Synthesis, antimycobacterial evaluation and pharmacophore modeling of analogues of the natural product formononetin

Article abstract of DOI:10.1016/j.bmcl.2015.04.064

Abstract The synthesis and antimycobacterial activity of formononetin analogues is hereby reported. Formononetin and its analogue 11E showed 88% and 95% growth inhibition, respectively, against the H37Rv strain of Mycobacterium tuberculosis. Pharmacophore modeling studies indicated that the presence of a hydroxyl group in formononetin and its analogues, is crucial for maintaining activity.

Full text of DOI:10.1016/j.bmcl.2015.04.064

Accepted Manuscript
Synthesis, antimycobacterial evaluation and pharmacophore modeling of ana-
logues of the natural product formononetin
Peggoty Mutai, Elumalai Pavadai, Ian Wiid, Andile Ngwane, Bienyameen
Baker, Kelly Chibale
PII:
S0960-894X(15)00396-0
DOI:
http://dx.doi.org/10.1016/j.bmcl.2015.04.064
BMCL 22647
Reference:
Bioorganic & Medicinal Chemistry Letters
To appear in:
Received Date:
Revised Date:
Accepted Date:
24 February 2015
17 April 2015
20 April 2015
Please cite this article as: Mutai, P., Pavadai, E., Wiid, I., Ngwane, A., Baker, B., Chibale, K., Synthesis,
antimycobacterial evaluation and pharmacophore modeling of analogues of the natural product formononetin,
Bioorganic & Medicinal Chemistry Letters (2015), doi: http://dx.doi.org/10.1016/j.bmcl.2015.04.064
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers
we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and
review of the resulting proof before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Synthesis, antimycobacterial evaluation and pharmacophore modeling of
analogues of the natural product formononetin
Peggoty Mutaia, Elumalai Pavadaia, Ian Wiidb, Andile Ngwaneb, Bienyameen Bakerb, Kelly
Chibalea,c,d*
aDepartment of Chemistry, University of Cape Town, Rondebosch 7701, South Africa
bDST-NRF Centre of Excellence for Biomedical Tuberculosis Research, SAMRC Centre for
Tuberculosis Research, Division of Molecular Biology and Human Genetics, Faculty of Medicine and
Health Sciences, Stellenbosch University, PO Box 19063, Tygerberg 7505, South Africa
cSouth African Medical Research Council Drug Discovery and Development Research Unit,
University of Cape Town, Rondebosch 7701, South Africa
dInstitute of Infectious Disease and Molecular Medicine, University of Cape Town, Rondebosch 7701,
South Africa
*Corresponding author
Email address: Kelly.Chibale@uct.ac.za (K. Chibale)
Abstract
The synthesis and antimycobacterial activity of formononetin analogues is hereby reported.
Formononetin and its analogue 11E showed 88% and 95% growth inhibition, respectively,
against the H37Rv strain of Mycobacterium tuberculosis. Pharmacophore modeling studies
indicated that the presence of a hydroxyl group in formononetin and its analogues, is crucial
for maintaining activity.
Keywords Antimycobacterial, Formononetin, Pharmacophore, Structure, Activity
Tuberculosis (TB) is a leading cause of mortality and morbidity, being estimated to infect
about one third of the world’s population 1. The Global tuberculosis report by WHO in 2013
reported that there were an estimated 8.6 million incidents of TB in 2012 and 1.3 million TB
related deaths 2.The HIV scourge has made the TB situation worse, especially in sub-Saharan
Africa, with the co-epidemic being particularly concentrated in the Southern African
countries where HIV prevalence is high 3. Increasing rates of drug-resistant tuberculosis are a
significant concern and pose serious implications for current and future treatment of the
disease4. Resistance has been reported for new drugs such as bedaquiline5, indicating that
there is an urgent need for more new drug candidates to raise the probability of developing a
novel, short-course and safe ‘universal’ regimen applicable to drug-susceptible and all forms
of drug-resistant TB6.

Natural products have played a prominent role in the management of tuberculosis. A number
of phytochemicals have been found to possess antimycobacterial activity while many of the
currently used drugs in the treatment of TB such as streptomycin and rifampicin are derived
from microbial natural sources 7,8. Formononetin is an isoflavone that is widespread in the
Leguminoseae family. It has previously been reported to possess various pharmacological
activities such as the inhibition of proliferation of prostate, breast and colon cancer cells,
wound and bone fracture healing, antihypertensive activity as well as anti-infective activities
especially against giardiasis where it is reported to be more potent than metronidazole, the
drug of choice in the treatment of giardiasis9–12. In addition, formononetin has been reported
to have strong multi-drug resistance reversal effects by inhibiting the P-gp efflux pump that is
13
responsible for drug resistance . This is of particular interest in tuberculosis as drug efflux
has been highlighted as an important mechanism of drug resistance by Mycobacterium
tuberculosis14.
This paper reports the antimycobacterial activity of formononetin and derives a tentative
structure activity relationship of its analogues based on the percent inhibition of the H37Rv
strain of Mycobacterium tuberculosis. Further, pharmacophore modeling studies on the
formononetin analogues were performed to elucidate the structural determinant responsible
for the biological activity.
Synthesis of formononetin analogues was carried out by modification of rings A and B as
well as the benzopyranone linker. Modifications on ring A and ring B were achieved through
similar synthetic schemes. For ring A modifications, the starting materials had varied
substituents at either position 5 or 6 (compounds 1-8) while for ring B modifications, the
scheme started with protection of the para-hydroxyl group on ring A (compounds 9-14)
(Scheme 1). Compounds 15 and 16 were synthesized using scheme 1 but varying both R1 and
R2.

Scheme 1: Reagents and conditions: (i) DMF-DMA, 95°C, 3h (ii) I , pyridine, CHCl , rt, 12h, (iii)
2
3
ArB(OH) , Pd/C, Na CO , DME/H2O, 45°C, 4h (iv) DHP, PPTS, DCM, rt, 4h (v) p-TsOH, MeOH,
2
2
3
THF, 60°C, 2h
Modifications on the benzopyrone linker were achieved through various synthetic schemes.
(Scheme 2)

Scheme 2: Reagents and conditions: (i)DHP, PPTS, DCM, rt, 4h; (ii) EtOH, aq. K2CO3, rt, 30 min;
(iii) p-methoxybenzaldehyde, 80°C reflux, 8h (iv) p-TsOH, MeOH, THF, 60°C, 2h (v) Lawesson’s
reagent, Toluene, reflux, 110°C, 1h; DMF, 10h (vi) N2H4.H2O; EtOH, reflux, 60-80°C, 2h (vii)
Hydroxylamine hydrochloride, N-methylmorpholine, EtOH, Reflux, 80°C, 8h
Twenty analogues of formononetin were synthesized and tested for antimycobacterial activity
against Mycobacterium tuberculosis H37Rv (ATCC 27294) using the Microtiter Alamar Blue
assay (MABA) as previously described 15 (Table 1& 2).
Table 1: Antimycobacterial activity of formononetin analogues with modifications on rings A
and B

Compound
R1
7-OH
7-H
R2
4’-OCH3
4’-OCH3
4’-OCH3
4’-OCH3
4’-OCH3
4’-OCH3
4’-OCH3
4’-OCH3
4’-OCH3
4-CN
X
C
C
C
C
C
C
C
C
C
C
C
C
N
C
C
C
N
% inhibition at 10µM
88
0
Formononetin
1C
2C
3C
4C
5C
6C
7C
8C
9E
6-F
0
7-F
0
6-Cl
0
6-OCH3
6-Br
0
5
6-CH3
6-OH
7-OH
7-OH
7-OH
7-OH
7-OH
7-OH
7-H
7
28
16
54
95
0
4-Cl
10E
11E
12E
13E
14E
15
4-C(CH3)3
4’-OCH3
3’-F, 4’-OCH3
3’-OCH3, 4’-OCH3
4’-H
13
14
0
7-F
4’-OCH3
2
16
Table 2: Antimycobacterial activity of formononetin analogues:
Compound
Structure
% inhibition at 10µM
58
17

4
18
19
40
15
20
When all compounds were tested at a concentration of 10µM, the parent compound,
formononetin, exhibited 88% inhibition of the H37Rv strain of Mycobacterium tuberculosis.
All modifications on ring A led to a decrease or loss of activity. When the hydroxyl group
was moved from position 7 to position 6 (8C), there was a marked drop in activity from 88%
to 28% inhibition. When the OH was replaced by other groups such as F, Cl, Br, CH3, OCH3,
activity was lost. Replacing the OCH3 group in ring B of formononetin with a tert-butyl
group (11E) improved activity from 88% to 95% inhibition. When the hydrophobic electron
withdrawing chloro substituent was introduced (10E) activity dropped to 54% inhibition. A
change to hydrophilic groups such as CN (9E) drastically reduced activity to 16%. The
unsubstituted isoflavone skeleton (15) was devoid of antimycobacterial activity at 10µM.
Compound 16, which is a hybrid of compound 3C and 12E, did not display any
antimycobacterial activity, just like 3C and 12E. When the benzopyrone linker in
formononetin was replaced by an open chain linker, activity dropped from 88% inhibition to
58% inhibition. When the benzopyrone was replaced by an unfused heterocyclic ring, there
was a further drop in activity as is noted with the pyrazole linked compound (19) and the
isoxazole linked compound (20), which displayed 40% and 15% inhibition, respectively.
In order to elucidate the structural determinant(s) responsible for the biological activity of
formononetin analogues, pharmacophore modeling studies were performed on the 21
formononetin analogues using Catalyst/HipHop of Discovery Studio 4.0. (see supplementary
information). The HipHop run resulted in 10 ranked five-feature models, which are presented
in Supplementary Table S1. It was observed that all models have the same pharmacophore
features. However, the spatial locations and statistical parameters are different. A top-ranked
model selected as the best pharmacophore model on the basis of fit values of compounds,

which contains five features; two hydrogen-bond acceptors, a hydrogen-bond-donor, an
aromatic ring and a hydrophobic feature, as shown in Fig 1 (A). The model fits well on the
most active compound 11E and its fit value of 13.95 as expected (Table 3). The mapping on
11E shows that the two hydrogen-bond acceptors located on the oxygen atom and keto group
of the furan ring on the chromone moiety, one hydrogen bond donor located on the hydroxyl
group of chromone moiety, an aromatic feature located on the benzyl group as well as a
hydrophobic feature corresponding to the tertiary butyl substituent, as shown in Fig 1 (B). In
addition, the mapping of the model was performed on inactive compounds, which suggested
that the hydrogen-bond donor feature is missing in most of the inactive compounds,
indicating that the hydrogen-bond donor feature is important for biological activity.
Additionally, the pharmacophore fit values of inactive compounds are significantly reduced
in the absence of the hydrogen-bond donor feature (see Table 3). The mapping of one of the
inactive compounds 3C, with a fit value of 3.91, is shown in Fig 1 (C). The fit values for the
21 compounds obtained by the pharmacophore model are presented in Table 3. The
pharmacophore model distinguishes all active compounds from inactive ones with high fit
values. The model successfully predicted all 4 active compounds (100 % success, above the
fit value cut off of 13 for above the antibacterial activity of 50 %) whereas 13 out of the 17
inactive compounds are predicted by values below this threshold (76% success, below the fit
value cut off of 13 for below the antibacterial activity of 50 %).
Fig 1. Pharmacophore model and its mapping with active and inactive compounds. (A) Common feature
pharmacophore model generated by HipHop algorithm. Pharmacophore features are color-coded with green

spheres for hydrogen-bond acceptor (HBA), purple sphere for hydrogen-bond donor (HBD), orange sphere for
aromatic ring (RA) and cyan sphere for hydrophobic (HYD). (B) The model mapping with the most active
compound 11E and its pharmacophore fit value of 13.95. (C) Mapping of the model with one of the inactive
compounds 3C and its pharmacophore fit value of 3.91. (D) Mapping of the model with the parent compound
formononetin and its fit value of 14. (E) Mapping of the model with 1C and its fit value of 3.97. The missed
pharmacophore HBD feature is indicated by an arrow in (C) and (E).
Importantly, the pharmacophore model is capable of explaining the activity difference
between formononetin (active) and its most inactive analogues, for example 1C. These two
compounds differ by a single functional group, hydroxyl (-OH), in formononetin that maps
into the hydrogen-bond donor pharmacophore feature [Fig 1 (D)]. This feature is missing in
1C owing to absence of the hydroxyl group [Fig 1 (E)]. As a result, there is a big difference
in these fit values of the two compounds and further signifies that the hydrogen-bond donor
feature is responsible for activity differences between formononetin and 1C.
Table 3. Qualitative activity data (active/inactive) and fit values based on the best
pharmacophore model for all 21 formononetin analogues.
Compound
Name
Experimental
Activity (%)
95
Experiment
Estimated
Scale
+
Pharmacophore Fit
value
Scale
+
11E
13.95
Formononetin
88
58
54
40
28
16
15
14
13
7
+
+
+
-
+
+
+
-
14.00
13.35
13.62
10.69
12.79
12.96
10.74
13.87
13.75
3.97
17
10E
19
8C
9E
-
-
-
-
20
-
-
14E
13E
7C
6C
16
-
+
+
-
-
-
4
-
+
-
13.14
3.97
5
-
18
2
-
-
3.99
12E
5C
0
-
+
-
13.89
4.00
0
-

1C
4C
2C
3C
15
0
0
0
0
0
-
-
-
-
-
-
-
-
-
-
3.97
3.97
3.97
3.91
1.05
aExperimental activity scale: +, % of activity > 50 (active); - , % of activity < 50 (inactive).
bEstimated scale based on pharmacophore fit value: +, fit value > 13 (active); -, fit value < 13 (inactive).
cFit value indicates how well the features in the pharmacophore overlap the chemical features in the molecule
and was calculated by the equation (1).
In conclusion, formononetin analogues were synthesized and evaluated for their
antimycobacterial activity against M. tuberculosis. This is the first report of the
antimycobacterial activity of formononetin. All modifications on formononetin led to
decreased activity, apart from the ring B modification with a tert-butyl substituent, which
improved activity from 88% to 95% inhibition. Pharmacophore modeling analysis on
formononetin analogues showed that the presence of a hydroxyl group in formononetin is
essential for high potency. The model will be used as a guide to direct future synthetic efforts
aimed at developing more potent formononetin-based antimycobacterial compounds.
Acknowledgements
P.M. is grateful to the USHEPiA Programme, South Africa for funding this work. The
University of Cape Town, South African Medical Research Council, and South African
Research Chairs initiative of the Department of Science and Technology administered
through the South African National Research Foundation are gratefully acknowledged for
support (KC). The authors would like to acknowledge the Centre of High Performance
Computing, Cape Town (http://www.chpc.ac.za) for Discovery Studio 4.0 license.
References
1.
Koul, A.; Arnoult, E.; Lounis, N.; Guillemont, J.; Andries, K. Nature 2011, 469, 483–
90.
2.
3.
4.
WHO | Global tuberculosis report 2013.
Gray, J. M.; Cohn, D. L. Semin. Respir. Crit. Care Med. 2013, 34, 32–43.
Chan, B.; Khadem, T.; Brown, J. Am. J. Heal. Pharm. 2013, 70, 1984–94.

5.
6.
7.
8.
9.
Andries, K.; Villellas, C.; Coeck, N.; Thys, K.; Gevers, T.; Vranckx, L.; Lounis, N.;
C.de.Jong, B.; Koul, A. PLoS One 2014, 9, e102135.
Zumla, A.; Hafner, R.; Lienhardt, C.; Hoelscher, M.; Nunn, A. Nat. Rev. Drug Discov.
2012, 11, 171–2.
Ohnishi, Y.; Ishikawa, J.; Hara, H.; Suzuki, H.; Ikenoya, M.; Ikeda, H.; Yamashita, A.;
Hattori, M.; Horinouchi, S. J. Bacteriol. 2008, 190, 4050–60.
Tupin, A.; Gualtieri, M.; Roquet-Banères, F.; Morichaud, Z.; Brodolin, K.; Leonetti,
J.-P. Int. J. Antimicrob. Agents 2010, 35, 519–23.
Ye, Y.; Hou, R.; Chen, J.; Mo, L.; Zhang, J.; Huang, Y.; Mo, Z. Horm. Metab. Res.
2012, 44, 263–7.
10. Chen, J.; Zeng, J.; Xin, M.; Huang, W.; Chen, X. Horm. Metab. Res. 2011, 43, 681–6.
11. Auyeung, K. K.-W.; Ko, J. K.-S. Invest. New Drugs 2010, 28, 1–13.
12. Khan, I. A.; Avery, M. A.; Burandt, C. L.; Goins, D. K.; Mikell, J. R.; Nash, T. E.;
Azadegan, A.; Walker, L. A. J. Nat. Prod. 2000, 63, 1414–6.
13. Gyémánt, N.; Tanaka, M.; Antus, S.; Hohmann, J.; Csuka, O.; Mándoky, L.; Molnár, J.
In Vivo 2005, 19, 367–74.
14. Gupta, S.; Cohen, K. A.; Winglee, K.; Maiga, M.; Diarra, B.; Bishai, R.. Antimicrob.
Agents Chemother. 2014, 58, 574–576..
15. Collins, L. A.; Franzblau, S. G. Antimicrob. Agents Chemother. 1997, 41, 1004–1009.