Turnover Is rate-limited by deglycosylation for micromonospora viridifaciens sialidase-catalyzed hydrolyses: Conformational implications for the Michaelis complex

Article abstract of DOI:10.1021/ja109199p

A panel of seven isotopically substituted sialoside natural substrate analogues based on the core structure 7-(5-acetamido-3,5-dideoxy-d-glycero- α-d-galacto-non-2-ulopyranosylonic acid)-(2→6)-β-d- galactopyranosyloxy)-8-fluoro-4-methylcoumarin (1, Neu5Acα2,6GalβFMU) have been synthesized and used to probe the rate-limiting step for turnover by the M. viridifaciens sialidase. The derived kinetic isotope effects (KIEs) on k_cat for the ring oxygen (¹⁸V), leaving group oxygen (¹⁸V), anomeric carbon (¹³V), C3-carbon (¹³V), C3-R deuterium (^DV_R), C3-S deuterium (^DV _S), and C3-dideuterium (^D₂V) are 0.986 ± 0.003, 1.003 ± 0.005, 1.021 ± 0.006, 1.001 ± 0.008, 1.029 ± 0.007, 0.891 ± 0.008, and 0.890 ± 0.006, respectively. The solvent deuterium KIE (D₂OV) for the sialidase-catalyzed hydrolysis of 1 is 1.585 ± 0.004. In addition, a linear proton inventory was measured for the rate of hydrolysis, under saturating condition, as a function of n, the fraction of deuterium in the solvent. These KIEs are compatible with rate-determining cleavage of the enzymatic tyrosinyl β-sialoside intermediate. Moreover, the secondary deuterium KIEs are consistent with the accumulating Michaelis complex in which the sialosyl ring of the carbohydrate substrate is in a ⁶S₂ skew boat conformation. These KIE measurements are also consistent with the rate-determining deglycosylation reaction occurring via an exploded transition state in which synchronous charge delocalization is occurring onto the ring oxygen atom. Finally, the proton inventory and the magnitude of the solvent KIE are consistent with deglycosylation involving general acid-catalyzed protonation of the departing tyrosine residue rather than general base-assisted attack of the nucleophilic water.

Full text of DOI:10.1021/ja109199p

ARTICLE
pubs.acs.org/JACS
Turnover Is Rate-Limited by Deglycosylation for Micromonospora
viridifaciens Sialidase-Catalyzed Hydrolyses: Conformational
Implications for the Michaelis Complex
Jefferson Chan, April Lu, and Andrew J. Bennet*
Departments of Chemistry and Molecular Biology and Biochemistry, Simon Fraser University, 8888 University Drive, Burnaby, British
Columbia V5A 1S6, Canada
S Supporting Information
b
ABSTRACT: A panel of seven isotopically substituted sialoside natural substrate analogues based on
the core structure 7-(5-acetamido-3,5-dideoxy-D-glycero-R-D-galacto-non-2-ulopyranosylonic acid)-(2f6)-
β-D-galactopyranosyloxy)-8-fluoro-4-methylcoumarin (1, Neu5AcR2,6GalβFMU) have been synthesized
and used to probe the rate-limiting step for turnover by the M. viridifaciens sialidase. The derived kinetic
isotope effects (KIEs) on kcat for the ring oxygen (18V), leaving group oxygen (18V), anomeric carbon (13V),
C3-carbon (13V), C3-R deuterium (DVR), C3-S deuterium (DVS), and C3-dideuterium (D V) are 0.986 (
2
0.003, 1.003 ( 0.005, 1.021 ( 0.006, 1.001 ( 0.008, 1.029 ( 0.007, 0.891 ( 0.008, and 0.890 ( 0.006,
respectively. The solvent deuterium KIE (D OV) for the sialidase-catalyzed hydrolysis of 1 is 1.585 ( 0.004. In
2
addition, a linear proton inventory was measured for the rate of hydrolysis, under saturating condition, as a
function of n, the fraction of deuterium in the solvent. These KIEs are compatible with rate-determining
cleavage of the enzymatic tyrosinyl β-sialoside intermediate. Moreover, the secondary deuterium KIEs are consistent with the accumulating
Michaelis complex in which the sialosyl ring of the carbohydrate substrate is in a 6S2 skew boat conformation. These KIE measurements are
also consistent with the rate-determining deglycosylation reaction occurring via an exploded transition state in which synchronous charge
delocalization is occurring onto the ring oxygen atom. Finally, the proton inventory and the magnitude of the solvent KIE are consistent with
deglycosylation involving general acid-catalyzed protonation of the departing tyrosine residue rather than general base-assisted attack of the
nucleophilic water.
’ INTRODUCTION
rate-determining steps.6,16-18 Specifically, negative βlg values
suggest that glycosylation is at least partially rate-limiting for the
appropriate kinetic term, as is the case with Vibrio cholerae
sialidase (βlg(V/K) = -0.73; βlg(V) = -0.25).17 The βlg values
on V/K and V for Micromonospora viridifaciens sialidase-catalyzed
hydrolyses were reported to be -0.30 þ 0.04 and 0.02 þ 0.03,
respectively.6 Together, these results suggest that glycosidic
bond cleavage is at least partially rate-limiting for kcat/Km, but
that kcat is limited by a subsequent step. In other words, the
possible rate-determining steps for enzymatic turnover (kcat) are:
(i) deglycosylation of the sialosyl-enzyme intermediate; (ii) a
conformational change of the initial bound R-sialic acid:enzyme
product complex; or (iii) product release. Given that only
deglycosylation, of the possible rate-limiting steps, involves
covalent bond formation and cleavage, it is likely that kinetic
isotope effects (KIEs) on V will allow a distinction to be made
between these three possibilities.
exo-Sialidases (neuraminidases; EC 3.2.1.18) are a class of
glycosyl hydrolases (GH) that catalyze the hydrolysis of the R-
ketosidic bond linking terminal sialic acid residues to various
glycoconjugates. These enzymes have been identified to play
important roles in a number of human pathologies, including
cancer,1,2 cholera,3 and influenza.4 All exo-sialidases, character-
ized to date, are retaining glycosidases in which a double
inversion of configuration occurs so that the first released
product has the same anomeric stereochemistry as that of the
substrate.5 It has been established through mutagenesis6-9 and
fluoro-sugar labeling studies10,11 that an active-site tyrosine
residue acts as a nucleophile during the catalytic cycle to generate
a covalently linked β-sialosyl-enzyme intermediate. This glyco-
sylation step has been proposed to proceed via either concerted
SN2 (ANDN)12 or stepwise SN1 (DN þ AN)13 pathways. Sub-
sequent hydrolysis of the enzyme bound intermediate followed
by product release completes the catalytic cycle. The sialidase
from the soil bacterium Micromonospora viridifaciens displays a
very high catalytic efficiency (kcat/Km > 107 M-1 s-1, for
activated substrates),6,14 broad substrate specificity,14 and a
remarkable tolerance to mutation of its active site residues.6-9
The mechanism for sialidase-catalyzed hydrolysis has been
probed by Brønsted analyses,15 where the derived βlg values
have provided valuable information regarding the identity of
The current study expands upon a previously established
coupled-enzyme assay using natural sialoside analogues, which
was used to determine enzymatic rate constants,19 for the
measurement of a series of secondary deuterium and heavy-atom
Received: October 13, 2010
Published: February 15, 2011
r
2011 American Chemical Society
2989
dx.doi.org/10.1021/ja109199p J. Am. Chem. Soc. 2011, 133, 2989–2997
|

Journal of the American Chemical Society
ARTICLE
J = 5.0, 1H, H-1), 4.64 (dd, J = 7.9, 2.4, 1H, H-3), 4.37 (dd, J = 5.0, 2.4,
1H, H-2), 4.30 (dd, J = 7.9, 1.6, 1H, H-4), 3.95-3.86 (m, 2H, H-5, H-6),
3.78 (dd, J = 15.7, 7.8, 1H, H-60), 2.13 (bs, 1H, OH), 1.56 (s, 3H, CH3),
1.49 (s, 3H, CH3), 1.37 (s, 6H, 2 ꢀ CH3). 13C NMR (151 MHz,
CDCl3): δ 109.01, 108.21, 95.84, 71.18, 70.30, 70.11, 67.57, 61.94,
25.57, 25.46, 24.47, 23.83. Anal. Calcd for C12H20FO6: C, 55.37; H, 7.74.
Found: C, 55.17; H, 7.81.
6-O-(18O)-Benzoyl-1,2:3,4-di-O-isopropylidene-R-D-(6-18O)galactopy-
ranose (3). A flame-dried flask was charged with 2 (500 mg, 1.92 mmol),
(18O2)-benzoic acid (300 mg, 2.38 mmol),12 PPh3 (655 mg, 2.49 mmol),
and anhydrous THF (25 mL). The reaction mixture was cooled in an ice
bath and treated with a 40% solution of diethyl azodicarboxylate in toluene
(1.1 mL, 2.3 mmol). Afterbeing stirredat room temperature overnight, the
solvent was removedunder reduced pressure, and the resultant residue was
purified via flash chromatography (3:17 v/v EtOAc/hexanes) to afford the
desired product as a syrup (371 mg, 0.96 mmol, 50.0%). 1H NMR (600
MHz, CDCl3): δ 8.13-8.04 (m, 2H, Ar-H), 7.63-7.54 (m, 1H, Ar-H),
7.51-7.42 (m, 2H, Ar-H), 5.60 (d, J = 5.0, 1H, H-1), 4.68 (dd, J = 7.9, 2.5,
1H, H-3), 4.56 (dd, J = 11.5, 4.9, 1H), 4.46 (dd, J = 11.5, 7.6, 1H), 4.40-
4.34 (m, 2H, H-2), 4.26-4.17 (m, 1H, H-5), 1.55 (s, 3H, CH3), 1.50 (s,
3H, CH3), 1.39 (s, 3H, CH3), 1.36 (s, 3H, CH3). 13C NMR (151 MHz,
CDCl3): δ 165.92, 132.51, 129.60, 129.24, 127.87, 109.23, 108.35, 95.87,
70.68, 70.27, 70.08, 65.68, 63.37, 25.55, 25.51, 24.51, 24.03.
Figure 1. Isotopically labeled compounds used in the measurement of
KIEs on kcat for M. viridifaciens sialidase-catalyzed hydrolysis of the
natural substrate analogue Neu5AcR2,6GalβFMU.
KIEs using a panel of seven isotopically labeled substrates
(Figure 1).
’ MATERIALS AND METHODS
Materials. Cytidine 50-triphosphate disodium salt was purchased
from 3B Scientific Corp. All other reagents were purchased from Aldrich
and used without purification. Thin-layer chromatography (TLC) was
performed on aluminum-backed TLC plates precoated with Merck silica
gel 60 F254. Compounds were visualized with UV light and/or staining
with a p-anisaldehyde solution. Flash chromatography was performed
using silica gel 60 (230-400 mesh). Solvents used for anhydrous
reactions were dried and distilled immediately prior to use. THF was
dried and distilled over sodium metal, and dichloromethane was dried
and distilled over calcium hydride. For anhydrous reactions, all glassware
was flame-dried and cooled under a nitrogen atmosphere immediately
prior to use. NMR spectra were recorded on a Bruker 600 MHz
spectrometer. Chemical shifts (δ) are listed in ppm downfield from
1,2,3,4-Tetra-O-acetyl-6-O-(18O)benzoyl-D-(6-18O)galactopyranose
(4). A solution of 3 (300 mg, 0.82 mmol) in 80% aqueous AcOH
(10 mL) was stirred at 75 °C for 30 min. The solvent was evaporated
under reduced pressure to yield a syrup that was then treated with a 2:1
mixture of pyridine:Ac2O (20 mL) and stirred overnight at room
temperature. The reaction mixture was diluted with CH2Cl2 (50 mL)
and washed successively with water (50 mL), 10% H2SO4 (50 mL),
saturated NaHCO3 (mL), and brine (50 mL). The organic layer was
dried (Na2SO4) and concentrated. The resultant residue was purified via
flash chromatography (1:1 v/v EtOAc/hexanes) to afford the desired
product as a colorless syrup (357 mg, 0.7 mmol, 80%). 1H NMR (600
MHz, CDCl3): δ 8.05-7.98 (m, 2H, Ar-H), 7.63-7.57 (m, 1H, Ar-
H), 7.50-7.44 (m, 2H, Ar-H), 6.44 (d, J = 3.1, 1H, H-1), 5.67-5.62
(m, 1H, H-3), 5.41 (dd, J = 5.2, 3.0, 2H, H-2, H-4/H-5), 4.53-4.44 (m,
2H, H-4/H-5, H-6), 4.30 (dd, J = 10.9, 6.9, 1H, H-6), 2.21 (s, 3H, CH3),
2.19 (s, 3H, CH3), 2.05 (d, J = 3.9, 3H, CH3), 2.03 (d, J = 3.9, 3H, CH3).
13C NMR (151 MHz, CDCl3): δ 169.63, 169.61, 169.43, 168.47, 165.40,
132.90, 129.31, 129.27, 128.80, 128.03, 89.29, 68.34, 67.06, 66.88, 66.04,
60.99, 20.42, 20.18, 20.16, 20.08.
1
TMS. All NMR peak assignments are based on H-1H COSY and
1H-13C HMQC experiments; coupling constants are reported in Hz. E.
coli sialic acid (Neu5Ac) aldolase was purchased from Codexis. Neisseria
meningitidis, CMP-NeuAc synthase, was expressed and purified as
reported,20 and the R-2,6-sialyltransferase from Photobacterium sp. JT-
ISH-224 was kindly donated by M. Gilbert (National Research Council,
Ottawa). A. oryzae β-galactosidase was purchased from Sigma-Aldrich.
M. viridifaciens sialidase was expressed recombinantly in E. coli and
purified as described.6 The 13C-labeled sodium pyruvates (99.9 atom %)
were purchased from Cambridge Isotopes, and 18O-labeled H2O was
purchased from Marshall isotopes (95.1 atom % O-18, batch no.
020414nw). Labeled C3-deuterated Neu5Ac21 and 2-acetamido-2-
deoxy-D-(3-18O)mannose22 were synthesized as reported in the litera-
ture. The synthesis of 8-fluoro-4-methylumbellyferyl β-galactopyrano-
side, which upon β-galactosidase-catalyzed hydrolysis gives the
fluorescent 8-fluoro-7-hydroxy-4-methylcoumarin (8-fluoro-4-methy-
lumbellyfereone, FMU), is given in the Supporting Information. In
addition, the Supporting Information contains the 1H NMR spectra of 1
and all of its labeled isotopomers, the 13C NMR spectra of 1 and the two
13C-labeled substrates, and the HR-MS m/z data for all isotopomers.
Synthesis of (6-18O)Galactoside Precursor. Preparation of
compound 1b requires synthesis of 8-fluoro-4-methylumbellyferyl β-
D-(6-18O)galactopyranoside 18O-6, which can be accessed in six steps
beginning from D-galactose in 8.0% yield overall as detailed below.
1,2:3,4-Di-O-isopropylidene-R-D-galactopyranose (2). A suspen-
sion of D-galactose (4.9 g, 27.2 mmol) in acetone (100 mL) was treated
with ZnCl2 (3.70 g, 27.2 mmol) and concentrated H2SO4 (0.1 mL). The
resultant mixture was stirred at room temperature for 20 h. Potassium
carbonate (1.0 g) was added, and the resultant mixture was stirred for 10
min. The reaction mixture was filtered through a Celite plug and washed
with acetone. The filtrate and washings were concentrated under
reduced pressure to afford the desired product as a colorless syrup
(6.95 g, 26.7 mmol, 98% yield). 1H NMR (600 MHz, CDCl3): δ 5.60 (d,
8-Fluoro-4-methylumbelliferyl 2,3,4-Tri-O-acetyl-6-O-(18O)benzoyl-β-
D-(6-18O)galacto-pyranoside (5). A solution of 33% HBr in AcOH
(60 mL) and Ac2O (11 mL) was cooled to 0 °C in an ice bath and
treated with 4 (3.09 g, 6.8 mmol). After being warmed and stirred at
room temperature for 2 h, the reaction mixture was diluted with CH2Cl2
(150 mL) and poured into ice-water (∼100 mL). After separation of
the organic phase, it was washed with saturated NaHCO3 (150 mL),
dried (Na2SO4), and concentrated. The resultant residue was treated
with 8-fluoro-4-methylumbelliferone (1.31 g, 6.8 mmol), tetrabutylam-
monium bromide (1.95 g, 6.0 mmol), CH2Cl2 (285 mL), and 0.67 M
aqueous NaOH (38 mL, 25.4 mmol). After being stirred for 1.5 h, the
reaction was quenched by the addition of saturated NH4Cl (20 mL), and
the resulting solution was extracted with EtOAc (3 ꢀ 100 mL). The
organic fractions were dried (Na2SO4) and concentrated. The residue
was purified by flash chromatography (1:39 v/v MeOH:CH2Cl2) to
afford the desired product as a white solid (1.10 g, 1.9 mmol, 27.6% over
two steps). mp = 88-89 °C. 1H NMR (600 MHz, CDCl3): δ 8.08-8.00
(m, 2H, Ar-H), 7.66-7.59 (m, 1H, Ar-H), 7.54-7.43 (m, 2H, Ar-
H), 7.16-7.03 (m, 2H, Ar-H), 6.26 (d, J = 1.2, 1H, Ar-H), 5.63-5.55
(m, 2H), 5.23-5.15 (m, 1H), 5.09 (d, J = 7.9, 1H), 4.61 (dd, J = 11.4, 7.5,
1H), 4.38 (dd, J = 11.4, 5.8, 1H), 4.25-4.18 (m, 1H), 2.38 (s, 3H, CH3),
2.25 (s, 3H, CH3), 2.14 (s, 3H, CH3), 2.06 (s, 3H, CH3). 13C NMR (151
2990
dx.doi.org/10.1021/ja109199p |J. Am. Chem. Soc. 2011, 133, 2989–2997

Journal of the American Chemical Society
ARTICLE
Scheme 1. One-Pot Three-Enzyme Chemoenzymatic Synthesis of Neu5AcR2,6GalβFMU
MHz, CDCl3): δ 169.69, 169.58, 168.98, 165.32, 158.66, 151.27, 146.32,
146.26, 142.59, 141.33, 139.65, 132.98, 129.31, 128.83, 128.05, 118.53,
118.50, 116.90, 114.71, 113.76, 100.66, 71.14, 70.08, 67.93, 66.41, 61.22,
20.21, 20.16, 20.10, 18.25.
Neu5AcR2,6GalβFMU exhibited the predicted 1H and 13C NMR
spectra consistent with incorporation of isotopes (2H, 13C, and 18O)
at the expected positions. Unlabeled Neu5AcR2,6GalβFMU, 1H NMR
(600 MHz, D2O): δ 7.53 (d, J = 9.1, 1H, Ar-H), 7.25 (dd, J = 8.9, 7.5,
1H, Ar-H), 6.26 (d, J = 0.9, 1H, Ar-H), 5.10 (s, 1H, 1-H), 3.93-3.85
(m, 3H, 4-H, 50-H), 3.80 (m, 1H, 2-H), 3.74-3.70 (m, 3H, 3-H), 3.65-
3.54 (m, 4H, 40-H), 3.51 (m, 1H), 3.44 (m, 1H), 2.70-2.61 (m, 1H, 30-
Heq), 2.40 (s, 3H, Ar-CH3), 1.91 (s, 3H, CH3), 1.55 (t, J = 12.1, 1H, 30-
Hax). 13C NMR (151 MHz, D2O): 174.96 (CdO, amide), 173.51 (C-
100), 162.46 (Ar), 155.52 (Ar), 146.37 (d, JCF = 8.1 Hz, C-10), 141.48 (d,
8-Fluoro-4-methylumbelliferyl β-D-(6-18O)Galactopyranoside (18O-
6). A solution of 5 (100 mg, 0.169 mmol) in 0.5 M NaOMe in MeOH
(20 mL) was stirred at room temperature for 1.5 h. After 30 min, a white
solid began to precipitate. The precipitate was collected via filtration and
rinsed with Et2O. Upon drying, the desired product was obtained as a
white solid (45 mg, 0.126 mmol, 74.3%). mp = 181-182 °C. 1H NMR
(600 MHz, DMSO): δ 7.55 (dd, J = 9.1, 1.5, 1H, Ar-H), 7.30 (dd, J =
9.0, 7.5, 1H, Ar-H), 6.35 (d, J = 1.2, 1H, Ar-H), 5.34 (bs, 1H, OH),
5.10 (d, J = 7.7, 1H, H-1), 4.94 (bd, J = 4.5, 1H, OH), 4.67 (bs, 1H, OH),
4.59 (bs, 1H, OH), 3.73 (s, 1H), 3.70-3.60 (m, 2H), 3.60-3.51 (m,
1H), 3.51-3.40 (m, 2H), 2.43 (s, J = 1.1, 3H, CH3). 13C NMR (151
MHz, DMSO): δ 158.77, 153.38, 147.48, 147.43, 142.21, 142.15,
139.62, 137.98, 120.35, 115.03, 112.37, 112.11, 100.91, 75.80, 73.26,
70.02, 68.02, 60.22, 18.19. Anal. Calcd for C16H17FO718O1: C, 53.63; H,
4.78. Found: C, 53.37; H, 4.81.
J
CF = 8.8 Hz, C-7), 140.31 (Ar), 139.00 (d, JCF = 248.2 Hz, C-8), 119.19
(d, JCF = 4.1 Hz, C-6), 116.00, 112.41, 111.57, 100.78 (C-10), 99.04 (C-
200), 74.09 (C-40), 72.31 (C-20), 71.68, 70.27, 68.35, 68.13, 67.76, 62.98,
62.55 (C-500), 51.79 (C-400), 40.19 (C-300), 21.93 (CH3, FMU), 18.05
(CH3). HR-MS m/z (M þ Hþ), C27H34FNO16 requires 648.1934,
found 648.1929.
Enzyme Kinetics. Michaelis-Menten kinetic parameters for M.
viridifaciens sialidase and A. oryzae β-galactosidase were determined from
a minimum of seven initial rate measurements within a substrate
concentration range of at least Km/4 to 4Km at 25 °C. The progress
of the reactions was continuously monitored for 5 min using a Cary
Eclipse fluorescence spectrometer equipped with a temperature con-
troller. Each 500 μL reaction mixture was prepared by addition of the
appropriate volume of buffer, substrate, and enzyme. The rate versus
substrate data were fitted to the Michaelis-Menten equation using a
standard nonlinear least-squares program Prism 4.0.
Test for Possible Product Inhibition. Employing conditions
identical to those used during the coupled-enzyme assays, the initial rate
for 4-methylumbelliferyl N-acetyl-R-D-neuraminic acid (MU-NANA)
hydrolysis was determined in the presence of varying concentrations of
FMU, Neu5Ac, and galactose.
KIE Measurements. The stability of the enzyme solution
(containing 0.1 mg/mL BSA), incubated at 25 °C, was evaluated by
testing sialidase activity with the substrate MU-NANA at periodic
intervals. The measured enzyme activity remained invariant within
experimental error for over 48 h. Measurements of KIEs were performed
at 25 °C in 100 mM MES ([BSA] = 0.1 mg/mL), adjusted to pH = 5.85.
Concentrations of least 7 times the Km value of 1 were used. The fraction
of substrate hydrolyzed during the course of the KIE measurements did
not exceed 5%. The analysis order of the isotopically labeled compounds
was alternated (i.e., 1H 2H 2H 1H 1H 2H 2H 1H 1H 2H, etc.). Consecutive
pairs of kinetic runs were compared, and the mean and standard
deviation of at least seven such dyads were taken. A double-blind
procedure was used whereby the identities of the “labeled” and
“unlabeled” samples used in the experiments were unknown to both
Synthesis and Purification of Neu5Acr2,6GalβFMU. A 50 mL
autoclaved falcon tube was charged with N-acetyl-D-mannosamine (50.0
mg, 226 μmol), sodium pyruvate (37.3 mg, 339 μmol), H2O (2.5 mL),
and E. coli Neu5Ac aldolase (2 mg). The reaction vessel was incubated
and shaken at 37 °C for 3 h. Subsequently, cytidine 50-triphosphate
disodium salt (178.7 mg, 339 μmol), 1.0 M tris (2.5 mL), 1.0 M MgCl2
(1.0 mL), 1.0 M dithiothreitol (1.5 μL), and CMP-Neu5Ac synthase
(0.6 mL, 50 U/mL) were added to the reaction mixture. After a further
incubation at 37 °C for approximately 1.5 h, the reaction mixture was
centrifuged (10 min @ 2500 rpm). The supernatant was separated from
the pellet and transferred to a new 50 mL autoclaved falcon tube, after
which GalβFMU (6, 50.1 mg, 103 μmol) and R2,6-sialyl transferase (0.5
mL, 2.1 U/mL) were added to the solution, and the resultant mixture
was incubated at 37 °C overnight. The reaction mixture was loaded onto
a reversed-phased C18 sep-pak cartridge (20 cc, 5 g), and the cartridge
was eluted successively with H2O (∼50 mL) and 1:19 v/v MeCN/H2O
(∼150 mL). The fractions containing product were pooled and
lyophilized to afford the product as a white solid. Typical yields ranged
from 35% to 40%. Samples of Neu5AcR2,6GalβFMU in H2O were then
loaded onto a HP/Agilent 1100 equipped with a C6 reverse phase
HPLC column equilibrated in 95% solvent A (0.1% TFA in H2O) and
5% solvent B (100% MeCN-0.1% TFA). The desired analytically pure
substrate was eluted from the column by increasing the gradient of
solvent B over 30 min from 5% to 30%. The column flow rate was
maintained at 2 mL/min, and 300 μL fractions were collected. The
absorbance of the effluent was monitored at 340 nm. All labeled
2991
dx.doi.org/10.1021/ja109199p |J. Am. Chem. Soc. 2011, 133, 2989–2997

Journal of the American Chemical Society
ARTICLE
Scheme 2. Preparation of (6-18O)Galactoside Precursor 18O-6
Scheme 3. Coupled Enzyme Assay for the Continuous Measurement of Sialidase Activity
experimenters. Specifically, samples marked as “A” and “B” were
prepared by experimenter 1, and the relative concentrations were
adjusted to <1% difference based on UV-vis absorbance by experi-
menter 2, who then relabeled the samples “C” and “D”.
Solvent Kinetic Isotope Effect. Buffers were prepared by add-
ing equivalent amounts of NaOH or NaOD to MES in H2O or D2O,
respectively. Reaction rates were determined at varying deuterium
fractions by mixing the H2O and D2O buffers accordingly. Substrate
concentrations (7 ꢀ Km) in H2O and D2O buffers were adjusted to <1%
difference based on the UV-vis spectra of each stock solution. To
ensure that the enzyme concentrations do not impact the observed rates,
the same enzyme stock solution in H2O was added into each reaction.
After dilution, the total H2O content from the enzyme solution was less
than 2% of the total volume.
mediated synthesis of 1 was used for the current study because, in
contrast to most chemical syntheses, formation of the glycal
5-acetamido-2,6-anhydro-3,5-dideoxy-D-glycero-D-galacto-non-2-
enonic acid (Neu2en5Ac), a potent inhibitor of sialidases, does
not occur during the enzymatic coupling reaction.
Test for Possible Product Inhibition. To ensure that the
products released during the kinetic assays did not affect the
measured rate of reaction by lowering the sialidase activity, the
products of hydrolysis were screened for possible product
inhibition. No inhibition was observed up to at least 25 μM
FMU, 680 μM Neu5Ac, and 650 μM galactose.
Enzyme Kinetics. To measure KIEs on V with natural sub-
strate analogues, it is necessary to ensure that the coupled-
enzyme assay reports on the rate of the sialidase-catalyzed
reaction (Scheme 3).
Thus, it was checked that doubling the concentration of A.
oryzae β-galactosidase did not change the observed rate. Further-
more, at constant β-galactosidase concentrations, halving or
doubling the concentration of M. viridifaciens sialidase results
in a corresponding change of the observed rate. The fluorescence
intensity of the galactoside aglycone (FMU) was found to be
linear up to concentrations of at least 25 μM. This concentration
corresponds to ∼5% hydrolysis. No photobleaching of FMU was
observed under the experimental conditions over a time-course
of 60 min. KIE experiments were performed at substrate
’ RESULTS
Substrate Synthesis. Neu5AcR2,6GalβFMU (1) and the
necessary seven labeled isotopologues (1a-g) were prepared
chemoenzymatically from the β-D-galactopyranoside precursor
6, ManNAc, and pyruvate (Scheme 1). The labeled starting
materials for the seven isotopically substituted substrates were (i)
pyruvate (1a and 1c); (ii) deuterated sialic acid (1e, 1f, and
1g);21 (iii) N-acetyl-3-18O-mannosamine (1b);22 and (iv) 18O-6,
the synthesis of which is shown in Scheme 2 (1d). Enzyme-
2992
dx.doi.org/10.1021/ja109199p |J. Am. Chem. Soc. 2011, 133, 2989–2997

Journal of the American Chemical Society
ARTICLE
Table 1. Kinetic Isotope Effects on kcat for the M. viridifaciens
Sialidase-Catalyzed Hydrolysis of 1 in 100 mM MES Buffer
pH = 5.85, [BSA] = 0.1 mg/mL, and Temperature = 25 °C
compound site of substitution
KIEa
weighted averageb
1a
1b
anomeric 2-13
C
1.022 ( 0.008
1.020 ( 0.008
0.986 ( 0.004
0.984 ( 0.006
0.989 ( 0.010
1.002 ( 0.010
1.001 ( 0.012
1.021 ( 0.006
ring oxygen 6-18O
0.986 ( 0.003
1c
C3-13C
1.001 ( 0.008
1.003 ( 0.005
1d
leaving group 2-18O 1.003 ( 0.006
1.002 ( 0.010
1.003 ( 0.011
1e
1f
axial (3R)-2H
1.035 ( 0.012
1.027 ( 0.011
1.025 ( 0.018
0.890 ( 0.012
0.893 ( 0.014
0.890 ( 0.014
0.885 ( 0.014
0.897 ( 0.009
0.885 ( 0.011
0.881 ( 0.009c
1.002 ( 0.008d
1.029 ( 0.007
0.891 ( 0.008
0.890 ( 0.006
Figure 2. Variation of observed rate constant for M. viridifaciens
sialidase-catalyzed hydrolysis at saturating concentrations of
Neu5AcR2,6GalβFMU in mixtures of H2O:D2O; n is the fraction of
deuterium in the mixture.
equatorial (3S)-2H
dideutero 3-2H2
irreversible event is glycosylation to give the tyrosinyl-enzyme
bound intermediate (k5, Scheme 4).
1g
kcat
Km
k1k3k5
¼
ð2Þ
ð3Þ
k2ðk4 þ k5Þ þ k3k5
control
n/a
n/a
kgkdc
a Average and standard error from at least seven dyad measurements.
b Calculated according to ref 47. c Concentration of 1 was doubled (∼14
ꢀ Km), n = 3. d Control experiment represents a ratio of observed rates
between two separately prepared batches of unlabeled 1, n = 5.
kcat
¼
kg þ kdc
k3k5
kg ¼
ð4Þ
ð5Þ
k3 þ k4 þ k5
concentrations of at least 7 times the Km value of 1. Listed in
Table 1 are averaged KIE values from the last seven consecutive
dyad measurements that are corrected to account for incomplete
isotopic substitution according to either a literature procedure23
or, for the case of the β-secondary deuterium effects where
unlabeled, monodeuterated, and dideuterated isotopomers are
present, eq 1 where fHH, fHD, fDH, and fDD are the fractions of
each isotopomer present in the sample. For the dideuterio
isotopomer (1g), the deuteration at C3 was assumed to be
complete (i.e., fDD = 1 and fHH = fHD = fDH = 0).
k7k9k11
kdc
¼
k7ðk9 þ k10 þ k11Þ þ k9k11
The observed rate constants that are obtained from the Michae-
lis-Menten parameters for the mechanism shown in Scheme 2
are given by eqs 2 and 3, where kg (glycosylation) and kdc
(deglycosylation, conformational change and product release)
are given by eqs 4 and 5, respectively.
The sialidase from M. viridifaciens has been shown to exhibit
βlg values for hydrolysis on V/K and V of -0.30 ( 0.04 and 0.02
( 0.03, respectively.6 These observations require that a step
following glycosylation is rate-determining for V.6 Product
release can be ruled out as the rate-determining step on V for
the M. viridifaciens sialidase because this would require that
product inhibition be observed during catalysis. However, little
or no inhibition of catalysis is observed at sialic acid concentra-
tions of up to 680 μM. Thus, it can be concluded that k11 is much
greater than either k9 or k10, and kdc can be rewritten as k7k9/(k7
þ k9), and a series of KIE measurements will enable a distinction
between the two remaining possible rate-determining steps for
V: deglycosylation of the sialosyl-enzyme intermediate (k7) and a
conformational change of the initial enzyme product complex
(k9).
kHH
KIEobs
¼
ðfHH ꢀ kHHÞ þ ðfHD ꢀ kHDÞ þ ðfDH ꢀ kDHÞ þ ðfDD ꢀ kDD
Þ
ð1Þ
Solvent Isotope Effects. The solvent deuterium KIE (D OV)
for the sialidase-catalyzed hydrolysis of 1 was measured to be
1.585 ( 0.004. A proton inventory for this reaction displayed a
linearly correlation between the rate of hydrolysis and n, the
fraction of deuterium in the solvent (Figure 2).
2
’ DISCUSSION
Mechanism of Sialidase-Catalyzed Reactions. The prevail-
ing mechanism for exo-sialidases is shown in Scheme 4,16 where
RNeu5Ac-OR is the substrate, RNeu5Ac-OH is the sialic acid
product, and the pyranosyl conformations chair and boat are
indicated by the subscripts C and B, respectively. For sialidase
reactions that are followed under initial rate conditions, the first
Measurement of KIEs on V can only be made by a direct
comparison of individually measured rates of unlabeled and
labeled substrates at saturating levels.24,25 As a consequence,
such measurements are prone to greater systematic errors than is
the case for competitive KIE measurements made on V/K
2993
dx.doi.org/10.1021/ja109199p |J. Am. Chem. Soc. 2011, 133, 2989–2997

Journal of the American Chemical Society
ARTICLE
Scheme 4. Proposed Catalytic Cycle for Sialidase-Catalyzed Hydrolysis Reactions
(including varying substrate and enzyme concentrations, chan-
ging temperature, and the purity of both isotopic substrates).
Despite these innate problems that render the measurement of
KIEs on V more difficult, precise and reliable methods have
emerged. For example, computer-controlled continuous-flow
methods have been reported for the measurement of carbon-
13 KIEs (13V)26,27 for formate dehydrogenase28 and dialkylgly-
cine decarboxylase.29
Ideally, KIEs should be measured on natural substrates;
however, such accurate kinetic parameter measurements for
sialidases require the use of auxiliary enzymes.19,30,31 In a recent
report, Indurugalla et al. showed that accurate sialidase kinetic
parameters could be obtained from R-2,3- and 2,6-sialyl-galacto-
side linkages, where a chromophore is released in the presence of
an exo-β-galactosidase.19 The conditions that are necessary for
Figure 3. Postulated conformation of the natural substrate analogue in
the accumulating Michaelis complex where pro-S σ(C-H) f σ*(C-
O) hyperconjugation is maximized.
accurate coupled assays are: (i) the reaction of interest is
irreversible and zero-order with respect to the substrate; and
(ii) the coupled reaction is irreversible and first-order with
respect to the intermediate generated by the primary enzyme.32
To satisfy these requirements, high substrate concentrations
(∼7 ꢀ Km) are used and maintained throughout the assay by
acquiring data only during the initial 5% of substrate hydrolysis,
thus ensuring that the reaction rate is unchanged during the
experiment. Also, the use of high concentrations of coupling
enzyme (β-galactosidase) ensures that the concentration of the
intermediate formed is maintained well below its Km value for
this auxiliary enzyme, and thus ensuring that the reaction is first-
order with respect to the intermediate β-galactoside product
formed during the sialidase-catalyzed reaction. In the current
study, hydrolysis of Neu5AcR2,6GalβFMU (1) releases GalβF-
MU, which in the presence of excess β-galactosidase is rapidly
hydrolyzed to liberate the fluorescent aglycone, 8-fluoro-4-
methylumbelliferone (FMU), for detection (Scheme 3). Of note,
FMU was chosen for this study, rather than the previously used
4-methylumbelliferone leaving group (4-MU),19 because of its
lower pKa value (6.4 vs 7.8 for 4-MU),33 and thus at the assay pH,
a greater fraction of the liberated phenol is present as its highly
fluorescent anion.
Oxygen-18 KIEs. In principle, the measured endocyclic and
exocyclic 18O-KIEs for pyranoside and furanoside hydrolysis
reactions are complementary.34,35 That is, the expectation is that
one 18O-KIE will be normal (18k > 1, C-O bond being broken)
and the other will be inverse (18k < 1, oxygen atom stabilizing
charge development in the nascent oxacarbenium ion).34,36 The
leaving group 18O-KIE for the hydrolysis of 1d (18V = 1.003 (
0.005) supports the previous Brønsted analysis6 that cleavage of
the aglycone is not rate-determining. In contrast, the complemen-
tary endocyclic oxygen atom is associated with a notably inverse
KIE (1b; 18V = 0.986 ( 0.003), a result that is consistent with
significant charge delocalization occurring onto the pyranosyl
oxygen at the deglycosylation transition state (k7, Scheme 4).
β-Secondary Deuterium KIEs. Generally, solvolysis reac-
tions of carbohydrate derivatives exhibit secondary deuterium
KIEs that are normal (Dk > 1), and these effects result from a
hyperconjugative weakening of the C-H/D bond by overlap
with the nascent F-orbital at the anomeric center.34,37,38 As
hyperconjugation is an angular-dependent phenomenon, the
magnitude of β-SDKIEs has been used to evaluate transition
state (TS) geometry for the glycosylation step (k5, Scheme 4) for
various sialidase-catalyzed reactions.17,18
The magnitudes of the two diastereomeric β-secondary deu-
terium KIEs are markedly different (Table 1) with the 3S-
isotopomer displaying a large inverse KIE (1f; DVS = 0.891),
while the other diastereomer gives rise to a small normal effect
(1e; DVR = 1.029). Of note, the C3-dideutero β-SDKIE (DVR,S
=
0.890 ( 0.006) is, within experimental error, the product of the
individual axial and equatorial β-SDKIEs. Also, when the sub-
strate concentrations for both unlabeled and dideutero isotopo-
mers were doubled (to approximately 14 ꢀ Km), the measured
3-2H2 KIE was unaltered (DVR,S = 0.881 ( 0.009; n = 3).
The interpretation of the large inverse secondary deuterium
KIE for the hydrolysis of 1f on V is simplified because the
magnitude of this KIE (DVS = 0.891) is outside that normally
associated with either steric or inductive effects.25,39 Thus, the
KIE for the 3S-isotopomer requires that the rate-limiting step
involves a marked strengthening of this C-H/D bond at the TS
relative to the ground state. A rationale that requires the
accumulating Michaelis complex (reactant state), under the
2994
dx.doi.org/10.1021/ja109199p |J. Am. Chem. Soc. 2011, 133, 2989–2997

Journal of the American Chemical Society
ARTICLE
residue to an acceptor water molecule. During this migration,
the dihedral angle between the pro-S σ(C-H) bond and the
nascent F orbital at the anomeric center increases from 60° and,
based on the inverse secondary deuterium KIE, is likely between
75-105° at the deglycosylation TS (cos2 θ < 0.07).
The other measured secondary deuterium KIE (1e; DVR =
1.029, Table 1) is compatible with a small hyperconjugative
interaction at the deglycosylation TS, a result that is consistent
with the reported Dk values in solution where the pro-R value
(DkR = 1.073) is significantly larger than the corresponding pro-S
value (DkS = 1.000) for the spontaneous hydrolysis of 4-nitro-
phenyl R-D-sialoside.21
Carbon-13 KIEs. The anomeric 13C KIE for the M. viridifa-
ciens sialidase-catalyzed hydrolysis of 1a, 13V = 1.021 ( 0.006, is
only consistent, when taken with the measured 18V KIEs
(Table 1) and the reported βlg value of 0.02 ( 0.03,6 with the
rate-determining step being deglycosylation (k7, Scheme 4,
Figure 4). The measured secondary 13C KIE on carbon 3 (1c;
13V = 1.001 ( 0.008, Table 1) is equal to one, within experi-
mental error, and this measurement functions as an internal
control because it was viewed that 13C-labeling at this position
would likely not give rise to a significant KIE.
Figure 4. Possible transition state structure for deglycosylation of the β-
sialosyl-enzyme intermediate formed during the M. viridifaciens sialidase-
catalyzed reactions.
conditions of substrate saturation, adopts a conformation in
which only one of the two C-H/D bonds is manifestly
weakened by a hyperconjugative interaction and that this effect
either disappears or considerable weakens at the transition state
(Figure 3). That is, hyperconjugative weakening of the equatorial
σ(C-H/D) bond into the σ*(C-O) orbital of the aglycone
moiety is the most likely cause for the reduction in the ground-
state force constant of this C-H/D bond (Figure 3). Further-
more, there must be little hyperconjugation from the pro-S C-H
bond into the nascent F-orbital at the deglycosylation TS.
A current popular mechanistic rationale for glycosidase-cata-
lyzed reactions, based mainly on X-ray diffraction studies of
modified glycosidase:substrate complexes, fluorinated glycosyl-
enzyme intermediates, and presumed transition state analogues,40
is based on the idea that the pyranosyl ring conformation at the TS
is close to that of a glycopyranosylium ion (C5, O5, C1, and C2 are
coplanar),41 which in the case of sialidases is proposed to be a 5H4
half-chair (numbering adjusted from ref 40 to match that for sialic
acid).40 Moreover, the postulated reaction coordinate for glyco-
sylation in sialidase-catalysis involves the substrate adopting a 4S2
skew boat in the Michaelis complex and that this converts into a
2C5 chair conformation during formation of the tyrosinyl-bound
intermediate.11
Typical SN2 (ANDN) reactions of glycosides in solution
exhibit 13k values in the range of 1.03-1.08,37 while carbohydrate
hydrolysis reactions that proceed via nonequilibrated short-lived
glycopyranosylium ions (DN*AN) exhibit 13k values around
1.01.34,38 Thus, the magnitude of the anomeric 13C KIE (13V =
1.021 ( 0.006) does not allow an unambiguous distinction to be
made between the separate mechanistic possibilities. Even so, it is
clear that M. viridifaciens sialidase-catalyzed deglycosylation in-
volves an “exploded” late transition state39-41 with significant
cleavage of the tyrosinyl-sialoside C-O bond and delocalization
of positive charge from the anomeric center occurring via a
np(O)fF(C) interaction (1b, 18V = 0.986). The main uncer-
tainty is whether the water nucleophile has started its attack on
the anomeric center; indeed the measured anomeric 13C KIE in
this study, which is equivalent to a 14C value of around 1.04,25 is
bracketed by two reported 14C KIE measurements of 1.03442 and
1.041 in which the calculated transition state bond lengths are (i)
anomeric carbon-leaving group (C-N in these cases) of 2.65
and 2.80 Å; and (ii) anomeric carbon-oxygen nucleophile of
2.46 and 2.30 Å, respectively. In addition, other calculations have
suggested that, for 2-deoxyribosyl reactants, the lower limit for an
anomeric 13C-KIE in an ANDN mechanism is 1.013-1.040,
which encompasses the measured value of 1.021.43,44 Therefore,
it can be concluded that at the transition state for deglycosylation
the covalent bond between the anomeric carbon and the enzy-
matic tyrosine residue is largely broken, while bond formation to
the water nucleophile has either not yet begun or just started.
Solvent Deuterium KIE and Proton Inventory. The mea-
sured solvent deuterium KIE for the M. viridifaciens sialidase-
Of note, the limiting inverse KIE value should occur in cases
where the hyperconjugative weakening, which varies as a function
of cos2 θ, is maximal (180° dihedral angle) in the ground state and
absent in the TS (90° dihedral angle to the nascent F-orbital at the
anomeric center). In the context of the inverse KIE (DVS = 0.891)
reported herein, the magnitude of this value is consistent with an
accumulating Michaelis complex in which the substrate is posi-
6
tioned in a S2 skew-boat conformation (Figure 3) in which
maximal hyperconjugative weakening is possible as the dihedral
angle between the pro-S σ(C-H) bond and the σ*(C-O) orbital
is 180° (cos2 θ = 1). A conformational change of this 6S2 skew-
catalyzed hydrolysis of 1 (D OV = 1.585 ( 0.004) and the
2
associated linear proton inventory (Figure 4) are consistent with
a single proton “in flight” at the transition state. The two plausible
proton transfers are shown in Figure 4: (a) general acid-catalyzed
protonation of the tyrosine leaving group by Glu260; and (b)
general base-assisted nucleophilic attack of water by Asp92.
It can be concluded that the deglycosylation reaction involves
general-acid catalysis of tyrosine departure (Figure 4a) as the
anomeric carbon KIE is consistent with an exploded, SN1-like,
transition state. That is, without catalysis by Glu-260, the
enzymatic tyrosine residue would be required to depart as an
4
boat Michaelis complex into a S2 conformation, via the B2,5,
reduces the dihedral angle to around 150° and the interaction by
25% (cos2 θ = 0.75). Formation of the tyrosinyl-bound inter-
mediate moves the pro-S σ(C-H) bond to a dihedral angle of
about 60° to the σ*(C-O) orbital of the anomeric carbon to
tyrosine oxygen bond (2C5 chair conformation; cos2 θ = 0.25).
Breakdown of the enzyme-bound intermediate (k7, Scheme 4,
Figure 4) involves the anomeric carbon undergoing an oxygen to
oxygen electrophilic migration from the enzymatic tyrosine
2995
dx.doi.org/10.1021/ja109199p |J. Am. Chem. Soc. 2011, 133, 2989–2997

Journal of the American Chemical Society
ARTICLE
anion with the TS resembling an intimate ion pair, a high energy
situation that is ameliorated by general acid catalysis during the
rate-limiting cleavage of the sialosyl-tyrosine C-O bond. In
contrast, as little or no bond formation is occurring to the
nucleophilic water molecule at the TS, the requirement for
catalysis by Asp-92 (Figure 4b) is anticipated to be negligible.
Mechanistic Summary. The sialidase from Micromonospora
viridifaciens operates via a double inversion mechanism in which
the accumulating Michaelis complex positions the substrate in a
6S2 skew boat conformation, although, at the present time, it is
not certain whether the initial binding event occurs simulta-
neously with, or prior to, the requisite enzyme-mediated con-
formational change. However, given that the M. viridifaciens
sialidase is a remarkable proficient enzyme (kcat/Km > 107 M-
1 s-1 with activated substrates),14 this conformational change is
rapid. Next, catalysis occurs via two sequential acetal C-O bond
cleavage reactions that occur via an enzyme-bound tyrosinyl β-
sialoside intermediate. It is likely that the glycosylation TS is
similar to the exploded, SN1-like, TS noted above for the
deglycosylation reaction step. Finally, the product sialic acid
dissociates from the enzyme active site.
The possibility that the glycosylation reaction (k5, Scheme 4)
involves formation of an oxacarbenium ion intermediate that
reacts directly with water rather than tyrosine-370 deserves
comment. For retaining glycosidases, any enzyme-bound inter-
mediate must have a lifetime long enough to allow the aglycone
(leaving group) to diffuse from the active site, for a water
nucleophile to replace it, otherwise starting material will be
regenerated. If the sialidase-bound intermediate is an oxacarbe-
nium ion, rather than an enzymatic tyrosine adduct, then the
complex rate constant for dissociation of the carbohydrate
leaving group, diffusional association of water, and water attack
must be faster than nucleophilic capture of the cation by the
proximal tyrosine residue. This possibility seems remote given
that Ghanem et al. have estimated the lifetime of an enzyme-
bound ribosyl oxacarbenium ion to be less than 50 ps,45 a lifetime
that is too short to allow diffusional separation of the leaving
group prior to reaction of the cation.46
1a-1g, and 13C NMR spectra for compounds 1, 1a, and 1c. This
material is available free of charge via the Internet at http://pubs.
acs.org.
’ AUTHOR INFORMATION
Corresponding Author
bennet@sfu.ca
’ ACKNOWLEDGMENT
We thank the Natural Sciences and Engineering Research
Council of Canada for financial support and Dr. M. Gilbert
(NRC) for providing the 2,6-sialyltransferase. We also thank
three anonymous referees for their many helpful comments
concerning this manuscript.
’ REFERENCES
(1) Miyagi, T.; Kato, K.; Ueno, S.; Wada, T. Trends Glycosci.
Glycotechnol. 2004, 16, 371–381.
(2) Schauer, R. Trends Biochem. Sci. 1985, 10, 357–360.
(3) Corfield, T. Glycobiology 1992, 2, 509–521.
(4) Taylor, G. Curr. Opin. Struct. Biol. 1996, 6, 830–837.
(5) Davies, G.; Sinnott, M. L.; Withers, S. G. In Comprehensive
Biological Catalysis, 1st ed.; Sinnott, M. L., Ed.; Academic Press: San
Diego, CA, 1998; pp 119-209.
(6) Watson, J. N.; Dookhun, V.; Borgford, T. J.; Bennet, A. J.
Biochemistry 2003, 42, 12682–12690.
(7) Watson, J. N.; Newstead, S.; Dookhun, V.; Taylor, G.; Bennet,
A. J. FEBS Lett. 2004, 577, 265–269.
(8) Watson, J. N.; Newstead, S.; Narine, A.; Taylor, G.; Bennet, A. J.
ChemBioChem 2005, 6, 1999–2004.
(9) Newstead, S.; Watson, J. N.; Knoll, T. L.; Bennet, A. J.; Taylor, G.
Biochemistry 2005, 44, 9117–9122.
(10) Watts, A. G.; Damager, I.; Amaya, M. L.; Buschiazzo, A.; Alzari,
P.; Frasch, A. C.; Withers, S. G. J. Am. Chem. Soc. 2003, 125, 7532–7533.
(11) Watts, A. G.; Oppezzo, P.; Withers, S. G.; Alzari, P. M.;
Buschiazzo, A. J. Biol. Chem. 2006, 281, 4149–4155.
(12) Chan, J.; Lewis, A. R.; Gilbert, M.; Karwaski, M. F.; Bennet, A. J.
Nat. Chem. Biol. 2010, 6, 405–407.
(13) Chong, A. K. J.; Pegg, M. S.; Taylor, N. R.; von Itzstein, M. Eur.
J. Biochem. 1992, 207, 335–343.
(14) Narine, A. A.; Watson, J. N.; Bennet, A. J. Biochemistry 2006, 45,
9319–9326.
(15) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in
Organic Chemistry, 3rd ed.; Harper & Row: New York, 1987.
(16) Guo, X.; Laver, W. G.; Vimr, E.; Sinnott, M. L. J. Am. Chem. Soc.
1994, 116, 5572–5578.
(17) Guo, X.; Sinnott, M. L. Biochem. J. 1993, 294, 653–656.
(18) Sinnott, M. L.; Guo, X. M.; Li, S. C.; Li, Y. T. J. Am. Chem. Soc.
1993, 115, 3334–3335.
(19) Indurugalla, D.; Watson, J. N.; Bennet, A. J. Org. Biomol. Chem.
2006, 4, 4453–4459.
(20) Karwaski, M. F.; Wakarchuk, W. W.; Gilbert, M. Protein
Expression Purif. 2002, 25, 237–240.
(21) Ashwell, M.; Guo, X.; Sinnott, M. L. J. Am. Chem. Soc. 1992,
114, 10158–10166.
(22) Indurugalla, D.; Bennet, A. J. Can. J. Chem. 2008, 86,
’ CONCLUSIONS
A panel of seven isotopically substituted Neu5AcR2,6GalβF-
MU natural substrate analogues has been used to determine that
the rate-limiting step for turnover by the M. viridifaciens sialidase
is deglycosylation of the β-sialosyl-enzyme intermediate.
Furthermore, on the basis of the observed KIEs, it is proposed
that the accumulating Michaelis complex places the sialosyl ring
6
in a S2 skew boat conformation. The measured KIEs are also
consistent with the rate-determining deglycosylation reaction
occurring via an exploded transition state in which synchronous
charge delocalization is occurring onto the ring oxygen atom.
Finally, the proton inventory and the magnitude of the solvent
KIE are consistent with general acid-catalyzed protonation of the
departing tyrosine residue rather than general base-assisted
attack of the nucleophilic water.
1005–1009.
(23) Bergson, G.; Matsson, O.; Sjoberg, S. Chem. Scr. 1977, 11,
25–31.
’ ASSOCIATED CONTENT
(24) Cook, P. F.; Cleland, W. W. Enzyme Kinetics and Mechanism;
Garland Science: London; New York, 2007.
(25) Melander, L. C. S.; Saunders, W. H. J. Reaction Rates of Isotopic
Molecules; Wiley: New York, 1980.
S
Supporting Information. Procedures for the syntheses
b
of 8-fluoro-4-methylumbelliferone (FMU) and of 8-fluoro-4-
methylumbelliferyl β-D-galactopyranoside (GalβFMU, 6), H
1
NMR spectra and HR-MS molecular ion data for compounds 1,
(26) Cook, P. F.; Cleland, W. W. Biochemistry 1981, 20, 1790–1796.
2996
dx.doi.org/10.1021/ja109199p |J. Am. Chem. Soc. 2011, 133, 2989–2997

Journal of the American Chemical Society
ARTICLE
(27) Northrop, D. B. Biochemistry 1975, 14, 2644–2651.
(28) Xue, H.; Wu, X. W.; Huskey, W. P. J. Am. Chem. Soc. 1996, 118,
5804–5805.
(29) Zhou, X.; Toney, M. D. J. Am. Chem. Soc. 1998, 120, 13282–
13283.
(30) Alon, R.; Bayer, E. A.; Wilchek, M. J. Biochem. Biophys. Methods
1991, 22, 23–33.
(31) Ziegler, D. W.; Hutchinson, H. D. Appl. Microbiol. 1972, 23,
1060–1066.
(32) Segel, I. H. Enzyme Kinetics: Behavior and Analysis of Rapid
Equilibrium and Steady State Enzyme Systems; Wiley: New York, 1975.
(33) Sun, W. C.; Gee, K. R.; Haugland, R. P. Bioorg. Med. Chem. Lett.
1998, 8, 3107–3110.
(34) Bennet, A. J.; Sinnott, M. L. J. Am. Chem. Soc. 1986, 108, 7287–
7294.
(35) Bennet, A. J.; Sinnott, M. L.; Wijesundera, W. S. S. J. Chem. Soc.,
Perkin Trans. 2 1985, 1233–1236.
(36) Bennet, A. J.; Davis, A. J.; Hosie, L.; Sinnott, M. L. J. Chem. Soc.,
Perkin Trans. 2 1987, 581–584.
(37) Kresge, A. J.; Weeks, D. P. J. Am. Chem. Soc. 1984, 106, 7140–
7143.
(38) Tanaka, Y.; Tao, W.; Blanchard, J. S.; Hehre, E. J. J. Biol. Chem.
1994, 269, 32306–32312.
(39) Buncel, E.; Lee, C. C. Secondary and Solvent Isotope Effects;
Elsevier: Amsterdam; New York, 1987.
(40) Vocadlo, D. J.; Davies, G. J. Curr. Opin. Chem. Biol. 2008, 12,
539–555.
(41) Sinnott, M. Carbohydrate Chemistry and Biochemistry: Structure
and Mechanism; RSC Publishing: Cambridge, 2007.
(42) Berti, P. J.; Blanke, S. R.; Schramm, V. L. J. Am. Chem. Soc. 1997,
119, 12079–12088.
(43) Berti, P. J.; Tanaka, K. S. E. Adv. Phys. Org. Chem. 2002, 37, 239–
314.
(44) Chen, X.-Y.; Berti, P. J.; Schramm, V. L. J. Am. Chem. Soc. 2000,
122, 6527–6534.
(45) Ghanem, M.; Murkin, A. S.; Schramm, V. L. Chem. Biol. 2009,
16, 971–979.
(46) Richard, J. P.; Jencks, W. P. J. Am. Chem. Soc. 1984, 106, 1373–
1383.
(47) Taylor, J. R. An Introduction to Error Analysis: The Study of
Uncertainties in Physical Measurements; University Science Books: Mill
Valley, CA, 1982.
2997
dx.doi.org/10.1021/ja109199p |J. Am. Chem. Soc. 2011, 133, 2989–2997