The Mechanism of the Adenylosuccinate Synthetase Reaction as Studied by Positional Isotope Exchange*

In an attempt to gain insight into the mechanism of the rat muscle adenylosuccinate synthetase reaction, experiments using the technique of positional isotope exchange (isotope scrambling) were undertaken. [y'BO]GTP was prepared and incubated with Mg2+ and the synthetase in the presence of various ligands. Positional isotope exchange occurred, as measured by nuclear magnetic resonance spectroscopy, when IMP was present. In the absence of IMP, with or without aspartate or succinate, the [-Y-'~O]GTP did not exhibit scrambling. These results suggest that the adenylosuc- cinate synthetase reaction involves the participation of 6-phosphoryl-IMP as an obligatory intermediate. On the basis of experiments carried out in our labora-tory as well as in others, we believe the GDP remains bound to the enzyme until the product, adenylosuccin- ate, is formed. All products may then dissociate ran-domly from the enzyme. The positional isotope ex- change experiments, along with initial-rate experi- ments carried out in our laboratory, serve to explain the lack of partial exchange reactions associated with the synthetase (Fromm, H. J. Biochirn. Bio- phys. Acta 29, 265-262), as well as the net inversion of configuration when chiral thio-GTP is converted to thiophosphate Biol. Chem. 259,


3044-3046).
Adenylosuccinate synthetase (1MP:L-aspartate ligase (GDP-forming), EC 6.3.4.4) catalyzes the reaction: GTP + IMP + L-aspartate GDP + Pi i-adenylosuccinate Because this reaction is the first step in the biosynthesis of adenine nucleotides from IMP, the enzyme plays an important role in the regulation of purine nucleotide interconversion. For an extensive review of adenylosuccinate synthetase, see Stayton et al. (1983).
To date, there have been three mechanisms proposed for the adenylosuccinate synthetase reaction. The earliest was suggested by Lieberman (1956) and involves a 6-phosphoryl-10546 and CA-14030 from the National Institutes of Health, United *This research was supported in part by Research Grants NS-States Public Health Service, C-582 from the Robert A. Welch Foundation, and by Grant PCM-8101999 from the National Science Foundation. Journal Paper No. 5-11465 of the Iowa Agriculture and Home Economics Experiment Station, Ames, Ia, Project 2575. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. anan (1962), involves a concerted reaction in which all three substrates participate simultaneously. The third mechanism, proposed by Markham and Reed (1978), has aspartate attacking the C-6 of IMP in the first step of the reaction.

IMP intermediate. The second, proposed by Miller and Buch
Recently, Webb et al. (1984) have followed the stereochemical course of the adenylosuccinate synthetase reaction using chiral [160,'70,180]thiophosphate in the y position of GTP. They found that the reaction proceeds with net inversion of configuration. These results are consistent with all three mechanisms if one proposes direct phosphoryl transfer from the y position of GTP to the 0-6 of IMP, with subsequent cleavage of the carbon-oxygen bond. To postulate a phosphorylated enzyme intermediate, one would have to propose three phosphoryl transfers; however, it seems unlikely that there are two phosphoryl-enzyme intermediates.
Positional isotope exchange is technique by which one can investigate the existence of phosphoryl intermediates in a reaction mechanism (see Rose, 1979). By labeling the yphosphate of GTP with "0, one can observe the exchange of the label from the (3-7 bridge position to the nonbridge of GTP, if exchange does occur. If exchange does occur, it can be observed by two methods: mass spectroscopy and 31P NMR spectroscopy. With the method of 31P NMR spectroscopy, "0-labeled in the 0-y bridge position can be differentiated from "0-labeled in the @ nonbridge position. The difference between these two resonances was found to be 0.012 ppm for ATP (Cohn and Rao, 1979). If an exchange is observed, one can hypothesize a phosphoryl-enzyme or a phosphoryl-substrate as a reaction intermediate. Positional isotope exchange experiments were performed with adenylosuccinate synthetase to determine the reaction mechanism. We report that an exchange reaction occurs when [y-180]GTP is incubated with IMP. No exchange reaction was observed when IMP was absent. These findings support a reaction mechanism that involves a 6-phosphoryl-IMP intermediate.

EXPERIMENTAL PROCEDURES
Adenylosuccinate synthetase was purified from rat muscle by the method of Baugher (1980) and was dialyzed against a buffer containing 100 mM HEPES' (pH 7.0) and 5 mM dithiothreitol. ['*O]HzO (97 atom % purity) was obtained from MSD Isotopes (Merck and Co.).
[180]H3P04 was synthesized by the method given in Rhyu et al. (1984) from PC16 (which was previously purified by sublimation) and purified by the method of Hackney et al. (1980). [y-"O]GTP was synthesized by the method of Wehrli et al. (1965), except as noted below. GDP-morpholidate was eluted from a column (2.5 X 15 cm) of AG 1x2 (HCO; form) by a 4-liter linear gradient from 5 to 750 mM Et3NH. HC03. The [r-"O]GTP was purified by elution from a column (2 X 34 cm) of DEAE-cellulose (HCOi form) by a 3-liter linear gradient from 5 to 360 mM EtSNH. HCO,. The labeled GTP was assayed spectrophotometrically and enzymatically to give a yield of 62 pmol. This represents a yield of 52% from GDP-morpholidate: All the NMR experiments were performed on a Bruker 300 Fourier Transform NMR Spectrometer operating at 121.5 MHz. Data were collected over 1600 scans, with a pulse width of 24 ps (approximately a 60" pulse angle). The data were transformed by using a 0.2 Hz line broadening. All samples contained 1.5 mM [r-180]GTP, 100 mM HEPES (pH 7.0), 10% D20, 3.5 mM magnesium acetate, and 0.16 units of adenylosuccinate synthetase. The samples were incubated at 24 "C for 17 h. (The enzyme showed no loss of activity when incubated at 22 "C for 60 h.) The reactions were stopped by adding CDTA (adjusted to pH 7.0) to give a final concentration of 30 mM. The total volume was adjusted to 2.0 ml. The samples then were stored on ice until ready for NMR analysis. The samples also carried a capillary containing 10% H,PO, during data acquisition. The phosphoric acid resonance was used as an external standard and set to 0.0 ppm. equimolar unlabeled GTP added (Fig. 1B). The total shift between major peaks of the y31P spectrum was 0.105 ppm. The value reported for ATP is 0.085 ppm (Cohn and Rao, 1979). In the fl region of the spectrum, the difference was 0.018 ppm, compared with a literature value of 0.016 ppm for ATP. The minor resonance seen in the y spectrum was found to have a shift of 0.027 ppm. This compares quite well with the expected shift for a nonlabeled oxygen atom in the y nonbridge position. Using the integrated areas of the peaks, we found the isotopic enrichment of the y-phosphate to be 91.2%. Fig. 2 depicts the spectra of the @ peak of labeled GTP alone (A), after incubation with adenylosuccinate synthetase ( B ) , after incubation with 1.5 mM IMP ( C ) , and after incubation with 1.5 m M IMP and 10 mM succinate (D). Fig. 3 illustrates the y resonances for the same set of experiments. Succinate, a competitive inhibitor for aspartate (Rudolph and Fromm, 1969), decreased the extent of exchange as compared with incubation with IMP alone. Table I summarizes the differences in the peak positions for all the experiments conducted. From these data, it can be seen that there is no exchange in the absence of IMP or in the presence of ligands that are known to bind to the enzyme in the absence of IMP (Rudolph and Fromm, 1969). When IMP is present, there was an exchange of isotope between the fl-y bridge and the @ nonbridge positions of GTP. This exchange was decreased when succinate was added to the incubation medium. In the experiment in which all the substrates were present (data not presented), an equilibrium mixture resulted. This led to the generation of peaks corresponding to phosphate with 0, 1, 2, and 3 ' ' 0 atoms (see also Cohn and Hu, 1978).

DISCUSSION
Three reaction mechanisms for adenylosuccinate synthetase have been suggested. In 1956, Lieberman proposed a 6phosphoryl-IMP intermediate. This mechanism involves a nucleophilic attack by the 0-6 of IMP on the y-phosphorus of GTP, generating the 6-phosphoryl-IMP intermediate. To date, this intermediate has eluded chemical synthesis. In the next step, the a-nitrogen of aspartate makes a nucleophilic attack on the C-6 of IMP, leading to a tetrahedral transition state. This breaks down to yield adenylosuccinate and orthophosphate. Lieberman based this mechanism on the finding that all the label in [6-'*0]IMP ended up as ["0]Pi and because there was no GTP e ["PIPi exchange when aspartate was missing from the reaction mixture. Fromm (1958) generated results from isotope exchange at equilibrium experiments that also support the hypothesis of a 6-phosphoryl-IMP intermediate. Lieberman's mechanism is depicted in l a of Scheme 1.
In 1962, Miller and Buchanan suggested a concerted mechanism for adenylosuccinate synthetase. In this mechanism, the a-nitrogen of aspartate makes a nucleophilic attack on the C-6 of IMP at the same time that the 0 -6 of IMP makes its attack on the y-phosphorus of GTP, thus leading to the tetrahedral transition state. This mechanism was proposed to account for the finding that arsenolysis or phosphorolysis of adenylosuccinate does not occur when GDP is missing from the reaction mixture. It also is consistent with the lack of partial exchange reactions for adenylosuccinate synthetase. This mechanism is shown in part I b of Scheme 1.
A third mechanism was proposed by Markham and Reed in 1978. The first step of this mechanism has the a-nitrogen of aspartate attacking the C-6 of IMP creating an "oxy-anion'' intermediate. The activated oxygen then makes a nucleophilic attack on the y-phosphorus of GTP leading to the tetrahedral transition state. This study used GTPyS, which lowers the turnover number of the enzyme by 12.5-fold. At this lower rate, Markham and Reed (1978) were able to observe a transient in the UV spectrum. They attributed this transient to the oxy-anion intermediate, reasoning that phosphorylation of the N-6 of AMP leads to a spectral red shift that is in the opposite direction of the transient that was observed with GTPyS. The Markham and Reed mechanism is outlined in part I C of Scheme 1.
The data from the experiments reported in this paper show that positional isotope exchange did take place between the P-y bridge and the P nonbridge positions of GTP in the presence of adenylosuccinate synthetase. The finding that the exchange took place only in the presence of IMP is indicative of a 6-phosphoryl-IMP intermediate. This would support the mechanism originally proposed by Lieberman (1956). The other two mechanisms can be eliminated since they require aspartate be present before the GDPO-P03 bond is broken.
Data used to support other models can be interpreted to fit mechanism l a of Scheme 1 as follows. The finding that arsenolysis does not occur without GDP being present (Miller and Buchanan, 1962) can be interpreted as a need by the enzyme for the guanosine nucleotide for catalytic activity. In the absence of GDP, the enzyme cannot enter into a catalytic conformation. This is supported by the finding that GTP and IMP bind synergistically to the enzyme (Markham and Reed, 1978).
The sequential kinetics reported for adenylosuccinate synthetase from a variety of sources (Rudolph and Fromm, 1969;Nagy et al., 1973;Van Der Weyden and Kelly, 1974;Clark et al., 1977;Baugher, 1980) are consistent with the participation of 6-phosphoryl-IMP as an intermediate, if it is assumed that all substrates and products remain bound to the enzyme until adenylosuccinate is formed (Fromm, 1975). This proposal also would account for the lack of partial exchange reactions.
The results of this report on positional isotope exchange, along with the findings of Webb et al. (1984), which demon-

YOo"
' i : I CHCH,C@ RSP RSP R I P SCHEME 1 strate the inversion of configuration when GTP is cleaved, are consistent with 6-phosphoryl-IMP being an intermediate in the adenylosuccinate synthetase reaction. Other proposals advanced to explain the mechanism of the reaction seem not in harmony with the results of the present investigation.