Synthesis of novel tricyclic oxazolidinones by a tandem SN2 and SNAr reaction: SAR studies on conformationally constrained analogues of Linezolid

Article abstract of DOI:10.1016/j.bmcl.2006.05.071

A series of conformationally constrained analogues of Linezolid were synthesised by employing a tandem SN₂ and SNAr reaction as the key step and tested for antibacterial activity. While the hexahydroazolo-quinoxaline compounds were inactive, th

Full text of DOI:10.1016/j.bmcl.2006.05.071

Bioorganic & Medicinal Chemistry Letters 16 (2006) 4416–4419
Synthesis of novel tricyclic oxazolidinones by a tandem
SN2 and SNAr reaction: SAR studies on conformationally
constrained analogues of Linezolidq
N. Selvakumar,* B. Yadi Reddy, G. Sunil Kumar, Manoj Kumar Khera, D. Srinivas,
M. Sitaram Kumar, Jagattaran Das, Javed Iqbal and Sanjay Trehan
Anti-infectives Discovery Group, Discovery Research, Dr. Reddy’s Laboratories Ltd, Miyapur, Hyderabad 500 049, India
Received 28 November 2005; revised 22 April 2006; accepted 9 May 2006
Available online 14 June 2006
Abstract—A series of conformationally constrained analogues of Linezolid were synthesised by employing a tandem SN2 and SNAr
reaction as the key step and tested for antibacterial activity. While the hexahydroazolo-quinoxaline compounds were inactive, the
tetrahydroazolo-benzothiazine compounds exhibited interesting antibacterial activity. The introduction of fluorine in the aromatic
ring further made the compounds more potent in acetamide compounds resulting in an interesting analogue 32. However, the intro-
duction of fluorine (analogue 34) on the already potent non-fluorine thiocarbamate 21 did not have any influence on the activity.
Ó 2006 Elsevier Ltd. All rights reserved.
The emergence of multidrug-resistant Gram-positive
bacteria including methicillin-resistant Staphylococcus
aureus (MRSA) and Staphylococcus epidermitis
(MRSE) and vancomycin-resistant enterococci (VRE)
is of major concern.1–5 In addition, an alarming situ-
ation has been created by the emergence of strains
of glycopeptide-intermediate S. aureus (GISA) with
reduced susceptibility to vancomycin.6,7 Strains of
common intestinal bacteria Enterococcus faecalis and
E. faecium have been resistant to vancomycin for
several years. Oxazolidinones are a new class of
synthetic antibacterials with activity against Gram-
positive bacteria and anaerobic bacteria, including
resistant pathogens.8
constrained analogues of Linezolid (e.g., compound 1)
after an initial lead identification and SAR studies on
the right-hand side (Fig. 1).10 However, we have not
undertaken a SAR on the left-hand side of the molecule
1 to vary the oxygen heteroatom to other heteroatoms
such as sulfur and nitrogen. In continuation of our
efforts in this tricyclic group of oxazolidinones, we
now report the synthesis and the SAR studies on the left
hand side of the molecule. We have synthesised a num-
ber of tricyclic oxazolidinones analogous of 2 and 3 pos-
sessing substituted nitrogen and sulfur atom,
respectively, instead of oxygen on the tricycle.
Synthesis of hexahydroazolo-quinoxaline compounds:
Addition of L-prolinol onto 1-chloro-2,4-dinitrobenzene
afforded the dinitro-alcohol 4 (Scheme 1). The com-
pound 4 upon hydrogenation presumably resulted in
the corresponding diamine-alcohol, which posed diffi-
The discovery of Linezolid as a new class of antibacteri-
als particularly active against VREs has triggered wide-
spread research in this area.9 In an ongoing programme
on the discovery of novel second generation oxazolidi-
nones, we have earlier reported certain conformationally
O
O
SAR
Me
O
N
O
N
N
N
Keywords: Oxazolidinones; Antibacterial; Tricyclic; Linezolid.
q
DRL Publication No. 219-A, A part of this work was presented in
the 42nd Interscience Conference on Antimicrobial Agents and
Chemotherapy, September 27–30, 2002, San Diego, USA.
H
H
N
N
Me
O
X
O
2: X = NMe
3: X = S
O
1
*
Corresponding author. Tel.: +91 40 3045439; fax: +91 40 3045438;
e-mail: selvakumarn@drreddys.com
Figure 1.
0960-894X/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2006.05.071

N. Selvakumar et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4416–4419
4417
O2N
F
Cl
NO2
F
NO2
NH
+
+
NH
a
NO
2b,c
NO2
b
N
N
a
O2N
F
OH
OH
OH
F
OH
4
13
F
Cbz-HN
N
Cbz-HN
N
c
NO2
N
NO2
N
d
NH-Cbz
O
NH-Cbz
d
e
SAc
Br
F
Br
6
OH
15
14
5
NO2
N
N
NO2
O
N
f,g
N
NH-Cbz
N
N
Cbz
8: R = OH
e
N
f
R
SH
S
N
Cbz
17
16
g,h,i
O
7
9: R = NHAc
NH-Cbz
h,i,j
k
O
N
N
j
k
N3
O
S
O
S
18
19
O
N
N
O
N
N
H
l,m,n
S
H
N
O
O
N
N
CHO
N
R
O
O
H
N
N
O
N
N
O
N
O
H
10
N
OMe
11: R = COCH3
R
12: R = COCH2Cl
O
S
21
3: R = S
20: R = SO2
Scheme 1. Reagents and conditions: (a) Et3N, CH3CN, rt, 4 h, 94%;
(b) 10% Pd on C, H2, THF, 14 h; (c) Cbz-Cl, aq Na2CO3, acetone, rt,
2 h, 75% for two steps; (d) CBr4, PPh3, CH2Cl2, rt, 48 h, 70%; (e)
K2CO3, DMF, rt, 28 h, 62%; (f) BuLi, (R)-glycidyl butyrate, THF,
À78 °C to rt, 14 h, 58%; (g) MsCl, Et3N, CH2Cl2, 0 °C, 4 h, 96%; (h)
NaN3, DMF, 80 °C, 1 h, 91%; (i) MeCOSH, rt, 14 h, 68%; (j) 10% Pd
on C, H2 then HCOOEt, 41%; (k) 10% Pd on C, H2 then AcCl (62%)
or chloroacetyl chloride, Et3N (55%).
Scheme 2. Reagents and conditions: (a) Hunig’s base, CH3CN, rt, 5 h,
86%; (b) CBr4, PPh3, CH2Cl2, 0 to 45 °C, 5 h, 93%; (c) MeCOSH,
NaH, THF, 8 h, 100%; (d) K2CO3, MeOH, reflux, 3 h, 66%; (e) NaH,
DMF, 60 °C, 1 h, 46%; (f) Fe, HCl, EtOH, 0 °C to rt, 45 min.; (g) Cbz-
Cl, aq Na2CO3, acetone, 0 °C to rt, 5 h, 81% for two steps; (h) BuLi,
(R)-glycidyl butyrate, THF, À78 °C, 4 h, 56%; (i) MsCl, Et3N,
CH2Cl2, 0 °C to rt, 2 h, 90%; (j) NaN3, DMF, 80 °C, 2 h, 91%; (k)
MeCOSH, rt, 15 h, 65%; (l) PPh3, H2O, THF, 68%; (m) CSCl2, Et3N;
(n) MeOH, reflux, 56% for two steps; (o) NaIO4, CH2Cl2, 50%.
culty for isolation due to its instability and thus was
directly treated with Cbz chloride in situ to produce
the di-Cbz compound 5. The compound 5 was converted
into its bromo derivative 6 with CBr4 and PPh3, which
was cyclised with K2CO3 in DMF to result in the tricy-
clic Cbz compound 7. After this point, the compound 7
was converted into the acetamide 9 as per the usual
literature procedures.9 The attempts to deprotect the
Cbz group to the corresponding free amine or as a salt
failed to give the required material because of the insta-
bility of the product. Thus, we resorted to installing
smaller groups such as formate, acetate and chloroace-
tate employing usual procedures of treating the amine
in situ with the appropriate reagents to give the com-
pounds 10, 11 and 12, respectively.
the protocol described in Scheme 1.11 The acetamide 3
was further converted to the sulfone 20 with sodium
metaperiodate. The intermediate azide 19 could be
converted to the thiocarbamate 21 following usual
reaction conditions.
Synthesis of fluorine containing hexahydroazolo-quinox-
aline and tetrahydroazolo-benzothiazine compounds:
A new methodology was developed for the prepara-
tion of title compounds involving a tandem SN2 and
a SNAr reaction. The nitro-alcohol 22 (Scheme 3),
obtained as per our earlier method,10 was converted
to the bromo compound 23 as usual. The analogous
bromo compound 14 in Scheme 2 was initially con-
verted to the thiol 16, by a SN2 reaction of the anion
of thioacetic acid followed by basic hydrolysis, before
inducing the cyclisation by a SNAr reaction to result
in the tricyclic nitro compound 17. Instead of this
usual three-step protocol, we envisioned a single pot
tandem SN2 and a SNAr reaction involving thioacetic
acid and a base as it was anticipated that the anion
addition onto bromide followed by the hydrolysis
and the subsequent cyclisation should occur
Synthesis of tetrahydroazolo-benzothiazine compounds:
Addition of L-prolinol onto 3,4-difluoronitrobenzene
resulted in the nitro-alcohol 13 (Scheme 2). The bromo
compound 14, obtained from the nitro-alcohol 13 as
above, was converted to the thioacetate 15 with
thioacetic acid. The treatment of the thioacetate 15
with K2CO3 in MeOH resulted in the thiol 16 that
was cyclised with NaH in DMF to yield the tricyclic
nitro compound 17. The nitro compound 17 was
converted into the acetamide 3 by essentially following

4418
N. Selvakumar et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4416–4419
F
F
The constrained analogues of Linezolid prepared
F
above were screened for in vitro activity against a
panel of Gram-positive organisms and the results
are summarized in Table 1. The tricyclic oxazolidi-
none analogues 9–12, possessing nitrogen atom
(bearing different substituents) instead of oxygen in
the tricycle of compound 1, exhibited no antibacte-
rial activity. It is relevant to note that the earlier
compound 1, possessing an oxygen atom, had mod-
erate activity (Table 1). Similarly, the analogue 33
containing a N-methyl group instead of oxygen in
the compound 1 with an additional fluorine atom
in the aromatic ring was also found to be inactive.
The corresponding thiocarbamate compound 36 of
the acetamide 33 was also an inactive compound.
Thus, it was concluded that varying the oxygen
atom in the constrained analogue 1 to a substituted
nitrogen atom results in loss of antibacterial
activity.
F
NO2
NH
+
NO2
N
a
b
F
OH
F
HO
22
F
X
d
c
NO2
NO2
N
N
F
F
Br
24: X = S
23
25: X = NMe
F
O
e
f
NH-Cbz
O
N
N
N
N3
X
X
26: X = S
28: X = S
27: X = NMe
29: X = NMe
F
O
F
O
g
As the above SAR of varying the oxygen atom in
the constrained analogue 1 to substituted nitrogen
atom resulted in inactive compounds, we turned
our attention in introducing sulfur atom in place
O
i,j
N
N
O
N
N
H
NH2
N
X
X
30: X = S
32: X = S
O
31: X = NMe
33: X = NMe
of the oxygen. Thus, the acetamide analogue
3
exhibited moderate activity (8 lg/mL) against all
the organisms tested, which was more or less com-
parable to the activity of oxygen analogue 1 (4–
16 lg/mL). Encouraged by the observed results, the
thiocarbamate analogue 21 was prepared and was
found to be highly potent with in vitro activity
ranging from 0.25 to 1 lg/mL. But, the conversion
of the sulfur atom to the sulfone 20 resulted in
the loss of activity. With the view of further increas-
ing the potency, following our earlier report,10 the
tricyclic compounds having fluorine atom in the aro-
matic ring were synthesised. Thus, the acetamide
compound 32 exhibited two- to fourfold more poten-
cy (1–2 lg/mL) than the non-fluorine analogue 3.
However, the thiocarbamate analogue 34 was only
marginally more potent than the acetamide 32. In
addition, the thioacetamide 37 and the thiourea 35
also possessed similar activity as that of the thiocar-
bamate 34. This implies that the introduction of
fluorine did not lead to increased potency in the thi-
ocarbamate analogues as observed for the acetamide
compounds. The non-fluorine analogue 21 was
selected for further development based on in vitro
activity.
h
F
N
X
F
O
O
O
O
N
N
N
H
H
N
R
N
S
S
S
37
34: X = S, R = OMe
35: X = S, R = NHMe
36: X = NMe, R = OMe
Scheme 3. Reagents and conditions: (a) Ref. 10; (b) CBr4, PPh3,
CH2Cl2, 0 to 45 °C, 5 h, 87%; (c) MeCOSH (42%) or MeNH2 (38%),
KOH, DMSO; (d) as in Scheme 2; (e) as in Scheme 1; (f) PPh3, H2O,
THF, 68%; (g) AcCl, Et3N, CH2Cl2, 89%; (h) Lawesson’s reagent,
68%; (i) CSCl2, Et3N, 87%; (j) MeOH (86% and 82% for 34 and 36)
and MeNH2 (61% for 35), reflux.
sequentially. Thus, the treatment of the bromide 23
with thioacetic acid and KOH in DMSO afforded
the tricyclic nitro compound 24 in one step albeit in
modest yield. In a similar manner, employing methyl-
amine in place of thioacetic acid in the above reaction
resulted in the tricyclic nitro compound 25 in one pot.
Having identified an efficient methodology to prepare
the tricyclic nitro compounds 24 and 25 in gram
quantities, the synthesis of the corresponding azides
28 and 29 was carried out as discussed in the previous
schemes. The amines 30 and 31, derived from their
corresponding azides, were converted to the
acetamides 32 and 33, respectively, under standard
conditions. The acetamide 32 was treated with the
Lawesson’s reagent to get the thioacetamide 37. Final-
ly, the amines 30 and 31 afforded the thiocarbamates
34 and 36, respectively, upon treatment with thiophos-
gene followed by refluxing the intermediate isothiocy-
anate in methanol. The treatment of the above
isothiocyanate intermediate obtained from the amine
30 with methylamine resulted in the thiourea 35.
In conclusion, the SAR studies on the left hand
side of the tricyclic analogue 1 in varying the oxy-
gen atom in the tricycle to substituted nitrogen
atom and sulfur were completed. While the change
to substituted nitrogen resulted in complete loss of
activity, introducing sulfur generated some potent
compounds. Though the introduction of fluorine
in the aromatic ring increased the potency in the
acetamide analogue (analogue 32), the same was
not true in the thiocarbamate case (analogue 34).
The thiocarbamate analogue 21 was chosen as the
lead compound and further development is in
progress.

N. Selvakumar et al. / Bioorg. Med. Chem. Lett. 16 (2006) 4416–4419
Table 1. In vitro antibacterial activity (MIC, lg/mL)aof selected tricyclic oxazolidinones
4419
X
O
O
N
N
R2
R1
Compound No.
X
R1
R2
S.a 019
S.a 213
S.a 035
E.f 034
E.f 153
E.fm 154
9
10
H
H
H
H
F
N-COOBn
N-CHO
NHCOCH3
NHCOCH3
NHCOCH3
NHCOCH3
NHCOCH3
NHCSOCH3
NHCOCH3
NHCSOCH3
NHCOCH3
NHCOCH3
NHCSCH3
NHCSOCH3
NHCSNHMe
NHCOCH3
>32
>256
>256
>256
>32
>32
8
>32
>256
>256
>256
>32
>32
8
>32
>256
>256
>256
>32
>32
8
>32
>256
>256
>256
>32
>32
8
>32
>256
>256
>256
>32
>32
8
>32
>256
>256
>256
32
11
12
N-COCH3
N-COCH2Cl
33
36
N-Me
N-Me
F
32
3
21
H
H
H
F
S
8
0.5
S
0.25
1
0.5
>32
2
0.25
>32
2
0.25
>32
1
20
32
SO2
S
>32
1
>32
1
>32
2
37
34
F
S
0.25
0.5
1
0.5
1
0.5
1
0.25
0.5
0.25
F
S
0.5
0.5
4
0.5
1
35
1
F
S
0.5
4
2
1
16
0.5
H
O
8
16
16
Linezolid
Vancomycin
1
2
2
2
2
>32
2
>32
2
1
1
2
a S.a 019 = Staphylococcus aureus ATCC 33591 (methicillin-resistant); S.a 213 = Staphylococcus aureus ATCC 49951; S.a 035 = Staphylococcus
aureus ATCC 29213; E.f 034 = Enterococcus faecalis ATCC 29212 (vancomycin sensitive); E.f 153 = Enterococcus faecalis NCTC 12201 (vanco-
mycin resistant) and E.fm 154 = Enterococcus faecium ATCC 12202 (vancomycin resistant).
Grega, K. C.; Hendges, S. K.; Toops, D. S.; Ford, C. W.;
Zurenko, G. E. J. Med. Chem. 1996, 39, 673.
Acknowledgments
10. Selvakumar, N.; Srinivas, D.; Khera, M. K.; Kumar, M.
S.; Mamidi, N. V. S. R.; Sarnaik, H.; Chandrasekar, C.;
Rao, B. S.; Raheem, M. A.; Das, J.; Iqbal, J.; Rajagopa-
lan, R. J. Med. Chem. 2002, 45, 3953.
We are indebted to Dr. K. Anji Reddy for his constant
support and encouragement. The help extended by Dr.
R. Rajagopalan is greatly acknowledged. We appreciate
the services extended by the Analytical Research
Department of Discovery Research, Dr. Reddy’s Labo-
ratories Ltd, for having carried out all the analytical
work.
11. Analytical data for selected final compounds: Compound
3: IR (KBr) 3355, 1738, 1666, 1507 cmÀ1 1H NMR
.
(CDCl3 + DMSO-d6) d 7.65 (bt, 1H), 7.21 (d, J = 2.7 Hz,
1H), 7.10 (dd, J = 8.8, 2.7 Hz, 1H), 7.46 (d, J = 8.8 Hz,
1H), 4.80–4.65 (m, 1H), 3.98 (t, J = 7.6 Hz, 1H), 3.80–3.25
(m, 5H), 3.05 (dd, J = 12.2, 2.4 Hz, 1H), 2.70–2.60 (m,
1H), 2.35–1.85 (m, 4H), 2.01 (s, 3H), 1.70–1.45 (m, 1H).
MS (DIP-method) 347 (M+), 303, 219. HPLC (System 2)10
References and notes
94.04% purity. Compound 21: IR (KBr) 1744, 1509 cmÀ1
.
1. Service, R. F. Science 1995, 270, 724.
1H NMR (200 MHz, CDCl3) d 7.20 (s, 1H), 7.15 (d,
J = 8.8 Hz, 1H), 6.76 (bt, 1H), 6.53 (d, J = 8.8 Hz, 1H),
5.00–4.80 (m, 1H), 4.20–3.20 (m, 9H), 3.04 (dd, J = 12.2,
2.4 Hz, 1H), 2.72 (dd, J = 12.2, 9.8 Hz, 1H), 2.28–1.40 (m,
5H). MS (CI-method) 380 (M++1), 348, 335. HPLC
(System 2)10 91.95% purity. Compound 34: IR (KBr)
2. Swartz, M. N. Proc. Natl. Acad. Sci. U.S.A. 1994, 91, 2420.
3. Tomasz, A. A. N. Engl. J. Med. 1994, 330, 1247.
4. Brumfitt, W.; Hamilton-Miller, J. M. T. Drugs Exp. Clin.
Res. 1994, 215.
5. Spera, R. V., Jr.; Farber, B. F. Drugs 1994, 48, 678.
6. Hiramatsu, K.; Hanaki, H.; Ino, T.; Yabuta, K.; Oguri, T.;
Tenover, F. C. J. Antimicrob. Chemother. 1997, 40, 135.
7. Waldvogel, F. A. N. Engl. J. Med. 1999, 340, 556.
8. Brickner, S. J. Curr. Pharm. Des. 1996, 2, 175.
9. Brickner, S. J.; Hutchinson, D. K.; Barbachyn, M. R.;
Manninen, P. R.; Ulanowicz, D. A.; Garmon, S. A.;
3264, 1752, 1492 cmÀ1 1H NMR (200 MHz, CDCl3) d
.
7.17 (dd, J = 16.6, 2.0 Hz, 1H), 6.85 (s, 1H), 6.76 (bt, 1H),
5.00–4.80 (m, 1H), 4.20–3.10 (m, 10H), 2.85–2.60 (m, 2H),
2.40–1.55 (m, 4H). MS (CI-method) 398 (M++1), 366, 354.
HPLC (System 2)10 98.11% purity.