Synthesis and biological testing of non-fluorinated analogues of Levofloxacin

Article abstract of DOI:10.1016/S0960-894X(03)00399-8

Quinolones without the usual 6-fluorine substituent have been recently described as potent antibacterial agents. A series of non-fluorinated analogues of the antibacterial quinolone Levofloxacin were synthesized and tested.

Full text of DOI:10.1016/S0960-894X(03)00399-8

Bioorganic & Medicinal Chemistry Letters 13 (2003) 2373–2375
Synthesis and Biological Testing of Non-Fluorinated Analogues
of Levofloxacin
Jeffrey L. Gray,* Ji-In K. Almstead, Corey P. Gallagher, X. Eric Hu, Nick K. Kim,
Cynthia J. Taylor, Tracy L. Twinem, Cynthia D. Wallace and Benoit Ledoussal
Procter & Gamble Pharmaceuticals, 8700 Mason-Montgomery Rd, Mason, OH 45040, USA
Received 18 February 2003; revised 14 April 2003; accepted 15 April 2003
Abstract—Quinolones without the usual 6-fluorine substituent have been recently described as potent antibacterial agents. A series
of non-fluorinated analogues of the antibacterial quinolone Levofloxacin were synthesized and tested.
# 2003 Elsevier Science Ltd. All rights reserved.
The quinolone antibacterials are a highly successful
class of agents for the treatment serious infections.
These drugs exert their effect by the inhibition of type II
bacterial topoisomerases such as DNA gyrase and
Topoisomerase IV. The first generation quinolones,
such as Nalidixic Acid (Fig. 1), were characterized by
activity against Gram-negative pathogens and were
used primarily for the treatment urinary tract infections.
The second generation quinolones were based on the
finding of Koga that a fluorine at the C6 position of the
quinolone combined with an amine containing group at
C7 enhanced the spectrum and distribution of the drug.1
Following this report, thousands of quinolones were
synthesized that contained a C6 fluorine. Several very
successful drugs were launched based on these results.
Because of the prevalent C6 fluorine, the class is fre-
quently referred to as fluoroquinolones antibacterials.
Nearly ten years after Koga’s disclosure, Ledoussal
reported that potent quinolones with a C6 hydrogen
could be made if the appropriate C7 substituent was
chosen.2 Since that time we3 and others4À11 have repor-
ted on the activity of non-fluorinated quinolones that
have biological activity equivalent to the fluoro-
quinolones.
Figure 1. Structures of naladixic acid and levofloxacin.
This ring is connected with an ether moiety at C8. This
type of substitution has the potential benefit of reducing
any phototoxic problems seen with quinolones that
have a halogen in the C8 position.13 Several recently
approved fluoroquinolones have a methoxy group at
C8.
In the course of our exploration of non-fluorinated
quinolones, we prepared a series of Levofloxacin ana-
logues with a hydrogen at the C6 position. Interestingly,
the original report on Ofloxacin disclosed several non-
fluorinated analogues.14 Limited microbiological data
was presented on these racemic compounds and no
topoisomerase inhibition values were available. Also,
none of these analogues utilized the pyrrolidines and
piperidines that were later found to increase the activity
of many quinolones.
Levofloxacin,12 a chiral version of the earlier drug
Ofloxacin, is a successful fluoroquinolone antibacterial
that departs from the typical quinolones substructure by
having a fused ring connecting the N1 to C8 position.
The synthetic route to the requisite quinolone nucleus
followed an earlier report on the synthesis of the 6F
quinolone (Scheme 1).15 The keto-ester 1 was elaborated
from the benzoyl chloride by standard methodology.16
The ethoxy ethylene moiety was formed by treatment
with acetic anhydride and triethylorthoformate and
*Corresponding author. Tel.:+1-513-622-3856; fax:+1-513-622-1433;
e-mail: gray.jl.2@pg.com
0960-894X/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0960-894X(03)00399-8

2374
J. L. Gray et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2373–2375
Scheme 1.a (a) Ethyl hydrogen malonate, 2 equiv nBuLi, THF, À78 ꢀC to rt; (b) (EtO)3CH, Ac2O, 110 ꢀC; (c) S-(+)-alaninol, CH2Cl2, 0 ꢀC to rt; (d)
2 equiv NaH, THF, 0 ꢀC to rt; (e) 10% KOH, THF, reflux. For 5a and 5b (f) 6a or 6b, Et3N, NMP, 80 ꢀC; (g) HCl, EtOH. For 5c (h) BF3 Et2O,
THF, reflux; (i) 6c, Et3N, DMF, 50 ꢀC; (j) Et3N, MeOH, reflux. For 5d (k) pyridine, 6d, 110 ꢀC.
a5a and 5b were synthesized in 19 and 30% yield, respectively, from 4. New compounds gave satisfactory analytical data (1H NMR, MS, HPLC,
CHN analysis). 5c and 5d were synthesized in 10% and 20% yield, respectively, from 4. New compounds gave satisfactory analytical data (1H
NMR, MS, HPLC, HRMS analysis).
used without isolation. Addition of commercially avail-
able S-(+)-alaninol afforded the enamine intermediate 2.
Table 1. Minimum inhibtory concentrations of new compounds
Compd
5a
5b
5c
5d Levofloxacin
We found the closure of the N1 to C8 ring for benzoxazine
formation was best effected by continued treatment with
base in the quinolone ring formation step rather than with
aqueous KOH in the ester hydrolysis step. The amines 6a,
6b, and 6c were synthesized according to the reported
protocols.17À19 In the case of quinolones 5a, 5b, and 5d
coupling of the C7 side chain was performed on the
quinolone carboxylic acid. For quinolone 5c the boronate
complex of the quinolone carboxylate was used to
increase reactivity during coupling. The boronate was
removed with triethylamine in refluxing methanol.
Organism
Typea Strain
MIC (mg/mL)
S. aureus
S. aureus
S. aureus
S. aureus
S. aureus
S. saprophyticus
S. epidermidis MSCR SE48
E. faecium
E. faecium
E. faecalis
S. pneumoniae PS
S. pneumoniae PS
S. pneumoniae PR
S. pyogenes
S. viridans
E. cloacae
E. coli
E. coli
E. coli
MSCS MI246
MSCS MI273
MSCR MI345
MRCS MI300
MRCR MI339
SS276
0.031 0.063
0.125 0.125
2
4
4
8
>32
4
>32
4
>32
>32
NG
NG
16
16
32
4
32
0.12
0.12
2
0.12
2
0.12
2
0.25
1
0.25
0.25
0.25
0.5
1
0.063 0.063
2
>32
2
2
0.125 0.031 16
2
>32
1
1
1
1
1
1
>32
>32
NG
VS
VR
EF1
EF12
STD44
0.031 0.125 NG
STP6301 NG 0.016 >32
STP64
STP51
STA2
STV1
AE63
ES142
DC0b
DC2
MC2
KL21
KL328 >8
PR91
PS96
—
—
0.063 0.5
>32
>32
The biological evaluation of these compounds was car-
ried out by both standard bacterial minimum inhibitory
concentration (MIC) assays and DNA gyrase inhibition
determinations. Our panel of organisms contains Gram-
positive and Gram-negative bacteria (Table 1). Addi-
tionally, strains resistant to standard antibacterials were
included. Topoisomerase inhibition values were also
measured against wild-type, quinolone-resistant DNA
gyrase (E. coli) and mammalian topoisomerase II.20 The
new pyrrolidine containing non-fluorinated quinolones
5a and 5b maintain good antibacterial potency relative
to Levofloxacin. This result follows the general trends
we have observed with the non-fluorinated quinolones.
The direct 6-fluorinated congener of 5a has been repor-
ted.21 While direct comparisons cannot be made
because of the different strains used in testing, the
fluorinated version of this quinolone had few significant
differences in activity using Levofloxacin as a reference.
The aminopiperidine containing quinolone 5c was dra-
matically reduced in potency compared to the pyrro-
lidine analogues. We have previously noted that side
chains of this class are much more sensitive to substitution
on other areas of the quinolone.22 The non-fluorinated
version of Levofloxacin (5d) was likewise quite poor in
its antibacterial activity. This result was not unexpected
0.016 0.016 >32
0.125 >32
0.125 0.25
0.5
1
0.25
0.008
0.016
0.25
0.25
0.03
<0.004
4
0.25 NG
0.5
2
0.5
4
1
16
2
2
0.25
8
>32
16
4
1
>32
8
>32
0.063 0.125
0.125 0.063
0.016 0.031
M. catarrhalis
K. pneumoniae
K. pneumoniae CR
P. mirabilis
>8
4
>32
2
2
1
0.06
0.5
P. aeruginosa
4
>32
aMSCS=Methicillin sensitive, ciprofloxacin sensitive; MRCR=Methicillin
resistant, ciprofloxacin resistant; PS=Penicillin sensitive; PR=Penicillin resis-
tant; VS=Vancomycin sensitive; VR=Vancomycin resistant.
bDC2 is a permeable mutant. DC0 is the parent strain. NG=No growth.
based on our previous work with piperazine derivatives
with non-fluorinated quinolones and the previous report
on the racemic version of this compound.23 The topo-
isomerase inhibition values were in line with the MIC
results (Table 2). The pyrrolidine containing com-
pounds 5a and 5b were essentially the same as Levo-
floxacin against the wild-type DNA gyrase. They were
less effective than similar non-fluorinated quinolones
without the benzoxazine ring system against the quino-
lone resistant DNA gyrase. Quinolone 5a was more
potent than Levofloxacin against mammalian topo-
isomerase II, indicating potential detrimental activity.

J. L. Gray et al. / Bioorg. Med. Chem. Lett. 13 (2003) 2373–2375
2375
4. Cecchetti, V.; Fravolini, A.; Palumbo, M.; Sissi, C.;
Tabarrini, O.; Terni, P.; Xin, T. J. Med. Chem. 1996, 39, 4952.
5. Cecchetti, V.; Fravolini, A.; Lorenzini, M. C.; Tabarrini,
O.; Terni, P.; Xin, T. J. Med. Chem. 1996, 39, 436.
6. Cecchetti, V.; Filipponi, E.; Fravolini, A.; Tabarrini, O.;
Bonelli, D.; Clementi, M.; Cruciani, G.; Clementi, S. J. Med.
Chem. 1997, 40, 1698.
7. Cecchetti, V.; Tabarrini, O.; Sabatini, S.; Miao, H.; Filip-
poni, E.; Fravolini, A. Bioorg. Med. Chem. 1999, 7, 2465.
8. Fujita, M.; Egawa, H.; Katoka, M.; Miyamoto, T.; Nakano,
J.; Matsumoto, J. I. Chem. Pharm. Bull. 1995, 43, 2123.
9. Fujita, M.; Egawa, H.; Miyamoto, T.; Nakano, J.; Matsu-
moto, J. I. Chem. Pharm. Bull. 1996, 44, 987.
Table 2. Inhibition values for E. coli DNA gyrase (wild type and
fluoroquinolone resistanta) and mammalian topoisomerase II
Compd
Wild type
gyrase
IC50, mg/mL
Quinolone
resistant gyrase
IC50, mg/mL
Topo II
IC50, mg/mL
5a
5b
5c
5d
6.4
3.2
63
31
3.2
51.2
>102
>500
>500
12.8
38
150
>500
250
>600
Levofloxacin
aFluoroquinolone DNA gyrase was characterized as Ser83 to Trp
mutant.
10. Graul, A.; Rabasseda, X.; Castaner, J. Drugs Future 1999,
24, 1324.
11. Takahata, M.; Mitsuyama, J.; Yamashiro, Y.; Yonezawa,
M.; Araki, H.; Todo, Y.; Minami, S.; Watanabe, Y.; Narita,
H. Antimicrob. Agents Chemother. 1999, 43, 1077.
12. Hayakawa, I.; Atarashi, S.; Yokohama, S.; Imamura, M.;
Sakano, K.; Furukawa, M. Antimicrob. Agents Chemother.
1986, 29, 163.
13. Domagala, J. M. J. Antimicrob. Chemother. 1994, 33, 685.
14. Hayakawa, I.; Hiramitsu, T.; Tanaka, Y. Chem. Pharm.
Bull. 1984, 32, 4907.
15. Mitscher, L. A.; Sharma, P. N.; Chu, D. T. W.; Shen,
L. L.; Pernet, A. G. J. Med. Chem. 1987, 30, 2283.
16. Chu, D. T. W.; Maleczka, R. E., Jr. J. Heterocycl. Chem.
1987, 24, 453.
As seen in their antibacterial activity, analogues 5c and
5d were notably less active in inhibiting topoisomerases.
This non-fluorinated series of Levofloxacin analogues
are equivalent or less potent than the parent molecule.
In contrast to some non-fluorinated analogues where
the N1 moiety is a cyclopropyl, there is a substantial
decrease in activity versus known agents. The loss of
freedom of movement at N1 and C8 appears to be
responsible for these diminished results.
17. Johnson, D. R.; Szoteck, D. L.; Domagala, J. M.; Stick-
ney, T. M.; Michel, A.; Kampf, J. W. J. Heterocycl. Chem.
1992, 29, 1481.
Acknowledgements
18. Fedij, V.; Lenoir, E. A.III; Suto, M. J.; Zeller, J. R.;
Wemple, J. Tetrahedron: Asymmetry 1994, 5, 1131.
19. Moon, S. H.; Lee, S. Synth. Commun. 1998, 28, 3919.
20. Roychoudhury, S.; Twinem, T. L.; Makin, K. M.; McIn-
tosh, E. J.; Ledoussal, B.; Catrenich, C. E. Journal of Anti-
microbial Chemotherapy 2001, 48, 29.
21. Kimura, Y.; Atarashi, S.; Takahashi, M.; Hayakawa, I.
Chem. Pharm. Bull. 1994, 42, 1442.
22. Manuscript in prepation.
23. Ledoussal, B.; Almstead, J. I.; Flaim, S. M.; Gallagher, C.
P.; Gray, J. L.; Hu, E. X.; Kim, N. K.; McKeever, H. D.;
Miley, C. J.; Twinem, T. L.; Zheng, S. X. 37th Interscience
Conference on Antimicrobial Chemotherapy, San Francisco,
1999, Abstr. No. F0544.
We would like to thank Tiehong Huang for the pre-
parative HPLC purification of compounds 5c and 5d.
References and Notes
1. Koga, H.; Itoh, A.; Murayama, S.; Suzue, S.; Irikura, T. J.
Med. Chem. 1980, 23, 1358.
2. Ledoussal, B.; Bouzard, D.; Coroneos, E. J. Med. Chem.
1992, 35, 198.
3. Ledoussal, B.; Almstead, J. K.; Gray, J. L.; Hu, X. E. WO
Patent 9914214, 1999.