The synthesis and biological evaluation of a novel series of antimicrobials of the oxazolidinone class.

Article abstract of DOI:10.1016/S0960-894X(02)00352-9

A novel series of antimicrobials of the oxazolidinone class is disclosed. These compounds are characterized relative to previously described analogues by a 'halostilbene-derived' pharmacophore and demonstrate enhanced antimicrobial activity against key Gram-positive pathogens when compared to Linezolid.

Full text of DOI:10.1016/S0960-894X(02)00352-9

Bioorganic & Medicinal Chemistry Letters 12 (2002) 2121–2123
The Synthesis and Biological Evaluation of a Novel Series of
Antimicrobials of the Oxazolidinone Class
Richard J. Sciotti,* Marina Pliushchev, Paul E. Wiedeman, Darlene Balli,
Robert Flamm, Angela M. Nilius, Kennan Marsh, DeAnne Stolarik, Robert Jolly,
Roger Ulrich and Stevan W. Djuric*
Abbott Laboratories, Infectious Disease Research, Department 47N, Abbott Park, IL 60064, USA
Received 25 February 2002; accepted 8 May 2002
Abstract—A novel series of antimicrobials of the oxazolidinone class is disclosed. These compounds are characterized relative to
previously described analogues by a ‘halostilbene-derived’ pharmacophore and demonstrate enhanced antimicrobial activity against
key Gram-positive pathogens when compared to Linezolid. # 2002 Elsevier Science Ltd. All rights reserved.
The oxazolidinones represent a novel class of anti-
microbial agent that possess excellent activity against
antibiotic-sensitive and MIC90 resistant Gram-positive
bacteria.1 The prototype member of this class, Lin-
ezolid, has recently been approved for a number of
clinical applications in man including the treatment of
nosocomial and community-acquired pneumonia and
skin infections caused by Staphylococcus aureus/Methi-
cillin resistant S. aureus, Vancomycin resistant Enter-
ococci, and Streptococcus pneumoniae (Pen-S). We were
interested in identifying second generation compounds
that had greater antimicrobial potency and less poten-
tial for side effects such as myelosuppression, including
anaemia, and thrombocytopenia.2 To this end, we pur-
sued a series of stilbene/‘heterostilbene’-based analogues
of general structure A whose conception was based on
the structure and activity of a previously disclosed a,b
unsaturated ester3 and a series of amide based phenyl-
oxazolidinone derivatives.4 In this context, it should be
remembered that vinyl fluorides are a well-known class
of amide bond mimetics.5
The compounds wherein the halogen group is fluoro
were constructed as outlined in Scheme 1. Synthesis of
the corresponding bromo analogues is shown in Scheme 2.
The sequence to access the target vinyl fluorides fea-
tured as key components the palladium(0) catalyzed
oxazolidinone/aryl halide coupling reaction recently
disclosed by us6,7 and the known but critical aldehyde to
a-fluoro vinyl stannane conversion introduced by
McCarthy and co-workers.8À10 The a-fluoro vinyl stan-
nane could be subsequently coupled under Pd(0) cata-
lyzed conditions with the requisite aryl halide. This
reaction proved to be especially capricious with only 11
out of 45 reactions providing product in modest yield
(range 2–40%).
Scheme 1. Synthesis of fluorostilbenes: (a) p-TsCl, pyridine, À5 ꢀC, 2 h,
95%; (b) potassium phthalimide, DMF, 80 ꢀC, 5 h, 75%; (c) 4-bromo-
2-fluorobenzaldehyde, Pd2dba3, Cs2CO3, BINAP, toluene, 100 ꢀC, 16
h, 77%; (d) fluoromethyl phenyl sulfone, (EtO)2P(O)Cl, LiN(TMS)2,
THF, 93%; (e) (i) N2H2–H2O, EtOH; (ii) Ac2O, pyridine/CH2Cl2,
67%; (f) n-Bu3SnH (4 equiv), AIBN, benzene, 80 ꢀC, 81%; (g) R-X
(1.01 equiv), CuI (0.95 equiv), Pd2dba3 (0.05 equiv), Pt-Bu3 (0.2
equiv), dioxane (0.2 M), 2–8 h, 75 ꢀC.
*Corresponding author. Fax: +1-847-938-1674. E-mail: richard.
sciotti@pfizer.com (R. J. Sciotti); stevan.w.djuric@abbott.com (S. W.
Djuric).
0960-894X/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S0960-894X(02)00352-9

2122
R. J. Sciotti et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2121–2123
Scheme 2 featured a new and useful version of the
Corey–Fuchs11 aldehyde to dibromoolefin conversion
using polymer bound triphenylphosphine. This modifi-
cation avoids the tedious removal of triphenyphosphine
that is associated with the parent procedure. The 1,1-
dibromoolefins coupled effectively with a variety of aryl
halides to provide target analogues in poor to good
yields (range 12–82%).
Scheme 2. Synthesis of bromostilbenes: (a) polystyrene-PPh3 (6
equiv), CBr4 (3 equiv), CH2Cl2, À5 ꢀC, 1 h, 93%; (b) N2H4–H2O (3
equiv), THF/EtOH, 70 ꢀC, 3 h; (c) Ac2O, pyridine, rt, 5 min, 77%
total for (b)+(c); (d) R-B(OH)2, Pd(PPh3)4 (0.1 equiv), 10% Na2CO3/
H2O, DME, 70 ꢀC, 16 h.
The compounds were evaluated for antimicrobial
activity against a panel of representative Gram-positive
pathogens including clinically relevant strains. Many
compounds demonstrated significantly better activity
than Linezolid against several of these organisms. In
Table 1. Structure–activity relationships in halostilbene oxazolidinones
Compd
R1
R2
S. aureus S. epidermidis E. faecium
MICa
M. cat.
MICa
S. pneumo.
MICa
MPSb
IC50 (mM) IC50 (mM)
MAO Bc
MICa
MICa
(mg/mL)
(mg/mL)
(mg/mL)
(mg/mL)
(mg/mL)
Linezolid 1
2
3
4
5
6
7
8
9
4
4
1
1
4
4
2
8
1
2
0.5
0.25
2
0.25
1
4
2
2
4
16
4
2
16
8
1
16
12 (7)
9.4Æ1.4
<0.1 (2)
1.3Æ0.2
7.8Æ1.0
1.6Æ0.1
12 (2)
17 (5)
>300 (2)
>300 (2)
>300 (2)
59 (2)
5-CN-3-Pyridyl
5-CN-3-Pyridyl
5-CN-3-Pyridyl
3-Pyridyl
H
F
Br
H
1
<0.125
2
1
2
4
1
2
4
ꢁ0.125
ꢁ0.125
16
0.5
1
3-Pyridyl
3-Pyridyl
F
4-Pyridyl
F
98 (2)
121 (4)
Br
3-Pyridyl
Br
Br
Br
F
8
32
16
32
128
>128
1
>128
>128
4
128
64
1
16
>128
4
4
1
2
2
8
8
16
2
4
4
14(3)
186 (2)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
4-MeO-Pyridin-3-yl
2-CN-Ph
3-CN-Ph
36 (2)
>300 (2)
24 (2)
76 (2)
69Æ12
>150 (2)
>300 (2)
>150 (4)
>150 (4)
>300 (2)
>128
128
0.125
1
1
1
64
0.5
0.25
4
8
1
>128
0.125
2
2
3-CN-Ph
4-CN-Ph
3-Acetyl-Ph
3-Acetyl-Ph
4-Acetyl-Ph
3-CHO-Ph
Br
Br
F
Br
Br
F
Br
Br
Br
Br
Br
Br
Br
Br
F
F
Br
F
F
Br
F
F
F
Br
0.25
4
0.5
2
64
1
0.5
4
8
2
2
0.25
0.5
1
1
1
4
1
4
32
1
2
2
64
2
2
8
0.5
2
8
0.5
2
2
1
32
1
0.5
2
4
0.5
2
9.9Æ0.8
>300 (2)
>300 (2)
3-CHO-Ph
33 (2)
3-CONH2-Ph
3-CO2Me-Ph
3-CH2OH-Ph
3-NH2-Ph
150 (2)
8 (2)
>300 (2)
>150 (4)
1
2
8
3-N-Acetyl-Ph
5-iPr-2-MeO-Phenyl
Ph
4
>128
2
64
32
128
32
64
1
>128
16
8
>128
16
8
16
8
8
>300 (2)
<9.4 (2)
H
PhSO2–
2-Furyl
2-Thiazolyl
8
0.25
2
16
0.5
1
1
4
16
8
0.5
2
16
0.5
4
2
8
32
1
2
>128
8
<9.4 (1)
285 (2)
2-CN-Thiophen-3-yl
5-Ac-Thiophen-2-yl
2-Me-2H-Tetrazol-5-yl
5-Pyrimidinyl
2-Amino-5-pyrimidinyl
3,5-di-Me-Oxazol-4-yl
>128
2
>300 (2)
4.9Æ0.5
1
8
2
16
32
32
4
32
64
>300 (2)
475 (1)
1
8
8
54 (2)
21 (2)
aNational Committee for Clinical Laboratory Standards. 1997. Methods for Dilution Antimicrobial Susceptibility Tests for Bacteria that Grow
Aerobically: Approved Standard M7-A5. National Committee for Clinical Laboratory Standards, Wayne, PA, USA.
bMitochondrial Protein Synthesis (MPS) IC50 expressed as a mean ÆSTD or as a mean of (n) experiments. See ref 12 for assay protocol.
cMonoaminoxidase B IC50 expressed as a mean of (n) experiments. Assay adapted from a Molecular Probes kit. The assay was carried out using rat
primary hepatocytes because the rat was the primary model for toxicity used in assessing these compounds. We have reported the MAO B data
because, in all cases, the compounds were significantly more inhibitory using substrate for this isozyme than for the MAO A isozyme, where little if
any inhibition was noted.

R. J. Sciotti et al. / Bioorg. Med. Chem. Lett. 12 (2002) 2121–2123
2123
addition, the compounds were evaluated in two key
toxicity/selectivity assays (a) inhibition of mitochondrial
protein synthesis (this effect may be, at least in part,
responsible for the hematologic toxicities associated
with Linezolid)12 and (b) inhibition of monoamine
oxidase. Although this is not a significant clinical prob-
lem with Linezolid, it is a class effect for oxazolidi-
nones13 and needs to be avoided. The biological SAR
data is highlighted in Table 1. Several compounds from
either series, gratifyingly, showed increased potency,
reduced MPS and MAO inhibitory activity compared to
Linezolid. Although it was difficult to draw SAR con-
clusions on the series in a general sense (no clear cut
patterns or direct comparison between the bromo and
fluoro series being impossible due to the relative inac-
cessibility of target compounds) some general comments
are instructive. Compounds containing halogen-sub-
stituted double bonds were as a rule more potent than
their corresponding des-halo counterparts (2–4 and 5–
7). Within a halogenated series, the bromo analogues
were generally more potent than the corresponding
fluoro compounds (3–4, 6–7, 12–13, 15–16, and 18–19).
Activity of compounds containing the (Z)-fluorostilbene
unit was greater than that of the corresponding (E)-iso-
mer (6 and 8). A variety of R1 groups were well toler-
ated with a general trend that heterocycles bearing
electron-withdrawing groups (higher sigma values) were
the most active in the antimicrobial screens. There was a
potency enhancement associated with substituents on
the meta position of phenyl groups (13 and 14, 16 and
17) or with a 3-pyridyl moiety to form a ‘heterostilbene’
(7 and 9). This was not generally true for analogues that
bore a phenyl group substituted with an electron-with-
drawing group in the para position, for example 17.
2. SCRIP—World Pharmaceutical News, 7th March 2001.
3. Brittelli, D. R.; Gregory, W. A.; Corless, P. F.; Park, C. H.
US Patent 4, 977, 173, 1990; Chem. Abstr. 1989, 111, 232783.
4. Gordeev, M. F.; Luehr, G. W.; Patel, D.; Gordon, E. PCT
Patent Application WO 99/37630; Chem. Abstr. 1999, 131,
116228.
5. Allmendinger, T.; Furet, P.; Hungerbuhler, E. Tetrahedron
Lett. 1990, 31, 7279.
6. Madar, D. J.; Kopecka, H.; Pireh, D.; Pease, J.; Pliushchev,
M.; Sciotti, R. J.; Wiedeman, P. E.; Djuric, S. W. Tetrahedron
Lett. 2001, 42, 3681.
7. Danielmeier, K.; Steckhan, E. Tetrahedron: Asymmetry
1995, 6, 1181.
8. Matthews, D. P.; Persichetti, R. A.; McCarthy, J. R. Org.
Prep. Proced. Int. 1994, 26, 605.
9. Chen, C.; Wilcoxen, K.; Zhu, Y.-F.; Kim, K.; McCarthy,
J. R. J. Org. Chem. 1999, 64, 3476.
10. Huber, E. W.; Le, T.-B.; Laskovics, F. M.; Matthews,
D. P.; McCarthy, J. R. Tetrahedron 1996, 52, 45.
11. Corey, E. J.; Fuchs, P. L. Tetrahedron Lett. 1972, 3769.
12. Mitochondrial Protein Synthesis (MPS) Assay. Rat heart
mitochondria were isolated as described by McKee et al. (Am.
J. Physiol. 1990, 258, E492). Mitochondria were incubated at 1
mg/mL protein as determined by the Bradford assay in buffer
containing constituents essential for translation (ADP, amino
acids, etc.) in a total volume of 50 mL. Protein synthesis was
measured by following incorporation of S35 methionine at a
known specific activity for one h at 30 ꢀC. The reaction was
quenched by addition of 50 mL of 5% cold TCA/5 mM
methionine and transferring it to ice. 20 mL aliquots (in tripli-
cate) of each reaction were transferred to Millipore multi-
screen plates. The plates were washed five times using 5%
TCA/5 mM methionine. Plates were dried overnight, 50 mL of
Optimax scintillant was added and the plates were counted
using a Wallac Microbeta 1450 (Perkin-Elmer). Abbott com-
pounds as well as positive and negative controls were prepared
in DMSO as 100Â stocks and assessed at 6–7 varying con-
centrations by including them in the incubation and then
determining S35 incorporation relative to control. The percent
organic in the reaction did not exceed 1% of the total volume.
An IC50 based on percent of control was calculated from the
linear portion of the inhibition/concentration curve. Com-
pounds were analyzed in at least two separate experiments and
these data were summarized to give a mean IC50. We hypo-
thesized that the hematosuppression of linezolid was due to an
inhibition of mitochondrial function and protein synthesis
similar to what has been described with chloramphenicol (see
Holt, D.; Harvey, D.; Hurley, R. Adverse Drug React. Toxicol.
Rev. 1993, 12, 83. Yunis, A. A. Am. J. Med. 1989, 87, 44N),
and we used the MPS assay as an index of mitochondrial
function in these studies. Both chloramphenicol and linezolid
inhibit MPS at low to sub-micromolar concentrations in vitro
(IC50’s less than 1 and 12 mM, respectively). Both compounds
have been shown to inhibit MPS and reduce both cytochrome
C oxidase activity and mitochondrial protein synthesis in bone
marrow of treated rats implicating the loss of mitochondrial
function in the toxicity (data not shown).
Representative compounds with potency comparable to
1 were selected for pharmacokinetic evaluation in cas-
sette dosing format in the rat. Compound 6 displayed
characteristics that were considered to be appropriate
for in vivo evaluation in animal infection models. For
example, the PK profile of 6 following a 5 mpk iv dose
was characterized by
a
low plasma clearance
(CLp=0.25 L/h kg), with a volume of distribution of
0.42 L/kg and an apparent elimination half life of ꢂ1.2
h. When administered orally at the same dose, the
bioavailability of the compound was found to be 65%,
n=6). The outcome of these in vivo studies will be
reported upon in due course.
References and Notes
1. Ford, C.; Hamel, J.; Stapert, D.; Moerman, J.; Hutchinson,
D.; Barbachyn, M.; Zurenko, G. Infect. Med. 1999, 16, 435.
13. Gates, K. S.; Silverman, R. B. J. Am. Chem. Soc. 1989,
111, 8891.