C. Gao et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
3
These compounds exhibited strong activities overall but markedly
less active than compound 1. Comparison of 7a with 7b, and 8a
with 8b and 8c suggested that the potency of benzothiazinones
were affected by the length of the alkyl chain. The nitro substituted
benzothiazinone 7b and the corresponding compound 8c, which
activity against M. tuberculosis H37Rv.14,15 Compounds 8k and 8m
showed MIC of 0.08
than INH (MIC = 0.2
l
l
M and 0.03
lM, 2.5 and 6.7-fold more potent
M), respectively. Compound 8n was equipo-
tent to 1. Compound 8o was particularly notable for its excellent
efficacy (MIC = 0.1 nM), 20-fold and 2000-fold more effective than
1 and INH, respectively. This could be explained by the reason that
sulfur atom, the electron density of which is higher, is larger than
oxygen atom, and the replacement of oxygen by sulfur atom could
slightly adjust the volume and lipophilicity of the side chain, mak-
ing the molecule interact with active site of the DprE1 enzyme
preferably. The activity of compound 8o against M. smegmatis (Ta-
ble 2) was evaluated by a microplate Alamar Blue assay. 8o dis-
contained a trifluoromethyl, showed a same MIC of 0.8 lM.
Branched alkyl groups were slightly better tolerated than the un-
branched alkyl chain, as the more lipophilic compounds 7c and
8d had twice the anti-mycobacterial activities of compounds 7a
and 8b, respectively. Compound 7d (ClogP 2.5) with a tert-butyl
moiety displayed a MIC of 0.4 lM, which was 4-fold lower than
7c (ClogP 2.7). It indicated that the activity can be associated with
the size of alkyl chain. The introduction of cyclic alkyl groups did
not affect the activity (8f and 8g). Replacing the hydrogen atom
bound to nitrogen, such as in compound 8e, resulted in an appar-
ent increase in activity compared to compound 8a. In this series of
alkylamine-substituted benzothiazinones, the substituents at 6-
position of benzene ring had little effect, although the nitro-substi-
tuted compounds were more hydrophilic.
We then converted the spiroketal group of compound 1 to three
simplified substituents (pyrrolidine, piperidine and piperidine,
respectively). The activities of relevant compounds 8h–8j were sig-
nificantly weaker than 1. Compounds 8h (ClogP 2.84) and 8i (ClogP
played a MIC of 0.0063 lM, which was superior to 1. Based on
these MIC results, it seems that compound 8o was more selective
to Mtb strain H37Rv compared to 1.
Most of the compounds were screened for cytotoxicity using
Vero cells.16 In general, these compounds were nontoxic, suggest-
ing that their anti-TB activity was not due to general toxicity. And
the selectivity index (SI) of 8o was >1,000,000 (IC50 divided by the
MIC against Mtb H37Rv), which suggested a superior safety and
tolerance in in vivo studies. In addition, there was no death in an
acute single dose toxicity study in BALB/c mice and all the mice be-
haved normally when administering compound 8o by gavage at a
dose of 2 g/kg17 (data not shown).
3.39) showed a MIC of 0.8 lM, about 4-fold lower than 8j (ClogP
2.01) and as effective as INH. It revealed that substituents of the
heterocycles were well tolerated. And then, we introduced a lipo-
philic methyl and a hydrophilic carboxylic acid group to the para
position of piperidine, respectively. The more lipophilic compound
Due to the orthologous enzyme of Rv3790 (DprE1) possesses an
alanine rather than a cysteine, which is critical to the action of
compound 1, M. avium is naturally resistant to 1. Thus, compound
8o was selected to determine its activity against M. avium. As ex-
pected, compound 8o did not exhibit obvious activity against M.
avium (Table 2). It suggested that 8o retain the mode of action of
compound 1.
Physicochemical properties (LogP) of 8o determined by a shake
flask method18 (Table 2) and pharmacokinetic properties were
investigated in Sprague Dawley (SD) rats by single oral (p.o.) and
intravenous (i.v.) administration (Table 3). Compound 8o exhibited
an elimination t1/2 of 2.5 h (i.v.) and 5.4 h (p.o.). The Cmax obtained
8k (ClogP 3.91) showed good potency profile (MIC = 0.1
the contrary, for compound 8l (ClogP 1.88), the replacement of
methyl by carboxylic acid group led to loss of activity
(MIC = 3.1 M). The lack of activity may due to the highly hydro-
lM). On
a
l
philic carboxylic acid group, which may prevent its penetration
through the cell wall of M. tuberculosis. Next, a tert-butoxycarbonyl
substituted piperidine group wasintroduced. Surprisingly, the Boc-
protected piperidine analog 8m exhibited an enhanced activity with
a MIC of 0.02
lM. It revealed that the volume of the substitutents
by oral dosing was 0.56
l
g/ml (1.25
lM), 12,560 times higher than
may be also a crucial factor for the activity. But the related com-
pound 7f containing a nitro group did not show fairly good potency
profile. Together with the fact that compound 7e showed a slightly
reduced activity compared to 8k, it suggested that the substituents
at 6-position of benzene ring played a significant role when a large
group was introduced to the terminal heterocycle. With the above
SAR data, we then replaced one oxygen atom of spiroketal moiety
by a sulfur to give compound 8n (ClogP 2.39), which showed a
its in vitro MIC level (0.0001
i.v. AUC of 8o in plasma over time resulted in an oral bioavailability
of 26%.
l
M). The comparison of p.o. AUC with
Table 2
Anti-mycobacterial activity and physicochemical properties of 8o
Compd
MIC (
lM)
LogPc
M. smegmatisa
M. aviumb
MIC of 0.002 lM, 2-fold more active than 1 (ClogP 2.45). Next, when
another sulfur atom was introduced, compound 8o exhibit an evi-
dent 4-fold increase in potency compared to 1.
1
8o
0.013
0.0063
74
36
2.69
3.94
A selection of the more interesting compounds was assayed to
determine their detailed anti-bacterial activities. As shown in Ta-
ble 1, compounds 8i, 8k, 8m, 8n and 8o also exhibited remarkable
a
b
c
Mycobacterium smegmatis mc2 155.
M. avium ATCC 25291.
Determined by a shake flask method with HPLC.
Table 3
Pharmacokinetic parameters and oral bioavailability of 8o
O
S
F3C
N
N
S
S
NO2
8o
Route
Dose (mg/kg)
Cmax
(
lg/ml)
Tmax (h)
t1/2 (h)
AUC0–1
(
lg h/ml)
Fa (%)
Intravenous
Oral
10
20
1.72
0.56
—
0.5
2.5
5.4
1.48
0.78
—
26
a
Oral bioavailability.
Gao, Chao
