Synthesis and 3D-QSAR analysis of 2-chloroquinoline derivatives as H 37 RV MTB inhibitors

Article abstract of DOI:10.1111/cbdd.12178

Frequency of tuberculosis is progressively increasing worldwide. New emerging strains of bacilli that are emerging are resistant to the currently available drugs which make this issue more alarming. In this regard, a series of substituted quinolinyl chalcones, quinolinyl pyrimidines, and pyridines were synthesized and evaluated for their antitubercular activity in vitro against Mycobacterium tuberculosis H₃₇RV. To establish the role of the 2-chloroquinoline nucleus as a pharmacophoric group and study its influence on the antimycobacterial activity, a 3D-QSAR study based on CoMFA and CoMSIA was undertaken on this set of 2-chloroquinoline derivatives. Statistically significant models that are able to well correlate the antimycobacterial activity with the chemical structures of the 2-chloroquinolines have been developed. The contour maps resulting from the best CoMFA and CoMSIA models were used to identify the structural features relevant to the biological activity in this series of analogs. Further analysis of these interaction-field contour maps also showed a high level of internal consistency. The information obtained from the field 3-D contour maps may be fruitfully utilized in the design of more potent 2-chloroquinoline-based analogs as potential antitubercular candidates. A series of 1 chloroquinoline bearing pyridine and pyrimidine ring system and fluoro/bromo at C-3 and C-6 of the quinoline ring were synthesized and evaluated for their antimycobacterial activity. The MIC data of antitubercular activity was performed for 3D QSAR analysis.

Full text of DOI:10.1111/cbdd.12178

Chem Biol Drug Des 2013; 82: 669–684
Research Article
Synthesis and 3D-QSAR Analysis of 2-Chloroquinoline
Derivatives as H37RV MTB Inhibitors
Ranjan C. Khunt1,*, Vijay M. Khedkar2 and Evans
8–10 million people globally and causing 3 million deaths
every year according to a World Health Organization report
C. Coutinho2
(3). Owing to the rapid spread of mycobacterium strains
1Department of Chemistry, Saurashtra University, Rajkot,
resistant to all first-line antitubercular drugs. such as isoni-
360005, India
azid, rifampicin, and ethionamide, and due to the toxicity
of second-line drugs, such as ethionamide, aminosalicylic
2Department of Pharmaceutical Chemistry, Bombay
College of Pharmacy, Kalina, Santacruz (E), Mumbai, 400
098, India
*Corresponding author: Dr. Ranjan C. Khunt,
drrckhunt12@yahoo.com
acid, cycloserine, amikacin, kanamycin, and capreomycin,
the discovery of new antitubercular agents with improved
activity, lower toxicity, broader spectrum, and safer thera-
peutic profiles is a pressing need (4). The distribution of
tuberculosis is not uniform across the globe. About 80%
of the population in many Asian and African countries test
positive in the tuberculin test, while only 5–10% of the US
population test positive (5). Globally, 9.2 million new cases
and 1.7 million deaths from TB were reported in the year
2006, of which 0.7 million cases and 0.2 million deaths
were of HIV-positive people. Among the infected individu-
als approximately 8 million develop active TB and almost
2 million of these die from this deadly disease. Each year,
95% of new TB cases appear in developing countries (6).
Frequency of tuberculosis is progressively increasing
worldwide. New emerging strains of bacilli that are
emerging are resistant to the currently available drugs
which make this issue more alarming. In this regard, a
series of substituted quinolinyl chalcones, quinolinyl
pyrimidines, and pyridines were synthesized and eval-
uated for their antitubercular activity in vitro against
Mycobacterium tuberculosis H37RV. To establish the
role of the 2-chloroquinoline nucleus as a pharmaco-
phoric group and study its influence on the antimyco-
bacterial activity, a 3D-QSAR study based on CoMFA
and CoMSIA was undertaken on this set of 2-chloro-
quinoline derivatives. Statistically significant models
that are able to well correlate the antimycobacterial
activity with the chemical structures of the 2-chloro-
quinolines have been developed. The contour maps
resulting from the best CoMFA and CoMSIA models
were used to identify the structural features relevant to
the biological activity in this series of analogs. Further
analysis of these interaction-field contour maps also
showed a high level of internal consistency. The infor-
mation obtained from the field 3-D contour maps may
be fruitfully utilized in the design of more potent
2-chloroquinoline-based analogs as potential antituber-
cular candidates.
New antitubercular drug regimens are clearly needed to
reduce the time required for a durable cure and to treat
the ever expanding problem of drug- and multidrug-resis-
tant (MDR) M. tuberculosis strains (7).
Despite the efforts of agencies such as WHO; the Global
Fund for HIV, TB, and Malaria; and the Gates Foundation
which have galvanized multilateral support and initiated
public–private partnerships to increase resources to combat
this disease of poverty, only seven candidate TB drugs from
five different chemical classes are undergoing clinical trials
till very recently, these are the fluoroquinolones gatifloxacin
and moxifloxacin, a diarylquinoline (TMC207), the nitroimi-
dazoles-OPC67683 and PA824, a pyrrole (LL3858) and an
ethylenediamine (SQ109) (8). The need to develop less toxic
and more potent compounds against tuberculosis has led
to the discovery of diarylquinolines that target the mycobac-
terial F1F0 proton ATP synthase, a new drug target in myco-
Key words: 3D-QSAR, antitubercular activity, chloroquino-
lines
Received 26 September 2012, revised 3 May 2013 and
accepted for publication 28 May 2013
bacteria. Among diarylquinolines,
a modified form of
diarylquinoline R207910 has been identified as a promising
anti-TB drug with in vitro activity against M. smegmatis (MIC
of 0.003 lg/mL) and M. tuberculosis (MIC of 0.030 lg/mL)
and potent activity against MDR-TB strains (9). It has been
reported that C6-substituted pyrimidine derivatives display
antitubercular activity (10–14). Pyridines and pyrimidines
have also been identified with antitubercular activity and can
Tuberculosis is caused by several species of mycobacte-
ria, including Mycobacterium tuberculosis, M. bovis, M. af-
ricanum, M. microti, M. avium, and M. leprae that are
intracellular, gram-positive, non-motile, and rod-shaped
obligate aerobic pathogens of higher vertebrates (1,2).
Tuberculosis is an infectious pulmonary disease infecting
ª 2013 John Wiley & Sons A/S. doi: 10.1111/cbdd.12178
669

Khunt et al.
be developed as new structural classes of antitubercular
agents. Among the pyridines, N-alkyl-1,2-dihydro-2-thioxo-
3-pyridine carbothioamides have been recognized as a
novel class of compounds with good antitubercular activity.
Two compounds in this series have an MIC of 0.50 lg/mL
(15). Within the pyrimidine group, 2,4-diaminopyrimidines
have been identified with IC50 of 0.0058 lM and a safety
index >600 (16). The most effective analogs show more than
97% inhibition of mycobacterium at a concentration of
2.50 lg/mL and this series was examined in detail by QSAR
analysis (17). Thus, this structural class holds great hope for
development of new and effective antitubercular agents.
General method for preparation of 2-chloro-3-(1-
arylprop-2-en-1-one-3-yl)-6-fluoro/bromoquinoline
(2a–m)
To a solution of 2-chloro-6-fluoro/bromo-quinoline-3-car-
boxaldehyde (0.01 mol) and substituted acetophenone
(0.01 mol) in ethanol (25 mL), 40% NaOH solution was
added till the reaction mixture became basic. The reaction
mixture was stirred for 24 h. The content was poured onto
crushed ice and neutralized. The isolated product was
crystallized from ethanol.
2-chloro-3-(1-(3-Nitro)-phenyl-prop-2-en-1-one-3-
yl)-6-fluoroquinoline (2a)
Chalcones have been associated with a wide variety of
pharmacological activity and also act as a synthon for pre-
paring new heterocycles (18). Keeping in mind the bio-
medical applications and in continuation of our previous
study, (a)we envisioned the development of a novel series
of quinoline derivatives incorporating the pyrimidine in a
single molecular framework with a potential spectrum of
bioresponses. The antitubercular activity of these com-
pounds against M. tuberculosis H37Rv strain was studied
using the radiometric BACTEC and broth dilution assay
methods. The current work describes the synthesis of a
series of 2-chloroquinoline derivatives with encouraging
antimycobacterial activity against M. tuberculosis H37Rv.
Yield 65%, mp 120–121 °C; Rf 0.66; IR (KBr): 3065, 3010,
1668, 1630, 1610, 1585, 1491, 1150, 850, 735 per cm.
1H NMR (300 MHz, CDCl3) d (ppm): 6.70–7.80 (m, 11H,
vinyl + Ar–H), 7.83(d, 1H, J = 15.8 Hz, vinyl proton), 8.12
(t, 1H, J = 8.00 Hz, quinoline), 8.30 (s, 1H, NH), 8.85 (s,
1H, quinoline). 13C NMR (CDCl3) d (ppm): 109.8, 116.2,
120.0, 126.0, 127.4, 128.1, 130.2, 131.0, 141.0, 143.2,
144.7, 160.0, 191.0. MS m/z: 356 (M+), 358 (M+2). Anal.
Calcd. for C18H10ClFN2O3: C, 51.77%; H, 2.41%; N,
6.71%. Found: C, 51.67%; H, 2.35%; N, 6.65%.
2-chloro-3-(1-(3,4-dichloro)phenyl-prop-2-en-1-
one-3-yl)-6-fluoroquinoline (2b)
To deduce a correlation between structure and biological
activity of these 2-chloroquinoline derivatives as antituber-
cular agents, a 3D-QSAR study employing comparative
molecular field analysis (CoMFA) (19,20) and comparative
molecular similarity indices analysis (CoMSIA) (21–23) was
performed. The structural variations in the molecular fields
at particular regions in the space were investigated and the
resulting 3D-QSAR models could provide useful indicators
for further design of new drug candidates for tuberculosis.
Yield 61%, mp 70–72 °C; Rf 0.60; IR (KBr): 3067, 3015,
1670, 1628, 1615, 1580, 1492, 1150, 740, 670 per cm. 1H
NMR (300 MHz, CDCl3) d (ppm): 7.32–7.80 (m, 6H,
vinyl + Ar–H), 7.84 (d, 1H, J = 15.8 Hz, vinyl proton), 8.03
(t, 1H, J = 8.00 Hz, quinoline), 8.80 (s, 1H, quinoline). 13C
NMR (CDCl3) d (ppm): 109.7, 116.2, 120.8, 127.1, 129.9,
130.2, 131.6, 135.0, 136.5, 145.2, 148.7. MS m/z: 378.0
(M+), 380.0 (M+2). Anal. Calcd. for C18H9NOCl3F: C, 56.80%;
H, 2.38%; N, 3.68%. Found: C, 56.72%; H, 2.26%; N, 3.56%.
Methods and Materials
2-chloro-3-(1-(4-ethoxy)-phenyl-prop-2-en-1-one-
3-yl)-6-fluoroquinoline (2c)
Melting points were determined in open capillary tubes
and are uncorrected. Thin layer chromatography (TLC)
was performed on Silica gel G (Merck TLC Plate, Darm-
stadt, Germany) with the solvent system ethyl acetate:
hexane (3:7, v/v); iodine chamber was used for visualiza-
tion of the TLC spots. 1H NMR was determined in CDCl3
solution on a Bruker DPX 300 MHz spectrometer. 13C
NMR (75 and 125 MHz) spectra were registered on a Bru-
ker DPX 300 and ARX 500 at 25 °C in CDCl3. IR spectra
were recorded on a Shimadzu 8400 spectrometer in KBr
disks. Elemental analysis was carried out on a Carlo Erba
1108 analyzer. The starting material quinoline was pre-
pared using Vilsmeier-Haack synthesis.
Yield 55%, mp 66–68 °C; Rf 0.58; IR (KBr): 3015, 2975,
1670, 1628, 1610, 1590, 1450, 1250, 1075, 740, 650 per
cm. 1H NMR (300 MHz, CDCl3) d (ppm): 1.5 (t, 3H,
J = 6.98 Hz, -CH3), 4.0 (q, 2H, J = 6.96 Hz, -CH2), 6.96–
7.70 (m, 7H, vinyl + Ar–H), 7.85 (d, 1H, J = 15.8 Hz, vinyl
proton), 8.12 (t, 1H, J = 8.00 Hz, quinoline), 8.80 (s, 1H,
quinoline). 13C NMR (CDCl3) d (ppm): 14.8, 63.5, 114.8,
129.5, 130.5, 135.6, 146.6, 149.8, 161.8, 164.5, 190.1.
MS m/z: 353.0.0 (M+), 355.0 (M+2). Anal. Calcd. for
C
20H15NO2ClF: C, 67.52%.; H, 4.25%.; N, 3.94%. Found:
C, 67.44%; H, 4.18%; N, 3.86%.
Synthesis of 3-(aryl)-1-phenyl-1H-pyrazole-4-
carbaldehydes (1)
2-chloro-3-(1-(2-thienyl)-prop-2-en-1-one-3-yl)-6-
fluoroquinoline (2d)
Synthesis of 3-(aryl)-1-phenyl-1H-pyrazole-4-carbaldehy-
des was achieved by reported method (18).
Yield 58%, mp 144–146 °C; Rf 0.54; IR (KBr): 3020, 1665,
1630, 1600, 1585, 1475, 742, 635 per cm. 1H NMR
670
Chem Biol Drug Des 2013; 82: 669–684

Synthesis of Chloroquinoline Derivative
(300 MHz, CDCl3) d (ppm): 7.06–7.65 (m, 8H, vinyl + Ar–
H), 7.62 (d, 1H, J = 15.7 Hz, vinyl proton), 8.15 (t, 1H,
J = 8.20 Hz, quinoline), 8.79 (s, 1H, quinoline). 13C NMR
(CDCl3) d (ppm): 109.8, 120.4, 128.5, 129.4, 130.7,
131.0, 135.2, 137.2, 139.4, 141.2, 149.8, 161.3, 180.1.
MS m/z: 315.0 (M+), 317.0 (M+2). Anal. Calcd. for
C17H11NOClFS: C, 60.48%; H, 2.85%; N, 4.41%. Found:
C, 60.36%; H, 2.73%; N, 4.32%.
NMR (CDCl3) d (ppm): 118.9, 121.2, 124.7, 128.5, 129.5,
131.8, 135.6, 137.4, 147.7, 148.9, 151.2, 169.8, 189.7.
MS m/z: 388.0 (M+), 390.0 (M+2). Anal. Calcd. for
C
18H10NOClBrF: C, 55.34%; H, 2.58%; N, 3.59%. Found:
C, 55.28%; H, 2.45%; N, 3.48%.
2-chloro-3-(1-(4-amino)phenyl-prop-2-en-1-one-3-
yl)-6-bromoquinoline (2i)
Yield 55%, mp 118–120 °C; Rf 0.56; IR (KBr): 3400–3200,
3080, 3000, 1664, 1601, 1585, 1490, 1250, 780, 735 per
cm. 1H NMR (300 MHz, CDCl3) d (ppm): 4.01 (broad, 2H,
-NH2), 6.80–7.85 (m, 7H, vinyl + Ar–H), 7.85(d, 1H,
J = 15.65 Hz, vinyl proton), 8.05 (d, 1H, J = 7.95 Hz,
quinoline), 8.80 (s, 1H, quinoline). 13C NMR (CDCl3) d
(ppm): 118.7, 121.3, 128.7, 129.3, 131.5, 135.6, 136.7,
148.2, 149.4, 158.2, 190.3. MS m/z: 385.0 (M+), 387.0
(M+2). Anal. Calcd. for C18H12N2OCl2Br: C, 55.77%; H,
3.12%; N, 7.23%. Found: C, 55.64%; H, 3.01%; N,
7.16%.
2-chloro-3-(1-(4-fluoro)-phenyl-prop-2-en-1-one-3-
yl)-6-fluoroquinoline (2e)
Yield 61%, mp 130–132 °C; Rf 0.48; IR (KBr): 3031, 3080,
1670, 1628, 1616, 1580, 1450, 735, 651 per cm. 1H
NMR (300 MHz, CDCl3) d (ppm): 7.16–7.79 (m, 7H,
vinyl + Ar–H), 7.90 (d, 1H, J = 15.8 Hz,vinyl proton), 8.18
(t, 1H, J = 8.00 Hz, quinoline), 8.85 (s, 1H, quinoline). 13C
NMR (CDCl3) d (ppm): 110.8, 116.8, 121.4, 127.6, 129.2,
130.7, 132.5, 135.4, 136.7, 142.4, 143.3, 149.7, 161.4,
165.9, 190.1. MS m/z: 327.0 (M+), 329.0 (M+2). Anal.
Calcd. for C18H10NOClF2: C, 65.57%; H, 3.06%; N,
4.25%. Found: C, 65.43%; H, 2.96%; N, 4.15%.
2-chloro-3-(1-(2-amino)phenyl-prop-2-en-1-one-3-
yl)-6-bromoquinoline (2j)
2-chloro-3-(1-(4-bromo)-phenyl-prop-2-en-1-one-3-
yl)-6-fluoroquinoline (2f)
Yield 59%, mp 88–90 °C; Rf 0.58; IR (KBr): 3350–3250,
3070, 3010, 1660, 1628, 1604, 1590, 1450, 1220, 779,
740 per cm. 1H NMR (300 MHz, CDCl3) d (ppm): 3.98
(broad, 2H, -NH2), 7.10–7.75 (m, 8H, vinyl + Ar–H), 8.30
(d, 1H, J = 7.80 Hz, quinoline), 8.82 (s, 1H, quinoline). 13C
NMR (CDCl3) d (ppm): 116.5, 118.5, 121.8, 125.7, 127.2,
129.6, 134.4, 138.6, 141.2, 148.9, 149.4, 150.2, 190.5
MS m/z: 385.0 (M+), 387.0 (M+2). Anal. Calcd. for
C18H12N2OClBr: C, 55.77%; H, 3.12%; N, 7.23%. Found:
C, 55.63%; H, 3.05%; N, 7.13%.
Yield 62%, mp 134–136 °C; Rf 0.58; IR (KBr): 3110, 3080,
1668, 1628, 1600, 1585, 1450, 781, 685 per cm. 1H
NMR (300 MHz, CDCl3) d (ppm): 7.32–7.70 (m, 7H,
vinyl + Ar–H), 7.92 (d, 1H, J = 15.5 Hz, vinyl proton), 8.20
(t, 1H, J = 8.12 Hz, quinoline), 8.80 (s, 1H, quinoline). 13C
NMR (CDCl3) d (ppm): 111.0, 119.8, 128.7, 129.9, 131.5,
133.5, 134.5, 136.8, 138.4, 143.7, 145.6, 150.7, 161.2,
190.7. MS m/z: 388.0 (M+), 390.0 (M+2). Anal. Calcd. for
C
18H10NOClFBr: C, 55.34%; H, 2.58%; N, 3.59%. Found:
C, 55.22%; H, 2.47%; N, 3.45%.
2-chloro-3-(1-(2-indene-1-one-3-yl)-6-
bromoquinoline (2k)
2-chloro-3-(1-(3,4-dichloro)phenyl-prop-2-en-1-
one-3-yl)-6-bromoquinoline (2g)
Yield 44%, mp 143–145 °C; Rf 0.62; IR (KBr): 3140,
3010, 2975, 1662, 1630, 1598, 1491, 781, 740 per cm.
1H NMR (300 MHz, CDCl3) d (ppm): 4.10 (s, 2H, CH2),
7.01–7.75 (m, 7H, vinyl + Ar–H), 8.30 (d, 1H,
J = 7.80 Hz, quinoline), 8.82 (s, 1H, quinoline). 13C NMR
(CDCl3) d (ppm): 121.8, 125.7, 127.2, 129.6, 134.4,
138.6, 141.2, 148.9, 149.4, 150.2, 190.5. MS m/z:
382.0 (M+), 384.0 (M+2). Anal. Calcd. for C19H11NOClBr:
C, 61.11%; H, 3.66%; N, 3.39%. Found: C, 60.98%; H,
3.58%; N, 3.27%.
Yield 59%, mp 240–242 °C; Rf 0.55; IR (KBr): 3150, 3030,
1672, 1630, 1598, 1550, 1450, 781, 741 per cm. 1H
NMR (300 MHz, CDCl3) d (ppm): 7.40–7.91 (m, 7H,
vinyl + Ar–H), 8.01 (t, 1H, J = 8.12 Hz, quinoline), 8.87 (s,
1H, quinoline). 13C NMR (CDCl3) d (ppm): 122.7, 124.8,
129.7, 132.4, 136.8, 138.9, 139.8, 147.4, 148.2, 151.0,
191.0. MS m/z: 439.0 (M+), 441.0 (M+2). Anal. Calcd. for
C18H9NOCl3Br: C, 48.96%; H, 2.05%; N, 3.17%. Found:
C, 48.84%; H, 1.98%; N, 3.02%.
2-chloro-3-(1-(4-ethoxyphenyl)prop-2-en-1-one-3-
yl)-6-bromoquinoline (2l)
2-chloro-3-(1-(4-fluoro)phenyl-prop-2-en-1-one-3-
yl)-6-bromoquinoline (2h)
Yield 58%, mp 170–172 °C; Rf 0.49; IR (KBr): 3115, 3030,
2980, 1673, 1600, 1585, 1493, 1250, 1075, 780, 731 per
cm. 1H NMR (300 MHz, CDCl3) d (ppm):1.20 (t, 3H,
J = 6.90 Hz, CH3), 4.02 (q, 2H, J = 6.98 Hz, CH2), 6.96–
7.75 (m, 7H, vinyl + Ar–H), 7.85 (d, 1H, J = 15.8 Hz,
vinyl), 8.15 (d, 1H, J = 7.80 Hz, quinoline), 8.80 (s, 1H,
Yield 54%, mp 82–84 °C; Rf 0.61; IR (KBr): 3100, 3025,
1667, 1630, 1615, 1585, 1480, 771, 735, 681 per cm. 1H
NMR (300 MHz, CDCl3) d (ppm): 7.16–7.85 (m, 7H,
vinyl + Ar–H), 7.94(d, 1H, J = 15.8 Hz, vinyl proton), 8.35
(d, 1H, J = 8.52 Hz, quinoline), 8.83 (s, 1H, quinoline). 13C
Chem Biol Drug Des 2013; 82: 669–684
671

Khunt et al.
quinoline). 13C NMR (CDCl3) d (ppm): 15.8, 63.2, 116.7,
121.4, 123.2, 128.7, 130.1, 131.7, 132.3, 135.8, 147.6,
148.9, 150.4, 162.3, 191.8. MS m/z: 414.0 (M+), 416.0
(M+2). Anal. Calcd. for C20H15NO2ClBr: C, 57.65%; H,
3.63%; N, 3.36%. Found: C, 57.58%; H, 3.59%; N,
3.31%.
Calcd. for C21H10N4Cl3F: C, 56.85%; H, 2.27%; N,
12.63%. Found: C, 56.78%; H, 2.19%; N, 12.54%.
2-chloro-3-(2-amino-3-cyano-4-(4-flouorophenyl)-
pyridin-6-yl)-6-fluoroquinoline (3c)
Yield 61%, mp 269–270 °C; 0.60; Rf 0.60; IR (KBr): 3410–
3250, 3080, 2120, 1628, 1550, 1498, 1100, 740, 645 per
cm. 1HNMR (300 MHz, CDCl3) d (ppm): 4.01 (broad, 2H,
-NH2), 7.03–7.49 (m, 7H, Ar–H + quinoline + pyridine), 8.09
(t, 1H, J = 8.00 Hz, quinoline), 8.95 (s, 1H, quinoline). 13C
NMR (CDCl3) d (ppm): 88.7, 110.5, 115.8, 117.1, 121.0
125.8, 127.2, 130.0, 135.5, 137.8, 145.5, 151.0, 155.8,
157.4, 161.8, 163. MS m/z: 390.0 (M+), 392.0 (M+2). Anal.
Calcd. for C21H11N4ClF2: C, 64.21%; H, 2.82%; N, 14.26%.
Found: C, 64.13%; H, 2.75%; N, 14.18%.
2-chloro-3-(1-(3-nitrophenyl)prop-2-en-1-one-3-yl)-
6-bromoquinoline (2m)
Yield 52%, mp 154–156 °C; Rf 0.51; IR (KBr): 3110, 3010,
1660, 1628 1601, 1585, 1450, 1250, 1075, 850, 780,
735 per cm. 1H NMR (300 MHz, CDCl3) d (ppm): 7.75–
8.47 (m, 9H, vinyl + Ar–H, quinoline), 8.85 (s, 1H, quino-
line). 13C NMR (CDCl3) d (ppm): 121.8, 124.8, 129.0,
134.0, 136.0, 139.8, 149.8, 151.7, 153.6, 189.2. MS m/z:
415.0 (M+), 417.0 (M+2). Anal. Calcd. for C18H10N2O3ClBr:
C, 51.77%; H, 2.41%; N, 6.71%. Found: C, 51.65%; H,
2.35%; N, 6.62%.
2-chloro-3-(2-amino-3-cyano-4-(2-theinyl)-pyridin-
6-yl)-6-fluoroquinoline (3d)
Yield 55%, mp 110–55 °C; Rf 0.58; IR (KBr): 3415–3215,
3120, 2150, 1628, 1575, 1498, 1100, 740, 680 per cm.
1H NMR (300 MHz, CDCl3) d (ppm): 4.02 (broad, 2H,-
NH2), 7.01–7.47 (m, 6H, Ar–H + quinoline + pyridine), 8.15
(t, 1H, J = 8.00 Hz, quinoline), 8.90 (s, 1H, quinoline). 13C
NMR (CDCl3) d (ppm): 90.2, 110.2, 121.8, 127.6, 130.2,
133.4, 135.8, 139.7, 142.0, 147.9, 151.8, 157.9, 161.0.
MS m/z: 378.00 (M+), 380.0 (M+2). Anal. Calcd. for
C19H10N4ClFS: C, 59.92%; H, 2.65%; N, 14.71%. Found:
C, 59.81%; H, 2.58%; N, 14.63%.
General method for preparation of 2-chloro-3-(2-
amino-3-cyano-4-arylpyridin-6-yl)-6-fluoro/
bromoquinoline (3a–g)
An equimolar mixture of 2-chloro-3-(1-arylprop-2-en-1-
one-3-yl)-6-fluoro/bromoquinoline (compound II) and mal-
ononitrile with ammonium acetate (0.08 M) in 30 mL etha-
nol was heated under reflux for 6 h. The reaction mixture
was poured onto crushed ice, and the product was crys-
tallized from ethanol.
2-chloro-3-(2-amino-3-cyano-4-(4-ethoxyphenyl)
pyridin-6-yl)-6-fluoroquinoline (3a)
2-chloro-3-(2-amino-3-cyano-4-(4-bromophenyl)-
pyridin-6-yl)-6-fluoroquinoline (3e)
Yield 58%, mp 144–146 °C; Rf 0.55; IR (KBr): 3400–
3200, 3020, 2960, 2148, 1630, 1600, 1550, 1250, 1100,
1075, 740, 648 per cm. 1H NMR (300 MHz, CDCl3)
d (ppm): 1.33 (t, 3H, J = 6.98 Hz, -CH3), 3.99 (q, 2H,
J = 6.96 Hz, -CH2), 4.05 (broad, 2H, -NH2), 6.83–8.03
(m, 8H, Ar–H + quinoline + pyridine), 9.01 (s, 1H, quino-
line). 13C NMR (CDCl3) d (ppm): 14.9, 64.7, 87.0, 109.2,
115.0, 116.8, 121.2, 127.1, 128.7, 138.7, 139.5, 141.2,
152.0, 158.3, 159.4, 160.9. MS m/z: 402.0 (M+),
404.0 (M+2). Anal. Calcd. for C23H16N4OClF: C, 65.25%;
H, 3.49%; N, 13.14%. Found: C, 65.10%; H, 3.36%; N,
13.02%.
Yield 53%, mp 159–161 °C; Rf 0.53; IR (KBr): 3400–3200,
3031, 2130, 1630, 1575, 1498, 1100, 780, 735, 645 per
cm.1H NMR (300 MHz, CDCl3) d (ppm): 3.98(broad, 2H,
-NH2), 7.32–7.49 (m, 7H, Ar–H + quinoline + pyridine), 8.15
(t, 1H, J = 8.10 Hz, quinoline), 8.89 (s, 1H, quinoline). 13C
NMR (CDCl3) d (ppm): 89.7, 109.5, 110.2, 117.7, 121.8,
125.4, 128.2, 130.1, 132.3, 135.5, 136.8, 145.5, 151.1,
155.8, 159.7, 161.0. MS m/z: 449.0 (M+), 451.0 (M+2).
Anal. Calcd. for C21H11N4ClFBr: C, 55.59%; H, 2.44%; N,
12.35%. Found: C, 55.42%; H, 2.36%; N, 12.26%.
2-chloro-3-(2-amino-3-cyano-4-(3-nitrophenyl)-
pyridin-6-yl)-6-bromoquinoline (3f)
2-chloro-3-(2-amino-3-cyano-4-(2,4-
Yield 57%, mp 154–156 °C; Rf 0.49; IR (KBr): 3400–3200,
3080, 2150, 1628, 1575, 1580, 1497, 1132, 780, 735 per
cm. 1H NMR (300 MHz, CDCl3) d (ppm): 4.02 (broad, 2H,
-NH2), 7.47-7.91 (m, 7H, Ar–H), 8.15 (t, 1H, J = 8.00 Hz,
quinoline), 8.85 (s, 1H, quinoline). 13C NMR (CDCl3)
d (ppm): 89.4, 110.0, 115.8, 121.2, 125.7, 129.0, 131.1,
132.3, 135.5, 136.8, 147.5, 151.8, 155.7, 158.2, 162.0.
MS m/z: 478.0 (M+), 480.0 (M+2). Anal. Calcd. for
C21H11N5ClFBr: C, 52.47%; H, 2.31%; N, 14.57%. Found:
C, 52.38%; H, 2.25%; N, 14.46%.
dichlorophenyl)-pyridin-6-yl)-6-fluoroquinoline (3b)
Yield 56%, mp 129–131 °C; Rf 0.52; IR (KBr): 3410–3250,
3015, 2150, 1628, 1592, 1100, 735, 680 per cm. 1H
NMR (300 MHz, CDCl3) d (ppm): 3.92 (broad, 2H, -NH2),
7.21–7.47 (m, 6H, Ar–H + quinoline + pyridine), 8.15 (t,
1H, J = 8.00 Hz, quinoline), 8.85 (s, 1H, quinoline). 13C
NMR (CDCl3) d (ppm): 89.2, 19.5, 110.1, 115.7, 121.0,
121.7, 127.5, 130.5, 135.8, 137.0, 145.7, 151.8, 157.6,
158.9, 160.9.0 MS m/z: 439.00 (M+), 441.0 (M+2). Anal.
672
Chem Biol Drug Des 2013; 82: 669–684

Synthesis of Chloroquinoline Derivative
2-chloro-3-(2-amino-3-cyano-4-(3, 4-
1H NMR (300 MHz, CDCl3) d (ppm): 4.05 (broad, 2H,
-NH2), 7.10–7.46 (m, 7H, Ar–H + quinoline + pyrimidine),
8.15 (t, 1H, J = 8.00 Hz, quinoline), 8.83 (s, 1H, quinoline).
13C NMR (CDCl3) d (ppm): 94.1, 109.5, 116.4, 121.0,
128.5, 129.9, 130.7, 132.5, 135.8, 144.3, 150.0, 158.2,
161.8, 163.2, 164.7. MS m/z: 366.0 (M+), 368.0 (M+2).
Anal. Calcd. for C19H11N4ClF2: C, 61.88%; H, 3.01%; N,
15.19%. Found: C, 61.75%; H, 2.84%; N, 15.05%.
dichlorophenyl)-pyridin-6-yl)-6- bromoquinoline (3g)
Yield 50%, mp 159–161 °C; Rf 0.55; IR (KBr): 3450–3200,
3030, 2150, 1628, 1595, 1580, 1498, 1150, 780, 741 per
cm. 1H NMR (300 MHz, CDCl3) d (ppm): 4.01 (broad, 2H,
-NH2), 7.27-7.81 (m, 6H, Ar–H + quinoline + pyridine),
8.12 (d, 1H, J = 8.07 Hz, quinoline), 8.85 (s, 1H, quino-
line). 13C NMR (CDCl3) d (ppm): 89.0, 110.3, 115.0,
122.0, 126.1, 128.0, 133.8, 134.5, 138.8, 149.5, 151.8,
154.7, 156.2, 162.0. MS m/z: 499.0 (M+), 501.0 (M+2).
Anal. Calcd. for C21H10N4Cl3Br: C, 49.99%; H, 2.00%; N,
11.10%. Found: C, 49.85%; H, 1.87%; N, 11.01%.
2-chloro-3-(2-amino-4-(2-thienyl)- pyrimidin-6-yl)-
6-fluoroquinoline (4d)
Yield 55%, mp 284–286 °C; Rf 0.63; IR (KBr): 3415–3200,
1
1630, 1610, 1598, 1498, 1120, 735, 680, 670 per cm. H
General preparation of 2-chloro-3-(2-amino-4-
phenyl-pyrimidin–6-yl)-6-fluoroquinoline (4a–h)
An equimolar mixture of 2-chloro-3-(1-aryl-prop-2-en-1-
on-3yl)-6-fluoro/bromoquinoline (2), guanidine hydrochlo-
ride was irradiated with microwave radiation at 480 power
for 10 min in the presence of catalytically amount of
potassium carbonate. Then, the product was isolated with
ethanol and crystallized it.
NMR (300 MHz, CDCl3) d (ppm): 4.10 (broad, 2H, NH2),
7.10–7.41 (m, 6H, Ar–H + quinoline + pyrimidine), 8.12 (t,
1H, J = 8.00 Hz, quinoline), 8.92 (s, 1H, quinoline). 13C
NMR (CDCl3) d (ppm): 92.8, 109.0, 121.0, 127.6, 130.3,
131.8, 135.7, 140.2, 143.7, 149.7, 157.9, 160.2, 162.8.
MS m/z (%):354.0 (M+), 356.0 (M+2). Anal. Calcd. for
C17H10N4ClFS: C, 57.23%; H, 2.82%; N, 15.70%. Found:
C, 57.08%; H, 2.75%; N, 15.66%.
2-chloro-3-(2-amino-4-(4-ethoxyphenyl)pyrimidin-
6-yl)-6-fluoroquinoline (4a)
2-chloro-3-(2-amino-4-(9H-indeno[2,1-a]pyrimidin-
6-yl)-6-fluoroquinoline (4e)
Yield 60%, mp 279–281 °C; Rf 0.62; IR (KBr): 3430–3200,
3020, 2980, 1628, 1600, 1575, 1580, 1498, 1250, 1100,
1075, 741, 680 per cm. 1H NMR (300 MHz, CDCl3)
d (ppm): 1.35 (t, 3H, -CH3), 3.98 (q, 2H, -CH2), 4.01
(broad, 2H, -NH2), 6.98–7.48 (m, 7H, Ar–H + quino-
line + pyrimidine), 8.12 (t, 1H, J = 8.10 Hz, quinoline),
8.85 (s, 1H, Quinoline). 13C NMR (CDCl3) d (ppm): 14.6,
64.7, 93.7, 110.8, 114.7, 121.2, 123.5, 126.8, 127.9,
130.2, 132.5, 135.8, 142.7, 150.0, 158.7, 161.8, 163.9,
164.7. MS m/z: 392.0 (M+), 394.0 (M+2). Anal. Calcd. for
Yield 54%, mp 183–185 °C; Rf 0.56; IR (KBr): 3410-3210,
1629, 1608, 1597, 1495, 1110, 780, 741, 635 per cm. H
1
NMR (300 MHz, CDCl3) d (ppm): 3.81 (s, 2H, indonyl),
4.05 (broad, 2H, -NH2), 7.12–7.49 (m, 6H, Ar–H + quino-
line + pyrimidine), 8.18 (d, 1H, J = 8.03 Hz, quinoline),
8.93 (s, 1H, quinoline). 13C NMR (CDCl3) d (ppm): 94.1,
110.0, 121.0, 123.7, 128.2, 129.7, 130.3, 135.8, 160.9,
162.8. MS m/z: 427.0 (M+), 429.0 (M+2). Anal. Calcd. for
C
20H12N4ClFBr: C, 53.11%; H, 2.58%; N, 13.04%. Found:
C, 53.01%; H, 2.46%; N, 12.87%.
C
21H16N4ClFO: C, 63.88%; H, 4.08%; N, 14.19%. Found:
C, 63.76%; H, 3.97%; N, 14.03%.
2-chloro-3-(2-amino-4-(4-bromophenyl)pyrimidin-
6-yl)-6-bromoquinoline (4f)
2-chloro-3-(2-amino-4-(3,4-dichlorophenyl)
pyrimidin-6-yl)-6-fluoroquinoline (4b)
Yield 42%, mp 164–166 °C; Rf 0.58; IR (KBr): 3450–3210,
1630, 1611, 1598, 1497, 1140, 780, 741, 635 per cm. H
1
Yield 59%, mp 180–181 °C; Rf 0.52; IR (KBr): 3410–3215,
3030, 1630, 1600, 1556, 1498, 1115, 741, 680 per cm.
1H NMR (300 MHz, CDCl3) d (ppm): 4.05(broad, 2H,
-NH2), 6.96–7.43 (m, 6H, Ar–H + quinoline + pyrimidine),
8.12 (t, 1H, J = 8.20 Hz, quinoline), 8.85 (s, 1H, quinoline).
13C NMR (CDCl3) d (ppm): 93.4, 109.9, 121.0, 127.5,
130.3, 132.5, 135.8, 142.3, 149.3, 160.9, 162.8, 164.0.
MS m/z: 415.0 (M+), 417.0 (M+2). Anal. Calcd. for
NMR (300 MHz, CDCl3) d (ppm): 4.10 (broad, 2H, -NH2),
7.15–7.91 (m, 7H, Ar–H + quinoline + pyrimidine), 8.18 (d,
1H, J = 8.15 Hz, quinoline), 8.80 (s, 1H, quinoline). 13C
NMR (CDCl3) d (ppm): 29.9, 115.8, 121.8, 125.7, 127.9,
129.9, 131.2, 135.8, 138.9, 145.8, 151.2, 160.0, 161.2,
170.0. MS m/z: 421.0 (M+), 423.0 (M+2). Anal. Calcd. for
C
19H11N4ClBr2: C, 58.49%; H, 3.57%; N, 12.40%. Found:
C, 58.36%; H, 3.41%; N, 12.28%.
C
19H10N4Cl3F: C, 54.38%; H, 2.40%; N, 13.35%. Found:
C, C, 54.24%; H, 2.27%; N, 13.27%.
2-chloro-3-(2-amino-4-(3, 4 dichlorophenyl)
pyrimidin-6-yl)-6-bromoquinoline (4g)
2-chloro-3-(2-amino-4-(4-fluorophenyl)pyrimidin-6-
yl)-6-fluoroquinoline (4c)
Yield 62%, mp 114–116 °C; Rf 0.62; IR (KBr): 3415–3250,
3028, 1628, 1601, 1558, 1497, 1110, 735, 680 per cm.
Yield 53%, mp 164–166 °C; Rf 0.54; IR (KBr): 3450–3200,
1628, 1610, 1598, 1497, 1150, 780, 741 per cm. 1H
NMR (300 MHz, CDCl3) d (ppm): 4.10 (broad, 2H, -NH2),
7.01–7.91 (m, 6H, Ar–H + quinoline + pyrimidine), 8.18 (d,
Chem Biol Drug Des 2013; 82: 669–684
673

Khunt et al.
1H, J = 8.10 Hz, quinoline), 8.85 (s, 1H, quinoline). 13C
NMR (CDCl3) d (ppm): 93.4, 121.8, 127.0, 128.8, 129.1,
130.8, 133.9, 134.6, 135.3, 145.8, 151.2, 155.4, 162.3,
164.0. MS m/z (%):479.0 (M+), 480.0 (M+2). Anal. Calcd.
for C19H10N4Cl3Br: C, 47.49%; H, 2.10%; N, 11.66%.
Found: C, 47.38%; H, 2.02%; N, 11.58%.
3D-QSAR model. In this investigation, molecule 3e
(Table 1) was selected as the template molecule because
it is the most active compound in the data set. Because
the bioactive conformation of these 2-chloroquinoline
analogs is not known, the lowest energy conformation of
3e was searched using a simulated annealing approach
and used as a template to generate the conformation of
the remaining molecules. The ligand geometries were
optimized by energy minimization using the Powell gradient
2-chloro-3-(2-amino-4-(3-nitrophenyl)-3, 4-
€
dihydropyrimidin-1H-6-yl)-6-bromoquinoline (4h)
Yield 50%, mp 189–191 °C; Rf 0.52; IR (KBr): 3410–3200,
1635, 1600, 1590, 1497, 1140, 780, 741 per cm. 1H
NMR (300 MHz, CDCl3) d (ppm): 4.12 (broad, 2H, -NH2),
7.10 (s, 1H, pyrimidine) 7.58–7.80 (m, 6H, Ar–H + quino-
line), 8.15 (d, 1H, J = 8.57 Hz, quinoline), 8.85 (s, 1H,
quinoline). 13C NMR (CDCl3) d (ppm): 93.7, 121.3, 122.0,
129.9, 131.8, 134.6, 136.8, 144.7, 149.7, 150.1, 158.7,
164.3, 166.0. MS m/z (%):454.0 (M+), 456.0 (M+2). Anal.
Calcd. for C19H11N5O2ClBr: C, 49.97%; H, 2.43%; N,
15.34%. Found: C, 49.82%; H, 2.32%; N, 15.23%.
method, with the Gasteiger–Huckel charges for the atoms
and the standard Tripos force field, to a convergence
ꢀ
gradient of 0.001 kcal/mol/A.
Molecular alignment
The spatial alignment of molecules under study is one of
the most crucial and determining factors in obtaining a reli-
able model. Molecular alignment was performed using the
database alignment in Sybyl with the atom-fit superimposi-
tion technique. The lowest energy conformation of the
molecules was used to align on the template molecule 3e.
The atoms selected for molecular alignment were based
on the common substructure of the molecules, due to its
effectiveness and easy implementation.
Biological activity
All compounds discussed in the previous section were
screened for their antitubercular activity. The activity was
carried out against Mycobacterium tuberculosis H37Rv,
(ATCC-27294) strain using the broth dilution assay method
(24,25). The antitubercular activity is presented in Table 1.
The antitubercular activity was measured (by TAACF, USA
Tuberculosis Antimicrobial Acquisition and Coordinating
Facility) for all the compounds at a fixed concentration of
6.25 lg/mL. Following this, only those compounds show-
ing more than 90% inhibition at this concentration were
tested further for measuring the MIC values; these are
quoted in the Table 1. For the remaining molecules
because no MIC values were measured, we have used an
alternate strategy that has been well published and
accepted (12,13,26) to transform the activity data that
could be used in 3D-QSAR as the dependent variable.
The activity data were transformed as follows:
Training and test sets
The data set was segregated into a training set for gener-
ating 3D-QSAR models and a test set (27) for validating
the models. This was carried out on the basis of chemical
and biological diversity using similarity search techniques
viz. D-optimal design, Tanimoto similarity coefficient, and
the Euclidian distance matrix criteria defined in Cerius2.b
The selection of the training and test sets was carried out
such that the test-set compounds had structural diversity
and a range of biological activities similar to that of the
training set. This approach grouped 10 compounds into
the test set leaving other 18 compounds to form the train-
ing set. The test set was used to evaluate the predictive
power of the CoMFA and CoMSIA models. The structures
and biological activity of the compounds in the training
and test sets are listed in Table 1.
Biological activity ¼ ꢀ log c þ logit
where c is the molar concentration = concentration (lg/
mL) 9 0.001/(molecular weight) and logit = log[% inhibi-
tion/(100ꢀ% inhibition)].
CoMFA and CoMSIA setup
After alignment of the molecules in the data set, CoMFA
and CoMSIA studies were carried with the QSAR option in
Sybyl. CoMFA steric and electrostatic interaction fields
were calculated at each intersection point of a 3D cubic
Computational details
The 3D-QSAR studies (CoMFA and CoMSIA) were per-
formed with the QSAR module integrated in Sybyl 7.1a
molecular modeling software package from Sybyl.Sybyl,
Version 7.1, Tripos Associate Inc., St. Louis, MO, USA,
running on a Pentium IV computer under the Centos WS
4.8 as OS.
ꢀ
lattice with a regularly spaced grid of 2 A, extending to
ꢀ
4 A beyond the aligned molecules in all directions. The
van der Waals potentials and Coulombic terms represent-
ing the steric and electrostatic fields, respectively, were
calculated using the Tripos force field with a distance
dependent dielectric. A sp3 hybridized carbon atom with a
charge of +1 served as the probe atom to calculate steric
and electrostatic fields. The CoMFA steric and electrostatic
fields thus generated were scaled by the CoMFA-STD
Ligand preparation
The choice of the template molecule and its conformation
are of utmost importance in the development of
a
674
Chem Biol Drug Des 2013; 82: 669–684

Synthesis of Chloroquinoline Derivative
Table 1: The experimental and predicted activity values for the molecules
Mycobacterium
tuberculosis H37
Predicted
Predicted
Molecule ID Structure
% inhibition (6.25 lg)
Observed activity
2.88
activity (CoMFA) activity (CoMSIA) (SHD)
O
2a
9%
2.82
2.82
F
NO2
N
Cl
O
Cl
F
F
2b
2c
2d
2e
2f
13%
2.96
5.75
3.13
2.60
5.49
2.57
5.79
2.97
2.60
3.66
2.91
4.21
3.18
2.60
3.84
2.55
4.35
3.06
2.63
3.53
2.80
4.48
3.22
2.70
4.85
2.62
4.04
2.96
2.55
3.92
Cl
Cl
N
O
O
O
O
13%
N
Cl
OC2H5
F
S
21%
N
N
N
N
Cl
Cl
Cl
Cl
F
7%
F
F
98% (5.0)
Br
Cl
O
Br
2g
2h
2i
5%
Cl
O
Br
Br
99% (6.25)
13%
N
N
Cl
Cl
F
O
O
NH2
NH2
Br
Br
2j
6%
N
N
Cl
Cl
O
2k
41%
O
Br
2l
19%
3.19
3.19
3.27
N
Cl
OC2H5
Chem Biol Drug Des 2013; 82: 669–684
675

Khunt et al.
Table 1: continued
Mycobacterium
tuberculosis H37
% inhibition (6.25 lg)
Predicted
Predicted
Molecule ID Structure
Observed activity
2.92
activity (CoMFA) activity (CoMSIA) (SHD)
O
2m
11%
2.88
3.37
2.89
Br
NO2
N
Cl
OC2H5
3a
24%
3.31
3.71
CN
F
N
NH2
N
Cl
Cl
Cl
3b
8%
2.79
2.71
3.44
CN
F
N
F
NH2
N
Cl
3c
3d
3e
3f
99% (4.5)
5.79
3.21
5.86
2.69
2.40
5.86
3.27
5.75
2.95
2.48
5.32
3.76
5.97
3.71
3.58
CN
F
N
NH2
N
Cl
S
CN
21%
F
N
NH2
N
Cl
Br
99% (6.25)
CN
F
N
NH2
N
Cl
NO2
CN
6%
Br
N
NH2
N
Cl
Cl
Cl
3g
3%
CN
Br
N
NH2
N
Cl
676
Chem Biol Drug Des 2013; 82: 669–684

Synthesis of Chloroquinoline Derivative
Table 1: continued
Mycobacterium
tuberculosis H37
% inhibition (6.25 lg)
Predicted
Predicted
Molecule ID Structure
Observed activity
5.31
activity (CoMFA) activity (CoMSIA) (SHD)
H5C2O
4a
97% (5)
5.32
5.35
N
F
N
NH2
N
Cl
Cl
Cl
4b
4c
4d
4e
12%
36%
9%
2.96
3.52
2.75
2.33
3.10
3.52
2.72
2.66
3.07
3.48
2.43
3.61
N
F
N
NH2
N
Cl
F
N
F
N
NH2
N
Cl
S
N
F
N
NH2
N
Cl
Br
3%
N
F
N
NH2
N
Cl
N
H2N
N
4f
14%
3.07
3.09
3.13
Cl
N
Br
Chem Biol Drug Des 2013; 82: 669–684
677

Khunt et al.
Table 1: continued
Mycobacterium
tuberculosis H37
% inhibition (6.25 lg)
Predicted
Predicted
Molecule ID Structure
Observed activity
3.17
activity (CoMFA) activity (CoMSIA) (SHD)
Cl
N
4g
16%
3.00
3.12
Cl
N
Br
NH2
NO2
N
Cl
4h
99% (6.25)
5.86
4.65
5.97
N
Br
N
NH2
N
Cl
method and their energy values were truncated at
ꢁ30 kcal/mol.
analysis (31) using the optimum number of components
was performed. This procedure involves the generation of
multiple new data sets from the original data sets, after
random sampling from the original data set. In each run,
some molecules may be excluded in the PLS analysis,
whereas some others might be included more than once.
Also y-scrambling (32) (100-trials) was performed on the
chosen 3D-QSAR models to probe the dependence of
the model on chance correlations. External predictivity of
the models was determined by calculating the rp2red from a
test set of 10 compounds that were not included in the
model generation procedure.
As an extension to the CoMFA approach which has just
two fields, CoMSIA incorporates five different interaction
energy fields, namely steric, electrostatic, hydrophobic,
hydrogen bond donor, and hydrogen bond acceptor.
CoMSIA descriptors were calculated with the same lattice
box as that used for the CoMFA calculations, with a grid
ꢀ
spacing of 2 A and a sp3 hybridized carbon atom as the
ꢀ
probe with values for the radius as 1.0 A, charge +1.0,
hydrophobicity +1.0, and H-bond properties (donor and
acceptor) +1.0. A Gaussian type distance dependence
was used between the grid point and each atom in the
molecule with a default value of 0.3 as the attenuation
factor.
The results of the 3D-QSAR studies were visualized as 3D
‘coefficient contour maps’ (contoured in terms of contribu-
tion) generated by interpolation of the pairwise products
between the PLS coefficients (coeff) and the standard
deviations (stdev) of the corresponding CoMFA and CoM-
SIA descriptor values. These maps signify those areas in
3D space where variations in steric, electrostatic, and
hydrogen bonding features in the molecular structures cor-
relate strongly with corresponding changes in activities.
Partial least squares (PLS) (28) regression was used to
derive the 3D-QSAR models with the standard implemen-
tation in the Sybyl package. The CoMFA and CoMSIA de-
scriptors were used as independent variables, while the
biological activity data served as the dependent variable.
Initial PLS regression analyses were performed in conjunc-
tion with the cross-validation (leave-one-out method)
(29,30) option to obtain the optimal number of compo-
nents which were subsequently used in deriving the final
CoMFA and CoMSIA models. To avoid over-fitting of data,
the number of components corresponding to the highest
cross-validated q2 value and the smallest PRESS value
was used. The conventional correlation coefficient r2 and
the standard error of estimate (SEE) were subsequently
computed for the final models using the optimal number of
components.
Results and Discussion
Chemistry
2-Chloroquinoline-3-carbaldehyde has been synthesized
by Vilsmeier reaction on p-fluoro or bromo substituted
acetanilide (Scheme 1). The chalcones (2a–m) have been
synthesized by condensation of compound 1 with different
aromatic ketones using the Claisen–Schmidt condensa-
tion. The chalcones on cyclization with ammonium acetate
and malononitrile or with guanidine hydrochloride furnished
cyanopyridines (3a–e) and aminopyrimidines, respectively
(4a–g). The molecules (3a–m to 4a–g) (Table 1) were
To further assess the statistical confidence and robustness
of the derived 3D-QSAR models, a 100-cycle bootstrap
678
Chem Biol Drug Des 2013; 82: 669–684

Synthesis of Chloroquinoline Derivative
1
characterized by H NMR, 13C NMR, and mass spectrom-
sibility of chance correlation. The contributions of the steric
and electrostatic fields are 47% and 53%, respectively.
The statistical significance of these models is further sup-
ported by the ‘fitness plot’ which shows the observed ver-
sus predicted activity of training and test-set compounds
and provides an idea about how well the model was
trained and how well it predicts the activity of the external
test set (Figures 1A and 2A).
etry techniques and their purity established by elemental
analysis. IR spectra of molecules (3a–m) show the charac-
teristic band at 2150–2200 per cm for the cyano group
and the characteristic band at 3400–3200 per cm for the
amino group. 1H NMR spectra are also in agreement with
these structures.
Biological activity
The CoMSIA analysis was performed using the same
structural alignment, PLS protocol, and the training/ test
sets as defined in the CoMFA study. Several models were
developed considering steric, electrostatic, hydrogen
bonding, and hydrophobic fields either separately or in
combinations to determine which of the five fields are
actually able to explain the structure activity relationships.
It was observed that using the steric, hydrogen bond
donor, hydrogen bond acceptor, and hydrophobic fields
when used independently yielded models with a lower q2
value. However, the CoMSIA model derived from the com-
bination of steric, hydrophobic, and H-bond donor field
produced a model with the highest cross-validated coeffi-
cient (q2) of 0.61, indicating a good predictive capacity
with SEE of 0.12 and the F-test value of 147.47. The
squared correlation coefficient r2 of 0.96 with boot strap
rb2s of 0.97 suggests a good internal consistency exists
within the underlying data set. Furthermore, a low r2 of
0.15 obtained by y-scrambling eliminates the possibility of
chance correlation. The relationship between experimental
and predicted activities for the training/test-set compounds
is illustrated in Figures 1B and 2B. The contributions of
steric, hydrophobic, and hydrogen bond donor fields to
this model are 25%, 28%, and 47%, respectively.
The analogs 2h, 4a, and 4h exhibit promising activity and
inhibit the growth of mycobacterium to 97%, 98%, and
99%, respectively, at a concentration of 6.25 lg/mL. Ana-
logs 3c and 3e were found to be the most active of the
series inhibiting the growth of drug-sensitive bacteria to
99% at 4.25 and 6.25 lg/mL concentration, respectively.
The remaining analogs of the series produced moderate
inhibition in the range of 5–41% (Table 1). A visualizing
SAR suggests that R = 4-F-C6H5 is more active against
Mycobacterium HRV-37 in case of chalcone and cyano-
pyridine derivatives.
3D-QSAR studies
The major objective behind CoMFA and CoMSIA analysis
is to find features associated with activity within the sys-
tem. The CoMFA and CoMSIA models based on atom-fit
alignment are given in Table 2, which shows that all the
statistical indices are in an acceptable domain.
The CoMFA model generated for the 2-chloroquinolines
has (rc2v) of 0.69 and a conventional correlation coefficient
(r2) of 0.99 with a bootstrap correlation coefficient (rb2s) of
0.98. Also a low standard error of estimate (SEE) of 0.09
with an excellent F value of 384.18 suggests the statistical
significance of the derived CoMFA model. These values
suggest that the model should be able to predict the activ-
ity of compounds outside the training set but within this
structural class. Furthermore, a significantly low r2 of 0.11
obtained for y-scrambling (100 cycles) eliminates the pos-
The predictive ability of the 3D-QSAR models was gauged
from the rp2red calculated for each of these models. The
predicted r2 values for the CoMFA and CoMSIA models
were found to be 0.64 and 0.60, respectively. These val-
ues suggest that models can be used to predict the inhibi-
tory activity of untested compounds.
The major advantage of CoMFA and CoMSIA techniques
is that the information contained in these models can be
visualized as 3D contour plots which can be fruitfully used
to tune molecules to improve their activity. The contour
plots were generated as scalar products of coefficients
and standard deviation (stdev*coeff), associated with the
CoMFA or CoMSIA column. These contour maps indicate
regions where differences in steric, electrostatic, hydro-
phobic, and hydrogen bond fields may affect the biological
activity.
Table 2: Statistical results of the CoMFA and CoMSIA models
CoMFA
CoMSIA (SHD)
N
18
18
q2
0.69
0.99
0.64
0.98
0.11
384.18
0.09
0.61
0.96
0.60
0.97
0.15
147.47
0.12
r2
rp2red
rb2s
r2 y-scrambling
F
SE
The CoMFA steric and electrostatic contour maps
obtained from the final non-cross-validated analysis are
shown in Figures 3A and 3B, respectively, associated with
reference compound, 3e. Green contours indicate regions
where steric bulk is tolerable and can be exploited to
improve the activity while yellow contours indicate regions
Field contribution
Steric
Electrostatic
Hydrophobic
H-bond donor
0.47
0.53
0.25
0.28
0.47
Chem Biol Drug Des 2013; 82: 669–684
679

Khunt et al.
8.00
8.00
6.00
4.00
2.00
0.00
A
B
6.00
4.00
2.00
Figure 1: The ‘fitness plot’
showing the observed vs. predicted
activity of training-set compounds
for CoMFA (A) and CoMSIA (B)
models.
0.00
0.00
2.00
4.00
6.00
8.00
0.00
2.00
4.00
6.00
8.00
Experimental BA
Experimental BA
8.00
6.00
4.00
2.00
0.00
6.00
A
B
4.00
2.00
Figure 2: The ‘fitness plot’
showing the observed vs. predicted
activity of test-set compounds for
CoMFA (A) and CoMSIA (B)
models.
0.00
0.00
2.00
4.00
6.00
8.00
0.00
2.00
4.00
6.00
8.00
Experimental BA
Experimental BA
B
A
Figure 3: The CoMFA molecular
interaction fields around the most
active molecule (A) steric contours
– favored (green); disfavored
(yellow) (B) electrostatic contours –
electropositive (blue);
electronegative (red).
A
B
Figure 4: The CoMSIA molecular
interaction fields around most
active molecule (A) hydrogen bond
donor contours – favored (cyan);
disfavored (purple) (B) Hydrophobic
contours – favored (Yellow);
disfavored (white).
where steric bulk is detrimental to the activity. Large green
contours are found localized around the ortho and para
positions of ring (D) indicating that the steric bulk is
favored in this region. This is consistent with experimental
results where compounds 3c, 3e, and 4a with a para sub-
stituent are more active than those that lack substituent at
this position. As regards, compounds 2a–2m which do
not have the pyridine ring (C) at the 3rd position of the
quinoline nucleus, like 3e, the substituted aromatic ring
present at this position adopts an orientation analogous to
the most active compound 3e. The para substituent on
the phenyl ring of compounds 2f and 2h extends into the
green contour and as a result, these compounds are as
active as 3e. Yellow contours are observed around the
meta position of ring (D) indicating that any steric substitu-
ent at this position would greatly reduce the biological
activity, suggesting limited bulk tolerance. Less active
compounds, as exemplified by 3f, 3g, and 4g, are found
with their meta substituents oriented in the direction of the
yellow contour. Likewise, compounds 2g and 2m pos-
sessing a meta substituent on the phenyl ring also have
lower activity.
680
Chem Biol Drug Des 2013; 82: 669–684

Synthesis of Chloroquinoline Derivative
O
NH
CH3
X
DM F+POCl 3
X=F/Br
O
CH3
x
CHO
+
R
N
Cl
a
O
x
N
Cl
2
NH
N
+
+
N
H2N
c
NH2
b
R
R
N
N
x
x
N
NH2
N
H
NH
N
Cl
N
Cl
3
4
R= 4-OC2H5,4-F,4-Br,4-NH2. 2-NH2,2,-4,(Cl),3,4(Cl),2-Thenyl : X=F/Br.
Scheme 1: Synthesis of pyridine and pyrimidine; Reagents and condition: (a) 40% NaOH; (b) CH3COONH4; (c) K2CO3, Microwave
irradiation; Reflux..
The electrostatic contour maps of the CoMFA model are
characterized by red isopleths near the para and ortho
positions of ring (D) as well as around the cyano
substituent of pyridine ring (C). This indicates that increas-
ing electronegativity at these positions should lead to
increase in antitubercular activity. This is evident from
compounds 3c, 3e, and 4a bearing an electronegative
substituent which show good antitubercular activity. Even
compounds 2f and 2h bearing an electronegative substitu-
tion at the para position of the phenyl ring display high
Chem Biol Drug Des 2013; 82: 669–684
681

Khunt et al.
activity. A small blue contour is observed around the 5′
position of ring © indicating that a reduction in the electro-
negativity near this position should lead to increase in
activity.
almost similar fashion. While the CoMFA model could
point to the importance of steric and electrostatic fields,
the CoMSIA model could explain the contribution of
hydrophobic and hydrogen bond donor fields as well.
CoMFA/CoMSIA models possess good predictive ability
as discerned from the result for the external test set and
were used to guide the further development of some
new 2-chloroquinolines which are predicted with
improved activities.
The CoMSIA steric regions are in agreement with the
CoMFA steric contours and therefore for the sake of brev-
ity are not discussed again. The hydrophobic map of the
CoMSIA model cloaked around the most potent com-
pound 3e is displayed in Figure 4A. The presence of yel-
low contour around the ortho and para positions of ring
(D) suggests that occupancy by hydrophobic groups at
these sites would favor increase in the activity. Com-
pounds 3c, 3e, and 4a possessing such substituents at
these positions display better activity than compounds
lacking such substituents. Similarly, compounds 2f and 2h
with a para substituent on the phenyl ring also show
higher activity than compounds without a corresponding
group. Also a small white contour disfavoring hydrophobic
substitution is observed around the meta position of ring
(D). This is supported by compounds 3f, 3g, 4g, and 4h
with substituents at this position which have poor antitu-
bercular activity. As regards H-bonding, a cyan contour
favoring the presence of a hydrogen bond donor function-
ality is observed around the amino group on the pyridine
ring (C) while a purple contour disfavoring the presence of
an H-bond donor substituent is observed around the
cyano group placed at the 3′ position of the pyridine ring
(C) in the most active compound 3e.
Acknowledgments
RCK thanks the Department of Chemistry, Saurashtra Uni-
versity which is supported by SAP (UGC) for infrastructure
facilities and FIST(DST), New Delhi for financial support
under the Fast Track Young Scientist Scheme. We also
thank Dhanshukh Rajani, Microcare Laboratory, Surat for
the antitubercular screening. The computational facilities
were jointly provided by the All India Council of Technical
Education through grant (F. No. 8022/RID/NPROJ/RPS-5/
2003-04), the Department of Science and Technology
through their FIST program (SR/FST/LSI-163/2003),
Department of Biotechnology (BT/PR11810/BRB/10/690/
2009), and the Council of Scientific and Industrial
Research (01/23/99/10-EMR-II).
References
Thus, in the absence of information on the binding mode
of these 2-chloroquinoline derivatives, the putative interac-
tion fields obtained from this comparative 3D-QSAR study
will be helpful for improving the antitubercular activity in
this series of molecules. Considering this fact, a set of
new 2-chloroquinoline analogs has been proposed as dis-
cussed in the next section which display higher antituber-
cular activity.
1. Asgharzadeh M., Kafil H.S. (2007) Current trends in
molecular
epidemiology
studies
of
Mycobac-
terium tuberculosis. Biotechnol Mol Biol Rev;2:108–
115.
2. Chakraborty A.K. (2004) Epidemiology of tuberculosis:
current status in India. Indian J Med Res;120:248–
276.
3. World Health Organization. (2007) Global Tuberculosis
Control: Surveillance, Planning, Financing: WHO Report
2008. Geneva, Switzerland: World Health Organization.
4. Scior T., Meneses Morales I., Garces Eisele S.J., Do-
meyer D., Laufer S. (2002) Antitubercular isoniazid and
drug resistance of Mycobacterium tuberculosis–a
review. Arch Pharm (Weinheim);335:511–525.
5. Iseman M.D. (2002) Tuberculosis therapy: past, present
and future. Eur Respir J Suppl;36:87s–94s.
6. Ginsberg A.M. (2008) Emerging drugs for active tuber-
culosis. Semin Respir Crit Care Med;29:552–559.
7. Tripathi R.P., Tewari N., Dwivedi N., Tiwari V.K. (2005)
Fighting tuberculosis: an old disease with new chal-
lenges. Med Res Rev;25:93–131.
Conclusions
The synthesis and biological evaluation of a series of 2-
chloro-3-substitued 6-bromo/fluoroquinolines as antituber-
cular agents has been discussed. Most importantly, this
work validated our initial proposition that pyridine/pyrimi-
dine-substituted quinoline scaffolds would lead to mole-
cules with potent antitubercular activity. 3-Formylquinoline
synthesized by the Vilsmeier reaction, on Claisen con-
densation with substituted acetophenones yielded the
chalcones. These chalcones were converted to a variety
of 2-chloro-3-substitued-6-bromo/fluoro quinolines which
were then screened for antitubercular activity. The study
resulted in the identification of compounds 2f, 2h, 3c,
3e, 4a, and 4h as promising inhibitors of M. tuberculo-
sis. The q2 and r2 values of the CoMFA and CoMSIA
models were statistically sound and suggest that all the
reported inhibitors would bind to the receptor in an
8. Nikonenko B.V., Protopopova M., Samala R., Einck L.,
Nacy C.A. (2007) Drug therapy of experimental tuber-
culosis (TB): improved outcome by combining SQ109,
a new diamine antibiotic, with existing TB drugs. Anti-
microb Agents Chemother;51:1563–1565.
682
Chem Biol Drug Des 2013; 82: 669–684

Synthesis of Chloroquinoline Derivative
9. Gelber R., Andries K., Paredes R.M.D., Andaya C.E.S.,
Burgos J. (2009) The diarylquinoline R207910 is bacte-
ricidal against Mycobacterium leprae in mice at low
dose and administered intermittently. Antimicrob
Agents Chemother;53:3989.
21. Klebe G., Abraham U., Mietzner T. (1994) Molecular
similarity indices in a comparative analysis (CoMSIA) of
drug molecules to correlate and predict their biological
activity. J Med Chem;37:4130–4146.
22. Klebe G. (1998) Comparative molecular similarity indi-
ces analysis: CoMSIA. Perspect Drug Discov
Design;12:87–104.
23. Klebe G., Abraham U. (1999) Comparative molecular
similarity index analysis (CoMSIA) to study hydrogen-
bonding properties and to score combinatorial libraries.
J Comput Aided Mol Des;13:1–10.
ꢁ
10. Molina P., Aller E., Lorenzo A., Lopez-Cremades P.,
Rioja I., Ubeda A., Terencio M.C., Alcaraz M.J. (2001)
Solid-phase synthesis and inhibitory effects of some
pyrido[1,2-c]pyrimidine derivatives on leukocyte forma-
tions and experimental inflammation. J Med Chem;
44:1011–1014.
11. Bhardwaj S.D., Jolly V.S. (1997) Synthesis, anti-HIV
and anticancer activities of some new formazans.
Asian J Chem;9:48–51.
12. Nayyar A., Monga V., Malde A., Coutinho E., Jain R.
(2007) Synthesis, anti-tuberculosis activity, and 3D-
QSAR study of 4-(adamantan-1-yl)-2-substituted quin-
olines. Bioorg Med Chem;15:626–640.
13. Nayyar A., Malde A., Jain R., Coutinho E. (2006) 3D-
QSAR study of ring-substituted quinoline class of
anti-tuberculosis agents. Bioorg Med Chem;14:847–
856.
24. Yajko D.M., Madej J.J., Lancaster M.V., Sanders C.A.,
Cawthon V.L., Gee B., Babst A., Hadley W.K. (1995)
Colorimetric method for determining MICs of antimicro-
bial agents for Mycobacterium tuberculosis. J Clin
Microbiol;33:2324–2327.
25. Suling W.J., Seitz L.E., Pathak V., Westbrook L., Bar-
row E.W., Zywno-Van-Ginkel S., Reynolds R.C., Piper
J.R., Barrow W.W. (2000) Antimycobacterial activities
of 2,4-diamino-5-deazapteridine derivatives and effects
on mycobacterial dihydrofolate reductase. Antimicrob
Agents Chemother;44:2784–2793.
14. Nayyar A., Malde A., Coutinho E., Jain R. (2006) Syn-
thesis, anti-tuberculosis activity, and 3D-QSAR study
of ring-substituted-2/4-quinolinecarbaldehyde deriva-
tives. Bioorg Med Chem;14:7302–7310.
15. Ubiali D., Pagani G., Pregnolato M., Piersimoni C., Ped-
raz Munoz J.L., Rodriguez Gascon A., Terreni M. (2002)
New N-alkyl-1,2-dihydro-2-thioxo-3-pyridinecarbothioa-
mides as antituberculous agents with improved
pharmacokinetics. Bioorg Med Chem Lett;12:2541–
2544.
16. Rosowsky A., Forsch R.A., Queener S.F. (2002) Inhibi-
tion of Pneumocystis carinii, Toxoplasma gondii, and
Mycobacterium avium dihydrofolate reductases by 2,4-
diamino-5-[2-methoxy-5-(omega-carboxyalkyloxy)ben-
zyl]pyrimidines: marked improvement in potency
relative to trimethoprim and species selectivity relative
to piritrexim. J Med Chem;45:233–241.
26. Manvar A., Malde A., Verma J., Virsodia V., Mishra A.,
Upadhyay K., Acharya H., Coutinho E., Shah A. (2008)
Synthesis, anti-tubercular activity and 3D-QSAR study
of coumarin-4-acetic acid benzylidene hydrazides. Eur
J Med Chem;43:2395–2403.
27. Golbraikh A., Tropsha A. (2002) Predictive QSAR mod-
eling based on diversity sampling of experimental data-
sets for the training and test set selection. J Comput
Aided Mol Des;16:357–369.
28. Wold S., Johansson E., Cocchi M. (1993) PLS: partial
least squares projections to latent structures. In: Kub-
inyi H., editor. PLS: Partial Least Squares Projections
to Latent Structures. Leiden: ESCOM Science Publish-
ers; p. 523–550.
29. Stone M. (1974) Cross-validatory choice and assess-
ment of statistical predictions.
J Roy Stat Soc
B;36:111–147.
17. Desai B., Sureja D., Naliapara Y., Shah A., Saxena
A.K. (2001) Synthesis and QSAR studies of 4-substi-
30. Richard D., Cramer R.D. III, Bunce J.D., Patterson
D.E., Frank I.E. (1988) Crossvalidation, bootstrapping,
and partial least squares compared with multiple
regression in conventional QSAR studies. Quant
Struct-Act Relat;7:18–25.
31. Shao J. (1996) Bootstrap model selection. J Am Stat
Assoc;91:655–665.
32. Rucker C., Rucker G., Meringer M. (2007) y-Randomi-
zation and its variants in QSPR/QSAR. J Chem Inf
Model;47:2345–2357.
tuted
phenyl-2,6-dimethyl-3,
5-bis-N-(substituted
phenyl)carbamoyl-1,4-dihydropyridines as potential
antitubercular agents. Bioorg Med Chem;9:1993–
1998.
18. Khunt R.C., Khedkar V.M., Chawda R.S., Chauhan
N.A., Parikh A.R., Coutinho E.C. (2012) Synthe-
sis, antitubercular evaluation and 3D-QSAR study of
N-phenyl-3-(4-fluorophenyl)-4-substituted
pyrazole
derivatives. Bioorg Med Chem Lett;22:666–678.
19. Cramer R.D. III, Patterson D.E., Bunce J.D. (1988)
Comparative molecular field analysis (CoMFA). I. effect
of shape on binding of steroids to carrier proteins. J
Am Chem Soc;110:5959–5967.
20. Cramer R.D. III, Patterson D.E., Bunce J.D. (1989)
Recent advances in comparative molecular field analy-
sis (CoMFA). Prog Clin Biol Res;291:161–165.
Notes
aSybyl. (2005) Sybyl. In: editor^editors. Sybyl. 1699 S
Hanley Rd., St. Louis, MO 631444, USA: Tripos Associ-
ates Inc.; p.
bAccelrys I. (1998) Cerius2. In: editor^editors. Cerius2.
San Diego, CA, USA; p.

Khunt et al.
Figure S3. NMR spectra of compound 3b.
Figure S4. NMR spectra of compound 4a.
Figure S5. Mass spectra of compound 2a.
Figure S6. Mass spectra of compound 3a.
Figure S7. Mass spectra of compound 4a.
Supporting Information
Additional Supporting Information may be found in the
online version of this article:
Figure S1. NMR spectra of compound 2a.
Figure S2. Expanded NMR spectra of compound 2a.
684
Chem Biol Drug Des 2013; 82: 669–684