W. Wang et al. / Bioorg. Med. Chem. Lett. 13 (2003) 489–493
491
potency. Overall, compounds showing the best in vitro
and in vivo antifungal profiles also turned out to be the
most potent histamine-releasing agents. From these
studies, we found A-192411 (compound 1) to possess
acceptable in vitro, in vivo and CV safety profiles (Fig.
1). MIC data, generated using NCCLS method, are
presented on a small subset of derivatives around com-
pound 1, Table 1.
NaOH to give the corresponding pentyl ether in 90%
yield. The latter underwent palladium mediated Suzuki-
coupling with 4-borono-benzoic diacid (Pd(PPh3)4,
DMF-water, 100 ꢁC) to give the desired side chain C
(Fig. 1) in 82% yield.7
Final couplings
Tripeptides A (17) and B (20) were first coupled at the
hydroxyproline and homotyrosine sites to give the cor-
responding linear hexapeptide in 57% yield (i) Tripeptide
B:diethyl amine (DIEA), CH3CN; (ii) Tripeptide A, CMC-
HOBT; (iii) DIEA, CH3CN; (iv) LiOH, THF–water; and
(v) DPPA-NaHCO3-DMF]. The latter underwent suc-
cessive N-Fmoc deprotection (ornithine), ester hydro-
lysis (azidoproline) and ring-cyclization to the
corresponding peptide 21 in 60% overall yield. Azide
reduction and guanidylation proceeded to give the cor-
responding bis-Cbz-guanidine in 85% overall yield,
Scheme 3. N-Boc removal, under TFA conditions, fol-
lowed by N-acylation with the terphenyl side chain C and
hydrogenation to remove all three N-Cbz groups, in a
near quantitative yield, gave A-192411 as its acetic acid
salt after HPLC purification. This convergent synthesis
(27 steps) was used to prepare multi-gram quantities of
the antifungal agent A-192411 suitable for in vitro and in
vivo preclinical evaluation and characterization.
While most cyclic peptides maintained relatively inter-
esting potencies as anti-microbial agents (Table 1, 3, 9,
and 11), their assessment in mast cell-histamine release
directed us towards compound 1 as a preferred entity
for further in vivo characterization. A-192411 contains
two novel amino acids and a lipophilic side chain
attached at the l-ornithine a-amino position. This com-
pound was synthesized through a convergent approach
consisting of couplings between two tripeptides A and B
and the 400-O-pentyl terphenyl carboxylate side chain C;
(Fig. 1). Previous syntheses of Echinocandin B were
reported by the Ohfune and Evans groups.6
Tripeptide A contains two unnatural amino acids, an
aryl-N-glycine homotyrosine and the 4-(S)-azido-pro-
line (Fig. 1). The aryl-N-glycine homotyrosine moiety
14 was prepared from the reaction of 2-chloroanisole 12
with (l)-N-methylcarbamoyl aspartic anhydride, under
Friedel–Craft’s conditions, to give the corresponding
amino acid in 88% yield. Ketone-hydrogenation affor-
ded the desired homotyrosine in 90% yield (not shown).
The resulting N-methyl carbamate-amino acid (AA) was
further treated with HBr and reprotected as the
N-Fmoc-AA 13, obtained in 90% yield. The latter
underwent ortho-nitration (70% yield) followed by
hydrogenation and N-acylation with Cbz-protected gly-
cine afforded precursor 14 in 68% yield, (Scheme 1).
A-192411 showed solubility >3 mg/mL is water, and
0.77 mg/mL phosphate buffered saline. The measured
octanol:water partition coefficient (logD) is 2.45 and
2.47 at pH 6.5 and 7.4, respectively. The high aqueous
solubility allowed ready formulation for iv delivery, as
was used in the preclinical animal studies. The in vitro
rat, dog and human plasma protein binding at 5, 10,
and 50 mg/mL of A-192411 acetate showed no apparent
gender-related differences observed for all species.
Plasma protein binding did not appear to be concen-
tration-dependent (range 5–50 mg/mL, in rats and in
humans) except in dogs, the mean plasma protein bind-
ing of A-192411 was lower (91.0%) at 5 mg/mL when
compared to the 10 or 50 mg/mL (98.9 and 99.6%,
respectively). The pharmacokinetic evaluation of
A-192411 was conducted in mouse, rat, dog and mon-
key at 1 mg/kg and is summarized in Table 2. In mon-
keys the plasma concentration profile was fit to a two
compartmental model, characterized by a terminal
elimination half-life of 3.3 h (range 2.2–5.9 h) after a 1
mg/kg intravenous dose. The terminal phase volume of
distribution of A-192411 was 0.9 L/kg (range 0.6–1.3 L/
kg). A-192411 had a plasma clearance of 0.17 L/h kg
(range 0.11–0.26 L/h kg). In dogs, it was characterized by
a long terminal elimination half-life of 30.4 h (range 23.1–
36.9 h), with a terminal phase volume of the distribution
Separately, 4-(R)-hydroxyproline methyl ester 15,
which, was used in the preparation of both tripeptides A
and B, was coupled to (l)-N-Boc-threonine to give
dipeptide 16 in 98% yield. The latter underwent hydro-
lysis, N-Boc deprotection (84% yield) and coupling to
(L)-N-a-Boc-e-Fmoc-Ornithine to give tripeptide B (17)
in 50% yield, (Scheme 2). Similarly, ester 15 was N-Boc-
protected and treated, in a one-pot reaction, with
methanesulfonyl chloride followed by sodium azide, to
give the desired derivative 18 in 89% yield. The latter was
N-deprotected in TFA and coupled to N-Boc-threonine to
give the corresponding dipeptide 19 in 90% yield. Sub-
sequent TFA deprotection and coupling to ortho-N-glycyl
homotyrosine moiety 14 (Scheme 1), afforded tripeptide A
(20) in an overall yield of 72–82%. The terphenyl side
chain was prepared as follows: 4-(40-bromophenyl)-phenol
was O-alkylated with iodopentane in the presence of
Scheme 1. (a) AlCl3, (L)-N-methylcarbamoyl Asp. Anhydride; (b) H2, Pd/C; (c) (i) HBr/AcOH; (ii) FmocOSu; (d) NaNO2, AcOH; (e) (i). H2, Pd/C;
(ii) Z-GlyOSu, THF.