Studies on acyclic pyrimidines as inhibitors of mycobacteria

Article abstract of DOI:10.1016/j.bmc.2006.12.032

In vitro anti-mycobacterial activities of several 5-substituted acyclic pyrimidine nucleosides containing 1-(2-hydroxyethoxy)methyl and 1-[(2-hydroxy-1-(hydroxymethyl) ethoxy)methyl] acyclic moieties are investigated against three mycobacteria viz. Mycoba

Full text of DOI:10.1016/j.bmc.2006.12.032

Bioorganic & Medicinal Chemistry 15 (2007) 2045–2053
Studies on acyclic pyrimidines as inhibitors of mycobacteria
a
a
b
a,
*
Naveen C. Srivastav, Tracey Manning, Dennis Y. Kunimoto and Rakesh Kumar
a
Department of Laboratory Medicine and Pathology, 1-41 Medical Sciences Building, University of Alberta,
Edmonton, AB, Canada T6G 2H7
Department of Medicine, Faculty of Medicine and Dentistry, University of Alberta, Edmonton, AB, Canada T6G 2H7
b
Received 3 November 2006; revised 15 December 2006; accepted 21 December 2006
Available online 23 December 2006
Abstract—In vitro anti-mycobacterial activities of several 5-substituted acyclic pyrimidine nucleosides containing 1-(2-hydroxyeth-
oxy)methyl and 1-[(2-hydroxy-1-(hydroxymethyl) ethoxy)methyl] acyclic moieties are investigated against three mycobacteria viz.
Mycobacterium tuberculosis, Mycobacterium bovis, and Mycobacterium avium, which cause serious infections and mortality in
healthy people as well as patients with AIDS. 1-(2-Hydroxyethoxy)methyl-5-(1-azido-2-haloethyl or 1-azidovinyl) analogs (4–7),
1
-[(2-hydroxy-1-(hydroxymethyl)ethoxy)methyl]-5-decynyluracil (37), and 1-[(2-hydroxy-1-(hydroxymethyl)ethoxy)methyl]-5-dode-
cynyluracil (38) exhibited significant in vitro anti-tubercular activity against these mycobacteria.
2006 Elsevier Ltd. All rights reserved.
ꢀ
1
. Introduction
Two groups of mycobacteria, M. tuberculosis and
M. avium, pose a significant threat of TB in HIV-infect-
ed patients and are often responsible for their deaths.
4
At the present time, nearly two billion people worldwide
are infected with the tubercle bacillus, and the preva-
lence of active tuberculosis (TB) is increasing.
Clinical management of Mycobacterium avium complex
(MAC) infections is very difficult, because many of the
first-line anti-tuberculosis drugs are ineffective against
1
–3
The
latent infection in many of these individuals may reacti-
vate sometime later in life. Tubercle bacillus remains
contained in the presence of effective cellular immunity,
however, immunocompromised status such as human
immunodeficiency virus (HIV) infection, cancer chemo-
therapy, or use of immunosuppressive drugs in trans-
plantation provides the single most significant factor in
reactivation of the latent TB leading to full clinical
5
,6
it. New macrolides, such as clarithromycin and azith-
romycin, are used for the treatment of MAC, however,
fast development of resistance with macrolide therapy
7
,8
poses a significant medical challenge.
Therefore,
intensive research efforts are required to develop new
agents for the treatment of TB and multidrug resistant
TB (MDR-TB).
1
–3
disease.
Our recent studies showed that novel 5-(C-1 substituted)
alkyl side chains at the C-5 position of pyrimidine
nucleosides play a crucial role in contributing to their
Chemotherapy with anti-tuberculosis drugs such as
streptomycin, p-aminosalicylic acid, isoniazid, and rif-
ampicin revolutionized TB therapy in 1970s, and there
was a rapid decline in tuberculosis in many developed
countries. However, there is now an ever increasing
threat of drug-resistant TB appearing as an epidemic
in many countries, particularly because no new classes
of TB-specific drugs have been developed since the
9
anti-mycobacterial properties. We reported that 5-(1-
hydroxyethyl)-(1a) and 5-(1-fluoro-2-haloethyl)-(1b,c)
0
2 -deoxyuridines possess significant in vitro anti-myco-
9
bacterial activity against M. avium and M. bovis. These
studies prompted us to investigate and resynthesize
0
related 5-(1-hydroxy-2-haloethyl)-(1d–f) 2 -deoxyuri-
3
rifampicin.
dines to evaluate their anti-mycobacterial activity. We
also resynthesized the OAc derivatives (1g–i) of com-
pounds 1d–f to determine if the anti-mycobacterial
activity is influenced by the incorporation of the acetyl
groups. Unfortunately, compounds 1d–i were found to
be non-inhibitory against M. avium, M. bovis or
M. tuberculosis up to a concentration of 100 lg/mL
(unpublished results, R. Kumar et al).
Keywords: Tuberculosis; Heterocycle; Pyrimidine; Anti-mycobacterial
agent.
*
Corresponding author. Tel.: +1 780 492 7545; fax: +1 780 492
521; e-mail: rakesh.kumar@ualberta.ca
7
0
968-0896/$ - see front matter ꢀ 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2006.12.032

2046
N. C. Srivastav et al. / Bioorg. Med. Chem. 15 (2007) 2045–2053
O
N
1a, R = CH(OH)CH , R = OH
1i, R = CH(OH)CH2I, R1 = OAc
3
1
1
1
j, R = CH(N )CH Br, R = OH
3 2 1
1
1
1
1
1
1
1
b, R = CH(F)CH Br, R = OAc
2
1
HN
R
k, R = CH(N )CH Cl, R = OH
c, R = CH(F)CH Cl, R = OAc
3
2
1
2
1
O
1l, R = CH(N3)CH2I, R1 = OH
d, R = CH(OH)CH Br, R = OH
2
1
R1
O
1
1
1
m, R = CH(N )=CH , R = OH
e, R = CH(OH)CH Cl, R = OH
3
2
1
2
1
n, R = C=C-(CH ) CH , R = OH
f, R = CH(OH)CH I, R = OH
2 7
3
1
2
1
o, R = C=C-(CH ) CH , R = OH
g, R = CH(OH)CH Br, R = OAc
2 9
3
1
2
1
R1
1
h, R = CH(OH)CH Cl, R = OAc
2
1
As part of investigation to design more effective anti-
mycobacterial agents, an azido moiety located at C-1
position of 5-alkyl side chain of pyrimidine nucleosides
attracted our attention as a novel pharmacophore. The
azido substituent possesses physicochemical properties
relative to other substituents used routinely in struc-
ture–activity correlationship studies, such as electronic
In an effort to design novel anti-mycobacterial
nucleosides and further explore the structure–activity
relationships, in the present communication, we
synthesized and evaluated anti-mycobacterial activities
of 5-(1-azido-2-haloethyl)-, 5-(1-azidovinyl)-, 5-alkynyl,
as well as several other 5-substituted pyrimidine nucleo-
sides in which the furanose moiety is replaced with
different acyclic moieties. Interestingly, 1-(2-hydroxyeth-
oxy)methyl-5-(1-azido-2-bromoethyl)uracil (4), 1-(2-
hydroxyethoxy)methyl-5-(1-azido-2-iodoethyl)uracil
(5), 1-(2-hydroxyethoxy)methyl-5-(1-azido-2-chloroeth-
yl)uracil (6), 1-(2-hydroxyethoxy)methyl-5-(1-azidovi-
nyl)uracil (7), 1-[(2-hydroxy-1-(hydroxymethyl)ethoxy)-
methyl]-5-decynyluracil (37), and 1-[(2-hydroxy-1-
(hydroxymethyl)ethoxy)methyl]-5-dodecynyluracil (38)
were found to be significantly inhibitory to M. bovis,
M. tuberculosis, and M. avium replication. To our
knowledge, 5-substituted acyclic pyrimidine nucleosides
have not previously been reported to possess anti-myco-
bacterial activity against M. bovis, M. tuberculosis, and
M. avium.
(
inductive effect, F value), steric (molecular refractive
0
1
value), and lipophilic effect (p-value) viz: N , 0.30,
3
1
0
0.2, 0.46; F, 0.43, 0.92, 0.14; Cl, 0.41, 6.03, 0.71; Br,
.44, 8.88, 0.86; I, 0.40, 13.94, 1.12; OH, 0.29, 2.85,
ꢀ
0.67; Me, ꢀ0.04, 5.65, 0.56; H, 0.00, 1.03, 0.00, respec-
tively. Thus, the electronic effect of azido is between that
of OH and Cl, the steric effect is between that of Br and
I, and the lipophilic effect is between that of F and Me.
These physical data suggest an azido group may be a
good isostere of halogens and OH. In recent studies,
0
we observed that 5-(1-azido-2-haloethyl) analogs of 2 -
deoxyuridine (1j–l) possessed weak anti-mycobacterial
activity against M. avium, whereas 5-(1-azidovinyl)-2 -
deoxyuridine (1m) exhibited significant activity for M.
avium (MIC50 = 1–5 lg/mL). In another study, we
reported that 5-alkynyl-2 -deoxyuridines also exhibit
0
1
1
0
notable in vitro anti-mycobacterial activity. We ob-
0
2. Chemistry
served that 2 -deoxyuridines with 5-decynyl (1n) and
5-dodecynyl (1o) terminal alkynes at the C-5 positions
were important determinants of anti-mycobacterial
The target compounds, 1-(2-hydroxyethoxy)methyl-5-
(1-azido-2-haloethyl) uracils (4–6), were synthesized
by the reactions of 1-(2-hydroxyethoxy)methyl-5-vinyl-
uracil (3) with N-bromo(or chloro)succinimide or
iodine monochloride, and sodium azide at 25–45 ꢁC
1
2
activity.
Modified nucleosides have acquired an important role as
therapeutic agents in the treatment of patients with dev-
astating cancer and viral infections. One promising class
of nucleoside analogs for anti-microbial chemotherapy
belongs to a group in which the cyclic carbohydrate
moiety is replaced with open-chain ‘acyclic’ sugar moie-
ties such as 9-(2-hydroxyethoxymethyl)guanine (acyclo-
1
8
using procedures reported by us earlier (Scheme 1).
Reaction of benzoylated derivative of 4, 1-[(2-benzoyl-
oxyethoxy)methyl]-5-(1-azido-2-bromoethyl)uracil, with
t-BuOK in THF followed by debenzoylation using
methanolic ammonia provided the 5-(1-azidovinyl)
1
9
analog (7) (Scheme 1). Similarly, 1-[(2-hydroxy-1-
(hydroxymethyl) ethoxy)methyl]-5-(1-azido-2-haloeth-
yl)uracil derivatives (9–11) were synthesized by the
regiospecific addition of halogenoazides (XN3:
X = Br, Cl, I) to the vinyl substituent of 1-[(2-
hydroxy-1-(hydroxymethyl)ethoxy)methyl]-5-vinyluracil
1
3
vir). The effectiveness of acyclic nucleoside analogs as
a substrate or inhibitor is likely dependent on the ability
of the acyclic side chain to mimic the interaction of the
glycosyl portion of the natural substrate with the en-
zyme. The flexibility of the acyclic chain may allow it
to adopt a conformation favorable for enzyme interac-
tion as substrates or inhibitors. Further, it has been
demonstrated that acyclic derivatives of pyrimidine
nucleosides possess in vivo stabilization against phos-
1
7,18
(8).
The reaction of 1-[(2-benzoyloxy-1-(benzoyl-
oxymethyl)ethoxy)methyl]-5-(1-azido-2- bromoethyl)
uracil with t-BuOK in THF yielded the 5-(1-azidovinyl)
compound that was subsequently deprotected using a
saturated solution of NH in MeOH to obtain 5-(1-azi-
3
dovinyl)-analog of 1-[(2-hydroxy-1-(hydroxymethyl)
19
ethoxy)methyl]uracil (12).
1
4,15
0
phorolysis.
can be rapidly catabolized to the corresponding pyrimi-
In contrast, 2 -deoxyuridine analogs
1
6,17
dine bases by the action of bacterial phosphorylases.

N. C. Srivastav et al. / Bioorg. Med. Chem. 15 (2007) 2045–2053
2047
O
N
O
N
O
N
O
N
HN
R
HN
I
HN
CH=CH2
HN
R
i
O
O
O
ii
O
HO
HO
O
O
HO
O
HO
O
OH
2
3
8
9
1
1
, R = CH=CH2
4
5
6
, R = CH(N )CH Br
3 2
, R = CH(N )CH Br
, R = CH(N )CH I
3
2
3
2
0, R = CH(N )CH I
, R = CH(N )CH Cl
3
2
3
2
1, R = CH(N )CH Cl
3
2
O
O
N
O
HN
CH(N )CH Br
3 2
HN
C(N3)=CH2
HN
3
C(N )=CH2
O
N
iii, iv
BzO
O
O
O
N
HO
HO
O
O
R1
R = H or CH OBz
7
OH
1
2
1
2
Scheme 1. Reagents and conditions: (i) Pd(OAc)
N-chlorosuccinimide (6 and 11), sodium azide, DME–water, 0 ꢁC; iodine monochloride, sodium azide, MeCN (5 and 10); (iii) t-Bu-OK, THF,
ꢁC; (iv) methanolic ammonia, 0–25 ꢁC.
2 3
, Ph P, triethylamine, vinyl acetate, 70 ꢁC; (ii) N-bromosuccinimide (4 and 9) or
0
Acyclic pyrimidines 22–30 were obtained by a single and
convenient route using iodomethyl [(trimethylsi-
lyl)oxy]ethyl ether, prepared in situ from 1,3-dioxolane
and trimethyl chlorosilane, instead of acetoxyethyl acet-
yl]-uracil (26) with 1-heptyne. Bycyclic compounds 32,
33, 35 were obtained in one-pot reaction by the Pd-cat-
alyzed coupling reaction of respective terminal alkynes
and aryl acetylene with 26, followed by cyclization of
the intermediate 5-alkynyluracils with copper iodide
and triethylamine in methanol. Reaction of 26 with
1-dodecyne also produced a di-addition compound,
1-[(2-hydroxyethoxy)methyl]-5,N -didodecynyluracil (34),
3
unexpectedly. 5-Decynyl-(37) and 5-dodecynyl- (38)
pyrimidine analogs were prepared by the treatment of
2
0
oxymethyl ether as alkylating reagent. Thus, uracils
3–21 silylated with bis (trimethylsilyl)acetamide in dry
1
acetonitrile were reacted with trimethylchlorosilane,
potassium iodide, and 1,3-dioxolane at room tempera-
ture for 16–24 h to yield the desired compounds 22–30
in 34–97% yields (Scheme 2).
5-iodo-1-[(2-hydroxy-1-(hydroxymethyl)ethoxy)methyl]-
5-Alkynyl acyclic pyrimidine (31) was synthesized by the
coupling reaction of 5- iodo-1-[(2-hydroxyethoxy)meth-
uracil (36) with 1-decyne and 1-dodecyne terminal alky-
nes as described for compounds 31 and 32. Cyclization
X
X
XSiMe3
N
R1
HN
O
R1
ii, iii
HN
R1
i
O
N
O
N
Me SiO
N
HO
H
13-21
3
22-30
1
1
1
1
1
1
1
2
2
3 = R
1
1
1
1
1
1
1
1
1
= H, X = O
= F, X = O
= Cl, X = O
= Br, X = O
= I, X = O
22 = R
23 = R
24 = R
25 = R
26 = R
27 = R
28 = R
29 = R
30 = R
1
1
1
1
1
1
1
1
1
= H, X = O
= F, X = O
= Cl, X = O
= Br, X = O
= I, X = O
4 = R
5 = R
6 = R
7 = R
8 = R
9 = R
0 = R
1 = R
= CH
= C
3
, X = O
, X = S
= CH , X = O
2 5
= C H , X = S
3
2
H
5
= NO
= NH
2
, X = O
, X = O
= NO
= NH
2
, X = O
, X = O
2
2
Scheme 2. Reagents and conditions: (i) bis(trimethylsilyl)acetamide, dry acetonitrile; (ii) 1,3-dioxolane, KI, trimethyl chlorosilane, 25 ꢁC;
iii) quenched with MeOH and neutralized with NaHCO
(
3
.

2048
N. C. Srivastav et al. / Bioorg. Med. Chem. 15 (2007) 2045–2053
O
N
O
R
C
I
C
HN
O
i
HN
O
O
O
N
HO
HO
2
6
31, R = (CH ) CH
2
4
3
ii
R
O
N
R
O
R
O
C
C
C
C
R1
N
O
R1
N
O
N
O
O
O
N
O
N
HO
HO
HO
3
4, R = (CH ) CH ,
2 9 3
3
3
3
2, R = (CH ) CH
2
4
3
R = C=C-(CH ) CH
3
3, R = (CH ) CH
3
1
2 9
2
9
5, R = C H
6
5
Scheme 3. Reagents and conditions: (i) Tetrakis(triphenylphosphine)palladium (0), CuI, (i-Pr)
nylphosphine) palladium (0), CuI, (i-Pr)
reflux.
2
EtN, DMF, 25 ꢁC, 1-heptyne; (ii) tetrakis(triphe-
EtN, DMF, 25 ꢁC, 1-heptyne (32), 1-dodecyne (33 and 34), phenylacetylene (35), triethylamine, dry MeOH,
2
of 5-alkynyl uracils (37 and 38) in presence of copper io-
dide and triethylamine in dry methanol at reflux temper-
ature yielded compounds 39 and 40 (Scheme 3).
sponding 5-(1-azidovinyl) analog (7) provided com-
pound that retained broad spectrum anti-myco-
bacterial activity similar to compound 6. In contrast,
a 5-vinyl pyrimidine, with 1-[(2-hydroxyethoxy)methyl]
substituent (3), was not found to be inhibitory to any
of the mycobacteria. These results suggest that azido
substituent at the C-1 position can be determinant
of anti-mycobacterial activity in this series of com-
pounds. In comparison of the anti-mycobacterial
activity of acyclic compound 7 with its corresponding
3
. Results and discussion
Acyclic pyrimidine nucleosides (3–12 and 22–40) were
evaluated in vitro against M. bovis, M. tuberculosis
(
H37Ra), and M. avium by the microplate Alamar blue
0
assay (MABA) at 1–100 lg/mL concentrations. Rifam-
picin and clarithromycin were used as reference stan-
dards. The results are summarized in Table 1. In these
studies, 5-(1-azido-2-haloethyl) derivatives 4–6, with a
2 -deoxyribose analog (1m), we further note that 7
was active against all of the three mycobacteria tested
1
1
while 1m was only inhibitory against M. avium.
1
-[(2-hydroxyethoxy)methyl] moiety, emerged as inhibi-
tors of M. bovis, M. tuberculosis, and M. avium where
-(2-hydroxyethoxy)methyl-5-(1-azido-2-chloroethyl)ura-
The above results prompted us to study the related 5-(1-az-
ido-2-haloethyl)- (9–11) and 5-(1-azidovinyl)-(12) uracil
analogs possessing a different 1-[(2-hydroxy-1-(hydroxy-
methyl) ethoxy)methyl] acyclic chain in order to determine
their anti-mycobacterial activity and identify structure–
activity correlations. Surprisingly, compounds 9–12
did not inhibit mycobacterial replication, whereas the
1
cil (6) exhibited significant broad spectrum anti-tubercu-
lar activity against all three Mycobacteria investigated
(MIC50 = 10 lg/mL). Interestingly, compounds 4–6
were found to exhibit broad spectrum anti-mycobacteri-
al activity in contrast to their corresponding cyclic 2 -de-
oxyribose analogs (1j–l) that were only active for
M.avium (MIC50 = 50 lg/mL). Further, a significant
shift in the anti-mycobacterial activity of acyclic com-
pound 6 was noted as compared to its 2 -deoxyribose
analog (1k) which was devoid of anti-mycobacterial
activity against all three mycobacteria. Acyclopyrimi-
dines 4–6 also showed improved activity in inhibiting
the growth of M. bovis than 5-(1-halo) substituted ana-
0
0
corresponding 2 -deoxyribose analogs (1j–l and 1m) pos-
sessed activity against M. avium (MIC = 50 and 1–5 lg/
mL, respectively).
5
0
1
1
11
0
Substituents at the 5-position of the pyrimidine nucleo-
sides such as halogens, methyl, ethyl, nitro, and amino
have played important role in their biological proper-
1
1
2
1
ties. Based upon these observations, it was of interest
to examine the anti-mycobacterial activity of selected
acyclic pyrimidine nucleoside analogs (22–30) contain-
ing 1-[(2-hydroxyethoxy)methyl] substituent in our
cell-based assays of anti-mycobacterial replication.
Interestingly, none of these analogs showed inhibitory
effect against M. bovis, M. tuberculosis, and M. avium
0
logs, 5-(1-fluoro-2-haloethyl)-2 -deoxyuridines (1b,c),
9
where 0–25% inhibition was obtained at 50 lg/mL.
Modification of the 5-(1-azido-2-haloethyl) substituent
in the acyclic pyrimidine nucleosides 4–6 to the corre-

N. C. Srivastav et al. / Bioorg. Med. Chem. 15 (2007) 2045–2053
2049
Table 1. In vitro anti-mycobacterial activity of 5-substituted acyclic pyrimidine nucleosides against M. bovis, M. tuberculosis, and M. avium
R
O
O
N
O
R
O
N
O
CH3(CH2)9-C C
HN
R
HN
R
N
O
N
O
R
N
O
O
O
N
O
N
O
N
HO
O
HO
HO
O
HO
O
HO
OH
OH
3
-7, 22-31
32,33,35
34
8-12, 36-38
39, 40
a
Compound
R
Anti-microbial activity % inhibition (concentration lg/mL)
M. bovis (BCG)
M. tuberculosis (H37Ra)
M. avium (ATCC 25291)
3
4
5
6
7
8
9
CH@CH
2
0
100 (50)
0
100 (100, 50), 50 (10)
0
100 (50)
CH(N
3
3
3
3
)CH
2
2
2
Br
CH(N
CH(N
CH(N
)CH
)CH
I
Cl
50 (100, 50)
100 (100, 50) 50 (10)
100 (100, 50), 50 (10)
100 (100, 50), 50 (10)
50 (100, 50)
100 (100, 50) 50 (10)
)@CH
2
50 (100, 50, 10)
0
100 (100, 50), 25 (10)
0
50 (100, 50, 10)
0
CH@CH
2
CH(N
CH(N
CH(N
CH(N
H
3
3
3
3
)CH
)CH
)CH
2
2
2
Br
I
0
0
0
10
11
12
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
0
0
0
Cl
0
0
0
)@CH
2
0
0
0
0
0
0
F
Cl
0
0
0
0
0
0
Br
I
0
0
0
0
0
0
CH
3
0
0
0
C
H
2 5
0
0
0
NO
NH
2
0
50 (100,50)
0
0
2
0
0
CCꢁ(CH
2
)
4
CH
3
3
0
50 (100)
0
25 (100)
0
(CH
(CH
2
)
)
4
CH
CH
3
0
2
9
3
0
0
0
CCꢁ(CH
2
)
9
CH
0
0
0
C
I
H
6 5
0
0
0
0
0
0
CC„(CH
CC„(CH
2
)
)
7
CH
CH
3
100 (100,50), 60 (10)
100 (100,50), 80 (10), 50 (1)
90 (100), 50 (50)
90 (100, 50), 50 (10)
100 (1)
100 (100,50), 50 (10)
100 (100,50), 50 (10)
75 (100), 50 (50)
100 (100), 75 (50), 50 (10)
100 (1)
100 (100), 25 (50)
100 (100), 25 (50)
50 (100, 50)
25 (100)
2
9
3
(CH
(CH
—
2
)
7
CH
3
2
)
9
CH
3
b
Rifampicin
Clarithromycin
90 (2)
95 (2)
b
c
—
ND
ND
a
b
c
Anti-mycobacterial activity was determined at concentrations 100, 50, 10, and 1 lg/mL.
Positive control drugs.
ND, not determined.
up to 100 lg/ml concentration except 1-[(2-hydroxyeth-
oxy)methyl]-5-aminouracil (30) that exhibited moderate
oxy)methyl]-5-heptynyluracil (31) and 1-[(2-hydroxyeth-
oxy)methyl]-5,N -didodecynyluracil (34) were devoid of
3
anti-mycobacterial activity against M. bovis (MIC50
5
with an azido group at C-1 position are impor-
tant for broad spectrum anti-mycobacterial activity
=
activity suggesting acyclic moiety, 1-(2-hydroxyeth-
oxy)methyl, present at the N-1 position of 31 and 34 is
detrimental to anti-tuberculosis activity. Encouragingly,
anti-mycobacterial activity of 37 and 38 against M. bovis
and M. avium was improved as compared to their corre-
0 lg/mL). These studies suggest that 5-alkyl substituents
(
Scheme 4).
0
sponding 2 -deoxyuridine analogs [1n, MIC = 50 lg/
5
0
Among the 5-alkynyl acyclic pyrimidine nucleosides (31,
4, 37, 38), 1-[(2-hydroxy-1-(hydroxymethyl)ethoxy)-
methyl]-5-decynyluracil (37) and 1-[(2-hydroxy-1-
hydroxymethyl) ethoxy)methyl]-5-dodecynyluracil (38)
provided significant inhibition of all three mycobacteria
tested, M. bovis (MIC50 = 10 and 1 lg/mL, respectively),
M. tuberculosis (MIC50 = 10 lg/mL), and M. avium
mL (M. bovis), MIC50 = >100 lg/mL (M. avium)], [1o,
3
MIC50 = >1 lg/mL (M. bovis), MIC = >50 lg/mL
50
1
2
(M. avium)]. This observation was similar to that we
noted earlier with acyclic pyrimidines 4–6 in contrast
to their corresponding 2 -deoxyuridine analogs. Surpris-
(
0
ingly, modification of the 5-alkynyl substituent in the
acyclopyrimidines (37 and 38) to the corresponding
bicyclic analogs (39 and 40) provided compounds with
(
MIC50 = 25 lg/mL). In contrast, 1-[(2-hydroxyeth-

2050
N. C. Srivastav et al. / Bioorg. Med. Chem. 15 (2007) 2045–2053
R
O
R
O
N
C
O
I
C
HN
O
HN
ii
i
N
O
O
N
O
N
O
HO
HO
O
HO
OH
OH
OH
3
6
3
3
7, R = (CH ) CH
2 7
3
39, R = (CH ) CH
2 7
40, R = (CH ) CH
2 9
3
3
8, R = (CH ) CH
2
9
3
Scheme 4. Reagents and conditions: (i) Tetrakis(triphenylphosphine)palladium (0), CuI, (i-Pr) EtN, DMF, 25 ꢁC, 1-decyne (37), 1-dodecyne (38); (ii)
2
CuI, triethylamine, dry MeOH, reflux.
similar or slightly diminished anti-mycobacterial activity
against all of the three mycobacteria tested, whereas
gel column chromatography was carried out using
Merck 7734 silica gel (100–200 lM particle size). Thin-
layer chromatography was performed with Machery-
Nagel Alugam SiL G/UV silica gel slides (20 lM
thickness).
0
corresponding bicyclic derivatives in the 2 -deoxyuridine
1
2
series were devoid of any anti-mycobacterial activity.
These results suggest that an acyclic glycosyl moiety,
-(2-hydroxy-1-(hydroxymethyl)ethoxy)methyl, also con-
1
tributes to the superior anti-mycobacterial activity.
4.1. 1-[(2-Hydroxyethoxy)methyl]-5-heptynyluracil (31)
The precise mechanism of action of the active com-
pounds inhibiting mycobacterial multiplication in this
study is not clear. The complete genome sequence of
M. tuberculosis has been deciphered. It encodes many
of the enzymes required for DNA and RNA synthesis,
and pyrimidine and purine nucleoside biosynthesis. It
is postulated that active compounds may be inhibiting
mycobacterial DNA and/or RNA synthesis, by acting
as substrates and/or inhibitors of metabolic enzymes of
DNA and/or RNA synthesis.
To a stirred solution of 1-[(2-hydroxyethoxy)methyl]-5-
iodouracil (26, 200 mg, 0.64 mmol) in anhydrous
dimethylformamide (15 mL), tetrakis(triphenylphos-
phine)palladium (0) (74.07 mg, 0.064 mmol), copper (I)
iodide (24.5 mg, 0.13 mmol), diisopropylethylamine
(0.17 mL, 1.0 mmol), and 1-heptyne (0.18 mL,
1.37 mmol) were added. The reaction mixture was stir-
red at room temperature under nitrogen atmosphere;
the progress of the reaction was monitored by TLC in
MeOH/EtOAc (1:99, v/v). After 18 h, 15 drops of 5%
2
2
of disodium salt of EDTA/H O were added to the reac-
2
The compounds 3–12 and 22–40 were also evaluated
for their anti-bacterial activities against several
Gram-positive and Gram-negative (Staphylococcus
aureus, Staphylococcus epidermis, Enterococcus feacal-
is, Bacillus subtilis, Streptococcus pneumoniae, Salmo-
nella typhimurium, Escherichia coli, Proteus vulgaris,
Pseudomonas aeruginosa) bacteria. However, none of
these compounds exhibited any activity at concentra-
tions up to 100 lg/mL suggesting specific action of
the most active 5-substituted acyclic pyrimidines (4–7
and 37–40) against mycobacteria. The promising com-
pounds 4–7 and 37–40 were tested in vitro for their
toxicity against Vero cells and HepG2 cells up to
tion mixture, and the mixture was concentrated in
vacuo. The residue obtained was purified on silica
gel column using EtOAc as an eluent to yield 31
1
(50 mg, 28%) as a syrup. H NMR (DMSO-d ) d 0.97
6
(t, 3H, J = 6.96 Hz, CH ), 1.33–1.59 (m, 6H, 3· CH ),
3
2
2.46 (t, 2H, J = 6.96 Hz, a-CH ), 3.55–3.58 (m, 4H,
2
OCH CH O), 4.77 (t, J = 5.13 Hz, 1H, OH), 5.18
(s, 2H, NCH ), 8.10 (s, 1H, H-6), 11.68 (br s, 1H,
2
2
2
NH). Anal. Calcd for C H N O : C, 60.0; H, 7.14;
4
1
20
2
4
N, 10.0. Found: C, 60.38; H, 7.0; N, 10.33.
4.2. 3-[(2-Hydroxyethoxy)methyl]-6-pentyl-2,3-dihydro-
furo-[2,3-d]pyrimidin-2-one (32)
1
observed.
00 lg/mL concentration, and no toxicity was
To a stirred solution of 1-[(2-hydroxyethoxy)methyl]-5-
iodouracil (26, 400 mg, 1.28 mmol) in anhydrous
dimethylformamide (25 mL), tetrakis(triphenylphos-
phine)palladium (0) (150 mg, 0.13 mmol), copper (I)
iodide (49 mg, 0.26 mmol), diisopropylethylamine
(0.33 mL, 1.89 mmol), and 1-heptyne (0.37 mL,
2.82 mmol) were added. The reaction mixture was stir-
red at room temperature under nitrogen atmosphere.
After 20 h, reaction mixture was cooled down and cop-
per (I) iodide (43 mg, 0.23 mmol), triethylamine (9 mL),
and dry methanol (15 mL) were added. The reaction
mixture was then heated to reflux and stirred for 4 h.
4. Experimental
Melting points were determined with a Buchi capillary
apparatus and are uncorrected. H NMR spectra were
determined for solutions in Me SO-d or CDCl on a
2
1
6
3
Bruker AM 300 spectrometer using Me Si as an internal
4
standard. The assignment of all exchangeable protons
OH, NH) was confirmed by the addition of the D O.
2
(
Micro-analyses were within ±0.4% of theoretical values
for all elements listed, unless otherwise indicated. Silica
Fifteen drops of 5% of disodium salt of EDTA/H O
2

N. C. Srivastav et al. / Bioorg. Med. Chem. 15 (2007) 2045–2053
2051
were added to the reaction mixture, and the mixture was
concentrated in vacuo. The residue thus obtained was
redissolved in MeOH and filtered to remove the inor-
ganic impurities. The obtained filtrate was concentrated
and crude product was purified on silica gel column by
H-5), 7.44–7.53 (m, 3H, aromatic), 7.83–7.85 (m, 2H,
aromatic), 8.68 (s, 1H, H-4). Anal. Calcd for
C H N O : C, 62.93; H, 4.89; N, 9.79. Found: C,
1
5
14
2
4
63.27; H, 5.27; N, 10.1.
using CHCl /MeOH (96:4) as eluent to give 32
3
4.5. 1-[(2-Hydroxy-1-(hydroxymethyl)ethoxy)methyl]-5-
decynyluracil (37)
1
(
108 mg, 30%) as a syrup. H NMR (DMSO-d ) d 0.87
6
(
t, 3H, J = 6.41 Hz, CH ), 1.28–1.33 (m, 4H, 2· CH ),
3
2
1
CH2), 3.45–3.57 (m, 4H, OCH CH O), 4.68 (t,
.59–1.64 (m, 2H, b-CH ), 2.64 (t, 2H, J = 7.32 Hz, a-
To
a
stirred solution of 5-iodo-1-[(2-hydroxy-1-
2
(hydroxymethyl)ethoxy)methyl]-uracil (36, 200 mg,
0.58 mmol) in anhydrous dimethylformamide (20 mL),
tetrakis(triphenylphosphine) palladium(0) (67.5 mg,
0.06 mmol), copper (I) iodide (22.27 mg, 0.12 mmol),
diisopropyl ethylamine (0.15 mL, 0.86 mmol), and 1-de-
cyne (0.24 mL, 1.33 mmol) were added. The reaction
mixture was stirred at room temperature under nitrogen
atmosphere; the progress of the reaction was monitored
by TLC in MeOH/EtOAc (1:9, v/v). After 18 h, 15 drops
2
2
J = 5.19 Hz, 1H, OH), 5.34 (s, 2H, NCH ), 6.44 (s,
2
1
H, H-5), 8.45 (s, 1H, H-4). Anal. Calcd for
C H N O : C, 60.0; H, 7.14; N, 10.0. Found: C,
1
4
20
2
4
6
0.37; H, 7.46; N, 10.29.
4.3. 3-[(2-Hydroxyethoxy)methyl]-6-decyl-2,3-dihydro-
furo-[2,3-d]pyrimidin-2-one (33) and 1-[(2-hydroxyethoxy)-
methyl]-5,N -didodecynyluracil (34)
3
of 5% of disodium salt of EDTA/H O were added to the
2
To a stirred solution of 1-[(2-hydroxyethoxy)methyl]-5-
iodo-uracil (26, 400 mg, 1.28 mmol) in anhydrous
dimethylformamide (25 mL), tetrakis(triphenylphos-
phine)palladium (0) (150 mg, 0.13 mmol), copper (I)
iodide (49 mg, 0.26 mmol), diisopropylethylamine
reaction mixture, and then the mixture was concentrated
in vacuo. The residue obtained was purified on silica gel
column using an initial eluent of EtOAc, followed by an
eluent of EtOAc/MeOH (98:2, v/v). The appropriate
fractions were combined and the solvent was removed
in vacuo to yield 37 (60 mg, 29%) as a syrup.
1
(
0.33 mL, 1.89 mmol), and 1-dodecyne (0.64 mL,
.99 mmol) were added. The reaction mixture was stir-
H
2
NMR (DMSO-d ) d 0.85 (t, 3H, J = 7.02 Hz, CH ),
6
3
red at room temperature under nitrogen atmosphere;
the progress of the reaction was monitored by TLC in
MeOH/CHCl (5:95, v/v). After 18 h, reaction mixture
1.21–1.50 (m, 12H, 6· CH ), 2.35 (t, 2H, J = 7.32 Hz,
2
a-CH ), 3.29–3.53 (m, 5H, OCH CH(O)CH O), 4.62
2
2
2
(t, J = 5.5 Hz, 2H, 2· OH), 5.16 (s, 2H, NCH ), 7.96
3
2
was cooled down. To the same reaction mixture, copper
(
(s, 1H, H-6), 11.53 (br s, 1H, NH). Anal. Calcd for
C H N O : C, 61.36; H, 7.95; N, 7.95. Found: C,
61.63; H, 8.19; N, 7.88.
I) iodide (43 mg, 0.23 mmol), triethylamine (9 mL), and
1
8
28
2
5
dry methanol (15 mL) were added. The reaction mixture
was then heated to reflux and stirred for 4 h. Fifteen
drops of 5% of disodium salt of EDTA/H O were added
4.6. 1-[(2-Hydroxy-1-(hydroxymethyl)ethoxy)methyl]-5-
dodecynyluracil (38)
2
to the reaction mixture, and the mixture was concentrat-
ed in vacuo. The residue thus obtained was redissolved
in MeOH and filtered to remove the inorganic impuri-
ties. Filtrate was concentrated and crude product was
Compound 38 was prepared using the procedures as de-
1
scribed for 37. Yield (74 mg, 33%); H NMR (CDCl ) d
3
purified on silica gel column by using CHCl /MeOH
3
0.88 (t, 3H, J = 7.02 Hz, CH ), 1.27–1.58 (m, 16H, 8·
3
as eluent to give compounds 33 and 34 as syrups. Com-
pound 34, eluent CHCl /MeOH (98:2); yield (60 mg,
CH ), 2.37 (t, 2H, J = 7.33 Hz, a-CH ), 3.63–3.82 (m,
2
2
5H, OCH CH(O)CH O), 5.29 (br s, 2H, NCH ), 7.54
3
2 2 2
1
9
1
(
%); H NMR (DMSO-d ) d 0.85 (m, 6H, 2· CH ),
(s, 1H, H-6), 9.0–9.50 (br s, 1H, NH). Anal. Calcd for
C H N O : C, 63.15; H, 8.42; N, 7.36. Found: C,
62.89; H, 8.36; N, 7.76.
6
3
.22–1.68 (m, 32H, 16· CH ), 2.26 (m, 2H, CH ), 2.72
2
2
20 32
2
5
m, 2H, CH ), 3.45–3.58 (m, 4H, OCH CH O), 4.66
2
2
2
(
Anal. Calcd for C H N O : C, 72.37; H, 9.72; N,
m, 1H, OH), 5.38 (s, 2H, NCH ), 8.55 (s, 1H, H-6).
2
4.7. 3-[(2-Hydroxy-1-(hydroxymethyl)ethoxy)methyl-6-
octyl-2,3-dihydrofuro-[2,3-d]pyrimidin-2-one (39)
31
50
2
4
5
eluent CHCl /MeOH (96:4); yield (120 mg, 27%); H
.44. Found: C, 72.09; H, 9.38; N, 5.8. Compound 33,
1
3
NMR (DMSO-d ) d 0.85 (m, 3H, CH ), 1.22–1.32 (m,
To a stirred solution of 37 (200 mg, 0.57 mmol) in
methanol/triethylamine (7:3) (30 mL), copper (I) iodide
(25 mg, 0.13 mmol) was added at room temperature
under a nitrogen atmosphere. The reaction mixture
was then heated to reflux and stirred for 3 h. The sol-
vent was removed in vacuo. The solid thus obtained
was redissolved in MeOH and filtered to remove the
inorganic impurities. Filtrate was concentrated and
purification was conducted on two columns, the first
6
3
1
4H, 7· CH ), 1.58–1.64 (m, 2H, b-CH ), 2.61–2.66
2
2
(
m, 2H, a-CH ), 3.44–3.58 (m, 4H, OCH CH O), 4.66
2 2 2
(
8
m, 1H, OH), 5.35 (s, 2H, NCH ), 6.42 (s, 1H, H-5),
.45 (s, 1H, H-4). Anal. Calcd for C H N O : C,
19 30 2 4
2
6
5.14; H, 8.57; N, 8.0. Found: C, 65.44; H, 8.82; N, 7.69.
4.4. 3-[(2-Hydroxyethoxy)methyl]-6-phenyl-2,3-dihydro-
furo-[2,3-d]pyrimidin-2-one (35)
using an eluent CHCl /MeOH (85:15) and the second
3
Compound 35 was prepared by using the procedures as
described for 33, except that aryl acetylene was used as
an eluent of CHCl /MeOH (9:1). The appropriate
3
fractions were combined and the solvent was removed
H
1
1
terminal alkyne. Yield (80 mg, 22%);
H
NMR
in vacuo to yield 39 (66 mg, 33%) as a syrup.
NMR (DMSO-d ) d 0.85 (t, 3H, J = 7.33 Hz, CH ),
(
J = 5.5 Hz, 1 H, OH), 5.39 (s, 2H, NCH ), 7.31 (s, 1H,
DMSO-d ) d 3.47–3.61 (m, 4 H, OCH CH O), 4.72 (t,
6
2
2
6
3
1.23–1.64 (m, 12H, 6· CH2), 2.63 (t, 2H,
2

2052
N. C. Srivastav et al. / Bioorg. Med. Chem. 15 (2007) 2045–2053
J = 7.02 Hz, a-CH ), 3.36–3.62 (m, 5H, OCH CH(O)-
ments, the M wells gave OD of 3000–4000, and the B
wells had OD values 60,000–100,000.
2
2
CH O), 4.64 (t, J = 5.19 Hz, 2H, 2· OH), 5.42 (s,
2
2
Anal. Calcd for C H N O : C, 61.36; H, 7.95; N,
H, NCH ), 6.44 (s, 1H, H-5), 8.49 (s, 1H, H-4).
2
4.10. In vitro anti-bacterial activity assay
1
8
28
2
5
7
.95. Found: C, 61.0; H, 8.24; N, 8.21.
A total of nine bacterial strains were used for the deter-
mination of the in vitro anti-bacterial activity of the
studied compounds. The in vitro anti-bacterial activity
was studied by determining their minimum inhibitory
concentrations (MICs) by means of the broth microdilu-
tion method. Briefly, exponentially growing bacteria
were diluted in a liquid sterile medium to obtain a final
4.8. 3-[(2-Hydroxy-1-(hydroxymethyl)ethoxy)methyl-6-
decyl-2,3-dihydrofuro-[2,3-d]pyrimidin-2-one (40)
Compound 40 was prepared by the cyclization of 38 as
described above for 39, except that the refluxing time
1
was 4 h. Yield (35 mg, 29%); H NMR (DMSO-d ) d
6
4
0
.84 (t, 3H, J = 7.02 Hz, CH ), 1.23–1.32 (m, 14H, 7·
inoculum of 1 · 10 CFU/mL and subsequently cultured
3
CH ), 1.57–1.65 (m, 2H, b-CH ), 2.63 (t, 2H,
with varying dilutions of compounds for 16–20 h. The
minimum inhibitory concentrations were defined as the
lowest concentration at which bacterial growth was no
longer evident.
2
2
J = 7.33 Hz, a-CH ), 3.36–3.61 (m, 5H, OCH CH(O)-
2
2
CH O), 4.64 (m, 2H, 2· OH), 5.42 (s, 2H, NCH ),
2
2
6
C H N O : C, 63.15; H, 8.42; N, 7.36. Found: C,
.44 (s, 1H, H-5), 8.49 (s, 1H, H-4). Anal. Calcd for
2
0
32
2
5
6
3.43; H, 8.75; N, 7.59.
Acknowledgment
4.9. In vitro anti-mycobacterial activity assay (M. bovis,
M. tuberculosis, M. avium)
This work was supported by the Canadian Institutes of
Health Research (CIHR) operating grant (MOP-49415)
to R.K.
Mycobacterium bovis (BCG), M. tuberculosis (H Ra),
3
7
and M. avium (ATCC 25291) were obtained from the
American Type Culture Collection, Rockville, MD.
The anti-mycobacterial activity was determined using
the Microplate Alamar blue assay (MABA). Test com-
pounds were dissolved in DMSO at 100· of the highest
final concentration used and subsequent dilutions were
performed in 7H9GC (Difco Laboratories, Detroit,
Michigan) media in 96-well plates. For these experi-
ments, each compound was tested at 100, 50, 10, and
References and notes
1
2
. Pilheu, J. A. Int. J. Tuberc. Lung Dis. 1998, 2, 696–703.
. Raviglione, M. C.; Snider, D. E.; Kochi, A. J. Am. Med.
Assoc. 1995, 273, 220–226.
3. (a) Willcox, P. A. Curr. Opin. Pulm. Med. 2000, 6, 198–
202; (b) M e´ ndez, A. P.; Raviglione, M. C.; Laszlo, A.;
Binkin, N.; Rieder, H. L.; Bustreo, F.; Cohn, D. L.;
Weezenbeek, C. S. B. L-van.; Kim, S. J.; Chaulet, P.;
Nunn, P. N. Eng. J. Med. 1998, 338, 1641–1649.
1 l/mL in triplicate. The experiments were repeated
three times and the mean percent inhibition is reported
in the table. The standard deviations were within 10%.
Frozen mycobacterial inocula were diluted in medium
4
. Pozniak, A. J. HIV Ther. 2002, 7, 13–16.
5. Inderlied, C. B.; Kemper, C. A.; Bermudez, L. E. Clin.
Microbiol. Rev. 1993, 6, 266–310.
5
7
H9GC and added to each well at 2.5 · 10 CFU/mL fi-
nal concentration. Sixteen control wells consisted of 8
with bacteria alone (B) and 8 with media alone (M).
Plates were incubated for an initial 6 days and starting
from 6 days of incubation, 20 ll of 10· Alamar blue
and 12.5 ll of 20% Tween 80 were added to one M
and one B well. Wells were observed for 24–48 h for
visual color change from blue to pink and read by spec-
trophotometer (at excitation 530/525 and emission 590/
6
. Falkinham, J. O., III Clin. Microbiol. Rev. 1996, 9, 177–
15.
2
7. Dautzenberg, B. Res. Microbiol. 1994, 145, 197–206.
8. Ellner, J. J.; Goldberger, M. J.; Parenti, D. M. J. Infect.
Dis. 1991, 163, 1326–1335.
9
. Johar, M.; Manning, T.; Kunimoto, D. Y.; Kumar, R.
Bioorg. Med. Chem. 2005, 13, 6663–6671.
1
1
1
0. Hansch, C.; Leo, A.; Unger, S. H.; Kim, K. H.; Nikaitani,
D.; Lien, E. J. J. Med. Chem. 1973, 16, 1207–1216.
1. Srivastav, N. C.; Manning, T.; Kunimoto, D. Y.; Kumar,
R. Med. Chem. 2006, 2, 287–293.
2. Rai, D.; Johar, M.; Manning, T.; Agrawal, B.; Kunimoto,
D. Y.; Kumar, R. J. Med. Chem. 2005, 48, 7012–7017.
13. Chu, C. K.; Cutler, S. J. J. Heterocyc. Chem. 1986, 23,
289–319.
14. Freeman, S.; Gardiner, J. M. Mol. Biotechnol. 1996, 5,
5
35) to determine OD values. If the B well became pink
by 24 h (indicating growth), reagent was added to the
entire plate. If B well remained blue, additional M and
B wells were tested daily until bacterial growth could
be visualized by color change. After the addition of
the reagent to the plate, cultures were incubated for
2
4 h and plates were observed visually for color change
1
5. de Miranda, P.; Good, S. S.; Krasny, H. C.; Connor, J. D.;
25–137.
and also read by spectrophotometer. Visual MIC was
defined as the lowest concentration of a compound that
prevented a color change from blue to pink. Percent
inhibition was calculated as (test well-M bkg./B well-
M bkg.) · 100. Similar methodology was used for M.
bovis BCG, M. tuberculosis, and M. avium. Rifampicin
and clarithromycin were used as positive controls. As
negative controls, DMSO was added to the B well at
concentration similar to that of compound wells, M
wells served as negative controls. In most of the experi-
1
Laskin, O. L.; Lietman, P. S. Am. J. Med. 1982, 73, 215–
2
20.
6. Balzarini, J.; Morin, K. W.; Knaus, E. E.; Wiebe, L. I.;
De Clercq, E. Gene Ther. 1995, 2, 317–322.
1
1
7. Balzarini, J.; Sienaert, R.; Liekens, S.; Van Kuilenburg,
A.; Carangio, A.; Esnouf, R.; De Clercq, E.; McGuigan,
C. Mol. Pharmacol. 2002, 61, 1140–1145.
18. Kumar, R.; Sharma, N.; Nath, M.; Saffran, H.; Tyrrell,
D. L. J. J. Med. Chem. 2001, 44, 4225–4229.

N. C. Srivastav et al. / Bioorg. Med. Chem. 15 (2007) 2045–2053
2053
1
2
2
9. Kumar, R.; Nath, M.; Tyrrell, D. L. J. J. Med. Chem.
2
0. Kumar, R.; Semaine, W.; Johar, M.; Tyrrell, D. L. J.;
Agrawal, B. J. Med. Chem. 2006, 49, 2049–2054.
1. (a) De Clercq, E. Methods Find Exp. Clin. Pharmacol.
Barry, C. E.; Tekala, F.; Badcock, K.; Bashman, D.;
Brown, D.; Chillingworth, T.; Connor, R.; Davies, R.;
Devlin, K.; Feltwell, T.; Gentles, S.; Hamlin, N.; Holroyd,
S.; Hornsby, T.; Jagels, K.; Krogh, A.; Mclean, J.; Moule,
S.; Murphy, L.; Oliver, K.; Osborne, J.; Quail, M. A.;
Rajandream, M.-A.; Rogers, J.; Rutter, S.; Seeger, K.;
Skelton, J.; Squares, R.; Squares, S.; Sulston, J. E.; Taylor,
K.; Whitehead, S.; Barrell, B. G. Nature 1998, 393, 537–
544.
002, 45, 2032–2040.
1
1
980, 2, 253–267; (b) Kulikowski, T. Pharm. World Sci.
994, 16, 127–138.
2
2. Cole, S. T.; Brosch, R.; Parkhill, J.; Garnier, T.; Churcher,
C.; Harris, D.; Gordon, S. V.; Eiglmeier, K.; Gas, S.;