Biochemistry zyxwvu 1981, 20, 3719-3723 Stereochemical Outcome of Processing of Fluorinated Substrates by ATP Citrate Lyase and Malate Synthase+ Michael A. Marietta,: Paul A. Srere, and Christopher Walsh* ABSTRACT: The (-)-erythro- and (+)-erythro-2-fluorocitrate isomers are substrates for the cytoplasmic ATP citrate lyase from rat liver with zyxwvutsr V, values 0.13% and 2.3%, respectively, that of citrate when the keto acid products are measured by in situ reduction by NADH and malate dehydrogenase. Anticipated regiospecificity of cleavage by ATP citrate lyase is production of 2-fluoroacetyl-CoA and oxalacetate from a (2R,3R)-2-fluorocitrate isomer and production of acetyl-coA and (3S)-fluorooxalacetate from cleavage of the other erythro diastereomer, the zyxwvutsr (2S,3S)-2-fluorocitrate. The product com- binations were validated by in situ reduction with MDH and (4R)-[’H]NADH to yield ~-(2S)-[2-’H]rnalate from zyxwvut (-)- erythro-2-fluorocitrate and to yield ~-(2R,3S)-3-fluoro[2- ’HJrnalate only from the (+)-erythro-2-fluorocitrate. The data allow for the first time unambiguous assignment of absolute stereochemistry to the (+)-erythro-2-fluorocitrate (2S,3S) and thereby necessarily to the (-)-erythro-2-fluorocitrate as 2R,3R. s u b s t r a t e analogues with specific replacement of H by F have proved useful in enzymology since enzymes often proces~ such molecules catalytically and thereby reveal mechanistic facets of their action. The small size of fluorine and the short C-F bond are a close steric mimic to the C-H bond, but the high electronegativity of fluorine makes the polarization of the C-F bond more akin to that of a C-OH bond. There are enzymatic precedents for apparent recognition of C-F as either a C-H or C-OH substituent (Filler, 1976). One recent use of fluorine replacement for hydrogen has been to probe en- zymatic generation of carbanionic intermediates at an adjacent site by detection of net H-F elimination sequence (Silverman & Abeles, 1976, 1977; Wang & Walsh, 1978; Kollonitsch et al., 1978; Seiler et al., 1978). With several PLP enzymes, such fluorinated substituted molecules behave as suicide substrates when the subsequent enamine intermediate is captured by an enzymatic nucleophile. On the other hand, fluorine substi- tution in 5-FdUMP’ (Hartmann & Heidelberger, 1961) generates mechanism-based inactivation of thymidylate synthetase precisely because the enzyme cannot break the C5-F bond by F+ abstraction (Santi et al., 1976). We have been investigating the consequences of enzymatic processing of fluorinated analogues at the C-F locus in re- actions where the C-F bond is itself not broken. t From the Departments of Chemistry and Biology, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139 (M.A.M. and C.W.), and the Department of Biochemistry and Veterans Administra- tion Medical Center, The University of Texas Health Science Center at Dallas, Dallas, Texas 15235 (P.A.S.). Received April 29,1980; revised manuscript received February 12,1981. This investigation was supported in part by National Institutes of Health Grant 6M20011 and the Vet- erans Administration. *Address correspondence to this author at the Department of Chtm- istry, M.I.T. NIH Postdoctoral Fellow l-F32-6M-6429-01. Present address: Department of Nutrition and Food Science, M.I.T., Cambridge, MA 02139. 3719 This latter isomer is, of course, the toxic species generated by the “lethal synthesis” from fluoroacetyl-CoA and oxalacetate effected by citrate synthase. Assuming inversion in the citrate synthase reaction, one can now state that it processes 2- flugroacetyl-CoA chirally with specific zyxw pro4 hydrogen ab- straction in the condensation. Cleavage of (2R,3R)-2- fluorocitrate by ATP citrate lyase in ’H20 should yield a product that is chiral, (2S))-fluor0[2-~H]acetyl-CoA. Attempts to analyze this product involved in situ coupling with [I4C]- glyoxalate and yeast malate synthase. Surprisingly, both erythro- and threo-3-fluoromalates formed in about equal amounts, and each had both 3H and I4C radioactivity. These results suggest that malate synthase, in contrast to citrate synthase, processes one of its two substrates achirally at a prochiral center, either C2 of fluoroacetyl-CoA or, less likely, the trigonal aldehyde carbon of glyoxalate during fluoromalate biosynthesis. I I I ‘b ‘b In reversible interconversions of fluoromethyl and fluro- methylene groups, the stereochemical outcome of the reaction reflects chirality of enzymatic recognition. For example, fluoropyruvate is carboxylated to fluorooxalacetate by trans- carboxylase with stereospecific removal of only one enantio- topic hydrogen, but fluoropyruvate is enolized achirally by pyurvate kinase (Goldstein et al., 1978). The simplest stable fluorinated carboxylate is the naturally occurring fluoroacetate (Peters, 1972), which is toxic by virtue of its chiral recognition (as fluoroacetyl-CoA) and conden- sation with oxalacetate to a single 2-fluorocitrate2 diastereomer (Fanshier et al., 1964) [(-)-erythro], catalyzed by citrate synthase, the celebrated “lethal synthesis” reaction (Peters, 1972). In the experiments noted here, we report an apparent achiral processing of fluoroacetyl-CoA by malate synthase. The fluoroacetyl-CoA was generated in situ from cleavage of (-)-erythro-fluorocitrate by rat liver ATP citrate lyase (citrate Abbreviations used: 5-FdUMP, 5-fluorouridine 5’-monophosphate; CCE, citrate cleavage enzyme (ATP citrate lyase); MDH, malate de- hydrogenase; NADH, reduced nicotinamide adenine dinucleotide; ATP, adenosine 5’-triphosphate; CoA, coenzyme A; HPLC, high-performance liquid chromatography; Tris, tris(hydroxymethy1)aminomethane; t-Fm, threo-3-fluoromalate; e-Fm, erythro-3-fluoromalate; LDH, lactate de- hydrogenase; DTT, dithiothreitol; PLP, pyridoxal phosphate. The numbering system for the fluorwitrates zyx used in this paper is that which has developed historically and is the most widely quoted; however, IUPAC nomenclature states that the portion of the citrate backbone derived from acetate be numbered 4 and 5. Therefore, under this system, the (-)-erythro isomer would be 4-fluorocitrate and the (+)-erythro would be a 2-fluorwitrate. Another numbering system is that designated as parent numbering (pn) where the pro-R branch has lower numbering than the pro-S branch [(1979) Bioorg. Chem. 2, 301-3101. 0006-2960/81/0420-3719$01.25/0 B 1981 American Chemical Society