Semisynthesis of an antifungal lipopeptide echinocandin

Article abstract of DOI:10.1021/jo9822232

Compound 1 is a macrocyclic lipopeptide belonging to the echinocandin family, which has been effective in treating systemic fungal infections. In this paper, we report a unique regio-, chemo-, and stereoselective synthesis of 1 in four steps from the deacylated nucleus 3 in an impressive 83% overall yield. Highlights of this synthesis include a selective reduction of the amide to the amine and a highly stereoselective (99:1 α/β) introduction of the 2-aminoethyl hemiaminal. In addition, the synthesis of the naphthoyl side chain was accomplished in three steps in 79% overall isolated yield from commercially available 6-bromo-2-naphthol.

Full text of DOI:10.1021/jo9822232

J . Org. Chem. 1999, 64, 2411-2417
2411
Sem isyn th esis of a n An tifu n ga l Lip op ep tid e Ech in oca n d in
Michel J ournet,* Dongwei Cai,* Lisa M. DiMichele, David L. Hughes, Robert D. Larsen,
Thomas R. Verhoeven, and Paul J . Reider
Department of Process Research, Merck Research Laboratories, P.O. Box 2000, RY80E-100,
Rahway, New J ersey 07065-0900
Received November 6, 1998
Compound 1 is a macrocyclic lipopeptide belonging to the echinocandin family, which has been
effective in treating systemic fungal infections. In this paper, we report a unique regio-, chemo-,
and stereoselective synthesis of 1 in four steps from the deacylated nucleus 3 in an impressive
83% overall yield. Highlights of this synthesis include a selective reduction of the amide to the
amine and a highly stereoselective (99:1 R/â) introduction of the 2-aminoethyl hemiaminal. In
addition, the synthesis of the naphthoyl side chain was accomplished in three steps in 79% overall
isolated yield from commercially available 6-bromo-2-naphthol.
In tr od u ction
The key to the effective conversion of 2 to 1 was the
determination of the optimal point to remove the myristyl
side chain. With a prior reduction of the amide to the
amine at R1 and/or incorporation of the ethanolamine
hemiaminal at R2, a protection/deprotection sequence
would be necessary as these amines are more reactive
in the reacylation than the desired amine at R3. In fact,
protection of the amines at R1 and R3 as the Cbz groups
in an earlier approach gave a mixture of products that
required chromatography. With this scheme the overall
conversion to 1 proved to be a low-yielding process.
Instead, deacylation of 2 as the first step in the synthesis
affording 3 was the most efficient point, which obviated
the protection of the amines.
The need for new drugs to treat systemic fungal
infections has intensified due to the increase in the
immunocompromised-patient population.1 The echinocan-
dins are a class of fungicidal cell-wall active lipopeptides
that are specific inhibitors of â-(1,3)-D-glucan synthesis.2
Recently, the discovery of their activity against Pneu-
mocystis carinii3 and Aspergillus4 has broadened interest
in their development. The pneumocandins are a subset
of the echinocandins, which are produced by the fungus
Glarea lozoyensis. Their isolation, structure elucidation,
and biological evaluation have recently been reported.5
The fermentation product pneumocandin B0 26 has since
become the nucleus of structure-activity relationship
(SAR) studies at Merck with the identification of 1 as a
drug candidate.7 The structure is modified at three points
(Table 1): a naphthoyl moiety R3 replaces the 10,12-
dimethylmyristoyl side chain, the primary amide is
reduced to the primary amine (R1), and an ethanolamine
substituent R2 is present at the hemiaminal position.6
The deacylation was carried out by enzymatic hydroly-
sis. The details of this procedure will be published
elsewhere.8 After the reacylation of 3 reduction of the
primary amide was to follow. The one-step reduction of
the amide group to the amine with hydrides, such as Red-
Al, DIBALH, LAH, etc., failed to give a clean reaction.
Borane‚dimethyl sulfide afforded the desired primary
amine; however, an unstirrable gel was formed and only
a 50% conversion was obtained.9 Consequently, a two-
step, albeit more effective, procedure was developed
through dehydration to the nitrile followed by a catalytic
hydrogenation. As the introduction of the ethanolamine
at the hemiaminal position was best accomplished with
Cbz-protected ethanolamine, the hydrogenation was run
as the last step to cleave the Cbz and to reduce the nitrile
in a one-pot operation.
* To whom correspondence should be addressed. E-mail: michel_
journet@merck.com or dongwei_cai@merck.com.
(1) (a) Balkovec, J . M. Expert Opin. Invest. Drugs 1994, 3, 65. (b)
Clark, A. M. The Need for New Antifungal Drugs. In New Approaches
for Antifungal Drugs; Fernandes, P. B., Ed.; Birkha¨user: Boston, 1992;
pp 2-18. (c) Graybill, J . R. Clin. Infect. Dis. 1992, 14 (Suppl. 1), S170.
(2) (a) Hector, R. F. Clin. Microbiol. Rev. 1993, 6, 1. (b) Ruiz-Herrera,
J . Antonie van Leewenhoek 1991, 60, 73.
(3) Schmatz, D. M.; Romancheck, M. A.; Pittarelli, L. A.; Schwartz,
R. E.; Fromtling, R. A.; Nollstadt, K. H.; VanMiddlesworth, F. L.;
Wilson, K. E.; Turner, M. J . Proc. Natl. Acad. Sci. U.S.A. 1990, 87,
5950.
(4) (a) Beaulieu, D.; Tang, J .; Zeckner, D. J .; Parr, T. R., J r. FEMS
Microbiol. Lett. 1993, 108, 133. (b) Denning, D. W.; Stevens, D. A.
Antimicrob. Agents Chemother. 1991, 35, 1329.
(5) Schwartz, R. E.; Masurekar, P. S.; White, R. F. Discovery,
Production Process Development, and Isolation of Pneumocandin B0.
In Cutaneous Antifungal Agents; Rippon, J . W., Fromtling, R. A., Eds.;
Marcel Dekker: New York, 1993; pp 375-394.
Here, we wish to report these three regio-, chemo-, and
stereoselective transformations from the deacylated
nucleus 3 to produce 1 in four steps in 83% overall yield
(Scheme 1): (i) reacylation (98%); (ii) dehydration (92%);
(iii) Cbz-protected ethanolamine insertion (100%); and
(6) Bouffard, F. A.; Zambias, R. A.; Dropinski, J . F.; Balkovec, J .
M.; Hammond, M. L.; Abruzzo, G. K.; Bartizal, K. F.; Marrinan, J . A.;
Kurtz, M. B.; McFadden, D. C.; Nollstadt, K. H.; Powles, M. A.;
Schmatz, D. M. J . Med. Chem. 1994, 37, 222 and references therein.
(7) Dropinski, J . F.; Bouffard, F. A.; Balkovec, J . M.; Adefarati, A.;
Dreikorn, S.; Nollstadt, K.; Powles, M. A.; Hadju, R.; Abruzzo, G.;
Flattery, A.; Gill, C.; Kong, L.; Bartizal, K. Side Chain-Modified
Pneumocandins: Derivatives of L-733,560. Presented at the 214th
National Medicinal Chemistry Meeting, J une 14-18, 1998, VA;
Abstract No. B01.
(8) The enzymatic deacylation of the pneumocandin B0 2 will be
discussed in a forthcoming paper: Yamazaki, S.; Grover, N.; Stanik,
M.; Brinkerhoff, C.; Goklen, K.; Cai, D.; Kress, M.; Hughes, D.;
Robinson, D. Manuscript in preparation. For a similar deacylation, see:
Boeck, L. D.; Fukuda, D. S.; Abbott, B. J .; Debono, M. J . Antibiot. 1989,
42, 382.
(9) Belyk, K. M.; Bender, D. R.; Black, R. M.; Hughes, D. L.; Leonard,
W. US patent 5,552,521.
10.1021/jo9822232 CCC: $18.00 © 1999 American Chemical Society
Published on Web 03/11/1999

2412 J . Org. Chem., Vol. 64, No. 7, 1999
J ournet et al.
Ta ble 1
Sch em e 1a
a
Key: (a) enzymatic hydrolysis;8 (b) 7, Et3N, DMF; (c) cyanuric chloride, H2O (2.5 equiv), DMF, -30 °C; (d) (i) phenylboronic acid,
THF, (ii) benzyl N-hydroxyethylcarbamate, Cl3CCO2H, CH3CN, 5 °C; (e) NH4OAc, i-PrOH/H2O (9:1), 5% AcOH, Pd/Al2O3,Rh/Al2O3, H2(40
psi).
(iv) simultaneous catalytic hydrogenation of the nitrile
and the Cbz-protecting group (92%). In addition, the
preparation of the naphthoyl side chain 7 was ac-
complished in three steps off line in 79% overall isolated
yield from commercially available 6-bromo-2-naphthol.
Resu lts a n d Discu ssion
Syn th esis of th e Naph th oyl Side Ch ain an d Reacy-
la tion of 3. The preparation of the naphthoyl side chain
involved a three-step procedure from commercially avail-

Synthesis of an Antifungal Lipopeptide Echinocandin
J . Org. Chem., Vol. 64, No. 7, 1999 2413
Sch em e 2a
but unfortunately, compound 5 totally decomposed within
1 h. Obviously, the high reactivity of the reagent and the
associated instability of the product posed a real problem.
Two major factorsstemperature and watershad a tre-
mendous impact on the rate of the dehydration. Cooler
temperatures (-30 °C) resulted in a much slower reaction
(1 h), but again decomposition was observed upon further
aging. No reaction occurred at -45 °C. Interestingly, with
“wet” 4 and 2 equiv of cyanuric chloride at -30 °C, the
dehydration was achieved with >98% conversion over 30
h and minimal decomposition. The need for water (ca.
250 mol %) indicates that cyanuric chloride might be
converted in situ to a new reagent, such as 11, that is
less reactive. With this in mind, other chlorinated-
triazine derivatives 12-14 and a chloropyrimidine 15
were tested, but none of them gave the desired reaction,
only unreacted starting material.
a
Key: (a) 2.5 N NaOH, n-C7H15Br, DMSO, 75 °C, 3 h, 97%; (b)
3 mol % [Pd(OAc)2, dppp], DIPEA, CO (60 psi), DMSO/H2O (4:1),
80 °C, 12 h, 86%; (c) DCC, C6F5OH, THF, 5 °C, 16 h, 95%.
The reaction was optimized by using 2.0 equiv of
cyanuric chloride in the presence of water (1000-1100
µg of H2O/mL) in DMF (40 mL/g) at -30 °C over 30 h.
Under these conditions, a reproducible 92% isolated yield
of 5 was obtained. The product mixture contained 4% of
a byproduct that was isolated by preparative HPLC and
identified as the epi isomer 16 at the benzylic position.
F igu r e 1.
able 6-bromo-2-naphthol with 79% overall isolated yield
(Scheme 2). Etherification with 1-bromoheptane in DMSO
at 75 °C with sodium hydroxide as the base provided 8
in 97% isolated yield. Palladium-catalyzed carbonylation
in a 4:1 mixture of DMSO/water at 80 °C for 12 h gave
the carboxylic acid 9 in 86% yield with 5% of the reduced
derivative (H replaced Br).
The use of phenyl esters and mixed anhydrides for
problematic amide bond-forming reactions10 prompted
the preparation of the pentafluorophenyl ester, commonly
used in peptide bond formation. The crude acid was
reacted with pentafluorophenol in THF under standard
DCC coupling conditions to yield the pentafluorophenyl-
ester 7 in 95% yield.
The amine 3 as a TFA salt was reacylated with 3 equiv
each of 711 and triethylamine in DMF at room temper-
ature. The reaction was complete within 3 h and was
quenched with a 10% solution of aqueous acetic acid at
0 °C. Isolation of the reacylated cyclic hexapeptide 4 was
accomplished by solid-phase extraction (SPE) in 98%
overall yield. To circumvent precipitation of the side
chain, an extraction with hexane was necessary to
remove pentafluorophenol and the excess side chain.
After recrystallization, this afforded a 90% recovery of 7
in >99% purity, which could be recycled.
Am id e Deh yd r a tion . The dehydration of the primary
amide 4 to the nitrile 5 was carried out as part of the
reduction to the amine. Of the dehydrating reagents
tested only cyanuric chloride 1012 in DMF gave a clean
reaction (Figure 1). However, this dehydration proved to
be tricky and required the use of 2.5 equiv of cyanuric
chloride as opposed to the theoretical 0.33 equiv. At room
temperature, the reaction was complete within 5 min,
The formation of this byproduct increased substantially
when cyanuric chloride was used in a larger excess,
indicating that the formation of HCl was not responsible
for this side reaction but the reagent itself.13 For example,
a reaction mixture with no cyanuric chloride remaining
was completely stable for more than 1 week at -30 °C
(pH ∼2).14 To avoid increased epimerization, the reaction
mixture was quenched when 98% conversion was reached
with an equal volume of water at -30 °C. The product
mixture was warmed to room temperature and loaded
on the C-18 resin IMPAQ RG 10150 (10:1 loading) to give
5 in 92% yield along with 4% of the inseparable epi
isomer 16.
(12) Olah, G. A.; Narang, S. C.; Fung, A. P.; Gupta, B. G. B.
Synthesis 1980, 657.
(13) Cyanuric chloride was reported to be
a hydrochlorinating
reagent for alcohols: Sandler, S. R. J . Org. Chem. 1970, 35, 3967.
Chlorination could have occurred at the benzylic position to give the
epichlorohydrin. Solvolysis or epoxide formation followed by ring
opening with water would explain the epimerization of 5 to 16.
(14) This was demonstrated by the following set of experiments:
Pure 5 was treated in DMF at -30 °C with (i) HCl/H2O (6 equiv); (ii)
cyanuric acid (2 equiv); (iii) HCl/H2O (6 equiv) with cyanuric acid (2
equiv); or (iv) cyanuric chloride (2 equiv). Formation of the epi isomer
16 occurred only with the cyanuric chloride treatment (40% decomposi-
tion after 24 h); under the other conditions (i-iii) 5 was found to be
completely stable.
(10) (a) Kemp, D. S.; Carey, R. I. J . Org. Chem. 1989, 54, 3640. (b)
Kisfaludy, L.; Roberts, J . E.; J ohnson, R. H.; Mayers, G. L.; Kovacs, J .
J . Org. Chem. 1970, 35, 3563. (c) Kovacs, J .; Mayers, G. L.; J ohnson,
R. H.; Cover, R.; Ghatak, U. J . Org. Chem. 1970, 35, 1810.
(11) One equivalent of the side chain could be used. However, an
excess ensured complete conversion of the reaction in less than 3 h at
room temperature. In addition, the excess could be recovered after-
wards.

2414 J . Org. Chem., Vol. 64, No. 7, 1999
J ournet et al.
Sch em e 3
Eth a n ola m in e In ser tion . The control of the regio-,
chemo-, and stereoselectivity at the hemiaminal position
in the aminal step was challenging. The substitution
required acidic conditions. With ethanolamine as the
hydrochloride salt only a 3:1 R/â selectivity was achieved.
Also, the reaction took days with substantial decomposi-
tion. Protection of the nitrogen was considered. Of the
protecting groups tested (N-9-fluorenylmethoxycarbonyl-,
2,2,2-trichloroethoxycarbonyl-), the carbobenzyloxy (Cbz)
group gave the best results. This had further advantages,
since the deprotection could be achieved during the
reduction of the nitrile to the primary amine. In anhy-
drous acetonitrile with 15% TFA and 8 equiv of benzyl
N-(2-hydroxyethyl)carbamate at 5 °C the aminal 6 was
formed within 3.5 h as a 99:1 R/â mixture in 90% yield.
However, 10% of the bis-ethanolamine adduct 17, where
again the reactivity of the benzylic position has come into
play, was observed as a 2:1 mixture of diastereomers.
encouraging results, but the reduction did not reach more
than 70% conversion. Although rhodium on alumina was
the best catalyst, cleavage of the Cbz-protecting group
was quite sluggish. By using Pd/Al2O3 as a cocatalyst to
cleave the Cbz group with Rh/Al2O3 to reduce the nitrile
at 8 and 16 wt % (5 and 10 mol %) loading, respectively,
the transformation proceeded very efficiently.17
The solvent also proved to be a key in this reaction.
For instance, when 6 was hydrogenated in methanol only
a 50% conversion (30% yield) was observed. A large
amount of rhodium (up to 60 mol %) was needed to reach
completion with a moderate 75% yield. In addition,
formation of the epi isomer (at the benzylic position)
occurred upon aging the reaction. By adding water to an
alcohol solvent a good conversion and yield were achieved.
With optimization a 9:1 mixture of 2-propanol/water
offered the best solvent system. In particular, no epi
isomer was formed.
To overcome this reactivity, the syn 1,2-diol was
selectively and quantitatively protected as the phenylbo-
rate 18 (Scheme 3).15 Also, trichloroacetic acid gave
slightly better results than TFA. Under these new
conditions, crude nitrile 5, containing 4% of 16, was
converted to 6 in 100% isolated yield with 99:1 R/â
selectivity. The higher-than-expected yield is due to the
reequilibration16 of 16 in the reaction. Interestingly,
under the reaction conditions 75% of the epi impurity 16
was recovered during the aminal reaction to afford an
added 3% yield to the desired penultimate intermediate
6. Only 1% of the epi isomer 19 contaminated the product
mixture. Upon hydrolysis with water, the borate was
cleaved and the product was isolated again by SPE.
Hyd r ogen a tion . The reduction of the nitrile and
removal of the Cbz group remained to complete the
synthesis. Hydrogenation was expected to effect both
transformations, perhaps in one pot. Parameters, such
as catalyst, catalyst support, catalyst loading, solvent,
and pH, were important. As the support, alumina was
crucial for this transformation. For example, Rh/Al2O3
(5% Rh) reduced 6 efficiently, whereas Rh/C gave no
reaction at all. Ruthenium gave no reaction, and the
conversion was very low with PtO2. Palladium gave
The pH of the reaction was investigated as a means to
control the formation of a secondary amine byproduct (eq
1). This dimerization can be prevented or minimized by
carrying out the hydrogenation in the presence of acid
to trap the amine through salt formation (eq 2) or excess
ammonia (g6 equiv) to remove the imine intermediate
(15) Phenylboronic acid reacted selectively and quantitatively with
the syn 1,2-diol on the bottom part of the molecule within 5 min,
whereas the protection of the anti 1,2-diol on the top part of the
molecule required a 24 h reaction. Belyk, K. M.; Bender, D. R.; Hughes,
D. L.; Leonard, W. Unpublished results.
(16) Compound 5 reacts with PhB(OH)2 but 16 does not. As a result,
16 reequilibrates to 5 under the reaction conditions.
(17) Hydogen transfer (ammonium formate) with Pd/C gave
cleanly; however, 300 mol % of Pd/C was needed.
1

Synthesis of an Antifungal Lipopeptide Echinocandin
J . Org. Chem., Vol. 64, No. 7, 1999 2415
from the reaction mixture (eq 3).18 The substrate 6 was
not stable to strong mineral acids leading to epimeriza-
tion at the benzylic position and cleavage of the Cbz-
ethanolamine moiety. Acetic acid could be used, but was
not optimal. On the other hand, the intermediate is stable
to ammonia in alcohol solutions, but the hydrogenation
stopped at 20% conversion with Rh/Al2O3 as the cata-
lyst.19 Raney nickel reduced the nitrile, but a 100 wt %
loading was required and >20% of the secondary amine
was formed. Interestingly, the reaction rate increased in
the presence of ammonium acetate. Ammonium chloride
gave no reaction. By moderating the acid/base combina-
tion a suitable reaction was attained.
The yield of the reaction was highly dependent on the
ammonium acetate. Under the optimized conditions,
ammonium acetate (35 equiv) was added to 6 in a 9:1
mixture of 2-propanol/water (20 mL/g) with 5% acetic
acid. The catalyst system of 8 wt % of Pd/Al2O3 (5% Pd)
and 16 wt % of Rh/Al2O3 (5% Rh) was added. The
resulting black slurry was treated with hydrogen at 40
psi for 12 h at room temperature, reducing the nitrile
and removing the Cbz group. The crude product was
isolated by SPE in 92% yield with only 2% of the
secondary amine. A final preparative HPLC using a
Zorbax RX-C8 column and eluting with an 87:13 mixture
of water (0.15% AcOH)/acetonitrile removed the byprod-
uct. The rich cuts were lyophilized to give the pure 1 as
a bis-acetate salt in 95% recovery as an amorphous white
solid.
and using UV detection at 220 nm. The solvent systems were
(A) H2O (0.1% HClO4) and (B) acetonitrile. A pure sample of
each intermediate of the synthesis was prepared by prepara-
tive HPLC and used as a standard for the determination of
the weight %. Solid-phase extractions were performed with
C-18 resin IMPAQ RG10150 (100 Å, 150 µm) purchased from
Dupont and was reused after having been washed with 100%
methanol.
6-Br om o-2-n -h ep tyloxyn a p h th a len e (8). To a stirred
solution of 6-bromo-2-naphthol (100 g, 448 mmol) in DMSO
(1 L) was added sodium hydroxide (2.5 N aqueous, 206 mL,
515 mmol) over ca. 5 min, maintaining the temperature at 20-
25 °C. The cold bath was removed, and 1-bromoheptane (81
mL, 515 mmol) was added in one portion. The resulting yellow
solution was heated at 75 °C for 3 h and cooled to room
temperature, whereupon the ether precipitated at 45-50 °C.
The solid was filtered and washed with water (2 L) to give
143 g of 8 in 97% yield as an off-white solid. Analytical HPLC
conditions: isocratic elution 5% H2O (0.1% HClO4) in aceto-
nitrile; retention time of 6-bromo-2-naphthol, 2.55 min; 8, 9.25
min: mp 54-55 °C; 1H NMR (250 MHz, CDCl3) δ 7.90 (d, J )
1.8 Hz, 1 H), 7.65-7.56 (m, 2 H), 7.48 (dd, J ) 8.7, 1.9 Hz, 1
H), 7.15 (dd, J ) 8.9, 2.2 Hz, 1 H), 7.08 (d, J ) 2.2 Hz, 1 H),
4.05 (t, J ) 7.1 Hz, 2 H), 1.84 (quint, J ) 7.1 Hz, 2 H), 1.56-
1.33 (m, 8 H), 0.90 (t, J ) 7.1 Hz, 3 H); 13C NMR (62.5 MHz,
CDCl3) δ 157.3, 133.0, 129.7, 129.4, 129.3, 128.2, 119.9, 116.7,
106.3, 67.9, 31.7, 29.1, 29.0, 26.0, 22.5, 14.0. Anal. Calcd for
C
17H21BrO (321.26): C, 63.56; H, 6.59. Found: C, 63.52; H,
6.58.
6-n -Hep tyloxy-2-n a p h th oic Acid (9). To a stirred auto-
clave were added 6-bromo-2-heptyloxynaphthalene 8 (140 g,
436 mmol), palladium acetate (2.93 g, 13.1 mol), 1,3-bis-
(diphenylphosphino)propane (5.4 g, 13.1 mmol), DMSO/H2O
(4:1, 870 mL), and N,N-diisopropylethylamine (152 mL, 872
mmol). The reaction mixture was heated at 80 °C under 60
psi of CO for 12 h (monitored by CO uptake). The resulting
black mixture was cooled, partitioned between ethyl acetate
(2 L) and aqueous HCl (1.2 N, 1 L), and filtered through a
pad of solka-floc. The layers were separated, and the organic
extract was washed with brine (1 L). The organic layer was
dried (MgSO4) and concentrated under reduced pressure to
give 117 g of the crude acid 9 (107 g assay) in 86% yield as a
pale-yellow solid containing 5% of the reduced bromide.
Analytical HPLC conditions: isocratic elution 5% H2O (0.1%
HClO4) in acetonitrile; retention time of 9, 3.90 min; reduced
bromide, 8.70 min: mp 159-160 °C (standard); 1H NMR (250
MHz, DMSO-d6) δ 12.90 (br s, 1 H), 8.50 (d, J ) 1.7 Hz, 1 H),
7.99 (d, J ) 9.0 Hz, 1 H), 7.92 (d, J ) 1.7 Hz, 1 H), 7.90 (d, J
) 1.7 Hz, 1 H), 7.39 (d, J ) 2.5 Hz, 1 H), 7.22 (dd, J ) 9.0, 2.5
Hz, 1 H), 4.10 (t, J ) 6.9 Hz, 2 H), 1.77 (quint, J ) 6.9 Hz, 2
H), 1.51-1.30 (m, 8 H), 0.85 (t, J ) 6.9 Hz, 3 H); 13C NMR
(62.5 MHz, DMSO-d6) δ 167.6, 158.5, 136.8, 130.9, 130.4, 127.5,
126.9, 125.8, 125.7, 119.6, 106.5, 67.7, 31.3, 28.6, 28.5, 25.6,
22.1, 14.0. Anal. Calcd For C18H22O3 (286.37): C, 75.50; H,
7.74. Found: C, 75.42; H, 7.73 (standard).
P en ta flu or op h en yl 6-n -Hep tyloxy-2-n a p h th oa te (7). To
a stirred solution of the crude acid 9 (92 wt %, 107 g assay,
374 mmol) in THF (1.2 L) were added at 5 °C pentafluorophe-
nol (75.8 g, 412 mmol) and 1,3-dicyclohexylcarbodiimide (77.5
g, 374 mmol). The reaction was stirred at 5 °C for 16 h. The
mixture was filtered to remove the spent urea and concen-
trated under reduced pressure to give 7. The crude ester was
redissolved in methyl tert-butyl ether (MTBE) (500 mL). The
mixture was filtered to remove additional spent urea and
concentrated under reduced pressure to dryness to afford a
pale yellow solid. Recrystallyzation from a 2:1 mixture of
2-propanol/water at 20 mL/g afforded 160 g of 6 (99.5 wt %)
in 95% yield as a white shiny crystalline solid. Analytical
HPLC conditions: isocratic elution 5% H2O (0.1% HClO4) in
acetonitrile; retention time of 7, 10.35 min: mp 81-82 °C; 1H
NMR (250 MHz, CDCl3) δ 8.69 (d, J ) 1.7 Hz, 1 H), 8.09 (dd,
J ) 8.6, 1.8 Hz, 1 H), 7.87 (d, J ) 9.1 Hz, 1 H), 7.80 (d, J ) 8.8
Hz, 1 H), 7.22 (dd, J ) 9.0, 2.5 Hz, 1 H), 7.16 (d, J ) 2.5 Hz,
1 H), 4.10 (t, J ) 7.1 Hz, 2 H), 1.86 (quint, J ) 7.1 Hz, 2 H),
1.62-1.33 (m, 8 H), 0.88 (t, J ) 7.1 Hz, 3 H); 13C NMR (62.5
Con clu sion
In summary, an efficient regio-, chemo-, and stereose-
lective semisynthesis of the echinocandin analogue 1 has
been developed from the cyclic hexapeptide 3 in four
linear steps in 83% overall yield. A number of key
discoveries were critical to the successful transformations
of these very sensitive molecules. The preliminary de-
hydration of the amide to the cyano group overcame the
problematic direct reduction. A highly stereoselective (99:
1) and quantitative introduction of the ethanolamine at
the hemiaminal position was achieved by blocking the
nitrogen as the Cbz derivative. Through in situ protection
of the syn vicinal diol with phenylboronic acid the
epimerization of the benzylic position was overcome.
Finally, an interesting one-pot catalytic hydrogenation
of the penultimate intermediate was achieved through a
mixed catalyst of rhodium and palladium on alumina in
the presence of ammonium acetate to reduce the nitrile
and cleave the Cbz-protecting group.
Exp er im en ta l Section
Gen er a l Meth od s. Reactions were performed under a
positive atmosphere of dry nitrogen. Prior to use, the solvents
were dried over 4 Å molecular sieves to <100 µg H2O/mL.
Water content was determined by Karl Fisher titration (KF).
Commercially available reagents were purchased from Aldrich
Chemical Co. and used as received. Melting points are uncor-
rected. Elemental analyses were performed by Quantitative
Technologies Inc., Whitehouse, NJ . HR-MS analyses were
performed by electrospray ionization. All reactions were
monitored by reversed-phase HPLC with Metachem inertsil
ODS-3 column (250 × 4.6 mm) with a flow rate of 1.5 mL/min
(18) Rylander, P. N. Catalytic Hydrogenations over Platinum Metals;
Academic Press: New York and London, 1967; pp 203-226.
(19) Freifelder, M. J . Am. Chem. Soc. 1960, 82, 2386.

2416 J . Org. Chem., Vol. 64, No. 7, 1999
J ournet et al.
MHz, CDCl3) δ 162.9, 160.0, 138.2, 132.8, 131.2, 127.7, 127.4,
126.1, 121.5, 120.6, 106.5, 68.3, 31.8, 29.2, 29.1, 26.1, 22.7, 14.1,
MS m/z 453.1 (M + H)+. Anal. Calcd for C24H21F5O3 (452.42):
C, 63.72; H, 4.68. Found: C, 63.66; H, 4.75.
172.7, 172.6, 170.0, 168.5, 160.1, 158.5, 138.0, 133.0, 131.6,
129.9, 129.7, 129.2, 129.0, 127.9, 125.6, 120.8, 119.8 (CN),
116.2, 107.4, 77.2, 75.7, 74.2, 74.2, 71.4, 71.2, 69.9, 69.7, 69.6,
69.2, 68.5, 62.6, 58.8, 57.1, 56.1, 55.0, 52.4, 47.0, 38.6, 35.0,
34.5, 33.0, 30.4, 30.3, 27.2, 23.7, 23.6, 19.7, 14.4; HR-MS calcd
Rea cyla tion of 3 to 4. To a stirred solution of 38 (TFA salt,
55 wt %, 5.23 g assay, 5.56 mmol) in DMF (85 mL) were added
pentafluorophenyl 6-n-heptyloxy-2-naphthoate 7 (7.55 g, 16.7
mmol) and triethylamine (2.35 mL, 16.7 mmol). The resulting
homogeneous solution was stirred at room temperature for 3
h and cooled to 0-5 °C. The reaction mixture was quenched
with 10% aqueous acetic acid (125 mL), and the product was
extracted with hexane (120 mL). The aqueous phase (pH ∼5)
was loaded on the C-18 resin IMPAQ RG10150 (70 g). The
column was washed with three volumes of a 90:10 mixture of
water/methanol (650 mL). Compound 4 was recovered by
eluting with 2 volumes of methanol (400 mL). The methanolic
fraction was concentrated under reduced pressure to dryness
to give 6.86 g of crude 4 (95 A%, 87 wt %, 5.97 g assay) in 98%
yield as a white solid that was used in the next step without
further purification. Analytical HPLC conditions: isocratic
elution 57% H2O (0.1% HClO4) in acetonitrile; retention time
of 3, 1.85 min (overlaps with DMF); 4, 6.25 min: 1H NMR (400
MHz, CD3OD) δ 8.25 (d, J ) 0.9 Hz, 1 H), 7.78 (dd, J ) 8.6,
1.7 Hz, 1 H), 7.74 (d, J ) 9.1 Hz, 1 H), 7.65 (d, J ) 8.7 Hz, 1
H), 7.17-7.11 (m, 2 H), 7.15 (d, J ) 8.5 Hz, 2 H), 6.76 (d, J )
8.5 Hz, 2 H), 5.41 (d, J ) 2.4 Hz, 1 H), 5.11 (d, J ) 4.2 Hz, 1
H), 5.04 (d, J ) 3.1 Hz, 1 H), 4.75 (dd, J ) 12.7, 4.7 Hz, 1 H),
4.63-4.51 (m, 4 H), 4.43-4.39 (m, 1 H), 4.34-4.23 (m, 4 H),
4.21 (td, J ) 7.9, 2.2 Hz, 1 H), 4.08 (t, J ) 6.4 Hz, 2 H), 4.01-
3.91 (m, 2 H), 3.81-3.75 (m, 2 H), 2.91 (dd, J ) 17.0, 3.7 Hz,
1 H), 2.58 (dd, J ) 15.3, 9.9 Hz, 1 H), 2.45-2.41 (m, 1 H),
2.29-2.10 (m, 2 H), 2.04 (td, J ) 12.4 and 3.4 Hz, 1 H), 1.97-
1.93 (m, 1 H), 1.83 (quint, J ) 6.9 Hz, 2 H), 1.52 (quint, J )
8.2 Hz, 2 H),1.42-1.32 (m, 6 H), 1.22 (d, J ) 6.0 Hz, 3 H),
0.92 (t, J ) 6.9 Hz, 3 H); 13C NMR (100 MHz, CD3OD) δ 177.25
(CONH2), 174.7, 173.5, 172.9, 172.8, 172.5, 169.9, 169.2, 160.1,
158.5, 138.0, 133.0, 131.6, 129.7, 129.2, 129.0, 127.9, 125.5,
120.8, 116.3, 107.4, 77.0, 75.8, 74.3, 74.0, 71.4, 71.0, 70.8, 69.8,
69.2, 68.3, 62.5, 58.5, 57.1, 56.3, 55.7, 52.3, 47.1, 39.7, 38.5,
34.5, 33.0, 30.4, 30.3, 27.2, 23.7, 23.7, 19.8, 14.5; HR-MS calcd
for C52H70N8O18 1094.4808, found 1094.4773.
Deh yd r a tion of th e Am id e 4 to th e Nitr ile 5. A solution
of 4 (87 wt %, 5.97 g assay, 5.45 mmol) in dry DMF (250 mL)
was chilled to -30 °C. The water content was measured by
KF and was adjusted to ca. 1000 µg H2O/mL (ca. 0.25 g, 13.6
mmol). Cyanuric chloride (2.01 g, 10.9 mmol) was added in
one portion. The resulting pale yellow solution was stirred at
-30 °C. When 98% conversion (ca. 30 h) was reached (HPLC),
water (250 mL) was added over 10 min, and the mixture was
warmed to room temperature. The crude mixture (1:1 DMF/
H2O; pH ∼2; 500 mL) was loaded on the C-18 resin IMPAQ
RG10150 (70 g), and the column was washed with a 90:10
mixture of water/methanol (1.5 L). The nitrile 5 was eluted
by washing the column with methanol (500 mL). This fraction
was concentrated under reduced pressure to dryness to give
6.43 g of 5 (92 A%, 84 wt %, 5.40 g assay) in 92% yield as a
white solid containing 4% of epi isomer 16 and 2% of nonre-
acted 4. This mixture was used as is in the next step without
further purification. Analytical HPLC conditions: isocratic
elution 57% H2O (0.1% HClO4) in acetonitrile; retention time
of 5, 10.25 min; 16, 11.95 min: 1H NMR (400 MHz, CD3OD) δ
8.29 (d, J ) 1.3 Hz, 1 H), 7.93 (dd, J ) 8.6, 1.8 Hz, 1 H), 7.76
(d, J ) 9.1 Hz, 1 H), 7.69 (d, J ) 8.8 Hz, 1 H), 7.19 (d, J ) 2.2
Hz, 1 H), 7.14 (dd, J ) 9.0, 2.2 Hz, 1 H), 7.13 (d, J ) 8.6 Hz,
2 H), 6.75 (d, J ) 8.6 Hz, 2 H), 5.42 (d, J ) 2.3 Hz, 1 H), 5.04
(t, J ) 3.3 Hz, 2 H), 4.71 (dd, J ) 11.0, 6.4 Hz, 1 H), 4.61-
4.53 (m, 3 H), 4.38-4.33 (m, 3 H), 4.28 (qd, J ) 8.0, 1.4 Hz, 2
H), 4.22 (d, J ) 4.1 Hz, 1 H), 4.16 (td, J ) 7.9, 2.2 Hz, 1 H),
4.09 (t, J ) 6.4 Hz, 2 H), 3.98-3.86 (m, 2 H), 3.82-3.74 (m, 2
H), 2.88 (dd, J ) 17.0, 3.7 Hz, 1 H), 2.77 (dd, J ) 17.0, 8.4 Hz,
1 H), 2.46-2.41 (m, 1 H), 2.31-2.23 (m, 1 H), 2.19-2.14 (m, 2
H), 2.06 (td, J ) 12.4, 3.4 Hz, 1 H), 1.99-1.92 (m, 1 H), 1.84
(quint, J ) 6.9 Hz, 2 H), 1.52 (quint, J ) 8.2 Hz, 2 H),1.44-
1.32 (m, 6 H), 1.26 (d, J ) 6.2 Hz, 3 H), 0.91 (t, J ) 6.9 Hz, 3
H); 13C NMR (100 MHz, CD3OD) δ 174.6, 173.5, 173.0, 172.9,
for C52
H68N8O17 1076.4702, found 1076.4702.
Cbz-P r otected Eth a n ola m in e In ser tion . P r ep a r a tion
of 6. To a solution of 5 (84 wt %, 5.40 g assay, 5.01 mmol) in
THF (100 mL) was added phenylboronic acid (853 mg, 7.0
mmol) at room temperature. The solution was concentrated
under reduced pressure to dryness to remove the water. This
azeotropic distillation was repeated twice. The resulting borate
was slurried in dry CH3CN (175 mL, KF ∼50 µg H2O/mL) with
benzyl N-(2-hydroxyethyl)carbamate (7.8 g, 40 mmol), and the
reaction mixture was cooled to 5 °C. A solution of trichloro-
acetic acid (60 g, 367 mmol) in dry CH3CN (75 mL) was added
over 5 min at <5 °C. The solution was stirred for 3.5 h at 5 °C
and quenched into water at 5 °C (375 mL). The mixture was
loaded on a C-18 resin IMPAQ RG10150 (70 g). The column
was washed with a 60:40 mixture of water/acetonitrile (1.5 L)
to remove the excess benzyl N-(2-hydroxyethyl)carbamate and
phenylboronic acid. Compound 6 was then eluted with metha-
nol (600 mL). This fraction was concentrated under reduced
pressure to dryness to afford 7.31 g of 6 (91 A%, 86 wt %, 6.28
g assay) in 100% yield as a white solid containing 1% of the
epi isomer 19. The product was used as is in the next step
without further purification. Analytical HPLC conditions:
isocratic elution 45% H2O (0.1% HClO4) in acetonitrile; reten-
tion time of 6, 10.15 min; 19, 11.25 min: 1H NMR (400 MHz,
CD3OD) δ 8.34 (br s, 1 H), 7.86 (dd, J ) 8.6 and 1.7 Hz, 1 H),
7.79 (d, J ) 8.9 Hz, 1 H), 7.72 (d, J ) 8.6 Hz, 1 H), 7.28-7.21
(m, 6 H), 7.16-7.13 (m, 1 H), 7.13 (d, J ) 8.5 Hz, 2 H), 6.75
(d, J ) 8.5 Hz, 2 H), 5.33 (br s, 1 H), 5.05 (t, J ) 3.5 Hz, 2 H),
4.92 (br s, 2 H), 4.73 (dd, J ) 12.6, 4.8 Hz, 1 H), 4.61-4.55 (m,
3 H), 4.36-4.30 (m, 2 H), 4.26 (dd, J ) 12.6, 4.3 Hz, 2 H), 4.18
(td, J ) 7.9, 2.2 Hz, 1 H), 4.09 (t, J ) 6.4 Hz, 2 H), 3.98-3.87
(m, 2 H), 3.83-3.78 (m, 2 H), 3.63-3.59 (m, 1 H), 3.54-3.49
(m, 1 H), 3.29-3.23 (m, 2 H), 2.86 (dd, J ) 17.0, 3.8 Hz, 1 H),
2.77 (dd, J ) 17.0, 8.4 Hz, 1 H), 2.46-2.41 (m, 1 H), 2.28-
2.03 (m, 4 H), 2.00-1.93 (m, 1 H), 1.84 (quint, J ) 6.9 Hz, 2
H), 1.52 (quint, J ) 8.2 Hz, 2 H), 1.44-1.30 (m, 6 H), 1.24 (d,
J ) 6.2 Hz, 3 H), 0.90 (t, J ) 6.9 Hz, 3 H); 13C NMR (100
MHz, CD3OD) δ 174.2, 173.7, 173.5, 172.8, 172.8, 170.0, 168.4,
160.2, 158.5, 138.1, 133.1, 131.6, 129.9, 129.7, 129.4, 129.3,
129.0, 128.9, 128.7, 128.1, 125.6, 120.9, 119.7 (CN), 116.2,
107.4, 80.6, 77.2, 75.7, 74.7, 71.4, 70.3, 69.9, 69.5, 69.2, 68.4,
68.0, 67.4, 62.6, 58.8, 57.1, 56.1, 54.9, 51.8, 47.0, 41.7, 38.6,
35.1, 34.7, 33.0, 30.4, 30.3, 27.2, 23.7, 23.6, 19.6, 14.4; HR-MS
calcd for C62H79N9O19 1253.5492, found 1253.5534.
Ca ta lytic Hyd r ogen a tion of 6 to 1. To a solution of 6 (86
wt %, 6.28 g assay, 5.01 mmol) in a 9:1 mixture of 2-propanol/
water (130 mL) were added acetic acid (6.5 mL), ammonium
acetate (13.5 g, 175.0 mmol), Pd/Al2O3 (5% Pd, 525 mg, 0.25
mmol), and Rh/Al2O3 (5% Rh, 1.05 g, 0.5 mmol). The resulting
black slurry was treated with hydrogen (40 psi) at room
temperature for 12 h, diluted with water (400 mL), and filtered
through a pad of solka-floc. The filtrate was loaded on the C-18
resin IMPAQ RG10150 (70 g) and washed with 90:10 water/
methanol (1.5 L). The product was eluted with methanol (600
mL). This fraction was concentrated under reduced pressure
to dryness to give 7.26 g of crude 1 as the bis-acetate salt (86
A%, 79 wt %, 5.73 g assay) in 92% yield as a white solid. Crude
1 was dissolved in a 90:10 mixture of water (0.15% AcOH)/
CH3CN (250 mL), and the mixture was loaded on a preparative
HPLC column (75 × 500 mm E. Merck A/E column containing
1.1 kg of Zorbax 10 µm C8 material, detection at 220 nm, 100
mL/min). The product was eluted with 13% acetonitrile/water
(0.15% AcOH). The rich cuts were lyophilized to give 5.50 g of
the pure 1 as the bis-acetate salt (99.5 A%, 99 wt %, 5.45 g
assay) in 95% recovery as an amorphous white solid. Analytical
HPLC conditions: isocratic elution 65% H2O (0.1% HClO4) in
acetonitrile; retention time of 1, 9.80 min: 1H NMR (400 MHz,
CD3OD) δ 8.32 (br s, 1 H), 7.85-7.76 (m, 3 H), 7.25 (d, J ) 1.2
Hz, 1 H), 7.14 (dd, J ) 8.9, 1.2 Hz, 1 H), 7.13 (d, J ) 8.6 Hz,
2 H), 6.75 (d, J ) 8.6 Hz, 2 H), 5.25 (br. s, 1 H), 5.03-4.93 (m,

Synthesis of an Antifungal Lipopeptide Echinocandin
J . Org. Chem., Vol. 64, No. 7, 1999 2417
2 H), 4.75 (dd, J ) 12.7, 5.0 Hz, 1 H), 4.62-4.54 (m, 3 H),
4.34-4.23 (m, 7 H), 4.11 (t, J ) 6.4 Hz, 3 H), 3.99-3.57 (m, 7
H), 3.15-3.12 (m, 2 H), 2.95-2.93 (m, 2 H), 2.45-2.41 (m, 1
H), 2.29-1.95 (m, 7 H), 1.88 (s, 6 H), 1.87-1.82 (m, 2 H), 1.51
(quint, J ) 7.3 Hz, 2 H), 1.39-1.32 (m, 6 H), 1.22 (d, J ) 4.9
Hz, 3 H), 0.92 (t, J ) 6.9 Hz, 3 H); 13C NMR (100 MHz, CD3-
OD) δ 174.4, 174.2, 173.5, 172.8, 172.8, 170.3, 169.1, 160.3,
158.6, 138.1, 133.0, 131.6, 129.7, 129.6, 129.2, 129.0, 128.1,
125.4, 121.1, 116.3, 107.4, 80.4, 77.3, 75.6, 74.9, 72.3, 71.4, 69.9,
69.3, 69.1, 68.3, 66.7, 62.8, 58.4, 57.1, 56.4, 56.1, 51.8, 47.1,
41.2, 39.1, 38.5, 34.6, 33.0, 31.2, 30.4, 30.2, 27.2, 24.3, 23.7,
20.1, 14.4; HR-MS calcd for C54H77N9O17 1123.5437, found
1123.5428.
Ack n ow led gm en t. We wish to thank Dr. Zigiang
Guan for the HR-MS analyses.
Su p p or t in g In for m a t ion Ava ila b le: 1H and 13C NMR
spectra for compounds 1, 4, 5, and 6. This material is available
free of charge via the Internet at http://pubs.acs.org.
J O9822232