Discovery, synthesis and in combo studies of a tetrazole analogue of clofibric acid as a potent hypoglycemic agent

Article abstract of DOI:10.1016/j.bmcl.2013.03.122

A tetrazole isosteric analogue of clofibric acid (1) was prepared using a short synthetic route and was characterized by elemental analysis, NMR ( ¹H, ¹³C) spectroscopy, and single-crystal X-ray diffraction. The in vitro inhibitory activity of 1 against 11β- hydroxysteroid dehydrogenase type 1 (11β-HSD1) was evaluated, showing a moderate inhibitory enzyme activity (51.17% of inhibition at 10 μM), being more active than clofibrate and clofibric acid. The antidiabetic activity of compound 1 was determined at 50 mg/Kg single dose using a non insulin dependent diabetes mellitus rat model. The results indicated a significant decrease of plasma glucose levels, during the 7 h post-administration. Additionally, we performed a molecular docking of 1 into the ligand binding pocket of one subunit of human 11β-HSD1. In this model, compound 1 binds into the catalytic site of 11β-HSD1 in two different orientations. Both of them, show important short contacts with the catalytic residues Ser 170, Tyr 183, Asp 259 and also with the nicotinamide ring of NADP⁺.

Full text of DOI:10.1016/j.bmcl.2013.03.122

Bioorganic & Medicinal Chemistry Letters xxx (2013) xxx–xxx
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Discovery, synthesis and in combo studies of a tetrazole analogue
of clofibric acid as a potent hypoglycemic agent
a
a
Gabriel Navarrete-Vázquez a, , Alfredo Alaniz-Palacios , Sergio Hidalgo-Figueroa ,
Cristina González-Acevedo a, Gabriela Ávila-Villarreal a, Samuel Estrada-Soto a, Scott P. Webster b,
José L. Medina-Franco c, Fabian López-Vallejo c, Jorge Guerrero-Álvarez d, Hugo Tlahuext d
⇑
a Facultad de Farmacia, Universidad Autónoma del Estado de Morelos, Cuernavaca, Morelos 62209, Mexico
b Endocrinology Unit, Centre for Cardiovascular Science, The Queen’s Medical Research Institute, University of Edinburgh, Edinburgh EH16 4TJ, UK
c Torrey Pines Institute for Molecular Studies, Port St. Lucie, FL 34987, USA
d Centro de Investigaciones Químicas, Universidad Autónoma del Estado de Morelos, Cuernavaca, Morelos 62209, Mexico
a r t i c l e i n f o
a b s t r a c t
Article history:
A tetrazole isosteric analogue of clofibric acid (1) was prepared using a short synthetic route and was
characterized by elemental analysis, NMR (1H, 13C) spectroscopy, and single-crystal X-ray diffraction.
The in vitro inhibitory activity of 1 against 11b-hydroxysteroid dehydrogenase type 1 (11b-HSD1) was
Received 13 February 2013
Revised 22 March 2013
Accepted 27 March 2013
Available online xxxx
evaluated, showing a moderate inhibitory enzyme activity (51.17% of inhibition at 10 lM), being more
active than clofibrate and clofibric acid. The antidiabetic activity of compound 1 was determined at
50 mg/Kg single dose using a non insulin dependent diabetes mellitus rat model. The results indicated
a significant decrease of plasma glucose levels, during the 7 h post-administration. Additionally, we per-
formed a molecular docking of 1 into the ligand binding pocket of one subunit of human 11b-HSD1. In
this model, compound 1 binds into the catalytic site of 11b-HSD1 in two different orientations. Both of
them, show important short contacts with the catalytic residues Ser 170, Tyr 183, Asp 259 and also with
the nicotinamide ring of NADP+.
Keywords:
Diabetes
11b-HSD1
Fibrates
Molecular docking
Ó 2013 Elsevier Ltd. All rights reserved.
Fibrates, such as bezafibrate, clofibrate and fenofibrate are used
as therapeutic agents in the treatment of dyslipidemia, heart dis-
ease and diabetic complications in humans, due to the fact that
glucose concentration.8 11b-HSD1 catalyzes the conversion of
inactive cortisone into the active hormone cortisol (in rodents
11-dehydrocorticosterone to corticosterone). 11b-HSD1 inhibitors
are therefore of considerable interest as potential treatments for
a number of diseases including type 2 diabetes, obesity and meta-
bolic syndrome.9
The majority of fibrates are prodrugs, which are extensively
metabolized by hydrolysis, leading to the free carboxylic acid form.
Medicinal chemists have frequently faced the problem of develop-
ing surrogates for carboxylic acid groups. Planar acidic heterocy-
cles, such as tetrazole, are commonly used as carboxylic acid
bioisosteres.10 This replacement improves the bioavailability, in-
creases the potency and improves chemical stability.11
In our ongoing research on fibrate derivatives with antihyper-
lipidemic and antidiabetic activities, we have synthesized in this
work the compound 2-(4-chlorophenoxy)-2-methyl-N-(1H-tetra-
zol-5-yl) propanamide (1), which is an isostere of clofibric acid,
with a tetrazolamide group instead of carboxylic acid (Fig. 1).
The design of the compound 1 was based on the bioisosteric
replacement of the carboxylic acid moiety by a tetrazole ring. In or-
der to search for new compounds with improved biological activ-
ity, the present investigation deals with the synthesis of
compound 1, as outlined in Scheme 1, which might have useful
biological and therapeutic activities. The crystal structure of title
they are agonists of the nuclear receptor PPAR
a (Peroxisome Pro-
liferator-Activated Receptor).1,2 The fibrates are a widely used class
of lipid-modifying agents that decrease plasma triglycerides.3,4 The
fibrate pharmacophore has been of interest to medicinal chemists
ever since the initial discovery that ethyl chlorophenoxyisobuty-
rate (clofibrate) possessed hypolipidemic properties.5 Clofibrate
is a prodrug which is biotransformed into clofibric acid, the active
metabolite which binds to PPAR
a.
Recent studies showed that fenofibrate but not fenofibric acid,
exerted
a
potent inhibitory activity against the enzyme
11b-hydroxysteroid dehydrogenase type 1 (11b-HSD1),6 as well
as negatively modulated the expression of 11b-HSD1 mRNA in
hepatocytes.7
Several lines of evidence have implicated glucocorticoids and
11b-HSD1 activity in the etiology and/or maintenance of type 2
diabetes mellitus and metabolic syndrome. The glucocorticoids,
such as cortisol, are potent antagonists of insulin action and pro-
moters of gluconeogenesis in liver, leading to an increase in blood
⇑
Corresponding author. Tel./fax: +52 777 329 7089.
E-mail address: gabriel_navarrete@uaem.mx (G. Navarrete-Vázquez).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.03.122
Please cite this article in press as: Navarrete-Vázquez, G.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.03.122

2
G. Navarrete-Vázquez et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
O
O
OH
O
in vivo hydrolysis
Cl
O
Cl
O
Clofibric acid
Clofibrate
N
N
N
O
N
H
NH
O
Cl
1
Figure 2. A view of the supramolecular synthon I, showing the atom labelling
scheme. Bifurcated N–Hꢁ ꢁ ꢁN and N-Hꢁ ꢁ ꢁO hydrogen bonds are represented by
dashed lines.
Figure 1. Metabolism of fibrates leading to carboxylic acid form, and the tetrazole
isostere prepared in this work (1).
Table 1
Experimental in vitro inhibition effect of compounds against 11b-HSD1
N
N
N
Compound
% of Inhibition 11b-HSD1 @ [10 lM]
O
O
N
H
1
51.17
19.14
18.11
78.32
NH
OH
N
N
N
Clofibric acid
Clofibrate
Carbenoxolone
a
H2N
+
Cl
O
Cl
O
N
H
1
and showed 51.17%, 19.14% and 18% of inhibition at 10
tively (Table 1).
lM, respec-
Scheme 1. Preparation of 1, (a) DCC, DMAP (catalyst), methylene chloride, room
temperature.
Results presented in Table 1 show that 1 was 2.5 times more
effective inhibiting 11b-HSD1 than clofibric acid and clofibrate.
The formation of the amide and the replacement of carboxylic acid
by a tetrazole scaffold increased the potency against 11b-HSD1.
The inhibition of this enzyme could reduce the glucose serum lev-
els in diabetic patients due to the low production of cortisol.
In order to gain an insight into the binding mode of 1, it was
docked into the catalytic site of the human 11b-HSD1 (PDB code:
2-BEL) using the program AutoDock 4.0.14 According to the results
obtained from LigPrep, four tautomeric forms and one ionized state
coexist at pH 7.0 2.0. The pKa value calculated for the neutral
form of the tetrazole ring of compound 1 is 2.89, which means that
only the ionized state is possible at the pH experimental value
(7.0). Even though only the ionized state is possible at pH 7.0,
docking calculations were performed also on the tautomeric forms.
The lowest binding energies calculated by AutoDock for the tauto-
meric forms were ꢀ7.14, ꢀ7.62, ꢀ7.83, and ꢀ7.79 kcal/mol and for
the ionized state was ꢀ7.97 kcal/mol. Taken together, these results
suggest that the ionized state is the most favorable form that binds
into the catalytic site of 11b-HSD1. Analysis of the binding mode of
1 hereafter will be focused on the ionized state. Figure 3 shows the
two top ranked binding modes of 1 found by AutoDock with dock-
ing energies of ꢀ7.97 and ꢀ7.62 kcal/mol, respectively. The first
binding mode (Fig. 3A) is characterized by the ligand making con-
tacts with the catalytic residue Ser170; as well as the polar resi-
dues Tyr177, Gly216, His232, Gln234, Tyr258, and Asp259, which
makes a hydrogen bond with the tetrazole ring (Fig. 3A). Other
hydrophobic contacts are made with Leu171, Ala172, Leu217,
Ile218, Met233, Ala235, Ala236, and Leu262. In this binding model,
1 is blocking the entrance of the catalytic site. It is important to
mention that only 14.5% of the solutions found by AutoDock fell
into this orientation. The second binding mode in Figure 3B is dif-
ferentiated from the first one in that compound 1 is located at the
opposite side of the binding cavity making contacts with the cata-
lytic residues Ser170 and Tyr183; other short contacts are made
with the polar residues Tyr177, Gly216, Thr222 and hydrophobic
compound determined by single-crystal X-ray diffraction, the
in vitro inhibition of 11b-HSD1, the in vivo hypoglycemic effect,
as well as a theoretical and predictive mode of binding of com-
pound with the enzyme is also reported here.
In vitro screens combined with in silico and in vivo assessments
leading to the concept of in combo screening in drug discovery.12
Compound 1 was prepared in a single-step reaction, starting
from the 2-(4-chlorophenoxy)-2-methylpropionic acid and 1H-tet-
razol-5-amine, using dicyclohexylcarbodiimide as a coupling agent
(Scheme 1).
Compound 1 was purified by recrystallization. The chemical
structure was confirmed on the basis of its NMR (1H, 13C) spectro-
scopic data,13 as well as mass spectrometry, the crystal structure
was determined by single-crystal X-ray diffraction, and their purity
ascertained by microanalysis.
A single crystal of 1 was grown at room temperature by slow
evaporation of ethanol solution. Figure 2 shows the supramolecu-
lar synthon I with the atom labelling system. The most relevant
crystallographic data and hydrogen bond geometry are summa-
14
rized in the Supplementary data.
The X-ray structure revealed that in compound 1 the unit cell
contains two independent molecules with different conformation.
The dihedral angles involving atoms O1–C7–C8–N1 and O3–C18–
C19–N6 are ꢀ40.3 (2) and ꢀ10.3 (2)°. The conformers are coupled
through bifurcated N–H N and N–H O hydrogen bonds. In a recent
paper, we reported some other amide derivatives from clofibric
acid and 2-aminobenzothiazole and all of them crystallized show-
ing hypervalent contacts (C@Oꢁ ꢁ ꢁS, Sꢁ ꢁ ꢁS), hydrogen bonds Y–H X
(Y = O, N, C; X = O, N, Cl,
p
) and van der Waals contacts (Clꢁ ꢁ ꢁ
p,
Sꢁ ꢁ ꢁp
, Hꢁ ꢁ ꢁH).15
A preliminary in vitro compound screen to identify inhibitors of
human 11b-HSD1 was performed. Inhibition of 11b-HSD1 was
determined using a human embryonic kidney (HEK293) cell-based
assay.9,14,16 Compound 1, clofibric acid and clofibrate were tested
Please cite this article in press as: Navarrete-Vázquez, G.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.03.122

G. Navarrete-Vázquez et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
3
Figure 3. The two top ranked binding modes and ligand interaction diagrams of compound 1 into the catalytic site of 11b-HSD1. (A) Binding mode and ligand interaction
diagram of compound 1 with the lowest binding energy (ꢀ7.97 kcal/mol). (B) Binding mode of compound 1 with second lowest binding energy (ꢀ7.62 kcal/mol). The active
residues Ser170 and Tyr183 are represented in cobalt blue, NADP+ in green and 1 in silver. The ligand interaction diagrams are represented as follow: The ligand proximity
contour is depicted with a dotted line. The ligand solvent exposure is represented with blue circles; larger and darker circles on ligand atoms indicate more solvent exposure.
The receptor solvent exposure differences, in the presence and absence of the ligand, are represented by the size and intensity of the turquoise discs surrounding the residues;
larger and darker discs indicate residues highly exposed to solvent in the active site when the ligand is absent.
contacts with Ile121, Leu171, Ala172, Leu217, Ala223, Ala226, and
Val227. Also an important short contact is made with the nicotin-
20
amide ring of NADP+ (Fig. 3B). Interestingly, in the second binding
10
mode the tetrazole ring is occupying a similar position as com-
pared with the carboxylic group of carbenoxolone at position
C30. The oxygen atom of the ether group of compound 1 is 3.4 Å
away from C4 of the nicotinamide ring (where the proton transfer
takes place) and 3.9 Å away from the hydroxyl group of the cata-
lytic residue Tyr183. It should be noted that 80% of the solutions
found by AutoDock fell into this orientation.
Compound 1 was evaluated for in vivo hypoglycemic activity
using a STZ–nicotinamide rat model of non insulin dependent dia-
betes mellitus rat model.14,16,17 Glibenclamide (5 mg/Kg, p.o.) was
taken as positive control. The hypoglycemic activity of 1 was deter-
mined using a 50 mg/Kg intragastric single dose. Compound 1
demonstrated important hypoglycemic activity, by lowering glyce-
mia ranging from 16% to 77%. The effect was sustained during the
7 h of experiment and it was more pronounced than the hypogly-
cemic action showed by glibenclamide (Fig. 4). It is important to
mention that clofibric acid did not show any hypoglycemic effect.
These results agree with those reported by Kim et al, where fenof-
ibrate but not fenofibric acid, exerted a potent inhibitory activity
against 11b-HSD1.6
0
-10
-20
-30
-40
-50
-60
-70
-80
-90
*
*
ISS
1
Glibenclamide
*
*
0
2
4
6
8
Time (h)
Figure 4. Effect of a single dose of 1 (50 mg/Kg; intragastric, n = 5) in streptozo-
tocin–nicotinamide-induced diabetes rat model. ISS: Isotonic saline solution. ⁄p
<0.05 versus ISS group.
50% of the treated animals (LD50) of the compound 1, clofibric acid
and clofibrate was calculated through the ACD/ToxSuite software,
v. 2.95 (Table 2).
In silico prediction of the acute toxicity is performed during
drug design and development in order to avoid the experimental
study of potentially harmful substances. The dose that is lethal to
Please cite this article in press as: Navarrete-Vázquez, G.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.03.122

4
G. Navarrete-Vázquez et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
Table 2
Supplementary data
Toxicity profiles predicted for compound 1 and selected fibrates
Compd
LD50 (mg/Kg)
Mouse
ip
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2013.03.
122.
Rat
ip
po
po
1
830
290
540
1900
1170
1220
1700
727
910
2200
897
940
References and notes
Clofibric acid
Clofibrate
1. Forcheron, F.; Cachefo, A.; Thevenon, S.; Pinteur, C.; Beylot, M. Diabetes 2002,
51, 3486.
2. Nevin, D. K.; Lloyd, D. G.; Fayne, D. Curr. Med. Chem. 2011, 18, 5598.
3. Miller, D. B.; Spence, J. D. Clin. Pharmacokinet. 1998, 34, 155.
4. Singh, N.; Tamariz, J.; Chamorro, G.; Medina-Franco, J. L. Mini-Rev. Med. Chem.
2009, 9, 1272.
5. Thorp, J. M.; Waring, W. S. Nature 1962, 194, 948–949.
6. Kim, C. H.; Ramu, R.; Ahn, J. H.; Bae, M. A.; Cho, Y. S. Mol. Cell. Biochem. 2010,
344, 91.
7. Hermanowski-Vosatka, A.; Gerhold, D.; Mundt, S. S.; Loving, V. A.; Lu, M.; Chen,
Y.; Elbrecht, A.; Wu, M.; Doebber, T.; Kelly, L.; Milot, D.; Guo, Q.; Wang, P. R.;
Ippolito, M.; Chao, Y. S.; Wright, S. D.; Thieringer, R. Biochem. Biophys. Res.
Commun. 2000, 279, 330.
8. Alberts, P.; Engblom, L.; Edling, N.; Forsgren, M.; Klingstrom, G.; Larsson, C.;
Ronquist-Nii, Y.; Ohman, B.; Abrahmsen, L. Diabetologia 2002, 45, 1528.
9. Webster, S. P.; Ward, P.; Binnie, M.; Craigie, E.; McConnell, K. M. M.; Sooy, K.;
Vinter, A.; Seckl, J. R.; Walker, B. R. Bioorg. Med. Chem. Lett. 2007, 17, 2838.
10. Wermuth, C. G. The Practice of Medicinal Chemistry, 3rd ed.; Academic Press:
London, 2008. pp 283–287.
In these predictions, compound 1 was 2-times less toxic than
the other fibrates, showing calculated LD50 by different administra-
tion routes ranging from 830 to 2200 mg/Kg.
In conclusion, we discovered a tetrazole isosteric analogue (1)
of clofibric acid as a promising antidiabetic compound using in
combo studies (in vitro, in silico and in vivo). Compound 1 shows
significant 11b-HSD1 inhibitory activity, being 2.5-times more ac-
tive than clofibrate. The formation of the amide and the replace-
ment of carboxylic acid by the tetrazole scaffold in compound 1,
increased the potency against 11b-HSD1 and also improved the
in vivo hypoglycemic effect.
Even though compound 1 has four stable tautomeric forms, it is
suggested that the ionized state is the only one present at the
experimental conditions. This is in agreement with the calculated
binding energy for the ionized state. According to the docking
models, 1 may bind into the catalytic site of 11b-HSD1 in two dif-
ferent orientations. In one of them, 1 shows important short con-
tacts with catalytic residues and also with the nicotinamide ring
of NADP+. The high in vivo hypoglycemic activity showed by this
compound makes it a suitable lead to develop new chemical enti-
ties for potential use in the treatment of diabetes and metabolic
syndrome.
11. Biot, C.; Bauer, H.; Schirmer, R. H.; Davioud-Charvet, E. J. Med. Chem. 2004, 47,
5972.
12. van de Waterbeemd, H. Basic Clin. Pharmacol. Toxicol. 2005, 96, 162.
13. 2-(4-chlorophenoxy)-2-methyl-N-(1H-tetrazol-5-yl)propanamide
Recrystallized from ethanol. Yield: 74.5%, mp: 151.2–152.5 °C. Mp: IR (KBr):
max/cmꢀ1 3250 (N–H), 1690 (C@O), 1591 (N@C). 1H NMR (400 MHz, DMSO-
d6): dH 1.30 (s, 6H, gem-(CH3)2), 6.56–6.67 (dd, 2H, H-2, H-6, J = 8.5, J = 2.5 Hz),
(1).
m
6.88–6.94 (dd, 2H, H-3, H-5, J = 8.5, J = 2.5 Hz), 11.2 (s, 2H, NHamide,NHtetrazole
)
ppm. 13C NMR (100 MHz, DMSO-d6): d 24.3 (gem-(CH3)2), 80.8 (C–O), 121.9 (C-
2, C-6), 127.9 (C-4), 128.7 (C-3, C-5), 152.2 (C-1, C-100), 173.1 (C@O) ppm. MS/
FAB+: m/z 282 (M+H+). HRMS (FAB+): m/z 282.0749 [M+H]+ (Calcd for
C
11H12ClN5O2: 282.0752). Anal. Calcd for C11H12ClN5O2: C, 46.90; H, 4.29; N,
24.86. Found: C, 46.97; H, 4.23; N, 24.89.
14. For more details, see Supplementary data.
15. Navarrete-Vázquez, G.; Alaniz-Palacios, A.; Tlahuextl, M.; Bernal-Uruchurtu,
M.; Tlahuext, H. CrystEngComm 2012, 14, 1256.
16. Moreno-Díaz, H.; Villalobos-Molina, R.; Ortiz-Andrade, R.; Díaz-Coutiño, D.;
Medina-Franco, J. L.; Webster, S. P.; Binnie, M.; Estrada-Soto, S.; Ibarra-Barajas,
M.; León-Rivera, I.; Navarrete-Vázquez, G. Bioorg. Med. Chem. Lett. 2008, 18,
2871.
17. Torres-Piedra, M.; Ortiz-Andrade, R.; Villalobos-Molina, R.; Singh, N.; Medina-
Franco, J. L.; Webster, S. P.; Binnie, M.; Navarrete-Vázquez, G.; Estrada-Soto, S.
Eur. J. Med. Chem. 2010, 45, 2606.
Acknowledgement
This work was supported in part by the Consejo Nacional de
Ciencia y Tecnología (CONACyT) under Grant No. 100608 and Fac-
ultad de Farmacia internal grant. We are grateful to Juan José Ra-
mirez-Espinosa for technical assistance and to Jacob Waddell for
proofreading the manuscript. J.L.M.-F. thank to the State of Florida
for funding.
Please cite this article in press as: Navarrete-Vázquez, G.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.03.122