Pholipeptin
J . Org. Chem., Vol. 62, No. 1, 1997 107
1
5
and plipastatin.16 The former is a macrolide, and
Several long-range interresidue ROEs (Table 3) pro-
vided information on the conformation of the molecule.
The peptide chain of OHDa-Leu-4-Asp-1 must be folded
toward the Thr residue, the branching point of the cyclic
peptide lactone, based on the ROE correlations from a
â-H of Thr to H-2 and 3 of OHDa. Regarding the cyclic
moiety, the Asp-2 residue was considered to be located
spatially in the vicinity of Leu-5 and Ser-2, according to
the ROE correlations from R-H of Asp-2 to NH and R-H
of Ser-2 and to R-H of Leu-5 and from âa-H of Asp-2 to
âb-H of Ser-2.
Furthermore, absolute configurations of the constituent
amino acids were analyzed by chiral HPLC. Peaks of
hydrolysate were identified by coinjection of each au-
thentic optically pure sample. The configuration of Leu
was detected with a Crownpak CR(+) column, and those
of other amino acids were detected with a Chiralpak WH
column. The configurations of D- and L-Leu could not be
distinguished by the latter column. The molar ratio of
amino acids was calculated from the peak area from the
Chiralpak WH column. All of the Leu and Ser moieties
were shown to be in the D-configuration, and both Ile and
Thr were in the L-configuration. The peak area of D-Leu
was analyzed as 77.2 pmol × 5 mol/100 pmol pholipeptin,
because no L-Leu was detected when the Crownpak CR-
B
2
the latter, being a detergent peptide, inhibits phospho-
lipase both A and C. The inhibitory activity of 1 on PI-
PLC is comparable to that of fluvirucin B . The structure
2
of 1 is completely different from those of other PI-PLC
inhibitors. So 1 would be a new guide compound for
preparation of more potent PI-PLC inhibitors.
Exp er im en ta l Section
Mea su r em en t of P I-P LC Activity. PI-PLC activity was
assayed as described before15 by the method of Wahl et al.
with slight modifications.
Isola tion of P h olip ep tin (1). The producing strain has
been deposited in the National Institute of Bioscience and
Human Technology Agency, Ministry of International Trade
and Industry, Tsukuba, J apan, under the accession number
FERM P-14943. P. fluorescens BMJ 279-76F1 was cultured in
production medium (1.5% glycerol, 1.5% soluble starch, 0.5%
2
3
soy bean meal, 1.5% fish meal, and 0.2% CaCO , pH 7.4 before
autoclaving) at 27 °C for 96 h. The culture broth (2 L) was
centrifuged at 10 000 rpm, and the filtrate was applied onto a
Diaion HP-20 column. The column was washed with H O and
2
50% MeOH, and the active fraction was eluted with 100%
MeOH. After addition of water to the eluate, evaporation was
carried out in vacuo to remove the MeOH. The resultant
solution was adjusted to pH 2 and extracted with an equal
volume of EtOAc. The organic extract was concentrated to
dryness and dissolved in MeOH. The MeOH-insoluble fraction
was removed by filtration, and the soluble fraction was
fractionated on a silica gel column (stepwise gradient elution,
(+) column was used. The absolute intensity of L-Thr
was low (57.7 pmol for 100 pmol pholipeptin), as reported
for arthrofactin.4 The fact that the intensity became
lower as time passed (data not shown) suggests the
instability of Thr in 0.02 N HCl solution. Two Asp
moieties were contained as a racemic mixture, 76.8 and
100% CHCl
further chromatographed on a silica gel column (stepwise
gradient elution, BuOAc/BuOH/MeOH/H O (6/4/1/1) to (2/4/1/
)), again, and followed by Sephadex LH-20 column chroma-
tography (MeOH) to yield 1 (20 mg).
NMR Sp ectr oscop y. NMR spectra were recorded with a
-mm C/ H/ F triple-resonance probe or a 5-mm inverse
probe, operating at 500.13 MHz for H and 125.77 MHz for
C, and the sample temperature was maintained at 30 °C.
Chemical shifts (δ) were reported in ppm relative to internal
tetramethylsilane. Spectral simulations and determinations
of the sign of couplings (J , Hz) were not carried out. HOHAHA
spectra were recorded with mixing times of 55, 85, and 112
3 3
to CHCl /MeOH (10/5)). The active fraction was
2
1
9
0.5 pmol for 100 pmol of D-Asp and L-Asp, respectively,
so the configurations of Asp-1 and Asp-2 could not be
determined simply.
13
1
19
17
5
1
We tried to determine the absolute configurations of
1
3
1
3
Asp-1 and -2 by a biosynthetic method using C-labeled
1
3
material. We fed L-Asp-4- C to the production medium
and then subjected the purified 1 to 13C NMR analysis.
As a result, the â-carbonyl carbon signal of Asp-1 was
observed to be stronger than any other carbonyl carbon
signals including the â-carbonyl carbon of Asp-2. The
intensity was 1.4 times as strong as that of the adjacent
ms. ROESY spectra were measured with a spin-locking time
1
of 200 ms. HMQC experiments were acquired for J
C-H
) 145
Hz, whereas HMBC spectra were obtained for 2
5.6, 7.0, 8.3 Hz.
,3
J
C-H ) 4.0,
1
3
carbonyl signal at δ 168.6 (Thr). Therefore, L-Asp-4- C
was concluded to be incorporated selectively into the
position of Asp-1, and chirality of Asp-1 was determined
as the L-configuration. Thus, optical configurations of
all amino acids in 1 were determined as shown. Related
Da ta for p h olip ep tin (1): white powder; UV (MeOH) end
-
1
1
absorption; IR (KBr) 3295, 2957, 1657, 1547 cm ; H NMR
DMSO-d
with TFA, pH 4) see Table 1; 13C NMR (DMSO-d
with TFA, pH 4), see Table 1; FAB-MS, m/z ) 1376 (M + Na
H), 1374 (M + Na - H), 1352 (M - H); FAB-HRMS calcd
for C H O N (M - H) 1352.7929, found 1352.7922.
Am in o Acid An a lysis. For standard amino acid analysis,
1 mg of 1 dissolved in 1 mL of 6 N HCl was placed in a sealed
glass tube and heated at 110 °C for 20 h. After evaporation,
the residue was dissolved in 1 mL of 0.02 N HCl and subjected
to amino acid analysis on a Hitachi amino acid analyzer L-8500
system. Retention times in the amino acid analysis (min) were
as follows: Asp (5.58), Thr (6.38), Ser (6.97), Ile (16.37), Leu
(17.36).
Ch ir a l HP LC An a lysis. One mg of 1 was dissolved in 6
N HCl (1 mL) and heated at 120 °C for 20 h. The reaction
mixture was evaporated in vacuo, and the residue was
dissolved in 0.02 N HCl and analyzed by a Chiralpak WH (4.0
mm φ × 150 mm, Daicel Chemical Industries, LTD) column
and a Crownpak CR(+) (4.6 mm φ × 250 mm, Daicel Chemical
(
6
6
+
4
peptides include arthrofactin, acyclic peptide isolated
6
4
110 20 11
from Arthrobacter sp. in which, however, the ester bond
was formed between the C-terminal Asp and OHDa.
Pholipeptin (1) inhibited PI-PLC of human carcinoma
A431 cells with an IC50 of 5.8 µM. It inhibited the
enzyme noncompetitively with the substrate. The phos-
pholipase C inhibitor (PCI) peptide was also reported to
inhibit the enzyme in a noncompetitive manner.14 The
structure of PCI peptide is Tyr-Arg-Lys-Met-Arg-Leu-
Arg-Tyr, in which Tyr residues of the N- and C-terminus
are necessary for the inhibition. But there was no Tyr
residue in 1. Moreover, this peptide is rich in basic amino
acids, such as Arg and Lys, while 1 does not contain any
basic amino acid. The mechanism of PI-PLC inhibition
for 1 is being studied by preparation of the partial
structure. Other PI-PLC inhibitors include fluvirucin
(
15) Ui, H.; Imoto, M.; Umezawa, K. J . Antibiot. 1995, 48, 387-
390.
(16) Umezawa, H.; Aoyagi, T.; Nishikiori, T.; Okuyama, A.; Yam-
agishi, Y.; Hamada, M.; Takeuchi, T. J . Antibiot. 1986, 39, 737-744.
(17) For triple-resonance experiments using this probe, see: Miyake,
T.; Koyama, Y. Carbohydr. Res. 1994, 258, 11-26.
(
14) Homma, Y.; Takenawa, T. J . Biol. Chem. 1992, 267, 21844-
2
1849.