C-3 alkyl/arylalkyl-2,3-dideoxy hex-2-enopyranosides as antitubercular agents: Synthesis, biological evaluation, and QSAR study

Article abstract of DOI:10.1021/jm070110h

A series of C-3 alkyl and arylalkyl 2,3-dideoxy hex-2-enopyranoside derivatives were synthesized by Morita - Baylis-Hillman reaction using enulosides 4, 5, and 6 and various aliphatic and aromatic aldehydes. The compounds were evaluated in vitro for the c

Full text of DOI:10.1021/jm070110h

2942
J. Med. Chem. 2007, 50, 2942-2950
C-3 Alkyl/Arylalkyl-2,3-dideoxy Hex-2-enopyranosides as Antitubercular Agents: Synthesis,
Biological Evaluation, and QSAR Study
Mohammad Saquib, Manish K. Gupta, Ram Sagar, Yenamandra S. Prabhakar,* Arun K. Shaw,* Rishi Kumar,†
Prakas R. Maulik,† Anil N. Gaikwad, Sudhir Sinha, Anil K. Srivastava, Vinita Chaturvedi, Ranjana Srivastava, and
Brahm S. Srivastava
Medicinal and Process Chemistry DiVision, DiVision of Molecular and Structural Biology, Drug Target DiscoVery and DeVelopment DiVision
and DiVision of Microbiology, Central Drug Research Institute, Lucknow-226001, India
ReceiVed January 27, 2007
A series of C-3 alkyl and arylalkyl 2,3-dideoxy hex-2-enopyranoside derivatives were synthesized by Morita-
Baylis-Hillman reaction using enulosides 4, 5, and 6 and various aliphatic and aromatic aldehydes. The
compounds were evaluated in Vitro for the complete inhibition of growth of Mycobacterium tuberculosis
H37RV. They exhibited moderate to good activity in the range of 25-1.56 µg/mL. Among these, 4d, 4h, 5c,
and 4hr showed activity at minimum inhibitory concentrations, 3.12, 6.25, 1.56, and 1.56 µg/mL, respectively.
These compounds were safe against cytotoxicity in VERO cell line and mouse macrophage cell line J 744A.1.
A QSAR analysis by CP-MLR with alignment-free 3D-descriptors indicated the relevance of structure space
comparable to the minimum energy conformation (from conformational analysis) of 5c to the activity. The
study indicates that the compounds attaining the conformational space of 5c and reflecting some symmetry,
minimum eccentricity, and closely placed geometric and electronegativity centers therein are favorable for
activity.
Introduction
improved activity.24 In recent literature, some 2H-pyran-3(6H)-
one derivatives (Figure 1) are reported to exhibit significant
activity against Gram-positive bacteria.25 A variety of carbo-
hydrate derivatives are also known to interfere with the cell
wall biosynthesis of M. tuberculosis.6,19,20 These studies sug-
gested that arabinogalactan-linked disaccharides and decaprenol
phosphoarabinosides interfere with the mycobacterial arabino-
syltransferases. Moreover, some iminosugars are inhibitors of
UDPGalf transferase,a in mycobacterial cell wall biosynthesis.21
All these studies emphasize modified carbohydrates as potential
scaffolds to interfere with the components of mycobacterial cell
wall biosynthesis. In these background studies, C-3 alkyl and
arylalky 2,3-dideoxy hexenopyranosides were synthesized to
investigate their activity against M. tuberculosis and to develop
a rationale for the activity. In recent modeling studies, the
alignment-free 3D-descriptors from the molecular geometry
were increasingly put to use as an alternative to the grid-based
3D-QSAR studies.26-33 They offer insight to the receptor via
the conformational space of the probe molecules and are
successfully used in the development of rationales for divergent
biological activities including tuberculosis.27,29,31-35 In view of
this, a QSAR of antitubercular activity of these hexenopyrano-
sides is attempted using the alignment-free 3D-descriptors from
DRAGON software.36
Mycobacterium tuberculosis, the leading causative agent of
tuberculosis (TB), is responsible for the morbidity and mortality
of a large population worldwide. According to a WHO report,
by 2020 AD nearly one billion more people will be infected,
200 million people will get sick, and 70 million will die from
tuberculosis if proper steps are not taken to control it.1-3
Moreover, the opportunistic TB infection due to AIDS and
emergence of drug resistant TB strains have made its chemo-
therapeutic strategies increasingly ineffective.4 Traditionally, it
has relied heavily on a limited number of drugs such as
isonicotinic acid hydrazide, rifampicin, ethambutal, streptomy-
cin, ethionamide, pyrazinamide, fluoroquinolones, etc.5 How-
ever, many of these drugs have different disadvantages such as
prolonged treatment schedules, host toxicity, ineffectiveness
against resistant strains, etc. This has motivated the search for
new chemical prototypes capable of rapid mycobactericidal
action with shortened duration of therapy, reduced toxicity, and
enhanced activity against drug resistant strains and also against
the latent bacteria for its control.6-12 This paradigm shift in the
strategies has drawn attention toward the outer cell wall and
membranes of the pathogen as an attractive target for exploring
new drugs.13,14 In mycobacteria, the cell wall structure consists
of a dense network of cross-linked sugar residues esterified with
mycolic acid at the ends.15,16 This understanding has prompted
the investigation of various sugar prototypes with distinct
characteristics as potential antitubercular agents.17-21 Earlier,
we investigated some highly functionalized heptenol and octenol
derivatives from glucal and galactal for their antitubercular
activity.22,23 A quantitative structure-activity relationship (QSAR)
study of these analogues with topological descriptors has
suggested less branched and saturated structural templates for
Chemistry. The synthesis of prototype molecules in this study
was initiated with 3,4,6-tri-O-acetyl-R-D-glucal (1).37 Iodine-
catalyzed Ferrier rearrangement38,39 of 1 with isopropanol
furnished 4,6-di-O-acetyl-2,3-dideoxy hex-2-enopyranoside (2).
Its deacetylation yielded dihydroxy derivative 3, which upon
allylic oxidation40 followed by acylation of C-6 hydroxyl with
different acylchlorides resulted in 4, 5, and 6 in good yields
(Scheme 1).
* To whom correspondence should be addressed. (Y.S.P.) Phone:+91-
522-2612411; fax:+91-522-2623405; e-mail: yenpra@yahoo.com. (A.K.S.)
Phone:+91-522-2612411; fax: +91-522-2623405; e-mail: akshaw55@
yahoo.com.
a Abbreviations: UDPGalf transferase, uridine diphosphate galactofura-
nosyl transferase; CP-MLR, combinatorial protocol in multiple linear
regression; GEO, geometrical; WHIM, weighted holistic invariant molecular
descriptors; GETAWAY, geometry, topology, and atom-weights assembly.
† R.K. and P.R.M. provided X-ray crystal data.
10.1021/jm070110h CCC: $37.00 © 2007 American Chemical Society
Published on Web 06/02/2007

2,3-Dideoxy Hexenopyranosides as Antitubercular Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13 2943
cally segregates the descriptor classes into four categories
depending on the significance of information content to the
activity (Figure 4). The first three categories are primary
contributors (category I: a descriptor class forming a model
with its constituent descriptors), collective contributors (category
II: a descriptor class unable to form a model with its constituent
descriptors alone, but can form model(s) in combination with a
descriptor from another such class), and secondary contributors
(category III: a descriptor class forming model(s) only in
combination with the category I). The last category is noncon-
tributors, a descriptor class not forming a model in any manner
such as in categories I, II, or III. The descriptor classes provide
scope to understand the phenomenon under investigation in
relation to the concepts embedded in them.
Figure 1. Substituted-2H-pyran-3(6H)-ones associated with activity
against Gram-positive bacteria; R1 and R2 are alkyl and aryl groups,
respectively.
The Morita-Baylis-Hillman (MBH) reaction41-43 of 4, 5,
and 6 with various aliphatic and aromatic aldehydes in the
presence of TiCl4/TBAI in DCM at -78 to -30 °C yielded
highly diastereoenriched 4a-h, 5c, and 6c-h (Scheme 2). The
reaction is almost completely diastereoselective (>99%) without
using any chiral catalyst. It is due to the formation of an ortho-
ester intermediate during the reaction.44 Making use of our
previously reported procedure,44 the C-4 keto group of 4a-f,
4h, 5c, 6c, 6f, and 6h were stereoselectively reduced with
NaBH4 in the presence of CeCl3‚7H2O, leading to the formation
of their erythro derivatives, 4ar-fr, 4hr, 5cr, 6cr, 6fr, and 6hr,
respectively, in very good yields (Scheme 2). All the compounds
were purified by column chromatography and characterized by
elemental analyses, IR, NMR, and FABMAS spectra.
Among the analogues, the crystals of 4dr were obtained on
slow evaporation of its solution in acetone-hexane (1:3) at room
temperature. Figure 2 is the ORTEP diagram of 4dr from the
X-ray study. It crystallized in monoclinic form with P21 space
group. The central hexenopyranose ring is in a distorted half
chair conformation with unsaturation between C-2 and C-3. In
the Cahn-Ingold-Prelog system,45 the relative conformations
of C-1, C-4, C-5, and C-1′ are assigned as S, S, R, and R,
respectively.
Results and Discussion
Among the highly active analogues, 4d carries a trimethy-
lacetyl protection group at the C-6 position and a p-cyanophenyl
methyl side chain at the C-3 position of hexenopyranoside ring.
In 4d, the replacement of C-3 p-cyanophenyl methyl group by
a p-nitrophenyl methyl (4c) or decyl moiety (4h) has resulted
in lowering the activity of the corresponding analogues.
Compound 5c, a C-6 acetyl variant of 4c, showed remarkably
high activity (1.56 µg/mL). However, in these analogues,
replacement of the C-6 trimethylacetyl or acetyl by a p-
nitrobenzoyl group resulted in loss of their antitubercular activity
(6c-h).
The compound 4hr formed by the reduction of the C-4
carbonyl of 4h showed better antitubercular activity (1.56 µg/
mL) than 4h itself (6.25 µg/mL). However, the other reduced
compounds (e.g., 4dr and 5cr) were found to be far less active
than their precursors. In the case of compounds with a C-6
p-nitrobenzoyl group, the C-4 carbonyl reduction has little effect
on the activity. For example, 6cr, 6fr, and 6hr showed almost
the same activity as 6c, 6f, and 6h. These results thus suggest
that in 2,3-dideoxy hexenopyranosides, the nature of the side
chain at C-3 and the protection group at C-6 play an important
role in determining the antitubercular activity of these com-
pounds.
Biology. The compounds were evaluated for their in Vitro
antitubercular activity against M. tuberculosis H37RV in Agar
dilution assay.46 The minimum concentrations of the compounds
required for the complete inhibition of the bacterial growth per
spot (minimum inhibitory concentration, MIC) are summarized
in Table 1. Ofloxacin was used as the standard drug (MIC is
0.75 µg/mL). Among these analogues, 4d, 4h, 5c, and 4hr
showed activity at MICs 3.12, 6.25, 1.56, and 1.56 µg/mL,
respectively.
QSAR Analysis. The alignment-free 3D-descriptors of X-ray
crystal structure (4dr) database of hexenopyranosides (Table
1) did not lead to any model for the antitubercular activity.
Under identical conditions, three 3D-descriptors classes, namely
GEO (geometrical descriptors),36 WHIM (weighted holistic
invariant molecular descriptors),26 and GETAWAY (geometry,
topology, and atom-weights assembly descriptors)30,31 of the
minimum energy conformation of high active compound 5c have
evolved as collective contributors (category II) to explain the
activity. At the end of a search for four parameter models, 50
descriptors (Table 2) have emerged from these descriptor classes
to model the activity. The following are selected three- and four-
parameter models52 for the activity from a pool of equations.
Cytotoxicity was investigated for 4d, 4h, 5c, and 4hr in
VERO cell line47 and in mouse macrophage cell line J 744A.1.48
In VERO cell line, 4h, 5c, and 4hr showed cytotoxicity at 25
µg/mL and 4d showed no toxicity even at 100 µg/mL. In mouse
macrophage cell line J 744A.1 assay, the compounds were tested
at 12.5, 25, and 50 µg/mL, in parallel with standard antituber-
cular drugs, Rifampicin and Sparfloxacin. These compounds
did not show any toxicity.
QSAR. The QSAR study was attempted with two structure
databases representing the X-ray structure 4dr (Figures 2, 3a)
and the conformational analysis-derived minimum energy
structure of a high active compound 5c (Figure 3b). The
conformational space of these two structure databases differ in
their C-3 substitution region (Figure 3). Alignment-free 3D-
descriptors from DRAGON software36 were opted as indices
of the conformational space. It resulted in 681 and 679
descriptors, respectively, for 4dr and 5c databases. The com-
putational procedures of embedded structural information divide
these indices into eight descriptor classes.36
-log MIC ) -3.667(0.558)PJI3 - 5.204(2.110)E3v +
11.832(3.735)R1u+ + 7.654
n ) 23, r ) 0.835, Q2 ) 0.508, s ) 0.251, F ) 14.61 (1)
The QSAR models have been developed using CP-MLR
(combinatorial protocol in multiple linear regression) in con-
junction with a three-stage descriptor classification protocol.49,50
CP-MLR is a variable selection procedure for the model
development.49-51 The descriptor classification protocol50 typi-
-log MIC ) -3.571(0.518)PJI3 - 14.986(4.940)ISH +
40.044(10.726)HATS2p - 59.740(17.397)R6e+ + 19.290
n ) 23, r ) 0.879, Q2 ) 0.573, s ) 0.224, F ) 15.23 (2)

2944 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13
Saquib et al.
Scheme 1a
a Reagents: (a) (CH3)2CHOH, THF, I2, 3 h, rt; (b) NaOMe, MeOH, 2 h, rt; (c) MnO2, CHCl3, 48 h, rt; (d) (CH3)3CCOCl/(CH3CO)2O/p-O2N-Ph-COCl,
pyridine, 6-48 h, 0-5 °C.
Scheme 2a
a Reagents: (a) TiCl4, TBAI, R2CHO, DCM, 8-48 h, -78 °C f -30 °C; (b) NaBH4, CeCl3‚7H2O, C2H5OH, 2-5 h, rt.
Table 1. Observed and Predicted Antitubercular Activity of
2,3-Dideoxy Hexenopyranosides (Scheme 2)
-log MICb
compd
no.
MICa
(µg/mL)
obsd
eq 1
eq 2
PLS
4a
4b
4c
4d
4e
4f
4g
4h
5c
6c
6d
6e
6f
6g
6h
4ar
4br
4cr
4dr
4er
4fr
4hr
5cr
6cr
6fr
6hr
50
50
50
3.12
NA
50
3.93
3.93
3.93
5.11
4.05
4.06
4.29
4.54
4.15
4.12
3.98
5.02
4.94
4.08
4.46
4.16
4.13
4.47
4.38
4.28
3.87
4.44
4.60
4.45
4.08
5.00
4.90
4.17
4.14
4.36
3.78
4.2
4.03
3.87
4.11
4.83
4.15
4.04
4.34
5.02
5.03
4.38
4.66
4.16
4.36
4.04
4.25
4.00
3.94
4.3
4.85
4.28
4.07
5.19
4.46
4.56
4.26
4.39
4.29
4.66
4.27
4.08
4.03
5.05
4.99
4.1
4.78
4.08
4.18
4.17
4.49
3.99
3.97
4.55
4.76
4.35
4.31
4.92
4.46
4.54
4.03
4.38
3.95
4.14
4.83
5.39
4.29
4.27
4.26
4.31
4.21
4.29
4.23
4.23
-
25
6.25
1.56
25
25
25
25
25
25
25
Figure 2. The ORTEP diagram (30% probability) of 4dr with atomic
numbering scheme.
In the randomization study, none of the identified models
has shown any chance correlation. These are further validated
through two test sets corresponding to descriptor clustering and
random selection procedures with each one containing eight out
of twenty-three compounds of Table 1. The test sets predictions
are in agreement with their experimental values (Table 3;
Figure 5).
25
NA
NA
25
25
1.56
25
25
25
25
-
4.20
4.25
5.44
4.18
4.29
4.31
4.30
Among the descriptors, PJI3 (3D Petitjean shape index; GEO
class) has the highest influence on the activity of these
analogues. It is a measure of eccentricity in the molecules. Its
regression coefficient suggests that molecular structures with
compact conformational features are favorable for activity. In
eq 1, E3v is a directional WHIM descriptor for axial density
(v; weighted by atomic van der Waals volumes) along principal
axes 3. Its regression coefficient suggests in favor of reduced
axial density for enhancement in activity. The GETAWAY
descriptors ISH, R1u+, R6e+, and HATS2p (eqs 1 and 2) are
embedded with the information from the molecular influence
matrix (MIM, H-matrix). In this class, ISH (standardized
information content on the leverage equality) encodes molecular
symmetry information. It suggests in favor of symmetry in
molecules for better activity. The descriptors R1u+ and R6e+,
a NA: not active up to 50 µg/mL. b MIC is in moles per liter.
respectively, represent the influence of maximal autocorrelations
of lag 1 (u; unweighted) and lag 6 (e; weighted by atomic
Sanderson electronegativities) of R-matrix (MIM as a fraction
of distance matrix) on the activity. The descriptor HATS2p is
autocorrelation of lag 2 (p; weighted by atomic polarizabilities)
from the MIM. The regression coefficients of these descriptors
suggest in favor of increasing number of one- and two-centered
fragments in the chosen molecular conformation for better
activity. Besides the three- and four-descriptor equations, a

2,3-Dideoxy Hexenopyranosides as Antitubercular Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13 2945
Figure 4. Schematic representation of categorization of descriptor
classes. A, B, C, etc., in circles are descriptor classes, and ai, bi, ci,
etc., respectively, are their constituent descriptors. Depending on the
information content to correlate the activity, the descriptor classes are
put into four categories as primary contributors (category I), collective
contributors (category II), secondary contributors (category III), and
non-contributors (category IV). Coefficients of ai, bi, ci, etc., are not
shown.
pounds (Table 2; descriptor no. 3 in Figure 6). The other
significant descriptors in the PLS model are SPH, DISPe (GEO),
Gu, G1v, G3v, L3u, E3e (WHIM), ISH, and R3m+ (GET-
AWAY) (Table 2; descriptor nos. 2, 5, 18, 11, 12, 9, 14, 19,
and 40 in Figure 6). Among them, SPH (spherosity) is an
anisometry descriptor and DISPe is a COMMA228 descriptor.
The coefficient SPH suggests in favor of flat molecules for the
activity. DISPe is a measure of displacement between the
geometric and the electronegativity (e) centers in the molecule.
Its negative regression coefficient suggests that an increasing
separation between geometric and electronegativity centers is
detrimental to the activity.
Both ISH (GETAWAY) and Gu (global molecular symmetry,
unweighted; WHIM) infer about molecular symmetry (eq 2;
Table 2). They suggest in favor of symmetry in the compounds
for the activity. The directional WHIM descriptors G1v, G3v,
L3u, and E3e (weighted by v, u, or e) represent the projections
of symmetry (G), dimension (L), and density (E) of the atoms
along respective principal axis 1, 2, or 3. Among these, the
coefficients of G1v, G3v, and L3u (Table 2) suggest in favor
of larger 1 and 3 axial symmetries and dimension for better
activity. The coefficient of E3e (Table 2) suggests a reduced
axial density, weighed by electronegativity. The descriptor
R3m+ (GETAWAY) is R maximal autocorrelation of lag 3
weighted by atomic mass (m). Its coefficient (Table 2) suggests
three-centered structural fragments with minimum atomic
masses. Collectively, the analysis suggests that minimizing of
PJI3, SPH, DISPe, Gu, E3e, ISH, and R3m+ and maximizing
of G1v, G3v, and L3u in the compounds would lead to improved
activity. In terms of molecular features, an extended structural
frame, reduced symmetry, and separation in geometric and
electronegativity centers are detrimental to the activity. At the
same time the compounds with conformational features of
Figure 3. (a) X-ray structure of 4dr; (b) conformational analysis
derived minimum energy structure of 5c; (c) superimposition pose
of 4dr (carbons: green) and 5c (carbons: yellow); in all, oxygens
are in red and nitrogens are in blue; hydrogens are suppressed for
clarity.
number of higher models also exist among the identified
descriptors (Table 2).
A PLS (partial least-squares) analysis has been carried out
on these 50 CP-MLR identified descriptors (Table 2) to facilitate
the development of a ‘single window’ structure-activity model.
In the PLS cross-validation,53 three components are found to
be the optimum for these 50 descriptors (dataset-I) and they
explained 76% variance in the activity (r ) 0.872, Q2 ) 0.588,
s ) 0.223, F ) 20.20). A PLS analysis has also been carried
out on the dataset devoid of these 50 identified descriptors
(dataset-II; 629 descriptors). Under identical conditions, the
dataset-II could explain only 63% variance in the activity (r )
0.794, Q2 ) 0.345, s ) 0.278, F ) 4.55). The dataset of 681
descriptors of the X-ray structure (4dr) database has resulted
in a far inferior PLS model and explained only 57% variance
in the activity (r ) 0.758, Q2 ) 0.264, s ) 0.298, F ) 8.54).
These results clearly substantiate the relevance of the 50
identified descriptors (Table 2) in modeling the activity of these
compounds. The MLR-like PLS coefficients of these 50
descriptors are given in Table 2. Figure 6 shows a plot of the
fraction contribution of normalized regression coefficients of
these descriptors to the activity (Table 2).
The PLS analysis has also suggested that PJI3 is the most
determining descriptor for modeling the activity of the com-

2946 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13
Saquib et al.
Table 2. The Identified 3D-Descriptors of Minimum Energy
Conformation (5c)-Directed Structure Database of 2,3-Dideoxy
Hexenopyranosides (Table 1) for Modeling Their Antitubercular Activity
Table 3. Predicted Activity of Descriptor Cluster and Random Test Sets
(eight compounds each) of the Compounds of Table 1 and the
Corresponding Statistics
residualsa
no.
descriptora
regr coeff (fc) orderb
desc clustb
eq 2
randomc
eq 2
Geometrical
1
2
3
4
5
6
7
SPAM
SPH
PJI3
DISPm
DISPe
DISPp
G(N..F)
0.281(0.005)45
-1.760(-0.054)3
-1.411(-0.109)1
-0.006(-0.020)16
-0.441(-0.042)6
-0.046(-0.007)38
0.001(0.009)37
compd
eq 1
PLS
eq 1
PLS
4a
4b
4c
4f
6c
-0.17
-0.13
0.09
-0.32
-0.45
-0.13
-0.31
-0.10
-0.33
-0.07
-0.36
-0.31
-0.17
0.15
-0.12
0.17
-0.34
-0.01
WHIM
6e
0.10
-0.26
0.11
0.01
0.45
0.16
8
9
L1u
L3u
L1v
G1v
G3v
E3v
E3e
L1p
G2p
Tp
-0.010(-0.015)24
0.120(0.039)7
-0.008(-0.014)26
16.783(0.056)2
6g
6h
4br
4hr
5cr
6cr
6fr
r
-0.04
-0.11
0.01
10
11
12
13
14
15
16
17
18
0.33
0.43
0.24
0.46
0.25
0.33
0.30
-0.66
0.39
-0.26
0.50
-0.46
9.568(0.036)9
-0.622(-0.013)28
-1.148(-0.038)8
-0.008(-0.013)29
1.283(0.006)42
0.001(0.002)47
-16.095(-0.048)4
0.12
0.11
-0.25
0.19
0.891
0.381
0.688
0.22
-0.16
0.30
0.14
0.30
0.21
0.929
0.745
0.516
0.165
22.95
0.851
0.429
0.64
0.234
9.64
0.874
0.455
0.736
0.226
8.11
0.818
0.395
0.681
0.267
7.41
0.894
0.622
0.532
0.208
14.58
Q2
R2
Pred
Gu
s
F
GETAWAY
9.7
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
ISH
HIC
H6u
HTu
HATS0u
HATS6u
HATS0v
HATS1v
H6e
-6.043(-0.044)5
0.019(0.003)46
0.051(0.010)31
0.002(0.009)35
0.505(0.010)33
-0.844(-0.019)17
3.705(0.030)11
2.745(0.018)18
0.026(0.005)44
-0.669(-0.016)22
0.169(0.009)36
3.732(0.027)15
3.018(0.018)19
0.095(0.001)49
-2.052(-0.014)27
0.049(0.005)43
0.097(0.012)30
-0.023(-0.002)48
2.388(0.029)12
0.587(0.017)21
-0.562(-0.016)23
-5.194(-0.031)10
-1.785(-0.007)41
0.482(0.014)25
2.188(0.007)39
0.110(0.010)32
-0.0002(-0.0003)50
-0.826(-0.017)20
-3.823(-0.028)14
-8.606(-0.029)13
0.267(0.009)34
2.188(0.007)40
a Difference of observed and predicted -log MIC; each training set has
15 compounds. b Test set from the descriptors clustering. c Test set from
random selection.
enhanced dimension and symmetries along their principal axis
1 and 3 are favorable for the activity. In comparison to these
10 descriptors, the remaining ones appear in lower order of
significance to influence the activity of the compounds (Table
2; Figure 6). A brief description of all the 50 descriptors is given
in Table 2.
HATS6e
H1p
HATS0p
HATS1p
HATS2p
HATS3p
R2u
Conclusion
R3u
R6u
This study has provided a rational approach for the develop-
ment of novel antitubercular agents based on 2,3-dideoxy hex-
2-enopyranoside from carbohydrate synthon 3,4,6-tri-O-acetyl-
R-D-glucal (1). Employing the MBH reaction, we conveniently
introduced alkyl/arylalkyl groups at C-3 of hexenopyranosides,
leading to this new class of molecules. The synthesis is almost
completely diastereoselective without the usage of any external
chiral agents. The correlation of the activity of these analogues
with the 3D-descriptors emanating from the minimum energy
conformation of 5c suggests its relevance to the antitubercular
activity. The conformational space of the minimum energy
structure from conformational analysis (5c) and the X-ray
structure (4dr) differ in their C-3 substitution region (Figure
3). Collectively, the identified descriptors suggest in favor of
compounds embedded with some degree of symmetry, compact
conformational features, and closely placed geometric and
electronegativity centers for enhanced antitubercular activity.
The involvement of the directional WHIM descriptors in the
correlations suggests the complex nature of interaction between
the compounds and their target. The autocorrelation descriptors
from GETAWAY suggest in favor of three-centered structural
fragments. Among the high active compounds, 4d has shown
no toxicity. It is identified for further development.
R1u+
R1m+
R2m+
R3m+
R6m+
RTm+
R2v+
R3e
RTe
R2e+
R3e+
R6e+
R1p
R2p+
a Geometrical: SPAM, average span sphere radius; SPH, spherosity; PJI3,
3D Petitjean shape index; DISPw, displacement between the geometric
center and the center of the w property field, calculated with respect to the
molecular principal axes, the last character ‘w’ refers to the weighting
property which may be either ‘m’ for atomic masses or ‘V’ for atomic van
der Waals volumes or ‘e’ for Sanderson electronegativities or ‘p’ for atomic
polarizabilities or ‘u’ for unweighted; G(N..F) sum of geometrical distances
between N and F. WHIM: Lnw, Gnw, and Enw, directional indices of size,
symmetry, and accessibility, respectively; in this, n refers to the directional
component 1, 2, or 3, and ‘w’ refers to the weighting property (for ‘w’ see
DISPw); Tw and Gw, global indices of size and symmetry weighted by
property w. GETAWAY: ISH, standardized information content on the
leverage equality; HIC, mean information content on the leverage magnitude;
Hkw and Rkw, H (molecular influence matrix; MIM) and R (MIM/distance
matrix) autocorrelation descriptors of lag k weighted by property w; lag k
refers to the number of edges in the fragment unit considered in the
computation (for ‘w’ see DISPw); HTu, H total index unweighted; HATSkw,
leverage-weighted autocorrelation of lag k weighted by property w; Rkw+,
R maximal autocorrelation of lag k weighted by property w; see ref 36.
b MLR-like regression coefficient of three-component PLS model; (fc) is
fraction contribution of the regression coefficient to the activity; order
indicates the order of their significance in the PLS model; the constant
term of PLS model is 10.328; number of compounds are 23.
Experimental Section
General Methods. All the reactions were monitored by warming
the CeSO4 (1% in 1 M H2SO4) sprayed precoated silica gel TLC
plates at 100 °C. NMR spectra were recorded on Bruker Avance
DPX 200 FT, Bruker Robotics, and Bruker DRX 300 spectrometers
at 200, 300 (1H) and 50, 75 MHz (13C). For 13C NMR, reference
CDCl3 appeared at 77.4 ppm, unless otherwise stated. Mass spectra
were recorded on a JEOL SX 102/DA 6000 mass spectrometer

2,3-Dideoxy Hexenopyranosides as Antitubercular Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13 2947
Figure 5. Plots of training (2) and test sets (O) predicted activities versus observed activity corresponding to eqs 1 (a, b) and 2 (c, d). Number
of compounds in training set is 15 and in test set is 8.
Figure 6. Plot of fraction contribution of MLR-like PLS coefficients (normalized) of the 50 descriptors (Table 2) to the activity. The horizontal
axis refers to descriptors’ serial numbers. The ten most significant descriptors are identified by gray shaded lines.
using argon/xenon (6 kV, 100 mA) as the FAB gas. IR spectra
were recorded on Perkin-Elmer 881 and FTIR-8210 PC Shimadzu
spectrophotometers. Optical rotations were determined on an
Autopol III polarimeter using a 1 dm cell at 28 °C in methanol or
chloroform as the solvent; concentrations mentioned are in g/100
mL. Elemental analyses were carried out on Carlo-Erba-1108 and
Vario EL III instruments. A Bruker P4 diffractometer equipped
with graphite monochromted Mo KR radiation (0.71073 Å) at 293-
(2) K was used for the X-ray Crystallographic study. XSCANS
(Version 2.21. Siemens Analytical X-ray Instruments Inc., Madison,
WI, 1996) was used for the data collection and reduction.
SHELXTL-NT (Version 5.1. Bruker AXS Inc., Madison, WI, 1997)
was used for structure refinement. Organic solvents were dried by
standard methods. Aldehydes were purchased from Aldrich Chemi-
cal Co. and Fluka Chemical Co. Enones 4, 5, and 6 were synthesized
in the lab.
After stirring for 2 min, a mixture containing enuloside 4, 5, or 6
(1 mmol) and the respective aldehyde (2 mmol) in dry CH2Cl2 (5
mL) was added. The reaction mixture was slowly warmed to
-30 °C and stirred for the specified time depending on the
enulosides and aldehydes used. A saturated aqueous solution of
sodium bicarbonate was added, followed by filtration through a
celite pad. The organic layer from the filtrate was separated, and
the aqueous layer was again extracted with ethyl acetate. The
organic layers were then combined, washed with brine solution,
and dried over sodium sulfate. The crude product obtained after
evaporation of the solvent was chromatographed to yield the pure
compounds. Compounds 4a-c, 4e-h, 5c, and 6c are reported in
previous publication.44
General Procedure for the Preparation of 4ar-6hr. To a
stirred solution of 4a-6h in ethanol (10 mL) were added NaBH4
(0.5mmole) and CeCl3‚7H2O (1 mmol), and the reaction mixture
was stirred continuously at room temperature for the specified time
depending on the compounds to be reduced. After completion of
the reaction (TLC control), excess NaBH4 was neutralized with
General Procedure for the Preparation of 4a-6h. To a stirred
solution of tetrabutylammonium iodide, TBAI (0.2 mmol), in dry
CH2Cl2 (5 mL) at -78 °C was added TiCl4 (1.5 mmol) dropwise.

2948 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13
Saquib et al.
acetone, and the reaction mixture was concentrated in vacuo to get
crude product which upon column chromatography yielded the pure
4ar-6hr. Compounds 4ar, 4cr, and 4hr are reported in previous
publication.44
Descriptor Classification Protocol. The three-stage descriptor
classification protocol50 is implemented with the two-descriptor
combinations (baseline models), as they are the simplest to
understand and explain the activity. In the first stage of the
classification protocol, the correlations of the activity with two-
descriptor combinations from the individual descriptor classes (DCs)
of the dataset were used to sort the DCs into four categories. They
are primary contributors (category I: a DC forms a model with its
constituent descriptors), collective contributors (category II: a DC
unable to form a model with its constituent descriptors, but forms
model(s) in combination with a descriptor from another such DC),
secondary contributors (category III: a DC forms a model(s) only
in combination with category I) and noncontributors (category IV:
a DC unable to form a model(s) in any manner like that of category
I, II, or III). The sorted DCs were collated in the second stage to
identify all the 3-descriptor models across the categories. In the
last stage, the individual descriptors of all three-descriptor models
were pooled to discover the higher models for the activity.
All the identified models were reassessed for chance correlations
by repeated randomization of the activity.51 Each identified model
was subjected to 100 simulation runs with scrambled activity, and
the emerging correlations were counted to express the percent
chance correlation of the model under examination. The proposed
models were validated through two test sets, one from the single
linkage hierarchical clustering of all the descriptor of the compounds
and the other from the random selection procedure, with each
containing 8 out of 23 compounds in analysis. The descriptors
identified in the CP-MLR were evaluated through the PLS
procedure53 as well.
Biological Activity. Agar Dilution Method.46 Briefly, 2-fold
serial dilutions of each test compound/drug were incorporated into
7H10 agar. Inoculum of M. tuberculosis H37RV was prepared from
fresh Lowenstein-Jensen slants adjusted to 1 mg/mL (wet weight)
in Tween 80 (0.05%) saline and diluted to 10-2 to give a
concentration of approximately107 cfu/mL. A 5 mL amount of
bacterial suspension was spotted into 7H10 agar tubes containing
2-fold serial dilution of drugs per mL. The tubes were incubated at
37 °C, and final readings were recorded after 30 days. The MICs
were read as the minimum concentration of drugs/compounds that
completely inhibited the growth of M. tuberculosis per spot.
Ofloxacin was used as the standard drug.
Cytotoxicity Studies. The cytotoxicity study was carried out in
VERO cell line.47 VERO cells, originally obtained from ATCC and
maintained in CDRI, were seeded on a 96-well plate and incubated
at 37 °C in 5% CO2 and 95% air. Cells were exposed to varying
concentrations (25, 50, 100 µg/mL) of the compound for 72 h. MTT
[3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide]
assay was performed as per the standard protocol to assess cell
viability. It is based on the MTT reduction. Using the standard
protocol, cytotoxicity was also studied in mouse macrophage cell
line J 744A.1.48
QSAR Study. Dataset. Two structure databases of hexenopy-
ranoside derivatives (Table 1), corresponding to the X-ray structure
(4dr) and the minimum energy structure from the conformational
analysis of a high active compound (5c), were created for the
computation of molecular descriptors. The database structures of
X-ray structure (4dr) were generated in MOE (molecular operating
environment) software54 by making appropriate changes to the
crystal structure template followed by energy minimization in AM1
semiempirical method55 (rms gradient 0.001) as implemented in
the software to result in the final conformations. For the second
structure database, the minimum energy conformation of 5c
was identified through its systematic conformational analysis in
MOE with default parameter settings followed by energy
minimization (force field: MMFF94)56 and refinement of final
conformation in AM1 method. The remaining structures of the
database were generated from this template of 5c by appending
appropriate changes to it followed by structural refinement in AM1
method.
Acknowledgment. M.K.G. and R.S. thank CSIR, New
Delhi, India, for the financial support in the form of Senior
Research Fellowships. Authors thank Sophisticated Analytical
Instrument Facility, CDRI, for the spectral data, and Mr. Anoop
Kishore Pandey for technical assistance. CDRI Communication
No. 6717.
Supporting Information Available: Characterization data for
all new compounds 4br, 4d, 4dr, 4fr, 4er, 5cr, 6cr, 6d, 6e, 6f,
6fr, 6g, 6h, and 6hr; purity table showing elemental analyses of
1
listed above compounds; H NMR and 13C NMR spectra of 4d,
4dr, 5c, 5cr, 6h, and 6hr; crystallographic information file (CIF)
of 4dr; molecular modeling coordinates of 5c; complete data set
of alignment free 3D-molecular descriptors of structure databases
corresponding to the X-ray crystal structure of 4dr and the minimum
energy conformation of 5c; PLS loadings, weights, and sensitivity
of independent and dependent descriptors of PLS model. This
material is available free of charge via the Internet at http://
pubs.acs.org.
DRAGON software36 was used for the computation of alignment
free 3D-descriptors of the structure databases. This resulted in 681
and 679 descriptors, respectively, for the X-ray (4dr) and the
minimum energy conformation (5c) structure databases. The
antitubercular activity (MIC, in moles per liter) of the compounds
(Table 1) was considered after its transformation into logarithm of
the inverse of inhibitory concentration (-log MIC). The QSAR
models were generated using CP-MLR49 in conjunction with a three-
stage descriptor classification protocol50 and PLS analysis.53
CP-MLR. It is a ‘filter’-based variable selection procedure for
the model identification and development in QSAR studies.49-51 It
involves a combinatorial strategy with appropriately placed ‘filters’
interfaced with MLR and extracts diverse models having unique
combination of descriptors from the dataset. The filters set the
thresholds for the descriptors in terms of interparameter correlation
cutoff limits in subset regressions (filter-1), t-values of the regression
coefficients (filter-2), internal explanatory power (filter-3; square-
root of adjusted multiple correlation coefficient of regression
equation, r-bar), and the external consistency (filter-4; Q2 i.e. cross-
validated R2 from the leave-one-out procedure). Throughout this
study, for the filters-1, 2, and 4 the thresholds were assigned as
0.79, 2.0, and 0.3 e Q2 e 1.0, respectively. The filter-3 was
assigned an initial value of 0.71. In order to collect the descriptors
with higher information content, the threshold of filter-3 was
successively incremented with increasing number of descriptors (per
equation) by considering the r-bar value of the preceding optimum
model as the new threshold for next generation.
References
(1) (a) WHO Tuberculosis Fact Sheet, 2004. http://www.who.int/me-
diacentre/factsheets/fs104/en/. (b) WHO Tuberculosis Fact Sheet,
1998. http://www.who.int/gtb/publications/factsheet/index.htm.
(2) (a) Dolin, P. J.; Raviglione, M. C.; Kochi, A. Global tuberculosis
incidence and mortality during 1990-2000. Bull. WHO 1994, 72,
213-220. (b) Daffe, M.; Draper, P. The envelope layers of
mycobacteria with reference to their pathogenicity. AdV. Microb.
Physiol. 1998, 39, 131-203.
(3) Dye, C.; Scheele, S.; Dolin, P.; Pathania, V.; Raviglione, M. C. Global
burden of tuberculosis: Estimated incidence, prevalence, and mortal-
ity by country. JAMA 1999, 282, 677-686.
(4) Somoskovi, A.; Parsons, L. M.; Salfinger, M. The molecular basis
of resistance to Isoniazid, Rifampin, and Pyrazinamide in Mycobac-
terium tuberculosis. Respir. Res. 2001, 2, 164-168.
(5) Khasnobis, S.; Escuyer, V. E.; Chatterjee, D. Emerging therapeutic
targets in tuberculosis: Post-genomic era. Expert Opin. Ther. Targets
2002, 6, 21-40.
(6) Maddry, J. A.; Suling, W. J.; Reynolds, R. C. Glycosyltransferases
as targets for inhibition of cell wall synthesis in M. tuberculosis and
M. avium. Res. Microbiol. 1996, 147, 106-112.
(7) Reynolds, R. C.; Bansal, N.; Rose, J.; Friedrich, J.; Suling, W. J.;
Maddry, J. A. Ethambutol-sugar hybrids as potential inhibitors of
mycobacterial cell-wall biosynthesis. Carbohydr. Res. 1999, 317,
164-179.

2,3-Dideoxy Hexenopyranosides as Antitubercular Agents
Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13 2949
(8) Jones, P. B.; Parrish, N. M.; Houston, T. A.; Stapon, A.; Bansal, N.
P.; Dick, J. D.; Townsend, C. A. A new class of antituberculosis
agents. J. Med. Chem. 2000, 43, 3304-3314.
(29) Pastor, M.; Cruciani, G.; McLay, I.; Pickett, S.; Clementi, S. Grid-
independent descriptors (GRIND): a novel class of alignment-
independent three-dimensional molecular descriptors. J. Med. Chem.
2000, 3, 3233-3243.
(30) Consonni, V.; Todeschini, R.; Pavan, M. Structure/response correla-
tions and similarity/diversity analysis by GETAWAY descriptors.
1. Theory of the novel 3D molecular descriptors. J. Chem. Inf.
Comput. Sci. 2002, 42, 682-692.
(31) Consonni, V.; Todeschini, R.; Pavan, M.; Gramatica, P. Structure/
response correlations and similarity/diversity analysis by GETAWAY
descriptors. 2. Application of the novel 3D molecular descriptors to
QSAR/QSPR studies. J. Chem. Inf. Comput. Sci. 2002, 42, 693-
705.
(32) Bagchi, M. C.; Maiti, B. C.; Mills, D.; Basak, S. C. Usefulness of
graphical invariants in quantitative structure-activity correlations of
tuberculostatic drugs of the isonicotinic acid hydrazide type. J. Mol.
Model. 2004, 10, 102-111.
(33) Bagchi, M. C.; Mills, D.; Basak, S. C. Quantitative structure-activity
relationship (QSAR) studies of quinolone antibacterials against M.
fortuitum and M. smegmatis using theoretical molecular descriptors.
J. Mol. Model. 2007, 13, 111-120.
(34) Cratteri, P.; Romanelli, M. N.; Cruciani, G.; Bonaccini, C.; Melani,
F.; GRIND-derived pharmacophore model for a series of alpha-
tropanyl derivative ligands of the sigma-2 receptor. J. Comput.-Aided.
Mol. Des. 2004, 18, 361-374.
(35) Gonzalez, M. P.; Suarez, P. L.; Fall, Y.; Gomez, G. Quantitative
structure-activity relationship studies of vitamin D receptor affinity
for analogues of 1alpha,25-dihydroxyvitamin D3. 1: WHIM descrip-
tors. Bioorg. Med. Chem. Lett. 2005, 15, 5165-5169.
(36) Todeschini, R.; Consonni, V. Dragon software (version 3.0-2003),
Milano, Italy, and references cited therein. http://www.talete.mi.it.
(37) Roth, W.; Pigman, W. In Methods in Carbohydrate Chemistry;
Whistler, R. L., Wolfrom, M. L., Eds.; Academic Press Inc.: New
York, 1963; Vol. 11, pp 405-407.
(38) Ferrier, R. J. Substitution-with-allylic-rearrangement reactions of
glycal derivatives. In Topics in Current Chemistry; Springer-
Verlag: Berlin, 2001; Vol. 215, pp 153-175 and references cited
therein.
(39) Prevost, N.; Rouessac, F. A novel route to pyrimidine isodideoxy-
nucleosides via Michael-type addition on unsaturated modified sugars.
Synth. Commun. 27, 1997, 2325-2335.
(40) Fraser-Reid, B.; McLean, A.; Usherwood, E. W.; Yunker, M.
Pyranosiduloses. II. The synthesis and properties of some alkyl 2,3-
dideoxy-2-enopyranosid-4-uloses. Can. J. Chem. 1970, 48, 2877-
2884.
(41) (a) Morita, K-i.; Suzuki, Z.; Hirose, H. A Tertiary phosphine-catalyzed
reaction of acrylic compounds with aldehydes. Bull. Chem. Soc. Jpn.
1968, 41, 2815. (b) Baylis, A. B.; Hillman, M. E. D. German Patent
2155113, 1972; Chem. Abstr. 1972, 77, 34174q.
(42) (a) Basavaiah, D.; Rao, A. J.; Satyanarayana, T. Recent advances in
the Baylis-Hillman reaction and applications. Chem. ReV. 2003, 103,
811-891. (b) Langer, P. New strategies for the development of an
asymmetric version of the Baylis-Hillman reaction Angew. Chem.,
Int. Ed. 2000, 39, 3049-3052. (c) Ciganek, E. The Catalyzed
R-hydroxyalkylation and R-aminoalkylation of activated olefins
(The Morita-Baylis-Hillman Reaction) In Organic Reactions;
Paquette, L. A., Ed.; Wiley: New York; 1997; Vol. 51, pp 201-
350. (d) Basavaiah, D.; Rao, P. D.; Hyma, R. S. The Baylis-Hillman
(9) Pasqualoto, K. F. M.; Ferreira, E. I. An approach for the rational
design of new antituberculosis agents. Curr. Drug Targets 2001, 2,
427-437.
(10) Teodori, E.; Dei, S.; Scapecchi, S.; Gualtieri, F. The medicinal
chemistry of multidrug resistance (MDR) reversing drugs. Il Farmaco
2002, 57, 385-415.
(11) Frieden, T. R.; Sterling, T. R.; Munsiff, S. S.; Watt, C. J.; Dye, C.
Tuberculosis. Lancet 2003, 362, 887-899.
(12) Smith, C. V.; Sharma, V.; Sacchettini, J. C. TB drug discovery:
Addressing issues of persistence and resistance. Tuberculosis 2004,
84, 45-55.
(13) Bayles, K. W. The bactericidal action of penicillin: New clues to
an unsolved mystery. Trends Microbiol. 2000, 8, 274-278.
(14) Katz, A. H.; Caufield, C. E. Structure-based design approaches to
cell wall biosynthesis inhibitors. Curr. Pharm. Des. 2003, 9, 857-
866.
(15) McNeil, M.; Wallner, S. J.; Hunter, S. W.; Brennan, P. J. Demonstra-
tion that the galactosyl and arabinosyl residues in the cell-wall
arabinogalactan of Mycobacterium leprae and Myobacterium tuber-
culosis are furanoid. Carbohydr. Res. 1987, 166, 299-308.
(16) Daffe, M.; Brennan, P. J.; McNeil, M. Predominant structural features
of the cell wall arabinogalactan of Mycobacterium tuberculosis as
revealed through characterization of oligoglycosyl alditol fragments
by gas chromatography/mass spectrometry and by 1H and 13C NMR
analyses. J. Biol. Chem. 1990, 265, 6734-6743.
(17) Lee, R. E.; Smith, M. D.; Nash, R. J.; Griffiths, R. C.; McNeil, M.;
Grewal, R. K.; Yan, W.; Besra, G. S.; Brennan, P. J.; Fleet, G. W.
J. Inhibition of UDP-Gal mutase and mycobacterial galactan bio-
synthesis by pyrrolidine analogues of galactofuranose. Tetrahedron
Lett. 1997, 38, 6733-6736.
(18) (a) Pathak, A. K.; Pathak, V.; Seitz, L.; Maddry, J. A.; Gurcha, S.
S.; Besra, G. S.; Suling, W. J.; Reynolds, R. C. Studies on (â,1f5)
and (â,1f6) linked octyl Galfdisaccharides as substrates for myco-
bacterium galactosyltransferase activity. Bioorg. Med. Chem. 2001,
9, 3129-3143. (b) Pathak, A. K.; Pathak, V.; Maddry, J. A.; Suling,
W. J.; Gurcha, S. S.; Besra, G. S.; Reynolds, R. C. Studies on R-
(1f5) linked octyl arabinofuranosyl disaccharides for mycobacterial
arabinosyl transferase activity. Bioorg. Med. Chem. 2001, 9, 3145-
3151.
(19) Wen, X.; Crick, D. C.; Brennan, P. J.; Hultin, P. G. Analogues of
the mycobacterial arabinogalactan linkage disaccharide as cell wall
biosynthesis inhibitors. Bioorg. Med. Chem. 2003, 11, 3579-3587.
(20) Centrone, C. A.; Lowary, T. L. Sulfone and phosphinic acid analogs
of decaprenolphosphoarabinose as potential anti-tuberculosis agents.
Bioorg. Med. Chem. 2004, 12, 5495-5503.
(21) Cren, S.; Gurcha S. S.; Blake A. J.; Besra, G. S.; Thomas N. R.
Synthesis and biological evaluation of new inhibitors of UDPGalf
transferase-a key enzyme in M. tuberculosis cell wall biosynthesis.
Org. Biomol. Chem. 2004, 2, 2418-2420.
(22) Pathak, R.; Shaw, A. K.; Bhaduri, A. P.; Chandrasekhar, K. V. G.;
Srivastava, A.; Srivastava, K. K.; Chaturvedi, V.; Srivastava, R.;
Srivastava, B. S.; Arora, S.; Sinha, S. Higher acyclic nitrogen
containing deoxy sugar derivatives: A new lead in the generation
of antimycobacterial chemotherapeutics. Bioorg. Med. Chem. 2002,
10, 1695-1702.
(23) Pathak, R.; Pant, C. S.; Shaw, A. K.; Bhaduri, A. P.; Gaikwad, A.
N.; Sinha, S.; Srivastava, A.; Srivastava, K. K.; Chaturvedi, V.;
Srivastava, R.; Srivastava, B. S. Baylis-Hillman Reaction: Conve-
nient ascending syntheses and biological evaluation of acyclic deoxy
monosaccharides as potential antimycobacterial agents. Bioorg. Med.
Chem. 2002, 10, 3187-3196.
(24) Gupta, M. K.; Sagar, R.; Shaw, A. K.; Prabhakar, Y. S. CP-MLR
directed QSAR studies on the antimycobacterial activity of func-
tionalized alkenolsstopological descriptors in modeling the activity.
Bioorg. Med. Chem. 2005, 13, 343-351.
reaction:
A
novel carbon-carbon bond forming reaction.
Tetrahedron 1996, 52, 8001-8062. (e) Drewes, S. E.; Roos, G. H.
P. Synthetic potential of the tertiary-amine-catalysed reaction of
activated vinyl carbanions with aldehydes. Tetrahedron 1988, 44,
4653-4670.
(43) Pathak, R.; Shaw, A. K.; Bhaduri, A. P. Chain extension of acyclic
sugar derivatives via the Baylis-Hillman reaction. Tetrahedron 2002,
58, 3535-3541.
(25) (a) Couladouros, E. A.; Strongilos, A. T. Generation of libraries of
pharmacophoric structures with increased complexity and diversity
by employing polymorphic scaffolds. Angew. Chem., Int. Ed. 2002,
41, 3677-3680. (b) Georgiadis, M. P.; Couladouros, E. A.; Delitheos,
A. K. Synthesis and antimicrobial properties of 2H-pyran-3(6H)-
one derivatives and related compounds. J. Pharm. Sci. 1992, 81,
1126-1131.
(26) Todeschini, R.; Lasagni, M.; Marengo, E. New molecular descriptors
for 2D and 3D structures theory: Part 1. J. Chemom. 1994, 8, 263-
272.
(27) Silverman, B. D.; Platt, D. E. CoMMA: Comparative molecular
moment analysis. 3D-QSAR without molecular superposition. J. Med.
Chem. 1996, 39, 2129-2140.
(28) Silverman, B. D. Three-dimensional moments of molecular property
fields. J. Chem. Inf. Comput. Sci. 2000, 40, 1470-1476.
(44) (a) Sagar, R.; Pant, C. S.; Pathak, R.; Shaw, A. K. A substrate
controlled, very highly diastereoselective Morita-Baylis-Hillman
reaction: A remote activation of the diastereofacial selectivity in the
synthesis of C-3-branched deoxysugars. Tetrahedron 2004, 60,
11399-11406. (b) In 33a the MBH-adduct derived reduced products
were reported as in threo conformation. Now they are revised as in
erytho conformation based on the X-ray analysis of compound 4dr.
(c) Sagar, R., Saquib, M., Shaw, A. K., Gaikwad, A. N., Sinha, S.
K., Srivastava, A., Chaturvedi, V., Manju, Y. K., Srivastava, R.,
Srivastava, B. S. C-3 Alkyl or arylalkyl substituted 2,3-dideoxy
glucopyranosides and a process for preparation thereof. Indian Patent
0533DEL2006, Ref. No. 0210NF2005/IN, 2006.
(45) Cahn, R. S.; Ingold, C.; Prelog, V. Specification of molecular chirality.
Angew. Chem., Int. Ed. Engl. 1966, 5, 385-415.

2950 Journal of Medicinal Chemistry, 2007, Vol. 50, No. 13
Saquib et al.
(46) Saito, H.; Tomioka, H.; Sato, K.; Emori, M.; Yamane, T.; Yamashita,
K.; Hosoe, K.; Hidaka, T. In vitro antimycobacterial activities of
newly synthesized benzoxazinorifamycins. Antimicrob. Agents
Chemother. 1991, 35, 542-547.
(52) The accompanied statistics of the equation: n, r, Q2, s, and F
represent, respectively, the number of compounds, correlation coef-
ficient, cross-validated R2 from leave-one-out (LOO) procedure,
standard error of the estimate, and F-test value of the regression.
The values given in parentheses are the standard errors of the
regression coefficients.
(53) Wold, S. Cross-validatory estimation of the number of components
in factor and principal component models. Technometrics 1978, 20,
397-405.
(54) MOE: The Molecular Operating Environment from Chemical
Computing Group Inc., 1255 University St., Suite 1600, Montreal,
Quebec, Canada H3B 3X3. http://www.chemcomp.com.
(55) Dewar, M. J. S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J. J. P.
Development and use of quantum mechanical molecular models. 76.
AM1: a new general purpose quantum mechanical molecular model.
J. Am. Chem. Soc. 1985, 107, 3902-3909.
(56) Halgren, T. A. Merck molecular force field. I. Basis, form, scope,
parameterization, and performance of MMFF94. J. Comput. Chem.
1996, 17, 490-519.
(47) Cory, A. H.; Owen, T. C.; Barltrop, J. A. Use of an aqueous
tetrazolium/formazan assay for cell growth assays in culture. Cancer
Commun. 1991, 3, 207-212.
(48) Srivastava, R.; Deb, D. K.; Srivastava, K. K.; Locht, C.; Srivastava,
B. S. Green fluorescent protein as a reporter in rapid screening of
antituberculosis compounds in Vitro and in macrophages. Biochem.
Biophys. Res. Commun. 1998, 253, 431-436.
(49) Prabhakar, Y. S. A combinatorial approach to the variable selection
in multiple linear regression analysis of Selwood et al. data set- A
case study. QSAR Comb. Sci. 2003, 22, 583-595.
(50) Gupta, M. K.; Prabhakar, Y. S. Topological descriptors in modeling
the antimalarial activity of 4-(3′,5′-disubstituted anilino)quinolines.
J. Chem. Inf. Model. 2006, 46, 93-102.
(51) Prabhakar, Y. S.; Solomon, V. R.; Rawal, R. K.; Gupta, M. K.; Katti,
S. B. CP-MLR/PLS directed structure-activity modeling of the hiv-
1rt inhibitory activity of 2,3-diaryl-1,3-thiazolidin-4-ones. QSAR
Comb. Sci. 2004, 23, 234-244.
JM070110H