Synthesis of methyl 5-S-alkyl-5-thio-d-arabinofuranosides and evaluation of their antimycobacterial activity

Article abstract of DOI:10.1016/j.bmc.2008.03.062

The emergence of drug resistant tuberculosis necessitates a search for new antimycobacterial compounds. The antigen 85 (ag85) complex is a family of mycolyl transferases involved in the synthesis of trehalose-6,6′-dimycolate and the mycolated hexasacchari

Full text of DOI:10.1016/j.bmc.2008.03.062

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 5672–5682
Synthesis of methyl 5-S-alkyl-5-thio-D-arabinofuranosides
and evaluation of their antimycobacterial activity
Aditya K. Sanki, Julie Boucau, Parijat Srivastava, Samuel S. Adams,
Donald R. Ronning and Steven J. Sucheck*
Department of Chemistry, The University of Toledo, 2801 W. Bancroft Street, MS602, Toledo, OH 43606, United States
Received 21 January 2008; revised 22 March 2008; accepted 25 March 2008
Available online 30 March 2008
Abstract—The emergence of drug resistant tuberculosis necessitates a search for new antimycobacterial compounds. The antigen 85
(ag85) complex is a family of mycolyl transferases involved in the synthesis of trehalose-6,60-dimycolate and the mycolated hexa-
saccharide motif found at the terminus of the arabinogalactan in mycobacterium. Enzymes involved in the synthesis of cell wall
structures like these are potential targets for the development of new antiinfectives. To potentially inhibit the ag85 complex, methyl
5-S-alkyl-5-thio-arabinofuranoside analogues were designed based on docking studies with ag85C derived from Mycobacterium
tuberculosis. The target arabinofuranosides were then synthesized and the antibacterial activity evaluated against Mycobacterium
smegmatis ATCC 14468. Two of the compounds, 5-S-octyl-5-thio-a-D-arabinofuranoside (8) and 5-S-octyl-5-thio-b-D-arabinofu-
ranoside (11), showed MICs of 256 and 512 lg/mL, respectively. Attempts to directly evaluate acyltransferase inhibitory activity
of the arabinofuranosides against ag85C are also described. In conclusion, a new class of antimycobacterial arabinofuranosides
has been discovered.
Published by Elsevier Ltd.
1. Introduction
associated with the outer leaflet. It is the mycolates that
are credited for producing the hydrophobic character of
the cell envelope.
Mycobacterium, including the human pathogen Myco-
bacterium tuberculosis, possesses a cell wall that differs
significantly in structure from both Gram-negative and
Gram-positive bacteria.1 The unique hydrophobic prop-
erties of this envelope protect the bacterium from its
environment and provides a barrier to the diffusion of
commonly used hydrophilic antimicrobial agents.2 The
main structural element of the mycobacterial cell wall
is the mycolyl-arabinogalactan-peptidoglycan (mAGP)
complex.1 The arabinan portion of this structure con-
tains a 1,3-branched arabinofuranoside-based hexasac-
charide motif that is approximately two-thirds
mycolated via ester linkages at each of the four of the
primary hydroxyls to form the mycolyl-arabinan do-
main (1, Fig. 1a).3,4 The mycolyl moieties are high
molecular weight a-alkyl, b-hydroxy fatty acids (classi-
fied as a-, methoxy-, and keto- mycolates, Fig. 1b) that
form the inner leaflet of the outer lipid bilayer. Mycolyl
esters are also found as glycoconjugates non-covalently
Many antimycobacterial drugs work by interfering with
the biosynthesis of key components of the cell wall.5–7
This suggests that the other enzymes involved in the syn-
thesis of cell wall structures may serve as targets for anti-
mycobacterial drug development. Antigens 85A, B and
C,8 represent a family of homologous mycolyltransferas-
es that are responsible for the synthesis of trehalose-6,60-
dimycolate (TDM) from trehalose-6-monomycolate
(TMM)9,10 and for transfer of mycolic acids from
TMM to the arabinogalactan (AG).2c,11 In light of the
role ag85 plays in mycobacterial cell wall synthesis, we
have initiated studies to identify inhibitors of the ag85
complex based on substrate analogues of the arabinan.
The significance of these studies is heightened as a result
of the resurgence in cases of TB, in particular, those of
the multi-drug resistant variety.1b,12,13
An elaborated mechanism for mycolyl transfer to the
arabinan by ag85 is shown in Figure 2 based on the
work by Ronning et al.14 The key element of the mech-
anism is the attack on the carbonyl carbon of TMM
by Ser124. Trehalose leaves the binding site, but the
Keywords: Carbohydrates; Arabinofuranosides; Antigen 85; Mycobac-
terium tuberculosis; Carbohydrate-based antibiotics.
*
Corresponding author. Tel.: +1 419 530 1504; fax: +1 419 530
1990; e-mail: Steve.Sucheck@UToledo.edu
0968-0896/$ - see front matter Published by Elsevier Ltd.
doi:10.1016/j.bmc.2008.03.062

A. K. Sanki et al. / Bioorg. Med. Chem. 16 (2008) 5672–5682
TMM + Ag85
5673
HO
O
Mycolyl
Mycolyl
O
HO
O
O
O
O
O
Glu228
OH
OH
HO
HO
HO
O
OH
OH
O
TMM
HO
O
OH
O
H
N
O
O
TDM + Ag85
OH
O
O
O
Arabinan
Mycolyl
Mycolyl
O
OH
O
N
O
His260
O
HO
trehalose
H
O
OH
O
O
Ser124
HO
O
Mycolyl
COOH
(CH2)22CH3
AGP
α:
CH3(CH2)17
(CH2)10
(CH2)17
(CH2)17
(CH2)19
(CH2)18
cis
cis
OH
COOH
(CH2)22CH3
trehalose
O
H3C OMe
CH3(CH2)17
methoxy:
(CH2)10
(CH2)10
Glu228
cis /
trans
OH
COOH
(CH2)22CH3
OH
AGP
H3C
CH3(CH2)17
O
mAGP + Ag85
N
His260
OH
cis
N
OH
O
O
H
keto:
COOH
(CH2)22CH3
OH
O
Base:
CH3
H3C
CH3(CH2)17
O
H
(CH2)10
O
trans
Ser124
OH
O
Mycolyl
Figure 1. (a) The terminal arabinofuranosyl motif [b-D-Araf-(1 ! 2)-
a-D-Araf]2-3,5-a-D-Araf-(1 ! 5)-a-D-Araf displayed at the non-reduc-
ing end of the arabinogalactan. Each terminal O-5 hydroxyl is a
potential site for mycolyl ester formation. (b) The structures of the
three classes of mycolic acids found in Mycobacterium tuberculosis.
Figure 2. Proposed mechanism of mycolyl moiety (shown in red)
transfer to TMM or AGP to form TDM and mAGP, respectively.
Residues of the catalytic triad found in M. tuberculosis antigen 85C:
Ser124, His260, and Glu228 are shown.
mycolyl moiety remains as part of an acyl enzyme inter-
mediate. A second molecule of TMM, or presumably
the terminus of the AGP, enters the substrate-binding
pocket, which we refer to as carbohydrate-binding pock-
et I, and the mycolyl group is transferred producing
TDM or mAGP, respectively. A hydrophobic tunnel
adjacent to the catalytic site has also been proposed to
accommodate the a-chain of the mycolate ester. A more
extensive interfacial model for the mechanism has been
developed by Anderson et al. using ag85B.15 Aspects
of this model highlight the possible role of a second car-
bohydrate-binding site (carbohydrate-binding pocket II,
Fig. 3) and a hydrophobic channel which may assist the
entry of the a-chain into the hydrophobic tunnel.16
Using these studies as a guide we propose a model for
the tetrahedral intermediate resulting from the attack
of the terminal arabinofuranoside primary hydroxyl on
the ag85 acyl enzyme intermediate (Fig. 4) analogous
to the one proposed for the TMM-ag85 intermediate.14
This model is consistent with X-ray studies that show a
substrate-binding pocket large enough to accommodate
two residues and an alkyl chain.
logues of ag85 suggests that a single inhibitor will
likely inhibit all members of the ag85 family.14,15,17 Fi-
nally, known inhibitors of ag85 have been shown to
compromise the cell wall, which can allow chemother-
apy to be more effective.8,18 Attempts to produce inhib-
itors of ag85 are limited, but some compounds show
promise as leads. These include 6-azido-6-deoxytreha-
lose which also inhibits the growth of Mycobacterium
aurum.8 Alkyl phosphonates19 designed to mimic the
tetrahedral intermediate of ag85C also show activity
against Mycobacterium avium and inhibit mycolyltrans-
ferase activity20 with IC50s in the low micromolar range.
TDM mimics that contain long hydrocarbon chains
have displayed strong activity against Mycobacterium
smegmatis when combined with INH and are
presumed to inhibit ag85,21 while related compounds
consisting of 6,60-bis(sulfonamido) and N,N0-dialkyla-
mino derivatives of trehalose were active against
M. tuberculosis and M. avium.22 Since ag85 enzymes
are believed to recognize the terminal arabinofuranoside
motif of the AGP we envisioned that compounds con-
taining an arabinofuranoside and an alkyl component
could act as substrate analogues. The rationale is illus-
trated in Figure 4, which shows that compounds linked
through the primary carbon of an arabinofuranoside
potentially overlap a portion of tetrahedral intermediate
formed during mycolic acid transfer. A thioether was
Additional factors that favor ag85 as a target for the
development of antimycobacterials are dealt with by Be-
lisle et al. who show that the enzyme complex is required
for the viability of mycobacterium.8 Further, the con-
served nature of the catalytic site of all three homo-

5674
A. K. Sanki et al. / Bioorg. Med. Chem. 16 (2008) 5672–5682
carbohydrate
carbohydrate
binding pocket I
binding pocket II
Ser126
OH
Phe232
O
O
HO OH
O
HO
O
HO
HO
HO
OH
O
TMM
carbohydrate
carbohydrate
Ser126
O
Phe232
binding pocket I
binding pocket II
OH
HO OH
OH
HO
O
OH
O
HO
O
HO
HO
O
HO
O
HO
HO
HO
HO
O
O
O
OH
O
trehalose
TMM
Figure 3. Selected steps of the interfacial model (Ref. 15) proposed for ag85. (a) TMM bound to the second carbohydrate-binding site with the a-
chain located in the hydrophobic channel (parallel lines) of ag85, the a-chain is shown entering the hydrophobic tunnel (dashed circle). (b) The ag85
acyl-enzyme intermediate, with Phe232 covering the hydrophobic channel. The meromycolate branch of the mycolate (red) remains buried in the
hydrophobic cell wall. Note: residue numbering is taken from the M. tuberculosis ag85B.
of the xenobiotic nature of the arabinofuranoside moi-
ety which would minimize any potential non-specific
activity in a mammalian system.23
HN-Leu40
R
O
HN-Met125
2. Results
O
Ser124
O
hydrophobic
side chains
HO
2.1. Synthesis of 5-S-alkyl-5-thio-D-arabinofuranosides
α-chain
binding tunnel
HO
A synthetic strategy was developed to rapidly access a
library of 5-S-alkyl-modified a/b-D-arabinofuranosides
6–11 (Scheme 1). The synthesis was initiated by subject-
ing commercially available D-arabinose with in situ gen-
erated hydrochloric acid by means of addition of acetyl
chloride (0.22 N HCl) to anhydrous methanol, followed
by heating at 90 ꢀC (Scheme 1, Method A, Section 5) in
an attempt to generate the methyl furanosides, 3a and
3b.24 Under these glycosylation conditions, the reaction
was complete within 0.5 h. However, along with methyl
arabinofuranoside 3a (20% yield), an undesired mixture
of methyl arabinopyranosides 2 (64% yield), b/a = 2.3:1,
1H NMR (600 MHz, DMSO-d6 + D2O), da 3.97
(3JHꢀ1,Hꢀ2 = 6.6 Hz, H-1), db 4.52 (3JHꢀ1,Hꢀ2 = 3.0 Hz,
H-1), was also formed. The lack of the b-methyl
arabinofuranoside was a disappointment since we had
speculated the b-anomer would be a more potent inhib-
itor of ag85, being structurally similar to the b-linked
O
O
carbohydrate
binding pocket I
OH
O
O
AG
OH
Figure 4. A model of the tetrahedral intermediate resulting from the
attack of the terminal arabinofuranoside primary hydroxyl on the acyl
enzyme intermediate of ag85. AG = arabinogalactan. R = meromyco-
late branch of a mycolic acid. The 5-S-alkyl-5-thio-D-arabinofurano-
sides inhibitors were designed based on a minimal structural domain
defined by the atoms and bonds in bold.
chosen to conveniently link the arabinose and alkyl
components to form 5-S-alkyl-5-thio-D-arabinofurano-
sides. The proposed ag85 inhibitors also take advantage

A. K. Sanki et al. / Bioorg. Med. Chem. 16 (2008) 5672–5682
5675
inofuranoside 4 in 60% yield.27 Similarly, known
tosylate 5 was obtained in 50% yield.27 In both cases
the triplets assigned to primary hydroxyl disappeared
after tosylation. 5-O-Tosylate 4 was subjected to nucle-
ophilic displacement with several n-alkanethiols (C4, C6,
and C8) in the presence of sodium methoxide in anhy-
drous DMF to afford the desired 5-S-alkyl derivatives
6–8 in 92%, 50%, and 98% yields, respectively. Under
similar thiolation condition, 5-O-tosylate 5 generated
thioethers 9–11 in 73%, 60%, and 73% yields, respec-
tively. All compounds were characterized by 1H and
HO
HO
HOOMe
HO
OH
O
O
O
HO
+
+
OMe
OMe
OH
OH
OH
3β
3α
2
c
c
a, b
TsO
TsO
HOOMe
HO
OH
O
O
OH
4
O
OH
5
HO
OH
OMe
OH
13C NMR, H–1H gCOSY, and HRMS. JHꢀ1,Hꢀ2 val-
ues of the desired a-series 6–8 (0 Hz) and b-series 9–11
(4.8 Hz) are in line with those reported in the literature
as previously mentioned.26 Similarly, d(C-1) values for
compounds 6–8 (109.0–109.1 ppm) and 9–11 (102.12–
102.16 ppm) lie in the range reported for a- and b-arabi-
nofuranosides, respectively.26
1
3
1
RS
RS
HOOMe
OH
d
d
O
O
OMe
OH
OH
9 R = C4H9
10 R = C6H13
11 R = C8H17
6 R = C4H9
7 R = C6H13
8 R = C8H17
2.2. Screening of the compounds for antimycobacterial
activity
Scheme 1. Reagents and conditions: (a) MeOH, 0.22 N CH3COCl,
90 ꢀC, 0.5 h; (b) i—MeOH, HCl, 0–4 ꢀC, 2.5 days; ii—MeOH,
CH3COCl (0.8 equiv), 0–4 ꢀC, 18 h; iii—purified by gravity column
chromatography on silica gel using acetone/MeOH/CHCl3 (1:1:8); (c)
TsCl, Py, rt, 1.5 days; (d) appropriate alkane thiol, NaOMe, DMF, 1–
3.5 h, rt.
Arabinofuranosides 6–11 were screened for their ability
to inhibit the growth of M. smegmatis ATCC 14468
using a Kirby–Bauer disk diffusion assay.28 INH was
used as the positive control while DMSO, also the dilu-
ent, was used as a negative control. M. smegmatis was
used as a fast growing, non-pathogenic surrogate for
M. tuberculosis.29 The strain has been previously used
to screen TDM mimics and oxazolidinone congeners,
in lieu of M. tuberculosis.21,30 The diameters of the zones
of inhibition (DZIs) are summarized in Table 1. Similar
DZIs were obtained when the strains were grown using
both Lennox and Middlebrook 7H10-OADC + 0.5%
glycerol agars.
arabinofuranoside found at the terminus of the arabino-
galactan. Thus, in order to access the desired quantities
of diastereomerically pure a/b methyl arabinofurano-
sides, we resorted to the conditions reported by Lowary
and others,25 who also accessed the furanosides, 3a/3b,
in a ratio of 1:1 using hydrochloric acid in anhydrous
methanol at 0 ꢀC. Using those conditions, the desired
kinetically favored furanosides 3a and 3b were obtained
in 85% yield in a 1.4:1 ratio. Although the conditions
were effective at delivering the desired products, the
reaction required 2.5 days to reach 85% conversion.
The reaction time was shortened to 18 h by using acetyl
chloride (0.8 equiv) in anhydrous methanol, instead of
concentrated hydrochloric acid. The latter conditions
afforded 3a and 3b in a combined yield of 89% on a
10.5 g scale, again with the a-anomer formed in excess
of the b (Method B, Section 5). Formation of the fur-
anosides was confirmed by 1H NMR, where characteris-
tic anomeric protons of 3a and 3b appeared at d 4.58
with coupling constants of 2.0 and 4.4 Hz in DMSO-
d6 solvent, respectively. These coupling constants are
consistent with related a- and b-arabinofuranosides
In order to evaluate the selectivity of the compounds for
killing mycobacterium and obtain quantitative values
for the growth inhibition, we determined the minimum
inhibitory concentrations (MICs) of glycosides 6–1131
against Bacillus subtilis ATCC 6633, M. smegmatis
ATCC 14468, and Escherichia coli ATCC 23722. Serial
dilutions ranging from 0.25 lg/mL to 1.024 mg/mL were
used to determine MICs in comparison to INH. The
MICs are summarized in Table 2.
Table 1. Kirby–Bauer assay of glycosides 6–11 against Mycobacterium
smegmatis ATCC 14468a
Compound
DZI (mm) (20 mg/mL)
DZI (mm) (10 mg/mL)
3
where the JHꢀ1,Hꢀ2 of a-arabinofuranosides ranged
6
None
None
17.5
None
None
14.5
from 0–3.0 Hz and d(C-1) = 104–110 ppm and the
3JHꢀ1,Hꢀ2 for the b-isomers ranged from 3.0–5.0 Hz
and d(C-1) = 97–104 ppm.26 Further evidence for the
formation of the furanosides was provided by the triplet
centered at d 4.71 in 3a and d 4.61 in 3b arising from the
primary hydroxyl at C-5 (RCH2–OH), which makes the
furanosides straightforward to distinguish from the pyr-
anosides, which show doublets for all of the hydroxyls
(R2CH–OH) in DMSO-d6. The primary hydroxyl of
3a was selectively tosylated using tosyl chloride in anhy-
drous pyridine to yield known methyl 5-O-tosyl-D-arab-
7
8
9
10
None
None
13.5
None
None
7.5
11
DMSO
INH
None
23.0 (500 lg/mL) (regrowth noted)
a Solutions of 20 mg/mL and 10 mg/mL for compounds 6–11 and
500 lg/mL for INH were prepared in DMSO, 10 lL of each solution
was then applied to a susceptibility disk. DZI = diameter (in mm) of
the zone of inhibition.

5676
A. K. Sanki et al. / Bioorg. Med. Chem. 16 (2008) 5672–5682
Table 2. MICs (lg/mL) of glycosides 6–11 against a panel of bacteria
Compound M. smegmatis 14468 B. subtilis 6633 E. coli 23722
pected, no inhibitory activity against ag85C was noted
over a concentration range of 1–20 mM (Fig. 5).
6
>1024
1024
256
>1024
>1024
128
>1024
>1024
>1024
>1024
>1024
>1024
Not
7
3. Discussion
8
9
>1024
1024
512
>1024
>1024
512
The ag85 complex is currently considered a potential
target for the treatment of mycobacterial infection.8,19–22
In light of that, we have initiated efforts to develop carbo-
hydrate-based inhibitors of this enzyme. To date, the
design of these inhibitors has focused on the synthesis of
compounds related to trehalose, TMM, TDM, and small
molecules designed to mimic the aliphatic, disaccharide,
and tetrahedral intermediate domains present during
TDM synthesis. Based on the observed reduction of cell
wall bound mycolates after inhibition of ag85C and com-
plementation studies in Corynebacterium glutamicum, it
has been inferred that the terminus of the AG is a sub-
strate for ag85 complex. Therefore, we hypothesized that
arabinofuranosides which possess a C-5 aliphatic moiety
could be used as a template for the design of substrate-
based inhibitors of the ag85 complex. Our initial targets
were simple methyl glycosides, which were accessed by a
non-stereoselective kinetic glycosylation to afford a separa-
ble mixture of the a/b-methyl arabinofuranosides on a
preparative scale. During the course of the synthesis of
10
11
INH
2
Not
determined
determined
2.3. Screening of the compounds for antigen 85C inhib-
itory activity
Attempts to evaluate the inhibitory activity of the glyco-
sides against ag85C proved to be troublesome. Com-
pounds 7,
8
and 10, 11 which showed
antimycobacterial activity in the disk and broth dilution
assays were dissolved in DMSO and added to a newly
developed high-throughput enzymatic assay. The assay
evaluates the mycolyltransferase activity of ag85C, from
M. tuberculosis, using UV–vis absorption spectroscopy.
A glucose-based substrate analogue was designed and
synthesized with a chromophoric p-nitrophenol moiety
as a signaling molecule. The assay couples the catalytic
activities of ag85C and b-glucosidase to release the chro-
mophore, p-nitrophenolate. The rate of p-nitropheno-
late release is observed by direct measurement of the
absorbance at 405 nm over time. By comparing the slope
of the absorbance versus time curves, it is possible to
evaluate the inhibitory activity of these compounds on
ag85C. Unfortunately, under a variety of conditions
evaluated the lipophilic compounds immediately precip-
itated upon addition to the aqueous assay system. Fur-
thermore, increasing DMSO to a concentration required
to keep the compounds in solution was found to inhibit
ag85C activity. Compound 9, which showed no activity
in the antimicrobial assay, stayed in solution and could
be evaluated using the mycolyltransferase assay. As ex-
1
the arabinofuranosides we found acquiring the H NMR
spectra in dry DMSO-d6 and examining the C-5 hydroxyl
splitting to be an unambiguous aid in distinguishing furan-
oside from pyranoside isomers.
In our evaluation of the antimycobacterial activity of
5-S-alkyl-5-thio-D-arabinofuranosides, we made the
assumption that M. smegmatis contained ag85 homo-
logues sufficiently similar to ag85 from M. tuberculosis
that it could serve as a surrogate in a growth inhibition
assay. The presence of ag85A and ag85C in the closely
related M. smegmatis (mc2155) was identified using the
microbial genome database from the National Center
for Biotechnology Information. A multiple-sequence
Figure 5. Absorbance versus time curve for compound 9 at different concentrations (1–20 mM).

A. K. Sanki et al. / Bioorg. Med. Chem. 16 (2008) 5672–5682
5677
alignment was performed between full-length ag85C M.
tuberculosis (H37Rv) and ag85C and ag85A from M.
smegmatis (mc2155).32 Ag85C from M. tuberculosis
(H37Rv) was found to be 75.7% identical and 83.5%
similar to ag85C from M. smegmatis (mc2155). When
ag85C from M. tuberculosis (H37Rv) was compared
to ag85A from M. smegmatis (mc2155), the sequences
were 62.1% identical and 75.8% similar (Fig. 6). The
residues representing the catalytic triad (red) are con-
served for the three proteins compared. The carbohy-
lose-based antimycobacterial agents making compari-
son of the results straightforward.21
A structure–activity relationship emerged on screening
the
5-S-alkyl-5-thio-D-arabinofuranosides
against
M. smegmatis. Both the a and b glycosides with the C8
alkyl chains displayed the lowest MICs, 256 and
512 lg/mL, respectively. The C6 alkyl homologue also
displayed activity at the highest concentration tested.
The slightly higher activity for the a-anomer was sur-
prising since the natural linkage of the terminal arabino-
furanoside is a b-linkage. The results demonstrate a
clear relationship between chain length and antimyco-
bacterial activity. Longer alkyl chains were not
employed since solubility was anticipated to be an issue.
A hypothetical model for the interaction of arabinofu-
ranoside 11 with ag85C is shown in Figure 7. The model
was generated by manually overlaying the octyl chain of
11 onto the octyl chain of the octylthioglucoside present
in the crystal structure of 1DQY.14 It can be seen that
the octyl chain and carbohydrate moiety are easily
accommodated by the substrate-binding site of the
drate-binding pocket
I of M. smegmatis ag85A
exhibits changes at two residues (blue) when compared
to the M. tuberculosis ag85C proteins. However, these
changes are conservative and should have little effect
on binding site structure or electrostatics. Having
established that M. smegmatis possesses ag85 suffi-
ciently similar to M. tuberculosis to draw a potential
correlation between killing and ag85 inhibition, we
chose M. smegmatis 14468 as a non-pathogenic surro-
gate for M. tuberculosis to evaluate the structure–activ-
ity relationships of our antibacterial agents. This strain
has been used previously for the evaluation of treha-
M. tuberculosis Ag85C (H37Rv)
M. smegmatis Ag85C (mc(2)155)
M. smegmatis Ag85A (mc(2)155)
MTFFEQVRRLRSAATTLPRRLAIAAMGAVLVYGLVGTFGGPATAGAFSRP
--------MRGIAAWKALRRFVIGALAALMLPGLIGFAGGSASAGAFSRP
---MKFVGRMRGAAAGLSRRLTVAVAAAAVLPGLVGVVGGSATAGAWSRP
**
**:.:.. .* :: **:* **.*:***:***
M. tuberculosis Ag85C (H37Rv)
M. smegmatis Ag85C (mc(2)155)
M. smegmatis Ag85A (mc(2)155)
GLPVEYLQVPSASMGRDIKVQFQGGGP--HAVYLLDGLRAQDDYNGWDIN
GLPVEYLDVFSPSMNRDIRVQFQGGGP--HAVYLLDGLRAQDDYNGWDIN
GLPVEYLEVPSAAMGRDIRVEFQSGGPGAPALYLLDGMRAREDQNGWDIE
*******:* *.:*.***:*:**.*** *:*****:**::* *****:
M. tuberculosis Ag85C (H37Rv)
M. smegmatis Ag85C (mc(2)155)
M. smegmatis Ag85A (mc(2)155)
TPAFEEYYQSGLSVIMPVGGQSSFYTDWYQPSQSN--GQNYTYKWETFLT
TPAFEWFYQSGLSTIMPVGGQSSFYTDWYQPSKGN--GQDYTYKWETFLT
LPTFEWFLNSGISVVMPVGGQSSFYSDWYKPACGSKDGGCKTYKWETFLT
*:** : :**:*.:**********:***:*: .. * *********
M. tuberculosis Ag85C (H37Rv)
M. smegmatis Ag85C (mc(2)155)
M. smegmatis Ag85A (mc(2)155)
REMPAWLQANKGVSPTGNAAVGLSMSGGSALILAAYYPQQFPYAASLSGF
QELPAWLEANRGVSRTGNAVVGLSMAGSAALTYAIYHPEQFIYASTLSGF
QELPAWLAANRDVKPTGSAVVGLSMAGSSALMLAARHPQQFIYAASLSGT
:*:**** **:.*. **.*.*****:*.:** * :*:** **::***
M. tuberculosis Ag85C (H37Rv)
M. smegmatis Ag85C (mc(2)155)
M. smegmatis Ag85A (mc(2)155)
LNPSEGWWPTLIGLAMNDSGGYNANSMWGPSSDP--AWKRNDPMVQIPRL
LNPSEGWWPMLIGLAMNDAGGYNAESMWGPSTDP--AWKRNDPMVNINQL
LNPSEGWWPMLIGISMGDAGGYKADDMWGSTNDPNNAWKANDPTENVATI
********* ***::*.*:***:*:.***.:.** *** *** :: :
M. tuberculosis Ag85C (H37Rv)
M. smegmatis Ag85C (mc(2)155)
M. smegmatis Ag85A (mc(2)155)
VANNTRIWVYCGNGTPSDLGG-----DNIPAKFLEGLTLRTNQTFRDTYA
VANNTRLWVYCGTGTPSELDAGVSGGNLLAAQFLEGLTLRTNLTFRDQYI
ANNGTRIWVYCGNGKPGELGG-----TDLPAKFLEGFVCRTNSTFQEKYI
. *.**:*****.*.*.:*..
:.*:****:. *** **:: *
M. tuberculosis Ag85C (H37Rv)
M. smegmatis Ag85C (mc(2)155)
M. smegmatis Ag85A (mc(2)155)
ADGGRNGVFNFPPNGTHSWPYWNEQLVAMKADIQHVLNGATPPAAPAAPA
AAGGTNAVFNFPPNGTHTWNYWGQQLLEMKPDIQRVLGAQSAT-------
EAGGKNGVFNFPQSGTHNWAYWGQQLQAMKPDLQRVLGATPTA-------
** *.***** .***.* **.:** **.*:*:**.. ...
M. tuberculosis Ag85C (H37Rv)
M. smegmatis Ag85C (mc(2)155)
M. smegmatis Ag85A (mc(2)155)
A
-
-
Figure 6. Sequence alignment of M. tuberculosis Ag85C (H37Rv) with M. smegmatis Ag85C (mc2155) and M. smegmatis Ag85A (mc2155). Multiple-
sequence alignments were obtained with the ClustalW program (version 1.82).32 Catalytic triad residues Ser, Glu, and His are indicated in red and
selected carbohydrate-binding pocket I residues in blue. (*) indicates identical residues; (:) indicates residues that are conserved; (.) indicates residues
that are semi-conserved; and (-) represents a gaps introduced to maintain the alignment. Residues in gray represent the portion of the sequence that is
removed upon protein secretion.

5678
A. K. Sanki et al. / Bioorg. Med. Chem. 16 (2008) 5672–5682
We were surprised that both compounds 8 and 11
showed MICs against B. subtilis (128 and 512 lg/mL,
respectively) while the compounds showed no activity
against E. coli. This non-specific activity raises the pos-
sibility that 5-S-alkyl-5-thio-D-arabinofuranosides,
which have an amphipathic character, kill bacteria via
a detergent effect independent of ag85 inhibition. Fur-
ther, both mycobacterium33–35 and B. subtilis36 are re-
ported to be sensitive to non-ionic surfactants. While
detergency is one possible mode of action for the growth
inhibition of M. smegmatis, it cannot rule out the possi-
bility that the compounds act as inhibitors of ag85.
In an attempt to further address the mode of interaction
of the compounds, an attempt was made to screen the
arabinofuranoside library for the ability to inhibit
ag85C using our acyltransferase assay. Existing assays
to evaluate ag85 activity are cumbersome and require
the isolation of TDM from bacteria and 14C-labeled tre-
halose as an acceptor.8,20 However, the most potent of
our inhibitors (8 and 11) precipitated in the aqueous as-
say system; therefore, future efforts will focus on the
generation of compounds with higher water solubility
and have fewer structural features associated with
amphipathic character.
Figure 7. Methyl 5-S-octyl-5-thio-b-D-arabinofuranoside (11) docked
within the active site of ag85C. The octyl chain was superimposed on
the octyl chain of an octyl thioglucoside–ag85C complex (1DQY, Ref.
14). Additionally, hydrogen bonding seen in 1DQY between the
thioglucoside and ag85C was used as a basis for positioning the
arabinose moiety in the model. The catalytic nucleophile, Ser124, is
highlighted in blue.
enzyme. Similar to the work of Wang et al., we also
noted a lack of bacterial regrowth in the disk assay for
compounds 8 (Fig. 8) and 11 (not shown) in comparison
to INH.21 This potentially indicates that the compounds
were superior to INH in their ability to completely ster-
ilize the zone of inhibition. We then evaluated the selec-
tivity of the antimicrobial activity by including Gram-
positive (B. subtilis 6633) and Gram-negative (E. coli
23722) strains as controls. Neither of these strains pro-
cesses mycolic acids or have proteins similar to ag85.
4. Conclusion
We have synthesized a small library of arabinose-based
compounds using a Fisher glycosylation protocol to ob-
tain the kinetically formed methylfuranosides. Aliphatic
chains were incorporated at C-5 of the methyl glycosides
and their antimicrobial activity was evaluated against a
panel of bacteria. Compounds with a C8-chain were the
most effective at inhibiting the growth of mycobacterium
and activity was noted against a second Gram-positive
strain. We attempted to evaluate the inhibitory activity
of the compounds against ag85C and encountered solu-
bility issues for compounds containing C6 and C8 side
chains. Compound 9, a C4 congener, was examined over
a range of concentrations in an acyltransferase assay
and did not show inhibitory activity. Despite not having
established a potent lead against the mycolyltransferase
enzyme, the antibacterial nature of these compounds
and their ability to prevent bacterial regrowth may
prove useful for the development of a new class of car-
bohydrate-based anti-tubercular drugs. Work is under-
way to synthesize arabinose derived compounds that
mitigate the solubility problem and to enhance the
inhibitory activity against the ag85 complex.
5. Experimental
5.1. Materials
All fine chemicals such as D-arabinose, tosyl chloride, 1-
butanethiol, 1-hexanethiol, and 1-octanethiol, INH,
DMSO and anhydrous solvents such as anhydrous
methanol, and DMF were purchased from Acros and
used without further purification. Thiazolyl blue tetra-
zolium bromide (MTT) was purchased from Sigma–Al-
Figure 8. Kirby–Bauer diffusion assay after three days of
M. smegmatis incubation for compound 8 and INH against dissolved
in DMSO. (A) 8 20 (20 mg/mL); (B) 8 (10 mg/mL); (C) INH (500 lg/
mL); and (D) DMSO.

A. K. Sanki et al. / Bioorg. Med. Chem. 16 (2008) 5672–5682
5679
drich. All the other solvents were obtained from Fisher
and used as received except pyridine, which was dried
and distilled following the standard procedures.37 Silica
(230–400 mesh) for column chromatography was ob-
tained from Sorbent Technologies; thin-layer chroma-
tography (TLC) precoated plates were from EMD.
TLCs (silica gel 60, f254) were visualized under UV light
or by charring (5% H2SO4–MeOH). Flash column chro-
matography was performed on silica gel (230–400 mesh)
using solvents as received. 1H NMR were recorded
either on a Varian VXRS 400 MHz or an INOVA 600
MHz spectrometer in CDCl3 or DMSO-d6 using resid-
ual CHCl3 and DMSO as internal references, respec-
tively. 13C NMR were recorded on the Varian INOVA
150 MHz spectrometer in CDCl3 using the triplet cen-
tered at d 77.273 as internal reference. High resolution
mass spectrometry (HRMS) was performed on a Micro-
mass Q-TOF2 instrument. Petri dishes (100 · 15 mm)
were obtained from Fisher Scientific. LB broth, Middle-
brook 7H9 broth, Lennox and Middlebrook 7H10 agar,
glycerol, ADC and OADC enrichment, and blank sus-
ceptibility disks were obtained from Becton, Dickinson,
and Company (Franklin Lakes, NJ). B. subtilis ATCC
6633, E. coli ATCC 23722, and M. smegmatis ATCC
14468 were obtained from the American Type Culture
Collection (Manassas, VA).
gum), and 3b is 2.27 g (34.6%, yellow gum); silica gel
TLC Rf3a = 0.43, Rf3b = 0.30 (3:3:14 acetone/MeOH/
CHCl3). When the reaction was carried out in the pres-
ence of acetyl chloride in anhydrous methanol at 0 ꢀC
on a 10.5 g scale, 89% desired products were isolated
in the same way as described above in 18 h.
5.2.3. Methyl 5-O-p-toluenesulfonyl-a-D-arabinofurano-
side (4) and methyl 5-O-p-toluenesulfonyl-b-D-arabinofu-
ranoside (5). To a well-stirred solution of 3a (0.79 mg,
4.82 mmol) in anhydrous pyridine (15 mL) was added
a
solution of (p)-toluenesulfonyl chloride (0.97 g,
5.1 mmol) in anhydrous pyridine (15 mL) under N2
atmosphere. The resulting solution was stirred at ambi-
ent temperature and the reaction was monitored by TLC
and appeared to stop after 24 h. Excess tosyl chloride
was hydrolyzed by adding ice cubes (two pieces) and
stirring was continued for 0.5 h. The reaction mixture
was poured into water (80 mL) and extracted with ethyl
acetate (4· 30 mL). The combined organic phases were
washed with brine, and aqueous phase was back ex-
tracted. Resulting organic phases were dried (anhydrous
Na2SO4), filtered and the filtrate was concentrated under
reduced pressure to get the crude material. Purification
of the crude material by silica gel flash chromatography
(10 · 4.5 cm) with 3:3:14 acetone/CHCl3/hexanes affor-
ded tosylate 4 as a yellow gum; yield: 60% (0.92 g); silica
gel TLC Rf = 0.6 (1:4 MeOH/CHCl3). Similarly, 3b
(0.85 g, 5.18 mmol) was converted to corresponding tos-
ylate 5 in 50% yield (0.83 g) as a yellow gum; silica gel
TLC Rf = 0.56 (1:1:8 acetone/MeOH/CHCl3).
5.2. Synthesis
5.2.1. Method A: methyl a-D-arabinofuranoside (3a) and
methyl b-D-arabinofuranoside (3b). To a suspension of D-
arabinose (5 g, 33.3 mmol) in anhydrous MeOH
(60 mL), acetyl chloride (0.9 mL) was added dropwise
at ambient temperature under N2 atmosphere and the
resulting solution was stirred at 90 ꢀC. The reaction
was monitored by TLC and appeared to stop after
0.5 h. The solution was allowed to attain room temper-
ature and excess acid was neutralized by Na2CO3 (requi-
site amount, checked by pH-paper) and the resulting
suspension was filtered through Celite using MeOH as
solvent. Celite bed was washed successively with MeOH
(3· 60 mL). The combined filtrate was concentrated to a
small volume and the precipitated Na2CO3 was removed
by filtration through analytical filter paper. The thick
crude residue thus obtained was dissolved in MeOH, ab-
sorbed on silica gel and finally loaded over silica gel
flash column (10 · 6.5 cm). Elution with 1:1:8 acetone/
MeOH/CHCl3 generated the methyl glycosides 3a
(1.1 g, 20%) as a colorless thick gum and a mixture of
undesired pyranosides 2 (3.51 g, 64%) as colorless amor-
phous solid; silica gel TLC Rf = 0.3 (3:3:14 acetone/
MeOH/CHCl3).
5.2.4. General procedure for the synthesis of 6–11. To a
suspension of NaOMe (2.5 equiv/mmol of substrate: ob-
tained by the reaction of metallic sodium with little ex-
cess of anhydrous MeOH and removing excess solvent
under reduced pressure and thereby drying it in high
vacuum pump) in anhydrous DMF (2 mL/mmol) was
added appropriate alkanethiol (5 equiv/mmol of sub-
strate) under N2 atmosphere. The resulting mixture
was stirred at ambient temperature for 15 min to get a
clear solution. A solution of the appropriate tosylate
(starting material) in DMF (3 mL/mmol) was added
dropwise. The resulting solution was allowed to stir at
the same temperature under N2 atmosphere and moni-
tored by TLC and appeared to stop at 1–3.5 h. After
completion of the reaction (monitored by TLC), the
reaction mixture was poured into water and extracted
with ethyl acetate (4· 40 mL). The combined organic
layers were washed with brine, aqueous phases were
back extracted, and resulting organic phase was dried
(anhydrous Na2SO4), filtered and the filtrate was con-
centrated in vacuo to get the crude material. Purification
of the crude material by silica gel flash chromatography
(10 · 3.5 cm) with acetone/CHCl3/hexanes (3:3:14 for 7,
8, 10 and 11 or 2:2:3 for 6 and 9) produced the title
compounds.
5.2.2. Method B: methyl a-D-arabinofuranoside (3a) and
methyl b-D-arabinofuranoside (3b). To a suspension of D-
arabinose (6 g, 39.96 mmol) in anhydrous MeOH
(100 mL), HCl (2 mL) was added dropwise at 0–4 ꢀC un-
der N2 atmosphere and the resulting solution was stirred
at 0–4 ꢀC. The reaction was monitored by TLC and ap-
peared to stop after 2.5 days. Following the similar work
up and purification protocols as described in Method A,
desired methyl glycosides 3a and 3b were obtained in
85% yield, of which 3a is 3.25 g (49.6%, colorless
5.2.5. Methyl 5-S-butyl-5-thio-a-D-arabinofuranoside (6).
Compound 4 (0.25 g, 0.786 mmol) was reacted with
butanethiol (0.337 mL, 3.14 mmol) to afford 6 in 1 h as
a colorless gum following the general procedure de-
scribed above. Purification was accomplished by silica

5680
A. K. Sanki et al. / Bioorg. Med. Chem. 16 (2008) 5672–5682
gel flash column (10 · 3.5 cm) with 2:2:3 acetone/CHCl3/
hexanes; yield: 92% (0.17 g); silica gel TLC Rf = 0.2
(1:1:3 acetone/CHCl3/hexanes); 1H NMR (600 MHz,
CDCl3): d 0.92 (t, 3H, J = 7.8 Hz, CH3), 1.40 (sextet,
2H, CH2), 1.58 (quintet, 2H, CH2), 2.59 (d, 1H,
J = 9.6 Hz, OH-3), 2.65 (ddd, 2H, J = 3.0, 7.8,
10.2 Hz, SCH2–), 2.85 (dd, 1H, J = 4.2, 14.4 Hz, H-5),
3.27 (d, 1H, J = 9.0 Hz, OH-2), 3.40 (s, 3H, OCH3),
3.88 (dd, 1H, J = 3.0, 10.2 Hz, H-3), 4.05 (d, 1H,
J = 9.0 Hz, H-2), 4.25 (m, 1H, H-4), 4.90 (s, 1H, H-1);
13C NMR (150 MHz, CDCl3): d 13.8 (CH3), 22.10
(CH2), 31.9 (CH2), 33.5 (CH2), 34.87 (CH2), 55.2
(OCH3), 80.73 (CH), 80.8 (CH), 84.99 (CH), 109.1 (C-
1); mass spectrum (HRMS), m/z = 259.0980 (M+Na)+
(C10H20O4NaS requires 259.0980).
butanethiol (0.656 mL, 6.12 mmol) to afford 9 in 3.5 h
as a colorless gum following the general procedure
described above. Purification was accomplished by silica
gel flash column (10 · 3.5 cm) with 2:2:3 acetone/CHCl3/
hexanes; yield: 73% (0.212 g); silica gel TLC Rf = 0.37
(2:2:1 acetone/CHCl3/hexanes); 1H NMR (600 MHz,
CDCl3): d 0.92 (t, 3H, J = 7.2 Hz, CH3), 1.41 (sextet,
2H, CH2), 1.59 (m, 2H, CH2), 2.56 (d, 1H, J = 9.6 Hz,
OH-2), 2.59 (t, 1H, J = 7.2 Hz, SCH2–), 2.65 (dd, 1H,
J = 7.8, 13.2 Hz, H-5), 2.69 (d, 1H, J = 3.0 Hz, OH-3),
2.84 (dd, 1H, J = 6.0, 13.8 Hz, H-5), 3.44 (s, 3H,
OCH3), 3.91 (m, 1H, H-4), 4.03 (ddd, 1H, J = 3.0, 6.6,
9.6 Hz, H-3), 4.09 (m, 1H, H-2), 4.80 (d, 1H,
J = 4.8 Hz, H-1); 13C NMR (150 MHz, CDCl3): d
13.77 (CH3), 22.04 (CH2), 31.78 (CH2), 32.20 (CH2),
36.76 (CH2), 55.38 (OCH3), 78.06 (CH), 79.63
(CH), 81.17 (CH), 102.14 (C-1); mass spectrum
(HRMS), m/z = 259.0975 (M+Na)+(C10H20O4NaS re-
quires 259.0980).
5.2.6. Methyl 5-S-hexyl-5-thio-a-D-arabinofuranoside (7).
Compound 4 (0.25 g, 0.786 mmol) was reacted with
hexanethiol (0.442 mL, 3.14 mmol) to afford 7 in 1 h
as a colorless gum following the general procedure de-
scribed above. Purification was accomplished by silica
gel flash column (10 · 3.5 cm) with 3:3:14 acetone/
CHCl3/hexanes; yield: 50% (0.104 g); silica gel TLC
Rf = 0.21 (1:1:3 acetone/CHCl3/hexanes); 1H NMR
(600 MHz, CDCl3): d 0.89 (t, 3H, J = 7.2 Hz, CH3),
1.29 (m, 4H, 2· CH2), 1.37 (quintet, 2H, CH2), 1.59
(quintet, 2 H, CH2), 2.63 (d, 1H, J = 9.6 Hz, OH-3),
2.64 (m, 2H, S–CH2–), 2.84 (dd, 1H, J = 3.6, 14.4 Hz,
H-5), 2.90 (m, 1H, H-5), 3.30 (d, 1H, J = 9.0 Hz, OH-
2), 3.40 (s, 3H, OCH3), 3.87 (dd, 1H, J = 3.0, 9.6 Hz,
H-3), 4.04 (d, 1H, J = 8.4 Hz, H-2), 4.24 (m, 1H, H-4),
4.90 (s, 1H, H-1); 13C NMR (150 MHz, CDCl3): d
14.20 (CH3), 22.70 (CH2), 28.65 (CH2), 29.77 (CH2),
31.56 (CH2), 33.66 (CH2), 34.84 (CH2), 55.17 (OCH3),
80.69 (CH), 80.98 (CH), 84.62 (CH), 109.03 (C-1); mass
5.2.9. Methyl 5-S-hexyl-5-thio-b-D-arabinofuranoside
(10). Compound 5 (0.458 g, 1.44 mmol) was reacted with
hexanethiol (1.01 mL, 7.20 mmol) to afford 10 in 3.5 h
as a colorless gum following the general procedure de-
scribed above. Purification was accomplished by silica
gel flash column (10 · 3.5 cm) with 3:3:14 acetone/
CHCl3/hexanes; yield: 60% (0.227 g); silica gel TLC
Rf = 0.45 (2:2:1 acetone/CHCl3/hexanes); 1H NMR
(600 MHz, CDCl3): d 0.89 (t, 3H, J = 6.6 Hz, CH3),
1.29 (m, 4H, 2· CH2), 1.38 (quintet, 2H, CH2), 1.60
(m, 2H, CH2), 2.47 (d,1H, J = 9.6 Hz, OH-2), 2.54 (d,
1H, J = 3.0 Hz, OH-3), 2.58 (t, 2H, J = 7.2 Hz, SCH2–
), 2.64 (dd, 1H, J = 8.4, 13.2 Hz, H-5), 2.85 (dd, 1H,
J = 6.0, 13.8 Hz, H-5), 3.45 (s, 3H, OCH3), 3.91 (m,
1H, H-4), 4.02 (ddd, 1H, J = 3.0, 6.6, 10.2 Hz, H-3),
4.09 (m, 1H, H-2), 4.81 (d, 1H, J = 4.8 Hz, H-1); 13C
NMR (150 MHz, CDCl3): d 14.08 (CH3), 22.59 (CH2),
28.60 (CH2), 29.66 (CH2), 31.48 (CH2), 32.51 (CH2),
36.74 (CH2), 55.32 (OCH3), 78.01 (CH), 79.51
(CH), 81.18 (CH), 102.12 (C-1); mass spectrum
(HRMS), m/z = 287.1297 (M+Na)+(C12H24O4NaS
requires 287.1293).
spectrum
(C12H24O4NaS requires 287.1293).
(HRMS),
m/z = 287.1299
(M+Na)+
5.2.7. Methyl 5-S-octyl-5-thio-a-D-arabinofuranoside (8).
Compound 4 (0.15 g, 0.471 mmol) was reacted with
octanethiol (0.287 mL, 1.65 mmol) to afford 8 in 1 h as
a colorless gum following the general procedure de-
scribed above. Purification was accomplished by silica
gel flash column (10 · 3.5 cm) with 3:3:14 acetone/
CHCl3/hexanes; yield: 98% (0.154 g); silica gel TLC
Rf = 0.20 (1:1:3 acetone/CHCl3/hexanes); 1H NMR
(600 MHz, CDCl3): d 0.88 (t, 3H, J = 7.2 Hz, CH3),
1.28 (m, 8H, 4· CH2), 1.37 (quintet, 2H, CH2), 1.59
(quintet, 2H, CH2), 2.60 (d, 1H, J = 10.2 Hz, OH-3),
2.64 (ddd, 2H, J = 1.8, 7.2, 9.0 Hz, SCH2–), 2.85 (dd,
1H, J = 4.2, 14.4 Hz, H-5), 2.90 (dd, 1H, J = 5.4,
14.4 Hz, H-5), 3.27 (d, 1H, J = 9.0 Hz, OH-2), 3.40 (s,
3H, OCH3), 3.87 (dd, 1H, J = 3.0, 9.6 Hz, H-3), 4.04
(d, 1H, J = 8.4 Hz, H-2), 4.24 (m, 1H, H-4), 4.90 (s,
1H, H-1); 13C NMR (150 MHz, CDCl3): d 14.27
(CH3), 22.81 (CH2), 29.0 (CH2), 29.35 (2· CH2), 29.82
(CH2), 31.98 (CH2), 33.67 (CH2), 34.82 (CH2), 55.18
(OCH3), 80.68 (CH), 80.98 (CH), 84.53 (CH), 109.0
(C-1); mass spectrum (HRMS), m/z = 315.1602
(M+Na)+(C14H28O4NaS requires 315.1606).
5.2.10. Methyl 5-S-octyl-5-thio-b-D-arabinofuranoside
(11). Compound 5 (0.325 g, 1.02 mmol) was reacted with
octanethiol (0.889 mL, 5.11 mmol) to afford 11 in 2.5 h
as a colorless gum following the general procedure de-
scribed above. Purification was accomplished by silica
gel flash column (10 · 3.5 cm) with 3:3:14 acetone/
CHCl3/hexanes; yield: 73% (0.219 g); silica gel TLC
Rf = 0.39 (1:1:3 acetone/CHCl3/hexanes); 1H NMR
(600 MHz, CDCl3): d 0.88 (t, 3H, J = 6.6 Hz, CH3),
1.28 (m, 8H, 4· CH2), 1.37 (quintet, 2H, CH2), 1.59
(m, 2H, CH2), 2.49 (d, 1H, J = 9.6 Hz, OH-2), 2.57 (d,
1H, J = 2.4 Hz, OH-3), 2.58 (t, 2H, J = 7.8 Hz, SCH2–),
2.64 (dd, 1H, J = 8.4, 13.2 Hz, H-5), 2.85 (dd, 1H,
J = 6.0, 13.2 Hz, H-5), 3.45 (s, 3H, OCH3), 3.91 (m,
1H, H-4), 4.03 (ddd, 1H, J = 3.0, 6.6, 9.6 Hz, H-3),
4.09 (m, 1H, H-2), 4.81 (d, 1H, J = 4.8 Hz, H-1); 13C
NMR (150 MHz, CDCl3): d 14.18 (CH3), 22.73 (CH2),
29.00 (CH2), 29.29 (CH2), 29.31(CH2), 29.75 (CH2),
31.90 (CH2), 32.55 (CH2), 36.78 (CH2), 55.35 (OCH3),
78.04 (CH), 79.58 (CH), 81.21 (CH), 102.16 (C-1);
5.2.8. Methyl 5-S-butyl-5-thio-b-D-arabinofuranoside (9).
Compound 5 (0.389 g, 1.223 mmol) was reacted with

A. K. Sanki et al. / Bioorg. Med. Chem. 16 (2008) 5672–5682
5681
mass spectrum (HRMS), m/z = 315.1607 (M+Na)+
(C14H28O0NaS requires 315.1606).
5.6. Modeling studies
The putative inhibitor 11 was created and minimized
using Chem3D Ultra (CambridgeSoft). This molecule
was positioned in the active site of ag85C using the pro-
gram Coot.39 The manual positioning of the inhibitor
within the active site was based on the crystal structure
of n-octyl-b-D-thioglucoside bound to ag85C (accession
code 1VA5).14 Care was taken to conserve the hydro-
gen-bonding pattern between the carbohydrate moiety
and residues L40, R41, and W262 forming the carbohy-
drate-binding site. The non-specific hydrophobic inter-
actions were less of a concern, but attempts were made
when positioning the carbons of the aliphatic chain to
maintain interactions with residues F150, W158, I162,
A165, and L227 of the acyl chain binding pocket.
5.3. Kirby–Bauer disk diffusion assay
Disk diffusion assays were adapted from methods out-
lined by the NCCLS and Wang et al.21,28 M. smegmatis
ATCC 14468 was inoculated into Middlebrook 7H9
containing 0.2% glycerol and ADC enrichment, respec-
tively. The inocula were incubated for approximately
24 h at 37.5 ꢀC and 160 rpm to yield an OD600 = 0.27.
The bacteria were plated on agar (Lennox or Middle-
brook 7H10 containing 0.5% glycerol and OADC
Enrichment) using sterile cotton-tipped applicators. Ali-
quots (10 lL) of arabinose derivatives dissolved in
DMSO (20 mg/mL and 10 mg/mL) and INH (500 lg/
mL) were applied to 6 mm diameter sterile paper disks.
The plates were incubated at 37 ꢀC for 24 h and then
analyzed. The plates were returned to the incubator
and then analyzed six days after the initial incubation
to check for additional changes. The DZIs (mm) were
recorded using a ruler.
Acknowledgments
We thank the University of Toledo for financial support
through New Faculty Research Grants, an Interdisci-
plinary Research Initiation Award (to S.J.S.), and a
deArce Memorial Fellowship (to D.R.R.).
5.4. Broth macrodilution assay
MIC values against three test organisms, E. coli ATCC
23722, B. subtilis ATCC 6633, and M. smegmatis ATCC
14468, were obtained using broth macrodilution assay
using either Mueller-Hinton or Middlebrook 7H9 con-
taining 0.2% glycerol and ADC enrichment (M.
smegmatis ATCC 14468), adapted from NCCLS proce-
dures.31 Bacterial cultures equivalent to a 0.5 McFar-
land turbidity standard were prepared in the
appropriate broth to give a final inoculum density of
5.0 · 105 cfu/mL and the solutions incubated at 37 ꢀC
for 36 h. To aid in the visualization of the bacteria,
20 lL of 10 mg/mL solution (MeOH solvent) of thiazol-
yl blue tetrazolium bromide (MTT) was added.38 All as-
says were run in duplicate.
References and notes
1. Reviews: (a) Minnikin, D. E. In The Biology of the
Mycobacteria Ratledge; Stanford, C. J., Ed.; Academic
Press: London, 1982; pp 95–184; (b) Crick, D. C.;
Mahapatra, S.; Brennan, P. J. Glycobiology 2001, 11,
´
107R; (c) Dover, L. G.; Cerden˜o-Tarraga, A. M.; Pallen,
M. J.; Parkhill, J.; Besra, G. S. FEMS Microbiol. Rev.
2004, 28, 225.
2. (a) Jarlier, V.; Nikaido, H. FEMS Microbiol. Lett. 1994,
´
123, 11; (b) Daffe, M.; Draper, P. Adv. Microb. Phys. 1998,
39, 131; (c) Jackson, M.; Raynaud, C.; Laneelle, M.-A.;
´
Guilhot, C.; Laurent-Winter, C.; Ensergueix, D.; Gicquel,
´
B.; Daffe, M. Mol. Microbiol. 1999, 31, 1573.
3. Daffe, M.; Brennan, P. J.; McNeil, M. J. Biol. Chem. 1990,
265, 6734.
5.5. Antigen 85C acyltransferase assay
4. McNeil, M. R.; Robuck, K. G.; Harter, M.; Brennan, P. J.
Glycobiology 1994, 4, 165.
A master mix of sodium phosphate buffer C, b-gluco-
sidase, and dextrose (0.1 M, 3 lL per reaction) was
prepared. Master mix (85.84 lL) was added to each
well of the assay, followed by 10 lL of a 10% v/v di-
methyl sulfoxide stock solution of substrate (0.1–
2.5 mM), and 1 lL of a dimethyl sulfoxide solution
of arabinofuranoside 9. The C6 and C8 homologues
were not soluble in the aqueous well solution and
therefore could not be assayed. To initiate catalysis
4.16 lL of a 45 lg/mL solution of purified antigen
85C in 10 mM Tris–HCl buffer, pH 7.5, 2 mM ethyl-
enediaminetetraacetic acid and 1 mM dithiothreitol
were added. The change of absorbance of the reaction
solution at 405 nm was followed by absorbance mea-
surement (absorbance plate reader Spectramaxꢁ
340PC by Molecular Devices) every 20 s for 1 h. All
assays were prepared and performed at room temper-
ature. The data were exported from the Molecular De-
vices SoftMaxꢁ Pro5 software and analyzed using
Microsoft Excelꢁ to obtain initial velocity values for
each reaction.
5. (a) Belanger, A. E.; Besra, G. S.; Ford, M. E.; Mikusova,
K.; Belisle, J. T.; Brennan, P. J.; Inamine, J. M. Proc. Natl.
Acad. Sci. U.S.A. 1996, 93, 11919; (b) Telenti, A.; Philipp,
W. J.; Sreevatsan, S.; Bernasconi, C.; Stockbauer, K. E.;
Wieles, B.; Musser, J. M.; Jacobs, W. R., Jr Nat. Med.
1997, 3, 567.
6. Takayama, K.; Wang, L.; David, H. L. Antimicrob. Agents
Chemother. 1972, 2, 29.
7. Schroeder, E. K.; de Souza, O. N.; Santos, D. S.;
Blanchard, J. S.; Basso, L. A. Curr. Pharm. Biotechnol.
2002, 3, 197.
8. Belisle, J. T.; Vissa, V. D.; Sievert, T.; Takayama, K.;
Brennan, P. J.; Besra, G. S. Science 1997, 276, 1420.
9. Takayama, K.; Wang, C.; Besra, G. S. Clin. Microbiol.
Rev. 2005, 18, 81.
10. (a) Fukui, Y.; Hirai, T.; Uchida, T.; Yoneda, M. Biken J.
1965, 8, 189; (b) Kilburn, J. O.; Takayama, K.; Arm-
strong, E. L. Biochem. Biophys. Res. Commun. 1982, 108,
132; (c) Sathyamoorthy, N.; Takayama, K. J. Biol. Chem.
1987, 262, 13417.
´
11. Puech, V.; Bayan, N.; Salim, K.; Leblon, G.; Daffe, M.
Mol. Microbiol. 2000, 35, 1026.

5682
A. K. Sanki et al. / Bioorg. Med. Chem. 16 (2008) 5672–5682
12. Dye, C.; Scheele, S.; Dolin, P.; Pathania, V.; Raviglione,
M. C. J. Am. Med. Assoc. 1999, 282, 677.
13. (a) Cohen, J. Science 2006, 313, 1554; (b) Johar, M.;
Manning, T.; Tse, C.; Desroches, N.; Agrawal, B.;
Kunimoto, D. Y.; Kumar, R. J. Med. Chem. 2007, 50,
3696.
(d) Bock, K.; Pederson, C. Carbohydr. Res. 1979, 73, 85;
(e) Casini, G.; Goodman, L. J. Am. Chem. Soc. 1964, 86,
1427; (f) Su, T.-L.; Klein, R. S.; Fox, J. J. J. Org. Chem.
1981, 46, 1790.
27. Hughes, N. A.; Kuhajda, K.-M.; Miljkovic, D. A.
Carbohydr. Res. 1994, 257, 299.
14. Ronning, D. R.; Klabunde, T.; Besra, G. S.; Vissa, V. D.;
Belisle, J. T.; Sacchettini, J. C. Nat. Struct. Biol. 2000, 7,
141.
15. Anderson, D. H.; Harth, G.; Horwitz, M. A.; Eisenberg,
D. J. Mol. Biol. 2001, 307, 671.
28. National Committee for Clinical Laboratory Standards.
Approved Standard: M2-A7. Performance Standards for
Antimicrobial Disk Susceptibility Tests, 7th ed.; National
Committee for Clinical Laboratory Standards: Wayne,
PA, 2000.
16. Gelb, M. H.; Cho, W.; Wilton, D. C. Curr. Opin. Struct.
Biol. 1999, 9, 428.
29. Pierre-Audigier, C.; Jouanguy, E.; Lamhamedi, S.; Altare,
F.; Rauzier, J.; Vincent, V.; Canioni, D.; Emile, J. F.;
Fischer, A.; Blanche, S.; Gaillard, J. L.; Casanova, J. L.
Clin. Infect. Dis. 1997, 24, 982.
30. Zhang, J.; Chen, H.-N.; Chiang, F.-I.; Takemoto, J. Y.;
Bensaci, M.; Chang, C.-W. T. J. Comb. Chem. 2007, 9, 17.
31. National Committee for Clinical Laboratory Standards.
Approved Standard: M7-A6. Methods for Dilution Anti-
microbial Susceptibility Tests for Bacteria that Grow
Aerobically, 6th ed.; National Committee for Clinical
Laboratory Standards: Wayne, PA, 2003.
17. Ronning, D. R.; Vissa, V.; Besra, G. S.; Belisle, J. T.;
Sacchettini, J. C. J. Biol. Chem. 2004, 279, 36771.
18. Mizuguchi, Y.; Udou, T.; Yamada, T. Microbiol. Immu-
nol. 1983, 27, 425.
ˇ
19. Gobec, S.; Plantan, I.; Mravljak, J.; Savajger, U.; Wilson,
R. A.; Besra, G. S.; Soares, S. L.; Appelberg, R.; Kikelj,
D. Eur. J. Med. Chem. 2007, 42, 54.
20. Kremer, L.; Maughan, W. N.; Wilson, R. A.; Dover, L.
G.; Besra, G. S. Lett. Appl. Microbiol. 2002, 34, 233.
21. Wang, J.; Elchert, B.; Hui, Y.; Takemoto, J. Y.; Bensaci,
M.; Wennergren, J.; Chang, H.; Rai, R.; Chang, C.-W. T.
Bioorg. Med. Chem. 2004, 12, 6397.
22. Rose, J. D.; Maddry, J. A.; Comber, R. N.; Suling, W. J.;
Wilson, L. N.; Reynolds, R. C. Carbohydr. Res. 2002, 337,
105.
23. Lowary, T. L. Mini-Rev. Med. Chem. 2003, 3, 689.
24. Augestad, I.; Berner, E. Acta Chem. Scand. 1954, 8, 251.
25. (a) Ayers, J. D.; Lowary, T. L.; Morehouse, C. B.; Besra,
G. S. Bioorg. Med. Chem. Lett. 1998, 8, 437; (b) Kam, B.
L.; Barascut, J.-L.; Imbach, J.-L. Carbohydr. Res. 1979,
69, 135.
26. (a) Kawana, M.; Kuzuhara, H.; Emoto, S. Bull. Chem.
Soc. Jpn. 1981, 54, 1492; (b) Montgomery, J. N.; Thorpe,
M. C.; Clayton, S. D.; Thomas, H. J. Carbohydr. Res.
1974, 32, 404; (c) Mizutani, K.; Kasai, R.; Nakamura, M.;
Tanaka, O.; Matsuura, H. Carbohydr. Res. 1989, 185, 27;
32. Thomson, J. D.; Higgins, D. G.; Gibson, T. J. Nucleic
Acids Res. 1997, 22, 4673.
33. Kabara, J. J.; Swieczkowski, D. M.; Conley, A. J.; Truant,
J. P. Antimicrob. Agents Chemother. 1972, 2, 23.
34. Kondo, E.; Kanai, K. Jpn. J. Med. Sci. Biol. 1972, 25, 1.
35. Thormar, H.; Bergsson, G. Recent Dev. Antiviral Res.
2001, 1, 157.
36. Rauter, A. P.; Lucas, S.; Almeida, T.; Sacoto, D.; Ribeiro,
V.; Justino, J.; Neves, A.; Silva, F. V. M.; Oliveira, M. C.;
Ferreira, M. J.; Santosa, M.-S.; Barbosa, E. Carbohydr.
Res. 2005, 340, 191.
37. Armarego, W. L. F.; Chai, C. L. L. In Purification of
Laboratory Chemicals, 5th ed.; Butterworth-Heinemann:
New York, 2003; pp 80–388.
38. Mosmann, T. J. Immunol. Methods 1983, 65, 55.
39. Emsley, P.; Cowtan, K. Acta Crystallogr., Sect. D 2004,
60, 2126.