Biol. Chem. Hoppe-Seyler Vol. 371, pp. 707-713, August 1990
On the Action of Carboxy Groups in the Citrate Synthase Reaction JürgenWiLDE, Ute LILL and Hermann EGGERER Institut für Physiologische Chemie der Technischen Universität München
(Received 4 May 1990)
Summary: The aza analogue (/?5')-3-hydroxy-2,5-pyrrolidinedione-3-acetic acid (6) of the five-membered citric anhydride (2) was prepared in the sequence citric acid —» 2-phenyl-l,3-dioxolan-4-one-5,5-diacetic acid (1) —> citric acid /3-amide (3) -> 6 and used to resolve ambiguities in the mechanism of the citrate synthase reaction. The results yield no indication for the formation of anhydride 2 on the enzyme and favour the direct hydrolysis of the intermediate (351)citryl-CoA. Ammonolysis of the dioxolanone 1 in the reaction sequence described above produced not only citric acid
/3-amide but also the -isomer. This is shown to originate in the transient formation of anhydride 2. Hydrolysis of the dioxolanone 1 under "physiological conditions" occurs via anhydride 2, generated in intramolecular bifunctional catalysis by a protonated and a deprotonated carboxyl group. The catalytic residue Asp375 of citrate synthase is considered to operate on the enzyme as does the protonated carboxyl group in the chemical reaction and to generate enolic acetyl-CoA in cooperative catalysis with His274. This reaction of Asp375 may also facilitate the hydrolysis of citryl-CoA.
Von der mechanistischen Sonde 3-Hydroxy-2,5-pyrrolidindion-3-essigsäure zu bifunktioneller Katalyse von Citrat-Synthase Zusammenfassung: Das Aza-Analoge (RS)-3-Hydroxy-2,5-pyrrolidindion-3-essigsäure (6) von fünfgliedrigem Citronensäureanhydrid (2) wurde in der Sequenz Citronensäure —» 2-Phenyl-l,3-dioxolan-4on-5,5-diessigsäure (1) —» Citronensäure ß-amid (3) —> 6 erhalten und zur Klärung mechanistischer Details der Citrat-Synthase-Reaktion verwendet. Die Ergebnisse liefern keinen Hinweis für die enzymatische Bildung des Anhydrids 2 und begünstigen die direkte Hydrolyse des Zwischenprodukts (3S)-CitrylCoA. Die Ammonolyse des Dioxolanons l in der voranstehend beschriebenen Reaktionsfolge ergab nicht nur
Citronensäure-ß-amid, sondern auch das -Isomere. Es entsteht aus transient gebildetem Anhydrid 2. Die Hydrolyse des Dioxolanons l unter „physiologischen Bedingungen" erfolgt über das Anhydrid 2 in intramolekularer bifunktioneller Katalyse mittels einer protonierten und einer deprotonierten Carboxylgruppe. Die Wirkung der protonierten Carboxylgruppe in der chemischen Reaktion liefert ein Modell für die Funktion der katalytischen Gruppe Asp375 der Citrat-Synthase. In bifunktioneller Katalyse mit His274 kann Asp375 enolisches Acetyl-Co A bilden und die Hydrolyse von Citryl-CoA einleiten.
Enzymes: ATP citrate (pro-35)-lyase, ATPrcitrate oxaloacetate-lyase (EC 4.1.3.8); Citrate (pro-35)-lyase, citrate oxaloacetate-lyase ((pro-35)-CH2COf —> acetate) (EC 4.1.3.6); Citrate (si')-synthase, citrate oxaloacetate-lyase ((pro-35)-CH2COf -> acetyl-CoA) (EC 4.1.3.7). Abbreviations: BCHAsalt, bis(cyclohexylammonium) salt;TLC, thin-layer chromatography.
Brought to you by | Purdue University Libraries Copyright © by WalterAuthenticated de Gruyter & Co · Berlin · New York
Download Date | 6/1/15 3:38 AM
708
J.Wilde, U. Lill and H. Eggerer
Vol. 371 (1990)
Key words: Citrate synthase, bifunctional catalysis, citric anhydride, aza analogue.
Citrate synthase of the tricarboxylic acid cycle catalyses the formation of citrate from acetyl-CoA and oxaloacetate.The enzyme (pig heart) also transforms (3,S)-citryl-CoA to the substrates and products of the physiological reaction and, in the presence of (S)-malate in place of oxaloacetate, liberates a proton from the methyl group of acetyl-CoA^'.The Claisen type of condensation (cf. ref.'21 for definition) indicated by these results could not be considered proven initially because the rate of product formation from citryl-CoA was lower than the rate of the physiological reaction. Pursuit of this unexpected result revealed complicated inhibition kinetics'31 which were thoroughly investigated'4'81 and could be described quantitatively on basis of a nonproductive enzyme· citryl-CoA complex'91. The actual initial rate of citrylCo A hydrolysis satisfies the kinetic criterion for an intermediate; with a source variant of the enzyme the rate was even twofold faster than the rate of the physiological re action'81. The experimental results are now in complete harmony with the outlined mechanism. Characteristic traits of Claisen (condensation) enzymes are the requirement of a cosubstrate for proton exchange of the nucleophilic reactand with the solvent'1·101, and the stereochemical course of the reaction^1 14] (inversion of configuration at the methyl group of acetyl-Co A which becomes methyleneofthe pro-S carboxymethyl group of citrate), arguments considered in favour of a concerted reaction'2'15'161. The obligatory formation of a ternary enzyme-substrate complex'171 as well as the enzymic activities required for C-C condensation and thioester hydrolysis residing in a single site of citrate synthase'18·191, are consistent with this view. Actually, however, and in agreement with the behaviour of citryl-CoA as an intermediate, enzymic Claisen condensations, like the familiar aldol condensations, have been shown to proceed in a stepwise manner'15'161 (see ref.'151 for a lucid discussion on the two types of condensation reactions). In the light of these results a citryl-CoA-derived five-membered and hydrolysis-sensitive citric anhydride (3-carboxymethyl-3-hydroxysuccinic anhydride) gains renewed interest. Its formation from citryl-thioesters by intramolecular attack of the C-6 carboxylate anion at the thioester carbonyl occurs under "physiological conditions" and effects hydrolysis of the thioester group that is otherwise stable under these conditions'201. The same reactions may initiate the hydrolysis of citryl-CoA on the enzyme but so far no experimental evidence supports this viewlu -1J. Here we probe for a new approach.The af-
finity of citrate synthase to the anhydride, if used as a substrate, would be expected to be high only if the anhydride is formed in the course of the physiological reaction. In the same selective sense an analogue of the anhydride would be expected to inhibit the enzyme only if it were an intermediate analogue. 3-Hydroxy-2,5-pyrrolidinedione-3-acetic acid (6in Scheme 1 shown below) in which the central oxygen of citric anhydride is replaced by nitrogen would appear to represent a suitable probe because the NH group is inert and the amide structure stable under physiological conditions. We have, therefore, prepared this substituted succinimide derivative and determined its interaction with citrate synthase. The results are presented in this paper.
Experimental Elemental analyses were performed by H. Schulz (Mikroanalytisches Laboratorium) and IR spectra of solid samples measured by H. Huber (Physikalisch-Chemisches Laboratorium), both Organisch-Chemisches Laboratorium der Universität München. NMR spectra were executed by Drs. Zetl and Sonnenbichler at the Max-Planck-Institut für Biochemie, Martinsried. We thank them all for their help. General Solvents were dried over Na2SO4 and evaporated under reduced pressure in a rotary evaporator. Titrations were performed with phenolphthalein as indicator. Melting points were measured in capillary tubes (Büchi 510 apparatus) and are uncorrected. Acids were recovered from salts by passage through Dowex-50 (H®; 50100 mesh).Thin-layer chromatography (TLC) was executed on DC Plastikfolie Kieselgel 60/F254, Merck, in the solvent ethanol/conc. NH 3 /water = 7.8:1.25:0.95 (v/v). Acids were detected with ethanolic kresol green, pH 8; compounds containing the benzene ring were visualized at 254 nm and by reaction with ethanolic 10mM 2,4-dinitrophenylhydrazine containing 5mM HC1 and liberating benzaldehyde. IR spectra of liquid and solid samples were measured with a Perkin-Elmer Infrared Spectrometer 710 B, and in KBr pellets with a Perkin-Elmer Infrared Spectrophotometer Model 1420 Ratio Recording, respectively. -NMR spectra were taken at 500 MHz on a Bruker AM 500 Spectrometer and are recorded in parts per million downfield from internal tetramethylsilane; coupling constants are expressed in Hz. Preparation of compounds (RS)-2-Phenyl-l,3-dioxolan-4-one-5,5-diaceticacid (1): The previously used condensation agent P4Oi0[22) was replaced by boron trifluoridet23].To a solution of water-free citric acid (22.4 g; 0.12 mol) and freshly distilled benzaldehyde (10.1 m/; 0.1 mol) in 100 ml abs. dioxane was added at room temperature dropwise with stirring 14 ml of boron trifluoride etherate within 5 min. The immediately formed cherry-red and clear solution was stirred over night, then evaporated and the oily brownish residue dissolved in 200 ml ethyl acetate. This was extracted with 3 100 ml sat. KHCO3 solution and the aqueous phase, pH ca. 7.5, cooled with ice, acidified to pH 3 with OM HC1 and extracted four times with a total of 300 ml ethyl acetate. The combined extracts were dried, then evaporated to yield 1 (20.6 g; 74%) of m.p. 171-173 °C (decomp.). Recrystalliza-
Brought to you by | Purdue University Libraries Authenticated Download Date | 6/1/15 3:38 AM
Vol.371 (1990)
Bifunctional Catalysis of Citrate Synthase
709
tion from ethyl acetate/light petroleum (60-80°) gave 19.7 g (71 %) of m.p. 173-175 °C (decomp.).TLC in methanol as solvent: only one spot, Rf = 0.54, containing acidic and aldehydic (after treatment with acid) functional groups (benzaldehyde: Rf = 0.32). Ή-NMR (acetone-Do): δ = 3.04 (d, 2 J(H,H) = 16.8, I H · pro-R or pro-S C//2CO2H), 3.14 (d, 2/(H,H) = 16.8, 1H; pro-R or pro-S C//2C02H), 3~.12 (d, 2J(H,H) = 16.8, 1H; pro-S or pro-R C//2CO2H), 3.16 (d, 2J(H,H) = 16.8, 1H; pro-S or pro-R C//2C02H); IR(KBr): ν (cm'1) = 1805-1810 (C = O); UVabsorbance (ethanol): £26i = 0.26mM~' cm"1. The anhydride of compd. 1 was prepared as in ref.' 24 '. Ή-NMR (acetone-D6): δ = 3.50 (d, 2J = (H,H) = 17.4, 1H; pro-R or pro-S C/Y 2 -CO-O),3.63(d, 2 /(H,H) = 17.4,1H; pro-R or pro-S CH2-COO), 3.56 (dd, 2 /(H,H) = 17.4 and 47(H,H) - 2.4,1H; pro-S or pro- R C// r CO-O),3.76(dd, 2 /(H,H) = 17.4 and 47(H,H) - 2.4, !H;proS or pro-R CH2-CO-0).
(65%) needles, m.p. 170-172 °C (decomp.). A mixed m.p. with 4 isolated from the reaction of 1 with ammonia gave no depression. TLC: only one compound, Rf = 0.19, which was inseparable from 4 isolated as described above under a).The free amic acid 4 was obtained from the BCHA salt after treatment with Dowex-50 ( 1 x 7 cm) and addition of acetone to the desalted and evaporated solution. Needles (340 mg), m.p. 143 °C (decomp.); after recrystallization from 95% ethanol: 270 mg (71%) m.p. 144-146°C (decomp.). Equiv. weight found: 97; calculated 96;TLC: only one acid, Rf = 0.19. C6H906N (191.1) Calc.: C37.7 H 4.8 Ν 7.3 Found: C37.8 H 5.0 Ν 7.6 Ή-NMR (D2O): δ - 2.77 (d, 2./(H,H) = 14.8: C//2CO2H), 2.88 (d, 2 J(H,H) - 14.8, 1H; C//2CO2H); 2.87 (d, 27(H,H) = 16.2, 1H; C// 2 CONH 2 ), 3.09 (d, 2 /(H,H) = 16.2, 1H; Ctf 2 CONH 2 ).
Citric acid -amide (3): To a solution of 5.6 g (20 mmol) 1 in 25 ml ethanol was added 25 ml (ca. 400 mmol) cone, ammonia and the reaction mixture kept at room temperature over night. Precipitated crystals (benzaldehyde ammonia) were isolated by suction filtration and washed with aqueous ethanol.The elutate was evaporated for removal of ammonia and ethanol, and the aqueous phase extracted with ethyl acetate (3 x 10 m/) for removal of residual benzaldehyde. The desalted solution (Dowex-50; 2 x 10 cm) was titrated with cyclohexylamine (5 ml), then evaporated and the residue crystallized from water-ethanol to yield 4.6 g (57%) BCHA salt, needles, m.p. 186-188 °C (decomp.) (the mother liquor of this crystallization contained 4, isolated as described below). TLC yielded one acid spot, RF= 0.32 (citric acid: RF= 0.06). A sample after recrystallization gave m.p. 186-188 °C (decomp.). C 6 H 9 O 6 N-2C 6 H 13 N-H 2 0 (407.5) Calc.: C53.1 H 9.2 N 10.3 Found: C53.3 H 9.2 N 10.3
(RS)-3-Hydroxy-2,5-pyrrolidinedione-3-acetic acid (6): The solution of the diammonium salt 3 (255 mg, 1.33 mmol) and 8 mg 4-toluenesulfonic acid in 30 ml dimethylformamide was heated under nitrogen for 16 h to 100-110 °C until the liberation of ammonia had ceased. The solvent was evaporated at 50 °C (0.1 Torr) and the yellowish residue desalted (Dowex-50; 0.5 x 6 cm), then neutralized with cyclohexylamine (0.15 mi) and brought todryness. Crystallization of the oily residue from ethanol-light petroleum (60-80 °C) gave 315 mg (87%) needles, m.p. 170-172 °C (decomp.). TLC: only one spot, Rp — 0.54; after recrystallization of a sample: m.p. 170-172 °C (decomp.). C6H7N05-C6H13N (272.3) Calc.: C52.9 H 7.4 N 10.3 Found: C52.7 H7.7 N 10.0 Ή-NMR (free acid in acetone-Do): δ = 4.69 (d, 2/(H,H) = 18.2, 1H; ring C//2CONH), 5.12 (d, 2/(H,H) = 18.2, 1H; ring C//2CONH), 4.97 (d, 2/(H,H) = 16.8, 1H; C//2CO2H), 5.09 (d, 2 7(H,H) = 16.8, 1H; C//2CO2H). On treatment with alkali (!M NaOH, 37 °C, 120 min) 6 was completely hydrolysed to yield according toTLC the α-amide 4 (major product) and the /3-amide 3 (minor product); free citric acid was not detectable.
Diammonium salt: BCHA salt (1.63 g; 4 mmol) was dissolved in 15 ml water under gentle warming, then desalted (Dowex-50; 1 x 7 cm) and the effluent carefully evaporated.The oily residue was dissolved in 1.5 ml water plus 1 ml one. ammonia and crystallized at 0 °C by repeated addition of ethanol to commencing turbidity. The isolated diammonium salt, prisms (750 mg; 83%), m.p. 170-172 °C (decomp.), gave a single spot onTLC, Rf = 0.32. C 6 H 9 O f t N-2NH 3 (225.2) Calc.: C32.0 H 6.7 N 18.7 Found: C32.2 H 6.9 N 18.7 Ή-NMR (D20): δ = 2.50 (d, 2/(H,H) = 15.1, 2H; pro-R or pro-S C//2C02H), 2.70 (d, 2J = (H,H) = 15.1, 2H; pro-R or pro-S Cf/2CO2H). The free acid 3 crystallized on treatment of the oily residue described above with ether; needles, m.p. 99-101 °C (decomp.). Recrystallization was successful from ethanol only but not without esterification: prisms, m.p. 93-96 °C (decomp.) which according to titration, elemental analysis and NMR-spectroscopy contained an ethyl ester of 3 as an impurity. Citric acid a-amide (4): a) From the dioxolanone l:The mother liquor of the BCH Asalt isolation of 3 described above was evaporated and the residue crystallized from water-acetone to yield 2.94 g (35.0%) BCHA salt, needles, m.p. 170-172 °C (decomp.). TLC: only one spot, Rf = 0.19. C 6 H 9 O 6 N-2C 6 Hi 3 N-1.75H 2 O (403.0) Calc.: C53.7 H 9.1 N 10.4 Found: C53.8 H 9.1 N 10.3 b) From (R.S)-citryl-N-octanoylcysteamine 5.The suspension of 5'22' (K®, H®; 210 mg; 0.5 mmol) in 1 ml water was neutralized with 0.25 m/2M NH 3 ; cone, ammonia (1 m/, ca. 16 mmol) was added without delay and the mixture kept at room temperature for 3 h. Precipitated N-octanoylcysteamine was extracted with ether and the clear aqueous solution evaporated for removal of ammonia, then desalted (Dowex-50; 1 x 3 cm).and the effluent titrated with cyclohexylamine (130 μ/). Isolation as indicated above gave 130 mg
Attempts to prepare the di-i-butylester of I: Application of established methods for the preparation of ί-butylesters yielded the anhydride of 1 (75% yield from 1 and f-butylisonitril [251 ; 70% from 1 and oxalylchloride followed by triethylamine and /-butanol126'; 75% from 1, dicyclohexylcarbodiimide and /-butanol in the presence of 4-(l-pyrrolidinyl)pyridin' 27 '). No evidence for formation of a neutral ester could be obtained in the reaction of the Cs salt 1 with2-bromo-2-methylpropanet 281 . Control experiment on the formation of 4 by transamidation: The solution of the BCHA salt of 3 (500 mg; 1.23 mmol) in 10 ml 50% aqueous ethanol and 1.25 ml (20 mmol) cone, ammonia was kept at room temperature for 2 days. Using the procedures described above, the BCHA salt was reisolated (456 mg; 91% needles; m.p. 185-187 °C (decomp.) and shown byTLC to contain only the amide 3, Rf= 0.32. No trace amount of 4 was detectable either in the crystals or in the mother liquor. Enzymes Citrate synthase and citrate lyase were purchased from BoehringerMannheim, Mannheim. ATP citrate lyase was purified from rat liver as described'29'. Enzymic assays All assays are described in ref.'29·30' and were performed at 25 °C in a total volume of 1.0 ml (d = 1 cm; Shimadzu UV-100-2 spectrophotometer). Citrate synthase was measured in the physiological reaction (50μ.Μ acetyl-CoA; 50μ,Μ oxaloacetate; 200μ,Μ 5,5dithiobis(2-nitrobenzoate); 60 mU citrate synthase and compound
Brought to you by | Purdue University Libraries Authenticated Download Date | 6/1/15 3:38 AM
710
J.Wilde, U. Lill and H. Eggerer
3 or 4 or 6 at O-lmw) and its reversal (0.5mM citrate; 0.5mM Co ASH; O.lSmM NADH; 200 U malate dehydrogenase; 2.7 U citrate synthase and compound 3 or 4 or 6 at 0-10mM). Citrate lyase (90 mU) and ATP citrate lyase (43 mU) were measured in the presence of amides 3 or 4 at 0-20mM. Concentrations of citrate were suboptimal with respect to determination of enzymic activity.
Results and Discussion Preparative approach Reaction of the lactone carbonyl of (/?S)-2-phenyll,3-dioxolan-4-one-5,5-diacetic acid (1) with ammonia should yield the citric acid /3-amide 3, required for the formation of the succinimide derivative 6 by cyclization. Similar as observed previously in the reaction of 1 with glycine ethyl ester[20], the ammonolysis of 1 produced not only the /3-amide but additionally the α-amide 4. In a representative reaction with 20 mmol 1 and 0.4 mol NH3, 56% (48%) of total substrate were isolated as the bis(cyclohexylammonium) salt (BCHA salt) of /3-amide 3, and 35% (35%) as that of α-amide 4. The structural assignment was made through ammonolysis of the thioester 5 yielding an amide which was identical (TLC; m.p.; mixed m.p.) with amide 4 derived from l.The overall procedure for the preparation and isolation of the two amides is simple and considered superior to their described modes of formation[31]. A solution of the diammonium salt of citric acid /3-amide 3 on heating liberated ammonia and yielded the desired succinimide derivative 6 (Scheme 1).
HO S X CH 2 C0 2 H
Υ. 0-C CH C0 H X V
2
H0 2 C' N CH 2 COSR
2
1 0
NH 3
NH 3
m^
X CH 2 C0 2 H
Η2ΝΟΓ/ X CH 2 C0 2 H
3
H0 2 C X " S CH 2 CONH 2
•O· OH X CH 2 C0 2 H
**
6 Η
Scheme 1. Formation of products on ammonolysis of 2-phenyll,3-dioxolan-4-one-5,5-diacetic acid (1).
Vol. 371 (1990)
Are the citric acid a- and -amides substrates of citrate synthase? Citrate synthase, in the reversed reaction, could convert amide 3 and half of the racemic amide 4 into acetyl-CoA and the a- and -amide of oxaloacetate, respectively. As judged from kinetic studies performed under usual assay conditions, neither of the two amides was a substrate or an inhibitor. These results were confirmed with citrate lyase and ATP citrate lyase. Citric anhydride (2) is transiently formed on solvolysis of the dioxolanone 1 The formation of the two amides from compound 1 would be expected if the previously assumed'20^ anhydride 2 were formed transiently as shown in Scheme 1, but could also result if the educt were actually a mixture of 1 and the six-membered isomer 2phenyl-l,3-dioxan-4-one-5-acetic-5-carboxylic acid, or if initially exclusively formed amide 3 could yield amide 4 by transamidation. Treatment of isolated amide 3 with ammonia under conditions as for 1 produced no trace of amide 4, thus excluding transamidation. The m.p. of 1 did not change on repeated recrystallization nor did TLC performed in different solvents reveal the presence of an impurity. The exclusive presence of the five-membered ring was ascertained from the Ή-ΝΜΚ spectra of 1 (4d, 2 / = 16.8 Hz) and its anhydride (2d, 2dd, 'J = 17.4 Hz each) indicating the presence of only one type of methylene groups. If the ring were six-membered in both compounds, the methylene groups would be different (one within and the other one outside the ring) with different coupling constants similar as shown for 3-hydroxy-2,5-pyrrolidinedione-3-acetic acid (6). Within the series citric acid, 1 and 1-anhydride the coupling constants increase in the sequence citric acid (2d, 2J = 15.8 Hz), 1 (4d, 2J = 16.8 Hz each) and 1-anhydride (2dd, 2/ = 17.4 Hz and 4J = 2.4 Hz; 2d, 2J 17.4 Hz). The results are in agreement with increasing steric hindrance of the methylene groups; the data of anhydride 1 are probably due to " W-coupling" of a pair of protons at the 3' and 5' positions of the anhydride ring. All of these results indirectly prove that amides 3 and 4 are formed from 1 via anhydride 2. In the reaction of 1 with ammonia, therefore, the direct ammonolysis of 1 yielding 3 must compete with the intramolecular cyclization reaction yielding anhydride 2 (and from this amides 3 and 4). Since 1 is racemic at C-2, both acetate side chains at the prochiral C-5 of 1 will participate equally well in the formation of 2. Assuming additionally an approximately equal attack of NH3 at both anhydride carbonyl Brought to you by | Purdue University Libraries Authenticated Download Date | 6/1/15 3:38 AM
Vol. 371 (1990)
711
Bifunctional Catalysis of Citrate Synthase
groups of 2, the ratio of products, 3/4, must exceed unity, as was found in the representative reaction, 3/4 = 1.6(1.4). Attempts were made to obtain the ammonolysisstable di-f-butyl ester'32' of 1 in which the action of a carboxylate anion would be abolished. Ammonolysis of the neutral ester should show the expected reaction of the lactone group with formation of only one product, the diester of amide 3. Application of methods established to prepare f-butyl esters, however, in the reaction of 1 were unsuccessful and yielded the anhydride of 1. Intramolecular bifunctional catalysis on hydrolysis ofl Convincing evidence for the formation of anhydride 2 was obtained from the pH-rate profile of the hydrolysis of 1 which shows a maximum near neutrality (see figure). Intramolecular anhydride-forming reactions of carboxylate anions have been established on hydrolysis of acetyl salicylic acid as well as of half esters of mainly succinic acid and of compounds containing a succinic acid half ester substructure'33"35', citryl-CoA included'20'. In all of these cases the increase in rate
-2.1
- -2.3
-2.4 pH
10
Figure. pH-Rate profile of the hydrolysis of 1 at 34 °C. The incubation mixture which was kept at 34 °C in a total volume of 1.2 m/83mM buffer solution* contained 16.7% ethanol (v/v) and 25mM dioxolanone 1. Samples (50 μ/; 1.25 μ,ιτιοί 1 initially) were taken at / = 1, 5, 10 and 30 min and immediately used to determine enzymically the hydrolysis product citrate. The bars shown in the figure indicate the results obtained in 4 independent series of measurements. * Glycine/HCl (pH 1.2-3.2); Tris/Mes (pH 3.2-9.0); NaHCO3/ Na 2 CO 3 (pH 10.7-12.5).
observed near neutrality is related to the dissociation of a carboxylic acid, the rate reaching a plateau after the ionization is complete. Examples are known additionally where the presence of two appropriately positioned carboxyl groups within the molecule cooperatively facilitates the hydrolysis of an ester bond in intramolecular bifunctional catalysis'33"33'. The plateau of the pH-rate profile described above is then replaced by a maximum, the position of which is given by the relationship (pK{ + p/C2)/2[33"35].This is observed on hydrolysis of 1 (figure; assumed: pK\ = 4.7, pK2 = 5.4, calculated maximum at pH = 5.1; Found: maximum at pH = 5.5) and indicates neighbouring group participation by a carboxyl group and its conjugated base as depicted in Scheme 2. H.
o^X^XXx^o 1
OH
2
Scheme 2. Cooperative catalysis on hydrolysis of 1.
Formation of anhydride 2 or direct hydrolysis of citryl-CoA on citrate synthase? Of the three catalytic groups of citrate synthase'191 His274 and His320 were suggested to accept a proton from the methyl group of acetyl-Co A and to donate a proton to the keto group of oxaloacetate, respectively, in "monofunctional catalysis"'19'. The action of the third catalytic residue, Asp375, was less clear due to less well defined structural data and proposed by us to polarize the thioester carbonyl of acetyl-CoA to facilitate its enolization in bifunctional catalysis'3'. In favour of this idea it has been shown recently by sitedirected mutagenesis of citrate synthase that Asp375 is required for the binding of acetyl-Co A'36'. The proposed action of the protonated catalytic carboxyl group of the enzyme is very similar indeed to the action of the protonated carboxyl group in the chemical reaction shown in Scheme 2. This action of Asp375 in combination with the previously suggested function of His274 and His320'191 is presented in Scheme 3. The intermediate enolic acetyl-CoA (demonstrated recently with dithio acetyl-CoA'15'), is generated in cooperative acid and base catalysis executed by Asp375 and His274, respectively. Addition of the enol(ate) to the si-face of the keto carbonyl of oxaloacetate satisfies the stereochemical course of the reaction and yields (SS^-citryl-CoA that is polarized at the thioester carbonyl. Citrate could now Brought to you by | Purdue University Libraries Authenticated Download Date | 6/1/15 3:38 AM
712
J.Wilde, U. Lill and H. Eggerer 4
(His320)N-H
5 6 7 0 2 C(Asp375)
8 9 10 11
O^C-0—H—0 2 C(Asp375) ^"•- -\(T} H20
12 13 14
,C0 2 H
•W > °
Scheme 3. Suggested role of Asp375 in the formation and hydrolysis of (3.S)-citryl-CoA with citrate synthase.
15 16 17 18 19
be formed in bifunctional catalysis via the anhydride (pathway 2 in Scheme 3), or by direct attack of a water molecule at the activated thioester group (pathway 1 in Scheme 3).These mechanistic ambiguities are similar as discussed for carboxypeptidase A[36"38l To probe for a solution we investigated the influence of the aza analogue of the anhydride, (ÄS)-3-hydroxy2,5- pyrrolidinedione-3-acetic acid, on the enzymic reaction. As judged from enzymic assays performed in the forward reaction of citrate synthase and its reversal, 6 was no inhibitor of the enzyme, thus favouring the direct hydrolysis of citryl-CoA in the alternative pathway. This work was supported by the Fonds der Chemischen Industrie. Note: An X-ray-crystallographic study of the ternary complex citrate synthase-S-carboxymethyl-Co A-oxaloacetate, assumed to represent a model for the condensation step, appeared'39' while our manuscript was in preparation. Bifunctional catalysis is supposed to generate enolic acetyl-CoA as in our paper but with the functions of His274 and Asp375 interchanged.
20 21 22 23 24 25 26 27 28 29 30
31 32
References 1 2 3
Eggerer, H. (1965) Biochem. Z. 343, 111-138. Bruice, T.C. & Benkovic, S.J. (1966) Bioorganic Mechanisms, vol. 1, p. 294, W. A. Benjamin, New York. Löhlein-Werhahn, G., Bayer, E., Bauer, B. & Eggerer, H. (1983) Eur. J. Biochem. 133, 665-672.
33 34
Vol.371 (1990)
Lill, U, Schreil, A., Henschen, A. & Eggerer, H. (1984) Eur. J. Biochem. 143, 205-212. Löhlein-Werhahn, G., Goepfert, P. & Eggerer, H. (1985) Eur. J. Biochem. 150, 79-84. Lill, U., Bibinger, A. & Eggerer, H. (1987) Eur. J. Biochem. 162, 683-689. Lill, U., Bibinger, A. & Eggerer, H. (1987) Eur. J. Biochem. 163, 599-607. Löhlein-Werhahn, G., Goepfert, P., Kollmann-Koch, A. & Eggerer, H. (1988) Biol. Chem. Hoppe-Seyler 369.417424. Pettersson, G., Lill, U. & Eggerer, H. (1989) Eur. J. Biochem. 182, 119-124. Eggerer, H. & Klette, A. (1967) Eur. J. Biochem. 1, 447475. Eggerer, H., Buckel, W, Lenz, H., Wunderwald, P., Gottschalk, G., Cornforth, J.W., Donninger, C, Mallaby, R., Redmond, J.W. (1970) Nature (London) 226,517-519. Lenz, H., Buckel, W, Wunderwald, R, Biedermann, G.. Buschmeier, V., Eggerer, H., Cornforth, J.W. & Mallaby. R. (1971) Eur. J. Biochem. 24, 207-215. Retey, J., Lüthy, J. & Arigoni, D. (1970) Nature (London) 226, 519-521. Klinman, J. & Rose, I.A. (1971) Biochemistry 10, 22672272. Clark, J.D., CTKeefe, S.J. & Knowles, J.R. (1988) Biochemistry 27, 5961-5971. Wlassics, J.D. & Anderson, V.E. (1989) Biochemistry 28. 1627-1633. Johansson, CJ. & Pettersson, G. (1974) Eur. J. Biochem. 46,5-11. Bayer, E., Bauer, B. & Eggerer, H. (1981) FEBS Lett. 127, 101-104. Remington, S., Wiegand, G. & Huber, R. (1982) J. Mol. Biol. 158, 111-152. Buckel, W. & Eggerer, H. (1969) Hoppe-Seyler's Z. Physiol. Chem. 350, 1367-1376. Wunderwald. P. & Eggerer. H. (1969) Eur. J. Biochem. 11. 97-105. Eggerer, H., Giesemann, W. & Aigner, H. (1983) Liebigs Ann. Chem. 1551-1560. Farines, M. & Soulier, J. (1970) Bull. Soc. Chim. Fr. l, 332-340. Nau, C.A., Brown, E.B. & Bailey, J.R. (1925) J. Am. Chem. Soc. 47, 2596-2606. Rehn, D. & Ugi, I. (1977)7. Chem. Res. (M) 1501-1506; S (Synopsis) 119. Kaiser, G.V, Cooper, R.D.G., Koehler, R.E., Murphy, C.F., Webber, J.A., Wright, LG. & van Heyningen, E.M. (1970)7. Org. Chem. 35, 2429-2430. Hassner, A. & Alexanian, V. (1978) Tetrahedron Lett. 46, 4475-4478. Wang, S.S., Gisin, B.F., Winter, D.P., Makofske, R., Kuleska, I.D.,Tzougraki, C. & Meienhofer, J. (1977) J. Org. Chem. 42,1286-1290. Linn,T.C. & Srere, P.A. (1979) J. Biol. Chem. 254, 16911698. Bergmeyer, H.U., Graßl, M. & Walter, H.-E. (1983) in Methods of Enzymatic Analysis, 3rd edn. (Bergmeyer, H.U, Bergmeyer, J. & Graßl, M., eds.) vol. 2, pp. 173175, VCH Verlagsgesellschaft, Weinheim. Schroeter, G. (1905) Ber. Dtsch. Chem. Ges. 38, 31903210. Sustmann, R. & Korth, H.-G. (1985) in Houben-Weyl, Methoden der Organischen Chemie, vol. E5, part l (J. Falbe, ed.)Thieme Verlag, Stuttgart 496-504. Bruice, T.C. & Benkovic, S.J. (1966) Bioorganic Mechanisms, vol. I, pp. 173-186, W.A. Benjamin, New York. Morawetz, H. & Oreskes, I. (1958) J. Am. Chem. Soc. 80, 2591-2592.
Brought to you by | Purdue University Libraries Authenticated Download Date | 6/1/15 3:38 AM
Vex. 371 (1990)
Bifunctional Catalysis of Citrate Synthase
35 Morawetz, H. & Shafer, J. (1962) J. Am. Chem. Soc. 84, 3783-3784. 36 Handford, P., Ner, S.S., Bloxham, P.D. & Wilton, C.D. (1988) Biochim. Biophys. Acta 953, 232-240.
713
37 Lipscomb,W.N. (1982) Ace. Chem. Res. 15, 232-238. 38 Breslow, R. (1986) in Adv. Eniymol. 58, 1-60. 39 Karpusas. M., Branchaud, B. & Remington, S.J. (1990) Biochemistry 29. 2213-2219.
JJ. Wilde. U. Lill, H. Eggerer*, Institut für Physiologische Chemie der Technischen Universität München, IBiedersteiner Str. 29, D-8000 München 40. :
* To whom correspondence should be atddressed.
Brought to you by | Purdue University Libraries Authenticated Download Date | 6/1/15 3:38 AM
Brought to you by | Purdue University Libraries Authenticated Download Date | 6/1/15 3:38 AM
On the Action of Carboxy Groups in the Citrate Synthase Reaction JürgenWiLDE, Ute LILL and Hermann EGGERER Institut für Physiologische Chemie der Technischen Universität München
(Received 4 May 1990)
Summary: The aza analogue (/?5')-3-hydroxy-2,5-pyrrolidinedione-3-acetic acid (6) of the five-membered citric anhydride (2) was prepared in the sequence citric acid —» 2-phenyl-l,3-dioxolan-4-one-5,5-diacetic acid (1) —> citric acid /3-amide (3) -> 6 and used to resolve ambiguities in the mechanism of the citrate synthase reaction. The results yield no indication for the formation of anhydride 2 on the enzyme and favour the direct hydrolysis of the intermediate (351)citryl-CoA. Ammonolysis of the dioxolanone 1 in the reaction sequence described above produced not only citric acid
/3-amide but also the -isomer. This is shown to originate in the transient formation of anhydride 2. Hydrolysis of the dioxolanone 1 under "physiological conditions" occurs via anhydride 2, generated in intramolecular bifunctional catalysis by a protonated and a deprotonated carboxyl group. The catalytic residue Asp375 of citrate synthase is considered to operate on the enzyme as does the protonated carboxyl group in the chemical reaction and to generate enolic acetyl-CoA in cooperative catalysis with His274. This reaction of Asp375 may also facilitate the hydrolysis of citryl-CoA.
Von der mechanistischen Sonde 3-Hydroxy-2,5-pyrrolidindion-3-essigsäure zu bifunktioneller Katalyse von Citrat-Synthase Zusammenfassung: Das Aza-Analoge (RS)-3-Hydroxy-2,5-pyrrolidindion-3-essigsäure (6) von fünfgliedrigem Citronensäureanhydrid (2) wurde in der Sequenz Citronensäure —» 2-Phenyl-l,3-dioxolan-4on-5,5-diessigsäure (1) —» Citronensäure ß-amid (3) —> 6 erhalten und zur Klärung mechanistischer Details der Citrat-Synthase-Reaktion verwendet. Die Ergebnisse liefern keinen Hinweis für die enzymatische Bildung des Anhydrids 2 und begünstigen die direkte Hydrolyse des Zwischenprodukts (3S)-CitrylCoA. Die Ammonolyse des Dioxolanons l in der voranstehend beschriebenen Reaktionsfolge ergab nicht nur
Citronensäure-ß-amid, sondern auch das -Isomere. Es entsteht aus transient gebildetem Anhydrid 2. Die Hydrolyse des Dioxolanons l unter „physiologischen Bedingungen" erfolgt über das Anhydrid 2 in intramolekularer bifunktioneller Katalyse mittels einer protonierten und einer deprotonierten Carboxylgruppe. Die Wirkung der protonierten Carboxylgruppe in der chemischen Reaktion liefert ein Modell für die Funktion der katalytischen Gruppe Asp375 der Citrat-Synthase. In bifunktioneller Katalyse mit His274 kann Asp375 enolisches Acetyl-Co A bilden und die Hydrolyse von Citryl-CoA einleiten.
Enzymes: ATP citrate (pro-35)-lyase, ATPrcitrate oxaloacetate-lyase (EC 4.1.3.8); Citrate (pro-35)-lyase, citrate oxaloacetate-lyase ((pro-35)-CH2COf —> acetate) (EC 4.1.3.6); Citrate (si')-synthase, citrate oxaloacetate-lyase ((pro-35)-CH2COf -> acetyl-CoA) (EC 4.1.3.7). Abbreviations: BCHAsalt, bis(cyclohexylammonium) salt;TLC, thin-layer chromatography.
Brought to you by | Purdue University Libraries Copyright © by WalterAuthenticated de Gruyter & Co · Berlin · New York
Download Date | 6/1/15 3:38 AM
708
J.Wilde, U. Lill and H. Eggerer
Vol. 371 (1990)
Key words: Citrate synthase, bifunctional catalysis, citric anhydride, aza analogue.
Citrate synthase of the tricarboxylic acid cycle catalyses the formation of citrate from acetyl-CoA and oxaloacetate.The enzyme (pig heart) also transforms (3,S)-citryl-CoA to the substrates and products of the physiological reaction and, in the presence of (S)-malate in place of oxaloacetate, liberates a proton from the methyl group of acetyl-CoA^'.The Claisen type of condensation (cf. ref.'21 for definition) indicated by these results could not be considered proven initially because the rate of product formation from citryl-CoA was lower than the rate of the physiological reaction. Pursuit of this unexpected result revealed complicated inhibition kinetics'31 which were thoroughly investigated'4'81 and could be described quantitatively on basis of a nonproductive enzyme· citryl-CoA complex'91. The actual initial rate of citrylCo A hydrolysis satisfies the kinetic criterion for an intermediate; with a source variant of the enzyme the rate was even twofold faster than the rate of the physiological re action'81. The experimental results are now in complete harmony with the outlined mechanism. Characteristic traits of Claisen (condensation) enzymes are the requirement of a cosubstrate for proton exchange of the nucleophilic reactand with the solvent'1·101, and the stereochemical course of the reaction^1 14] (inversion of configuration at the methyl group of acetyl-Co A which becomes methyleneofthe pro-S carboxymethyl group of citrate), arguments considered in favour of a concerted reaction'2'15'161. The obligatory formation of a ternary enzyme-substrate complex'171 as well as the enzymic activities required for C-C condensation and thioester hydrolysis residing in a single site of citrate synthase'18·191, are consistent with this view. Actually, however, and in agreement with the behaviour of citryl-CoA as an intermediate, enzymic Claisen condensations, like the familiar aldol condensations, have been shown to proceed in a stepwise manner'15'161 (see ref.'151 for a lucid discussion on the two types of condensation reactions). In the light of these results a citryl-CoA-derived five-membered and hydrolysis-sensitive citric anhydride (3-carboxymethyl-3-hydroxysuccinic anhydride) gains renewed interest. Its formation from citryl-thioesters by intramolecular attack of the C-6 carboxylate anion at the thioester carbonyl occurs under "physiological conditions" and effects hydrolysis of the thioester group that is otherwise stable under these conditions'201. The same reactions may initiate the hydrolysis of citryl-CoA on the enzyme but so far no experimental evidence supports this viewlu -1J. Here we probe for a new approach.The af-
finity of citrate synthase to the anhydride, if used as a substrate, would be expected to be high only if the anhydride is formed in the course of the physiological reaction. In the same selective sense an analogue of the anhydride would be expected to inhibit the enzyme only if it were an intermediate analogue. 3-Hydroxy-2,5-pyrrolidinedione-3-acetic acid (6in Scheme 1 shown below) in which the central oxygen of citric anhydride is replaced by nitrogen would appear to represent a suitable probe because the NH group is inert and the amide structure stable under physiological conditions. We have, therefore, prepared this substituted succinimide derivative and determined its interaction with citrate synthase. The results are presented in this paper.
Experimental Elemental analyses were performed by H. Schulz (Mikroanalytisches Laboratorium) and IR spectra of solid samples measured by H. Huber (Physikalisch-Chemisches Laboratorium), both Organisch-Chemisches Laboratorium der Universität München. NMR spectra were executed by Drs. Zetl and Sonnenbichler at the Max-Planck-Institut für Biochemie, Martinsried. We thank them all for their help. General Solvents were dried over Na2SO4 and evaporated under reduced pressure in a rotary evaporator. Titrations were performed with phenolphthalein as indicator. Melting points were measured in capillary tubes (Büchi 510 apparatus) and are uncorrected. Acids were recovered from salts by passage through Dowex-50 (H®; 50100 mesh).Thin-layer chromatography (TLC) was executed on DC Plastikfolie Kieselgel 60/F254, Merck, in the solvent ethanol/conc. NH 3 /water = 7.8:1.25:0.95 (v/v). Acids were detected with ethanolic kresol green, pH 8; compounds containing the benzene ring were visualized at 254 nm and by reaction with ethanolic 10mM 2,4-dinitrophenylhydrazine containing 5mM HC1 and liberating benzaldehyde. IR spectra of liquid and solid samples were measured with a Perkin-Elmer Infrared Spectrometer 710 B, and in KBr pellets with a Perkin-Elmer Infrared Spectrophotometer Model 1420 Ratio Recording, respectively. -NMR spectra were taken at 500 MHz on a Bruker AM 500 Spectrometer and are recorded in parts per million downfield from internal tetramethylsilane; coupling constants are expressed in Hz. Preparation of compounds (RS)-2-Phenyl-l,3-dioxolan-4-one-5,5-diaceticacid (1): The previously used condensation agent P4Oi0[22) was replaced by boron trifluoridet23].To a solution of water-free citric acid (22.4 g; 0.12 mol) and freshly distilled benzaldehyde (10.1 m/; 0.1 mol) in 100 ml abs. dioxane was added at room temperature dropwise with stirring 14 ml of boron trifluoride etherate within 5 min. The immediately formed cherry-red and clear solution was stirred over night, then evaporated and the oily brownish residue dissolved in 200 ml ethyl acetate. This was extracted with 3 100 ml sat. KHCO3 solution and the aqueous phase, pH ca. 7.5, cooled with ice, acidified to pH 3 with OM HC1 and extracted four times with a total of 300 ml ethyl acetate. The combined extracts were dried, then evaporated to yield 1 (20.6 g; 74%) of m.p. 171-173 °C (decomp.). Recrystalliza-
Brought to you by | Purdue University Libraries Authenticated Download Date | 6/1/15 3:38 AM
Vol.371 (1990)
Bifunctional Catalysis of Citrate Synthase
709
tion from ethyl acetate/light petroleum (60-80°) gave 19.7 g (71 %) of m.p. 173-175 °C (decomp.).TLC in methanol as solvent: only one spot, Rf = 0.54, containing acidic and aldehydic (after treatment with acid) functional groups (benzaldehyde: Rf = 0.32). Ή-NMR (acetone-Do): δ = 3.04 (d, 2 J(H,H) = 16.8, I H · pro-R or pro-S C//2CO2H), 3.14 (d, 2/(H,H) = 16.8, 1H; pro-R or pro-S C//2C02H), 3~.12 (d, 2J(H,H) = 16.8, 1H; pro-S or pro-R C//2CO2H), 3.16 (d, 2J(H,H) = 16.8, 1H; pro-S or pro-R C//2C02H); IR(KBr): ν (cm'1) = 1805-1810 (C = O); UVabsorbance (ethanol): £26i = 0.26mM~' cm"1. The anhydride of compd. 1 was prepared as in ref.' 24 '. Ή-NMR (acetone-D6): δ = 3.50 (d, 2J = (H,H) = 17.4, 1H; pro-R or pro-S C/Y 2 -CO-O),3.63(d, 2 /(H,H) = 17.4,1H; pro-R or pro-S CH2-COO), 3.56 (dd, 2 /(H,H) = 17.4 and 47(H,H) - 2.4,1H; pro-S or pro- R C// r CO-O),3.76(dd, 2 /(H,H) = 17.4 and 47(H,H) - 2.4, !H;proS or pro-R CH2-CO-0).
(65%) needles, m.p. 170-172 °C (decomp.). A mixed m.p. with 4 isolated from the reaction of 1 with ammonia gave no depression. TLC: only one compound, Rf = 0.19, which was inseparable from 4 isolated as described above under a).The free amic acid 4 was obtained from the BCHA salt after treatment with Dowex-50 ( 1 x 7 cm) and addition of acetone to the desalted and evaporated solution. Needles (340 mg), m.p. 143 °C (decomp.); after recrystallization from 95% ethanol: 270 mg (71%) m.p. 144-146°C (decomp.). Equiv. weight found: 97; calculated 96;TLC: only one acid, Rf = 0.19. C6H906N (191.1) Calc.: C37.7 H 4.8 Ν 7.3 Found: C37.8 H 5.0 Ν 7.6 Ή-NMR (D2O): δ - 2.77 (d, 2./(H,H) = 14.8: C//2CO2H), 2.88 (d, 2 J(H,H) - 14.8, 1H; C//2CO2H); 2.87 (d, 27(H,H) = 16.2, 1H; C// 2 CONH 2 ), 3.09 (d, 2 /(H,H) = 16.2, 1H; Ctf 2 CONH 2 ).
Citric acid -amide (3): To a solution of 5.6 g (20 mmol) 1 in 25 ml ethanol was added 25 ml (ca. 400 mmol) cone, ammonia and the reaction mixture kept at room temperature over night. Precipitated crystals (benzaldehyde ammonia) were isolated by suction filtration and washed with aqueous ethanol.The elutate was evaporated for removal of ammonia and ethanol, and the aqueous phase extracted with ethyl acetate (3 x 10 m/) for removal of residual benzaldehyde. The desalted solution (Dowex-50; 2 x 10 cm) was titrated with cyclohexylamine (5 ml), then evaporated and the residue crystallized from water-ethanol to yield 4.6 g (57%) BCHA salt, needles, m.p. 186-188 °C (decomp.) (the mother liquor of this crystallization contained 4, isolated as described below). TLC yielded one acid spot, RF= 0.32 (citric acid: RF= 0.06). A sample after recrystallization gave m.p. 186-188 °C (decomp.). C 6 H 9 O 6 N-2C 6 H 13 N-H 2 0 (407.5) Calc.: C53.1 H 9.2 N 10.3 Found: C53.3 H 9.2 N 10.3
(RS)-3-Hydroxy-2,5-pyrrolidinedione-3-acetic acid (6): The solution of the diammonium salt 3 (255 mg, 1.33 mmol) and 8 mg 4-toluenesulfonic acid in 30 ml dimethylformamide was heated under nitrogen for 16 h to 100-110 °C until the liberation of ammonia had ceased. The solvent was evaporated at 50 °C (0.1 Torr) and the yellowish residue desalted (Dowex-50; 0.5 x 6 cm), then neutralized with cyclohexylamine (0.15 mi) and brought todryness. Crystallization of the oily residue from ethanol-light petroleum (60-80 °C) gave 315 mg (87%) needles, m.p. 170-172 °C (decomp.). TLC: only one spot, Rp — 0.54; after recrystallization of a sample: m.p. 170-172 °C (decomp.). C6H7N05-C6H13N (272.3) Calc.: C52.9 H 7.4 N 10.3 Found: C52.7 H7.7 N 10.0 Ή-NMR (free acid in acetone-Do): δ = 4.69 (d, 2/(H,H) = 18.2, 1H; ring C//2CONH), 5.12 (d, 2/(H,H) = 18.2, 1H; ring C//2CONH), 4.97 (d, 2/(H,H) = 16.8, 1H; C//2CO2H), 5.09 (d, 2 7(H,H) = 16.8, 1H; C//2CO2H). On treatment with alkali (!M NaOH, 37 °C, 120 min) 6 was completely hydrolysed to yield according toTLC the α-amide 4 (major product) and the /3-amide 3 (minor product); free citric acid was not detectable.
Diammonium salt: BCHA salt (1.63 g; 4 mmol) was dissolved in 15 ml water under gentle warming, then desalted (Dowex-50; 1 x 7 cm) and the effluent carefully evaporated.The oily residue was dissolved in 1.5 ml water plus 1 ml one. ammonia and crystallized at 0 °C by repeated addition of ethanol to commencing turbidity. The isolated diammonium salt, prisms (750 mg; 83%), m.p. 170-172 °C (decomp.), gave a single spot onTLC, Rf = 0.32. C 6 H 9 O f t N-2NH 3 (225.2) Calc.: C32.0 H 6.7 N 18.7 Found: C32.2 H 6.9 N 18.7 Ή-NMR (D20): δ = 2.50 (d, 2/(H,H) = 15.1, 2H; pro-R or pro-S C//2C02H), 2.70 (d, 2J = (H,H) = 15.1, 2H; pro-R or pro-S Cf/2CO2H). The free acid 3 crystallized on treatment of the oily residue described above with ether; needles, m.p. 99-101 °C (decomp.). Recrystallization was successful from ethanol only but not without esterification: prisms, m.p. 93-96 °C (decomp.) which according to titration, elemental analysis and NMR-spectroscopy contained an ethyl ester of 3 as an impurity. Citric acid a-amide (4): a) From the dioxolanone l:The mother liquor of the BCH Asalt isolation of 3 described above was evaporated and the residue crystallized from water-acetone to yield 2.94 g (35.0%) BCHA salt, needles, m.p. 170-172 °C (decomp.). TLC: only one spot, Rf = 0.19. C 6 H 9 O 6 N-2C 6 Hi 3 N-1.75H 2 O (403.0) Calc.: C53.7 H 9.1 N 10.4 Found: C53.8 H 9.1 N 10.3 b) From (R.S)-citryl-N-octanoylcysteamine 5.The suspension of 5'22' (K®, H®; 210 mg; 0.5 mmol) in 1 ml water was neutralized with 0.25 m/2M NH 3 ; cone, ammonia (1 m/, ca. 16 mmol) was added without delay and the mixture kept at room temperature for 3 h. Precipitated N-octanoylcysteamine was extracted with ether and the clear aqueous solution evaporated for removal of ammonia, then desalted (Dowex-50; 1 x 3 cm).and the effluent titrated with cyclohexylamine (130 μ/). Isolation as indicated above gave 130 mg
Attempts to prepare the di-i-butylester of I: Application of established methods for the preparation of ί-butylesters yielded the anhydride of 1 (75% yield from 1 and f-butylisonitril [251 ; 70% from 1 and oxalylchloride followed by triethylamine and /-butanol126'; 75% from 1, dicyclohexylcarbodiimide and /-butanol in the presence of 4-(l-pyrrolidinyl)pyridin' 27 '). No evidence for formation of a neutral ester could be obtained in the reaction of the Cs salt 1 with2-bromo-2-methylpropanet 281 . Control experiment on the formation of 4 by transamidation: The solution of the BCHA salt of 3 (500 mg; 1.23 mmol) in 10 ml 50% aqueous ethanol and 1.25 ml (20 mmol) cone, ammonia was kept at room temperature for 2 days. Using the procedures described above, the BCHA salt was reisolated (456 mg; 91% needles; m.p. 185-187 °C (decomp.) and shown byTLC to contain only the amide 3, Rf= 0.32. No trace amount of 4 was detectable either in the crystals or in the mother liquor. Enzymes Citrate synthase and citrate lyase were purchased from BoehringerMannheim, Mannheim. ATP citrate lyase was purified from rat liver as described'29'. Enzymic assays All assays are described in ref.'29·30' and were performed at 25 °C in a total volume of 1.0 ml (d = 1 cm; Shimadzu UV-100-2 spectrophotometer). Citrate synthase was measured in the physiological reaction (50μ.Μ acetyl-CoA; 50μ,Μ oxaloacetate; 200μ,Μ 5,5dithiobis(2-nitrobenzoate); 60 mU citrate synthase and compound
Brought to you by | Purdue University Libraries Authenticated Download Date | 6/1/15 3:38 AM
710
J.Wilde, U. Lill and H. Eggerer
3 or 4 or 6 at O-lmw) and its reversal (0.5mM citrate; 0.5mM Co ASH; O.lSmM NADH; 200 U malate dehydrogenase; 2.7 U citrate synthase and compound 3 or 4 or 6 at 0-10mM). Citrate lyase (90 mU) and ATP citrate lyase (43 mU) were measured in the presence of amides 3 or 4 at 0-20mM. Concentrations of citrate were suboptimal with respect to determination of enzymic activity.
Results and Discussion Preparative approach Reaction of the lactone carbonyl of (/?S)-2-phenyll,3-dioxolan-4-one-5,5-diacetic acid (1) with ammonia should yield the citric acid /3-amide 3, required for the formation of the succinimide derivative 6 by cyclization. Similar as observed previously in the reaction of 1 with glycine ethyl ester[20], the ammonolysis of 1 produced not only the /3-amide but additionally the α-amide 4. In a representative reaction with 20 mmol 1 and 0.4 mol NH3, 56% (48%) of total substrate were isolated as the bis(cyclohexylammonium) salt (BCHA salt) of /3-amide 3, and 35% (35%) as that of α-amide 4. The structural assignment was made through ammonolysis of the thioester 5 yielding an amide which was identical (TLC; m.p.; mixed m.p.) with amide 4 derived from l.The overall procedure for the preparation and isolation of the two amides is simple and considered superior to their described modes of formation[31]. A solution of the diammonium salt of citric acid /3-amide 3 on heating liberated ammonia and yielded the desired succinimide derivative 6 (Scheme 1).
HO S X CH 2 C0 2 H
Υ. 0-C CH C0 H X V
2
H0 2 C' N CH 2 COSR
2
1 0
NH 3
NH 3
m^
X CH 2 C0 2 H
Η2ΝΟΓ/ X CH 2 C0 2 H
3
H0 2 C X " S CH 2 CONH 2
•O· OH X CH 2 C0 2 H
**
6 Η
Scheme 1. Formation of products on ammonolysis of 2-phenyll,3-dioxolan-4-one-5,5-diacetic acid (1).
Vol. 371 (1990)
Are the citric acid a- and -amides substrates of citrate synthase? Citrate synthase, in the reversed reaction, could convert amide 3 and half of the racemic amide 4 into acetyl-CoA and the a- and -amide of oxaloacetate, respectively. As judged from kinetic studies performed under usual assay conditions, neither of the two amides was a substrate or an inhibitor. These results were confirmed with citrate lyase and ATP citrate lyase. Citric anhydride (2) is transiently formed on solvolysis of the dioxolanone 1 The formation of the two amides from compound 1 would be expected if the previously assumed'20^ anhydride 2 were formed transiently as shown in Scheme 1, but could also result if the educt were actually a mixture of 1 and the six-membered isomer 2phenyl-l,3-dioxan-4-one-5-acetic-5-carboxylic acid, or if initially exclusively formed amide 3 could yield amide 4 by transamidation. Treatment of isolated amide 3 with ammonia under conditions as for 1 produced no trace of amide 4, thus excluding transamidation. The m.p. of 1 did not change on repeated recrystallization nor did TLC performed in different solvents reveal the presence of an impurity. The exclusive presence of the five-membered ring was ascertained from the Ή-ΝΜΚ spectra of 1 (4d, 2 / = 16.8 Hz) and its anhydride (2d, 2dd, 'J = 17.4 Hz each) indicating the presence of only one type of methylene groups. If the ring were six-membered in both compounds, the methylene groups would be different (one within and the other one outside the ring) with different coupling constants similar as shown for 3-hydroxy-2,5-pyrrolidinedione-3-acetic acid (6). Within the series citric acid, 1 and 1-anhydride the coupling constants increase in the sequence citric acid (2d, 2J = 15.8 Hz), 1 (4d, 2J = 16.8 Hz each) and 1-anhydride (2dd, 2/ = 17.4 Hz and 4J = 2.4 Hz; 2d, 2J 17.4 Hz). The results are in agreement with increasing steric hindrance of the methylene groups; the data of anhydride 1 are probably due to " W-coupling" of a pair of protons at the 3' and 5' positions of the anhydride ring. All of these results indirectly prove that amides 3 and 4 are formed from 1 via anhydride 2. In the reaction of 1 with ammonia, therefore, the direct ammonolysis of 1 yielding 3 must compete with the intramolecular cyclization reaction yielding anhydride 2 (and from this amides 3 and 4). Since 1 is racemic at C-2, both acetate side chains at the prochiral C-5 of 1 will participate equally well in the formation of 2. Assuming additionally an approximately equal attack of NH3 at both anhydride carbonyl Brought to you by | Purdue University Libraries Authenticated Download Date | 6/1/15 3:38 AM
Vol. 371 (1990)
711
Bifunctional Catalysis of Citrate Synthase
groups of 2, the ratio of products, 3/4, must exceed unity, as was found in the representative reaction, 3/4 = 1.6(1.4). Attempts were made to obtain the ammonolysisstable di-f-butyl ester'32' of 1 in which the action of a carboxylate anion would be abolished. Ammonolysis of the neutral ester should show the expected reaction of the lactone group with formation of only one product, the diester of amide 3. Application of methods established to prepare f-butyl esters, however, in the reaction of 1 were unsuccessful and yielded the anhydride of 1. Intramolecular bifunctional catalysis on hydrolysis ofl Convincing evidence for the formation of anhydride 2 was obtained from the pH-rate profile of the hydrolysis of 1 which shows a maximum near neutrality (see figure). Intramolecular anhydride-forming reactions of carboxylate anions have been established on hydrolysis of acetyl salicylic acid as well as of half esters of mainly succinic acid and of compounds containing a succinic acid half ester substructure'33"35', citryl-CoA included'20'. In all of these cases the increase in rate
-2.1
- -2.3
-2.4 pH
10
Figure. pH-Rate profile of the hydrolysis of 1 at 34 °C. The incubation mixture which was kept at 34 °C in a total volume of 1.2 m/83mM buffer solution* contained 16.7% ethanol (v/v) and 25mM dioxolanone 1. Samples (50 μ/; 1.25 μ,ιτιοί 1 initially) were taken at / = 1, 5, 10 and 30 min and immediately used to determine enzymically the hydrolysis product citrate. The bars shown in the figure indicate the results obtained in 4 independent series of measurements. * Glycine/HCl (pH 1.2-3.2); Tris/Mes (pH 3.2-9.0); NaHCO3/ Na 2 CO 3 (pH 10.7-12.5).
observed near neutrality is related to the dissociation of a carboxylic acid, the rate reaching a plateau after the ionization is complete. Examples are known additionally where the presence of two appropriately positioned carboxyl groups within the molecule cooperatively facilitates the hydrolysis of an ester bond in intramolecular bifunctional catalysis'33"33'. The plateau of the pH-rate profile described above is then replaced by a maximum, the position of which is given by the relationship (pK{ + p/C2)/2[33"35].This is observed on hydrolysis of 1 (figure; assumed: pK\ = 4.7, pK2 = 5.4, calculated maximum at pH = 5.1; Found: maximum at pH = 5.5) and indicates neighbouring group participation by a carboxyl group and its conjugated base as depicted in Scheme 2. H.
o^X^XXx^o 1
OH
2
Scheme 2. Cooperative catalysis on hydrolysis of 1.
Formation of anhydride 2 or direct hydrolysis of citryl-CoA on citrate synthase? Of the three catalytic groups of citrate synthase'191 His274 and His320 were suggested to accept a proton from the methyl group of acetyl-Co A and to donate a proton to the keto group of oxaloacetate, respectively, in "monofunctional catalysis"'19'. The action of the third catalytic residue, Asp375, was less clear due to less well defined structural data and proposed by us to polarize the thioester carbonyl of acetyl-CoA to facilitate its enolization in bifunctional catalysis'3'. In favour of this idea it has been shown recently by sitedirected mutagenesis of citrate synthase that Asp375 is required for the binding of acetyl-Co A'36'. The proposed action of the protonated catalytic carboxyl group of the enzyme is very similar indeed to the action of the protonated carboxyl group in the chemical reaction shown in Scheme 2. This action of Asp375 in combination with the previously suggested function of His274 and His320'191 is presented in Scheme 3. The intermediate enolic acetyl-CoA (demonstrated recently with dithio acetyl-CoA'15'), is generated in cooperative acid and base catalysis executed by Asp375 and His274, respectively. Addition of the enol(ate) to the si-face of the keto carbonyl of oxaloacetate satisfies the stereochemical course of the reaction and yields (SS^-citryl-CoA that is polarized at the thioester carbonyl. Citrate could now Brought to you by | Purdue University Libraries Authenticated Download Date | 6/1/15 3:38 AM
712
J.Wilde, U. Lill and H. Eggerer 4
(His320)N-H
5 6 7 0 2 C(Asp375)
8 9 10 11
O^C-0—H—0 2 C(Asp375) ^"•- -\(T} H20
12 13 14
,C0 2 H
•W > °
Scheme 3. Suggested role of Asp375 in the formation and hydrolysis of (3.S)-citryl-CoA with citrate synthase.
15 16 17 18 19
be formed in bifunctional catalysis via the anhydride (pathway 2 in Scheme 3), or by direct attack of a water molecule at the activated thioester group (pathway 1 in Scheme 3).These mechanistic ambiguities are similar as discussed for carboxypeptidase A[36"38l To probe for a solution we investigated the influence of the aza analogue of the anhydride, (ÄS)-3-hydroxy2,5- pyrrolidinedione-3-acetic acid, on the enzymic reaction. As judged from enzymic assays performed in the forward reaction of citrate synthase and its reversal, 6 was no inhibitor of the enzyme, thus favouring the direct hydrolysis of citryl-CoA in the alternative pathway. This work was supported by the Fonds der Chemischen Industrie. Note: An X-ray-crystallographic study of the ternary complex citrate synthase-S-carboxymethyl-Co A-oxaloacetate, assumed to represent a model for the condensation step, appeared'39' while our manuscript was in preparation. Bifunctional catalysis is supposed to generate enolic acetyl-CoA as in our paper but with the functions of His274 and Asp375 interchanged.
20 21 22 23 24 25 26 27 28 29 30
31 32
References 1 2 3
Eggerer, H. (1965) Biochem. Z. 343, 111-138. Bruice, T.C. & Benkovic, S.J. (1966) Bioorganic Mechanisms, vol. 1, p. 294, W. A. Benjamin, New York. Löhlein-Werhahn, G., Bayer, E., Bauer, B. & Eggerer, H. (1983) Eur. J. Biochem. 133, 665-672.
33 34
Vol.371 (1990)
Lill, U, Schreil, A., Henschen, A. & Eggerer, H. (1984) Eur. J. Biochem. 143, 205-212. Löhlein-Werhahn, G., Goepfert, P. & Eggerer, H. (1985) Eur. J. Biochem. 150, 79-84. Lill, U., Bibinger, A. & Eggerer, H. (1987) Eur. J. Biochem. 162, 683-689. Lill, U., Bibinger, A. & Eggerer, H. (1987) Eur. J. Biochem. 163, 599-607. Löhlein-Werhahn, G., Goepfert, P., Kollmann-Koch, A. & Eggerer, H. (1988) Biol. Chem. Hoppe-Seyler 369.417424. Pettersson, G., Lill, U. & Eggerer, H. (1989) Eur. J. Biochem. 182, 119-124. Eggerer, H. & Klette, A. (1967) Eur. J. Biochem. 1, 447475. Eggerer, H., Buckel, W, Lenz, H., Wunderwald, P., Gottschalk, G., Cornforth, J.W., Donninger, C, Mallaby, R., Redmond, J.W. (1970) Nature (London) 226,517-519. Lenz, H., Buckel, W, Wunderwald, R, Biedermann, G.. Buschmeier, V., Eggerer, H., Cornforth, J.W. & Mallaby. R. (1971) Eur. J. Biochem. 24, 207-215. Retey, J., Lüthy, J. & Arigoni, D. (1970) Nature (London) 226, 519-521. Klinman, J. & Rose, I.A. (1971) Biochemistry 10, 22672272. Clark, J.D., CTKeefe, S.J. & Knowles, J.R. (1988) Biochemistry 27, 5961-5971. Wlassics, J.D. & Anderson, V.E. (1989) Biochemistry 28. 1627-1633. Johansson, CJ. & Pettersson, G. (1974) Eur. J. Biochem. 46,5-11. Bayer, E., Bauer, B. & Eggerer, H. (1981) FEBS Lett. 127, 101-104. Remington, S., Wiegand, G. & Huber, R. (1982) J. Mol. Biol. 158, 111-152. Buckel, W. & Eggerer, H. (1969) Hoppe-Seyler's Z. Physiol. Chem. 350, 1367-1376. Wunderwald. P. & Eggerer. H. (1969) Eur. J. Biochem. 11. 97-105. Eggerer, H., Giesemann, W. & Aigner, H. (1983) Liebigs Ann. Chem. 1551-1560. Farines, M. & Soulier, J. (1970) Bull. Soc. Chim. Fr. l, 332-340. Nau, C.A., Brown, E.B. & Bailey, J.R. (1925) J. Am. Chem. Soc. 47, 2596-2606. Rehn, D. & Ugi, I. (1977)7. Chem. Res. (M) 1501-1506; S (Synopsis) 119. Kaiser, G.V, Cooper, R.D.G., Koehler, R.E., Murphy, C.F., Webber, J.A., Wright, LG. & van Heyningen, E.M. (1970)7. Org. Chem. 35, 2429-2430. Hassner, A. & Alexanian, V. (1978) Tetrahedron Lett. 46, 4475-4478. Wang, S.S., Gisin, B.F., Winter, D.P., Makofske, R., Kuleska, I.D.,Tzougraki, C. & Meienhofer, J. (1977) J. Org. Chem. 42,1286-1290. Linn,T.C. & Srere, P.A. (1979) J. Biol. Chem. 254, 16911698. Bergmeyer, H.U., Graßl, M. & Walter, H.-E. (1983) in Methods of Enzymatic Analysis, 3rd edn. (Bergmeyer, H.U, Bergmeyer, J. & Graßl, M., eds.) vol. 2, pp. 173175, VCH Verlagsgesellschaft, Weinheim. Schroeter, G. (1905) Ber. Dtsch. Chem. Ges. 38, 31903210. Sustmann, R. & Korth, H.-G. (1985) in Houben-Weyl, Methoden der Organischen Chemie, vol. E5, part l (J. Falbe, ed.)Thieme Verlag, Stuttgart 496-504. Bruice, T.C. & Benkovic, S.J. (1966) Bioorganic Mechanisms, vol. I, pp. 173-186, W.A. Benjamin, New York. Morawetz, H. & Oreskes, I. (1958) J. Am. Chem. Soc. 80, 2591-2592.
Brought to you by | Purdue University Libraries Authenticated Download Date | 6/1/15 3:38 AM
Vex. 371 (1990)
Bifunctional Catalysis of Citrate Synthase
35 Morawetz, H. & Shafer, J. (1962) J. Am. Chem. Soc. 84, 3783-3784. 36 Handford, P., Ner, S.S., Bloxham, P.D. & Wilton, C.D. (1988) Biochim. Biophys. Acta 953, 232-240.
713
37 Lipscomb,W.N. (1982) Ace. Chem. Res. 15, 232-238. 38 Breslow, R. (1986) in Adv. Eniymol. 58, 1-60. 39 Karpusas. M., Branchaud, B. & Remington, S.J. (1990) Biochemistry 29. 2213-2219.
JJ. Wilde. U. Lill, H. Eggerer*, Institut für Physiologische Chemie der Technischen Universität München, IBiedersteiner Str. 29, D-8000 München 40. :
* To whom correspondence should be atddressed.
Brought to you by | Purdue University Libraries Authenticated Download Date | 6/1/15 3:38 AM
Brought to you by | Purdue University Libraries Authenticated Download Date | 6/1/15 3:38 AM
