Pleuromutilin derivatives having a purine ring. Part 2: Influence of the central spacer on the antibacterial activity against Gram-positive pathogens

Article abstract of DOI:10.1016/j.bmcl.2008.10.123

Structural modification of the 4-piperidinethio moiety, as a spacer of the first pleuromutilin analogues 2A and 2B having a purine ring, led to discovery of the novel pleuromutilin derivatives 14B and 17B. These compounds with good solubility in water showed promising in vitro antibacterial activity against various Gram-positive bacteria including MRSA, PRSP, and VRE and have potent in vivo efficacy.

Full text of DOI:10.1016/j.bmcl.2008.10.123

Bioorganic & Medicinal Chemistry Letters 19 (2009) 170–174
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Pleuromutilin derivatives having a purine ring. Part 2: Influence of the central
spacer on the antibacterial activity against Gram-positive pathogens
a
a
Yoshimi Hirokawa a, , Hironori Kinoshita a, , Tomoyuki Tanaka , Takanori Nakamura ,
*
Koichi Fujimoto b, Shigeki Kashimoto b, Tsuyoshi Kojima b, Shiro Kato a
a Chemistry Research Laboratories, Dainippon Sumitomo Pharma Co., Ltd, Enoki 33-94, Suita 564-0053, Japan
b Pharmacology Research Laboratories, Dainippon Sumitomo Pharma Co., Ltd, 3-1-98 Kasugade Naka, Konohana-ku, Osaka 554-0022, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Structural modification of the 4-piperidinethio moiety, as a spacer of the first pleuromutilin analogues 2A
and 2B having a purine ring, led to discovery of the novel pleuromutilin derivatives 14B and 17B. These
compounds with good solubility in water showed promising in vitro antibacterial activity against various
Gram-positive bacteria including MRSA, PRSP, and VRE and have potent in vivo efficacy.
Ó 2008 Elsevier Ltd. All rights reserved.
Received 26 August 2008
Revised 23 October 2008
Accepted 28 October 2008
Available online 5 November 2008
Keywords:
Pleuromutilin
Gram-positive bacteria
MRSA
PRSP
VRE
Purine ring
Spacer
MIC
Sulfide
Piperazine ring
The search for new antibacterial agents is important due to
increasing multi-drug resistant Gram-positive bacteria including
methicillin-resistant Staphylococcus aureus (MRSA), penicillin-
resistant Streptococcus pneumoniae (PRSP), and vancomycin-resis-
tant enterococci (VRE).1–3 We have focused our attention on the
natural product pleuromutilin4–7 (1) with unique mechanism of
action and consequently no cross-resistance to the marketed anti-
bacterial agents. In part 1 of this series,8 we identified the structur-
ally novel pleuromutilin analogues 2A9 and 2B (Table 1) having a
purine ring as a polar and water solubilizing group. These com-
pounds, with good solubility in water, displayed potent in vitro
antibacterial activity against various Gram-positive bacteria
including MRSA, PRSP, and VRE. Furthermore, their in vivo efficacy
against S. aureus Smith systemic infection model in mice was
essentially equipotent to that of the marketed antibiotic vancomy-
cin. As an extension to that study, we have directed our efforts to
understanding the influence of the 4-piperidinethio moiety as a
central spacer in 2A and 2B on the in vitro and in vivo antibacterial
activities. The present communication deals with the structure-
activity relationships (SARs) of a novel series of pleuromutilin ana-
logues 3A–17B with several spacer moieties as shown in Table 1.
The synthesis of several nitrogen-containing heteroalicycles
and the alkylene chains 26a–e, 26g,h, 27c, and 28c (the optically
active pyrrolidine; R and S forms, respectively), and 29f having a
sulfide linkage is illustrated in Scheme 1. The thiol derivatives were
prepared via the thioesters 21a–h according to a method described
in the patent literature9,10 but using appropriate nitrogen-contain-
ing heteroalicycles (20a–f) and alkylene chains (20g,h), all of
which are either commercially available or prepared according to
a published procedure.11 Reaction of the resultant thiol derivatives,
without isolation, with the mutilin 14-tosyloxyacetate 1812 or the
19,20-dihydromutilin 14-tosyloxyacetate 19, which was prepared
from the 19,20-dihydromutilin13 using a method described in the
literature12 for 18, followed by deprotection of the corresponding
N-Boc derivatives 22a–e, 22g,h, 23c, 24c, and 25f gave the desired
sulfur-linked derivatives 26a–e, 26g,h, 27c, 28c, and 29f as a
hydrochloride (see Fig. 1).
The corresponding ether-linked analogue 32 of 29f having a
sulfide linkage was prepared as follows. O-alkylation of 1-benzyl-
oxycarbonyl-4-piperidinol, which was obtained by reaction of
4-piperidinol with benzyl chloroformate, with ethyl bromoacetate
* Corresponding author. Tel.: +81 6 6337 5906; fax: +81 6 6337 6010.
E-mail address: hironori-kinoshita@ds-pharma.co.jp (H. Kinoshita).
Present address: Faculty of Pharmacy, Osaka Ohtani University, 3-11-1 Nishikiori-
kita, Tondabayashi, Osaka 584-8540, Japan.
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.10.123

Y. Hirokawa et al. / Bioorg. Med. Chem. Lett. 19 (2009) 170–174
171
Table 1
In vitro and in vivo antibacterial activities of 2–17.
Me
R1
OH
N
O
R2
N
SPACER
N
N
Me
O
O
N
Me
Me
•nHCl
R3
O
Compounda
R1
n
MICb
(lg/mL)
MSSAc
NSPACER
MSSAc
MRSAd
PSSPe
PRSPf
S. p.g
VREh
M. c.i
H. j.j
ED50 (mg/kg, iv)
k
2A
2B
Vinyl
Vinyl
1
1
0.25
0.5
0.063
0.063
0.032
0.016
0.125
0.25
1
1.86
N
S
S
0.063
0.5
0.125
1
0.016
0.032
0.032
0.032
0.032
0.25
0.032
0.125
1
2
1.47
3.13
N
S
3A
3B
4A
Vinyl
Vinyl
Vinyl
1
1
1
0.016
0.008
N
S
0.25
0.25
0.5
1
0.016
0.063
0.016
0.125
0.063
0.25
0.063
0.125
2
1
2.02
2.61
N
N
N
S
50.004
S
S
5A
6A
Vinyl
Vinyl
1
1
0.05
0.5
2
1
0.063
0.063
0.125
0.125
0.032
0.016
0.5
0.25
0.25
2
2
1.28
3.16
0.25
N
N
7A
8A
Vinyl
Vinyl
1
1
0.5
2
0.125
0.032
0.25
50.004
0.5
0.125
0.125
2
4
4.07
2.77
S
0.25
0.5
0.032
0.016
0.063
N
CH2S
S
9A
Vinyl
1
1
2
0.063
0.125
0.032
0.5
0.25
8
4.81
N
O
10A
11B
12B
13B
Ethyl
Ethyl
Ethyl
Vinyl
1
1
1
1
0.25
0.25
0.5
0.25
0.5
2
0.063
0.032
0.063
0.125
0.063
0.063
0.125
0.125
0.032
0.032
0.032
0.063
0.125
0.125
1
0.125
0.125
0.125
0.25
1
4
4
2
1.66
2.04
1.79
1.75
N
N
N
N
S
O
SO2
0.5
2
0.5
NH
S
Me
N
14B
15B
Vinyl
Vinyl
1
1
0.125
0.25
0.25
0.5
0.016
0.016
0.016
0.016
0.008
0.063
0.063
0.063
0.125
1
2
0.86
1.76
Me
N
S
50.004
Me
Me
N
16B
Vinyl
Vinyl
2
2
1
4
1
0.063
0.063
0.125
0.063
0.063
0.032
1
0.25
0.25
8
4
>3.13
0.80
N
N
17B
0.25
0.5
N
a
A; R2 = 1-piperazinyl, R3 = NH2, B; R2 = ( )-3-amino-1-pyrrolidinyl, R3 = H.
b
c
d
e
f
Minimum inhibitory concentration (MIC): lowest concentration of compound that inhibits visible growth of the organism.
MSSA, methicillin-susceptible S. aureus Smith.
MRSA, S. aureus KMP9.
Penicillin-susceptible S. pneumoniae ATCC49619.
S. pneumoniae KT2524.
S. pyogenes ATCC12344.
VRE, E. faecium KU1778.
M. catarrhalis K1209.
H. influenzae TH13.
g
h
i
j
k
The efficacy criterion, ED50, was calculated as the dose at which mice survival rate was 50%. Mice were inoculated with each organism intraperitoneally. Medication was
given intravenously once, 1 h after inoculation.

172
Y. Hirokawa et al. / Bioorg. Med. Chem. Lett. 19 (2009) 170–174
a; l = 1, m = 2, n = 1, X = CH2, d; l = 0, m = 1, n = 0, X = CH2, g; m = 2
b
e
h
; m = 1
; l = 2, m = 1, n = 0, X = CH2, ; l = 2, m = n = 1, X = O,
23c 27c
22c
,
; R form of
c; l = m = 1, n = 0, X = CH2,
f; l = 1, m = 2, n = 0, X = CH2
24c, 28c; S form of 22c
(CH2)nOH
(CH2)nSCOMe
i)
ii)
X
X
1
1
(CH2)m
NH
(CH2)m
22a-e
; Y = S, P = Boc, R = CH=CH2
22g,h; Y = S, P1 = Boc, R1 = CH=CH2
23c, 24c; Y = S, P1 = Boc, R1 = CH=CH2
25f; Y = S, P1 = Boc, R1 = Et
(CH2)l
(CH2)l
iii)
N
Boc
20a-f
21a-f
MeOCS
Me
iv)
1
1
26a-e
; Y = S, P = H, R = CH=CH2
(CH2)m
i)
ii)
HO
26g,h; Y = S, P1 = H, R1 = CH=CH2
27c, 28c; Y = S, P1 = H, R1 = CH=CH2
29f; Y = S, P1 = H, R1 = Et
(CH2)m
N
Me NH
Boc
21g,h
20g,h
Scheme 1. Reagents and conditions: (i) Boc2O, CHCl3, rt; (ii) (iPrCO2N@)2, Ph3P, THF, ꢁ20 °C, 0.5 h then MeCOSH, ꢁ10 °C to rt, 0.5 h; (iii) tBuOK, MeOH, reflux, 2 h then 18 or
19, 0 °C to rt, 18 h; (iv) 30% HCl/EtOH, rt, 2 h.
trifluroacetic acid mutilin 11-ester14 to afford the mutilin 11,14-
diesters 31. Deprotection of trifluoroacetyl group at the 11-posi-
tion in 31 by aqueous ammonia and successive hydrogenation with
palladium on carbon produced the 19,20-dihydromutilin deriva-
tive 32 (Scheme 2).
Me
R1
OP
O
12
11
Me
R-H2CCO
14
The SO2-linked derivative 34 was prepared as illustrated in
Scheme 3. Oxidation of 25f by m-chloroperbenzoic acid (MCPBA),
followed by Boc-deprotection of the sulfone analogue 33 led to
the desired compound 34 as hydrochloride in an excellent yield.
N-alkylation of 4-amino-1-(tert-butoxycarbonyl)piperidine, [N-
(tert-butoxycarbonyl)-N,N0-dimethyl]ethylenediamine, and tert-butyl
1-piperazinecarboxylate with 18 and successive N-Boc-deprotection
of the resultants 35, 37, and 39 gave the corresponding N-linked
derivatives 36, 38, and 40, respectively, as di-hydrochlorides
(Schemes 4 and 5).
Me
Me
O
Pleuromutilin (1); R = OH, R1 = CH(19)=CH2(20), P = H
18; R = OTs, R1 = CH=CH2, P = H
19; R = OTs, R1 = Et, P = H
P1
(CH2)m
22a-e, 23c, 24c, 25f,
N
26a-e, 27c, 28c, 29f; R =
Y(CH2)n
Me
P = H
The pleuromutilin analogues 3A–17B shown in Table 1 were
prepared as illustrated in Scheme 6. Condensation of the 3-(pur-
in-9-yl)propionic acids8 41 and 42 with several amines 26a–e,
26g,h, 27c, 28c, 29f, 32, 34, 36, 38, and 40 in the presence of ben-
zotriazole-1-yloxytris(pyrrolidino)phosphonium hexafluorophos-
phate as a coupling agent and Et3N, followed by successive Boc-
deprotection led to the target compounds 3A–17B in moderate to
good yields. The chemical structures of 3A–17B were confirmed
by 1H NMR and mass spectra and the purity was demonstrated
by HPLC analysis. Compounds 3A–17B, which were obtained as
mono- or di-hydrochlorides, showed good solubility in water
(ꢀ50 mg/mL).
(CH2)l
X
N
22g,h, 26g,h; R =
P = H
Y
P1
(CH2)m
Y
N P1
31-36; R =
Me
Me
N (CH2)2 N P1 R1 = CH = CH2, P = H
37, 38; R =
39, 40; R =
N
P1
R1 = CH = CH2, P = H
N
Figure 1. Structure of pleuromutilin derivatives.
The pleuromutilin analogues 3A–17B shown in Table 1 were
tested for in vitro antibacterial activity15 against well characterized
drug-susceptible and -resistant Gram-positive and -negative bac-
teria including methicillin-susceptible S. aureus (MSSA), MRSA,
penicillin-susceptible S. pneumoniae (PSSP), PRSP, VRE, Streptococcus
in the presence of NaH, followed by successive alkaline hydrolysis
produced the glycolic acid 30. Reaction of 30 with thionyl chloride
gave the corresponding acid chloride, which was treated with the
OH
O
OH
O
i)
ii)
iii)
31; Y = O, P1 = Cbz, R1 = CH=CH2, P = COCF3
iv)
N
32; Y = O, P1 = H, R1 = Et, P = H
N
H
Cbz
30
Scheme 2. Reagents and conditions: (i) ClCO2CH2Ph, Et3N, CH2Cl2, rt, 4 h; (ii) BrCH2CO2Et, DMF, NaH, rt to 50 °C, 3 h then 2 N NaOH/MeOH, reflux, 3 h, 23%; (iii) SOCl2, CHCl3,
reflux, 2 h then trifluoroacetic acid mutilin 11-ester, DMAP, pyridine, reflux, 18 h, 20%; (iv) 28% NH4OH, MeOH, rt, 3 h, 98% then 10% Pd/C, H2, EtOH, rt, quant.

Y. Hirokawa et al. / Bioorg. Med. Chem. Lett. 19 (2009) 170–174
173
reduced by using the ( )-3-piperidinythio spacer (3A and 3B).
The in vivo efficacy of Type B as in 2B and 3B was more potent than
that of Type A as in 2A and 3A. Compound 4A bearing the pyrrolid-
inyl spacer exhibited in vitro antibacterial activity comparable to
that of 2A, while its in vivo efficacy was lower than that of 2A.
The in vitro antibacterial activity of the optically active derivatives
5A and 6A of 4A was very similar to that of 4A. On the other hand, R
configuration at the 3-position on the pyrrolidine ring, such as in
5A, increased the in vivo efficacy, while the S configuration (6A)
showed slightly worse efficacy. Although replacement of the pipe-
ridinyl spacer of 2A with an azetidinyl or a 2-morpholinyl spacer
(yielding 7A or 9A, respectively) caused a significant or slight de-
crease in in vitro and in vivo activities, the in vitro antibacterial
activity of 8A, containing the 4-piperidinylmethyl spacer, was
essentially the same as that of 2A. Compound 8A in vivo efficacy,
however, was weaker than that of 2A.
i)
33; Y = SO2, P1 = Boc, R1 = Et, P = H
34; Y = SO2, P1 = H, R1 = Et, P = H
25f
ii)
Scheme 3. Reagents and conditions: (i) MCPBA, CH2Cl2, rt, 24 h, 90%; (ii) 30% HCl/
EtOH, rt, 2 h, quant.
NH2
i)
35; Y = NH, P1 = Boc, R1 = CH=CH2, P = H
ii)
N
36; Y = NH, P1 = H, R1 = CH=CH2, P = H
Boc
Scheme 4. Reagents and conditions: (i) 18, K2CO3, NaI, MeCN, reflux, 16 h, 76%; (ii)
30% HCl/EtOH, rt, 2 h, quant.
As described in a previous paper,16 hydrogenation of the vinyl
group in the 12-position on the mutilin ring of 2A (giving an ethyl
group as in 10A) did not negatively affect the in vitro or in vivo
activity when compared to that of 2A. Replacement of the sulfur
atom in 2B by an oxygen atom, sulfone linkage, or nitrogen atom
(yielding 11B, 12B, or 13B, respectively) did not, in general, in-
crease the in vitro or in vivo activity. Furthermore, removal of
the methylene or ethylene moiety from the piperidine ring of 2B
provided the 3-methylaminopropylsufide 14B and 2-methylami-
noethylsulfide 15B, both of which exhibited very good in vitro anti-
bacterial activity against all strains with MIC values between 2 and
Me
NH
H
N
i)
37, 39; P1 = Boc
38, 40; P1 =
Me
ii)
N
Boc
N
Boc
H
Scheme 5. Reagents and conditions: (i) 18, K2CO3, NaI, MeCN, reflux, 16 h, (37; 97%,
39; 95%); (ii) 30% HCl/EtOH, rt, 2 h, quant.
60.004 lg/mL. In particular, compound 14B showed the most
favorable in vitro and in vivo activities in this series, while the
in vivo efficacy of 15B was less than that of 2B. Replacement of
the sulfur atom of 15B with a methylamino group (giving 16B)
considerably reduced both the vitro and in vivo activities. A similar
finding has been reported with the addition of a basic nitrogen in
the spacer.16 Quite surprisingly, formation of the piperazine ring
from 16B as in 17B led to a considerable increase in both the
in vitro and in vivo activities. Compound 17B as well as the 3-
methylaminopropylsulfide 14B conferred the highest in vivo effi-
cacy. The excellent in vivo efficacy of 14B and 17B, which also have
good solubility in water, may reflect improved pharmacokinetics
and ADME properties compared to derivatives 2A and 2B. Further
SARs of a novel series of pleuromutilin derivatives containing the
piperazine ring as a central spacer will be reported in due course.
With the purpose of finding novel antibacterial agents for use in
human, the new pleuromutilin analogues 2A and 2B with good sol-
ubility in water and excellent in vitro and in vivo activities were
identified. Modification of the sulfur-linked piperidinyl spacer of
2A and 2B resulted in the discovery of the novel pleuromutilin
derivatives 14B and 17B having the 3-methylaminopropylsulfide
and the piperazine ring, respectively, as a central spacer. Both com-
pounds showed much higher in vivo efficacy than the parent com-
R2
N
N
ii)
i)
N
3A-17B
R3
N
CO2H
41; R2 =
42; R2 =
R3 = NH2
N
N
N
Boc,
NHBoc
R3 = H
,
Scheme 6. Reagents and conditions: (i) 26a–e, 26g,h, 27c, 28c, 29f, 32, 34, 36, 38,
and 40, benzotriazole-1-yloxytris(pyrrolidino)phosphonium hexafluorophosphate,
Et3N, DMF, rt, 2 h; (ii) 30% HCl/EtOH, rt, 2 h.
pyogenes, Moraxella catarrhalis, and Haemophilus influenzae, all of
which are common serious respiratory tract pathogens. Further-
more, 3A–17B were evaluated for in vivo efficacy against S. aureus
Smith (MSSA) systemic infection model in mice. For comparison,
the previously reported pleuromutilin analogues8,9 2A and 2B were
used. All compounds were found to possess potent in vitro antibac-
terial activity against all strains (Table 1). In general, PSSP, PRSP,
and S. pyogenes strains were highly susceptible, while MRSA and
H. influenzae strains had somewhat higher MIC values. For these
derivatives, no significant differences in minimum inhibitory con-
centration (MIC) for drug-susceptible versus -resistant strains was
observed.
pound 2B and appeared to have
a well-balanced in vitro
antibacterial activity profile against MRSA, PRSP, VRE, S. pyogenes,
M. catarrhalis, and H. influenzae, all of which are known to cause
respiratory tract infections. The promising pleuromutilin deriva-
tives 14B and 17B may serve as useful lead compounds for the dis-
covery of new antibiotics.
Influence of the sulfide linkage of 2A and 2B on the antibacterial
activity was first examined. Compounds 3A and 3B, bearing the 3-
piperidinylthio ring as a central spacer, while keeping the mutilin
framework with its 2-amino-6-(piperazin-1-yl)purine (Type A) or
6-[( )-3-aminopyrrolidin-1-yl]purine (Type B) ring intact, showed
in vitro antibacterial activity practically comparable to that of 2A
and 2B against all strains, but no improvement in in vivo efficacy.
It was subsequently found that the direction of the connecting vec-
tor on the piperidine ring is important, since in vivo efficacy was
References and notes
1. Projan, S. J.; Youngman, P. J. Curr. Opin. Microbiol. 2002, 5, 463.
2. Livermore, D. M. Clin. Infect. Dis. 2003, 36, 1058.
3. Ohno, A. Infect. Control 2004, 13, 1012.
4. Hunt, E. Drug Future 2000, 25, 1163.
5. Anchel, M. J. Biol. Chem. 1952, 199, 133.
6. Arigoni, D. Gazz. Chim. Ital. 1962, 92, 884.
7. Birch, A. J.; Cameron, D. W.; Holzapfel, C. W.; Rihards, R. W. Chem. Ind. 1963,
374.

174
Y. Hirokawa et al. / Bioorg. Med. Chem. Lett. 19 (2009) 170–174
8. Hirokawa, Y.; Kinoshita, H.; Tanaka, T.; Nakamura, T.; Fujimoto, K.; Kashimoto,
13. (a) Birch, A. J.; Holzapfel, C. W.; Richards, R. W. Tetrahedron 1966, (Suppl.
8), part II, 359.; (b) Berner, H.; Schulz, G.; Fischer, G. Monat. Chem. 1981,
112, 1441.
S.; Kojima, T.; Kato, S. BioMed. Chem. Lett. 2008, 18, 3556.
9. Hirokawa, Y.; Kinoshita, H.; Tanaka, T.; Nakata, K.; Kitadai, N.; Fujimoto, K.;
Kashimoto, S.; Kojima, T.; Kato, S. J. Med. Chem. 2008, 51, 1991.
10. WO 99/21855.
11. (a) Sum, F.-W.; Wong, V.; Han, S.; Largis, E.; Mulvey, R.; Tillett, J. BioMed. Chem.
Lett. 2003, 13, 2191; (b) Kojima, T.; Niigata, K.; Fujikura, T.; Tachikawa, S.;
Nozaki, Y.; Kagami, S. Chem. Pharm. Bull. 1985, 33, 3766.
12. (a) Egger, H.; Reinshagen, H. J. Antibiotics 1976, 29, 915; (b) Riedl, K. J.
Antibiotics 1976, 29, 132.
14. (a) WO 97/25309.; (b) WO-2000/73287.
15. National Committee for Clinical Laboratory Standards 2000. Methods for Dilution
Antimicrobial Susceptibility Tests for Bacteria That Grow Aerobically, 5th ed.;
Approved Standard M7 A5; National Committee for Clinical Laboratory
Standards: Wayne, PA, 2000.
16. Egger, H.; Reinshagen, H. J. Antibiotics 1976, 29, 923.