Aspartate transcarbamylase (ATCase; EC 2.1.3.2) cata- lyses the condensation of carbamyl phosphate (CP) with the amino group of aspartate to form carbamyl aspartate and inorganic phosphate. In several organ- isms, this reaction is the first committed step of the de novo synthesis of pyrimidine nucleotides and, as such, it is subject to extensive control: transcriptional repres- sion of ATCase synthesis by the pyrimidines, and allosteric regulation of ATCase activity by nucleotide effectors. ATCase activity is present in almost all organisms from the three kingdoms of life, Bacteria, Eukarya and Archaea, under a variety of molecular forms. The best-studied ATCase is that from the bac- terium Escherichia coli, which has become a paradigm of cooperative and allosteric enzymes [1–5]. The discovery of microorganisms living in extreme conditions of, for example temperature and ⁄ or pH, prompted investigations into the structure and Keywords allostery; aspartate transcarbamylase; cooperativity; inhibition by analogues; Pyrococcus abyssi Correspondence R. Cunin, Erfelijkheidsleer en Microbiologie, Vrije Universiteit Brussel, Pleinlaan 2, 1050 Brussels, Belgium Fax: 32 2 629 1473 Tel: 32 2 629 1341 E-mail: rcunin@vub.ac.be (Received 15 December 2004, revised 1 March 2005, accepted 22 March 2005) doi:10.1111/j.1742-4658.2005.04678.x Aspartate transcarbamylase (ATCase) (EC 2.1.3.2) from the hyperthermo- philic archaeon Pyrococcus abyssi was purified from recombinant Escheri- chia coli cells. The enzyme has the molecular organization of class B microbial aspartate transcarbamylases whose prototype is the E. coli enzyme. P. abyssi ATCase is cooperative towards aspartate. Despite con- straints imposed by adaptation to high temperature, the transition between T- and R-states involves significant changes in the quaternary structure, which were detected by analytical ultracentrifugation. The enzyme is allos- terically regulated by ATP (activator) and by CTP and UTP (inhibitors). Nucleotide competition experiments showed that these effectors compete for the same sites. At least two regulatory properties distinguish P. abyssi ATCase from E. coli ATCase: (a) UTP by itself is an inhibitor; (b) whereas ATP and UTP act at millimolar concentrations, CTP inhibits at micro- molar concentrations, suggesting that in P. abyssi, inhibition by CTP is the major control of enzyme activity. While V max increased with temperature, cooperative and allosteric effects were little or not affected, showing that molecular adaptation to high temperature allows the flexibility required to form the appropriate networks of interactions. In contrast to the same enzyme in P. abyssi cellular extracts, the pure enzyme is inhibited by the carbamyl phosphate analogue phosphonacetate; this difference supports the idea that in native cells ATCase interacts with carbamyl phosphate synthe- tase to channel the highly thermolabile carbamyl phosphate. Abbreviations AEBSF, aminoethylbenzylfluoride; ATCase, aspartate transcarbamylase; CP, carbamyl phosphate; CPSase, carbamyl phosphate synthetase; CK, carbamate kinase; IPTG, isopropyl-thio-b-galactopyranoside; PAGE, polyacrylamide gel electrophoresis; PALA, N-phosphonacetyl- L- aspartate; SDS, sodium dodecylsulfate. 2670 FEBS Journal 272 (2005) 2670–2683 ª 2005 FEBS mechanisms of their ATCases. ATCase from Pyrococ- cus abyssi was the first archaeal hyperthermophilic ATCase to be characterized and expressed in recom- binant E. coli cells. It was found to be a cooperative and allosteric enzyme consisting of catalytic and regu- latory subunits. Recently, the crystal structure of the catalytic subunit was solved to a resolution of 1.8 A ˚ [6]. First attempts to purify this ATCase from either native or recombinant cells resulted in the irreversible loss of the regulatory properties, probably due to dis- sociation of the holoenzyme [7,8]. This study describes purification of the P. abyssi ATCase holoenzyme from recombinant E. coli cells without significant alteration of its regulatory properties, which allowed major questions regarding its mechanism and regulation of activity to be addressed. (a) Is the homotropic cooper- ativity toward aspartate of this enzyme and the attend- ant transition from a low affinity T-state to a higher affinity R-state accompanied by a significant structural change as in the case of the E. coli ATCase? (b) To what extent is cooperativity affected by temperature? (c) What is the mode of action of the allosteric effec- tors? Do activator and inhibitors bind competitively to the same sites on the regulatory subunits? Is there instead an additive effect of inhibitors, or even a syn- ergy as in E. coli? Furthermore, comparison of the properties of the pure enzyme with those observed in native P. abyssi cell-free extracts may reveal the influence of inter- actions with other cellular constituents such as other enzymes. Indeed, a channelling of the highly thermo- labile substrate CP between carbamyl phosphate syn- thase and ATCase has been reported for P. abyssi [9] as for other thermophilic and hyperthermophilic microorganisms [10,11]. Results Expression, purification and molecular mass determination P. abyssi pyrBI genes were expressed from the isopro- pyl-thio-b-galactopyranoside (IPTG)-inducible trc pro- moter on the pTrc99A plasmid transformed into the E. coli ATCase-deficient strain C600 ATC– . The enzyme was purified to homogeneity. The different purification stages, as analysed by sodium dodecyl sulfate–poly- acrylamide gel electrophoresis (SDS–PAGE), are shown in Fig. 1. Twenty grams of wet cells yielded 3 mg of pure ATCase holoenzyme. The molecular mass of the P. abyssi ATCase mole- cule was estimated to be 301 ± 15 kDa by calibrated gel filtration on Superdex 200 pg, using two XK 16 ⁄ 60 columns in series for greater accuracy. An independ- ent estimate of 304 ± 10 kDa for the molecular mass of the holoenzyme was obtained by analytical ultracentifugation. Taken together with the known structure of the trimeric catalytic subunit (c 3 ) [6], and the molecular masses for the catalytic and regu- latory polypeptides calculated from the sequence (34 879 and 16 947 Da, respectively), this value cor- responds to a [2(c 3 ):3(r 2 )] molecular architecture, typ- ical for class B ATCases. Thermal stability of activity The kinetic stability of the P. abyssi holoenzyme was measured by determining the residual activity at 37 °C after incubation for increasing periods at tem- peratures between 90 and 98 °C. The measured half- lives (s 0.5 ) are given in Table 1. For a comparison, the half-life of the isolated catalytic subunit was 80 min at 90 ° C and 5 min at 98 °C under the same conditions [6]. Fig. 1. Denaturing polyacrylamide gel, presenting successive purifi- cation stages of Pyrococcus abyssi ATCase. Lane 1: cell-free extract; lane 2: after the heat-purification step (20 min, 85 °C); lane 3: after anion-exchange chromatography (ResourceQ and MonoQ); lane 4: after size-exclusion chromatography (Superdex 200 pg); lane 5: molecular mass markers. Table 1. Thermostability of Pyrococcus abyssi ATCase. Half-life of activity (s 0.5 ) at different temperatures. T (°C) s 0.5 (min) 90 260 92 240 94 190 96 190 98 100 ATCase was at 100 lgÆmL )1 in 50 mM phosphate pH 7.5. S. Van Boxstael et al. Aspartate transcarbamylase from Pyrococcus abyssi FEBS Journal 272 (2005) 2670–2683 ª 2005 FEBS 2671 Saturation by the substrates Saturation by aspartate was measured at 30, 37, 45 and 55 °C, in the presence of saturating amounts of CP. The values obtained for the kinetic parameters are given in Table 2. Although the optimal growth tem- perature of P. abyssi is 96 °C under atmospheric pres- sure, 55 °C was the highest temperature at which complete saturation experiments were performed because of the high thermolability of the other sub- strate CP (a half-life of < 2 s at 100 °C, 4 min at 55 °C) [10]. The enzyme is stable for hours at that temperature. The saturation curves were sigmoidal at all temperatures, indicative of a cooperativity toward aspartate (Fig. 2). The curvature shown by the Eadie– Hofstee plot (Fig. 2) is characteristic of cooperative enzymes [12]. The Hill coefficient, taken as an index of cooperativity, did not vary significantly in the range of temperatures tested. V max increased sixfold with increasing temperature in the range tested. In contrast, temperature did not influence the affinity of the enzyme for aspartate as indicated by S 0.5 . As a result, the catalytic efficiency, expressed as V max ⁄ S 0.5 , also increased by about sixfold over the 30–55 °C range. For a comparison, the kinetic parameters of the isolated catalytic subunit measured at 55 °C are also given (Table 2, bottom line). Because of the thermolability of CP, saturation by this substrate was measured at 37 °C only. In the pres- ence of saturating aspartate (15 mm), the apparent K m for CP is 5 ± 2 lm and V max is 3.2 ± 0.3 mmolÆ h )1 Æmg )1 ATCase. The saturation curve was hyperbolic at low (1 mm) and saturating (15 mm) aspartate con- centrations (Fig. 3). Under the same conditions, E. coli ATCase shows cooperativity toward CP at high, but not at low, aspartate concentrations, a consequence of the cooperativity toward aspartate and of the ordered binding of the substrates, first CP then aspartate [13]. These results raised the possibility that binding of the substrates does not follow the same ordered mechan- ism as in the case of E. coli ATCase, especially as a loop (80s loop) which contains two residues inter- acting with the substrates is shorter in P. abyssi ATCase [6]. Order of substrate binding In an attempt to see whether the lack of apparent coop- erativity towards CP reflects a mode of substrate bind- ing different from that of E. coli ATCase, the inhibition patterns of N-phosphonacetyl-l-aspartate (PALA) toward CP and aspartate were analysed. PALA is a transition state analogue with chemical groups similar Table 2. Kinetic parameters of the Pyrococcus abyssi enzyme as a function of temperature. V max is the maximal velocity, S 0.5 the con- centration of aspartate at half the V max , n H the Hill coefficient and V max ⁄ S 0.5 the catalytic efficiency. Units (U) are mmoles carbamyl aspartate formed per hour and per mg ATCase. T (°C) V max (U) S 0.5 (mM) V max ⁄ S 0.5 n H P. abyssi holoenzyme 30 1.5 ± 0.3 2.6 ± 0.3 0.6 1.5 ± 0.1 37 2.9 ± 0.5 3.0 ± 0.5 1.0 1.6 ± 0.2 45 5.5 ± 0.7 2.5 ± 0.3 2.2 1.6 ± 0.2 55 9.9 ± 1.5 2.7 ± 0.4 3.7 1.7 ± 0.2 P. abyssi catalytic subunit 55 71.0 ± 4.0 19.7 ± 1.5 3.6 1.0 ± 0.1 Fig. 2. Saturation of Pyrococcus abyssi ATCase by aspartate at 55 °C(5m M CP, 50 mM Tris ⁄ HCl pH 8.0). (Inset: corresponding Eadie–Hofstee plot.) Fig. 3. Saturation of Pyrococcus abyssi ATCase by CP in the pres- ence of a saturating aspartate concentration (15 m M aspartate, 50 m M Tris-acetate pH 8.0). (Inset: corresponding Eadie–Hofstee plot.) Aspartate transcarbamylase from Pyrococcus abyssi S. Van Boxstael et al. 2672 FEBS Journal 272 (2005) 2670–2683 ª 2005 FEBS to those of the two substrates [14]. CP and aspartate saturations were performed in the presence of various concentrations of PALA. The results, presented as dou- ble reciprocal plots (Fig. 4A,B), showed that PALA behaves as a competitive inhibitor toward CP and as a noncompetitive inhibitor toward aspartate. If the bind- ing were random, PALA would behave as a linear non- competitive inhibitor toward both substrates. The results clearly indicate an ordered binding mechanism, first CP, then aspartate, and the lack of cooperativity toward CP at high aspartate concentration cannot be explained by a random binding mechanism. Effects of inhibitors PALA In the presence of a low aspartate concentration, PALA stimulates the activity of E. coli ATCase by promoting transition from the low-affinity T-state to the high-affinity R-state. This reflects cooperativity: while the inhibitor blocks the sites which it occupies, it converts the remaining sites to the R-state, resulting in an increase in the activity [14–16]. The effect of PALA on P. abyssi ATCase was tested. Figure 5 shows that the enzyme is stimulated by PALA at low aspartate concentrations. The amplitude of the activation decrea- ses as the aspartate concentration increases: 65 ± 15% at 0.1 mm aspartate; 20 ± 5% at 0.5 mm aspartate. At 3 mm aspartate, only direct inhibition by PALA can be observed. These results confirm the existence of a cooperative mechanism of aspartate binding. Phosphonacetate Native P. abyssi ATCase in crude extracts was repor- ted to be insensitive to the CP analogues phosphonace- tate and pyrophosphate [17]. These molecules are competitive inhibitors of CP in E. coli ATCase [18]. It was suggested that the P. abyssi CP binding site is shielded to some extent from the bulk solvent and that CP may be sequestered by a complex [17]. The isolated recombinant holoenzyme was tested for sensitivity to the CP analogue phosphonacetate. The activity was inhibited almost completely at 90 mm phosphonacetate (an E 50 value of 18 mm in the presence of 0.05 mm CP compared with 35 mm in the presence of 0.5 mm CP). Sensitivity to the inhibitor increased with decreasing CP concentration, suggesting that phosphonacetate acts in competition with CP. The observed sensitivity of the isolated recombinant ATCase to phosphonace- tate contrasts with the insensitivity of the enzyme in native P. abyssi ATCase extracts. Fig. 4. (A) Inhibition by PALA towards CP. Double reciprocal plot: no PALA (j), 0.25 l M PALA (d), 0.5 lM PALA (m)and1lM PALA (.)(37°C, 15 m M aspartate, 50 mM Tris-acetate pH 8.0). (B) Inhibi- tion by PALA towards aspartate. Double reciprocal plot: no PALA (j), 1 l M PALA (.), 5 lM PALA (d)and10lM PALA (m) (37 °C, 5m M CP, 50 mM Tris-acetate pH 8.0). Fig. 5. Saturation of Pyrococcus abyssi ATCase with PALA in the presence of different aspartate concentrations: (j) 0.1 m M,(n) 0.5 m M and (d)3mM (37 °C, 0.5 mM CP, 50 mM Tris ⁄ HCl pH 8.0). Relative activity defined as A ⁄ A 0 *100; where A is the activity in the presence of PALA and A 0 the activity in the absence of PALA. S. Van Boxstael et al. Aspartate transcarbamylase from Pyrococcus abyssi FEBS Journal 272 (2005) 2670–2683 ª 2005 FEBS 2673 Sedimentation velocity experiments in the presence or the absence of PALA Titration with PALA supports the existence of T- and R-states with a PALA- or aspartate-induced transition. However, this does not necessarily reflect the existence of a major difference in the quaternary structure between the T- and R-states. In E. coli ATCase, the quaternary structure transition involves a conforma- tional change of such amplitude that it can be detected by sedimentation studies [19–21]. Sedimentation velo- city experiments by analytical ultracentrifugation were performed on P. abyssi ATCase both unliganded and liganded with saturating concentrations of PALA and the relative difference in sedimentation coefficient Ds ⁄ s 0 , induced by the binding of PALA was calculated (Table 3). For comparison, similar experiments were performed with E. coli ATCase. PALA binding to P. abyssi ATCase results in a 2.6% relative decrease of the sedimentation coeffi- cient. As expected, the respective effects of 0.5 and 1mm PALA (both saturating) on the sedimentation coefficient are identical. The relative difference in sedimentation coefficient of P. abyssi ATCase is iden- tical to that observed for E. coli ATCase. Thus, PALA binding to P. abyssi ATCase induces swelling of the enzyme indicative of a significant conforma- tional change. Allosteric regulation Influence of the nucleotide effectors on aspartate saturation Purified recombinant enzyme is inhibited by CTP and UTP and activated by ATP. Aspartate saturation of the P. abyssi holoenzyme in the presence of saturating amounts of nucleotides was studied at 37 and 55 °C (Fig. 6A,B). The values of the kinetic parameters are given in Table 4. The nucleotides have a pronounced effect on the affinity for aspartate. ATP decreases the [S 0.5 ] Asp four- to fivefold and CTP increases it two- to threefold. The effect of UTP on the [S 0.5 ] Asp is less pronounced. ATP reduces significantly the Hill coefficient. This is similar to what is observed for E. coli ATCase. How- ever, an opposite effect of CTP and UTP is not observed. CTP and UTP do not affect the maximal activity. Remarkably, ATP provokes an increase of V max of % 35%. In E. coli ATCase no such effect of ATP on V max is observed. Saturation by the nucleotide effectors Saturation by the nucleotide effectors was studied at 1.5 mm aspartate, a concentration corresponding to Table 3. Effect of PALA on the sedimentation coefficients of Escherichia coli and Pyrococcus abyssi ATCase. N is the number of measurements, s 0 is the sedimentation coefficient of the unligan- ded enzyme. Ds is defined as the difference between the sedi- mentation coefficient of the unliganded enzyme (s 0 ) and the sedimentation coefficient of the PALA-liganded enzyme. SEM values are given. PALA (m M) N s (sedimentation coefficient) Ds ⁄ s 0 (%) P. abyssi ATCase 0 5 10.23 ± 0.05 0.5 2 9.97 ± 0.07 )2.6 ± 0.8 1 2 9.95 ± 0.13 )2.7 ± 1.4 E. coli ATCase 0 6 11.19 ± 0.03 0.3 4 10.89 ± 0.04 )2.7 ± 0.4 Fig. 6. Saturation of Pyrococcus abyssi ATCase by aspartate at (A) 37 °C and (B) 55 °C(5m M CP, 50 mM Tris ⁄ HCl pH 8.0) in the pres- ence of nucleotide effectors: no effectors (j), 2 m M ATP (d), 5m M UTP (,) and 0.3 mM CTP (m). Aspartate transcarbamylase from Pyrococcus abyssi S. Van Boxstael et al. 2674 FEBS Journal 272 (2005) 2670–2683 ª 2005 FEBS half the [S 0.5 ] Asp and in the presence of saturating CP. These are standard conditions to observe the maximal amplitude of effector action. Saturations were performed at 37 and 55 °C. At both tempera- tures, ATP increased ATCase activity up to 370%, whereas CTP and UTP inhibited ATCase activity, respectively, by $ 50 and 40% (Table 5). The con- centrations of nucleotides at which half the maximal effect is observed (E NTP 50 ) were slightly lower at 37 °C than at 55 °C, which is in agreement with the nucleotide interactions with the protein being mainly ionic or polar. Such interactions are weakened when the temperature is increased. However, the maximal amplitudes of the effects were affected little by tem- perature. Remarkably, the CTP concentration at which half the maximal inhibition is observed (E CTP 50 ) is two orders of magnitude lower than the E UTP 50 or the E ATP 50 . Increasing the CTP concentration to 5 mm did not increase the amplitude of the inhibition, indi- cating that if P. abyssi ATCase contains two differ- ent classes of nucleotide binding sites, like E. coli ATCase [22,23], the difference in their binding con- stants is too small to be detected by measurements of enzyme activity. Nucleotide effectors competition In E. coli ATCase, the activator ATP and inhibitor CTP bind competitively to the same sites on the regu- latory subunits. UTP, which has little effect by itself, is a synergistic inhibitor with CTP [24], an effect which can be mostly ascribed to a positive interaction between nucleotide binding sites in a regulatory dimer [22]. Nucleotide competition experiments were per- formed on P. abyssi ATCase to determine whether ATP, CTP and UTP have additive, antagonistic or synergistic effects on the catalytic activity. The princi- ple of these experiments is to determine if, in the pres- ence of a fixed amount of effector A, the addition of increasing concentrations of effector B can suppress the response to effector A and lead to the response observed when only effector B is present. First, the two inhibitors CTP and UTP were tested separately against the activator ATP, they were then tested against each other. CTP versus ATP Saturation by CTP was performed in the presence or absence of 0.2 mm ATP (Fig. 7A). In the presence of ATP, 10 lm CTP is required to reduce the relative activity from 280 to 100%. A further increase in CTP concentration reduces the activity to the same inhibited level as in the absence of ATP. This experiment shows an antagonistic effect of CTP on the activation of ATCase by ATP. In a reverse experiment, ATP saturation was studied in the presence or absence of CTP (Fig. 7B). In the presence of 2 lm CTP, the relative activity was 70%; 0.1 mm ATP was required to increase the relative activity from 70 to 100%. A further increase in ATP concentration brought the activity to the same activa- ted level as in the absence of CTP, thus ATP counter- acts completely the effect of CTP. Taken together, these results demonstrate that ATP and CTP have ant- agonistic effects on activity. UTP versus ATP At 0.1 mm, ATP elicited a 110% activation of the ATCase activity. Increasing the UTP concentration from 0 to 5 mm resulted in a 50% inhibition of the activity. This is the maximal inhibition level by UTP Table 4. Effect of nucleotides on the kinetic properties of the ATCase. V max is the maximal velocity; S 0.5 the concentration at half saturation and n H the Hill coefficient. Activities were determined at 5m M CP, 50 mM Tris ⁄ HCl pH 8.0. Units (U) are mmoles carbamyl aspartate formed per hour and per mg ATCase. V max (U) S 0.5 (mM) n H 37 °C no effector 2.9 ± 0.5 3.0 ± 0.5 1.6 ± 0.2 +2 mm ATP 4.5 ± 0.3 0.7 ± 0.2 1.1 ± 0.1 +0.3 mm CTP 2.6 ± 0.4 7.0 ± 0.3 1.3 ± 0.2 +5 mm UTP 2.4 ± 0.5 5.0 ± 0.4 1.4 ± 0.2 55 °C no effector 9.9 ± 1.0 2.5 ± 0.5 1.6 ± 0.1 +2 mm ATP 12.5 ± 1.0 0.5 ± 0.3 1.1 ± 0.2 +0.3 mm CTP 9.0 ± 0.8 7.5 ± 0.5 1.5 ± 0.1 +5 m M UTP 8.5 ± 1.0 4.0 ± 0.5 1.4 ± 0.2 Table 5. Effect of allosteric effectors on the activity of the ATCase at 37 and 55 °C. Activities were determined at 1.5 m M aspartate, 50 m M Tris ⁄ HCl pH 8.0, 5 mM CP. One hundred per cent is the activity in the absence of nucleotide effectors. Relative activity defined as A ⁄ A 0 *100; where A is the activity in the presence of the effector and A 0 the activity in its absence. E 50 is the concentra- tion of effector at half maximal effect. Nucleotides Relative activity (%) 37 °C E 50 (mM) 37 °C Relative activity (%) 55 °C E 50 (mM) 55 °C ATP 370 ± 60 0.20 ± 0.10 350 ± 50 0.25 ± 0.10 CTP 55 ± 10 0.002 ± 0.001 50 ± 10 0.003 ± 0.001 UTP 60 ± 10 0.30 ± 0.20 65 ± 10 0.60 ± 0.20 S. Van Boxstael et al. Aspartate transcarbamylase from Pyrococcus abyssi FEBS Journal 272 (2005) 2670–2683 ª 2005 FEBS 2675 alone (Fig. 8). The reverse experiment, saturation by ATP in the presence of 0.5 mm UTP (not shown) confirmed that the effects ATP and UTP are also antagonistic. CTP versus UTP At 0.5 lm concentration, CTP elicits 25% inhibition (Fig. 9). Increasing the UTP concentration from 0 to 5mm resulted in 50% inhibition of the enzyme, the maximal inhibition caused by UTP alone (Fig. 9), showing that the effects of UTP and CTP are antag- onistic. The reverse experiment, saturation by CTP in the presence of 0.5 mm UTP confirmed the com- petitive inhibitory effects of CTP and UTP (not shown). Taken together, these results show that the effects of the nucleotide effectors are neither additive nor syner- gistic but antagonistic. CTP and ATP act in competi- tion with each other, as do UTP and ATP. This suggests that they bind competitively to the regulatory sites. In this case, UTP and CTP are expected to inhi- bit in competition with each other too, which is what is observed, thereby confirming that the nucleotides bind to the same regulatory sites. Webb’s formalism allows us to distinguish quantita- tively among antagonism, additivity and synergism of inhibitor effects [25]. Here the inhibition observed when both inhibitors are present is smaller than the sum of the inhibitions elicited by each inhibitor indi- vidually minus their product (i 1,2 < i 1 + i 2 ) i 1 * i 2 ): 0.54 < 0.55 +0.60 ) 0.33 (values taken from Table 5 and Figs 7 and 9). This demonstrates the competition between the inhibitors. Fig. 7. Effect of the simultaneous presence of CTP and ATP. (A) Saturation by CTP alone (s), saturation by CTP in the presence of 0.2 m M ATP (j) (1.5 mM aspartate, 5 mM CP, 50 mM Tris ⁄ HCl pH 8.0, 37 °C). (B) Saturation by ATP alone (s), saturation by ATP in the presence of 2 l M CTP (j)(1.5mM aspartate, 5 mM CP, 50 m M Tris ⁄ HCl pH 8.0, 37 °C). Fig. 8. Effect of simultaneous presence of UTP and ATP. Saturation by UTP alone (s), saturation by UTP in the presence of 0.1 m M ATP (j) (1.5 mM aspartate, 5 mM CP, 50 mM Tris ⁄ HCl pH 8.0, 37 °C). Fig. 9. Effect of simultaneous presence of UTP and CTP. Saturation by UTP alone (s), saturation by UTP in the presence of 0.5 l M CTP (j)(2m M aspartate, 5 mM CP, 50 mM Tris ⁄ HCl pH 8.0, 37 °C). Aspartate transcarbamylase from Pyrococcus abyssi S. Van Boxstael et al. 2676 FEBS Journal 272 (2005) 2670–2683 ª 2005 FEBS Discussion Pyrococcus abyssi ATCase is a class B ATCase: evolutionary implications Calibrated size-exclusion chromatography and sedi- mentation equilibrium experiments give similar molecular mass estimates for the holoenzyme of 301 ± 15 and 304 ± 10 kDa. Taken together with the known trimeric [c 3 ] structure of the catalytic subunit [6], and with the molecular masses for the catalytic and regulatory polypeptides calculated from the sequence (34.9 and 17.0 kDa, respectively), these results are consistent with a [2(c 3 ):3(r 2 )] molecular architecture, typical of class B ATCases, whose proto- type is E. coli ATCase. Prokaryotic ATCases fall into three classes, A, B and C, according to their molecular mass and molecu- lar organization [26,27]. The limited number of Arch- aea for which a pyrB (catalytic chain) sequence is available all have a matching pyrI gene (coding for a regulatory chain). Thus far, a class B architecture appears characteristic of archaeal ATCases [28]. On phylogenetic grounds, two families of ATCases, ATC I and ATC II, have been recognized, which would both have been present in the last universal common ances- tor and later inherited differently in the ancestors of present-day organisms [29,30]. All class B ATCases form a clade in the ATC II family [31], and a coevolu- tion scheme of the pyrB and the pyrI genes, in response to a need for the conservation of the inter- actions between their polypeptide products in the holo- enzyme was proposed. This study supports this hypothesis and suggests that for archaeal hyperthermo- philic ATCases, one of the constraints may have been adaptation to high temperature: indeed, where tested, the association of the catalytic subunits with regulatory subunits is a major factor in thermostability [6,32]. The association between catalytic and regulatory subunits also imposes strong conformational con- straints on the catalytic sites. At 55 °C, the maximal observed activity of the holoenzyme is sevenfold lower than that of the catalytic subunit (Table 2). These con- straints are, in part, responsible for cooperativity. A similar phenomenon is observed with Sulfolobus acido- caldarius ATCase – a threefold difference [32], and E. coli ATCase – a twofold difference. Cooperativity towards substrates P. abyssi ATCase is cooperative toward aspartate. The cooperativity, expressed by the Hill coefficient, is less pronounced than in the case of E. coli ATCase (1.7 compared with 2.2). This lower cooperativity might result from changes imposed by adaptation to tem- perature, but also from a slightly different folding of the enzyme in the recombinant E. coli host. Indeed, a Hill coefficient of 2.2 was calculated in native cell extracts [17], where an association with the carbamate kinase-like CP synthetase (CK-like CPSase) involved in the channelling of CP might assist the folding of ATCase and ⁄ or affect its cooperative behaviour. How- ever, the high intrinsic thermostability of the recom- binant enzyme strongly suggests a correct folding in the mesophilic host and, besides, cooperativity toward aspartate was found to vary as much between ATCases from different mesophilic enterobacterial spe- cies [27] as between native and recombinant P. abyssi ATCases. The cooperativity of P. abyssi ATCase appears little or not affected by temperature, at least in the range 30–55 °C. The half-saturating aspartate concentration also showed very little variation, whereas maximal velocity increased sixfold in this same temperature range. Thus, increasing temperature increases the rate of the reaction without much affecting homotropic interactions. That cooperativity reflects the existence of T- and R-states characterized by active sites with different affinity and ⁄ or catalytic velocity is clearly indicated by the stimulation of activity by PALA at subsaturating aspartate concentrations: while this bisubstrate ana- logue and inhibitor blocks the sites which it occupies, it converts the remaining sites to the R-state. At aspar- tate concentrations higher than [S 0.5 ] Asp (3 mm), PALA behaved simply as an inhibitor. The concentration range in which P. abyssi ATCase is activated by PALA is much broader and lower than that in which E. coli ATCase is activated by PALA (between 10 )10 and 10 )8 m PALA instead of between 1 and 8 lm for E. coli ATCase). This is in agreement with the higher affinity of P. abyssi ATCase for CP compared with E. coli ATCase (a K m of 5 ± 2 lm instead of 600 lm). Our study makes clear that activation by PALA, and thus cooperativity exists even in the absence of the CK-like CPSase. The question arises whether because of specific con- straints imposed by adaptation to high temperature, presumably an increased rigidity, the T- and R-states deduced from the kinetic data correspond to different quaternary states and if a global quaternary transition with an amplitude comparable with that observed in the case of E. coli ATCase effectively occurs with the hyperthermophilic enzyme. E. coli ATCase undergoes major structural rearrangements during the T-to-R transition which result in a global expansion of the S. Van Boxstael et al. Aspartate transcarbamylase from Pyrococcus abyssi FEBS Journal 272 (2005) 2670–2683 ª 2005 FEBS 2677 molecule, as documented by X-ray crystallography studies [5,33], small-angle X-ray scattering [34–36] and sedimentation studies [20,37]. The sedimentation velo- city experiments performed in this study on both P. abyssi and E. coli ATCases showed that the binding of saturating amounts of PALA induced similar decreases in their sedimentation coefficients (Ds ⁄ s o of 2.7%), showing that P. abyssi ATCase undergoes a conformational change – an increase in volume – of an amplitude comparable with that of the E. coli enzyme. It is worth noting that our results for E. coli ATCase are in quantitative agreement with those recently pub- lished by Schachman’s laboratory, a decrease of 2.9 and 2.6% in the presence of saturating PALA concen- trations [19,21]. The existence of structurally distinct T- and R-states implies the existence of interactions stabilizing these states. Alignment of the P. abyssi ATCase amino acid sequence with that of the E. coli enzyme shows that the residues involved in polar interactions which are of major importance to stabilize the T- and R-states of E. coli ATCase [4] are nearly all conserved, with the exception of rLys 143 which makes a crucial salt link with cAsp 236 in the E. coli C4–R1 interface. However, rLys 143 is replaced by an arginine which makes the for- mation of a salt link possible. This conservation sug- gests that similar polar interactions stabilize the T- and R-state of P. abyssi ATCase. In contrast with E. coli ATCase, P. abyssi ATCase shows no cooperativity toward CP. The CP coopera- tivity of the E. coli enzyme is apparent and reflects the ordered binding of substrates, first CP, then aspartate, and the cooperativity toward the second substrate [22]. A shorter 80S loop in P. abyssi ATCase, missing one of the residues between a serine and a lysine which make interactions with both substrates, might affect the mechanism of substrate binding. Competitions with PALA showed, however, that the shorter loop does not affect the ordered binding of the substrates. The cause of the lack of apparent cooperativity toward CP must therefore be looked for elsewhere. The lack of cooperativity for CP of the pure P. abyssi ATCase contrasts significantly with the cooperativity observed at both low and high aspar- tate concentrations in determinations performed on native crude extracts [17]. A possible explanation for this discrepancy could be an interaction of ATCase with one or more proteins present in the crude P. abyssi extract, linked to the channelling of CP. This would correlate with the different response to the CP analogue phosphonacetate of the pure ATCase and the ATCase in native extracts. Another, methodological explanation could be that given the high affinity for CP, a ‘false’ cooperativity was observed in the CP saturations performed on crude extracts: due to CP exhaustion in the lower range of concentrations, initial velocity conditions would not have been obtained [38]. Allosteric regulation by nucleotide effectors P. abyssi ATCase is activated by ATP (270%) and inhibited by CTP and UTP (50 and 40%, respectively). The amplitudes of the responses to the different allo- steric effectors do not change between 37 and 55 °C. Competition experiments showed that the nucleotide effectors bind competitively to the same regulatory sites. Their effects on ATCase activity are neither additive nor synergistic, but antagonistic. Remarkably, CTP inhibition is already maximal in the micromolar range, whereas the effects of ATP and UTP reach their maximum in the millimolar range. This suggests that CTP is the major physiological regulator of P. abyssi ATCase activity. In E. coli ATCase, all nucleotides act in the millimolar range. Alignment of the P. abyssi and E. coli regulatory chains shows that only two of the residues involved in interactions with ATP and CTP in E. coli ATCase are not conserved in P. abyssi, whereas 13 are conserved [39]. The two nonconserved residues are involved specifically in ATP binding. It can thus be suggested that the high affinity of P. abyssi ATCase for CTP requires extra interactions with CTP, which do not occur in E. coli ATCase. Class B ATCases exhibit a varied pattern of responses to nucleotide effectors (for a review of mesophilic enzymes, see Wild and Wales [27]). The few archaeal ATCases studied so far also show diverse allosteric regulatory patterns: for instance, the ATCase of S. acidocaldarius is activated by the four nucleoside triphosphates ATP, GTP, CTP and UTP [32]. Single- residue changes or the modification of discrete secon- dary structure elements can dramatically affect the allosteric response of ATCase, showing that the latter depends on very subtle networks of intramolecular interactions [40–43]. A response similar to that of P. abyssi ATCase is found in the mesophile Yersinia intermedia [27]. Clearly, adaptation to high tempera- ture has little or no impact on the patterns of allosteric response of ATCases. Although little is understood about the constraints which have led to the acquisition and conservation of specific patterns of regulation, it should be con- sidered that under physiological conditions, class B ATCases are liganded by nucleotides, and that, because ATP is a general activator [27], the compet- itive binding of other nucleotides, inhibitors or less Aspartate transcarbamylase from Pyrococcus abyssi S. Van Boxstael et al. 2678 FEBS Journal 272 (2005) 2670–2683 ª 2005 FEBS efficient activators than ATP, appears as a conserved mechanism of modulation of catalytic activity. Several models have been proposed to explain the molecular mechanism of effector action [5,44,45]. Glo- bally, effectors appear to modulate the stability of interfaces between domains and between subunits, thereby facilitating (activators) or hindering (inhibi- tors) the quaternary structure transition from T- to R-state. Different properties of the pure ATCase and ATCase in native extracts may reflect the channelling of the thermolabile substrate CP At the optimal growth temperature of P. abyssi (96 °C), CP turns out to be very unstable with a half-life < 2 s [10]. The decomposition of CP leads to the accumulation of toxic amounts of cyanate, a powerful and indiscriminate carbamylating agent. Metabolic channelling, a process by which the prod- uct of an enzyme is directly transferred to the next enzyme in the pathway without being released in the bulk solvent, could provide a means to minimize the thermal decomposition of CP at high temperature. Indeed, isotopic dilution experiments and coupled reaction kinetics with P. abyssi extracts showed the existence of an imperfect channelling between the CK-like CPSase and ATCase, although no physical interaction between the two enzymes could be dem- onstrated [9]. The lack of sensitivity of native P. abyssi ATCase in cellular extracts to phosphon- acetate, a CP analogue and an inhibitor of E. coli ATCase, was proposed to result from a shielding of the CP binding site from the bulk solvent by an interaction with another cellular component, possibly CPSase [17]. The observation that purified P. abyssi ATCase is normally sensitive to phosphonacetate supports this hypothesis. It should be mentioned that a channelling of CP was also observed between the CPSase and the ornithine transcarbamylase of Pyrococcus furiosus [10] and that a physical inter- action between these enzymes could be demon- strated [11]. Experimental procedures Strain and plasmids The E. coli host strain was C600 ATC À (F ) , supE44, hsdR, endA, thi, D(lac-proAB), D pyrB) (Microbiology, VUB). The expression vector was pTrc99A (Amersham Pharmacia Bio- tech, Ghent, Belgium). pSJS 1240 [46] was a gift from S.J. Sandler. Chemicals and enzymes CP (lithium salt), l-aspartate, CTP (sodium salt), UTP (sodium salt), IPTG, ammonium sulfate, zinc acetate, anti- pyrine and diacetylmonoxime were purchased from Sigma (Bornem, Belgium). Sulfuric acid was purchased from Pan- Reac. ATP sodium salt was purchased from Boehringer- Mannheim (Brussels, Belgium). Tris-base was purchased from Invitrogen (Bruges, Belgium). Leupeptine and amino- ethylbenzylfluoride (AEBSF) were purchased from ICN. PALA was obtained from the Drug Synthesis and Chem- istry Branch, Developmental Therapeutics Program, Divi- sion of Cancer Treatment and Diagnosis, National Cancer Institute (Bethesda, MD, USA). Restriction enzymes were from Amersham Pharmacia Biotech. Cloning and expression pyrBI genes from P. abyssi were amplified by PCR using primers designed to generate a NcoI restriction site coinci- ding with the ATG initiation codon of pyrB and a BamHI site after the end of pyrI. The PCR product was cloned in the pTrc99A vector and transformed into competent cells of the ATCase-deficient strain C600 ATC . In order to improve expression, the cells were cotransformed with the pSJS1240-vector [46]. The P. abyssi pyrBI genes were expressed from the IPTG-inducible trc promotor of pTrc99A. Cells were grown in a 12 L Biolafitte fermentor, in rich 853 medium [47] containing 50 lgÆmL )1 ampicilline and 50 lgÆmL )1 spectinomycine. IPTG was added to a final concentration of 2 mm at A 600 ¼ 2. Cells were harvested at A 600 ¼ 6 and kept at )80 °C. Purification of P. abyssi ATCase holoenzyme Twenty grams of cells were resuspended in 100 mL 100 mm Tris ⁄ HCl, pH 8.2 containing the protease inhibitors leu- peptine (1 lgÆmL )1 ) and AEBSF (100 lgÆmL )1 ). Cells were lysed by cooled sonication in a Heat System Branson soni- cator (model W-225R; Fisher Block, Kortrijk, Belgium) or 30 min. The cell extract was centrifuged (30 min, 10 000 g). Heat denaturation The supernatant was incubated for 20 min at 85 °C after which it was cooled on ice. In order to remove denatured proteins, the protein solution was centrifuged (10 min, 10 000 g). First anion-exchange chromatography The supernatant was applied on an Amersham Pharmacia Biotech 26 ⁄ 60 column packed with Source 15Q medium. The column was equilibrated with 20 mm Tris ⁄ HCl (pH 8.2), 2 mm b-mercaptoethanol, 0.1 mm zinc acetate. S. Van Boxstael et al. Aspartate transcarbamylase from Pyrococcus abyssi FEBS Journal 272 (2005) 2670–2683 ª 2005 FEBS 2679 [...]... Khan S & Maes D (2003) Aspartate transcarbamylase from the hyperthermophilic ˚ archaeon Pyrococcus abyssi: thermostability and 1.8 A resolution crystal structure of the catalytic subunit complexed with the bisubstrate analogue N-phopsphonacetyl-l -aspartate J Mol Biol 326, 203–216 7 Purcarea C (1995) Enzymes involved in carbamoyl phosphate metabolism from the hyperthermophilic and barophilic marine archaebacterium... Escherichia coli aspartate transcarbamylase: the relations between structure and function Science 241, 669–674 4 Lipscomb W (1994) Aspartate transcarbamylase from Escherichia coli: activity and regulation Adv Enzymol 68, 67–152 5 Lipscomb WN (1995) Activity and allosteric regulation in mammalian fructose-1,6-biphosphatase and E coli aspartate transcarbamylase Biopolymers and Bioproducts, Proceedings of the 11th... scans taken at 5 h intervals) The equilibrated scans were analysed with the same Beckman software and the mass was determined Aspartate transcarbamylase from Pyrococcus abyssi 9 Acknowledgements This work was supported by the Flemish Science Foundation (FWO, grant G 0448.99), by the Flanders Interuniversity Institute for Biotechnology (VIB) and by the Research Council (OZR) of the Vrije Universiteit Brussel... Evans DR & Herve G (1999) Channeling of carbamoyl phosphate to the pyrimidine and arginine biosynthetic pathways in the deep sea hyperthermophilic Archaeon Pyrococcus abyssi J Biol Chem 274, 6122– 6129 ´ Legrain C, Demarez M, Glansdorff N & Pierard A (1995) Ammonia-dependent synthesis and metabolic channelling of carbamoyl phosphate in the hyperthermophilic archaeon Pyrococcus furiosus Microbiology 141,... within and near a helices in the catalytic chains of aspartate transcarbamoylase: effects on assembly, stability and function Protein Sci 10, 528–537 Howlett GJ & Schachman HK (1977) Allosteric regulation of aspartate transcarbamoylase Changes in the sedimentation coefficient promoted by the bisubstrate analogue N-(phosphonacetyl)-l -aspartate Biochemistry 16, 5077–5083 2681 Aspartate transcarbamylase from. .. independent experiments at 37 and 55 °C) Aspartate saturations in the presence of nucleotides were performed twice at both 37 and 55 °C The Hill coefficient was obtained by determination of the slope in the Hill plot: log(V ⁄ (Vmax ) V)) vs log( [aspartate] ) All other curves were fitted manually Analytical ultracentrifugation Determination of ATCase concentration The concentration of the pure holoenzyme was... thyroglobuline (660 kDa), ferritine (440 kDa), catalase (232 kDa), aldolase (158 kDa) and bovine serum albumin (67 kDa) were determined A standard curve was made with the logarithm of the molecular mass of the proteins in the y-axis and the elution volumes (Ve) in the x-axis Comparison of the elution 2680 Sedimentation velocity and sedimentation equilibrium experiments were performed in a Beckman Optima XL-A... specified otherwise, assays were performed in 50 mm Tris ⁄ HCl, pH 8.0 Aspartate saturations were performed in the presence of 5 mm CP at 30, 37, 45 and 55 °C, taking into account the large DpH ⁄ °C (0.28 ⁄ 10 °C) of Tris buffer For aspartate saturations at 45 and 55 °C, a blank value was subtracted for each different aspartate concentration to account for chemical carbamylation At 30 and 37 °C the chemical... Herve G (1994) The catalytic and regulatory properties of aspartate transcarbamoylase from Pyrococcus abyssi, a new deep-sea hyperthermophilic archaeobacterium Microbiology 140, 1967–1975 Porter RW, Modebe MO & Stark GR (1969) Aspartate transcarbamylase Kinetic studies of the catalytic subunit J Biol Chem 214, 1846–1859 Beernink PT, Yang YR, Graf R, King DS, Shah SS & Schachman HK (2001) Random circular... gratefully acknowledges the Flemish Institute for the Improvement of Scientific and Technological Research in Industry (IWT) for a specialization grant The authors are grateful to Tony Aerts (Biomedical Sciences, University of Antwerp) for performing the AUC experiments and to the Drug Synthesis and Chemistry Branch, Developmental Therapeutics Program, Division of Cancer Treatment and Diagnosis, National . Aspartate transcarbamylase from the hyperthermophilic archaeon Pyrococcus abyssi Insights into cooperative and allosteric mechanisms Sigrid. toward aspartate and of the ordered binding of the substrates, first CP then aspartate [13]. These results raised the possibility that binding of the substrates