Synthesis and activity of a C-8 keto pleuromutilin derivative

Article abstract of DOI:10.1016/S0960-894X(03)00297-X

A C-8 keto pleuromutilin derivative has been synthesized from the biotransformation product 8-hydroxy mutilin. A key step in the process was the selective oxidation at C-8 of 8-hydroxy mutilin using tetrapropylammonium perruthenate. The presence of the C-8 keto group precipitated interesting intramolecular chemistry to afford a compound (10) with a novel pleuromutilin-derived ring system.

Full text of DOI:10.1016/S0960-894X(03)00297-X

Bioorganic & Medicinal Chemistry Letters 13 (2003) 1751–1753
Synthesis and Activity of a C-8 Keto Pleuromutilin Derivative
Dane M. Springer,a,* Margaret E. Sorenson,a Stella Huang,b Timothy P. Connolly,a
Joanne J. Bronson,a James A. Matson,c Ronald L. Hanson,d David B. Brzozowski,d
Thomas L. LaPorted and Ramesh N. Pateld
aDepartment of Anti-infective Chemistry, 5 Research Parkway, PO Box 5100, Wallingford, CT 06492, USA
bDiscovery Analytical Sciences, 5 Research Parkway, PO Box 5100, Wallingford, CT 06492, USA
cNatural Products Chemistry, 5 Research Parkway, PO Box 5100, Wallingford, CT 06492, USA
dProcess Research and Development, Bristol-Myers Squibb Pharmaceutical Research Institute, One Squibb Drive,
New Brunswick, NJ 08903, USA
Received 9 September 2002; accepted 27 February 2003
Abstract—A C-8 keto pleuromutilin derivative has been synthesized from the biotransformation product 8-hydroxy mutilin. A key
step in the process was the selective oxidation at C-8 of 8-hydroxy mutilin using tetrapropylammonium perruthenate. The presence
of the C-8 keto group precipitated interesting intramolecular chemistry to afford a compound (10) with a novel pleuromutilin-
derived ring system.
# 2003 Elsevier Science Ltd. All rights reserved.
deaths due to gastrointestinal tract infections such as
dysentery, and respiratory tract infections.3 Azamulin
(3) was under development by Sandoz in the 1980’s for
potential use in humans. Unfortunately, while deter-
mined to be safe in healthy volunteers, azamulin dis-
played poor oral absorption and a short half-life due to
rapid metabolism and subsequent excretion.4 These
poor PK characteristics are the bane of the pleuro-
mutilin class, and efforts to overcome them have been
largely fruitless to date.
A significant contributor to the metabolism of pleuro-
mutilin derivatives is cytochrome P450-mediated hydro-
xylation at C-2 and C-8 of the core structure. These
Pleuromutilin (1) derivatives have attracted considerable
pharmaceutical and academic attention due to their
potent activity against drug-resistant Gram-positive bac-
teria, and the fascinating chemistry exhibited by the tri-
cyclic nucleus.1,2 Tiamulin (2) is a derivative discovered
by Sandoz that is currently employed in veterinary medi-
cine for promoting the growth of turkey poults, suckling
pigs, and other newborn animals. Tiamulin is added to
foodstuffs or injected ip in these animals to prevent
metabolites are described in the pleuromutilin literature
as being poorly active molecules.2c,5 However, it is not
clear from the reports if the inactivity is due to lack of
intrinsic antibacterial activity in vitro (MIC), or a
reflection of poor in vivo activity due to conjugation of
the hydroxyls at C-2 and C-8 and subsequent elimina-
tion. The Sandoz group attacked the metabolism prob-
lem by synthesizing a number of C-2 and C-8 analogues
designed to either sterically shield these sites, or electro-
nically alter them to disfavor hydrogen atom abstraction
by the iron–oxo complex of the cytochrome. Their che-
mical efforts at C-2 utilized pleuromutilin as starting
material, readily available in bulk quantities from fungal
*Corresponding author at current address: Rib-X Pharmaceuticals,
300 George St., Room 301, New Haven, CT 06511, USA. Fax: +1-
203-624-5627; e-mail: dspringer@rib-x.com
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00297-X

1752
D. M. Springer et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1751–1753
cultures. Their C-8 work proceeded from 4a (described as
being a by-product in the fermentation leading to pleuro-
mutilin), and 4b, available presumably by hydrogena-
tion of 4a.6 While the Sandoz group utilized the C-8
hydroxyl functional group of 4a and 4b to synthesize a
significant number of derivatives with alterations in the
cyclohexyl ring of the system, no biological data has
emerged from these derivatives.6,7
tetrapropylammonium perruthenate gave the best yield of
desired 8 (48%) relative to other oxidized products. We
next attempted a familiar stratagem of pleuromutilin
chemistry, the acid-catalyzed conversion of 8 to 9. This
process works well in most pleuromutilin derivatives,
and serves to protect the C-3 ketone and C-11 hydroxyl
groups during further transformations.10 [The intramo-
lecular hydride migration process is reversed using
Lukas’ reagent (concd HCl saturated with ZnCl2) to
reform the ketone at C-3 and the hydroxyl group at
C-11.] Disconcertingly, we observed that the expected
1H NMRresonances for a derivative such as 9 were not
present in the major product of the reaction. After
As part of a program at Bristol-Myers Squibb to
develop a novel pleuromutilin derivative for human use,
we wished to have chemical access to the six-membered
ring to manipulate this under-exploited region for the
purpose of extending pleuromutilin SAR. In one
approach to this goal, we decided to produce C-8 oxi-
dized pleuromutilin derivatives by initial bio-
transformation of mutilin (5) and subsequent
conversion to the desired targets (Scheme 1). The suc-
cessful biotransformation efforts generated very good
yields of C-8 hydroxy mutilin 6, and depending on the
exact conditions, significant amounts of C-7 hydroxy
mutilin 7 as well.8 With ample quantities of 6 in hand
we decided to synthesize the C-8 keto, C-14 carbamate
analogue 11 (Scheme 2) as one of our initial targets. A
group at GlaxoSmithKline has recently published a
series of C-14 carbamates including carbamate 12.9 We
had previously evaluated 12 in vitro and found it to be
extremely potent. We believed a head-to-head compar-
ison of 11 and 12 might prove useful in assessing the
effect of the C-8 keto group on antibacterial potency.
1
scrutiny of H–1H COSY, HMBC, and HMQC NMR
spectra, we were able to assign the structure of 10 to the
product of this reaction.11 The formation of this novel
ring structure is a direct consequence of the presence of
a C-8 keto group in 8. Once a methyl oxonium ion is
formed at C-3, it is trapped by a C-7/C-8 enol to form
the norbornane substructure of 10. Hand-held models
are useful in revealing that the cyclooctane ring of such
a norbornane derivative has considerably more flex than
a normal pleuromutilin. The normal pleuromutilin skel-
eton restricts the cyclooctane ring to predominately a
boat-chair conformation. The boat-boat conformer is
precluded by an intense ‘prow-like’ interaction between
the C-4 proton protruding out over the ring, and the
methyl group at C-12. Crown-like cyclooctane ring
structures in this system are disfavored by a steric
interaction between the C-10 methyl group and the C-4
methine proton. However, the norbornane structure of
10 causes the C-4 carbon atom (and its proton) to be
splayed back away from the cyclooctane ring, allowing
increased mobility of C-10 and C-12 towards C-4. A
We began our synthesis of 11 by exploring the selective
oxidization of the C-8 hydroxyl of 6 to provide diketone 8.
After trying a number of oxidizing agents we found that
Scheme 1.
Scheme 2. Reaction conditions: (a) NMO, TRAP, CH3CN, 4 A sieves, 0 ꢀC, 42%; (b) HC(OMe)3, MeOH, concd H2SO4, 51%; (c) Ac2O, DMAP,
pyr, rt, 54%; (d) 4-MeO-C6H4NCO, i-Pr2NEt, CHCl3, reflux, 60%; (e) 1 N NaOH, rt, 30%.

D. M. Springer et al. / Bioorg. Med. Chem. Lett. 13 (2003) 1751–1753
1753
Table 1.
References and Notes
Organism
A no. Mic of compounds (mg/mL)
1. (a) Kavanagh, F.; Hervey, A.; Robbins, W. J. Proc. Natl.
Acad. Sci. U.S.A. 1951, 37, 570. (b) Egger, H.; Reinshagen, H.
J. Antibiot. 1976, 29, 923.
2. (a) Arigoni, D. Gazz. Chim. Ital. 1962, 92, 884. (b) Birch,
A. J.; Holzapfel, C. W.; Rickards, R. W. Tetrahedron 1966,
Suppl. 8, 359. (c) Berner, H.; Vyplel, H.; Schulz, G. Tetra-
hedron 1987, 43, 765 and references cited therein.
11
12
1
2
3
4
S. pneumoniae/Pen. Resist. A28272
S. aureus/Homo MRA27223
32
0.003
32
0.003
M. catarrhalis/Pen. +
H. influenzae/Pen. +
A22344
A21515
64
64
0.125
4
3. (a) Anderson, M. D. Vet. Med. Small Anim. Clin. 1983, 78,
98. (b) Burch, D. G. S.; Jones, G. T.; Heard, T. W.; Tuck,
R . E.Veterinary Record 1986, 119, 108.
conformation can now be accessed that allows the
hydroxyl at C-11 to approach the backside of the C-14
carbon’s hydroxyl bond to displace it via acid catalysis,
forming the cyclic ether between C-11 and C-14.
4. (a) Ganzinger, U.; Stephen, A.; Obenaus, H.; Baumgartner,
R.; Walzl, H.; Bruggemann, S.; Schmid, B.; Racine, R.;
Schatz, F.; Haberl, H. In Proc. Int. Congr. Chemother. 13th;
Spitzey, K. H., Karrer, K., eds.; H. Egermann: Vienna, Aus-
tria, 1983, Vol. 5, p 108/53. (b) Nefzger, M.; Battig, F.; Czok,
R . InProc. Int. Congr. Chemother. 13th; Spitzey, K. H., Kar-
rer, K., eds.; H. Egermann: Vienna, Austria, 1983; Vol. 5, p
108/47. (c) Schatz, F.; Haberl, Czok, R. In Proc. Int. Congr.
Chemother. 13th; Spitzey, K. H., Karrer, K., eds.; H. Eger-
mann: Vienna, Austria, 1983, Vol. 5, p. 108/50.
5. Berner, H.; Hildebrandt, J.; Schuster, I. In Proc. Int. Congr.
Chemother. 13th; Spitzey, K. H., Karrer, K., Eds.; H. Eger-
mann: Vienna, Austria, 1983; Vol. 5, p. 108/20, and references
cited.
6. (a) An experimental description of the efforts to produce 4a
and 4b is absent from the literature; however, the approach is
alluded to in the following reports: Berner, H.; Vyplel, H.;
Schulz, G.; Stuchlik, P. Tetrahedron 1983, 39, 1317. (b) Berner,
H.; Vyplel, H.; Schulz, G. Tetrahedron 1987, 43, 765.
7. Berner, H.; Vyplel, H.; Schulz, G.; Stuchlik, P. Monatsh.
Chem. 1983, 114, 1125.
8. Hanson, R. L.; Matson, J. A.; Brzozowski, D. B.;
LaPorte, T. L.; Springer, D. M.; Patel, R. N. Organic Process
Research and Development 2002, 6, 482.
9. Brooks, G.; Burgess, W.; Colthurst, D.; Hinks, J. D.; Hunt,
E.; Pearson, M. J.; Shea, B.; Takle, A. K.; Wilson, J. M.;
Woodnutt, G. Bioorg. Med. Chem. Lett. 2001, 9, 1221.
10. Berner, H.; Schulz, G.; Schneider, H. Tetrahedron 1980,
36, 1807.
We employed new tactics to arrive at the desired C-8
keto derivative 11. We found that diol 8 could be selec-
tively acylated at C-11. Introduction of the carbamate
at C-14 was then followed by hydrolysis of the acetate
at C-11 to deliver 11.12 Quite surprisingly, C-8 keto
derivative 11 was found to be ꢁ10,000 times less active
than C-8 methylene derivative 12 in vitro against bac-
teria of interest to our program (see Table 1). This
staggering erosion of activity after such a slight struc-
tural change, coupled with the observations that
8-hydroxy azamulin 13 and the C-6 epimeric pleuro-
mutilin 14 are poorly active, clearly indicates that the
environment of the cyclohexyl ring of the system is cru-
cial to the molecule’s interactions with the ribosomal
complex. Altering this portion of the molecule (to
improve pharmacokinetic parameters and avoid meta-
bolic liabilities) while maintaining antibacterial potency
will be a significant challenge to further investigations in
the pleuromutilin class.
11. 1H NMRdata for 10: (500 MHz, CDCl3, partial) d 5.91
(dd, 1H, J=11, 17 Hz), 4.92 (d, 1H, J=17 Hz), 4.84 (d, 1H,
J=11 Hz), 3.89 (dd, 1H, J1=J2=8 Hz), 3.86 (d, 1H, J=10
Hz), 3.23–3.16 (m, 1H), 3.22 (s, 3H), 2.50–2.47 (m, 1H), 2.45–
2.38 (m, 1H), 2.09 (d, 1H, J=2 Hz), 2.05–2.00 (m, 1H), 1.84–
1.68 (m, 3H), 1.48–1.44 (m, 1H), 1.23 (s, 3H), 1.04 (d, 3H,
J=7 Hz), 0.99 (s, 3H), 0.92 (d, 3H, J=8 Hz).
12. 1H NMRdata for 11: (300 MHz, CDCl3) d 7.36–7.22 (m,
2H), 6.94–6.82 (m, 2H), 6.55 (dd, 1H, J=11, 18 Hz), 5.27 (d,
1H, J=18 Hz), 3.80 (s, 3H), 3.45–3.35 (m, 1H), 2.91 (dd, 1H,
J1=J2=14 Hz), 2.65–2.50 (m, 2H), 2.42 (br s, 1H), 2.31–2.17
(m, 3H), 2.17–1.93 (m, 2H), 1.69–1.15 (m, 8H), 1.24 (s, 3H),
0.95–0.86 (m, 3H), 0.80 (d, 3H, J=7 Hz).
Acknowledgements
We thank our colleagues in the Department of Micro-
biology in Wallingford, CT for the biological data
reported in this work.