The synthesis and antimicrobial evaluation of a new series of isoxazolinyl oxazolidinones

Article abstract of DOI:10.1016/j.bmcl.2004.04.037

A series of oxazolidinone antibacterial agents containing a 5-substituted isoxazol-3-yl moiety were synthesized via a nitrile oxide [3+2] dipolar cycloaddition reaction. These compounds were screened against a panel of susceptible and resistant Gram-positive organisms. Several analogs from this series were comparable to or more potent than linezolid in vitro.

Full text of DOI:10.1016/j.bmcl.2004.04.037

Bioorganic & Medicinal Chemistry Letters 14 (2004) 3069–3072
The synthesis and antimicrobial evaluation
of a new series of isoxazolinyl oxazolidinones
Michele A. Weidner-Wells,* Harvey M. Werblood, Raul Goldschmidt,
Karen Bush, Barbara D. Foleno, Jamese J. Hilliard, John Melton,
Ellyn Wira and Mark J. Macielag
Antimicrobial Agents Research Team, Drug Discovery, Johnson & Johnson Pharmaceutical Research and Development,
L. L. C., 1000 Route 202, Raritan, NJ 08869, USA
Received 12 February 2004; accepted 12 April 2004
Abstract—A series of oxazolidinone antibacterial agents containing a 5-substituted isoxazol-3-yl moiety were synthesized via a
nitrile oxide [3+2] dipolar cycloaddition reaction. These compounds were screened against a panel of susceptible and resistant
Gram-positive organisms. Several analogs from this series were comparable to or more potent than linezolid in vitro.
ꢀ 2004 Elsevier Ltd. All rights reserved.
The oxazolidinones, a new class of synthetic antibacter-
ial agents, possess useful activity against a variety of
susceptible and resistant Gram-positive organisms such
as methicillin-resistant Staphylococcus aureus (MRSA),
vancomycin-resistant enterococci (VRE), and penicillin-
resistant Streptococcus pneumoniae (PRSP).1;2 This class
inhibits bacterial protein synthesis by binding to the 50S
ribosomal subunit and prevents the formation of a
functional 70S initiation complex. Without this com-
plex, protein synthesis does not commence.3 Due to this
unique mechanism of action, there is no cross-resistance
between the oxazolidinones and other families of anti-
bacterial agents. Linezolid (Zyvoxꢁ, 1) is the only oxazol-
idinone to have been approved for the treatment of
hospital- and community-acquired pneumonia as well as
skin infections.
pounds 2 containing a C-linked heteroaromatic ring
attached to the phenyl ring are active against susceptible
and resistant Gram-positive organisms.6 Several patents
from Pharmacia have generically disclosed (though not
exemplified) isoxazole analogs where the pyrazole ring
of 2 is replaced with an isoxazole (the point of attach-
ment was not specified).7;8 To our knowledge, the isoxa-
zolines have not been investigated.
O
O
R
4'
1'
O
5
N
4'
3'
1'
N
O
5
O
N
N
N
NHAc
3'
F
NHAc
F
linezolid (1)
2
Previous literature on the structure–activity relation-
ships for the oxazolidinones has demonstrated that the
S-configuration at the C-5 position of the oxazolidinone
ring is essential and the acetamide is usually optimal for
activity.4 In addition, there is a high tolerance for a wide
variety of substituents at the 40-position of the phenyl
ring.5 For example, Pharmacia has reported that com-
O
R
4'
1'
O
5
N
O
NHAc
N
3'
F
3
This paper describes the synthesis and microbiological
evaluation of a series of 3-isoxazolinyl oxazolidinones 3.
The key step is a late stage nitrile oxide dipolar [3+2]
cycloaddition reaction with an appropriately substituted
olefinic dipolarophile, which enables access to a diverse
array of analogs.
Keywords: Antibacterial; Oxazolidinone; Resistant bacteria.
* Corresponding author. Tel.: +1-908-704-5933; fax: +1-908-203-8109;
e-mail: mwells@prdus.jnj.com
0960-894X/$ - see front matter ꢀ 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2004.04.037

3070
M. A. Weidner-Wells et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3069–3072
O
O
Hc
F
Hb
Ha
1.HMT, TFA,
reflux
O
O
N
OHC
N
NHAc
NHAc
2. H2SO4, H2O,
85%
F
4
5
O
OH
N
1. NH2OH, pyridine,
2. NCS, DMF,
O
N
NHAc
Cl
F
6
R
R
TEA, DMF
TEA, DMF
O
N
O
R
R
O
O
N
O
O
NHAc
N
NHAc
N
F
F
3a-m
7a, b
Scheme 1. Synthesis of 3-isoxazolinyl and 3-isoxazolyloxazolidinones.
Compounds 3a–m were readily synthesized in several
steps from the known intermediate 4 (Scheme 1).9
Formylation of 4, utilizing modified Duff conditions,
occurs regioselectively to afford aldehyde 5 in excellent
yield.10 Examination of the splitting pattern of the
protons in the aromatic region indicated that formyl-
ation occurred exclusively para to the oxazolidinone
ring. Proton Hc is a triplet at d 7.88 (J ¼ 8:5 Hz) indi-
cating meta fluorine and ortho hydrogen coupling. Pro-
ton Ha is a doublet of doublets at d 7.63 (J ¼ 13:6,
1.9 Hz) indicating ortho fluorine and meta hydrogen
coupling. Lastly, Hb is a doublet of doublets at d 7.28
(J ¼ 8:5, 1.9 Hz) indicating ortho and meta proton
couplings. This is the typical coupling pattern seen in
other series of oxazolidinones. Additional evidence for
this regiochemical assignment is the appearance of the
aldehyde proton as a singlet indicating that the fluorine
is an even number of bonds removed since 5-bond
fluorine coupling to aldehydic protons has been
observed in other systems.10
analytical method to determine a more accurate ratio or
to separate the diastereomers. Isoxazole cycloadducts 7
were produced when an alkynyl dipolarophile was uti-
lized in the cycloaddition reaction.
Oxazolidinones 3 and 7 were characterized in vitro
against a representative methicillin-susceptible S. aureus
strain in broth, as well as in the presence of 50% mouse
serum, using a twofold dilution series of concentrations
for testing.12 The minimum inhibitory concentrations
(MIC) for these compounds are reported in Table 1. A
fourfold or greater increase in the MIC in the presence
of serum suggests that the compound binds to serum
proteins or is inactivated by components of the serum.
As a result, there may no longer be a sufficient con-
centration of free drug in the blood to inhibit the growth
of bacteria effectively in vivo.
Isoxazolinyl ester 3a was fourfold less potent than line-
zolid and suffered a dramatic increase in MIC in the
presence of mouse serum, such that it no longer inhibi-
ted bacterial growth at concentrations as high as 128 lg/
mL. The esterases in serum may have hydrolyzed the
ester to the corresponding acid, which may no longer
be able to enter the bacterial cell. In contrast, the cor-
responding primary amide 3b and the nitrile analog 3c
were equipotent to linezolid in broth and twofold more
potent in the presence of serum. In order to increase the
aqueous solubility of amide 3b, secondary amide 3d, in
which an alkyl ether is appended to the amide nitrogen
was prepared. This compound was less potent than both
3b and the secondary amide analog 3e. Tertiary amide 3f
was also significantly less active than 3b, suggesting
limited steric bulk tolerance in this region of the mole-
cule.
Conversion to the oxime followed by chlorination with
N-chlorosuccinimide produced the nitrile oxide precur-
sor 6.11 Dehydrohalogenation of chlorooxime 6 with
base in the presence of a dipolarophile afforded the
5-substituted cycloadducts 3 in good yields with none of
the regioisomeric 4-substituted cycloadduct being
observed by NMR or LC/MS. Isoxazolines 3 were pre-
sumably formed as a diastereomeric mixture although
diastereomers were not detected by proton NMR, LC/
MS (several solvent systems), or TLC. HPLC analysis of
the crude reaction mixture utilizing a variety of solvent
systems as well as chiral stationary phases suggested that
both diastereomers were present in an approximate 1:1
ratio. However, we were unable to develop a practical

M. A. Weidner-Wells et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3069–3072
3071
Table 1. MIC values (lg/mL) for oxazolidinones 3a–m and 7a,ba
Table 2. Spectrum of activity against susceptible and resistant
organisms (MIC (lg/mL))
O
Compd MRSA
E. faecalis E. faecium E. coli
OC 3312
(VRE)
E. coli
OC 2530
(HS)a
R
O
N
OC2878 ATCC
29212
OC
2605
O
NHAc
N
F
3a
3b
3c
3d
3e
3f
4
2
16
4
8
4
>128
>64
128
128
32
16
64
32
32
32
>32
64
32
8
Compds
R
Bondb
S. aureus
1
2
2
OC 4172
16
4
16
4
16
4
>64
>128
>128
64
Without With
serum
8
8
8
serum
3g
3h
3i
8
16
>32
8
32
>32
8
3a
3b
3c
3d
3e
CO2Et
C(O)NH2
CN
s
s
s
s
s
8
2
2
8
4
>128
>32
8
>32
128
1
1
8
4
3j
1
4
4
>64
>64
>64
>32
>128
>128
>32
C(O)NHCH2CH2OMe
C(O)NHMe
3k
3l
2
2
2
16
8
16
8
16
16
8
64
16
64
64
16
O
3m
7a
7b
1
3f
s
8
8
N
O
O
4
16
4
2
4
2
4
4
3g
3h
s
s
16
16
N
N
a HS: hypersensitive.
N
>32
>32
3i
3j
3k
3l
CH2NMe2
4-Pyridyl
2-Pyridyl
s
s
s
s
8
4
2
8
2
4
4
8
3k, and 7b exhibit microbiological activity comparable
to linezolid against MRSA, E. faecalis, and VRE. Like
linezolid, this series did not have an effect on an E. coli
wild type strain (OC 2605), but did exhibit marginal
activity against a mutant E. coli strain (OC 2530) that is
hypersensitive to antibacterial agents due to a defective
outer membrane. The lack of activity for this class
against Gram-negative organisms may be due in part to
active efflux, effectively lowering the intracellular con-
centration of test compound.13
CH2NHAc
3m
s
8
8
N
NAc
7a
7b
1
CO2Me
2-Pyridyl
––
d
8
4
2
>128
d
––
8
2
a The variance in the determination of MIC values is twofold such that
an MIC difference of more than fourfold is significant.
b s indicates a single bond (isoxazoline); d indicates a double bond
(isoxazole).
Compounds 3b, 3c, 3k, and 7b were tested in a lethal
murine systemic infection model with S. aureus Smith
(OC 4172) (Table 3). The isoxazolinyl compounds 3b,
3c, and 3k protected the mice from infection following
oral or subcutaneous administration, however, they
were several fold less potent than linezolid 1 despite
comparable in vitro activity against this organism. This
difference in in vivo potency may be due to poor solu-
bility in the vehicle or suboptimal pharmacokinetics.
Isoxazole 7b was inactive when tested subcutaneously.
Morpholine analog 3g and imidazole analog 3h were
significantly less potent than linezolid, thus indicating
that a sterically demanding basic functionality is not
tolerated. Although, the less bulky dimethylamino
analog 3i was slightly more potent than either 3g or 3h,
the MIC value of 3i was fourfold higher than that of
linezolid. Incorporation of a pyridyl group onto the
isoxazoline ring enhanced the in vitro potency of the
series. Regioisomeric analogs 3j and 3k were two- to
fourfold more potent than dimethylamino analog 3i
with the 2-pyridyl isomer 3k being optimal.
In conclusion, we have identified a series of isoxazolinyl
oxazolidinones with in vitro microbiological activity
against resistant and susceptible Gram-positive organ-
isms comparable to linezolid. Several analogs from this
series also exhibited in vivo activity in a lethal murine
infection model when administered either subcutane-
ously or orally, but were less effective than linezolid.
In light of the above structure–activity relationships, the
antimicrobial activity of several acylated amines was
investigated. Compounds 3l and 3m were equipotent to
dimethylamino analog 3i against the S. aureus strain,
but were fourfold less potent than linezolid.
From the limited number of examples, there appears to
be little difference in the in vitro activity between the
aromatic and the dihydro species (3a vs 7a and 3k vs 7b).
Table 3. In vivo efficacy in a murine lethal infection model
Compd
ED50 (sc) (mg/kg/day)
ED50 (po) (mg/kg/day)
3b
3c
3k
7b
1
11.8
22.4
22.8
>40
5.4
40
22.1
This series of oxazolidinone antimicrobial agents was
screened against an expanded panel of susceptible and
resistant Gram-positive and Gram-negative organisms
(Table 2). As is evident from the data, analogs 3b, 3c, 3j,
31.6
Not tested
6.8

3072
M. A. Weidner-Wells et al. / Bioorg. Med. Chem. Lett. 14 (2004) 3069–3072
4. Genin, M. J. Exp. Opin. Ther. Patents 2000, 10, 1405–
1414.
5. Hutchinson, D. K. Curr. Top. Med. Chem. 2003, 3, 1021–
1042.
6. Lee, C. S.; Allwine, D. A.; Barbachyn, M. R.; Grega, K.
C.; Dolak, L. A.; Ford, C. W.; Jensen, R. M.; Seest, E. P.;
Hamel, J. C.; Schaadt, R. D.; Stapert, D.; Yagi, B. H.;
Zurenko, G. E.; Genin, M. J. Bioorg. Med. Chem. 2001, 9,
3243–3253.
7. Hester, J. B., Jr.; Nidy, E. G.; Perricone, S. C.; Poel, T.-J.
U.S. 6,255,304 (Pharmacia), 2001.
8. Barbachyn, M. R.; Brickner, S. J. U.S. 5,801,246 (Phar-
macia), 1998.
9. Barbachyn, M. R.; Brickner, S. J. WO 93/09103 (Upjohn),
1993.
10. Weidner-Wells, M. A.; Fraga-Spano, S. A. Synth. Com-
mun. 1996, 14, 3775–3783.
Minimum inhibitory concentration (MIC) determina-
tions: Antibacterial susceptibility testing was performed
according to the broth microdilution method of the
National Committee for Clinical Laboratory Stan-
dards.12
Mouse protection assay: In vivo efficacy was assessed in a
murine septicemia model of infection caused by S. aur-
eus Smith. Female Swiss-Webster mice were infected ip
with approximately 6 · 105 CFU/mL of the challenging
strain. Protecting compounds were administered sub-
cutaneously in 40% hydroxypropyl-b-cyclodextrin or
orally in 0.5% Methocel at 1 and 3 h after infection. The
dose allowing survival of 50% of the animals (ED50) was
calculated using the logistic routine of the SAS suite of
statistical programs.
11. Liu, K.-C.; Shelton, B. R.; Howe, R. K. J. Org. Chem.
1980, 45, 3916–3918.
12. National Committee for Clinical Laboratory Standards,
1997. Methods for dilution antimicrobial susceptibility
tests for bacteria that grow aerobically, 4th ed. Approved
standard. NCCLS Document M7-A4. National Commit-
tee for Clinical Laboratory Standards, Villanova, PA.
13. Buysse, J. M.; Demyan, W. F.; Dunyak, D. S.; Stapert, D.;
Hamel, J. C.; Ford, C. W. 36th Interscience Conference on
Antimicrobial Agents and Chemotherapy (ICAAC), C-42,
1996, New Orleans, LA.
References and notes
1. Diekema, D. J.; Jones, R. N. Drugs 2000, 59, 7–16.
2. Kaatz, G. W.; Rybak, M. J. Emerging Drugs 2001, 6, 43–
55.
3. Barbachyn, M. R.; Ford, C. W. Angew. Chem., Int. Ed.
2003, 42, 2010–2023.