Phosphonooxymethyl prodrugs of the broad spectrum antifungal azole, ravuconazole: Synthesis and biological properties

Article abstract of DOI:10.1016/j.bmcl.2003.08.029

Synthesis of phosphonooxymethyl derivatives of ravuconazole, 2 (BMS-379224) and 3 (BMS-315801) and their biological evaluation as potential water-soluble prodrugs of ravuconazole are described. The phosphonooxymethyl ether analogue 2 (BMS-379224) and N-ph

Full text of DOI:10.1016/j.bmcl.2003.08.029

Bioorganic & Medicinal Chemistry Letters 13 (2003) 3669–3672
Phosphonooxymethyl Prodrugs of the Broad Spectrum Antifungal
Azole, Ravuconazole: Synthesis and Biological Properties
Yasutsugu Ueda,a,* John D. Matiskella,a Jerzy Golik,a Timothy P. Connolly,a
Thomas W. Hudyma,a Srini Venkatesh,a Mandar Dali,b Shin-Hong Kang,a
Nancy Barbour,b Ravi Tejwani,b Sailesh Varia,b Jay Knipe,a Ming Zheng,a
Marina Mathew,a Kathy Mosure,a Junius Clark,a Lucinda Lamb,a Ivette Medin,a
Qi Gao,a Stella Huang,a Chung-Pin Chena and Joanne J. Bronsona
aBristol-Myers Squibb Company, Pharmaceutical Research Institute, Wallingford, CT 06492-7660, USA
bBristol-Myers Squibb Company, Pharmaceutical Research Institute, New Brunswick, NJ 08903-0191, USA
Received 22 April 2000; accepted 12 August 2003
Abstract—Synthesis of phosphonooxymethyl derivatives of ravuconazole, 2 (BMS-379224) and 3 (BMS-315801) and their biolog-
ical evaluation as potential water-soluble prodrugs of ravuconazole are described. The phosphonooxymethyl ether analogue 2
(BMS-379224) and N-phosphonooxymethyl triazolium salt 3 (BMS-315801) were both prepared from ravuconazole (1) and bis-tert-
butyl chloromethylphosphate, but under two different conditions. Both derivatives were highly soluble in water and converted to
the parent in alkaline phosphatase, and also in vivo (rat). However, BMS-315801 was found to be less stable than BMS-379224 in
water at neutral pH. BMS-379224 (2) has proved to be one of the most promising prodrugs of ravuconazole that we tested, and it is
currently in clinical evaluation as an intravenous formulation of the broad spectrum antifungal azole, ravuconazole.
# 2003 Elsevier Ltd. All rights reserved.
Ravuconazole 1 (BMS-207147) licensed to Bristol-
Myers Squibb Company from Eisai (ER-30346) is a
potent and broad spectrum antifungal agent, exhibiting
excellent antifungal activity against fungal pathogens
such as Candida albicans, Cryptococcus neformans with
exceptional activity against Aspergillus species.1 It is
currently under clinical evaluation by BMSas an oral
agent, but to optimize its therapeutic potential, devel-
opment of an intravenous (iv) formulation was deemed
necessary. This is essential for therapy of serious sys-
temic fungal infections such as pulmonary Aspergillus
infections, particularly for patients to whom oral medi-
cation cannot be readily administered. The poor aqu-
eous solubility of ravuconazole (0.6 mg/mL) precludes
its development for intravenous administration.2 Thus,
an approach of a water-soluble prodrug was employed
to develop an iv formulation.
The hydroxyl group and the triazole moiety present in
ravuconazole (1) were thought to be most logical sites
for derivatization, leading to potentially water-soluble
prodrugs. Phosphonooxymethyl derivatives of alcohols
and cyclic imides have shown to be successful water-
soluble prodrugs for poorly water-soluble drugs such as
paclitaxel and phenytoin.3 These phosphonooxymethyl
prodrugs are known to be hydrolyzed by the ubiquitous
mammalian enzyme, alkaline phosphatase (ALP) to
chemically labile hemiacetal derivatives which are
spontaneously cleaved to the parent alcohols or cyclic
imides. However, the sterically hindered hydroxyl group
in ravuconazole 1, coupled with the presence of the
reactive neighboring groups posed a formidable chal-
lenge for functionalizing this hydroxyl group. The pre-
paration of quaternary ammonium salts and triazolium
salts as potential prodrugs of tertiary amine- or triazole-
containing drugs has been reported,4 but there was no
report on the preparation of quaternary phosphonooxy-
methyl triazolium salt of antifungal triazoles. Here, we
would like to report our unique synthesis of phosphono-
oxymethyl ether of ravuconazole, 2 (BMS-379224) and
the preparation of N-quaternary phosphonooxymethyl
*Corresponding author. Tel.:+1-203-677-6303; fax: +1-203-677-
7702; e-mail: yasutsugu.ueda@bms.com
0960-894X/$ - see front matter # 2003 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2003.08.029

3670
Y. Ueda et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3669–3672
triazolium derivative of ravuconazole, 3 (BMS-315801)
as well as their physico-chemical and biological proper-
ties as potential prodrugs.
phase column. The structure of 2 was confirmed by single-
crystal X-ray analysis of the bis-lysine salt (Scheme 1).8
BMS-315801, an N-phosphonooxymethyl quaternary
derivative of ravuconazole, 3 was prepared from ravu-
conazole (1) by direct N-alkylation with bis-tert-butyl
chloromethylphosphate 4 at ꢁ80 ꢀC (neat)10 as shown
in Scheme 2. Interestingly, the tertiary butyl groups
were cleaved concomitantly during this process to pro-
vide modest yield (ca. 40%) of 311 as the quaternary salt
after C-18 reverse-phase column chromatography. The
1
structure of 3 was supported by its H NMR (DMSO-
The phosphonooxymethyl ether derivative of ravuco-
nazole 2 (BMS-379224) was prepared in two steps from
ravuconazole (1) by O-alkylation with di-tert-butyl
chloromethyl phosphate 4 followed by cleavage of the
protecting tert-butyl ester moiety. The first step, O-alkyl-
ation was successfully achieved under rather unconven-
tional conditions. The methods reported for the
preparation of O-phosphonooxymethyl ether of alco-
hols3a were not successful for this particular alcohol-con-
taining ravuconazole. After many attempts to alkylate
this hydroxyl moiety, a direct alkylation reaction of
oxy-anion, which was generated from ravuconazole and
NaH in THF, with a semi-purified di-tert-butyl chloro-
methylphosphate 4, at room temperature to 50 ꢀC, pro-
vided a modest but variable yield (10–55%) of the
desired tert-butyl phosphate ester product 6.5 The
phosphate 4 was prepared by the reaction of chloro-
methyliodide with ammonium salt of di-tert-butyl
phosphoric acid 5.6 The yield of 6 was dependent on the
purity of the alkylating agent 4. When highly purified
di-tert-butyl chloromethylphosphate 4 was used, very
little or no desired product was observed. Later, it was
found that addition of iodine and an excess amount of
NaH as an iodide source provided more consistent and
much higher yield of the phosphonooxymethyl ether
derivative 6.8 Deprotection of the tert-butyl group was
accomplished by use of trifluoroacetic acid in CH2Cl2 to
produce the phosphonooxymethyl ether 29 in good yield
(>80%). The sodium salt was prepared by treatment
with sodium carbonate and purified by C-18 reverse
d6) which indicated the presence of the methylene pro-
tons (–OCH2N+) at 5.60–5.76 ppm as a multiplet and
its MSwhich indicated m/z 548 (M+H). The location
of the quaternary group at the 4-nitrogen was supported
by the NOE experiment where NOEs were observed
between the quaternary methylene protons and both
triazole protons (C-3 and C-5).
Solubility and Stability
The amorphous sodium salt and the crystalline bis-
lysine salt of 2 were soluble in water at pH 7 (>30 mg/
mL). The amorphous zwitterion 3 was also soluble in
water (>30 mg/mL) at pH 6.6.
The phosphonooxymethyl ether 2 was stable in solution
with concentration of 1–10 mg/mL at pH ranges of 3.6–
8.9, generating less than 1% of ravuconazole after 13
days at room temperature. However, quaternary tria-
zolium salt 3 was not as stable as the phosphonoooxy-
methyl ether 2 in solution at pH 7, generating 5–6% of
ravuconazole at room temperature during 24-h period.
It also became cloudy during this period due to pre-
cipitation of insoluble ravuconazole.
The phosphonooxymethyl ether 2 was also stable as
crystalline bis-lysine salt at 40 ꢀC and 75% relative
humidity for 3 weeks. No degradation and no genera-
tion of ravuconazole were observed. Whereas, the N-
quaternary phosphonooxymethyl triazolium salt 3 was
hygroscopic with moisture uptake of 5–7% at 25 ꢀC and
40% relative humidity. It was also susceptible to
hydrolysis under accelerated conditions of heat and
humidity (e.g., 40 ꢀC and 75% relative humidity).
These results on the relative stability in solution and
solid-state indicate the phosphonooxymethyl ether 2
(BMS-379224) is much more promising than the qua-
ternary triazolium salt 3 (BMS-315801).
Scheme 1.
In Vitro Conversion
Both compounds, 2 (BMS-379224) and 3 (BMS-315801)
were readily converted to ravuconazole (1) by ALP. In a
typical experiment, the disappearance of 3 (BMS-
315801) and formation of ravuconazole (1) after incu-
bation of 3 (BMS-315801) with human placentral ALP
is shown in Figure 1. The results demonstrate that these
phosphates are substrates for ALP, an enzyme of wide
Scheme 2. (a) ClCH2OP(O)(OtBu)2, 4, THF; (b) neat, 75–85 ꢀC;
(c) C-18 column.

Y. Ueda et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3669–3672
3671
Figure 1. Disappearance of BMS-315801 (&) and formation of ravu-
conazole (BMS-207147) (^) as a function of time after incubation of
BMS-315801 in alkaline phosphatase solution. BMS-315801 in buffer
solution (~) is shown as a control.
Figure 3. Plasma concentrations of BMS-315801 (&) and ravucona-
zole (BMS-207147) (^) in rats following iv infusion of BMS-315801 at
5.9 mg/kg (average of two rats).
Figure 2. Plasma concentrations of BMS-379224 (&) and ravucona-
zole (BMS-207147) (^) in rats following iv administration of BMS-
379224 at 6.76 mg/kg (meanÆS.D., n=3).
distribution in various mammalian tissues. These phos-
phate prodrugs were also converted to ravuconazole
after incubation in liver S9 homogenates from rat dog
and human tissues.12 However, these prodrugs were not
rapidly converted to ravuconazole in rat, dog or human
plasma. We believe this is due to high binding of plasma
proteins such as albumin to these phosphates of highly
lipophilic molecules.13
Figure 4. Efficacy of intravenous BMS-379224 and oral ravuconazole
in a systemic Candida albicans infection model in mice.14
potential water-soluble prodrugs. Both compounds, the
phosphonooxymethyl ether analogue, 2 (BMS-379224)
and the N-quaternary phosphonooxymethyl triazolium
analogue, 3 (BMS-315801) were soluble in water at neu-
tral pH and both phosphates generated parent ravucona-
zole (1) in in vitro systems (ALP or liver S9) and also in
vivo in rats, dogs and in monkeys. However, the prodrug
2 (BMS-379224) was much more stable than the qua-
ternary phosphate 3 (BMS-315801), meeting our require-
ment of solution and solid-state chemical stability.
In Vivo Conversion
The rapid in vivo conversion of the phosphate prodrugs
2 (BMS-379224) and 3 (BMS-315801) to ravuconazole
(1) upon iv administration to rats is shown in Figures 2
and 3, respectively. A similar rapid conversion of 2
(BMS-379224) to ravuconazole was also demonstrated
in dogs and monkeys.
The results from pre-clinical evaluations, including ani-
mal safety indicate that this phosphonooxymethyl ether
derivative 2 (BMS-379224) is one of the most promising
prodrugs of ravuconazole we tested, and 2 (BMS-
379224) has now advanced to its clinical study as an iv
prodrug of ravuconazole.
In vivo antifungal efficacy
In vivo efficacy of the water-soluble phosphate 2
(BMS-379224) was evaluated in the systemic C. albicans
infection model in mice. The results are summarized in
Figure 4. The iv administered prodrug 2 (BMS-379224)
showed comparable efficacy as orally administered
ravuconazole.
Acknowledgements
In summary, two phosphonooxymethyl derivatives of
ravuconazole, 2 and 3 were synthesized and evaluated as
We would like to thank other members of the Micro-
biology Research group, the Discovery Analytical

3672
Y. Ueda et al. / Bioorg. Med. Chem. Lett. 13 (2003) 3669–3672
Science group and the Chemical Process Research
group for their support in this program.
1H, J=9, 9, 2 Hz), 6.91 (ddd, 1H, J=9, 9, 2 Hz), 5.35 (dd, 1H,
J=6, 6 Hz), 5.29 (d, 1H, J=15 Hz), 5.21 (dd, 1H, J=6, 6 Hz),
5.19 (d, 1H, J=15 Hz), 3.86 (q, 1H, J=7 Hz), and 1.35 (d, 3H,
J=7 Hz). MS(EIS À) m/z 546 [M–H]À. Anal. calcd for
C23H18F2N5O5SP/Na2/3.5H2O: C, 42.21: H, 3.85: N, 10.70:
Na, 7.03. Found: C, 42.32: H, 3.83: N, 10.60: Na, 7.04; pKa
(potentiometric, water): 2.2, 6.1.
References and Notes
1. (a) Fung-Tomc, J.; Huczko, E.; Minassian, B.; Bonner, D.
Antimicrob. Agents Chemother. 1998, 42, 313. (b) Naito, T.;
Hata, K.; Tsuruoka, A. Drugs Future 1996, 21, 20. (c) For a
recent review article on ravuconazole, see Arikan, S.; Rex,
J. H. Curr. Opin. Investig. Drugs 2002, 3, 555.
10. The mixture dissolved originally in THF was concentrated
and maintained at 80 ꢀC, removing THF. See also, Golik, J.;
Matiskella, J. D.; Ueda, Y. U.S. Patent, US 6,235,728 (May
22, 2001). Chem. Abstr. 2001, 135, 5703.
1
11. Compound 3: H NMR (DMSO-d6) d 10.21 (s, 1H), 8.98
2. Our efforts to formulate parenteral ravuconazole were not
successful even by use of solubilizing co-solvents, including
cyclodextrin-based agents. It is believed that this is primarily
due to its exceptionally poor aqueous solubility.
3. (a) For paclitaxel Golik, J.; Wong, H. S. L.; Chen, S. H.;
Doyle, T. W.; Wright, J. J. K.; Knipe, J.; Rose, W. C.;
Casazza, A. M.; Vyas, D. M. Bioorg. Med. Chem. Lett. 1996,
6, 1837. (b) For phenytoin, Varia, S. A.; Stella, V. J. J. Phar-
maceut. Sci. 1984, 73, 1087. Stella, V. J. Adv. Drug Deliv. Rev.
1996, 19, 311.
(s, 1H), 8.39 (s, 1H), 8.17 (d, 2H, J=9 Hz), 7.90 (d, 2H, J=9
Hz), 7.38–7.25 (m, 2H), 6.97–6.91 (m, 1H), 5.76–5.60 (m, 2H),
5.00 (d, 1H, J=14 Hz), 4.75 (d, 1H, J=14 Hz), 4.04 (q, 1H,
J=7 Hz), 1.16 (d, 3H, J=7 Hz). MS(MH +=548). 19F NMR
(DMSO-d6) d À73.87 (s, 0.1F), À107.5 (s, 1F), À111.3 (s, 1F).
Anal. calcd for C23H20F2N5O5SP/0.05CF3CO2H/0.4H2O: C
49.51, H 3.75, N 12.50, F 7.29. Found: C 49.39, H3.71, N
+
12.42, F 7.98 (H2O 1.21%, Karl Fisher Method). MS(ESI
m/z 548 [M+H]+. pKa (potentiometric, water): 2.0, 4.6.
)
12. For ALP incubation studies, prodrugs (25 mg/mL, final
concentration) were incubated with a solution of human pla-
cental ALP (Sigma #P-1391, 50 U/L, final concentration) in
TRISbuffer, pH 7.4. For liver homogenate studies, prodrugs
(25 mg/mL, final concentration) were incubated with commer-
cially available (GenTest Corporation, Woburn, MA, USA)
9000 g supernatant fractions (S9) from rat, dog and human
livers fortified with cofactors (NADPH, MgCl2, glucose-6-
phosphate, glucose-6-phosphate dehydrogenase). Aliquots
were removed at various times, extracted with CH3CN and
analyzed for prodrug and parent using a specific LC/MS
method.
13. Ueda, Y.; Mikkilineni, A. B.; Knipe, J.; Rose, W. C.;
Casazza, A. M.; Vyas, D. M. Bioorg. Med. Chem. Lett. 1993, 3,
1761 A separate experiment indicated that the conversion of
BMS-315801 to ravuconazole in alkaline phosphatase was sig-
nificantly retarded by addition of albumin (from 10 to 90%).
14. Female ICR mice (Harlan–Sprague Dawley) were used in
these studies. Mice were infected systemically by iv injection of
C. albicans SC 5314 (inoculum size of 105 per animal). Ther-
apy with intravenously administered BMS-379224 or orally
administered ravuconazole was begun 2 days post-infection
and given once a day for 7 consecutive days. Both compounds
were administered at daily doses equivalent to 5 mg/kg ravu-
conazole. Efficacy was determined by survival for 30 days fol-
lowing infection.
4. (a) Ohwada, J.; Murasaki, C.; Yamazaki, T.; Ichihara, S.;
Umeda, I.; Shimma, N. Bioorg. Med. Chem. Lett. 2002, 12,
2775. (b) Ichikawa, T.; Kitazaki, T.; Matsushita, Y.; Yamada,
M.; Hayashi, R.; Yamaguchi, M.; Kiyota, Y.; Okonogi, K.; Itoh,
K. Chem. Pharm. Bull. 2001, 49, 1102. (c) Krise, J. P.; Zygmunt,
J.; Georg, G. I.; Stella, V. J. J. Med. Chem. 1999, 42, 3094.
1
5. Compound 6: H NMR (300 MHz, CDCl3) d 8.35 (s, 1H),
7.98 (d, 2H, J=9 Hz), 7.76 (s, 1H), 7.71 (d, 2H, J=9 Hz), 7.63
(s, 1H), 7.36–7.27 (m, 1H), 6.86–6.78 (m, 2H), 5.53 (dd, 1H,
J=28, 6 Hz), 5.53 (dd, 1H, J=9, 6 Hz), 5.17 (d, 1H, J=15
Hz), 5.03 (d, 1H, J=15 Hz), 4.01 (q, 1H, J=7 Hz), 1.47 (s,
9H), 1.45 (s, 9H), 1.37 (d, 3H, J=7 Hz). MS(ESI +) m/z 660
[M+H]+.
6. Compound 5 was prepared from di-tert-butyl phosphoric
acid7 and tetrabutyl ammonium hydroxide in methanol.
7. Zwierzak, A.; Kluba, M. Tetrahedron 1971, 27, 3163.
8. (a) Ueda, Y.; Matiskella, J. D.; Golik, J.; Hudyma, T. W.;
Chen, C.-P. U.S. Patent, US 6,362,172 (March 26, 2002).
Chem. Abstr. 2001, 135, 122628. (b) Chen, C.-P.; Connolly,
T. P.; Kolla, L. R.; Matiskella, J. D.; Mueller, R. H.; Pendri,
Y.; Petsch, D. T. U.S. Patent, US 6,448,401 (September 10,
2002). Chem. Abstr. 2002, 136, 386259.
9. Compound 2 (sodium salt): 1H NMR (500 MHz, D2O) d
8.91 (s, 1H), 7.92 (s, 1H), 7.81 (d, 2H, J=8 Hz), 7.80 (s, 1H),
7.77 (d, 2H, J=8 Hz), 7.21 (dd, 1H, J=15, 9 Hz), 6.99 (ddd,