Molecular modeling and antimycobacterial studies of Mannich bases: 5-hydroxy-2-methyl-4H-pyran-4-ones

Article abstract of DOI:10.3906/kim-1006-689

The World Health Organization lists tuberculosis among the top 3 leading causes of death from a single infectious agent, and reported cases of multidrug-resistant tuberculosis (MDR-TB) are on the rise. In an attempt to improve MDR-TB drug-directed therapy

Full text of DOI:10.3906/kim-1006-689

Turk J Chem
35 (2011) , 317 – 330.
¨
˙
c
ꢀ TUBITAK
doi:10.3906/kim-1006-689
Molecular modeling and antimycobacterial studies of
Mannich bases: 5-hydroxy-2-methyl-4H-pyran-4-ones
1
1
2
2
¨
¨
Barkın BERK , Demet US , Sinem OKTEM , Z. Tanıl KOCAGOZ ,
3
3
1,∗
˘
˙
Berrak C¸ AGLAYAN , I¸sıl AKSAN KURNAZ , Dilek DEMIR EROL
1Yeditepe University, Faculty of Pharmacy, Department of Pharmaceutical Chemistry,
˙
34755, Kayı¸sda˘gı, Istanbul-TURKEY
e-mail: derol@yeditepe.edu.tr.
2Acıbadem University, Faculty of Medicine, Department of Microbiology,
˙
Istanbul-TURKEY
3Yeditepe University, Faculty of Engineering and Architecture, Department of Genetics and Bioengineering,
˙
34755, Kayı¸sda˘gı, Istanbul-TURKEY
Received 19.06.2010
The World Health Organization lists tuberculosis among the top 3 leading causes of death from a single
infectious agent, and reported cases of multidrug-resistant tuberculosis (MDR-TB) are on the rise.
In an attempt to improve MDR-TB drug-directed therapy, we synthesized 11 4-substituted piper-
azine derivatives of 3-hydroxy-6-methyl-4H -pyran-4-one pharmacophore by reacting 5-hydroxy-2-methyl-
4H -pyran-4-one with suitable piperazine derivatives under Mannich reaction conditions.
Inhibitory effects of the 11 compounds on Escherichia coli DNA gyrase were evaluated via DNA gyrase
supercoiling assay. The minimum inhibitory concentrations (MIC) of the 11 compounds and 41 compounds
from our previous studies against Mycobacterium tuberculosis H37RV were assessed, in vitro, by a broth
dilution method. To determine the interaction pattern between active site amino acids and all 52 compounds,
homology modeling for the construction of M. tuberculosis DNA gyrase B subunit was performed, followed
by a docking study.
The data presented here could prove useful in future studies on interaction field analysis and high
throughput virtual screening of the derivatives of the 3-hydroxy-6-methyl-4H -pyran-4-one pharmacophore
toward the development of more clinically applicable compounds.
Key Words: Antimycobacterial activity, DNA gyrase activity, hydroxy-4H -pyran-4-one, homology mod-
eling
∗Corresponding author
317

Molecular modeling and antimycobacterial studies of Mannich bases:..., B. BERK, et al.,
Introduction
The World Health Organization estimates that about 8-10 million new tuberculosis (TB) cases occur annually,
worldwide. Tuberculosis is listed among the top 3 leading causes of death from a single infectious agent, along
with malaria and the human immunodeficiency virus (HIV).1 In the past decade, it has been reported that
the incidence of multidrug-resistant TB (MDR-TB) is on the rise, not only in developing countries but also
in industrialized nations.1 Moreover, enhanced susceptibility to TB in HIV-infected populations and increased
cases of the highly contagious and life-threatening pulmonary form of TB pose a serious worldwide health
concern. This underscores the importance of the development of new nonresistant anti-TB drugs and new
protocols for the efficacious clinical control of TB patients using the antimycobacterial drugs available today.1
The primary cause of MDR-TB stems from the sophisticated genomic structure of Mycobacterium tu-
berculosis. The M. tuberculosis genome comprises 4,411,529 base pairs, contains around 4000 genes, and has a
very high guanine and cytosine content that is reflected in the biased amino acid protein content.2 M. tuber-
culosis differs radically from other bacteria in that a very large portion of its coding capacity is devoted to the
production of enzymes, and it is highly flexible to adaptation and variation.2
The M. tuberculosis genome encodes only 2 topoisomerase-producing genes, one for DNA gyrase, a type
II DNA topoisomerase, and the other for topoisomerase I, excluding the topoisomerase IV gene, another type
II DNA topoisomerase involved in the DNA replication process. In such cases, DNA gyrase is the only type II
enzyme responsible for the negative supercoiling of DNA by relaxing and decatenating positively supercoiled
DNA.3
DNA gyrase, which is found in all bacteria, is an adenosine triphosphate (ATP)-dependant hydrolytic
enzyme, a proven target for antibacterial chemotherapy.4 DNA gyrase comprises two subunits, GyrA and GyrB,
forming functional heterodimer A2B2. The GyrA subunit is responsible for DNA breakage and reunion. This
catalytic process involves large conformational changes in the enzyme, which are triggered by the binding and
hydrolysis of ATP on the GyrB subunit.5
Presently, fluoroquinolones are the only inhibitors of DNA gyrase A commonly used by physicians to
treat bacterial infections.6−8 However, reports of fluoroquinolone side effects and toxicity,9−11 as well as the
emergence of bacterial fluoroquinolone resistance,7,12 have revived a growing interest in alternative compounds
such as inhibitors of the ATPase catalytic domain of DNA gyrase.13 The most studied and characterized
inhibitors of GyrB are coumarin-like natural antibiotics such as novobiocin, clorobiocin, and cyclothialidines,
isolated from Streptomyces microorganisms, and synthetic derivatives of triazine and indazole ring structures. 5
However, all were reported to have either high toxicity or low solubility and permeability during in vivo tests.14,15
X-ray crystallographic studies on DNA gyrase B structural motifs in complex with inhibitors of microor-
ganisms such as Thermus thermophilus, Escherichia coli, and Staphylococcus aureus have been published.16−24
Although these homologs have similar catalytic activities compared to M. tuberculosis, the crystallographic
structures of M. tuberculosis DNA gyrases have not been reported yet.
Compounds derived from 3-hydroxy-4(1H)-pyranone structurally resemble the coumarin ring system37,38
and exhibit antimicrobial,25−29 antitumor,30,31 anticonvulsant,32−34 and tyrosinase inhibitory effects.35,36 Pre-
viously, various 2-substituted piperazine/piperidine-1-yl-methyl derivatives of 3-hydroxy-6-methyl-4H-pyran-4-
ones37,38 were coupled in compounds 5a-w and 6a-s, shown in Tables 1 and 2, respectively,33,34,39 and were
318

Molecular modeling and antimycobacterial studies of Mannich bases:..., B. BERK, et al.,
Table 1. MIC (μg/mL) values determined for the piperidine derivatives against M. tuberculosis H37RV.
R
MIC (μg/mL)
R
MIC (μg/mL)
6a
6b
6c
6d
6e
6f
hydrogen
2-methyl
3-methyl
4-methyl
2,6-dimethyl
3,5-dimethyl
16
64
64
16
64
8
6j
6k
6l
4-hydroxy
16
32
32
16
16
16
16
16
8
3-(hydroxymethyl)
4-(hydroxymethyl)
4-phenyl
6m
6n 4-(4-hydroxyphenyl)
6o
6p
6r
6s
4-piperidin-1-yl
4-(4-morpholino)
4-benzyl
6g 5-ethyl-2-methyl
64
16
32
6h
6i
4-propyl
3-hydroxy
4-ethoxycarbonyl
Table 2. MIC (μg/mL) values determined for the 4-substituted piperazine derivatives against M. tuberculosis H37RV.
R
MIC (μg/mL)
R
MIC (μg/mL)
4a
4b
2-cyanophenyl
2-cyanoethyl
32
32
32
16
32
64
128
32
64
128
128
8
5g
5h
5i
4-methylphenyl
4-chlorophenyl
4-(trifluoromethyl)phenyl
4-cyanophenyl
4-nitrophenyl
16
8
4c 3,4-dichlorophenyl
8
4d
4e
4f
4-chlorobenzyl
4-hydroxyphenyl
2-phenylethyl
2-ethoxyethyl
2-methylphenyl
prop-2-en-1-yl
2-methoxyethyl
propan-2-yl
5j
16
16
16
16
16
16
16
16
16
8
5k
5l
cyclohexyl
4g
4h
4i
5m
5n
5o
5p
pyridine-2-yl
pyridine-4-yl
phenylcarbonyl
furan-2-ylcarbonyl
4j
4k
5a
5b
5c
5q tetrahydrofuran-2-ylmethyl
phenyl
5r
5s
cyclohexyl methyl
N-methylpiperidin-4-yl
tert-butoxycarbonyl
ethoxycarbonyl
2-chlorophenyl
2-fluorophenyl
4
16
16
8
5t
8
5d 2-methoxyphenyl
5u
5v
5w
16
16
16
5e
5f
3-chlorophenyl
2-hydroxyethyl
3-methoxyphenyl
16
2-(dimethylamino)ethyl
319

Molecular modeling and antimycobacterial studies of Mannich bases:..., B. BERK, et al.,
then screened in vitro for their antimicrobial activities, quantified in MIC (μg/mL), against E. coli, S. paraty-
phi, S. flexneri, E. gergoviae, and Mycobacterium smegmatis via a broth dilution technique. The MIC results
obtained, as low as 4 μg/mL against M. smegmatis, were found to be promising, indicating that these structures
could potentially have high efficacy in low concentrations against M. tuberculosis.37,38
In this study, 11 2-[(4-substitutedpiperazin-1-yl-methyl] derivatives of 3-hydroxy-6-methyl-4H-pyran-4-
one (compounds 4a-k, Figure 1 and Table 2) were synthesized and characterized by physicochemical, spectral
(IR, 1 H-NMR), and elemental analysis methods. The 11 newly synthesized compounds were screened for
supercoiling inhibitory activity by a commercial E. coli DNA gyrase supercoiling assay (Figure 2). These
compounds were also coupled with 41 previously synthesized 2-substituted piperazine/piperidine-1-yl-methyl
derivatives of 3-hydroxy-6-methyl-4H-pyran-4-ones.33,34,37−39 The desired products were then tested for their
antimycobacterial activities against M. tuberculosis H37RV via broth dilution to determine their MICs (Tables
1 and 2).
O
O
O
O
OH
R
OH
OH
OH
N
(a)
(b)
(c)
Cl
N
OH
O
H3C
H3C
O
O
O
(1)
(2)
(3)
(4a-k)
Figure 1. General synthesis of 3-hydroxy-6-methyl-2-[(4-substitutedpiperazin-1-yl)methyl]-4H-pyran-4-one derivatives
(compounds 4a-k in Table 2).
Figure 2. Effects of 11 newly synthesized 3-hydroxy-6-methyl-2-((4-substitutedpiperazin-1-yl)methyl)-4H -pyran-4-one
derivatives (compounds 4a-k in Table 2) on DNA gyrase supercoiling activity. The bands in the agarose gel pictures
indicate the topoisomers of the pBR322 plasmid. OC and S indicate the open-circular and supercoiled pBR322 plasmid
DNA, respectively.
320

Molecular modeling and antimycobacterial studies of Mannich bases:..., B. BERK, et al.,
The structure of the unknown DNA gyrase B chain for M. tuberculosis was constructed by the known
amino acid sequence of DNA gyrase B for M. tuberculosis, available in the UniProtKB database (UniProtKB/Swiss-
Prot P0C5C5), and a similar template X-ray crystal structure of the 43 K ATPase domain of Thermus ther-
mophilus gyrase B in complex with novobiocin (PDB Code: 1KIJ).40 Finally, we performed core-constrained
rigid docking experiments using all of the new and previously synthesized compounds to establish structure-
activity relations based on activity and modeling data.
Experimental methods
All reagents were obtained from commercial sources. Solvents were dried and purified by known conventional
methods. DNA Gyrase Assay Kit 3 (1000U-K0003) and DNA gyrase enzyme (100U-G1001) were purchased
from Inspiralis (Norwich, UK). Molecular Operating Environment (MOE 2009.10) (Chemical Computing Group,
Montreal, Canada) and Maestro (Schro¨dinger Corp., Portland, USA) software were used for modeling purposes.
Melting points were detected with a Mettler-Toledo FP-62 melting point apparatus (Columbus, USA) and are
uncorrected. IR spectra (KBr) were recorded on a Perkin Elmer 1720X FT-IR spectrometer (Beaconsfield,
UK). 1 H-NMR spectra were obtained with a Varian Mercury 400, 400 MHz, High Performance Digital FT-
NMR using DMSO-d6 , CDCl3 , and CD3 OD as solvents and tetramethylsilane as an internal standard. All
chemical shift values (δ) were recorded in ppm. Compound purity was monitored by thin-layer chromatography
on silica gel-coated aluminum sheets (Merck, 1.005554, silica gel HF254-361, Type 60, 0.25 mm; Darmstadt,
Germany). Elemental analysis of compounds was performed with a LECO CHNS 932 analyzer (LECO Corp.,
Michigan, USA). Elemental analysis for C, H, and N were within 0.4% of theoretical values. 1 H-NMR spectra
and elemental analysis were performed at the Central Analysis Laboratory of Ankara University, Faculty of
Pharmacy, in Ankara, Turkey.
Chemistry
General method for synthesis of 2-(chloromethyl)-5-hydroxy-4H-pyran-4-one (compound 2)
Kojic acid (142 g, 2 mol) (compound 1, Figure 1) was dissolved in thionyl chloride (237 g, 2 mol), followed by
stirring for 2 h at room temperature. The resultant yellow solid was filtered and washed with cold petroleum
ether. Recrystallization from water gave light yellow crystals (115 g, 72%), mp 146-147 ◦ C (lit. value41 146-
147 ◦ C). 1 H-NMR δ (ppm, DMSO-d6), 9.32 (s, 1H), 8.13 (s, 1H), 6.57 (s, 1H), 4.66 (s, 2H). Anal. Cal. For.
C6 H5 ClO3 : C 44.88, H 3.14. Found: C 44.62, H 3.03.
General method for synthesis of 5-hydroxy-2-methyl-4H-pyran-4-one (compound 3)
Compound 2 (20 g, 0.12 mol) was suspended in water (500 mL). The temperature of the reaction mixture
was raised to 50 ◦ C. Zinc dust (16 g, 0.24 mol) was added and stirred at 70 ◦ C for 30 min. Concentrated
hydrochloric acid (13.6 g, 3 mol) was added dropwise, followed by stirring for 3 h at 70 ◦ C. The solution
was filtered, poured into ice water, extracted with dichloromethane, dried with anhydrous sodium sulfate, and
evaporated to dryness. Recrystallization of the resulting yellow solid from isopropanol yielded compound 3 as
41
light yellow needles (10.2 g, 64%), mp 125-127 ◦ C (lit. value
125-127 ◦ C). 1 H-NMR δ (ppm, DMSO-d6),
321

Molecular modeling and antimycobacterial studies of Mannich bases:..., B. BERK, et al.,
8.96 (s, 1H), 7.97 (s, 1H), 6.24 (s, 1H), 2.24 (s, 3H). Anal. Cal. For. C6 H6 O3 : C 57.14, H 4.80. Found: C
57.35, H 4.67.
General method for synthesis of 3-hydroxy-6-methyl-2-[(4-substitutedpiperazin-1-yl)methyl]-4H-
pyran-4-one derivatives (compounds 4a-k, Figure 1, Table 2)
A mixture of 0.01 mol of 4-substituted piperazine and 0.01 mol of compound 3 (Figure 1) in 20 mL of methanol
with 0.012 mol of 37% formalin was stirred at room temperature until a solid mass precipitated. The compound
was filtered and washed with cold water, dried under a vacuum, and recrystallized from suitable solvents.
2-{4-[(3-Hydroxy-6-methyl-4-oxo-4H-pyran-2-yl)methyl]piperazin-1-yl}benzonitrile (compound 4a)
Recrystallized from methanol. Mp 158.0 ◦ C. IR (KBr) (υ, cm−1 , stretching), 2218 (C≡N), 1614 (C=O), 1452
(C=C), 1218 (C-O); 1 H-NMR δ (ppm, CDCl3), 7.55-7.49 (m, 4H), 6.23 (s, 1H), 3.72 (s, 2H), 3.26 (t, J = 4.4
Hz, 4H), 2.83 (t, J = 4.8 Hz, 4H), 2.32 (s, 3H). Anal. Cal. For. C18 H19 N3 O3 : C 66.45, H 5.89, N 12.91.
Found: C 66.34, H 5.85, N 12.65.
3-{4-[(3-Hydroxy-6-methyl-4-oxo-4H-pyran-2-yl)methyl]piperazin-1-yl}propanenitrile (compound
4b)
Recrystallized from methanol. Mp 120.4 ◦ C. IR (KBr) (υ, cm−1 , stretching), 2244 (C≡N), 1620 (C=O), 1454
(C=C), 1197 (C-O); 1 H-NMR δ (ppm, CDCl3), 6.21 (s, 1H), 3.65 (s, 2H), 2.72-2.49 (m, 12H), 2.30 (s, 3H).
Anal. Cal. For. C14 H19 N3 O3 : C 60.63, H 6.91, N 15.15. Found: C 60.73, H 6.80, N 14.96.
2-{[4-(3,4-Dichlorophenyl)piperazin-1-yl]methyl}-3-hydroxy-6-methyl-4H-pyran-4-one (compound
4c)
Recrystallized from methanol. Mp 178.5 ◦ C. IR (KBr) (υ, cm−1 , stretching), 1626 (C=O), 1455 (C=C), 1222
(C-O); 1 H-NMR δ (ppm, CDCl3), 7.27 (m, 1H), 6.94 (d, J = 2.8 Hz, 1H), 6.72 (m, 1H), 6.24 (s, 1H), 3.69
(s, 2H), 3.19 (t, J = 4.8 Hz, 4H), 2.73 (t, J = 4.8 Hz, 4H), 2.32 (s, 3H). Anal. Cal. For. C17 H18 Cl2 N2 O3 -
0.25HOH: C 54.63, H 4.99, N 7.50. Found: C 54.64, H 5.20, N 7.69.
2-{[4-(4-Chlorobenzyl)piperazin-1-yl]methyl}-3-hydroxy-6-methyl-4H-pyran-4-one (compound 4d)
Recrystallized from methanol/diethyl ether. Mp 159.9 ◦ C. IR (KBr) (υ, cm−1 , stretching), 1619 (C=O), 1456
(C=C), 1199 (C-O); 1 H-NMR δ (ppm, CDCl3), 7.26 (dd, J = 8.2 Hz, 4H), 6.20 (s, 1H), 3.64 (s, 2H), 3.47
(s, 2H), 2.63 (s, 4H), 2.49 (s, 4H), 2.84 (s, 3H). Anal. Cal. For. C18 H21 ClN2 O3 -0.2HOH: C 61.34, H 6.12, N
7.95. Found: C 61.38, H 6.24, N 8.00.
322

Molecular modeling and antimycobacterial studies of Mannich bases:..., B. BERK, et al.,
3-Hydroxy-2-{[4-(4-hydroxyphenyl)piperazin-1-yl]methyl}-6-methyl-4H-pyran-4-one (compound
4e)
Recrystallized from methanol. Mp 178.9◦ C. IR (KBr) (υ, cm−1 , stretching), 1624 (C=O), 1510 (C=C), 1220
(C-O); 1 H-NMR δ (ppm, CD3 OD), 6.78 (dd, J = 9.2 Hz, 4H), 6.26 (s, 1H), 3.73 (s, 2H), 3.06 (t, J = 4.8 Hz,
4H), 2.76 (t, J = 4.8 Hz, 4H), 2.34 (s, 3H). Anal. Cal. For. C17 H20 N2 O4 : C 64.54, H 6.37, N 8.86. Found:
C 64.19, H 6.34, N 8.82.
3-Hydroxy-6-methyl-2-{[4-(2-phenylethyl)piperazin-1-yl]methyl}-4H-pyran-4-one (compound 4f)
Recrystallized from chloroform/n-hexane. Mp 182.9 ◦ C. IR (KBr) (υ, cm−1 , stretching), 1626 (C=O), 1432
(C=C), 1223 (C-O); 1 H-NMR δ (ppm, CDCl3), 7.24 (m, 5H), 6.20 (s, 1H), 3.67 (s, 2H), 2.80-2.60 (m, 12H),
2.29 (s, 3H). Anal. Cal. For. C19 H24 N2 O3 : C 69.49, H 7.37, N 8.53. Found: C 69.13, H 7.35, N 8.33.
2-{[4-(2-Ethoxyethyl)piperazin-1-yl]methyl}-3-hydroxy-6-methyl-4H-pyran-4-one (compound 4g)
Recrystallized from methanol/diethyl ether. Mp 133.0 ◦ C. IR (KBr) (υ, cm−1 , stretching), 1626 (C=O), 1450
(C=C), 1108 (C-O); 1 H-NMR δ (ppm, CDCl3), 6.20 (s, 1H), 3.66 (s, 2H), 3.56-3.46 (m, 4H), 2.66-2.58 (m,
10H), 2.29 (s, 3H), 1.19 (t, J = 7.2 Hz, 3H). Anal. Cal. For. C15 H24 N2 O4 : C 60.79, H 8.16, N 9.45. Found:
C 60.58, H 8.34, N 9.16.
3-Hydroxy-6-methyl-2-{[4-(2-methylphenyl)piperazin-1-yl]methyl}4H-pyran-4-one (compound 4h)
Recrystallized from ethyl acetate/diethyl ether. Mp 220.7 ◦ C. IR (KBr) (υ, cm−1 , stretching), 1617 (C=O),
1437 (C=C), 1224 (C-O); 1 H-NMR δ (ppm, CDCl3), 7.15 (m, 4H), 6.33 (s, 1H), 4.36 (s, 2H), 3.65 (s, 4H), 3.17
(s, 4H), 2.44 (s, 3H), 2.27 (s, 3H). Anal. Cal. For. C18 H22 N2 O3 -2.5HOH: C 60.15, H 7.57, N 7.79. Found: C
60.27, H 7.05, N 7.63.
3-Hydroxy-6-methyl-2-{[4-(prop-2-en-1-yl)piperazin-1-yl]methyl}-4H-pyran-4-one (compound 4i)
Recrystallized from methanol/diethyl ether. Mp 152.2 ◦ C. IR (KBr) (υ, cm−1 , stretching), 1616 (C=O), 1456
(C=C), 1201 (C-O); 1 H-NMR δ (ppm, CDCl3), 6.20 (s, 1H), 5.87-5.80 (m, 1H), 5.21-5.14 (m, 2H), 3.66 (s,
2H), 3.01 (d, J = 6.4 Hz, 2H), 2.70-2.50 (m, 8H), 2.29 (s, 3H). Anal. Cal. For. C14 H20 N2 O3 : C 63.62, H
7.63, N 10.60. Found: C 63.38, H 7.89, N 10.37.
3-Hydroxy-2-{[4-(2-methoxyethyl)piperazin-1-yl]methyl}-6-methyl-4H-pyran-4-one (compound 4j)
Recrystallized from methanol/diethyl ether. Mp 135.0 ◦ C. IR (KBr) (υ, cm−1 , stretching), 1660 (C=O), 1458
(C=C), 1111 (C-O); 1 H-NMR δ (ppm, CDCl3), 6.20 (s, 1H), 3.66 (s, 2H), 3.50 (t, J = 5.6 Hz, 2H), 3.45 (s,
3H), 2.59 (t, J = 5.6 Hz, 2H), 2.70-2.60 (m, 8H), 2.29 (s, 3H). Anal. Cal. For. C14 H22 N2 O4 -0.17HOH: C
58.93, H 7.89, N 9.82. Found: C 58.97, H 7.74, N 9.75.
323

Molecular modeling and antimycobacterial studies of Mannich bases:..., B. BERK, et al.,
3-Hydroxy-6-methyl-2-{[4-(propan-2-yl)piperazin-1-yl]methyl}-4H-pyran-4-one (compound 4k)
Recrystallized from ethyl acetate/diethyl ether. Mp 153.2 ◦ C. IR (KBr) (υ, cm−1 , stretching), 1626 (C=O),
1453 (C=C), 1199 (C-O); 1 H-NMR δ (ppm, CDCl3), 6.19 (s, 1H), 3.66 (s, 2H), 2.67 (m, 1H), 2.65-2.55 (m,
8H), 2.28 (s, 3H), 1.04 (d, J = 6.8 Hz, 6H). Anal. Cal. For. C14 H22 N2 O3 : C 63.13, H 8.33, N 10.52. Found:
C 63.26, H 7.91, N 10.46.
Molecular modeling
Homology modeling of M. tuberculosis DNA gyrase subunit B was carried out by Chemical Computing Group
MOE software, version 2009.10, based on the crystal structure of the 43 K ATPase domain of Thermus
˚
thermophilus gyrase B in complex with novobiocin (PDB code 1KIJ, 2.30 A resolution, 43% identity among 420
aligned residues).40
The target sequence was located in the UniProtKB/Swiss-Prot database (access number P0C5C5). The
sequence alignment and homology modeling were performed using MOE-align and MOE-homology model
modules. During the modeling process, the position of crystallized novobiocin and a water molecule, which
was believed to be responsible for key interactions, were kept in their original positions for rebuilding the
binding cavity in the specific and original size. Subsequently, hydrogen ligands were removed from this model
file and subjected to a protein preparation wizard: Maestro for hydrogen adjustment and rotamers, and H-
bond optimization using OPLS 2005 as the energy parameters. Grid files were prepared without restrictions
in the Glide-Grid application of Maestro. Ligand structures enriched with novobiocin were prepared for their
tautomeric and ionized forms at various pH levels using the Lig Prep function of Maestro, followed by Glide
docking using XP mode with the restricted coumarin core of novobiocin. Compound poses with the most
negative e-model scores, which consider hydrophobic parameters extensively, were selected and analyzed visually.
Microbiology
Broth dilution method for Minimum Inhibitory Concentration (MIC)
MICs of compound preparations were determined by broth dilution using Middlebrook 7H10 agar supplemented
with 10% oleic acid–albumin–dextrose–catalase for mycobacteria and tryptic soy broth for other species. The
tested dilutions ranged from 128 to 0.5 μg/mL using dimethyl sulfoxide (DMSO) as the solvent for all com-
pounds. Mycobacteria were suspended in Middlebrook 7H10 broth and the others in tryptic soy broth to match
the turbidity of 1 McFarland standard (2 × 108 cfu/mL). Tube slants were inoculated with undiluted or 1/100
diluted bacterial suspensions and incubated at 37 ◦ C. The slants were examined until visible colonies were ob-
served in the control tube. Controls prepared with concentrations of DMSO used in the dilutions did not show
any inhibitory activity under these circumstances. The MIC value of each isolate was the lowest concentration
of the compound that inhibited visible bacterial growth.
324

Molecular modeling and antimycobacterial studies of Mannich bases:..., B. BERK, et al.,
DNA gyrase supercoiling assay
DNA gyrase supercoiling assays were performed with a Gyrase Supercoiling Assay Kit (Inspiralis) according to
the manufacturer’s instructions and analyzed by monitoring the conversion of relaxed pBR322 plasmid to its
supercoiled form using DNA gel electrophoresis. Essentially, 1 U of E. coli DNA gyrase was first diluted in 5×
gyrase buffer and incubated in an assay buffer (35 mM Tris HCl (pH 7.5), 24 mM KCl, 4 mM MgCl2 , 2 mM
DTT, 1.8 mM spermidine, 1 mM ATP, 6.5% (w/v) glycerol, and 0.1 mg/mL BSA), with 0.5 μg of pBR322
plasmid and a series of synthesized compound dilutions at 37 ◦ C for 30 min. Reactions were stopped with the
addition of stop dye (40% sucrose, 100 mM Tris HCl (pH 7.5), 1 mM EDTA, and 0.5 mg/mL bromophenol
blue) and loaded onto TAE agarose gel (1%). Gels were visualized using a gel documentation system (Bio-Rad
ChemiDoc). Since high levels of DMSO are known to affect DNA gyrase activity, titration was used to determine
the minimum amount of DMSO to be used in the assays, and 5% DMSO (with negligible or no effect on the
gyrase) was chosen to dilute the compounds (data not shown); see Tables 1 and 2.
Results and discussion
The synthesis of key intermediate 5-hydroxy-2-methyl-4H-pyran-4-one (compound 3), available from the exten-
sively studied methyl side chain chlorination of kojic acid with thionyl chloride followed by selective reduction
of the chloro atom in the presence of zinc dust, was undertaken.32−34,41,42
With the process outlined above, compound 3 (allomaltol, in nonsystematic nomenclature) was produced,
without affecting the hydroxyl group at the fifth position of 2-(chloromethyl)-5-hydroxy-4H-pyran-4-one (com-
pound 2), in moderate yields. Finally, the targeted 3-hydroxy-6-methyl-2-((4-substitutedpiperazin-1-yl)methyl)-
4H -pyran-4-one derivatives (compounds 4a-k) shaped the reacting allomaltol with suitable piperazine deriva-
tives in a methanol/formalin medium until formation of the desired precipitates at room temperature was
achieved. In the Mannich reaction, substitution for the second position of the ring system is preferred due
to the presence of a free enolic hydroxyl group in the third position.43 Compounds were crystallized from
appropriate solvents in moderate yields.
By IR, 1 H-NMR, and elemental analysis, 11 compounds were characterized. In the IR spectra, the C=O
stretch band belonging to the fourth position of the pyran ring was generally seen between 1614 and 1660 cm−1
.
C-H stretching belonging to the methyl group at the sixth position of the pyran and piperazine ring system was
observed between 2800 and 2900 cm−1 . In combination with the side chains, the pyran ring’s C=C stretching
bands were observed at 1432-1510 cm−1 and the C-O-C stretching of the pyran system was noticeable between
1108 and 1224 cm−1 . The C≡N stretching in the functional side of compounds 6a and 6b was observed at
2214-2244 cm−1
.
In the 1 H-NMR spectra of the compounds, piperazine ring protons were seen at δ 2.49-3.65 ppm. Side
chain hydrogens and sixth-position methyl protons were seen at δ 1.04-2.84 ppm in consensus with the total
number of hydrogens. Vinylic protons of the pyran ring system were observed as singlets at δ 6.20-6.33 ppm.
Hydroxyl protons of the ring system changed with deuterated solvents and were not observed. The methylene
linkage protons between the pyran and piperazine rings were between δ 3.64 and 4.36 ppm as singlets. The
removal of the singlet around 9-9.5 ppm belonging to the proton at the sixth position of the allomaltol structure
and the formation of these new singlets confirmed the desired 3-hydroxy-6-methyl-2-[(4-substitutedpiperazin-
325

Molecular modeling and antimycobacterial studies of Mannich bases:..., B. BERK, et al.,
1-yl)methyl]-4H-pyran-4-one structures. All other protons were observed according to the expected chemical
shift and integral values.32,33,44
Based on previously published procedures, we screened the synthesized compounds’ inhibitory effects on
supercoiling by performing DNA gyrase supercoiling assays, using gel electrophoresis and E. coli gyrase,which
can convert relaxed pBR322 plasmids to their supercoiled topoisomer.45,46 In these procedures, DNA gyrase
activity can by monitored as 2 or more bands on an agarose gel where the upper band is an open-circular (nicked)
plasmid and the faster running band is a negatively supercoiled (closed-circular) plasmid. The inhibition of
gyrase supercoiling was visually assessed by a 50% reduction of the supercoiled band and the appearance of a
spread of slower migrating plasmids. In our previous studies, we reported the inhibitory effects of compounds
5a-6s.37,38 Here we report the results for the inhibition of gyrase supercoiling assay of newly synthesized
compounds 4a-k (Figure 2). All screened piperidine and piperazine derivatives of 3-hydroxy-6-methyl-4H-
pyran-4-one derivatives, including those newly synthesized, revealed mild to moderate inhibitor activity on E.
coli DNA gyrase supercoiling.
Sequence similarities between DNA gyrase B subunits of E. coli and M. tuberculosis (UniProtKB/Swiss-
Prot database accession numbers P0AES7 and P0C5C5, respectively) suggested a 48% overall score after
alignment with ClustalW2.47 The combination of DNA gyrase supercoiling assay results and this finding
provoked us to investigate the inhibition potencies of these compounds in vitro.
Minimum inhibitory concentrations values for the 52 compounds were assessed with M. tuberculosis
H37RV, using a gradient-based decreasing concentration technique by the broth dilution method. The lowest
MIC (4 μg/mL) was attributed to compound 5b, structured as 2-((4-(2-chlorophenyl)piperazin-1-yl)methyl)-3-
hydroxy-6-methyl-4H -pyran-4-one. For other compounds, the MIC results were also promising: between 8 and
128 μg/mL weighted at a 16 μg/mL concentration.
4-Substituted-piperazinyl derivatives (5a-v and 4a-k) varied in their antimycobacterial effects with
diverse substitutions. 4-Phenyl (5a), 4-(N-methylpiperidin-4-yl) (5s), and 4-tert-butoxycarbonyl (5t) substi-
tutions of the piperazine core resulted in MIC values of 8 μg/mL. Various 2-, 3-, 4-, and 5-phenyl side chain
substitutions revealed decreasing potency, whereas, on the contrary, the 4-(2-chlorophenyl) derivative 5b acted
as an inhibitor. Placing a linker of 2 or 3 carbons between the piperazine core and the phenyl group (4d-f) led to
a decline in inhibitor activity. Interestingly, the largest decrease in potency was observed with 4-(2-ethoxyethyl)
(4g) and 4-(2-methoxymethyl) (4j) substituted piperazine derivatives.
For substituted piperidine derivatives 6a-s, 2- or 6-positioned side chain substitutions mainly reduced
the inhibitory activity against M. tuberculosis, while 3- and 5-positioned side chain substitutions revealed
higher activity than a nonsubstituted piperidine ring. When compared to 3- and 5-substituted piperidines,
substitution from the fourth position of the piperidine ring revealed a potency of 8 μg/mL, similar to 4-
ethoxycarbonylpiperidine derivative 6s.
To examine the possible interactions between ligands and the active site of the Mycobacterium DNA
gyrase B subunit, homology modeling and a docking process were undertaken.
For homology modeling of the DNA gyrase subunit B of M. tuberculosis, the crystal structure of the
gyrase B 43 K ATPase domain complex with the potent inhibitor novobiocin (PDB code 1KIJ)23 was selected
as the template. This “open” conformation of the active site in the absence of ATP is unique, as it clearly
demonstrates large conformational changes during inhibition processes.
326

Molecular modeling and antimycobacterial studies of Mannich bases:..., B. BERK, et al.,
This crystal structure has a special water molecule (HOH 539) trapped in the hydrophobic cavity that
shows a clear H-bonding network with Asp 72 (a key amino acid responsible for the activity), Gly 76, Thr
166, and the ligand itself. We selected this template to preserve this particular H-bonding network of amino
acid-water molecule-ligand within the modeling process.17 Even if there were other better E-valued crystal
structures according to the compositional matrix adjusted alignment of BLAST-p, we would still have selected
this particular template.
During homology modeling, after the heavy atoms were modeled and all hydrogen atoms were added,
the protein coordinates were minimized with MOE using the AMBER94 force field.48 Protein stereochemistry
evaluation was performed using several tools (Ramachandran and Chi plots to measure phi/psi and chi1/chi2
angles and clash contact reports) implemented in the MOE suite.49 The pair-wise percentage residue identity
was determined as 46.4 between 2 chains, where the pair-wise RMSD values for Cα and the main chain backbone
˚
atoms of the superimposed model and template were 0.64 and 0.70 A, respectively.
Briefly, the structure comprised a compact single domain with an 8-stranded beta sheet, which was
flanked on 1 side by 3 alpha-helices and random coils. By keeping the conserved water molecule and novobiocin
during the modeling process, we not only preserved the appropriate distances and hydrogen networking between
the key amino acids of the active site, but also tried to steadily preserve the binding mode of novobiocin. The
˚
RMSD value difference of 0.84 A of the pose from the nonrestricted redocking of the novobiocin structure itself
also confirmed the approach.
The coiled “open” protein structure gave the active site a considerable amount of flexibility, which could
also affect the binding mode of the ligands. Considering the structural resemblance between 3-hydroxy-4(1H)-
pyranone and the coumarin ring system, we performed the docking core under restriction. The docking poses
belonging to ligands suggested that Arg 82 stabilized the sixth-position methyl group over the pyranone ring
system with hydrophobic interactions, whereas bridge methyl, piperidine, and piperazine carbons were stabilized
between Phe 109 and Ile 84 with the same kind of interactions. The substituted phenyl structures were generally
positioned in the hydrophobic space surrounded by Val 99, Val 125, Val 123, Ile 84, and Asn 52. If there is
a hydrogen donating group in the para position of phenyl, then there is a chance of a hydrogen bond existing
with that group and the Val 123 carbonyl. The small groups, such as hydroxymethyl, in the third position of
the piperidine ring have a tendency to form a hydrogen bond with the carbon Asp 52 backbone carbonyl and
the conserved water molecule (Figure 3).
Although the generated docking poses illustrated a parallelism between MIC values against M. tuberculosis
and compound interactions with the surrounding environment regarding side chain changes in general, further
molecular dynamics simulations for this computationally estimated crystal structure of the complex, under
physiological conditions, would be the best option for verification.
Both piperidine and piperazine derivatives of 3-hydroxy-6-methyl-4H-pyran-4-one pharmacophore are
promising compounds for MDR-TB; their high efficiency toward particular M. tuberculosis strains is comparable
to standard novobiocin.50 An alternative to finding a core other than 5-hydroxy-2-methyl-4H-pyran-4-one
would be to perform high throughput virtual screening methods using a homology model of M. tuberculosis
DNA gyrase B together with side-chain interaction field analysis over existing piperidine and piperazine side
chains to evaluate more clinically applicable compounds.
327

Molecular modeling and antimycobacterial studies of Mannich bases:..., B. BERK, et al.,
Figure 3. Docked coordinates of compound 5b and novobiocin in the active site (circled in the upper middle) of the
homology model of M. tuberculosis DNA gyrase B. The circle on the left shows compound 5 (yellow) superposed with
novobiocin (green). The circle on the right shows hydrophobic (green dotted lines) and electrostatic (blue dotted lines)
interactions between compound 5 and the amino acid residues of the binding site.
Acknowledgement
We thank Shirley McCartney, Ph.D., for proofreading and discussion.
References
1. Tomioka, H.; Namba, K. Kekkaku 2006, 81, 753-774.
2. Cole, S.; Brosch, R.; Parkhill, J.; Garnier, T.; Churcher, C.; Harris, D.; et al. Nature 1998, 393, 537-544.
3. Champoux, J. J. Annu. Rev. Biochem. 2001, 70, 369-413.
4. Ferrero, L.; Cameron, B.; Manse, B.; Lagneaux, D.; Crouzet, J.; Famechon, A.; Blanche, F. Mol. Microbiol. 1994,
13, 641-653.
5. Maxwell, A. Trends Microbiol. 1997, 5, 102-109.
6. Drlica, K.; Malik, M. Curr. Top. Med. Chem. 2003, 3, 249-282.
7. Mitscher, L. A. Chem. Rev. 2005, 105, 559-592.
8. Sifaoui, F.; Lamour, V.; Varon, E.; Moras, D.; Gutmann, L. J. Bacteriol. 2003, 185, 6137-6146.
9. Stahlmann, R. Toxicol. Lett. 2002, 127, 269-277.
328

Molecular modeling and antimycobacterial studies of Mannich bases:..., B. BERK, et al.,
10. Leone, R.; Venegoni, M.; Motola, D.; Moretti, U.; Piazzetta, V.; Cocci, A.; Resi, D.; Mozzo, F.; Velo, G.; Burzilleri,
L.; Montanaro, N.; Conforti, A. Drug Saf. 2003, 26, 109-120.
11. Oh, Y. R.; Carr-Lopez, S. M.; Probasco, J. M.; Crawley, P. G. Ann. Pharmacother. 2003, 37, 1010-1013.
12. Piddock, L. J. V. Drugs 1999, 58, 11-18.
13. Maxwell, A.; Lawson, D. M. Curr. Top. Med. Chem. 2003, 3, 283-303.
14. Boehm, H.; Boehringer, M.; Bur, D.; Gmuender, H.; Huber, W.; Klaus, W.; Kostrewa, D.; Kuehne, H.; Luebbers,
T.; Meunier-Keller, N.; Mueller, F. J. Med. Chem. 2000, 43, 2664-2674.
15. Poyser, J. P.; Telford, B.; Timms, D.; Block, M. H.; Hales, N. J. WO Patent 99/01442, 1999.
16. Wigley, D.; Davies, G.; Dodson, E.; Maxwell, A.; Dodson, G. Nature 1991, 351, 624-629.
17. Lewis, R.; Singh, O.; Smith, C.; Skarzynski, T.; Maxwell, A.; Wonacott, A.; Wigley, D. EMBO J. 1996, 15,
1412-1420.
18. Holdgate, G.; Tunnicliffe, A.; Ward, W.; Weston, S.; Rosenbrock, G.; Barth, P.; Taylor, I.; Pauptit, R.; Timms, D.
Biochemistry 1997, 36, 9663-9673.
19. Tsai, F.; Singh, O.; Skarzynski, T.; Wonacott, A.; Weston, S.; Tucker, A.; Pauptit, R.; Breeze, A.; Poyser, J.;
O’Brien, R.; Ladbury, J.; Wigley, D. Proteins 1997, 28, 41-52.
20. Brino, L.; Urzhumtsev, A.; Mousli, M.; Bronner, C.; Mitschler, A.; Oudet, P.; Moras, D. J. Biol. Chem. 2000, 275,
9468-9475.
21. Ward, W.; Holdgate, G. Prog. Med. Chem. 2001, 38, 309-376.
22. Lafitte, D.; Lamour, V.; Tsvetkov, P. O.; Makarov, A. A.; Klich, M.; Deprez, P.; Moras, D.; Briand, C.; Gilli, R.
Biochemistry 2002, 41, 7217-7223.
23. Lamour, V.; Hoermann, L.; Jeltsch, J.; Oudet, P.; Moras, D. J. Biol. Chem. 2002, 277, 18947-18953.
24. Bellon, S.; Parsons, J.; Wei, Y.; Hayakawa, K.; Swenson, L.; Charifson, P.; Lippke, J.; Aldape, R.; Gross, C.
Antimicrob. Agents Chemother. 2004, 48, 1856-1864.
25. Kayahara, H.; Shibata, N.; Tadasa, K.; Maeda, H.; Kotani, T.; Ichimoto, I. Agric. Biol. Chem. 1990, 54, 2441-2442.
26. Kotani, T.; Ichimoto, I.; Tatsumi, C.; Fujita, T. Agric. Biol. Chem. 1975, 39, 1311-1317.
¨
27. Aytemir, M. D.; Hider, R. C.; Erol, D. D.; Ozalp, M.; Ekizo˘glu, M. Turk. J. Chem. 2003, 27, 445-452.
¨
28. Aytemir, M. D.; Erol, D. D.; Hider, R. C.; Ozalp, M. Turk. J. Chem. 2003, 27, 757-764.
29. Wan, H. M.; Chen, C. C.; Giridhar, R.; Chang, T. S.; Wu, W. T. J. Ind. Microbiol. Biotechnol. 2005, 32, 227-233.
30. Yamato, M.; Yasumoto, Y.; Sakai, J.; Luduena, R. F.; Baneerjee, A.; Tashiro, T. J. Med. Chem. 1987, 30, 1897-
1900.
31. Tamura, T.; Mitsumori, K.; Totsuka, Y.; Wakabayashi, K.; Kido, R.; Kasai, H.; Nasu, M.; Hirose, M. Toxicology
2006, 222, 213-224.
¨
32. Aytemir, M. D., C¸alı¸s, U.; Ozalp, M. Arch. Pharm. Pharm. Med. Chem. 2004, 337, 281-288.
¨
¨
33. Aytemir, M. D.; C¸alı¸s, U. Hacettepe University Journal of the Faculty of Pharmacy 2007, 27, 1-10.
¨
34. Aytemir, M. D.; C¸alı¸s, U. FABAD J. Pharm. Sci. 2006, 31, 23-29.
35. Cabanes, J.; Chazarra, S.; Garcia-Carmona, F. J. Pharm. Pharmacol. 1994, 46, 982-985.
36. Kim, H.; Choi, J.; Cho, J. K.; Kim, S. Y.; Lee, Y. S. Bioorg. Med. Chem. Lett. 2004, 14, 2843-2846.
329

Molecular modeling and antimycobacterial studies of Mannich bases:..., B. BERK, et al.,
¨
37. Us, D.; Gu¨rdal, E.; Berk, B.; Oktem, S.; Kocag¨oz, T.; C¸ag˘layan, B.; Kurnaz, I. A.; Erol, D. D. Turk. J. Chem.
2009, 33, 803812.
38. Us, D.; Berk, B.; Gu¨rdal, E.; Aytekin, N.; Kocag¨oz, T.; C¸ag˘layan, B.; Kurnaz, I. A.; Erol, D. D. Turk. J. Chem.
2010, 34, 447-456.
˙
39. Iskeleli, N. O.; Isıka, S.; Aytemir, M. D. Acta Crystallogr. Sect. E 2005, 61, o1947o1949.
40. Lamour, V.; Hoermann, L.; Jeltsch, J.; Oudet, P.; Moras, D. J. Biol. Chem. 2002, 277, 18947-18953.
41. Erol, D. D.; Yulu˘g, N. Eur. J. Med. Chem. 1994 29 893-897.
42. Yabuta, T. J. Chem. Soc. 1924, 125, 575-587.
43. O’Brien, G.; Patterson, J. M.; Meadow, J. R. J. Org. Chem. 1960, 25, 86-89.
44. Liu, Z. D.; Piyamongkol, S.; Liu, D. Y.; Khodr, H. H.; Lu, S. H.; Hider, R. C. Bioorg. Med. Chem. 2001, 9, 563-573.
45. Maxwell, A.; Burton, N. P.; O’Hagan, N. Nucleic Acids Res. 2006, 34, e104.
46. Reece, R. J.; Maxwell A. J. Biol. Chem. 1989, 264, 19648-19653.
47. Larkin M. A.; Blackshields, G.; Brown, N. P.; Chenna, R.; McGettigan, P. A.; McWilliam, H.; Valentin, F.; Wallace,
I. M.; Wilm, A.; Lopez, R.; Thompson, J. D.; Gibson, T. J.; Higgins, D. G. Bioinformatics 2007, 23, 2947-2948.
48. Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; Merz, K. M. Jr.; Ferguson, D. M.; Spellmeyer, D. C.; Fox,
T.; Caldwell, J. W.; Kollman, P. A. J. Am. Chem. Soc. 1995, 117, 5179-5197.
49. MOE (The Molecular Operating Environment) Version 2009.10, software available from Chemical Computing
Group Inc., 1010 Sherbrooke Street West, Suite 910, Montreal, Canada H3A 2R7. http://www.chemcomp.com.
50. Olitzki, A. L.; Godinger, D.; Israeli, M.; Honigman A. Appl Microbiol. 1967, 15, 994-1001.
330