Syntheses and antifungal activity of pseudomycin side-chain analogues. Part 1

Article abstract of DOI:10.1016/S0960-894X(00)00423-6

We have described herein the syntheses of three novel series of aromatic ring containing pseudomycin side-chain analogues. Preliminary biological evaluations of these analogues clearly indicate that it is possible to synthesize rigid pseudomycin side-chain analogues without compromising in vitro antifungal activity. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0960-894X(00)00423-6

Bioorganic & Medicinal Chemistry Letters 10 (2000) 2101±2105
Syntheses and Antifungal Activity of Pseudomycin Side-Chain
Analogues. Part 1
James Jamison, Stuart Levy, Xicheng Sun, Doug Zeckner, William Current,
Mark Zweifel, Michael Rodriguez, William Turner and Shu-Hui Chen*
Lilly Research Laboratories, A Division of Eli Lilly and Company, Lilly Corporate Center, Indianapolis, IN 46285, USA
Received 1 June 2000; accepted 12 July 2000
AbstractÐWe have described herein the syntheses of three novel series of aromatic ring containing pseudomycin side-chain analo-
gues. Preliminary biological evaluations of these analogues clearly indicate that it is possible to synthesize rigid pseudomycin side-
chain analogues without compromising in vitro antifungal activity. # 2000 Elsevier Science Ltd. All rights reserved.
Introduction
bearing analogues of cyclic lipopeptide echinocandin B
(ECB) (e.g., aromatic, unsaturated, etc.) exhibited
improved antifungal potency in comparison to the par-
ent compound. Bearing these considerations in mind,
we decided to prepare a series of aromatic ring con-
taining pseudomycin side-chain analogues in search for
compounds possessing improved antifungal potency. In
this paper, we wish to report the synthesis and our pre-
liminary biological evaluation of three types of aromatic
ring bearing side-chain analogues as exempli®ed by 6, 7,
and 8 shown in Figure 2.
Pseudomycins, produced as metabolites by Pseudomo-
nas syringae in plants,1 represent a novel series of nat-
ural products possessing antifungal activities against a
number of fungi. Structurally, the pseudomycins share
similarity with several other lipodepsinonapeptides gen-
erated by P. syringae in plants including syringomycins,2
syringotoxin3 and syringostatins.4 Depending on the
nature of side chain, pseudomycins are subdivided into
pseudomycin A, B, and C0, etc. In 1994 the gross struc-
ture of pseudomycin A (PSA) was reported by Ballio et
al.5 The structures of pseudomycin B and C0 (PSB and
PSC0) were fully established through extensive NMR
analysis and later con®rmed by a pseudomycin semi-
synthetic e€ort.6 When evaluated against Candida albi-
cans and Cryptococcus neoformans, two major fungal
pathogens responsible for systemic fungal infection,
PSB exhibited superior activity to that achieved by
amphotericin B (AMB), the commonly used antifungal
drug.
Chemical Synthesis
The synthetic scheme employed for the preparation of
6a±c is shown in Scheme 1. Palladium mediated cou-
pling8 of 3-bromobenzaldehyde 9 with 1-octyne or 1-
dodecyne a€orded the desired acetylene bearing pro-
ducts 10a (51%) and 10b (51%), respectively. Saturation
of the triple bond moieties in 10a and 10b was a€ected
by standard hydrogenolysis, leading to 11a and 11b in
good yields. Treatment of 11a and 11b with the lithium
enolate generated in situ from t-butylacetate provided
the corresponding b-hydroxyesters 12a (67%) and 12b
(33%), respectively. Alternatively, O-alkylation of 3-
hydroxybenyaldehyde with dodecyl bromide a€orded
10c (15%), which was further converted to the requisite
b-hydroxyester 12c (77%) using the same chemistry as
described for 12a. Treatment of 12a±c with TFA yielded
the desired b-hydroxyacids, all of which were further
coupled with ZPSB (5a) to provide the corresponding
three pairs of 30-diastereomers 13a±c. The resulting dia-
stereomers 13a±c were separated using reverse-phase
Previous work from this institution demonstrated that
certain pseudomycin side-chain analogues can be pre-
pared via N-acylation of the amino function of Residue
1 within the pseudomycin core 5a or 5b with various
b-hydroxyacids (see Fig. 1 for structures).6 It was fur-
ther demonstrated by Rodriguez et al. that the 30-
hydroxyl group was essential for the optimal antifungal
activity of PSC0.6 Additionally, a recent publication by
Debono et al.7 documented that many rigid side-chain
*Corresponding author. Tel.: +1-317-276-2076; fax +1-317-276-
5431; e-mail: chen_shu-hui@lilly.com
0960-894X/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(00)00423-6

2102
J. Jamison et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2101±2105
Figure 1. Representative pseudomycin structures.
Figure 2. Pseudomycin side-chain analogues.
Scheme 1. Synthesis of pseudomycin side-chain analogues 6a±6c.

J. Jamison et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2101±2105
2103
HPLC to give the respective 30-enantiomers. As expec-
ted, the 30a-hydroxyl group bearing isomer had a rela-
tively shorter retention time than that observed with
their 30b-hydroxyl bearing counterpart.6 Final removal
of the N-protective groups (Cbz) in 13a±c (from either
30a or 30b enantiomers) thus completed the syntheses of
pseudomycin side-chain analogues 6a±c. The structures
of ®nal products were con®rmed by spectroscopic NMR
analyses and mass spectra.
and lyophilization, were secured by their mass spectro-
metry and 1H NMR spectroscopic analyses.
The synthetic sequence employed for pseudomycin ana-
logues 8a±c is outlined in Scheme 3. The starting mate-
rials 21a±c used in this case were obtained via standard
Mitsunobu reaction from 3-hydroxy-benzaldehyde. The
reaction of 21a with triphenylphosphoranylidene-acet-
aldehyde gave the unsaturated aldehyde 22a (47%),
which was then converted to 24a (70%) via the usual
palladium catalyzed hydrogenation. Alternatively, 21b
and 21c were ®rst converted to their unsaturated esters
22b (70%) and 22c (80%), and were in turn converted to
the corresponding aldehydes 24b (60%) and 24c (43%)
via intermediates 23b and 23c, respectively. Titanium
tetrachloride promoted allylsilane addition to 24a±c
a€orded the homoallylic alcohols 25a±c in yields rang-
ing from 61±78%. The carbinols 25a±25c were next
transformed, via the same three-step sequence as descri-
bed for 19a±c in Scheme 2, to the requisite acids 26a±c in
excellent overall yields. Standard coupling of 26a±c with
5a or 5b provided the expected 30-racemic products 27a±c.
Compound 27a was deprotected to give 8a via catalytic
hydrogenation in 17% overall yield from 26a. The alloc-
protected 30-racemic adducts 27a were separated by
semipreparative HPLC to yield the corresponding two
stereoisomers 27a (30a) and 27a (30b). Alloc deprotec-
tion9 gave the desired products 8a (30a) and 8a (30b) in
21 and 25%, respectively. Similarly, the alloc protecting
groups in 27b and 27c were removed to give 8b and 8c in
As shown in Scheme 2, the synthesis of 7a±7c began
with conversion of 3-hydroxybenzaldehyde 14 to the
ethers 15a±c under Mitsunobu conditions. Compounds
15a±c were then transformed, via the Wittig reaction,
into the corresponding enolethers 16a±c, and thereafter,
the desired aldehydes 17a±c, upon exposure to aqueous
acid. Allylmagnesium bromide addition to 17a±c furn-
ished the carbinols 18a±c in modest yields, which were
converted to the requisite b-hydroxyl acids 19(a±c) via a
sequence consisting of osmylation, vicinol diol cleavage
(NaIO4) and subsequent oxidation (NaClO2). The acids
thus obtained were activated as their HOBT esters, and
then coupled with either 5a (for 19a and 19c) or 5b (for
19b) to a€ord the expected products 20a±c in 23±35%
yield. Finally, 20b was deallylated using tributyltin
hydride and Pd(PPh3)2Cl2 to provide 7b in 44% yield.9
Intermediates 20a and 20c were deprotected under
standard hydrogenation conditions,10 to the desired
products 7a (85%) and 7c (50%). The structures of 7a±c,
obtained through semipreparative HPLC puri®cation
Scheme 2. Synthesis of side-chain analogues 7a±7c.

2104
J. Jamison et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2101±2105
Scheme 3. Syntheses of side-chain analogues 8a±8c.
low overall yields (16±24%). Again, the structures of
these pseudomycin side-chain analogues were con®rmed
by their mass spectrometry and proton NMR spectro-
scopy.
yield a starting concentration of 20 mg/mL in the ®rst
wells following the addition of inoculum. Serial 2-fold
dilutions in 100 mL aliquots were made in 96-well
microtiter plates using a QuadFlex automatic pipeting
liquid delivery instrument (Titertek Instruments Inc.,
Huntville, AL).
Biological Evaluation
The fungal yeast isolates used for in vitro assay were
grown on Sabouraud dextrose agar slants at 35 ꢀC for
24 h. C. albicans and C. neoformans were suspended in
saline and adjusted to 2Â105 conidia/mL in Sabourauds
dextrose broth (DIFCO, Detroit, MI). A. fumigatus
spores were gently teased from mycelia grown on Potato
dextrose agar (DIFCO, Detroit, MI) and suspended in
0.1% Tween 80 and saline.
The newly synthesized pseudomycin analogues (6, 7,
and 8) were subjected to a robotic microdilution micro-
titer testing system to determine their MICs. Candida
albicans, Cryptococuus neoformans andAspergillus fumi-
gatus account for about 90% of systemic fungal infec-
tions in immunocompromised patients. Therefore, all
new pseudomycin side-chain analogues described herein
were tested against these fungi. Pseudomycin B was
included as a positive control.
The results of in vitro evaluation of 6±8 are detailed in
Table 1. Pseudomycin B (PSB), the positive control,
exhibited excellent MIC's against C. albicans (0.312±
0.625 mg/mL) and C. neoformans (0.039 mg/mL). When
tested against A. fumigates, PSB showed only weak
activity (MIC >20 mg/mL).
The samples used for this in vitro assay were prepared
according to the following procedure: Pseudomycin
analogues (6±8) were dissolved in 100% DMSO. Com-
pound solutions were diluted in Sabourauds broth to

J. Jamison et al. / Bioorg. Med. Chem. Lett. 10 (2000) 2101±2105
Table 1. Biological evaluation of pseudomycin analogues 6±8
2105
incorporation of an aromatic ring, located one or two
carbons away from 30-hydroxyl group on the side chain
(e.g., 7a±c and 8a±c), is detrimental to the antifungal
activity associated with these pseudomycin analogues.
Furthermore, in view of the promising antifungal activ-
ities displayed by 6b (a) and 6c (a), it is evident that it is
possible to synthesize novel rigid pseudomycin side-
chain analogues without compromising in vitro poten-
cies. Being mindful of the fact that compounds 6b (a)
and 6c (a) were also capable of inducing irritation, we
continue our search for novel pseudomycin side-chain
analogues endowed with good antifungal activities and
acceptable toxicity pro®les. The results of these investi-
gations will be detailed in the near future.
Compounds
30-OH
con®guration C. albicans C. neoformans A. fumigates
MICa
(In vitro)
(mg/mL)
PSB
6a (ꢀ)
6a(ꢁ)
6b(ꢀ)
6b(ꢁ)
6c(ꢀ)
6c(ꢁ)
7a
7b
7c
8a
8a(ꢀ)
8a(ꢁ)
8b
a
a
b
a
b
a
b
0.312±0.625
5.0
>2 0
0.039
0.312
5 . 0
0.01
1.25
0.02
0.625
>20
>20
>2 0
1.25
5.0
0.625
2.5
>20
>20
>20
>20
>20
>20
>20
>2 0
>2 0
>20
>20
10
Racemic >20
>20
>20
1.25
>20
>2 0
>2 0
20
Racemic >20
Racemic 15
Racemic >20
a
>2 0
>2 0
10
b
Racemic
Racemic
8c
20
2.5
Acknowledgements
aMIC, Lowest drug concentration required to inhibit 90±100% of
visible growth compared to controls.
We are indebted to Drs. J. Munroe and B. Laguzza for
their supports and encouragement.
In vitro evaluation of 6a±c revealed the following
trends: (1) 30-a stereochemistry is required for the opti-
mal antifungal activity. Thus, 6a (a) was found to be
more potent than 6a (b) when evaluated against Can-
dida and Cryptococcus. Likewise, 6b (a) exhibited
greater activity against all three major fungi than its 30-b
counterpart 6b (b). This observation is in agreement
with that discovered with PSC0.6 (2) C-18 side-chain
bearing analogues, 6b and 6c, proved to be more potent
than the C-14 side-chain counterpart 6a and (3) gen-
erally speaking, all analogues within this subgroup
demonstrated excellent activity against Cryptococcus.
(4) With the exception of 6b (a), the remaining analo-
gues within this class exhibited weak activity towards
Aspergillus. (5) All six analogues within this class were
tested in the mice tail vein irritation/toxicity assay.11
Although two C-14 side-chain bearing analogues, 6a (a)
and 6a (b), were devoid of tail vein irritation, four
remaining analogues (including more potent analogues
6c (a) and 6b (a)) were all capable of inducing irritation
in this assay.
References and Notes
1. Harrison, L.; Teplow, D. B.; Rinald, M.; Strobel, G. J.
Gen. Microbiol 1991, 137, 2857.
2. Segre, A.; Bachmann, R. C.; Ballio, A.; Bossa, F.; Grgur-
ina, I.; Iacobellis, N. S.; Marino, G.; Pucci, P.; Simmaco, M.;
Takemoto, J. Y. FEBS Lett. 1989, 255, 27.
3. Ballio, A.; Bossa, F.; Collina, A.; Gallo, M.; Iacobellis, N.
S.; Paci, M.; Pucci, P.; Scaloni, A.; Serge, A.; Simmaco, M.
FEBS Lett. 1990, 269, 269377.
4. Isogai, A.; Fukuchi, N.; Yamashita, S.; Suyama, K.;
Suzuki, A. Tetrahedron Lett. 1990, 31, 695.
5. Ballio, A.; Bossa, F.; Giorgio, D. D.; Ferranti, P.; Paci, M.;
Pucci, P.; Scaloni, A.; Serge, A.; Strobel, G. A. FEBS Lett.
1994, 355, 96.
6. Rodriguez, M.; Belvo, M.; Moris, R.; Zeckner, D.; Current,
W.; Kulanthaivel, P.; Zweifel, M. Submitted to Bioorg. Med.
Chem. Lett. 2000.
7. Debono, M.; Turner, W. W.; LaGrandeur, L.; Burkhardt,
F. J.; Nissen, J. S.; Nichols, K. K.; Rodriguez, M.; Zweifel, M.
J.; Zeckner, D. J.; Gordee, R. S.; Tang, J.; Parr, T. R., Jr. J.
Med. Chem. 1995, 38, 3271. Recently Kasibhatla et al. repor-
ted a similar approach to the design of AMP deaminase inhi-
bitors. J. Med. Chem. 2000, 43, 1508.
8. Heck, R. F. Palladium Reagents in Organic Syntheses;
Academic: San Diego, 1990; pp 299±306.
9. General deallylation procedure: Dangles, O.; Guibe, F.;
Balavoine, G.; Lavielle, S.; Marquet, A. J. Org. Chem. 1987,
52, 4984.
10. The hydrogenation reactions were conducted in 10%
HOAc/MeOH for about 40 min. Prolonged hydrogenation
may lead to saturation of the double bond presented on pseu-
domycin core.
11. General procedure for performing tail vein toxicity assay:
Mice (Outbred, male ICR mice about 18±20 g; harlan Spran-
gue Dawley, Indianapolis, IN) were treated intravenously (IV)
through the lateral tail vein with 0.1 mL of testing compounds
(20 mg/kg) at 0, 24, 48, and 72 h. Two mice were included in
each group. Compounds were formulated in 5.0% dextrose
and sterile water for injection. Mice were monitored closely for
signs of irritation including erythema, swelling, discoloration,
necrosis and tail loss, etc., for a total of 7 days following ®rst
treatment. Mice were also observed for any other signs of
adverse e€ects that are indicative of toxicity.
We next examined the antifungal activity of the novel
side-chain analogues 7a±c as their 30-racemates. None of
these analogues exhibited good activity against Candida
and Aspergillus. Furthermore, 7a and 7b were totally
devoid of activity against Cryptococcus. The C-18 side-
chain bearing analogue, 7c, showed relatively weaker
activity against this Cryptococcus in comparison to PSB.
When side-chain analogues 8a±c were evaluated, we
observed the following trends: (1) all analogues within
this subset exhibited marginal activities against Candida
and Cryptococcus. (2) The best analogue within this
class, 8c, showed relatively weak activity against Cryp-
tococcus, with a MIC value of 2.5 mg/mL. All other
analogues failed to inhibit the growth of Cryptococcus
at the dose of up to 20 mg/mL.
In conclusion, we have described herein the syntheses of
three types of rigid side-chain containing pseudomycin
analogues. In light of our testing results, it is clear that