Ursolic acid derivatives as potential antidiabetic agents: In vitro, in vivo, and in silico studies

Article abstract of DOI:10.1002/ddr.21422

(Table presented.). Protein tyrosine phosphatase 1B (PTP-1B) has attracted interest as a novel target for the treatment of type 2 diabetes, this because its role in the insulin-signaling pathway as a negative regulator. Thus, the aim of current work was to obtain seven ursolic acid derivatives as potential antidiabetic agents with PTP-1B inhibition as main mechanism of action. Furthermore, derivatives 1–7 were submitted in vitro to enzymatic PTP-1B inhibition being 3, 5, and 7 the most active compounds (IC₅₀?=?5.6, 4.7, and 4.6?μM, respectively). In addition, results were corroborated with in silico docking studies with PTP-1B orthosteric site A and extended binding site B, showed that 3 had polar and Van der Waals interactions in both sites with Lys120, Tyr46, Ser216, Ala217, Ile219, Asp181, Phe182, Gln262, Val49, Met258, and Gly259, showing a docking score value of ?7.48?Kcal/mol, being more specific for site A. Moreover, compound 7 showed polar interaction with Gln262 and Van der Waals interactions with Ala217, Phe182, Ile219, Arg45, Tyr46, Arg47, Asp48, and Val49 with a predictive docking score of ?6.43?kcal/mol, suggesting that the potential binding site could be localized in the site B adjacent to the catalytic site A. Finally, derivatives 2 and 7 (50?mg/kg) were selected to establish their in vivo antidiabetic effect using a noninsulin-dependent diabetes mice model, showing significant blood glucose lowering compared with control group (p?<.05).

Full text of DOI:10.1002/ddr.21422

Received: 14 November 2017
DOI: 10.1002/ddr.21422
Revised: 4 January 2018
Accepted: 5 January 2018
|
|
R E S E A R C H A R T I C L E
Ursolic acid derivatives as potential antidiabetic agents: In vitro,
in vivo, and in silico studies
ꢀ
1
Virginia Flores-Morales2*
Paolo Paoli3
ꢀ
|
|
|
Ricardo Guzman-Avila
4
1
Guido Camici3
Juan Jose Ramírez-Espinosa
Litzia Ceron-Romero
|
ꢀ
|
ꢀ
|
1
Sergio Hidalgo-Figueroa5
Maria Yolanda Rios6
ꢀ
|
|
|
Gabriel Navarrete-Vazquez
Rafael Villalobos-Molina7
Samuel Estrada-Soto1
|
Enabling Technologies
Strategy, Management & Health Policy
Hit, Lead &
Candidate Discovery
Preclinical Research
& Development
Clinical
Research
Post-Market
Research
1
2
ꢀ
Facultad de Farmacia, Universidad Autonoma del Estado de Morelos, Cuernavaca, Morelos 62209, Mexico
ꢀ
ꢀ
ꢀ
ꢀ
Laboratorio de Síntesis Asimetrica y Bioenergetica (LSAyB), Unidad Academica de Ciencias Químicas, Universidad Autonoma de Zacatecas, Zacatecas, Zacatecas
98160, Mexico
3
ꢀ
Dipartimento di Scienze Biomediche Sperimentali e Cliniche “Mario Serio”, Sezione di Scienze Biochimiche, Universita degli Studi di Firenze, Firenze 50134, Italy
4
5
6
7
ꢀ
ꢀ
ꢀ
ꢀ
Instituto de Ciencias Biomedicas, Universidad Autonoma de Ciudad Juarez, Juarez, Chihuahua 32310, Mexico
ꢀ
ꢀ
ꢀ
ꢀ
Catedras CONACyT IPICYT/Consorcio de Investigacion, Innovacion y Desarrollo para las Zonas Aridas, San Luis Potosí 78216, Mexico
ꢀ
Centro de Investigaciones Químicas, IICBA, Universidad Autonoma del Estado de Morelos, Cuernavaca, Morelos 62209, Mexico
ꢀ
ꢀ
ꢀ
ꢀ
Unidad de Biomedicina, Facultad de Estudios Superiores Iztacala, Universidad Nacional Autonoma de Mexico, Tlalnepantla, Estado de Mexico 54090, Mexico
Correspondence
Abstract
Dr. Samuel Estrada-Soto
Email: enoch@uaem.mx
Protein tyrosine phosphatase 1B (PTP-1B) has attracted interest as a novel target for the treatment
of type 2 diabetes, this because its role in the insulin-signaling pathway as a negative regulator. Thus,
the aim of current work was to obtain seven ursolic acid derivatives as potential antidiabetic agents
with PTP-1B inhibition as main mechanism of action. Furthermore, derivatives 1–7 were submitted
in vitro to enzymatic PTP-1B inhibition being 3, 5, and 7 the most active compounds (IC50 5 5.6, 4.7,
and 4.6 lM, respectively). In addition, results were corroborated with in silico docking studies with
PTP-1B orthosteric site A and extended binding site B, showed that 3 had polar and Van der Waals
interactions in both sites with Lys120, Tyr46, Ser216, Ala217, Ile219, Asp181, Phe182, Gln262,
Val49, Met258, and Gly259, showing a docking score value of 27.48 Kcal/mol, being more specific
for site A. Moreover, compound 7 showed polar interaction with Gln262 and Van der Waals interac-
tions with Ala217, Phe182, Ile219, Arg45, Tyr46, Arg47, Asp48, and Val49 with a predictive docking
score of 26.43 kcal/mol, suggesting that the potential binding site could be localized in the site B
adjacent to the catalytic site A. Finally, derivatives 2 and 7 (50 mg/kg) were selected to establish
their in vivo antidiabetic effect using a noninsulin-dependent diabetes mice model, showing signifi-
cant blood glucose lowering compared with control group (p < .05).
K E Y W O R D S
antidiabetic agents, docking, pentacyclic acid triterpenes, PTP-1B inhibition, ursolic acid derivatives
*The same level of participation as the principal and correspondence author is considered.
|
1
C
V
Drug Dev. Res.. 2018;1–11.
wileyonlinelibrary.com/journal/ddr
2018 Wiley Periodicals, Inc..

ꢀ
ꢀ
2
GUZMAN-AVILA ET AL.
|
2014), antidepressant (Colla et al., 2014; Machado et al., 2012), antioxi-
|
1
INTRODUCTION
ꢀ
dant and antiatherogenic effects (Allouche, Beltran, Gaforio, Uceda, &
Mesa, 2010), among others, which makes it an important scaffold in
medicinal chemistry. Currently, is known that the ursolic acid antidia-
betic effect is mediated by its interaction with different targets: UA
acts as insulinomimetic (Jung et al., 2007), increasing the autophospho-
rylation and activation of insulin receptor, and also proceeds as insulin
sensitizer since it is able to inhibit the PTP-1B enzyme (Ramírez-Espi-
nosa et al., 2011; Zhang et al., 2006). A recent report showed that UA
augments GLUT4 as a result of increasing intracellular Ca21 concentra-
tion, which augments movement and translocation of GLUT4 trans-
porter unto membrane (Castro et al., 2015). Other UA direct and
indirect antidiabetic mechanisms are enzymatic inhibition of: 11b-
HSD1 (Rollinger et al., 2010), glycogen phosphorylase (Jang, Kim, Choi,
Kwon, & Lee, 2010), a-glucosidase (Wu, Zhang, Lu, He, & Zhao, 2014),
and aldose reductase (Lee et al., 2014). Thus, the aim of current work
was to synthesize new UA derivatives based on this scaffold in order
to discover more potent, specific and effective antidiabetic bioactive
compounds.
According to the World Health Organization, the chronic degenerative
diseases represent the main causes of death in the world. Diabetes is a
metabolic disease, which is characterized by an impaired regulation of
glucose, and additionally, lipids and proteins (Mahapatra, Asati, &
Bharti, 2015). This because insulin is not able to perform its function as
main metabolic regulator in peripheral tissues, because of lack of insulin
(type 1 diabetes) or because of resistance to insulin action and/or
impaired insulin secretion (type 2 diabetes). Diabetes increases the risk
to develop micro and macrovascular complications such as neuropathy,
nephropathy and retinopathy, triggering consequences like cardiovas-
cular diseases (myocardial infarction, stroke, among others), blindness,
limb amputation, and end-stage renal disease (Brownlee, 2001). More-
over, diabetes has been linked to central nervous system diseases like
depression and cognition disturbances (Li, Zhang, & Sima, 2007).
Despite the availability of drugs for the treatment of diabetes,
many of them fail because two frequent reasons (Tahrani, Bailey, Del
Prato, & Barnett, 2011): (a) many drugs cause diverse side-effects, that
is, hypoglycemia, edema, weight gain, increasing cardiovascular risk,
which can limit its use for the patient; (b) the improvement in the con-
trol of glycaemia is not continuous, due to the degenerative features
related with the disease, thus, eventually it is necessary to increase the
dose of the drug used or combinatorial therapy is needed to increase
control. Thus, currently it is necessary to develop new drugs that coun-
teract these points, and/or to obtain novel drugs which act at new tar-
gets. In this context, protein tyrosine phosphatase 1B (PTP-1B) has
recently shown an important influence over insulin sensitivity, since it
is implicated in modulating insulin signal transduction becoming a key
regulator of insulin-receptor activity, and downstream signaling path-
ways (Johnson, Ermolieff, & Jirousek, 2002). This is related with a
dephosphorylation of active insulin receptor (IR) (Romsicki, Reece,
Gauthier, Asante-Appiah, & Kennedy, 2004). Also, PTP-1B is implicated
in the regulation of the leptin receptor, which is involved in food intake
regulation by CNS (Klok, Jakobsdottir, & Drent, 2007). PTP-1B is an
enzyme formed by 435 amino acids residues, which is anchored into
cytoplasmic face of endoplasmic reticulum (Johnson et al., 2002) and
presents two important regions, sites A (catalytic) and B (extended
binding pocket). The A site is formed by 214–221 residues and cata-
lyzes the dephosphorilation of tyrosine arylphosphate residues (Tonks,
2003). The second region (B site), is an extended binding site but inac-
tive region next to A site, which only binds to arylphosphate groups
(Puius et al., 1997). Then, inhibition of PTP-1B could be a very impor-
tant target to develop new antidiabetic drugs. There are several syn-
thetic and natural derived compounds which are described as PTP-1B
inhibitors (Verma, Gupta, Chaudhary, & Garg, 2017).
|
2
METHODS AND MATERIALS
|
2.1 General
The preparation of 1–6 UA derivatives was carried out in a CEM Dis-
covery BenchMate system with a potency of 100 W. The temperature
reaction was set at 608C under open vessel system. Reaction progress
and purity of products was monitored by thin layer chromatography
(TLC) using silica gel 60 F254 aluminum sheets, visualizing the plates
under UV-light and then spraying with (NH4)Ce(SO4)4 in 2N H2SO4
and heating. The products were purified by flash column chromatogra-
phy (FCC) using 40–63 lm silica gel (230–440 mesh). Melting points
were determined on a Fisher–Johns apparatus and are reported
uncorrected.
1H, 13C, and bidimensional spectra were recorded on Varian unity
400TM spectrometer equipped with a 5 mm inverse detection pulse
field gradient probe at 258C at 400 MHz for 1H and 100 MHz for 13C.
Chemical shifts are given in values of part per million, referenced to tet-
ramethylsilane (TMS) as an internal standard.
|
2.2 Synthesis of ursolic acid derivatives
|
2.2.1 Starting material
Ursolic acid was purchased from Sigma-Aldrich (St. Louis, MO). Chemi-
cal reagents were purchased analytical grade from Sigma-Aldrich, Mex-
ico. Commercially available solvents (n-hexane, ethyl acetate, and
acetone) were used after purification and dried according to standard
procedures.
Natural products represent a great source for to study and develop
new drugs candidates for the treatment of diabetes (Liu, Chen, Hu,
Guo, & Shen, 2010). Ursolic acid (UA) is a triterpene that has shown
metabolic (Sheng & Sun, 2011), cardiovascular (Aguirre-Crespo et al.,
2006; Rios et al., 2012; Somova, Nadar, Rammanan, and Shode, 2003a;
Somova, Shode, Ramnanan, & Nadar, 2003b), antibacterial (Nascimento
et al., 2014), antifungal (Innocente et al., 2014), cytotoxic (Li et al.,
|
2.2.2 Synthesis procedure for derivatives 1, 2, and 6
In a well-stirred and cooled solution of UA (1.0 equiv.) in dry THF was
added 2.1 equiv. of the corresponding base. Mixture reaction was
stirred at 08C for 30 min. Alkyl halide (2.1 equiv.) was added dropwise

ꢀ
ꢀ
GUZMAN-AVILA ET AL.
3
|
over period of 5 min, then the reaction mixture was placed under
microwave radiation in CEM equipment at 608C. Solvent was elimi-
nated under vacuum and the reaction crude was purified by FCC.
(C-10), 36.8 (C-22), 33.2 (C-7), 30.9 (C-21), 28.3 (C-23), 28.2 (C-15),
27.4 (C-2), 24.4 (C-16), 23.8 (C-27), 23.5 (C-11), 21.4 (C-30), 18.5 (C-6),
17.2 (C-26), 17.1 (C-29), 15.8 (C-25), 15.6 (C-24).
(3b)-3-methoxyursen-12-en-28-oic acid (1)
|
2.2.3 Synthesis procedure for compound 3, 4, and 5
UA 120 mg (0.263 mmol), THF (2 mL), Na 15.1 mg (0.657 mmol),
methyl iodide 35 lL (0.552 mmol), 2 h reaction. FCC (n-hexane/ace-
tone 90:10 v/v) affording 80.4 mg product, 65% yield, TLC:
Rf 5 0.41 (n-hexane/acetone 80:20 v/v); white powder; m.p. 165–
1678C. 1H-NMR (400 MHz, CDCl3) d 5.24 (t, 3J 5 3.6 Hz, 1H), 3.60
(s, 3H), 3.21 (dd, 3J 5 10.8, 4.8 Hz, 1H), 2.22 (d, 3J 5 10.7 Hz, 1H),
1.99 (td, 3J 5 13.3, 4.5 Hz, 1H), 1.90 (dd, 3J 5 8.9, 3.6 Hz, 2H),
1.76 (td, 3J 5 13.6, 4.6 Hz, 1H), 1.57 (m, 12H), 1.31 (m, 5H), 1.06
(s, 4H), 0.99 (d, 4H), 0.93 (dd, 6H), 0.85 (d, 3J 5 6.5 Hz, 3H), 0.73
(dd, 7H). 13C-NMR (100 MHz, CDCl3) d 178.1 (C-28), 138.3 (C-13),
125.7 (C-12), 79.1 (C-3), 55.4 (C-5), 53.0 (C-18), 51.5 (C-A), 48.2 (C-
17), 47.7 (C-9), 42.1 (C-14), 39.7 (C-8), 39.2 (C-19), 39.0 (C-20),
38.9 (C-4), 38.8 (C-1), 37.1 (C-10), 36.8 (C-22), 33.1 (C-7), 30.8 (C-
21), 28.3 (C-23), 28.2 (C-15), 27.4 (C-2), 24.4 (C-16), 23.8 (C-27),
23.4 (C-11), 21.3 (C-30), 18.5 (C-6), 17.2 (C-26), 17.1 (C-29), 15.8
(C-25), 15.6 (C-24).
In a well-stirred and cooled solution of UA (1.0 equiv.) in dry CH2Cl2
and DMF was added 2.1 equiv. of the corresponding base. Mixture
reaction was stirred at 08C for 30 min. Acyl halide (2.1 equiv.) was
added dropwise over a 5 min period, then, the reaction mixture was
placed under microwave radiation in CEM equipment at 608C. Solvent
was eliminated under vacuum and the reaction crude was purified by
FCC.
(3b)-3-(acetyloxy)ursen-12-en-28-oic acid (3)
UA 61.3 mg (0.134 mmol), DCM (2 mL), DMF (1 mL), Et3N 40 lL
(0.276 mmol), acetyl chloride 10 lL (0.144 mmol), 2 h reaction. FCC (n-
hexane/ethyl acetate 90:10 v/v) affording 31.5 mg product, 48% yield,
TLC: Rf 5 0.35 (n-hexane/ethyl acetate 85:15 v/v); white powder; m.p.
185–1878C. [Lit. 175–1788C] (Collins et al., 2002). 1H-NMR (400 MHz,
CDCl3): d 5.16 (t, 3J 5 3.4 Hz, 1H), 4.43 (dd, 3J 5 9.3, 6.7 Hz, 1H), 2.11
(d, 3J 5 11.3, Hz, 1H), 1.98 (s, 3H), 1.92 (dd, 3J 5 13.2, 4.2 Hz, 1H),
1.84 (dd, 3J 5 8.8, 3.3 Hz, 2H), 1.78 (dd, 3J 5 13.6, 4.4 Hz, 1H), 1.61
(m, 6H), 1.43 (m, 4H), 1.24 (m, 5H), 1.02 (m, 5H), 0.88 (m, 6H), 0.79 (d,
3J 5 7.7 Hz, 9H), 0.69 (s, 3H). 13C-NMR (100 MHz, CDCl3): d 184.3 (C-
28), 171.3 (C-A), 138.1 (C-13), 125.9 (C-12), 81.1 (C-3), 55.5 (C-5), 52.7
(C-18), 48.2 (C-17), 47.6 (C-9), 42.1 (C-14), 39.7 (C-8), 39.2 (C-19), 39.0
(C-20), 38.4 (C-4), 37.9 (C-1), 37.1 (C-10), 36.9 (C-22), 33.0 (C-7), 30.8
(C-21), 28.3 (C-23), 28.2 (C-15), 24.2 (C-16), 23.8 (C-27), 23.5 (C-11),
21.5 (C-B), 21.4 (C-30), 18.3 (C-6), 17.3 (C-26), 17.2 (C-29), 16.9 (C-
25), 15.7 (C-24).
(3b)-3-ethoxyursen-12-en-28-oic acid (2)
UA 120 mg (0.263 mmol), THF (2 mL), Na8 15.1 mg (0.657 mmol), ethyl
bromide 50 lL (0.552 mmol), 2 hours reaction. FCC (n-hexane/acetone
95:05 v/v) affording 113.4 mg product, 89% yield, TLC: Rf 5 0.67 (n-
hexane/acetone 80:20 v/v); white powder; m.p. 135–1378C. 1H-NMR
(400 MHz, CDCl3) d 5.23 (t, 3J 5 3.6 Hz, 1H), 4.04 (q, 3J 5 7.1 Hz, 2H),
3.20 (dd, 3J 5 10.9, 4.9 Hz, 1H), 2.23 (d, 3J 5 10.7 Hz, 1H), 1.97 (td,
J 5 13.3, 4.5 Hz, 1H), 1.89 (dt, 3J 5 8.8, 3.7 Hz, 2H), 1.77 (td,
3J 5 13.7, 4.6 Hz, 1H), 1.58 (m, 11H), 1.30 (m, 5H), 1.21 (t, 3J 5 7.1 Hz,
3H), 1.07 (s, 3H), 0.98 (s, 4H), 0.92 (m, 6H), 0.84 (d, 3J 5 6.5 Hz, 3H),
0.73 (dd, 7H). 13C-NMR (100 MHz, CDCl3) d 177.7 (C-28), 138.4 (C-
13), 125.7 (C-12), 79.3 (C-3), 60.2 (C-A), 55.5 (C-5), 53.1 (C-18), 48.1
(C-17), 47.8 (C-9), 42.3 (C-14), 39.8 (C-8), 39.3 (C-19), 39.1 (C-20), 39.0
(C-4), 38.9 (C-1), 37.2 (C-10), 36.9 (C-22), 33.3 (C-7), 30.9 (C-21), 28.4
(C-23), 28.2 (C-15), 27.5 (C-2), 24.4 (C-16), 23.7 (C-27), 23.5 (C-11),
21.4 (C-30), 18.5 (C-6), 17.3 (C-26), 17.2 (C-29), 15.8 (C-25), 15.7 (C-
24), 14.4 (C-B).
(3b)-3-(formyloxy)ursen-12-en-28-oic acid (4)
UA 60.3 mg (0.134 mmol), DCM (2 mL), DMF (1 mL), K2CO3
36.3 mg (0.263 mmol), benzoyl chloride 20 lL (0.144 mmol), 2 h
reaction. FCC (n-hexane/acetone 90:10 v/v) affording 37.5 mg prod-
uct, 58% yield, TLC: Rf 5 0.31 (n-hexane/acetone 90:10 v/v); white
crystals; m.p. 120–1218C. [Lit. 117–1198C] (Collins et al., 2002). 1H-
NMR (400 MHz, CDCl3) d 8.11 (s, 1H), 5.23 (t, 3J 5 3.8 Hz, 1H),
4.62 (dd, 3J 5 9.8, 6.1 Hz, 1H), 2.18 (d, 3J 5 11.8 Hz, 1H), 2.01 (m,
1H), 1.92 (dd, 3J 5 8.8, 3.7 Hz, 2H), 1.85 (dd, 3J 5 13.6, 5.0 Hz,
1H), 1.68 (m, 6H), 1.50 (m, 4H), 1.32 (m, 7H), 1.09 (m, 5H), 0.96 (m,
6H), 0.86 (m, 9H), 0.77 (s, 2H). 13C NMR (100 MHz, CDCl3) d
184.1 (C-28), 161.3 (C-A), 138.2 (C-13), 125.9 (C-12), 81.3 (C-3),
55.5 (C-5), 52.7 (C-18), 48.2 (C-17), 47.5 (C-9), 42.1 (C-14), 39.7 (C-
8), 39.2 (C-19), 39.0 (C-20), 38.4 (C-4), 37.8 (C-1), 37.1 (C-10), 36.9
(C-22), 33.0 (C-7), 30.8 (C-21), 29.9 (C-23), 28.2 (C-15), 24.2 (C-2),
23.9 (C-16), 23.8 (C-27), 23.5 (C-11), 21.4 (C-30), 18.4 (C-6), 17.3
(C-26), 17.2 (C-29), 16.9 (C-25), 15.7 (C-24).
Methyl (3b)-3-hydroxyursen-12-en-28-oate (6)
UA 120 mg (0.263 mmol), DCM (2 mL), DMF (1 mL), K2CO3 36.3 mg
(0.263 mmol), methyl iodine 35 lL (0.552 mmol), 3 h reaction. FCC (n-
hexane/ethyl acetate 90:10 v/v) affording 94.1 mg product, 76% yield;
TLC: Rf 5 0.55 (n-hexane/acetone 80:20 v/v); white powder; m.p.
1728C [Lit. 166–1688C] (Ceron-Romero et al., 2016). 1H-NMR
ꢀ
(400 MHz, CDCl3) d 5.24 (t, 3J 5 3.6 Hz, 1H), 3.60 (s, 3H), 3.21 (dd,
3J 5 10.9, 4.9 Hz, 1H), 2.23 (d, 3J 5 11.3 Hz, 1H), 2.00 (td, 3J 5 13.3,
4.5 Hz, 1H), 1.91 (dd, 3J 5 8.9, 3.7 Hz, 2H), 1.76 (td, 3J 5 13.7, 4.5 Hz,
1H), 1.58 (m, 12H), 1.31 (m, 5H), 1.05 (m, 4H), 0.98 (d, 4H), 0.92 (dd,
6H), 0.85 (t, 3J 5 6.5 Hz, 3H), 0.77 (d, 3H), 0.72 (m, 4H). 13C-NMR
(100 MHz, CDCl3) d 178.2 (C-28), 138.3 (C-13), 125.8 (C-12), 79.2
(C-3), 55.4 (C-5), 53.1 (C-18), 51.6 (C-A), 48.3 (C-17), 47.8 (C-9), 42.2
(C-14), 39.7 (C-8), 39.3 (C-19), 39.1 (C-20), 38.9 (C-4), 38.8 (C-1), 37.2
(3b)-3-[(2E)-but-2-enoyloxy]ursen-12-en-28-oic acid (5)
UA 120.6 mg (0.364 mmol), DCM (2 mL), DMF (1 mL), K2CO3 72.6 mg
(0.526 mmol), crotonyl chloride 40 lL (0.288 mmol), 3 h reaction. FCC
(n-hexane/ethyl acetate 90:10 v/v) affording 33.2 mg product, 24%
yield, TLC: Rf 5 0.52 (n-hexane/acetone 80:20 v/v); white powder; m.p.

ꢀ
ꢀ
4
GUZMAN-AVILA ET AL.
|
189–1928C. 1H-NMR (400 MHz, CDCl3) d 6.95 (dd, 3J 5 15.5, 6.9 Hz,
1H), 5.85 (dd, 3J 5 15.5, Hz, 1H), 5.24 (t, 3J 5 3.3 Hz, 1H), 4.55 (m,
1H), 1.87 (m, 5H), 1.68 (m, 6H), 1.55 (m, 4H), 1.34 (m, 5H), 1.25 (s, 3H),
1.17 (d, 1H), 1.08 (s, 4H), 0.96 (m, 7H), 0.87 (m, 10H), 0.78 (m, 3H).
13C-NMR (100 MHz, CDCl3) d 183.6 (C-28), 166.6 (C-A), 144.2 (C-C),
138.2 (C-13), 126.0 (C-12), 123.6 (C-B), 80.8 (C-3), 55.5 (C-5), 52.8 (C-
18), 48.2 (C-17), 47.7 (C-9), 42.1 (C-14), 39.7 (C-8), 39.2 (C-19), 39.0
(C-20), 38.5 (C-4), 38.1 (C-1), 37.1 (C-10), 37.0 (C-22), 33.1 (C-7), 30.7
(C-21), 29.9 (C-23), 28.3 (C-15), 24.3 (C-13), 24.1 (C-16), 23.8 (C-27),
23.5 (C-11), 21.4 (C-30), 18.4 (C-16), 17.3 (C-D), 17.2 (C-26), 17.0 (C-
29), 16.9 (C-25), 15.8 (C-24).
with KOH 0.1M (4 mL) at the final time. The released p-nitrophenolate
ion was determined by reading the absorbance at 400 nm
(E 5 18,000 M21 cm21).
To verify if compounds were reversible inhibitors, aliquots of PTP-
1B were incubated in presence of at least 25-fold molar excess of
inhibitor for 1 h at 378C. Control experiments were performed adding
DMSO instead of inhibitor. After this interval time, the enzyme solu-
tions were diluted 400-fold and the residual enzyme activity was
assayed.
|
2.4 Animals
Male CD1 mice weighing 30–40 g body weight were housed at stand-
ard laboratory conditions and fed with a rodent pellet diet and water
ad libitum. They were maintained at room temperature and a photoper-
iod of 12 h day/night cycle. Animals described as fasted were deprived
of food for 16 h but had free access to water. All animals’ procedures
were conducted in accordance with Federal Regulations for Animals
Experimentation and Care (SAGARPA, NOM-062-ZOO-1999),
and approved by the Institutional Animal Health Care and Use Commit-
tee based on US National for Health publication (No. 85-23, revised
1985).
(3b)-ursen-12-en-3,28-diol (7)
A cooled solution of LiAlH4 (180 mg, 4.743 mmol) in dry THF (10 mL)
was added dropwise to a mixture of UA (500 mg, 1.085 mmol) and dry
THF (5 mL) under N2 atmosphere. The mixture reaction was stirred at
08C for 1 h. The reaction was refluxed 7 h. Reaction was quenched by
addition of NaOH 10% (10 mL). Precipitate was filtered and washed
with AcOEt (3 3 5 mL). The combined organic layers were dried over
anhydrous Na2SO4, and evaporated under reduced pressure. FCC: (n-
hexane:2-propanol, 98:02) affording 73.6 mg product, 15% yield. TLC:
Rf 5 0.47 (n-hexane: 2-propanol, 95:05); white powder; m.p. 218–
2218C [Lit. 222–2248C] (Nascimento et al., 2014). 1H-NMR (400 MHz,
CDCl3) d 5.24 (t, 3J 5 3.7 Hz, 1H), 4.47 (dd, 3J 5 9.2, 3.4 Hz, 1H), 3.37
(m, 1H), 3.21 (m, 1H), 2.23 (m, 1H), 2.03 (m, 1H), 1.90 (m, 2H), 1.60 (m,
11H), 1.34 (m, 5H), 1.26 (s, 2H), 1.17 (s, 1H), 1.09 (m, 3H), 1.04 (s, 1H),
0.96 (m, 11H), 0.86 (m, 3H), 0.81 (s, 2H), 0.78 (m, 4H), 0.72 (m, 1H).
13C-NMR (100 MHz, CDCl3) d 138.14 (C-13), 126.1 (C-12), 79.3 (C-3),
77.4 (C-28), 55.4 (C-5), 52.9 (C-18), 48.1 (C-17), 47.8 (C-9), 42.2 (C-14),
39.7 (C-8), 39.3 (C-19), 39.0 (C-20), 38.8 (C-1), 37.2 (C-10), 36.9 (C-22),
33.2 (C-7), 30.8 (C-21), 28.4 (C-23), 28.2 (C-15), 27.4 (C-2), 24.4 (C-16),
23.8 (C-27), 23.5 (C-11), 21.4 (C-30), 18.5 (C-6), 17.3 (C-26), 17.2 (C-
29), 15.8 (C-25), 15.7 (C-24).
|
2.5 Induction of NIDDM model
Streptozotocin (STZ, Sigma-Aldrich, St. Louis, MO) was dissolved in
citrate buffer (pH 4.5), and nicotinamide (NAD) was dissolved in normal
physiological saline solution. NIDDM was induced in overnight fasted
mice by a single intraperitoneal injection of STZ (100 mg/kg), 15 min
after the i.p. administration of NAD (40 mg/kg). After 10 days of STZ
administration, blood glucose levels in each mouse was determined.
Mice with >140 mg/dL glycaemia were considered diabetics (Hidalgo-
Figueroa et al., 2017).
|
2.6 In vivo antidiabetic assay
|
2.3 Biological assays
ꢀ
A described protocol was used (Chavez-Silva et al., 2018). In brief,
|
2.3.1 Expression and purification of recombinant PTP-1B
fasted diabetic mice were divided into groups of six animals each. The
groups were orally administered with a suspension of compounds 2
and 7 prepared in 10% Tween 80 (50 mg/kg). Glibenclamide was used
as hypoglycemic reference drug (3 mg/kg). Control group was treated
with 10% Tween 80. Glucose concentration was determined from mice
tail at 0, 1, 3, 5, and 7 h after administration with a glucometer (Roche
AccuCheck Performa, Mexico City, Mexico). The percentage of glycae-
mia was calculated by comparing a selected time glycaemia (Gx) with
The human PTP-1B enzyme was purified as fusion protein from bacte-
rial cell lines. The complete sequence of PTP-1B was cloned in the
pGEX-2T bacterial expression vector downstream the GST sequence,
and this vector was used to transform E. coli into TB1 strain. The
recombinant fusion protein was purified from bacterial lysate using a
single step affinity chromatography on glutathione-agarose. The solu-
tion with fusion protein was treated with thrombin for 3 h at 378C.
Then the active phosphatase was purified from GST and thrombin by
gel filtration on a Superdex G75 column. The enzyme purity was deter-
mined by SDS-PAGE.
initial value (G0) using the formula: %Variation of glycaemia
5
[(Gx – G0)/G0] 3 100. All values are expressed as mean 6 SEM. Statisti-
cal significance was estimated by analysis of variance (ANOVA),
p < .05 and p < .01 implies significance.
|
2.3.2 PTP-1B inhibition assay
The assays were carried out at 378C. The substrate (pNPP) was dis-
solved in 0.075 M of b,b-dimethyl glutarate pH 7.0 buffer, containing
1 mM EDTA and 1 mM dithiothreitol. The final volume was 1 mL. The
reactions were started by adding aliquots of the enzyme and stopped
|
2.7 Docking
Compounds 1-7 were docked with the crystallographic structure of
PTP-1B obtained from Protein Data Bank (PDB code 1C83) resolution

ꢀ
ꢀ
GUZMAN-AVILA ET AL.
5
|
SCHEME 1 Reagents and conditions for the synthesis of ursolic acid derivatives: (a) base, 08C, CH2Cl2 or THF; then acyl/alkyl halide, MW
irradiation, reflux. (b) K2CO3, 08C, CH2Cl2; then methyl iodide, MW irradiation, reflux
of 1.8, and co-crystallized with 6-(oxalylamino)-1H-indole-5-carboxilic
acid. Docking studies were developed by selecting the A and B sites,
described as important regions for PTP-1B inhibition (Ramírez-Espinosa
et al., 2011). The A site or catalytic site is conformed by Asp181,
Cys215, and Arg221, also Lys120, Ser216, Phe182, Gly220, all of them
are important residues for interaction; meanwhile the region B or
extended binding site is composed by Arg24, Arg254, Gln262, Tyr46,
Asp48, Val49, Ile219, and Met258.
|
3
RESULTS AND DISCUSSION
|
3.1 Chemistry
In our research, seven semisynthetic UA derivatives were prepared
(Schemes 1 and 2), consisting in two ethers and three esters on C-3
hydroxyl group, the methyl ester and the corresponding alcohol from
the reduction of C-28 carboxylic acid group. From them, ester and
ether derivatives (compounds 1–6) were prepared using microwave
(MW) assisted synthesis procedure according to reports for some olea-
Docking calculations were conducted with AutoDock 4.2, the aux-
iliary program AutoGrid generate the grid maps centered at
X 5 48.136, Y 5 7.931, and Z 5 2.961, and the grid dimensions were
60 3 60 3 60 with points separated by 0.375 for including both
sites.
ꢀ
nolic and moronic acids derivatives (Ceron-Romero et al., 2016;
Ramírez-Espinosa et al., 2014), and a conventional synthesis for com-
pound 7. The 3-acyloxy derivatives 3 and 5 were prepared through the
reaction of UA with the respective acyl chlorides, meanwhile the ethers
derivatives (1 and 2) and the ursolic acid methyl ester (6) were obtained
by reaction with alkyl halide, in the presence of the corresponding
base. Compound 4 was prepared by reaction between DMF in pres-
ence of benzoyl chloride as catalyst through Vilsmeier reaction. These
reactions were subjected to microwave irradiation at 608C. This meth-
odology allowed an improved performance in the reaction times, that
is, less than 3 h, although the yields were variable (Table 1). The deriva-
tive 7 was prepared by a conventional method, through UA reduction
by use of LiAlH4 in reflux conditions for 8 h. All compounds were char-
acterized by NMR spectroscopy (Supporting Information).
Before docking experiments, the protocol was validated by
docking the co-crystalized ligand 6-(oxalylamino)-1H-indole-5-carbox-
ilic acid and obtaining a RMSD of 0.33. Briefly, AutoDock performs
an automated docking of the ligand with user-specified dihedral flex-
ibility within a protein rigid binding site. The program performs sev-
eral runs in each docking experiment, each run provides one
predicted binding mode. All water molecules and 6-(oxalylamino)-1H-
indole-5-carboxilic acid were removed from the crystallographic
structure, and all hydrogen atoms were added. For all ligands and
protein, Gasteiger charges were assigned and nonpolar hydrogen
atoms were merged, all torsions were allowed to rotate during dock-
ing. Lamarkian genetic algorithm was applied for the search using
default parameters.
The number of docking runs was 100, all solutions were clustered
into groups with RMS lower than 1 after docking and the clusters were
ranked by the lowest energy representative of each cluster. In order to
describe the ligand–binding pocket interactions, the top ranked binding
mode found by AutoDock in complex with the selected binding region
of PTP-1B was subject to full energy minimization using MMFF94
force field. PyMOL 1.0 was used to generate the molecular surface of
docking models.
SCHEME 2 Reagents and conditions for the synthesis of
derivative 7

ꢀ
ꢀ
6
GUZMAN-AVILA ET AL.
|
TABLE 1 Summary of ursolic acid derivatives synthesis
Espinosa et al., 2011; Zhang et al., 2006). All of these results correlate
with the complete PTP-1B inhibition of the most active compounds
with their corresponding low IC50 values, and also explain in part, the
reason why the remaining compounds did not show important inhibi-
tion values over PTP-1B. Also, in a previous work with oleanolic acid
derivatives, we described that is necessary the presence of the carbox-
ylic acid and/or its corresponding reduction product carbinol derivative
(H-bond donor) in C-28 to maintain the inhibitory activity (Ramírez-
Espinosa et al., 2014). Thus, the fact that introducing a methyl ester in
C-28 in UA to obtain compound 6, led to a weaker inhibition of enzy-
matic activity in current investigation.
MW reaction
Derivative
Base
Alkyl/acyl halide
Iodomethane
time (hours)
Yield
65%
89%
48%
58%
24%
76%
1
2
3
4
5
6
Na
2
2
2
2
3
3
Na
Ethyl bromide
K2CO3
K2CO3
K2CO3
K2CO3
Acetyl chloride
Dimethylformamide
Crotonyl chloride
Iodomethane
Finally, for a better understanding of the type of enzymatic inhibi-
tion, an additional reversibility tests were performed. For that, the
most active derivatives were incubated with an aliquot of PTP-1B dur-
ing one hour. It is known that UA inhibits in a complete reversible man-
ner (Ramírez-Espinosa et al., 2011). From the data in Figure 3, we
argued that all compounds tested were PTP-1B reversible inhibitors,
because the enzyme recovered its complete in vitro activity after com-
|
3.2 In vitro PTP-1B inhibition assay
The inhibitory properties of these compounds against PTP-1B was
screened. The initial enzymatic assays were carried out at fixed concen-
tration of p-nitrophenylphosphate (2.5 mM) in the presence of 25 lM
of each UA derivative. As can be seen in Figure 1, compounds 3, 5, and
7 were the most active; in fact, they strongly inhibit 95–99% of the
enzyme activity respect to the control test DMSO; meanwhile, com-
pound 2 was a medium inhibitor (<60%). On the other hand, com-
pounds 1, 4, and 6 were weaker inhibitors with enzyme inhibition
values minor than 30%.
pounds were eliminated, showing
a similar performance as that
reported for the parent compound UA (Ramírez-Espinosa et al., 2011).
The potential of some triterpenic acids as promising PTP-1B inhibi-
tors has been reported; however, these showed two main disadvan-
tages: (1) they are not selective on this target, since these compounds
inhibit other phosphatases, sharing 50–80% of homology with PTP-1B
in their catalytic site, which represents a great challenge for to develop
selective compounds; (2) they have not the same potency/activity as
the parent compound UA, which also has more selectivity to PTP-1B
over other phosphatases (Zhang et al., 2006). Owing to this, some UA
derivatives have been developed in order to achieve better com-
pounds. In our case, we found that three of the seven synthesized
compounds showed important activity over PTP-1B with low IC50 val-
ues; also, the inhibition caused by these derivatives was reversible. This
is important since the irreversible inhibition of PTP-1B may cause over-
stimulation of insulin receptor, which finally can lead the desensitiza-
tion of the receptor and/or side effects such as a drastic hypoglycemic
effect.
In order to obtain the IC50 value for each UA derivative, the enzy-
matic activity of PTP-1B in the presence of increasing concentrations
of UA derivatives was determined (Figure 2). The normalized residual
activity values were reported in graphic versus inhibition concentra-
tions and fitted by Equation 1.
Max2Min
y5
slope 1Min
(1)
ꢀ
ꢁ
x
11
IC50
From that, it was found that 3, 5, and 7 were the most potent
compounds (Figure 2), with IC50 values minor to 10 lM (Table 2);
which allowed us to classify these compounds as hits, meanwhile com-
pound 2 IC50 was around 20 lM. The remaining compounds exhibited
IC50 values higher than 100 lM (graphics not shown). So, these values
are similar to that described for UA (IC50 5 3.08 lM) (Ramírez-
To date, only one of the UA derivatives with significant antidia-
betic effect has been reported, named UA0713, mediating PTP-1B
inhibition (Zhang et al., 2006). However, this compound was not a
selective enzymatic inhibitor, because it displayed similar IC50 values
among PTP-1B and other phosphatases.
|
3.3 Molecular docking with PTP-1B
In order to gain insight into the putative binding mode of compounds
1–7 with PTP-1B, their structures were docked with a crystallographic
structure of human PTP-1B, which was obtained from the Protein Data
Bank (accession code 1C83). Before docking the UA derivatives, the
protocol was validated by predicting the binding mode of 6-(oxalyl-
amino)-1H-indole-5-carboxilic acid, a competitive inhibitor of PTP-1B.
Figure 4 shows a comparison between binding mode of the crystallo-
graphic ligand and the mode predicted by AutoDock. This figure clearly
shows that AutoDock successfully predicted the binding mode of
FIGURE 1 PTP-1B inhibition activity assay of UA derivatives at
25 lM. The data reported represent the mean 6 SEM. All
experiments were performed in triplicate

ꢀ
ꢀ
GUZMAN-AVILA ET AL.
7
|
FIGURE 2 IC50 determination for derivatives 2, 3, 5, and 7. These values were determined by plotting the data relative to the residual
activity of PTP-1B against the inhibitor concentration. 10–15 different inhibitor concentrations were used for every derivative. All assays
were performed in quadruplicate. Data represent the mean 6 SEM [Color figure can be viewed at wileyonlinelibrary.com]
crystallographic ligand with a root-mean square deviation (RMSD) of
0.33 Å. Predicted Binding energies, Ki’s and binding sites determined
by AutoDock for the seven compounds are summarized in Table 3. As
it can be seen, compounds 3 and 5 showed the better D8G and Ki’s cal-
culated, which clearly correlates with its binding location that is the cat-
alytic site. Moreover, both 3 and 5 were two of the three most in vitro
active compounds of the entire series. Therefore, there exist an
important correlation between docking prediction interactions with in
vitro inhibitory activity on PTP-1B. In this context, the remaining com-
pounds interacted with the extended B-binding site, and their D8G and
Ki’s calculated were less than the most active compounds 3 and 5, sub-
sequently, the in vitro inhibitory activity shown by 1, 2, 4, and 6 were
weaker or inactive. With exception of compound 7 that showed
TABLE 2 Assay for reversibility of inhibition and IC50 values against
PTP-1B
Compound
% Reversibility
100.0 6 9.8
n.t.
IC50 (lM)
–
DMSO
1
2
3
4
5
6
7
>100
94.3 6 0.5
96.3 6 5.0
n.t.
20.6 6 0.2
5.6 6 0.2
>100
99.8 6 1.0
n.t.
4.7 6 0.2
>100
FIGURE 3 Inhibition reversibility test. PTP-1B aliquots were incu-
bated in presence of fixed concentration of each compound for 1 h
at 378C, then the enzyme was diluted with assay solution to mea-
sure the residual activity [Color figure can be viewed at wileyonli-
nelibrary.com]
88.3 6 1.1
4.6 6 0.2
The data reported represent the value 6 SEM of triplicates.
n.t., not tested.

ꢀ
ꢀ
8
GUZMAN-AVILA ET AL.
|
FIGURE 4 Surface representation of the binding sites of PTP-1B.
Catalytic binding site or site A, is shown in red, extended binding
site B is in blue. A comparison between the binding position of the
co-crystal ligand (gray) and the binding position predicted by Auto-
Dock (pink) is also shown (RMSD 5 0.33 Å) [Color figure can be
viewed at wileyonlinelibrary.com]
interactions with extended site B with D8G 5 –6.43 Kcal/mol and
Ki 5 18 lM, however, was one of the most active compound in in vitro
inhibition. Further in silico experiments are necessary to clarify this
result.
FIGURE 5 3D binding mode of compound 3 [Color figure can be
viewed at wileyonlinelibrary.com]
|
3.4 In vivo antidiabetic effect
To illustrate different predicted ways in which compound 3 (cata-
lytic site) and 7 (extended binding site) are interacting with PTP-1B, in
Figures 5–8 are shown their interactions. Docking score of derivatives
with PTP-1B active site A and extended binding site B, showed that 3
(Figures 5 and 7) had polar and Van der Waals interactions in both sites
with Lys120, Tyr46, Ser216, Ala217, Ile219, Asp181, Phe182, Gln262,
Val49, Met258, and Gly259, showing a docking score value of
27.48 kcal/mol, being more specific for site B, explaining the reversible
inhibition of PTP-1B. On the other hand, Compound 7 (Figures 6 and
8) showed polar interaction with Gln262 and Van der Waals with
Ala217, Phe182, Ile219, Arg45, Tyr46, Arg47, Asp48, and Val49 with a
predictive docking score of 26.43 kcal/mol, suggesting that the poten-
tial binding site could be localized in an extended binding pocket (site
B) adjacent to the catalytic site A. The fact that compound 7 showed
interactions with extended binding site B, and the docking score value
could explain the lesser experimental PTP-1B inhibition percentage in
comparison to compound 3.
Based on the PTP-1B IC50 values and inhibitory pattern, compounds 2
and 7 were selected to be evaluated through acute evaluation, on
experimental noninsulin diabetic mice model (Figure 9). Compound 7
was selected because of its potent IC50 value and its pattern as a nearly
complete PTP-1B inhibitor, whilst derivative 2 was tested since its
PTP-1B inhibition was moderated, in order to prove the potential cor-
relation between in vitro and in vivo experiments.
The acute administration of 50 mg/kg of 7 showed a significant
decrease in the variation of glycaemia (p < .05) from the first hour, and
this effect was sustained until the end of the assay (7 h). This hypogly-
cemic effect over 70% was similar to the control drug glibenclamide,
TABLE 3 Predicted binding site, docking score, and affinity con-
stant for UA and compounds 1–7 over PTP-1B
Compound
Binding site
D8G (kcal/mol)
27.15
Ki (lM)
5.49
UA
1
Allosteric (extended)
Allosteric (extended)
Allosteric (extended)
Catalytic
26.82
9.96
2
26.72
11.88
3.14
3
27.48
4
Allosteric (extended)
Catalytic
27.03
7.0
5
28.23
0.88
6
Allosteric (extended)
Allosteric (extended)
27.16
5.67
FIGURE 6 3D binding mode of compound 7 [Color figure can be
viewed at wileyonlinelibrary.com]
7
26.43
18.06

ꢀ
ꢀ
GUZMAN-AVILA ET AL.
9
|
FIGURE 7 Two-dimensional interaction diagram of the predicted binding mode of 3 with PTP-1B [Color figure can be viewed at wileyonli-
nelibrary.com]
and better than the parent compound UA. Meanwhile, compound 2 did
not show a significant effect within the three first hours of the acute
assay; however, an important hypoglycemic effect was determined
after fifth hour and until the end of the experiment, achieving a hypo-
glycemic effect value around 55%.
explored that possibility yet. A similar approach has been reported with
a different compound by Boutselis, Yu, Zhang, and Borch (2007). It is
important to show that both compounds had better pattern that the
parent compound UA, and this poor hypoglycemic effect over this
NIDDM model could be due to poor gut absorption.
From these results, we conclude that derivative 7 presents a better
in vivo acute hypoglycemic effect in comparison with 2. This observa-
tion can be explained by two points: (1) the derivative 7 IC50 value was
lower than 2, which means a better potency and inhibition of PTP-1B
by compound 7, and therefore better sensibility to insulin action in the
corresponding treated animals; and (2) it is proposed that 2 could be a
prodrug, that is why we mentioned that the delayed effect (i.e., after
3 h of administration an effect was observed) could be due if 2 acts as
a prodrug, this because after metabolism it could be transformed into
UA, and then UA exerts its antidiabetic action; however, we have not
|
4
CONCLUSION
We prepared seven ursolic acid semisynthetic derivatives. Six com-
pounds were prepared by microwave-assisted synthesis method, which
allowed to improve reaction times although yields were variable. Com-
pounds 3, 5, and 7 showed significant inhibitory activity on PTP-1B
enzyme in a reversible manner. The most active compound was 7
showing important in vitro and in vivo effects. Furthermore, acetyl and
crotonyl esters were the most active derivatives through in vitro
FIGURE 9 Percentage of variation of blood glucose concentration
on noninsulin dependent diabetic mice model treated with UA, 2 or
7. Values represent the mean 6 SEM (n 5 6). *p < .05 compared
with control group [Color figure can be viewed at
wileyonlinelibrary.com]
FIGURE 8 Two-dimensional interaction diagram of the predicted
binding mode of compound 7 with PTP-1B [Color figure can be
viewed at wileyonlinelibrary.com]

ꢀ
ꢀ
10
GUZMAN-AVILA ET AL.
|
moronic acid derivatives. Bioorganic Medicine & Chemical Letters, 26,
2018–2022.
studies; however, further experiments are necessary to determine the
in vivo effects for these compounds. The molecular docking studies
showed that acetyl and crotonyl derivatives had better binding scores
compared to the parent compound, ursolic acid.
ꢀ
ꢀ
ꢀ
ꢀ
Chavez-Silva, F., Ceron-Romero, L., Arias-Duran, L., Navarrete-Vazquez,
ꢀ
ꢀ
G., Almanza-Perez, J., Roman-Ramos, R., . . . Estrada-Soto, S. (2018).
Antidiabetic effect of Achillea millefollium through multitarget interac-
tions: a-glucosidase inhibition, insulin sensitization and insulin secre-
tagogue activities. Journal of Ethnopharmacology, 212, 1–7.
ACKNOWLEDGMENTS
ꢀ
Colla, A. R. S., Oliveira, A., Pazini, F. L., Rosa, J. M., Manosso, L. M.,
This study was supported by Consejo Nacional de Ciencia y Tecno-
Cunha, M. P., & Rodrigues, A. L. S. (2014). Serotonergic and norad-
renergic systems are implicated in the antidepressant-like effect of
ursolic acid in mice. Pharmacology Biochemical & Behaviour, 124, 108–
116.
ꢀ
ꢀ
logía (CONACyT) Proyecto de Ciencia Basica 167044. R. Guzman-
ꢀ
Avila is grateful with CONACyT for the Pharm. M. fellowship grant.
ꢀ
ꢀ
Taken in part from the Master in Pharmacy of R. Guzman-Avila.
Collins, D. O., Ruddock, P. L. D., Chiverton, J., Grasse, D., Reynolds, W.
F., & Reese, P. B. (2002). Microbial transformation of cadina-4,10
(15)-dien-3-one, aromadendr-1(10)-en-9-one and methyl ursolate by
Mucor plumbeus ATCC 4740. Phytochemistry, 59, 479–488.
CONFLICT OF INTEREST
The authors report no conflicts of interest.
ꢀ
Hidalgo Figueroa, S., Navarrete-Vazquez, G., Estrada-Soto, S., Giles-Rivas,
ꢀ
ꢀ
ꢀ
D., Alarcon-Aguilar, F. J., Leon-Rivera, I., . . . Almanza-Perez, J. (2017).
Discovery of new dual PPAR g-GPR40 agonists with robust antidia-
betic activity: Design, synthesis and in combo drug evaluation. Bio-
medical Pharmacotherapy, 90, 53–61.
AUTHOR CONTRIBUTIONS
Study design, coordination and in vivo studies: S. Estrada-Soto, R.
ꢀ
Villalobos-Molina, L. Ceron-Romero; J.J. Ramírez-Espinosa; in vitro
Innocente, A., Casanova, B. B., Klein, F., Lana, A. D., Pereira, D., Muniz,
M. N., . . . Gnoatto, S. C. B. (2014). Synthesis of isosteric triterpenoid
derivatives and antifungal activity. Chemical Biology & Drug Design,
83, 344–349.
PTP-1B inhibition: G. Camici and P. Paoli; preparation of the UA
derivatives, purification and structural elucidation; V. Flores-Morales;
ꢀ
ꢀ
M.Y. Rios and R. Guzman-Avila; in silico studies G. Navarrete-
ꢀ
Jang, S. M., Kim, M. J., Choi, M. S., Kwon, E. Y., & Lee, M. K. (2010).
Inhibitory effects of ursolic acid on hepatic polyol pathway and glu-
cose production in streptozotocin-induced diabetic mice. Metabolism,
59, 512–519.
ꢀ
ꢀ
Vazquez; S. Hidalgo-Figueroa and R. Guzman-Avila; preparation and
ꢀ
ꢀ
writing of the manuscript R. Guzman-Avila, S. Estrada-Soto, and R.
Villalobos-Molina; Finally, all authors contributed to the writing and
revised the manuscript.
Johnson, T. O., Ermolieff, J., & Jirousek, M. R. (2002). Protein tyrosine
phosphatase 1B inhibitors for diabetes. Nature Review & Drug Discov-
ery, 1, 696–709.
ORCID
Jung, S. H., Ha, Y. J., Shim, E. K., Choi, S. Y., Jin, J. L., Yun-Choi, H. S., &
Lee, J. R. (2007). Insulin-mimetic and insulin-sensitizing activities of a
pentacyclic triterpenoid insulin receptor activator. Biochemical Journal,
250, 243–250.
Virginia Flores-Morales
Samuel Estrada-Soto
http://orcid.org/0000-0003-4540-1011
http://orcid.org/0000-0003-1860-0744
ꢀ
Gabriel Navarrete-Vazquez
http://orcid.org/0000-0001-5344-4762
Klok, M. D., Jakobsdottir, S., & Drent, M. L. (2007). The role of leptin
and ghrelin in the regulation of food intake and body weight in
humans: A review. Obesity Review, 8, 21–34.
REFERENCES
ꢀ
Aguirre-Crespo, F., Vergara-Galicia, J., Villalobos-Molina, R., Lopez-Guer-
Lee, J., Lee, H., Seo, K., Cho, H. W., Kim, M., Park, E., & Lee, M. (2014).
Effects of ursolic acid on glucose metabolism, the polyol pathway
and dyslipidemia in non-obese type 2 diabetic mice. Indian Journal of
Experimental Biology, 52, 683–691.
ꢀ
rero, J. J., Navarrete-Vazquez, G., & Estrada-Soto, S. (2006). Ursolic
acid mediates the vasorelaxant activity of Lepechinia caulescens via
NO release in isolated rat thoracic aorta. Life Science, 79, 1062–
1068.
Li, Y., Lu, X., Qi, H., Li, X., Xiao, X., & Gao, J. (2014). Ursolic acid induces
apoptosis through mitochondrial intrinsic pathway and suppression of
ERK1/2 MAPK in HeLa cells. Journal of Pharmacology Science, 125,
202–210.
ꢀ
Allouche, Y., Beltran, G., Gaforio, J. J., Uceda, M., & Mesa, M. D. (2010).
Antioxidant and antiatherogenic activities of pentacyclic triterpenic
diols and acids. Food Chemical & Toxicology, 48, 2885–2890.
Boutselis, I. G., Yu, X., Zhang, Z. Y., & Borch, R. F. (2007). Synthesis and
cell-based activity of a potent and selective protein tyrosine phos-
phatase 1B inhibitor prodrug. Journal of Medical Chemistry, 50, 856–
864.
Li, Z., Zhang, W., & Sima, A. A. F. (2007). Diabetes. Biomedical Research,
56, 1817–1824.
Liu, Q., Chen, L., Hu, L., Guo, Y., & Shen, X. (2010). Small molecules from
natural sources, targeting signaling pathways in diabetes. Biochimical
Biophysical Acta, 1799, 854–865.
Brownlee, M. (2001). Biochemistry and molecular cell biology of diabetic
complications. Nature, 414, 813–820.
Machado, D. G., Neis, V. B., Balen, G. O., Colla, A., Cunha, M. P., Dal-
marco, J. B., . . . Rodrigues, A. L. S. (2012). Antidepressant-like effect
of ursolic acid isolated from Rosmarinus officinalis L. in mice: Evidence
for the involvement of the dopaminergic system. Pharmacology Bio-
chemical & Behaviour, 103, 204–211.
Castro, A. J. G., Frederico, M. J. S., Cazarolli, L. H., Mendes, C. P., Breta-
ꢀ
nha, L. C., Schmidt, E. C., . . . Silva, F. R. M. B. (2015). The mechanism
of action of ursolic acid as insulin secretagogue and insulinomimetic
is mediated by cross-talk between calcium and kinases to regulate
glucose balance. Biochemical Biophysical Acta, 1850, 51–61.
Mahapatra, D. K., Asati, V., & Bharti, S. K. (2015). Chalcones and their
therapeutic targets for the management of diabetes: Structural and
pharmacological perspectives. European Journal of Medical Chemistry,
92, 839–865.
ꢀ
Ceron-Romero, L., Paoli, P., Camici, G., Flores-Morales, V., Rios, M. Y.,
Ramírez-Espinosa, J. J., . . . Estrada-Soto, S. (2016). In vitro and in sil-
ico PTP-1B inhibition and in vivo antidiabetic activity of semisynthetic

ꢀ
ꢀ
GUZMAN-AVILA ET AL.
11
|
^
Nascimento, P. G. G., Lemos, T. L. G., Bizerra, A. M. C., Arriaga, A. M. C.,
Ferreira, D. A., Santiago, G. M. P., . . . Costa, J. G. M. (2014). Antibac-
terial and antioxidant activities of ursolic acid. Molecules, 19, 1317–
1327.
Somova, L. O., Nadar, A., Rammanan, P., & Shode, F. O. (2003a). Cardiovas-
cular, antihyperlipidemic and antioxidant effects of oleanolic and ursolic
acids in experimental hypertension. Phytomedicine, 10, 115–121.
Somova, L. I., Shode, F. O., Ramnanan, P., & Nadar, A. (2003b). Antihy-
pertensive, antiatherosclerotic and antioxidant activity of triterpe-
noids isolated from Olea europaea, subspecies africana leaves.
Journal of Ethnopharmacology, 84, 299–305.
Puius, Y. A., Zhao, Y., Sullivan, M., Lawrence, D. S., Almo, S. C., & Zhang,
Z. Y. (1997). Identification of a second aryl phosphate-binding site in
protein-tyrosine phosphatase 1B: a paradigm for inhibitor design. Pro-
ceedings of National Academic Sciences USA, 94, 13420–13425.
Tahrani, A. A., Bailey, C. J., Del Prato, S., & Barnett, A. H. (2011). Man-
agement of type 2 diabetes: New and future developments in treat-
ment. Lancet, 378, 182–197.
ꢀ
ꢀ
Ramírez-Espinosa, J. J., Rios, M. Y., Lopez-Martínez, S., Lopez-Vallejo, F.,
Medina-Franco, J. L., Paoli, P., . . . Estrada-Soto, S. (2011). Antidia-
betic activity of some pentacyclic acid triterpenoids, role of PTP–1B:
In vitro, in silico, and in vivo approaches. European Journal of Medical
Chemistry, 46, 2243–2251.
Tonks, N. K. (2003). PTP1B: From the sidelines to the front lines! FEBS
Letters, 546, 140–148.
Verma, M., Gupta, S. J., Chaudhary, A., & Garg, V. K. (2017). Protein
tyrosine phosphatase 1B inhibitors as antidiabetic agents: A brief
review. Bioorganic Chemistry, 70, 267–283.
Ramírez-Espinosa, J. J., Rios, M. Y., Paoli, P., Flores-Morales, V., Camici,
G., La Rosa-Lugo, V. D., . . . Estrada-Soto, S. (2014). Synthesis of olea-
nolic acid derivatives: In vitro, in vivo and in silico studies for PTP-1B
inhibition. European Journal of Medical Chemistry, 87, 316–327.
Wu, P. P., Zhang, K., Lu, Y. J., He, P., & Zhao, S. Q. (2014). In vitro and in
vivo evaluation of the antidiabetic activity of ursolic acid derivative.
European Journal of Medical Chemistry, 80, 502–508.
ꢀ
ꢀ
Rios, M. Y., Lopez-Martínez, S., Lopez-Vallejo, F., Medina-Franco, J. L.,
Villalobos-Molina, R., Ibarra-Barajas, M., . . . Estrada-Soto, S. (2012).
Vasorelaxant activity of some structurally-related triterpenic acids
from Phoradendron reichenbachianum (Viscaceae) mainly by NO pro-
duction: ex vivo and in silico studies. Fitoterapia, 83, 1023–1029.
Zhang, W., Hong, D., Zhou, Y., Zhang, Y., Shen, Q., Li, J.-Y., . . . Li, J.
(2006). Ursolic acid and its derivative inhibit protein tyrosine phos-
phatase 1B, enhancing insulin receptor phosphorylation and stimulat-
ing glucose uptake. Biochimical Biophysical Acta, 1760, 1505–1512.
Rollinger, J. M., Kratschmar, D. V., Schuster, D., Pfisterer, P. H., Gumy,
C., Aubry, E. M., . . . Odermatt, A. (2010). 11b-Hydroxysteroid dehy-
drogenase 1 inhibiting constituents from Eriobotrya japonica revealed
by bioactivity-guided isolation and computational approaches. Bioor-
ganic Medical Chemistry, 18, 1507–1515.
SUPPORTING INFORMATION
Additional Supporting Information may be found online in the sup-
porting information tab for this article.
Romsicki, Y., Reece, M., Gauthier, J. Y., Asante-Appiah, E., & Kennedy, B.
P. (2004). Protein tyrosine phosphatase dephosphorylation of the
insulin receptor occurs in a perinuclear endosome compartment in
human embryonic kidney 293 cells. Journal of Biological Chemistry,
279, 12868–12875.
ꢀ
ꢀ
How to cite this article: Guzman-Avila R, Flores-Morales V,
Paoli P, et al. Ursolic acid derivatives as potential antidiabetic
agents: In vitro, in vivo, and in silico studies. Drug Dev. Res.
2018;00:1–11. https://doi.org/10.1002/ddr.21422
Sheng, H., & Sun, H. (2011). Synthesis, biology and clinical significance
of pentacyclic triterpenes: a multi-target approach to prevention and
treatment of metabolic and vascular diseases. Nature Products &
Reports, 28, 543–593.