The mechanism of the reaction catalyzed by ADP-β-L-glycero-D-manno- heptose 6-epimerase

Article abstract of DOI:10.1021/ja0485659

ADP-L-glycero-D-manno-heptose 6-epimerase (AGME, RfaD) is a bacterial enzyme that is involved in lipopolysaccharide biosynthesis and interconverts ADP-β-L-glycero-D-manno-heptose (ADP-L,D-Hep) with ADP-β-D-glycero-D-manno-heptose (ADP-D,D-Hep). AGME is kn

Full text of DOI:10.1021/ja0485659

Published on Web 06/30/2004
The Mechanism of the Reaction Catalyzed by
ADP-â-L-glycero-D-manno-heptose 6-Epimerase
Jay A. Read,† Raef A. Ahmed,† James P. Morrison,† William G. Coleman, Jr.,‡ and
Martin E. Tanner*,†
Department of Chemistry, UniVersity of British Columbia, VancouVer, British Columbia, Canada, and
the Laboratory of Biochemistry and Genetics, National Institute of Diabetes and DigestiVe and Kidney Diseases,
National Institutes of Health, Bethesda, Maryland
Received March 12, 2004; E-mail: mtanner@chem.ubc.ca
ADP-L-glycero-D-manno-heptose 6-epimerase (AGME, RfaD) is
a bacterial enzyme that interconverts the unusual seven-carbon
â-linked sugar nucleotides ADP-â-L-glycero-D-manno-heptose (ADP-
L,D-Hep) and ADP-â-D-glycero-D-manno-heptose (ADP-D,D-Hep)
(Figure 1).1 In Gram-negative bacteria, AGME plays a key role in
the biosynthesis of cell surface lipopolysaccharide (LPS) as it
provides the L,D-Hep that is a component of the inner core
tetrasaccharide.2 Mutant strains of Escherichia coli that lack a
functional AGME produce truncated LPS chains, have an increased
susceptibility to hydrophobic antibiotics, and show decreased
pathogenicity.3-5
Figure 1. Reaction catalyzed by AGME and structure of deoxy analogues
(inset).
The mechanism by which AGME catalyzes the inversion of
stereochemistry is of interest since the C-6′′ stereocenter of the
substrate lacks an acidic C-H bond and a direct deprotonation/
reprotonation mechanism cannot be employed.6,7 While few
mechanistic studies have been undertaken, it is known that the
enzyme bears a tightly bound NADP+ cofactor that is presumably
utilized to transiently oxidize the substrate.8-10 Sequence alignments
and X-ray crystallographic studies support this hypothesis by
showing that AGME is a member of the short-chain dehydrogenase/
reductase (SDR) superfamily.9,11 Several members of this family
are known to catalyze reactions on sugar nucleotides that involve
transient oxidations of the sugar moiety (usually at C-4′′).12,13 One
potential mechanism for the AGME reaction involves an initial
oxidation at C-7′′ that serves to acidify the proton at C-6′′ (Scheme
1, path A). A subsequent deprotonation at C-6′′, followed by
reprotonation on the opposite face of the enol(ate) intermediate and
reduction of the C-7′′ aldehyde, generates the epimeric product.
An alternative mechanism involves oxidation directly at C-6′′
(Scheme 1, path B). A conformational change would then serve to
expose the opposite face of the carbonyl to the cofactor, and
reduction would give the epimeric product. Last, the transient
oxidation may take place at C-4′′ (Scheme 1, path C). A nonste-
reospecific dehydration/rehydration process could invert the ster-
eochemistry at C-6′′, and subsequent reduction of the C-4 carbonyl
would generate the product. A nonstereospecific retroaldol/aldol
process could also be used to invert the stereochemistry at C-6′′ of
the 4-keto intermediate (not shown).
with the recombinant E. coli AGME and the reactions were
monitored by both 1H- and 31P NMR spectroscopy, an equilibrium
mixture of 2.3:1 ADP-L,D-Hep/ADP-D,D-Hep was observed. Samples
of the epimeric mixture formed under multiple turnover conditions
were analyzed for deuterium content by both 1H NMR spectroscopy
and mass spectrometry. No detectable levels of deuterium were
observed, indicating that no more than 1% of the enzymatic
turnovers results in solvent isotope incorporation. A similar
experiment was run using a 50:50 mixture of H218O/H216O as
solvent, and no 18O-label could be detected in the mass spectrum
of the resulting epimeric mixture. The absence of deuterium
incorporation argues against a mechanism involving deprotonation/
reprotonation at C-6 (Scheme 1, path A). In all known racemases
and epimerases that catalyze nonstereospecific proton transfer steps,
a two-base mechanism is employed, and solvent deuterium is
stoichiometrically incorporated into the product.6,16,17 Similarly, the
absence of 18O-isotope incorporation argues against the dehydration/
rehydration mechanism (Scheme 1, path C) since the same molecule
of water formed upon dehydration would have to be efficiently
redelivered to the opposite face of the enone intermediate.
In further studies, the 7-deoxy and 4-deoxy analogues of ADP-
L,D-Hep (1a and 2a, respectively) were synthesized.15 These
compounds were incubated with AGME, and the resulting reactions
were monitored using both 1H- and 31P NMR spectroscopy. In each
case, a new set of signals emerged that ultimately equilibrated to
an approximately equimolar mixture of two species (Figure 2). The
31P NMR spectra identified the new compounds as sugar nucle-
otides, and the mass spectra of the isolated mixtures were
indistinguishable from those of the starting compounds. These
results are consistent with the notion that AGME interconverts both
substrate analogues with epimeric sugar nucleotides of similar
stability. In the case of compound 1a, the H-5′′ signals for the
starting material and product were sufficiently well separated so
that a 1D-TOCSY experiment could be used to fully assign the
spectrum of the new heptose moiety. Coupling constants were
consistent with a mannose configuration for the stereochemistry at
In this communication we report evidence in favor of a
mechanism involving transient oxidation directly at C-6′′ (Scheme
1, path B). Epimerization in either D2O or H218O was not
accompanied by solvent isotope incorporation, and the 7-deoxy and
4-deoxy analogues of ADP-L,D-Hep (Figure 1, 1a and 2a, respec-
tively) were found to be alternate substrates for the enzyme.
ADP-L,D-Hep and ADP-D,D-Hep were obtained via chemical
synthesis.14,15 When either substrate was incubated in buffered D2O
† Department of Chemistry, University of British Columbia.
‡ Laboratory of Biochemistry and Genetics, National Institute of Diabetes and
Digestive and Kidney Diseases, National Institutes of Health.
9
8878
J. AM. CHEM. SOC. 2004, 126, 8878-8879
10.1021/ja0485659 CCC: $27.50 © 2004 American Chemical Society

C O M M U N I C A T I O N S
Scheme 1
carbons 2′′-5′′, indicating that epimerization had occurred at C-6′′
and the new compound was 1b.15 In the case of compound 2a,
however, the 1H NMR signals of the two species were not
sufficiently separated to employ this technique. Instead, a nonste-
reoselective synthesis was used to generate an authentic 1.8:1
mixture of 2a/2b.15 A comparison of the 1H- and 31P NMR spectra
of the enzymatic mixture to that of the synthetic mixture confirmed
that the new product was 2b (Figure 2). Furthermore, when the
synthetic 1.8:1 mixture of 2a/2b was incubated with AGME, it was
converted into an equimolar mixture. The ability of 1 and 2 to serve
as substrates for AGME rules out all mechanisms that require
oxidation at C-4′′ or C-7′′ (paths A and C).
flexibility with regard to the position of oxidation is somewhat
reminiscent of the reaction catalyzed by CDP-tyvelose 2-epimerase,
an SDR family member that is thought to employ transient oxidation
at C-2′′.25,26
Acknowledgment. This work was supported by the Canadian
Institutes of Health Research (M.E.T.) and NCMHD/NIH (W.G.C.).
Supporting Information Available: Experimental procedures, 1H
NMR spectra of sugar nucleotides, and time courses of enzymatic
reactions. This material is available free of charge via the Internet at
http://pubs.acs.org.
As AGME is a member of the SDR superfamily, one might
expect it would utilize a strategy involving transient oxidation at
C-4′′.12,13 In particular, a notable similarity exists between the first
two steps in path C and those employed by the sugar nucleotide
4,6-dehydratases.18-21 However, both the solvent isotope incorpora-
tion studies and the deoxy-analogue studies indicate that transient
oxidation actually occurs at C-6′′. Thus, AGME appears to employ
a nonstereospecific oxidation/reduction mechanism (path B). As
with other SDR family members, at least one of the required acid/
base residues would likely be supplied by the conserved triad,
Ser116, Tyr140, Lys144.9 This strategy is conceptually similar to
that of the SDR enzyme UDP-galactose 4-epimerase; however, both
the position of oxidation and the required reorientation of the
oxidized intermediate are quite different.6,22-24 This display of
References
(1) Valvano, M. A.; Messner, P.; Kosma, P. Microbiology 2002, 148, 1979-
1989.
(2) Raetz, C. R. H.; Whitfield, C. Annu. ReV. Biochem. 2002, 71, 635-700.
(3) Coleman, W. G., Jr.; Leive, L. J. Bacteriol. 1979, 139, 899-910.
(4) Vaara, M. Antimicrob. Agents Chemother. 1993, 37, 354-356.
(5) Taylor, P. W. Microbiol. ReV. 1983, 47, 46-83.
(6) Samuel, J.; Tanner, M. E. Nat. Prod. Rep. 2002, 19, 261-277.
(7) Allard, S. T. M.; Giraud, M.-F.; Naismith, J. H. Cell. Mol. Life Sci. 2001,
58, 1650-1665.
(8) Ni, Y.; McPhie, P.; Deacon, A.; Ealick, S.; Coleman, W. G., Jr. J. Biol.
Chem. 2001, 276, 27329-27334.
(9) Deacon, A. M.; Ni, Y. S.; Coleman, W. G., Jr.; Ealick, S. E. Structure
2000, 8, 453-461.
(10) Ding, L.; Seto, B. L.; Ahmed, S. A.; Coleman, W. G., Jr. J. Biol. Chem.
1994, 269, 24384-24390.
(11) A structure of AGME complexed with ADP-R-D-glucose is available (ref
9); however, the electron density of the hexose moiety was disordered
and difficult to interpret. This is likely the result of having the incorrect
stereochemistry at both C-1′′ and C-2′′ of the substrate analogue, and it
complicates any mechanistic interpretations regarding the position of
transient oxidation.
(12) Field, R. A.; Naismith, J. H. Biochemistry 2003, 42, 7637-7647.
(13) Tanner, M. E. Curr. Org. Chem. 2001, 5, 169-192.
(14) Zamayatina, A.; Gronow, S.; Puchberger, M.; Graziani, A.; Hofinger, A.;
Kosma, P. Carbohydr. Res. 2003, 338, 2571-2589.
(15) See Supporting Information for details.
(16) Tanner, M. E. Acc. Chem. Res. 2002, 35, 237-246.
(17) Tanner, M. E.; Kenyon, G. L. In ComprehensiVe Biological Catalysis;
Sinnott, M., Ed.; Academic Press: San Diego, 1998; Vol. II, pp 7-41.
(18) He, X. M.; Agnihotri, G.; Liu, H.-W. Chem. ReV. 2000, 100, 4615-4661.
(19) Somoza, J. R.; Menon, S.; Schmidt, H.; Joseph-McCarthy, D.; Dessen,
A.; Stahl, M. L.; Somers, W. S.; Sullivan, F. X. Structure 2000, 8, 123-
135.
(20) Hegeman, A. D.; Gross, J. W.; Frey, P. A. Biochemistry 2002, 41, 2797-
2804.
(21) Beis, K.; Allard, S. T. M.; Hegeman, A. D.; Murshudov, G.; Philp, D.;
Naismith, J. H. J. Am. Chem. Soc. 2003, 125, 11872-11878.
(22) Wei, Y.; Lin, J.; Frey, P. A. Biochemistry 2001, 40, 11279-11287.
(23) Berger, E.; Arabshahi, A.; Wei, Y.; Schilling, J. F.; Frey, P. A.
Biochemistry 2001, 40, 6699-6705.
(24) Thoden, J. B.; Holden, H. M. Biochemistry 1998, 37, 11469-11477.
(25) Koropatkin, N. M.; Liu, H.-W.; Holden, H. M. J. Biol. Chem. 2003, 278,
20847-20881.
(26) Hallis, T. M.; Zhao, Z.; Liu, H.-W. J. Am. Chem. Soc. 2000, 10493-10503.
Figure 2. 31P-NMR spectra illustrating the epimerization of 2a.
JA0485659
9
J. AM. CHEM. SOC. VOL. 126, NO. 29, 2004 8879