5264 J. Am. Chem. Soc., Vol. 121, No. 22, 1999
O’Brien et al.
indicate the remarkable enhancement of binding that is achieved
due to concentrating and cooperativity effects.
Glu-Lys(N-ꢀ-acetyl)-D-Ala-D-lactate (analogue 1)] such that when 400
µL of the ligand solution was added to the NMR tube, the antibiotic
would be >90% bound by ligand based on an estimated binding
constant for the association. The pH of the ligand solution was
readjusted to 4.5 with NaOH and HCl. Aliquots of the ligand solution
were added to the NMR tube (rising from 5 µL initially, to 100 µL at
the end of the titration) and the change in chemical shift of the antibiotic
Experimental Section
Materials. Chloroeremomycin was provided by Eli Lilly and Co;
XV1-E L-R-phosphatidylcholine and acetic anhydride were purchased
from Sigma; 1,3-diisopropylcarbodiimide (DIC), diisopropylethyamine
20
proton ω
was determined by performing a least-squares curve fit on a plot of
change in chemical shift of ω vs ligand concentration using the
commercial software package Kaleidagraph version 3.1.2 (Abelbeck
2
was recorded after each addition. The association constant
(DIEA), 4-(dimethylamino)pyridine (DMAP), and triisoproplsilane
(TIPS) were purchased from Aldrich; 2-chlorotrityl chloride resin, N-R-
2
Fmoc-D-alanine, N-R-Fmoc-alanine, N-R-Fmoc-N-ꢀ-acetyl-L-lysine,
N-R-Fmoc-D-glutamic acid-γ-tert-butyl ester, N-R-Fmoc-glycine, 1-hy-
droxybenzotriazole (HOBt), and benzotriazole-1-yl-oxy-tris-pyrrolidino-
phosphonium hexafluorophosphate (PyBOP) were purchased from
Novabiochem; lithium-D-lactate, dimethylformamide, dimethyl sulfox-
ide, and tetrakis(triphenylphosphine) palladium (0) were purchased from
Fluka; N-docosanoyl-Gly-Ala-D-γ-Glu-Lys(N-ꢀ-acetyl)-D-Ala-D-lactate
was synthesized according to the previously published procedure.15
Preparation of Phosphatidylcholine Vesicles. Typical Preparation
of Vesicle Suspension, Phosphatidylcholine Concentration 10 mM.
Type XV1-E L-R-phosphatidylcholine from fresh egg yolk (Sigma, 64
mg) was dissolved in chloroform (10 mL) which had been rendered
ethanol-free by passage through a column of activated alumina. The
solution was then evaporated under reduced pressure to yield a thin
film on the wall of the flask which was dried under high vacuum for
-
1
Software). The value obtained (2000 ( 200 M ) lies within a factor
of 2 of the values obtained for other -D-Lac terminating ligands, both
at ph 4.5 and pH 7.9
,15
Synthesis of Precursor for Solid-Phase Synthesis. N-r-Fmoc-D-
glutamic acid-γ-tert-butyl ester-r-allyl ester. N-R-Fmoc-D-glutamic
acid-γ-tert-butyl ester (0.5 g, 1.16 mmol), HOBt (0.2 g, 1.47 mmol),
and PyBOP (0.77 g, 1.47 mmol) were dissolved in allyl alcohol (10
mL). After stirring for 15 min, DIEA (0.2 g, 1.47 mmol) was added
and the solution stirred for 15 h. Excess allyl alcohol was removed
under reduced pressure and the crude mixture chromatographed over
silica (chloroform) to afford pure N-R-Fmoc-D-glutamic acid-γ-tert-
butyl ester-R-allyl ester (520 mg, 96%) as a white solid. The product
gave the expected spectra by FT-ICR electrospray mass spectrometry
+
1
(C27
H
32
O
6
N, [M + H] found: 466.19792, calculated: 466.2224). H
2
h. The flask was flushed with argon, and D
2
O or pH 7.4 NaH
2
PO
4
NMR (500 MHz; CDCl ), 1.95-1.99
3
, 300K) δ 1.44 (9H, s, 3 × CH
3
buffer (50 mM, 8 mL) was added. The mixture was shaken for 30 min
and sonicated for 30 min to yield a slightly turbid suspension of vesicles.
This suspension was then passed 17 times through a 100-nm pore size
polycarbonate filter in an Avestin Lipofast Basic extrusion apparatus,
to yield a clear suspension of vesicles, phosphatidylcholine concentra-
(1H, m, Glu-â), 2.12-2.20 (1H, m, Glu-â), 2.41-2.50 (2H, m, 2 ×
Glu-γ), 4.20 (1H, t, J 7.2 Hz, Fmoc CH), 4.35-4.42 (1H, m, Glu-R),
2 2
4.48 (2H, d, J 7.2 Hz, Fmoc CH ), 4.64 (2H, d, J 7.0 Hz, CH ), 5.25
(1H, d, J 9.0 Hz, CHcis), 5.38 (1H, d, J 14.0 Hz, CHtrans), 5.40 (1H, d,
J 7.2 Hz, Glu NH), 5.93-5.99 (1H, m, CH), 7.35 (2H, t, J 6.1 Hz
Fmoc), 7.40 (2H, t, J 6.1 Hz, Fmoc), 7.61 (2H, d, J 6.1 Hz, Fmoc),
7.79 (2H, d, J 6.1 Hz, Fmoc).
tion 10 mM. The samples in D
2
O were adjusted to pH 7.0 with NaOD/
D
2
O and DCl/D O solutions, and these deuterated samples were used
2
1
for the H experiments. Control experiments showed that the competi-
tion experiments were pH insensitive in the range pH 7.0-7.4. All
quoted pD values were measured using a Corning pH meter fitted with
a combination glass electrode, and no attempt was made to correct for
isotope effects.
N-r-Fmoc-D-glutamic acid-r-allyl Ester. N-R-Fmoc-D-glutamic
acid-γ-tert-butyl ester-R-allyl ester (520 mg, 1.12 mmol) was dissolved
in dichloromethane (10 mL), and triflouroacetic acid (10 mL) added.
The mixture was stirred for 4 h, and the solvent was evaporated under
reduced pressure. The crude mixture was chromatographed over silica
(9:1 chloroform:methanol) to afford pure N-R-Fmoc-D-glutamic acid-
R-allyl ester as a white solid (410 mg, 89%). The final product gave
1
NMR Spectroscopy. All H NMR experiments were performed on
a Bruker DRX-500 spectrometer at 300 K. Suppression of solvent was
achieved using presaturation for samples dissolved in D
2
O or H
2
O.
the expected spectra by FT-ICR electrospray mass spectrometry
+
One-dimensional spectra were recorded using 32k complex data points.
(C23
H
23
O
6
NNa, [M + Na] found: 432.1395, calculated: 432.1418).
1
1
All H NMR spectra were referenced to 3-trimethylsilyl-2,2,3,3-d
propionic acid, sodium salt (TSP, δ ) 0.00 ppm).
4
-
3
H NMR (500 MHz; CDCl , 300K) δ 1.95-1.99 (1H, m, Glu-â), 2.12-
2.20 (1H, m, Glu-â), 2.41-2.50 (2H, m, 2 × Glu-γ), 4.20 (1H, t, J 7.2
1
9
All F NMR experiments were performed on a Bruker AM-400
Hz, Fmoc), 4.35-4.42 (1H, m, Glu-R), 4.48 (2H, d, 7.2 Hz, Fmoc),
1
9
1
spectrometer equipped with a F/ H probe at 300 K. One-dimensional
spectra were typically recorded at 376.47 MHz using 8k data points
over a spectral width of 6.9 ppm. TfOH was used as an external
reference.
2
4.64 (2H, d, J 7.0 Hz, CH ), 5.25 (1H, d, J 9.0 Hz, CHcis), 5.38 (1H,
d, J 14.0 Hz, CHtrans), 5.40 (1H, d, J 7.2 Hz, Glu NH), 5.93-5.99 (1H,
m, CH), 7.35 (2H, t, J 6.1 Hz, Fmoc), 7.40 (2H, t, J 6.1 Hz, Fmoc),
7.61 (2H, d, J 6.1 Hz, Fmoc), 7.79 (2H, d, J 6.1 Hz, Fmoc).
Solid-Phase Synthesis of N-acetyl-Gly-Ala-D-γ-Glu-Lys(N-E-
acetyl)-D-Ala-D-lactate. Solid-phase synthesis was carried out on
2-chlorotrityl chloride resin (Nova-Biochem).
General Procedure for NMR Titrations. Titrations with Vesicles.
All titrations were performed at 300 K. The antibiotic concentration
used in all the titrations was 0.2 mM, and the lipopeptide concentration
1
.0 mM. The lipopeptide was added to the vesicle solution and the
D-Lactate-2-chlorotrityl Chloride Resin. Lithium D-Lactate (0.38
g, 3.9 mmol) was dissolved in a mixture of dimethyl sulfoxide/
dichloromethane (1:1 v/v, 10 mL) and added to the 2-chlorotrityl
chloride resin (1.3 mmol, 1 g resin), and the resulting mixture was
resulting solution was mixed using agitation only; no sonication was
used. The competing ligand (Ac-KAA for H NMR and TFAc-KAA
1
1
9
for F NMR) was prepared from a solution identical to that in the
NMR tube, so as not to change the pH or concentration of the other
components in the tube (vesicles, antibiotic, lipopeptide). During the
titrations, a total of up to 100 µL of the competing solution was added
in 10 µL aliquots to 500 µL of the antibiotic lipopeptide solution. The
2
agitated overnight (15 h, under N ) and then washed successively with
dimethysulfoxide (15 mL × 6), dimethylformamide (15 mL × 6) and
dichloromethane (15 mL × 6) to afford D-Lactate-P.
D-Alanyl-D-lactate-P. N-R-Fmoc-D-alanine (1.214 g, 3.9 mmol) was
dissolved in dimethylformamide (6 mL) and then added to the resin,
and the resulting mixture was cooled to 0 °C. DIC (492 mg, 3.9 mmol)
and 4-(dimethylamino)pyridine (24 mg, 0.20 mmol) were dissolved in
dimethylformamide (4 mL) and added dropwise to the resin mixture
over 30 min (0 °C, under N ), and the mixture was agitated (under N )
1
H NMR method used TSP in water (10 mM) as an external reference
(in a glass insert in the NMR tube) for both chemical shift and relative
19
peak integrals. The F NMR method used TfOH in water as an external
reference for chemical shift and relative peak integrals. The procedures
for calculating K using the H NMR and F NMR methods have
v
1
19
2
2
1
0,15
previously been published.
Determination of Ksol. A solution of antibiotic (5 mM, 1.5 mL) in
:1 H O:D O was prepared and adjusted to pH 4.5 with HCl and NaOH.
at room temperature for 6 h. The resin was washed successfully with
dimethylformamide (15 mL × 6) and dichloromethane (15 mL × 6),
and dried under vacuum for 2 h. Two samples of the resin complex (1
mg, 1 mmol of original resin loading) were placed in solutions of 20%
piperidine in dimethylformamide (3 mL) at 290 nm. The average
absorbance of 0.9 gave an estimated loading of 0.9 mmol of D-lactate/
9
2
2
1
This solution (600 µL) was added to an NMR tube and a 1-D H NMR
spectrum was recorded. A part of the remaining antibiotic solution (600
µL) was used to dissolve an amount of ligand [N-acetyl-Gly-Ala-D-γ-