pentasaccharide 11 (45%) as a mixture (∼1:1) of R- and
â-anomers at the newly generated glycosidic bond. Finally,
one-step removal of all the ester and phthalimido protecting
groups by treatment of 11 with methylamine (33% solution
in EtOH) furnished the targeted pentasaccharide 1 (R/â ∼
1:1) with 2.2% overall yield for the total of 31 chemical
steps from GlcN‚HCl and KDO‚NH4 as starting materials.
region between -40 and 40 ppm,12 suggesting that no side
reaction other than the hydrolysis of GTP occurred. Only a
small amount of GMP formation (2.9 ppm) was detected
after prolonged incubation of the reaction products (spectrum
after 18 h in Figure 1), indicating that 1 has a substantial
selectivity to GTP as a substrate in comparison to that to
GDP as a substrate.
The pseudo-first-order rate constant kobs for the hydrolysis
of GTP in the presence of 1 was determined from plots of
ln[GTP] vs time (Figure 2) and found to be 3.98 × 10-3
1
The H, 13C, and 2D-COSY NMR and mass spectroscopic
properties of the purified 1 were consistent with the expected
structure.10
Initially, the pure 1 (mixture of R- and â-anomers) was
tested for DNAse, phosphodiesterase, and GTPase activi-
ties.11 Among these, the observed relatively high GTPase
activity was particularly intriguing and was therefore more
extensively investigated. The rates of GTP hydrolysis were
measured at pH 7.1 and 50 °C by direct observation of the
reaction course by 31P NMR spectroscopy. Figure 1 illustrates
Figure 2. Rate profiles of GTP hydrolysis in the presence of
different oligosaccharides as indicated. The ln of the remaining GTP
normalized as a percentage as a function of time is plotted for each
oligosaccharide. The data were obtained from 31P NMR experiments
conducted in the presence of GTP (15.0 mM), Mg2+ (15.0 mM),
and the particular oligosaccharide (4.2 mM) as indicated at an
apparent pH of 7.1 in D2O at 50 °C. The curve referred to as GTP
was obtained in the presence of GTP (15.0 mM) and Mg2+ (15.0
mM) under the same conditions.
min-1. This corresponds to a rate acceleration of about 500-
fold compared with that of the same reaction in the absence
of 1 (kobs ) 8.3 × 10-6 min-1). It turns out that Mg2+ ions
(9) The R- or â-stereochemistry at the glycosidic linkages of KDO was
determined either by the chemical shift (δ) of H-4 proton (δ > 5 for
R-anomers and δ < 5 for â-anomers: Kosma, P.; Strobl, M.; Allmaier, G.;
Schmid, E.; Brade, H. Carbohydr. Res. 1994, 254, 105) or by the chemical
shift difference (∆δ) between H-3ax and H-3eq (small ∆δ for R-anomers
and larger ∆δ for â-anomers: Imoto, M.; Kusunose, N.; Matsuura, Y.;
Kusumoto, S.; Shiba, T. Tetrahedron Lett. 1987, 28, 6277). Complementary
results were obtained by both methods in all molecules containing KDO
(see Supporting Information).
(10) Complete NMR assignments for the monosaccharides 2-5 along
with the protected oligosaccharides 7-8, 10, 11R, 11â and selected data
for 1R and 1â are given in Supporting Information.
Figure 1. Time course of 31P NMR spectra (proton decoupled)
showing the conversion of GTP to GDP and OP (1.5 ppm) in the
presence of 1 (4.2 mM), GTP (15.0 mM), and Mg2+ (15.0 mM) at
an apparent pH of 7.1 in D2O at 50 °C. Spectra were acquired on
a Bruker WH-200 operating at 81.03 MHz (referenced to 200 mM
D3PO4 at 0.0 ppm). TR (-9.8 ppm), Tâ (-17.9 ppm), and Tγ (-4.5
ppm) refer to the R-, â-, and γ-phosphate signals of GTP,
respectively; DR (-9.3 ppm) and Dâ (-5.6 ppm) refer to R- and
â-phosphate signals of GDP, respectively; IS refers to trimethyl
phosphate (4.7 ppm), used as an internal standard.
(11) DNAse activity was examined by using DNA plasmid pSL301
(Brosius, J. DNA 1989, 8, 759) as a substrate. The extent of DNA hydrolysis
was monitored [in Tris/HCl (50 mM), DNA (16.7 µg/mL), 1 (6.2 mM),
pH 7.5, 2 h, 55 °C] by measuring the relative quantities of the supercoiled
(ccc), open-circular (oc), and linear (l) DNA forms by using agarose gel
electrophoresis. The bis(p-nitrophenyl)phosphate was used as a substrate
to determine the phosphodiesterase activity; the reaction progress was
followed [in Tris/HCl (50 mM), bis (p-nitrophenyl) phosphate (0-20 mM),
1 (1.9 mM), pH 7.2, 50 °C] by UV spectroscopy (400 nm). In both cases,
only moderate activities were determined (see Figures 1S and 2S in
Supporting Information).
(12) We note that the integral of the inorganic phosphate signal in Figure
1 (the spectra were acquired at a relaxation delay time of 0.5 s, 320 scans)
appears to be less than that for either of the phosphate groups of GDP.
This is attributed to the longer spin lattice relaxation time (T1) of inorganic
phosphate than that for either phosphate group of GDP. Indeed, when the
same spectra were recorded at the relaxation delay time of 10 s, 320 scans,
integration of 1:1 was observed.
the typical time course of 31P NMR spectra for the conversion
of GTP to GDP and OP in the presence of 1 (4.2 mM), GTP
(15 mM), and Mg2+ ions (15 mM). No other phosphate
peaks, except those shown in Figure 1, were detected in the
Org. Lett., Vol. 3, No. 26, 2001
4313
Solomon
