Water-soluble phosphate prodrugs of pleuromutilin analogues with potent in vivo antibacterial activity against Gram-positive pathogens

Article abstract of DOI:10.1016/j.bmcl.2009.07.115

A phosphate prodrug strategy was investigated to address the problem of poor aqueous solubility of pleuromutilin analogues. Water-soluble phosphate prodrugs 6a, 6b and 6c of pleuromutilin analogues were designed and synthesized. Three compounds all exhibi

Full text of DOI:10.1016/j.bmcl.2009.07.115

Bioorganic & Medicinal Chemistry Letters 19 (2009) 5407–5410
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Water-soluble phosphate prodrugs of pleuromutilin analogues with potent
in vivo antibacterial activity against Gram-positive pathogens
*
Liqiang Fu , Zhiteng Jiang , Zhan cai, Xin Liu, Huili He, Yushe Yang
State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica, Shanghai Institute for Biological Sciences, Chinese Academy of Sciences, Shanghai 20103, China
a r t i c l e i n f o
a b s t r a c t
Article history:
A phosphate prodrug strategy was investigated to address the problem of poor aqueous solubility of pleu-
romutilin analogues. Water-soluble phosphate prodrugs 6a, 6b and 6c of pleuromutilin analogues were
designed and synthesized. Three compounds all exhibited excellent aqueous solubility (>50 mg/mL) at
near-neutral pH and sufficient stability in buffer solution. In particular, the phenol pleuromutilin prodrug
6c displayed favourable pharmacokinetic profiles and comparable potency with vancomycin against
MSSA and MRSA strains in vivo.
Received 29 May 2009
Revised 23 July 2009
Accepted 24 July 2009
Available online 28 July 2009
Keywords:
Pleuromutilin
Ó 2009 Elsevier Ltd. All rights reserved.
Gram-positive bacteria
Water-soluble phosphate prodrugs
Due to rapid emergence of multi-drug-resistant Gram-positive
bacteria including methicillin-resistant Staphylococcus aureus
(MRSA), penicillin-resistant Streptococcus pneumoniae (PRSP), and
vancomycin-resistant enterococci (VRE),1,2 there is an urgent need
to identify and develop new antibacterial agents with novel mech-
anisms of action against drug-resistant bacterial strains. One strat-
egy is to re-evaluate old generation of antibiotics that have not
been widely used in clinical. The natural product pleuromutilin
(1) belongs to this class of under-exploited antibiotics.3
Another series of pleuromutilin analogues attracted great atten-
tion is the C-14 acyl-carbamates derivatives. Like thioether deriva-
tives, these carbamates show broad-spectrum antibacterial
activities, however, are much more metabolically stable.3 SmithK-
line recently reported two potent pleuromutilin analogues with
C-14 aroyl-carbamates side chain, 5a (SB-225586) and 5b (SB-
222734) (Fig. 2), which had excellent antibacterial potency but fur-
Pleuromutilin (1) was first isolated in 1951 from basidiomyce-
tes Pleurotus and Pleurotus passeckerianus. Further studies have
shown that this class of antibiotics selectively inhibit bacterial pro-
tein synthesis by specifically targeting the 50s subunit of the bac-
terial ribosome and display modest activity against Gram-positive
organisms in vitro and relatively weak activity in vivo.4–6 Over the
past years, the Sandoz group studied structure–activity relation-
ships (SARs) of pleuromutilin analogues focused on variations in
the C-14 glycolic acid side chain. As a result of these efforts, tiam-
ulin (2) and valnemulin (3) were successfully developed as veteri-
nary medicine to treat serious infections in swine and poultry.7,8 In
2007, GlaxoSmithKline’s novel pleuromutilin analogue retapamu-
lin (4) (Fig. 1), with excellent activity in vitro, was first approved
for human use as a topical antimicrobial agent to treat skin infec-
tions.9 Although these semi-synthetic pleuromutilins with thioe-
ther substituents in the C-14 ester side chain are generally
potent in vitro, they suffer from rapid and extensive metabolism
in vivo because of their strong hydrophobic nature.10
* Corresponding author. Tel./fax: +86 21 50806786.
E-mail address: ysyang@mail.shcnc.ac.cn (Y. Yang).
Both authors contributed equally to this work.
Figure 1. Pleuromutilin and its derivatives.
0960-894X/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2009.07.115

5408
L. Fu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5407–5410
Figure 2. Pleuromutilin analogues with C-14 aroyl-carbamates side chain.
ther development was hindered by their atrocious solubility in
water.3,11 An effective approach to overcome poor solubility is to
prepare water-soluble prodrugs.12 Hydroxyl group and phenol pre-
sented in 5a and 5b provide ideal sites for modification to generate
prodrugs. Phosphorylation of alcohols and phenols have been suc-
cessfully employed to produce water-soluble prodrugs for poorly
soluble drugs such as amprenavir, etoposide, fluconazole, propofol
and combretastatin A-4.13–16 The soluble phosphates can be
cleaved by alkaline phosphatase (ALP) in vivo to release parent
drugs. In this letter, we report our efforts in the development of no-
vel pleuromutilin phosphate prodrugs with sufficient solubility in
water, great stability in buffer and high efficacy in vivo.
Compounds 5a–d (Fig. 2) were prepared in our lab following the
method reported by Gerald Brooks.10 The in vitro antibacterial
activity of the compounds against a spectrum of resistant and sus-
ceptible Gram-positive bacteria was tested with retapamulin and
linezolid as positive controls. Minimum inhibitory concentration
(MIC) values were determined using agar dilution method accord-
ing to NCCLS. The result is summarized in Table 1. It clearly
showed that four compounds displayed excellent antibacterial
activities and the reduced compounds 5c and 5d are more potent
than the corresponding vinyl analogues 5a and 5b. To improve
their water solubility, the most potent compounds 5a, 5c and 5d
were selected as the parent drugs to generate phosphate prodrugs.
The synthesis of prodrugs 6a, 6b and 6c is outlined in Schemes 1
and 2. Reacting 5a with di-benzyl N,N-diisopropyl-phosphorami-
dite in the presence of 1H-tetrazole followed by oxidation with
mCPBA afforded dibenzyl-phosphate ester 7 in high yield (91.8%).
Deprotection of the benzyl group was accomplished by use of trim-
ethylbromosilane and methanol, and subsequent treatment with
sodium carbonate gave the crude sodium salt 6a, which was puri-
fied by C-18 reverse phase column in moderate isolate yield
(73.2%).17 Hydrogenolysis of the vinyl moieties of 6a in methanol
provided the bis-sodium salt 6b in good yield (98.2%).
Scheme 1. Reagents and conditions: (a) (1) (BnO)2PN(i-Pr)2, 1H-tetrazole, CH2Cl2,
rt, 2 h; (2) mCPBA, ꢀ40 °C, 1 h, 91.8%; (b) (1) TMSBr, CH2Cl2, 0 °C to rt, 1 h; (2)
MeOH, 2 h; (3) Na2CO3, 30 min, 73.2%; (c) Pd/C, H2, MeOH, 98.2%.
phosphite successfully produced the phosphoric acid dibenzyl es-
ter compound 8 in high yield (81.8%). Hydrogenolysis of the benzyl
and vinyl moieties of 8 followed by addition of sodium carbonate
furnished the desired product 6c (91.7% yield).19
Not surprisingly, all of the di-sodium salts of phosphate deriva-
tives were found to have significantly improved aqueous solubility
(>50 mg/mL) at near-neutral pH (Table 2). Compound 6c was
shown to be more soluble than the rest of two compounds. Stabil-
ities of 6a–c were also investigated in buffer solution at pH 7.4
with concentration of 10 mg/mL. No degradation was observed at
room temperature in a period of 8 h, and less than 0.5% of parent
drugs were detected in buffer solution after 14 days standing at
room temperature.
The regioselective phosphorylation of the phenolic hydroxyl
group of 5b at the C-14 side chain without protecting the C-11
skeleton hydroxyl group relied on the higher acidity of the pheno-
lic proton and the steric hindrance of the C-11 hydroxyl group.18
Treatment of 5b with N,N-dimethylaminopyridine (DMAP), carbon
tetrachloride, N,N-diisopropylethylamine (DIPEA) and dibenzyl
The pharmacokinetic profiles of prodrug 6a–c were evaluated in
male Sprauge-Dawley rats after oral and intravenous (iv) admin-
stration of the drugs (10 mg/kg bodyweight). Following iv admin-
Table 1
MIC ranges (
l
g/mL) of 5a–d against Gram-positive clinical isolates
MSSAa n = 5
Compounds
MRSAb n = 6
MSSEc n = 5
MRSEd n = 5
S.p.e n = 3
5a
5b
5c
5d
0.0625
0.125
0.031
0.031
0.031
0.5
0.0625
0.125
0.031
0.031
0.031
1
0.0625
0.0625
0.031
0.031
0.031
0.5
0.0625
0.125
0.031
0.031
0.031
1
0.125
0.25
0.125
0.0625
0.031
1
Retapamulin
Linezolid
a
MSSA = methicillin-susceptible Staphylococcus aureus.
MRSA = methicillin-resistant Staphylococcus aureus.
MSSE = methicillin-susceptible Staphylococcus epidermidis.
MRSE = methicillin-resistant Staphylococcus epidermidis.
S.p. = Streptococcus pneumoniae.
b
c
d
e

L. Fu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5407–5410
5409
Table 4
Efficacy of intravenous 6c in a systemic S. aureus infection model in mice
a
Organism
ED50 (mg/kg, iv)
Vancomycin
6c
MSSA 232-51
MRSA 236-29
17.38 (10.29–29.36)b
20.89 (11.11–39.30)
13.49 (8.87–20.52)
15.85 (10.30–24.39)
a
The efficacy criterion, ED50, was calculated as the dose at which mice survival
rate was 50%. Mice were inoculated intraperitoneally. Medication was given
intravenously twice, 1 h and 4 h after infection.
b
Numbers in parentheses are 95% confidence ranges.
tected at the earliest time point, indicating a rapid conversion of
the prodrug 6c to 5d. Within three dosages tested, Cmax and AUC
increased dose-dependently and approximate twofold of increase
were observed at the dose of 100 mg/kg.
The in vivo efficacy (ED50) of the water-soluble phosphate pro-
drug 6c was evaluated along with the positive control vancomycin
in lethal systemic S. aureus (MSSA and MRSA) infection model in
mice following iv administration. The results are summarized in
Table 4. The iv administered prodrug 6c exhibited potent protec-
tive effects against MSSA and MRSA strains with ED50 of
17.38 mg/kg and 20.89 mg/kg, respectively, which are comparable
to the potent antibiotic vancomycin. In addition, the acute toxicity
studies of 6c after iv administration in SPF level ICR mice (n = 50)
showed a LD50 of 1009 mg/kg, which was much greater than its
ED50
.
Scheme 2. Reagents and conditions: (a) (1) DMAP, CCl4, DIPEA, CH3CN, ꢀ10 °C; (2)
dibenzylphosphite, ꢀ10 °C to rt, 3 h, 81.8%; (b) (1) Pd/C, H2, MeOH, 40 psi, rt, 8 h;
(2) Na2CO3, 30 min, 91.7%.
In conclusion, a new serial of phosphate derivatives of pleurom-
utilin analogues were synthesized and evaluated as potential
water-soluble prodrugs. All compounds were found to be suffi-
ciently soluble and stable in water or buffer solution. More impor-
tantly, the phenolic phosphate prodrug 6c was able to efficiently
metabolize to the biologically active parent drug (5d) in vivo,
and showed great efficacy comparable to vancomycin (VCM) in
mice systemic infection model. These pre-clinical results indicate
that the prodrug 6c has potential as an promising injectable agent
for chemotherapy of systemic bacterial infections. A more compre-
hensive study of 6c will be reported in due course.
Table 2
In vitro stabilities and aqueous solubilities of phosphate derivatives 6a–c
Compound
Aqueous solubility
(mg/mL)
Buffer, pH 7.4a
(% remaining at 8 h)
Buffer, pH 7.4a
(% remaining at 14 d)
6a
6b
6c
57.6
52.3
96.8
100
100
100
99.8
99.5
99.6
a
in vitro stabilities as determined by UV/HPLC.
References and notes
istration, high level of the C-11 phosphate prodrug 6a was detected
in plasma (Cmax = 30650 ng/mL) with a slow conversion to the par-
ent drug 5a (Cmax < 5 ng/mL). The parent drug 5a was not detect-
able after oral administration of 6a.The similar results were
observed with compound 6b (data not shown). As a possible expla-
nation for this observation, the phosphate moieties attached to the
C-11 hydroxyl group may not be accessible for the alkaline phos-
phatase in a highly hindered steric environment. In contrast, the
phosphate 6c displayed satisfactory in vivo pharmacokinetic pro-
files after iv administration. As shown in Table 3, the maximum ob-
served plasma concentration of the parent drug 5d was achieved in
less than 5 min (Tmax < 5 min) and the prodrug 6c could not be de-
1. John, F. B. Exp. Opin. Ther. Target 2005, 9, 253.
2. Rice, L. B. Am. J. Infect. Control. 2006, 34, S11.
3. Hunt, E. Drugs Future 2000, 25, 1163.
4. Kavanagh, K.; Hervey, A.; Robbins, W. J. Proc. Natl. Acad. Sci. U.S.A. 1951, 37, 570.
5. Bich, A. J.; Holzapfel, C. W.; Rickards, R. W. Tetrahedron 1966, 8part 2, 359.
6. Davidovich, C.; Bashan, A.; Yaggie, R. D.; Gontarek, R. R. Proc. Natl. Acad. Sci.
U.S.A. 2007, 104, 4291.
7. Pickles, R. W. Vet. Rec. 1982, 110, 403.
8. Aitken, I. A.; Morgan, J. H.; Dalziel, R. Vet. Rec. 1999, 144, 128.
9. Parish, L. C.; Parish, J. L. Drugs Today 2008, 44, 91.
10. Brooks, G.; Burgess, W.; Colthurst, D.; Hinks, J. D. Bioorg. Med. Chem. 2001, 9,
1221.
11. Hinks, J. D.; Takle, A. K.; Hunt, E. WO9725309, 1997; Chem. Abstr. 1997, 127,
161997.
12. Stella, V. J.; NtiAddae, K. W. Adv. Drug Delivery Rev. 2007, 59, 677.
13. Corbett, A. H.; Kashuba, A. M. Curr. Opin. Invest. Drugs 2002, 3, 384.
14. Kreis, W.; Budman, D. R.; Vinciguerra, V.; Hock, K. J.; Baer, R.; Ingram, L. P.
Cancer Chemother. Pharmacol. 1996, 38, 378.
15. Bently, A.; Butters, M.; Green, S. P.; Learmonth, W. J. Org. Proc. Res. Dev. 2002, 6,
109.
16. Siemann, D. W.; Chaplin, D. J.; Walicke, P. A. Exp. Opin. Invest. Drugs. 2009 18,
189.
Table 3
Pharmacokinetic parameters of 5d derived from 6c after iv administration at the dose
of 10 mg/kg, 30 mg/kg and 100 mg/kg in male Sprauge-Dawley rats (n = 3)
a
b
c
d
17.
A
solution of 5a (200 mg, 0.4 mmol), di-benzyl N,N-diisopropyl-
Dose (mg/kg)
Tmax (h)
Cmax
(l
g/mL)
AUC0–1
(l
g h/mL)
t1/2 (h)
phosphoramidite (695 mg, 2.0 mmol), and 1H-tetrazole (140 mg, 2.0 mmol)
in dichloromethane (10 mL) was stirred at room temperature for 2 h. The
mixture was cooled to ꢀ40 °C, followed by dropwise addition of a solution of
m-chloroperoxybenzoic acid (500 mg, 2.0 mmol) in dichloromethane (10 mL).
The mixture was warmed to room temperature and stirred for 1 h. A 10%
solution of Na2S2O3 (15 mL) was added, and the mixture was stirred for 30 min.
The reaction mixture was extracted with dichloromethane and washed with
10% Na2S2O3 and then saturated NaHCO3. The organic phase was dried over
Na2SO4, filtered, and evaporated. The residue was purified by chromatography
10
30
100
0.083
0.083
0.083
3.39 0.4
11.3 4.8
67.5 11.0
1.29 0.13
5.04 0.88
32.6 1.2
0.51 0.07
0.75 0.14
0.63 0.01
a
Time at which Cmax achieved.
Maximum plasma concentration.
Area under the concentration–time curve.
Elimination half-life.
b
c
d

5410
L. Fu et al. / Bioorg. Med. Chem. Lett. 19 (2009) 5407–5410
on silica gel, to give 7 (280 mg, 91.8% yield). mp: 107–108 °C; 1H NMR (CDCl3,
syringe. After 10 min, dibenzyl phosphite (0.33 mL, 1.5 mmol) was added, and
the reaction mixture was allowed to warm to room temperature and stirred for
300 MHz) d ppm 0.81 (3H, d, J = 6.6 Hz), 0.88 (3H, d, J = 7.0 Hz), 1.10–1.81 (9H,
m), 1.15 (3H, s), 1.51 (3H, s), 2.09–2.26 (2H, m), 2.12 (1H, br s), 2.45 (1H,
quintet J = 6.9 Hz), 3.86 (3H, s), 4.36 (1H, m), 4.90–5.05 (4H, m), 5.22 (1H, dd,
J = 17.3, 1.5 Hz), 5.37 (1H, dd, J = 11.0, 1.5 Hz), 5.83 (1H, d, J = 8.5 Hz), 6.55 (1H,
dd, J = 11.0, 1.5 Hz), 6.95 (2H, d, J = 8.8 Hz), 7.28–7.40 (10H, m), 7.92 (2H, d,
J = 8.8 Hz), 7.88 (1H, br s); MS (ESI) m/z 780.4 (M+Na)+, 756.2 (MꢀH)ꢀ; 31P
NMR (121.5 MHz, CDCl3, 85% phosphoric acid standard) d –1.571 (s, 1P); Anal.
Calcd for C43H52NO9P: C, 68.15; H, 9.92; N, 1.85. Found: C, 68.20; H, 6.85; N,
1.79. To a solution of 7 (250 mg, 0.33 mmol) in dichloromethane (20 mL) at
0 °C was added trimethylsilyl bromide (111.2 mg, 0.73 mmol) via syringe, and
the mixture was stirred at 0 °C for 1 h. The solvent was evaporated, and the
residue was triturated with methanol (5 mL), stirred at rt for 2 h, then water
(10 mL) were added. The pH was adjusted to 9 by addition of 10% Na2CO3
solution, gave the crude sodium salts 6a. The residue was purified by
chromatography (C18), eluting with a gradient of 0–70% methanol in water
to give 6a (150 mg, 73.2%). mp: 197–199 °C; 1H NMR (CD3OD, 300 MHz) d ppm
0.81 (3H, d, J = 6.6 Hz), 0.88 (3H, d, J = 7.0 Hz), 1.10–1.81 (9H, m), 1.15 (3H, s),
1.51 (3H, s), 2.09–2.26 (2H, m), 2.12 (1H, br s), 2.50 (1H, quintet J = 6.9 Hz),
3.82 (3H, s), 4.25 (1H, m), 4.90–5.05 (4H, m), 5.22 (1H, dd, J = 17.3, 1.5 Hz), 5.37
(1H, dd, J = 11.0, 1.5 Hz), 5.83 (1H, d, J = 8.5 Hz), 6.55 (1H, dd, J = 11.0, 1.5 Hz),
7.05 (2H, d, J = 8.8 Hz), 7.28–7.40 (10H, m), 7.85 (2H, d, J = 8.8 Hz); 31P NMR
(121.5 MHz, CDCl3, 85% phosphoric acid standard) d 3.908 (s, 1P); the pure acid
of 6a: MS (ESI) m/z 602.0 (M+Na)+; HRMS (ESI) calcd for C29H39NO9 P (MꢀH)ꢀ
576.2362, found 576.2343.
3 h. The reaction mixture was poured into
a 0.5 M monobasic sodium
phosphate solution (10 mL), and the aqueous mixture was extracted with
ethyl acetate (3 ꢁ 20 mL). The combined organic layers were condensed in
vacuo and purified via flash chromatography (silica gel, hexane/ethyl acetate
3:1–1:1 gradient) to provide 8 (630 mg, 81.8%) as a white solid, mp: 122–
124 °C; 1H NMR (CDCl3, 300 MHz) d ppm 0.81 (3H, d, J = 6.6 Hz), 0.88 (3H, d,
J = 7.0 Hz), 1.10–1.81(9H, m), 1.15 (3H, s), 1.51 (3H, s) 2.09–2.26 (2H, m), 2.12
(1H, br s), 2.35 (1H, quintet J = 6.9 Hz), 3.36 (1H, d), 5.10–5.16 (4H, d,
J = 8.8 Hz), 5.22 (1H, dd, J = 17.3, 1.5 Hz), 5.37 (1H, dd, J = 11.0, 1.5 Hz), 5.80
(1H, d, J = 8.5 Hz), 6.60 (1H, dd, J = 11.0, 1.5 Hz), 7.20 (2H, d, J = 8.8 Hz), 7.30–
7.40 (10H, m), 7.70 (2H, d, J = 8.8 Hz), 8.0 (1H, br s); MS (ESI) m/z 766.80
(M+Na)+; 31P NMR (121.5 MHz, CDCl3, 85% phosphoric acid standard) d ꢀ6.826
(s, 1P); Anal. Calcd for C42H50NO9P: C, 67.82; H, 6.78; N, 1.88. Found: C, 67.94;
H, 6.77; N, 1.81. To a solution of 8 (400 mg, 0.54 mmol) in methanol (50 mL)
was added Pd on carbon (60 mg, 10% by wt Pd), and the mixture was shaken in
a Parr hydrogenator under a 40 psi atmosphere of hydrogen for 8 h. The
mixture was filtered through Celite, and the solvent was evaporated. Methanol
and water were added, and the pH was adjusted to 9 by addition of 10% Na2CO3
solution. The sample was purified by chromatography (C18), eluting with a
gradient of 0–70% methanol in water to give 6c (300 mg, 91.7% yield). mp:
200–202 °C; 1H NMR (CDCl3, 300 MHz) d ppm 0.81 (3H, d, J = 6.6 Hz), 0.88 (3H,
d, J = 7.0 Hz), 1.10–1.81 (12H, m), 1.15 (3H, s), 1.51 (3H, s), 2.09–2.26 (2H, m),
2.38 (1H, br s), 2.40 (1H, quintet J = 6.9 Hz), 3.45 (1H, d), 5.78 (1H, d, J = 8.5 Hz),
7.30 (2H, d, J = 8.8 Hz), 7.90 (2H, d, J = 8.8 Hz); 31P NMR (121.5 MHz, CDCl3, 85%
phosphoric acid standard) d -1.747 (s, 1P); the pure acid of 6c: MS (ESI) m/z
565.2 (MꢀH)ꢀ; HRMS (ESI) calcd for C28H39NNaO9 P (M+Na)+ 588.2316, found
588.2338.
18. Silverberg, L. J.; Dillon, J. L.; Vemishetti, P. Tetrahedron Lett. 1996, 37, 771.
19. Compound 5b (500 mg, 1.0 mmol) and dimethylaminopyridine (12.2 mg,
0.1 mmol) were dissolved in anhydrous acetonitrile (20 mL) under argon. The
reaction mixture was cooled to ꢀ10 °C, carbon tetrachloride (0.5 mL,
5.0 mmol) and diisopropylethylamine (0.37 mL, 2.1 mmol) were added via