Exploring bulky natural and natural-like periphery in the design of p-(benzyloxy)phenylpropionic acid agonists of free fatty acid receptor 1 (GPR40)

Article abstract of DOI:10.1016/j.bioorg.2020.103830

Six derivatives of 3-phenylpropionic acid bearing various natural and natural-like, spatially defined peripheral motifs have been synthesized and evaluated in vitro for free fatty acid receptor 1 (FFA1) activation. Two frontrunner compounds (bearing a bornyl and cytosine groups) were evaluated in an oral glucose tolerance test in mice where both demonstrated the ability to sustain blood glucose levels following a glucose challenge. The bornyl compound displayed a somewhat superior, dose-dependent efficacy and, therefore, can be regarded as a lead compounds for further development as a therapeutic agent for type 2 diabetes mellitus. Its high affinity to FFA1 was rationalized by docking experiments.

Full text of DOI:10.1016/j.bioorg.2020.103830

Journal Pre-proofs
Exploring bulky natural and natural-like periphery in the design of p-(benzy-
loxy)phenylpropionic acid agonists of free fatty acid receptor 1 (GPR40)
Sergey O. Kuranov, Olga A. Luzina, Oleksandra Onopchenko, Iryna Pishel,
Sergey Zozulya, Maxim Gureev, Nariman F. Salakhutdinov, Mikhail
Krasavin
PII:
S0045-2068(20)30504-6
DOI:
https://doi.org/10.1016/j.bioorg.2020.103830
YBIOO 103830
Reference:
To appear in:
Bioorganic Chemistry
Received Date:
Revised Date:
Accepted Date:
3 March 2020
31 March 2020
5 April 2020
Please cite this article as: S.O. Kuranov, O.A. Luzina, O. Onopchenko, I. Pishel, S. Zozulya, M. Gureev, N.F.
Salakhutdinov, M. Krasavin, Exploring bulky natural and natural-like periphery in the design of p-
(benzyloxy)phenylpropionic acid agonists of free fatty acid receptor 1 (GPR40), Bioorganic Chemistry (2020),
doi: https://doi.org/10.1016/j.bioorg.2020.103830
This is a PDF file of an article that has undergone enhancements after acceptance, such as the addition of a cover
page and metadata, and formatting for readability, but it is not yet the definitive version of record. This version
will undergo additional copyediting, typesetting and review before it is published in its final form, but we are
providing this version to give early visibility of the article. 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.
© 2020 Published by Elsevier Inc.

Exploring bulky natural and natural-like periphery in the design of p-
(benzyloxy)phenylpropionic acid agonists of free fatty acid receptor 1
(GPR40)
Sergey O. Kuranov,a Olga A. Luzina,a Oleksandra Onopchenko,b Iryna Pishel,b Sergey
Zozulya,b,c Maxim Gureev,d Nariman F. Salakhutdinov,a and Mikhail Krasavine,*
aN. N. Vorozhtsov Novosibirsk Institute of Organic Chemistry, Siberian Branch of the Russian
Academy of Sciences, Novosibirsk 630090, Russian Federation
bEnamine Ltd, Kyiv 02094, Ukraine
cTaras Shevchenko National University, 62 Volodymyrska, Kyiv 01033, Ukraine
dI.M. Sechenov First Moscow State Medical University, Moscow 119991, Russian Federation
eSaint Petersburg State University, Saint Petersburg 199034 Russian Federation
*Corresponding author. Tel.: +7 931 3617872; fax: +7 812 428 6939. E-mail address:
m.krasavin@spbu.ru (M. Krasavin).
Abstract: Six derivatives of 3-phenylpropionic acid bearing various natural and natural-like,
spatially defined peripheral motifs have been synthesized and evaluated in vitro for free fatty
acid receptor 1 (FFA1) activation. Two frontrunner compounds (bearing a bornyl and cytosine
groups) were evaluated in an oral glucose tolerance test in mice where both demonstrated the
ability to sustain blood glucose levels following a glucose challenge. The bornyl compound
displayed a somewhat superior, dose-dependent efficacy and, therefore, can be regarded as a lead
compounds for further development as a therapeutic agent for type 2 diabetes mellitus. Its high
affinity to FFA1 was rationalized by docking experiments.
Keywords: free fatty acid receptor 1; type 2 diabetes mellitus; natural products; natural-likeness;
three-dimensional structure; glucose tolerance.
1.
Introduction
Diabetes mellitus is a chronic metabolic disease affecting more than 400 million people
worldwide, more than 90 per cent of this patient population suffering from type 2 diabetes
mellitus (T2DM) [1]. T2DM is a complex pancreatic disorder characterized by impaired
carbohydrate metabolism, high blood sugar levels, as well as insulin resistance and/or insulin
deficiency [2]. The impact of T2DM is exacerbated by such complications as nephropathy [3],
retinopathy [4], atherosclerosis [5] and cardiovascular disease [6]. Currently, numerous drugs are

in clinical use against T2DM [7,8], however, their suboptimal efficacy and potential side effects
[9] warrant the development of new therapies.
One of the serious side-effects associated with antidiabetic glucose-lowering drugs in the risk of
hypoglycemia which can be as debilitating condition as hyperglycemia treated by the drug
themselves [10]. Free fatty acids have been discovered as potential targets for treatment of
T2DM and other metabolic diseases [11-12]. In particular, free fatty acid receptor 1 (FFA1, also
known as GPR40) expressed in pancreatic -cells and enteroendocrine cells was shown to
mediate glucose-stimulated insulin secretion (GSIS) [13]. The basal, healthy expression levels of
FFA1 are relatively low; however, they dramatically increase in hyperglycemic state. Activation
of FFA1 at this point by endogenous free fatty acids or therapeutic agonists leads triggers a
signaling cascade which, in turn, increases insulin secretion and ultimately lowers the glucose
levels. Once the latter return to normal, so does the expression of FFA1. Therefore, continued
exposure to FFA1 agonists will have no effect and carry no risk of developing hypoglycemia
[13-14]. Numerous FFA1 agonists have been developed over the last 15 years; these have been
comprehensively summarized in the recent excellent reviews [15-17]. While no drug targeting
FFA1 for diabetes is currently approved for clinical use, several drug candidates are noteworthy
(Figure 1) as they all contain a pharmacophoric 3-phenylpropanoic acid motif responsible for
mimicking the endogenous fatty acid ligands. The molecular periphery around this basic scaffold
is typically optimized so as to increase compounds’ affinity and specificity to the target [17].
O
O
OH
O
OH
O
N
H
S
O
O
Cl
1
, GW9508 EC50 = 0.9 µM
2
, TAK-875 EC50 = 9.1 nM
F
O
O
OH
OH
O
O
O
N
N
3
, LY2881835 EC50 = 164±94 nM
4
, EC50 = 410 nM
Figure 1. Drug candidates targeting FFA1 (at various stages of development).
An eloquent example of such elaborate periphery optimization is provided by the 100-fold
improvement of potency on going from the nearly micromolar agonist 1 or GW9508 (also
suffering from poor oral bioavailability) [18] to the advanced clinical candidate 2 or TAK-875

[19] which demonstrated promising efficacy in phase II clinical trials but failed in phase III due
to idiosyncratic liver toxicity [20].
Two drug candidates – Eli Lilly’s candidate LY2881835 (3) which was discontinued after phase
I clinical trials [21] and early preclinical candidate 4 earlier developed by us [22] – illustrate a
rather promising approach to the design of the 3-phenylpropanoic acid periphery (which is
particularly relevant to the present study), namely, the use of spirocyclic motifs (highlighted in
blue). The potential of spirocyclic motifs to provide a distinctly three-dimensional binding motifs
with high degree of complementarity to the protein target signified them as an emerging type of
privileged structures for drug design [23]. It is perhaps unsurprising that spirocycles frequently
occur in natural products [24] which makes design of new spirocycles highly important from the
standpoint of “natural-likeness” [25] of new chemical libraries for drug discovery.
Our recent success employing natural products (camphor and cytisine) in the design of dipeptidyl
peptidase-IV inhibitors with pronounced in vitro and in vivo hypoglycemic activities [26]
prompted us to consider investigating the applicability of various natural and natural-like
periphery motifs as a hitherto unexplored approach to the design of FFA1 agonists. The general
design idea would be to introduce bulky, spatially defined groups derived from the natural (5-8)
as well as natural-like adamantine (9-10) [27-28] polycyclic amines (Figure 2), very much in line
with the successful employment of the spirocyclic motifs in compounds 3 and 4. Herein, we
report reducing this idea to practice.
NH
NH2
N
R
H2N
H
O
6
(-)-Cytisine (R=H)
3-Nitrocytisine (R=NO2)
5
8
(-)-Bornylamine
Leelamine
7
NH2
NH2
9
10
2-Adamantylamine
1-Adamantylamine
Figure 2. Natural (5-8) and non-natural (9-10) bulky amines for the design of FFA1 agonists.
2. Results and discussion
2.1 Synthesis
The key 3-(p-benzyloxy)phenylpropiponic acid building block 15 was prepared from p-
hydroxybenzaldehyde (11) via condensation with malonic acid followed by decarboxylation
which gave p-hydroxycinnamic acid (12) [29]. The latter was esterified (MeOH/H2SO4) to 13
and reduced at the double bond to give methyl 3-(p-hydroxyphenyl)propionate (14). The

phenolic group was alkylated by 1,4-bis(bromomethyl)benzene (taken in excess to reduce bis-
substitution reaction) to give an excellent yield of the core building block 15. The latter reacted
with six amines shown in Figure 2 in acetonitrile in the presence of DIPEA as an HBr scavenger
to give esters 16a-f in modest to excellent yield. Alkaline hydrolysis of the latter followed by
acidification with hydrochloric acid afforded target compounds 17a-f as hydrochloride salts
(Scheme 1).
O
COOMe
a
c
COOH
b
COOMe
H
HO
HO
HO
HO
14
, 80%
11
12
13
, 92%
, 83%
d
COOH
COOMe
COOMe
f, g
e
HCl
O
O
O
R
R
Br
17a-f
, 40-72%
16a-f
, 39-92%
15
, 84%
H
N
H
N
N
H
N
NH
N
O
N
O
NH
R =
O2N
a
c
f
e
b
d
o
Reagents and conditions: a. CH2(COOH)2, piperidine, pyridine, 65 C; b. H2SO4, MeOH, reflux; c NaBH4,
o
NiCl2·6H2O, MeOH, 0 C; d. p-BrCH2C6H4CH2Br, K2CO3, MeCN, reflux; e. RH (or RH·HCl), DIPEA, CH3CN,
reflux; f. LiOH·H2O, MeOH/THF/H2O; g. HCl, H2O.
Scheme 1. Synthesis of target FFA1 agonists 17a-f.
2.2. In vitro biological evaluation
Compound 17a-f were tested for FFA1 activation in Chinese hamster ovary (CHO) cells stably
overexpressing the receptor, using calcium flux essay (Table 1). To our delight, four out of six
compounds (17a-d) displayed a full and dose-dependent agonism towards FFA1 as assessed
relative to maximum efficacy relative to GW9508 (taken as a reference in 10 M concentration).
Interestingly, compound 17f bearing the leelamine tail turned out to be virtually inactive. The
difference in activity between 1-adamantyl (17e) and 2-adamntyl (17d) compounds is quite
remarkable as well. The former compound, in contrast to the latter, displayed a 2.5-fold higher
potency and full agonism while compound 17e only gave a maximum of 53% FFA1 activation.
Effect of the cytisine core nitration on the potency is noticeably negative (though it did not affect
the maximum efficacy). Gratifyingly, the compounds decorated with bornyl (17a) and cytisinyl
(17b) groups displayed full agonism in the submicromolar concentration ranger, very much in

line with our previous findings in the area of DPP-IV inhibitors [26]. These two frontrunner
compounds were selected for in vivo efficacy experiments described below.
Table 1. In vitro agonistic activity of compounds 17a-f towards FFA1.
COOH
. HCl
O
R
FFA1 EC50
Compound
17a
R
% efficacya
(μM)
NH
108
94
0.75
N
17b
0.86
2.74
N
O
N
17c
116
N
O2N
O
H
N
17d
17e
17f
96
53
12
1.72
5.26
50
NH
H
H
N
a % relative to maximum agonism displayed by GW9508 at 10 M.
2.3 In vivo efficacy evaluation
Based on the high potency and the full agonistic profile displayed by compounds 17a and 17b in
the vitro screening experiments, these compounds have been selected for evaluation in vivo for
hypoglycemic effects in oral glucose tolerance test (oGTT) in mice. To this end, doses escalating
from 1 mg/kg to 25 mg/kg of these compounds were administered orally to C57BL/6NCrl mice
prior to an oral glucose challenge. The blood levels were monitored over 2 h relative to control.
As it is evident from the data, compound 17a slightly reduced the plasma glucose levels at 1
mg/kg. Further increase led to a dose-dependent effect and 25 mg/kg dose of this compound
essentially led to no noticeable blood glucose level increase following the glucose challenge
(Figure 3A). On the contrary, compound 17b proved less efficacious as even 25 mg/kg led to

noticeable (though certainly less significant than that of the control) increase in blood glucose
levels, up to roughly 40% of the control (Figure 3B).
25.00
A
20.00
Control PO
15.00
10.00
5.00
17a, 25 mg/kg
17a, 10 mg/kg
17a, 5 mg/kg
17a, 1 mg/kg
*
*
*
*
* - 5 mg/kg p=0.078
* - 25 mg/kg p=0.035
* - 10 mg/kg p=0.071
* - 5 mg/kg p=0.038
0.00
-30
0
15
30
60
120
25.00
B
20.00
15.00
10.00
5.00
17b, 25 mg/kg
Control PO
17b, 10 mg/kg
17b, 5 mg/kg
17b, 1 mg/kg
*
*
*
*
* - 10 mg/kg p=0.023
* - 25 mg/kg p=0.068
* - 1 mg/kg p=0.067
* - 10 mg/kg p=0.071
0.00
-30
0
15
30
60
120
*Figure 3. Plasma glucose concentration during the oGTT in C57BL/6NCrl mice (2 g/kg
glucose challenge after 6 h fasting) performed after administration of 1-25 mg/kg of compounds
17a (A) and 17b (B). Results are mean ± SE (n = 6). The level of statistical significance was
determined using Student’s two-tailed t-test where the difference between the mean valus for two
populations was considered.

2.4 In silico modeling
In order to rationalize the observed SAR trends for compounds 17a-f, we validated the docking
protocol by performing de novo docking on TAK-875 into the crystal structure of FFA1 with
bound TAK-875 (pdb code 4PHU). To our delight, the docking fully reproduced the binding
mode observed in the crystal structure with RMSD < 2.5 Ȧ (Figure 4A). Since GW9805 was
used as a reference FFA1 agonist in this study, we also docked it into the binding site of FFA1
and we happy to see it fully reproduce the key interactions with the target, in particular, the
variable -stacking interaction with W174/F87, as well as principal hydrogen bonding to
R183/R258(2258)/Y240 residues (Figure 4B).
А
В
Figure 4. (A) Overlay of the docking pose of TAK-875 (yellow) with its crystal structure
(green); (B) Docking of GW9805 into the binding site of FFA1.
Having validated the docking protocol, we proceeded to obtain Glide Score values for the
compounds 17a-f docked into the FFA1 binding site in comparison with reference ligands TAK-
875 and GW9805. Unfortunately, Glide Score alone did not display any correlation with the
observed EC50 values. Therefore, we calculated free energy values (G) associated with the
binding, using MMGBSA method which takes into account solvation energies of the protein-
ligand complexes. To our delight, the G values correlated well with the EC50 values with the
most negative value observed for the most active compound (17a). At the same time the inactive
compound (17f) did not even allow calculating relevant values associated with its binding to the
target as its docking resulted in a number of high-energy poses which were deemed irrelevant
(Table 2).

Table 2. EC50 values, Glide Score and DG values for compounds 17a-f as well as reference
ligands TAK-875 and GW9805 calculated based on docking experiments.
Compound % efficacy FFA1 EC50, μM GlideScore ∆G (kcal/mol)
GW9508
TAK875
17a
100
100
108
94
116
96
0.90
0.009
0.75
0.86
2.74
1.72
5.26
50
-7.45
-10.616
-8.86
-9.70
-9.90
-9.22
-8.92
n/a
-80.69
-87.36
-82.19
-74.27
-73.28
-69.51
-70.51
n/a
17b
17c
17d
17e
53
12
17f
Thus, compounds 17a displayed the free energy of binding closest, among the six compounds
studied in this work, to those of GW9508 and TAK-875. This finding was also commensurate
with the observation that 17a very closely reproduced the docking pose of both GW9807 and
TAK-875 in the FFA1 binding site (Figure 5).
A
B
Figure 5. Overlay of the docking poses of compound 17a (yellow) and (A) GW9805 (green) and
(B) TAK-875 in the FFA1 binding site.

3. Conclusions
We have explored several natural and natural-like peripheral motif in the design of 3-
pnehylpropionic acid-based agonists of free fatty acid receptor 1. The agonistic potency and the
maximum efficacy displayed by the compounds synthesized appeared to be rather sensitive to
the nature of the peripheral moiety. In particular, employing leelamine as a tail fragment,
rendered the compound virtually inactive. The best, submicromolar in vitro activity and full
agonism was displayed by two frontrunner compounds bearing a bornyl and cytisine moieties,
respectively. Evaluation of these compounds at escalating doses in mice for the ability to sustain
blood glucose levels following glucose challenge (i. e., the oral glucose tolerance test)
demonstrated somewhat superior efficacy of the bornyl compound which represents a promising
candidate for further evaluation as a therapeutic agent to treat type 2 diabetes mellitus.
4.
Experimental section
4.1. General experimental
GC: 7820A gas chromatograph (Agilent Tech., USA); flame-ionisation detector; HP-5 capillary
column, helium as carrier gas (flow rate 2 mL/min, flow division 99:1). Optical rotation: polar
1
3005 spectrometer; CHCl3 soln. H and 13C NMR spectra: Bruker spectrometers AV-400 at
400.13 MHz (1H) and 100.61 MHz (13C), AV-600 at 600.30 MHz (1H) and 150.95 MHz (13C) in
CDCl3 or DMSO-d6; chemical shifts δ in ppm relative to residual CHCl3 [δ(CHCl3) 7.26,
δ(CHCl3) 77.00 ppm] or DMSO-d6 [δ (DMSO-d6) 2.50, δ (DMSO-d6) 39.51 ppm], J in Hz. The
1
13
structure of the products was determined by analyzing H and C NMR spectra. HR-MS: DFS
Thermo Scientific spectrometer in a full scan mode (15–500 m/z, 70eV electron impact
ionisation, direct sample administration). Column chromatography was performed on silica gel
(60–200 l, Macherey-Nagel. All the target compounds reported in this paper have the purity of at
least 95%. Spectral and analytical studies were carried out at the Collective Chemical Service
Center of the Siberian Branch of the Russian Academy of Sciences. All chemicals were
analytically pure and were used as received without purification prior to use. Solvents were
purified according to literature procedures [30].
Leelamine 8 and 1-, 2-Adamantylamines 9, 10 were obtained from commercial sources.
4.2 Synthetic organic chemistry
4.2.1. (-)-Bornyl amine (5)

A mixture of (+)-camphor (13 mmol, 1.98 g), hydroxylamine hydrochloride (215 mmol, 1.77 g),
and water (7 ml) was heated to 80°C, and methanol (7 mL) was added to dissolve the camphor.
A solution of sodium acetate trihydrate (3.3 mmol, 4.5 g) in water (4 mL) was added, and the
reaction mixture was heated under reflux at 100°C for 12 h. Upon removal of the methanol under
reduced pressure, the white solid that precipitated was collected by filtration, washed with water
(3x10 ml), and dried in vacuo to afford 1.95 g (90% yields) camphor oxime.
The
sodium
borohydride (70.0 mmol , 2.66 g) was added portionwise to a solution of (1S)-camphor oxime
(12 mmol, 1.95 g) and nickel dichloride hexahydrate (23 mmol, 5.54 g) in anhydrous methanol
(40 ml) at –25°C over a period of 3 h. After completion of the addition, the resulting black slurry
was stirred at this temperature overnight. The reaction mixture was then warmed to room
temperature, and 25% ammonia solution (15 mL) in water (20 mL) was added with vigorous
stirring. The resulting slurry was extracted with diethyl ether (3x70 mL), and the combined
organic layers were washed with brine (20 mL), dried with anhydrous magnesium sulfate,
filtered, and concentrated in vacuo. The crude product was purified by column chromatography
(dichloromethane-methanol 100:0–90:10).
Foamy white solid, 33% yield. 1H NMR (400 MHz, CDCl3)  0.81 (s, 3H), 0.86 (s, 3H), 0.96 (s,
3H), 0.97 - 1.08 (m, 2H), 1.21 - 1.28 (m, 2H), 1.48 - 1.57 (m, 2H), 1.65 - 1.77 (m, 3H), 2.69 (dd,
J = 8.9 Hz and J = 5.1 Hz, 1H).
4.2.2. 3-Nitrocytisine (7)
(–)-Cytisine 6 (0.5 g, 2.6 mmol) was dissolved in conc. H2SO4 (5 mL), treated with NaNO3 (0.45
g, 5.26 mmol), stirred on a magnetic stirrer at room temperature until the starting material
disappeared completely (controlled by TLC), neutralized with anhydrous Na2CO3, and extracted
with ethyl acetate (3x5 mL). The extracts were combined, dried over Na2SO4, and evaporated.
The solid was chromatographed over silicagel (C6H6–MeOH eluent, 9:1).
Yellow crystals, 60% yield. 1H NMR (300 MHz, CDCl3)  1.49 (br.s, 1H), 1.90 - 2.02 (m, 2H),
2.42 (br.s, 1H), 2.96 - 3.16 (m, 5H), 3.93 - 4.03 (m, 1H), 4.17 - 4.26 (m, 1H), 6.11 (d, J = 8.1 Hz,
1H), 8.32 (d, J = 8.1 Hz, 1H).
4.2.3. (E)-3-(4-Hydroxyphenyl)acrylic acid (12)
p-Hydroxybenzaldehyde 11 (1.512 g, 12.3 mmol) and malonic acid (2.83 g, 27.2 mmol) were
placed in a 10 mL flask and dissolved in 6.8 ml of pyridine. Piperidine (0.115 mL, 1.2 mmol)
was added, and the reaction was placed in pre-heated to 72oC bath for 5 days (controlled by TLC
CH2Cl2-ethyl acetate 4:1). Then, the mixture was transferred to a beaker, and 100 mL of water

was added. Concentrated HCl was added drop-wise until pH 2-3. A white precipitate was
collected using a Hirsch funnel and dried under vacuum. White powder, 83% yield.
1H NMR (300 MHz, DMSO-d6)  6.27 (d, J = 15.9 Hz, 1H), 6.78 (d, J = 8.6 Hz, 2H), 7.40 - 7.56
(m, 3H), 9.84 - 10.27 (br.s, 1H)
4.2.4. (E)-Methyl 3-(4-hydroxyphenyl)acrylate (13)
Acid 12 (1.06 g 6.46 mmol) was dissolved in dry methanol. 2-3 drops of conc. sulfuric acid were
added and solution obtained was refluxed for a two days until disappearing of starting material
(controlled by TLC, CH2Cl2-acetic acid 100:1). After completion of the reaction methanol was
evaporated under vacuum. A residue was re-dissolved in ethyl acetate (30 mL) and washed with
water (10 mL), brine (10 mL) and dried over magnesium sulfate. Solvent was evaporated and the
residue was used in following step without further purification.
White-off powder, 92% yield. 1H NMR (400 MHz, CDCl3)  3.81 (s, 3H), 6.30 (d, J = 16.0 Hz,
1H), 6.53 (br.s, 1H), 6.82 - 6.91 (m, 2H), 7.41 (d, J = 8.6 Hz, 2H), 7.64 (d, J = 16.0 Hz, 1H).
4.2.5. Methyl 3-(4-hydroxyphenyl)propanoate (14)
To an ice-bath cooled solution of 13 (0.845 g, 5.1 mmol) and nickel dichloride hexahydrate (0.12
g, 0.5 mmol) in dry methanol (10 mL) a sodium borohydride (0.5 g, 13.2 mmol) was added in
small portion. After completion of reaction (controlled by GC-MS), saturated ammonium
chloride solution (2.5 mL) was added and methanol was evaporated under reduced pressure.
Water (5 mL) and aqueous ammonia (10mL) were added to the residue, and the product was
extracted with ethyl acetate (3x15 ml). The extracts were combined, dried over Na2SO4, and
evaporated. The crude product was purified by column chromatography (hexane-ethyl acetate
7:3).
1
White solid, 0.67 g 80%. H NMR (500 MHz, CDCl3)  2.60 (t, J = 7.70 Hz, 2H), 2.88 (t, J =
7.70 Hz, 2H), 3.67 (s, 3H), 5.55 (br.s, 1H), 6.70 - 6.80 (m, 1H), 7.00 - 7.09 (m, 1H).
4.2.6. Methyl 3-(4-(4-(bromomethyl)benzyloxy)phenyl)propanoate (15)
A solution of 14 (1.15 g, 6.4 mmol) and potassium carbonate (2.55 g, 18 mmol) in acetonitrile
(20 mL) was stirred for 1 hour at reflux. Then stirring was stopped, the reaction mixture was
allowed to cool down and a precipitate was allowed to settle. The solution was decanted into an
addition funnel with additional acetonitrile (40 mL). A solution of p-dibromoxylene (4.80 g, 18.2
mmol) in acetonitrile (20 mL) was charged to the same reaction flask. Then the solution of 14
was slowly added to the reaction flask under reflux for 4 hours. The mixture obtained was

refluxed for additional 2 hours and then was cooled down. A precipitate was filtered of and
washed with chloroform (3x30mL). Combined organic filtrate was evaporated and part of the
unreacted p-dibromoxylene was recrystallized from acetone. The residue was purified by column
chromatography with dry loading (hexane-ethyl acetate 20:1 to 10:1 after separation of
p-dibromoxylene).
White solid, 81% yield.1H NMR (400 MHz, CDCl3)  2.60 (t, J = 7.8 Hz, 2H), 2.89 (t, J = 7.8
Hz, 2H), 3.67 (s, 3H), 4.50 (s, 2H), 5.03 (s, 2H), 6.89 (d, J=8.7 Hz, 2H), 7.12 (d, J = 8.6 Hz, 2H),
7.41 (s, 4H).
4.2.5. General procedure for reaction with amines 16a-f
A solution of bromide 15 (1.18 mmol), amine (1.46 mmol) and N,N-diisopropylethylamine (2.01
mmol) in acetonitrile (7 mL) was refluxed for 2 hours. Ethyl acetate (20 mL) and 5% sodium
hydroxide (10 mL) solution were added. Organic layer was separated and water layer was
extracted with ethyl acetate (2x10 mL). Combined organic layers were washed with water (10
mL), brine (10 mL) and dried over magnesium sulfate. Solvent was evaporated under vacuum
and residue was purified over silica gel column chromatography using chloroform-methanol
100:1.
4.2.5.1.
Methyl
3-(4-(4-(((1R,2R,4R)-1,7,7-trimethylbicyclo[2.2.1]heptan-2-
ylamino)methyl)-benzyloxy)phenyl)propanoate (16a)
24
White solid 76% yield. M.p. 89.7-90.0°C. [α]D -46 (c 0.140, CHCl3). IR (KBr,) ν/cm-1: 831,
1
1012, 1178, 1232, 1516, 1729. H NMR (500 MHz, CDCl3)  0.82 (s, 3H), 0.90 (s, 3H), 1.01 -
1.12 (m, 5H), 1.44 - 1.76 (m, 6H), 2.56 - 2.66 (m, 3H), 2.90 (t, J = 7.8 Hz, 2H), 3.62 (d, J = 13.4
Hz, 1H), 3.67 (s, 3H), 3.79 (d, J = 13.4, 1H), 5.02 (s, 2H), 6.90 (d, J = 8.5 Hz, 2H), 7.12 (d, J =
13
8.5 Hz, 2H), 7.31 - 7.41 (m, 4H). C NMR (126 MHz, CDCl3)  12.2, 20.5, 20.6, 27.3, 30.1,
36.0, 36.8, 38.7, 45.3, 46.7, 48.4, 51.6, 52.2, 66.1, 69.9, 114.8 (2C), 127.5 (2C), 128.3 (2C),
+
129.2 (2C), 132.8, 135.4, 141.1, 157.3, 173.4. HRMS for C28H37O3N1 calcd 435.2768, found
435.2760 [M]+.
4.2.5.2.
Methyl 3-{4-[(4-{[(1S,9R)-6-oxo-7,11-diazatricyclo[7.3.1.0²,⁷]trideca-2,4-dien-
11-yl]methyl}phenyl)methoxy]phenyl}propanoate (16b)
24
Colorless oil, 85% yield. [α]D -168 (c 0.140, CHCl3). IR (neat), ν/cm-1: 798 1142, 1153, 1240,
1
1512, 1547, 1568, 1543, 1734. H NMR (300 MHz, CDCl3)  1.73 - 1.97 (m, 2H), 2.28 - 2.48
(m, 3H), 2.54 - 2.64 (m, 2H), 2.80 - 3.00 (m, 5H), 3.34 - 3.53 (m, 2H), 3.65 (s, 3H), 3.81 - 3.95
(m, 1H), 4.04 - 4.16 (m, 1H), 4.95 (s, 2H), 5.91 (dd, J = 6.9 and J = 1.2 Hz, 1H), 6.48 (dd, J =

9.0 and J = 1.4 Hz, 1H), 6.83 - 6.91 (m, 2H), 7.00 (d, J = 8.0, 2H), 7.07 - 7.14 (m, 2H), 7.22 -
13
7.31 (m, 3H). C NMR (75 MHz, CDCl3)  25.9, 28.0, 30.0, 35.4, 35.9, 49.9, 51.5, 59.9 (2C),
61.6, 69.8, 104.7, 114.8 (2C), 116.5, 127.4 (2C), 128.4 (2C), 129.2 (2C), 132.8, 135.7, 137.7,
+
138.6, 151.2, 157.2, 163.6, 173.3. HRMS for C29H32O4N2 calcd 472.2357, found 472.2362
[M]+.
4.2.5.3.
Methyl
3-{4-[(4-{[(1S,9R)-5-nitro-6-oxo-7,11-diazatricyclo[7.3.1.0²,⁷]trideca-
2,4-dien-11-yl]methyl}phenyl)methoxy]phenyl}propanoate (16c)
24
Orange oil, 82% yield. [α]D -186 (c 0.140, CHCl3). IR (neat), ν/cm-1: 1155, 1238, 1300, 1317,
1511, 1550, 1677, 1733. 1H NMR (400 MHz, CDCl3)  1.82 - 1.96 (2Н, м, Н-24), 2.39 (2Н, д, J
= 8.5), 2.51 (br.s, 1H), 2.59 (t, J = 7.8 Hz, 2H), 2.85 - 3.00 (m, 4H), 3.08 (br s, 1H), 3.43 (d, J =
12.0 Hz, 2H), 3.66 (s, 3H), 3.98 (dd, J = 15.8 Hz and J = 6.4 Hz, 1H ), 4.19 (d, J = 15.7 Hz, 1H),
4.96 (s, 2H), 6.04 (d, J = 8.1 Hz, 1H), 6.88 (d, J = 8.60 Hz, 2H), 6.99 (d, J = 5.4 Hz. 2H), 7.11
13
(d, J = 8.6, 2H), 7.24 - 7.32 (m, 2H), 8.33 (d, J = 8.1 Hz, 1H). C NMR (101 MHz, CDCl3) 
25.1, 27.6, 30.0, 35.8, 36.3, 51.1, 51.4, 59.0, 59.5, 61.5, 69.6, 102.6, 114.8 (2C), 127.4 (2C),
128.3 (2C), 129.1 (2C), 132.8, 134.4, 136.1, 137.1, 137.6, 155.1, 157.2, 160.5, 173.2. HRMS for
+
C29H31O6N3 calcd 517.2207, found 517.2222 [M]+
4.2.5.4.
Methyl
3-{4-[(4-{[(adamantan-1-yl)amino]methyl}phenyl)methoxy]phenyl}-
propanoate (16d)
White solid, 56% yield. M.p. 108.5°C. IR (KBr,) ν/cm-1: 1169, 1178, 1250, 1271, 1516, 1732. 1H
NMR (400 MHz, CDCl3)  1.57 - 1.78 (m, 13H), 2.10 (m, 3H), 2.59 (t, J = 7.7 Hz, 2H), 2.89 (t, J
= 7.7 Hz, 2H), 3.66 (s, 3H), 3.76 (s, 2H), 5.01 (s, 2H), 6.89 (d, J = 8.5 Hz, 2H), 7.10 (d, J = 8.5
Hz, 2H), 7.36 (s, 4H). 13C NMR (101 MHz, CDCl3)  29.6 (3C), 30.0, 35.9, 36.7 (3C), 42.7
(3C), 44.8, 50.9, 51.5, 69.8, 114.8 (2C), 127.5 (2C), 128.5 (2C), 129.2 (2C), 132.7, 135.5, 141.2,
157.2, 173.4. HRMS for C28H35O3N1 + calcd 433.2612, found 433.2610 [M]+.
4.2.5.5.
Methyl
3-{4-[(4-{[(adamantan-2-yl)amino]methyl}phenyl)methoxy]phenyl}-
propanoate (16e)
1
White solid, 69% yield. M.p. 86.3-86.9°C. IR (KBr,) ν/cm-1: 1178, 1244, 1514, 1726. H NMR
(400 MHz, CDCl3)  1.51 (d, J = 12.1 Hz, 4H), 1.64 - 1.76 (m, 4H), 1.77 - 1.96 (m, 7H), 2.04 (d,
J = 12.9 Hz, 2 H), 2.60 (t, J = 7.8 Hz, 2H), 2.79 (br.s, 1H), 2.90 (t, J = 7.8 Hz, 2H), 3.67 (s, 3H),
3.79 (s, 2H), 5.02 (s, 2H), 6.86 - 6.95 (m, 2H), 7.11 (d, J = 8.6 Hz, 2H), 7.32 - 7.45 (m, 4H). 13C
NMR (101 MHz, CDCl3)  27.6, 27.8, 30.1, 31.3 (2C), 32.0 (2C), 35.9, 37.5 (2C), 37.9, 50.7,

51.6, 61.1, 69.9, 114.8 (2C), 127.5 (2C), 128.3 (2C), 129.2 (2C), 132.8, 135.4, 141.1, 157.3,
+
173.4. HRMS for C28H35O3N1 calcd 433.2612, found 433.2604 [M]+.
4.2.5.6.
Methyl 3-(4-{[4-({[(1R,4aS)-1,4a-dimethyl-7-(propan-2-yl)-1,2,3,4,4a,9,10,10a-
octahydrophenanthren-1-yl]amino}methyl)phenyl]methoxy}phenyl)propanoate (16f)
26
White solid, 57% yield. M.p. 67.7°C. [α]D +21 (c 0.048, CHCl3). IR (neat), ν/cm-1: 822, 1014,
1178, 1242, 1514, 1728. 1H NMR (300 MHz, CDCl3)  0.92 (s, 3H), 1.20 - 1.28 (m, 9H), 1.36 -
1.53 (m, 4H), 1.61 - 1.87 (m, 5H), 2.24 - 2.33 (m, 2H), 2.54 (d, J = 11.8 Hz, 1H), 2.62 (t, J = 7.5
Hz, 2H), 2.78 - 2.95 (m, 5H), 3.68 (s, 3H), 3.73 - 3.84 (m, 2H), 5.03 (s, 2H), 6.88 - 6.95 (m, 3H),
7.00 (dd, J = 8.1 Hz and J = 1.6 Hz, 1H), 7.13 (d, J = 8.6 Hz, 2H), 7.19 (m, J = 8.2 Hz, 1H), 7.31
- 7.42 (m, 4H). 13C NMR (126 MHz, CDCl3)  18.8, 18.9, 19.3, 24.0 (2C), 25.4, 30.1, 30.3, 33.4,
35.9, 36.2, 37.1, 37.4, 38.5, 45.2, 51.5, 54.6, 61.0, 69.9, 114.9 (2C), 123.7, 124.3, 126.8, 127.5
(2C), 128.1 (2C), 129.2 (2C), 132.8, 134.8, 135.5, 140.9, 145.4, 147.6, 157.4, 173.4. HRMS for
C38H49O3N+ calcd 567.3707, found 567.3710 [M]+.
4.2.6. General procedure for hydrolysis
To an ice-cooled solution of methyl ester (0.195 mmol) in THF (2 mL) a solution of lithium
hydroxide monohydrate (0.390 mmol) in water (2 mL) was added. After completion of the
reaction, THF was evaporated under reduced pressure. 2M hydrochloric acid was added drop-
wise until pH 2-3. Precipitate was filtered and washed with water.
4.2.6.1.
3-(4-{[4-({[(1R,2R,4R)-1,7,7-trimethylbicyclo[2.2.1]heptan-2-yl]amino}methyl)-
phenyl]methoxy}phenyl)propanoic acid hydrochloride (17a)
26
White solid, 65% yield. M.p. 153.2°C. [α]D -35 (c 0.052, EtOH). IR (KBr,) ν/cm-1: 827, 1244,
1
1383, 1423, 1512, 1709. H NMR (500 MHz, DMSO-d6)  0.72 - 0.82 (m 3H), 0.85 - 1.04 (m,
8H), 1.41 - 1.54 (m, 2H), 1.55 - 1.65 (m, 1H), 1.71 (t, J = 4.1 Hz, 1H), 2.06 - 2.18 (m, 1H), 2.42 -
2.53 (m, 2H), 2.67 - 2.79 (m, 2H), 2.91 (dd, J = 8.5 Hz and J = 5.4, 1H), 4.05 - 4.21 (m, 2H),
5.09 (s, 2H), 6.89 (d, J = 8.5 Hz, 2H), 7.12 (d, J = 8.4 Hz, 2H), 7.47 (d, J = 7.9 Hz, 2H), 7.67 (d,
13
J = 7.9 Hz, 2H), 8.7 (br.s, 2H) 12.05 (br.s, 1H). C NMR (126 MHz, DMSO-d6)  11.8, 19.9,
20.4, 26.3, 29.6, 34.7, 35.7, 36.3, 44.3, 47.0, 48.6, 50.1, 64.2, 68.8, 114.8 (2C), 127.8 (2C), 129.3
+
(2C), 131.0 (2C), 133.2, 138.2, 156.6, 173.9. HRMS for C27H35O3N1 calcd 421.2615, found
421.2610 [M]+. Anal calcd for C27H36ClNO3: Cl 7.74. Found: Cl 8.06.
4.2.6.2.
3-{4-[(4-{[(1S,9R)-6-oxo-7,11-diazatricyclo[7.3.1.0²,⁷]trideca-2,4-dien-11-
yl]methyl}-phenyl)methoxy]phenyl}propanoic acid hydrochloride (17b)

26
White solid, 42% yield. M.p. 184.5°C. [α]D -123 (c 0.132, EtOH). IR (KBr,) ν/cm-1: 1236,
1
1510, 1549, 1647, 1724. H NMR (500 MHz, DMSO-d6)  1.79 - 1.95 (m, 2H), 2.42 - 2.51 (m,
2H), 2.69 - 2.81 (m, 3H), 3.09 - 3.54 (m, 5H), 3.77 - 3.89 (m, 1H), 3.92 - 4.04 (m, 1H), 4.19 -
4.42 (m, 2H), 5.08 (s, 2H), 6.28 (d, J = 6.6 Hz, 1H), 6.42 (d, J = 8.6 Hz, 1H), 6.91 (d, J = 8.6 Hz,
2H), 7.13 (d, J = 8.5, 2H), 7.43 (dd, J = 8.8 Hz and J = 7.1 Hz, 1H), 7.49 (d, J = 7.9 Hz, 2H),
13
7.58 (d, J = 7.9 Hz, 2H), 10.06 (br.s, 1H). C NMR (75 MHz, DMSO-d6)  22.5, 25.8, 29.5,
31.9, 35.6, 48.0, 55.8, 56.1, 59.7, 68.7, 106.5, 114.6 (2C), 116.9 , 127.8 (2C), 128.5, 129.3 (2C),
+
132.1 (2C), 133.2, 138.8, 139.6, 147.5, 156.6, 162.3, 173.8. HRMS for C28H30O4N2 calcd
458.2200, found 458.2195 [M]+. Anal calcd for C28H31ClN2O4: Cl 7.16. Found: Cl 7.37.
4.2.6.3.
3-{4-[(4-{[(1S,9R)-5-Nitro-6-oxo-7,11-diazatricyclo[7.3.1.0²,⁷]trideca-2,4-dien-
11-yl]methyl}phenyl)methoxy]phenyl}propanoic acid hydrochloride (17c)
26
Orange solid, 40% yield. M.p. 115.5°C. [α]D -173 (c 0.060, EtOH). IR (KBr,) ν/cm-1: 1304,
1
1321, 1510, 1552, 1680, 1730. H NMR (300 MHz, DMSO-d6): 1.90 (m, 2H), 2.43 - 2.52 (m,
2H), 2.70 - 2.86 (m, 3H), 3.16 - 3.55 (m, 5H), 3.84 - 4.09 (m, 2H), 4.30 (m, 2H), 5.07 (s, 2H),
6.40 (d, J = 7.8 Hz, 1H), 6.91 (d, J = 8.6 Hz, 2H), 7.14 (d, J = 8.6 Hz, 2H), 7.51 (m, 4H), 8.39 (d,
J = 8.1 Hz, 1H), 9.99 - 10.19 (m, 1H), 11.46 - 12.63 (br.s, 1H). 13C NMR (75 MHz, DMSO-d6)
 21.8, 25.5, 29.5, 32.6, 35.6, 49.1, 54.9, 55.7, 59.7, 68.7, 104.1, 114.6 (2C), 127.8 (2C), 128.3,
129.3 (2C), 132.1 (2C), 133.2, 135.4, 138.0, 138.9, 154.6, 156.3, 156.6, 173.8. Anal calcd for
C28H30ClN3O6: С 62.28; Н 5.60; N 7.78; Cl 6.57. Found: С 62.20; Н 5.64; N 7.77; Cl 6.61.
4.2.6.4.
3-{4-[(4-{[(Adamantan-1-yl)amino]methyl}phenyl)methoxy]phenyl}propanoic
acid hydrochloride (17d)
White solid, 42% yield. M.p. 264.7-268.1°C. IR (KBr,) ν/cm-1: 827, 1250, 1514, 1697. 1H NMR
(400 MHz, DMSO-d6)  1.56 - 1.72 (m, 6H), 1.99 (m, 6H), 2.13 (m, 3H), 2.43 - 2.53 (m, 2H),
2.70 - 2.78 (m, 2H), 4.07 (s, 2H), 5.10 (s, 2H), 6.90 (d, J = 8.5 Hz, 2H), 7.12 (d, J = 8.6 Hz, 2H),
13
7.47 (d, J = 7.9 Hz, 2H), 7.62 (d, J = 8.2 Hz, 2H), 9.18 (br.s, 2H), 12.12 (br.s, 1H). C NMR
(101 MHz, DMSO-d6)  28.5 (3C), 29.5, 35.3 (3C), 35.6, 37.4 (3C), 42.3, 57.0, 68.6, 114.7 (2C),
+
127.7 (2C), 129.3 (2C), 130.5 (2C), 132.2, 133.1, 138.0, 156.5, 173.8. HRMS for C27H35O3N1
calcd 419.2455, found 419.2447 [M]+. Anal calcd for C27H34ClNO3: Cl 7.77. Found: Cl 8.02.
4.2.6.5.
3-{4-[(4-{[(Adamantan-2-yl)amino]methyl}phenyl)methoxy]phenyl}propanoic
acid hydrochloride (17e)
White solid, 42% yield. M.p. 160.6.1°C. IR (KBr,) ν/cm-1: 825, 1244, 1290, 1418, 1462, 1512,
1585, 1705. 1H NMR (300 MHz, DMSO-d6)  1.46 - 1.64 (m, 4H), 1.68 (br.s, 2H), 1.81 (m, 4H),

2.09 - 2.29 (m, 4H), 2.40 - 2.58 (m, 2H), 2.74 (t, J = 7.5 Hz, 2H), 3.13 (br.s, 1H), 4.17 (s, 2H),
5.09 (s, 2H), 6.90 (d, J = 8.5 Hz, 2H), 7.12 (d, J = 8.5 Hz, 2H), 7.47 (d, J = 7.9 Hz, 2H), 7.66 (d,
J = 7.9 Hz, 2H), 9.31 (br.s, 2H), 11.63 - 12.33 (br.s, 1H). 13C NMR (101 MHz, DMSO-d6)
 26.2, 26.4, 28.2 (2C), 29.5, 29.6 (2C), 35.6, 36.3 (2C), 36.7, 47.4, 60.7, 68.7, 114.7 (2C), 127.7
+
(2C), 129.2 (2C), 130.6 (2C), 131.3, 133.1, 138.0, 156.5, 173.8. HRMS for C27H33O3N1 calcd
419.2455, found 419.2447 [M]+. Anal calcd for C27H34ClNO3: Cl 7.77. Found: Cl 8.07.
4.2.6.6.
3-(4-{[4-({[(1R,4aS)-1,4a-Dimethyl-7-(propan-2-yl)-1,2,3,4,4a,9,10,10a-
octahydrophenanthren-1-yl]amino}methyl)phenyl]methoxy}phenyl)propanoic
acid
hydrochloride (17f)
26
1
White solid, 58% yield. M.p. 187.3-189.2°C. [α]D +33 (c 0.157, EtOH). H NMR (400 MHz,
DMSO-d6)  0.90 (s, 3H), 1.05 - 1.19 (m, 9H), 1.21 - 1.76 (m, 8H), 2.24 (d, J = 12.5 Hz, 1H),
2.42 - 2.50 (m, 2H), 2.53 - 2.92 (m, 7H), 4.04 - 4.16 (m, 2H), 5.10 (s, 2H), 6.81 - 6.86 (m, 1H),
6.87 - 6.98 (m, 3H), 7.08 - 7.16 (m, 3H), 7.48 (d, J = 7.9 Hz, 2H), 7.62 (d, J = 7.8 Hz, 2H), 8.74 -
13
9.30 (m, 2H), 11.88 (br.s, 1H). C NMR (101 MHz, DMSO-d6)  17.9, 18.1, 18.3, 23.9 (2C),
24.9, 28.7, 29.5, 32.9, 35.2, 35.5, 35.9, 36.9, 37.3, 43.8, 50.6, 56.1, 68.7, 114.7 (2C), 123.5,
123.8, 126.3, 127.7 (2C), 129.2 (2C), 130.6 (2C), 132.7, 133.1, 134.3, 138.1, 145.1, 146.7,
156.5, 173.8. HRMS for C27H35O3N+ calcd 553.3551, found 553.3550 [M]+. Anal calcd for
C38H48ClNO3: Cl 6.01. Found: Cl 6.16
4.3 In vitro FFA1 activation assay
CHOcells stably expressing human FFA1 (stable CHO-GPR40 line created at Enamine Ltd.)
were seeded (12 500 cells/well) into 384- well black-wall, clear-bottom microtiter plates 24 h
prior to assay. Cells were loaded for 1 h with dye-loading solution containing Fluo-8AM
(Abcam, ab142773), Probenecid (Sigma, P8761), Pluronic F-127 (Sigma, P2443), Tartrazine
(Sigma, T0388), Amaranth (Sigma, A1016) and tested using fluorometric imaging plate reader
(FLIPR Tetra® High Throughput Cellular Screening System, Molecular De- vices Corp.).
Maximum change in fluorescence over base line was used to determine agonist response. A
potent and selective agonist for FFA1 GW9508 (Selleckchem, S8014) was tested with the test
compounds as a positive control. To obtain concentration response curve data nonlinear
regression with variable slope (four paramters) were fitted using GraphPad Prism version 7.03
for Windows, GraphPad Software, La Jolla California USA, www.graphpad.com.
4.4. In silico modeling

The structure of FFA1 receptor was downloaded from RCSB ProteinDataBank (Pdb id: 4PHU).
Using the ProteinPrep Wizard module within Schrödinger Suite, errors contained in the pdb file
(such as missing amino acids, incorrect bonds, irrelevant protonation) were corrected and the
structures were optimized using the Ramachandran plot and minimized for energy.
4.5. In vivo studies
4.5.1. Animals
Male C57BL/6NCrl mice were maintained on a 12:12-h light-dark cycle under standard animal
housing conditionswith temperature and humidity control. All animal procedures were
performedin accordance with the approved standard procedures and animal care guidelines of
Bioethics Commission of the Educational and Scientific Center "Institute of Biology" at
Shevchenko National University, Kyiv, Ukraine. The“3Rs”principles were applied for
conducting all animal studies.
4.5.2. Administration of test compounds
All compounds were formulated in Labrasol® (Gattefosse, France) solution in saline (50%-50%,
v/v). Compounds were administered to mice by oral gavage using 20 gastainless steel animal
feeding tube (InstechLaboratories, Inc., USA), at a volume dose of 5 ml/kg. Mice from the
control groups received the same volume of the vehicle alone. All mice were dosed at 30 min
before starting the OGTT test. GW9508 (Cayman Chemical) was used as reference compound at
the dose of 100 mg/kg.
4.5.3. Determination of blood glucose levels
Determining the level of blood glucose was performed after 6-hour fasting. On Call Plus
glucometer (Acon Laboratories, Inc., USA) and specific test strips (REF G133-111) were used
for determining glucose level. Tail vein blood sampling was done by incision of the tail tip; 5-6
μl of blood was used for each glucose assay.
4.5.3. Glucose tolerance test
For glucose tolerance test (OGTT), mice were orally treated with 2 g/kg of glucose in a volume
of 10 ml/kg after a 6-hour fasting. Glucose measurements were performed right before the
compound treatment (point “-30 min”) as well as immediately before and at 15, 30, 60 and 120
min after glucose administration.

Acknowledgements
Authors would like to acknowledge the Multi-Access Chemical Service Center of the Siberian
Branch of the Russian Academy of Science for spectral and analytical measurements.
Supplementary data
Supplementary data associated with this article can be found, in the online version, at
http://dx.doi.org/xxx.
References
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
WHO: 10 Facts of Diabetes. https://www.who.int/features/factfiles/diabetes/en/.
Wajchenberg, B. L. Endocr. Rev. 2007, 28, 18.
Lim, A. Int. J. Nephrol. Renovasc. Dis. 2014, 7, 361.
Netwich, M. M.; Ulbig, M. W. World J. Diabetes 2015, 6, 489.
Chait, A.; Bornfeldt, K. E. J. Lipid Res. 2009, 50, S335.
Leon, B. M.; Maddox, T. M. World J. Diabetes 2015, 6, 1246.
Safavi, M.; Foroumadi, A.; Abdollahi, M. Expert Opin Drug Dis. 2013, 8, 1339.
Bailey, C. J.; Day, C. Diabetes Res. Clin. Pract. 2019, 155, 107785.
Bailey, C. J. Clin. Pharmacol. Ther. 2015, 98, 185.
Murad, M. H.; Coto-Yglesias, F.; Wang, A. T.; Sheidaee, N.; Mullan, R. J.; Elamin, M.
B.; Erwin, P. J.; Montori, V. M. J. Clin. Endocrinol. Metabol. 2009, 94, 741.
11.
12.
13.
Stoddart, L. A.; Smith, N. J.; Milligan, G. Pharmacol. Rev. 2008, 60, 405.
Nakamura, M. T.; Yudell, B. E.; Loor, J. J. Prog. Lipid. Res. 2014, 53, 124.
Itoh, Y.; Kawamata, Y.; Harada, M.; Kobayashi, M.; Fujii, R.; Fukusumi, S.; Ogi, K.;
Hosoya, M.; Tanaka, Y.; Uejima, H.; Tanaka, H.; Maruyama, M.; Satoh, R.; Okubo, S.; Kizawa,
H.; Komatsu, H.; Matsumura, F.; Noguchi, Y.; Shinohara, T.; Hinuma, S.; Fujisawa, Y.; Fujino,
M. Nature 2003, 422, 173.
14.
Kebede, M.; Ferdaoussi, M.; Mancini, A.; Alquier, T.; Kulkarni, R. N.; Walker, M. D.;
Poitout, V. Proc. Natl. Acad. Sci. USA 2012, 109, 2376.
15.
16.
17.
18.
Milligan, G.; Shimpukade, B.; Ulven, T.; Hudson, B. D. Chem. Rev. 2017, 117, 67.
Li, Z.; Xu, X.; Huang, W.; Qian, H. Med. Res. Rev. 2018, 38, 381.
Li, Z.; Zhou, Z.; Zhang, L. Exp. Opin. Ther. Pat. 2020, 30, 27.
Garrido, D. M.; Corbett, D. F.; Dwornik, K. A.; Goetz, A. S.; Littleton, T. R.; McKeown,
S. C.; Mills, W. Y.; Smalley, T. L.; Briscoe, C. P.; Peat, A. J. Bioorg. Med. Chem. Lett. 2006, 16,
1840.
19.
Negoro, N.; Sasaki, S.; Mikami, S. ACS Med. Chem. Lett. 2010, 1, 290.

20.
21.
Mancini, A. D.; Poitout, V. Diabetes Obes. Metab. 2015, 17, 622.
Chen, Y.; Song, M.; Riley, J. P.; Hu, C. C.; Peng, X.; Scheuner, D.; Bokvist, K.; Maiti,
P.; Kahl, S. D.; Montrose-Rafizadeh, C.; Hamdouchi, C.; Miller, A. R. Pharmacol. Res.
Perspect. 2016, 4, e00278.
22.
Krasavin, M.; Lukin, A.; Bagnyukova, D.; Zhurilo, N.; Zahanich, I.; Zozulya, I.;
Ihalainen, J.; Forsberg, M. M.; Lehtonen, M.; Rautio, J.; Moore, D.; Tikhonova, I. G. Bioorg.
Med. Chem. 2016, 24, 5481.
23.
24.
Zheng, Y.; Tice, C. M.; Singh, S. B. Bioorg. Med. Chem. Lett. 2014, 24, 3673.
Chupakhin, E.; Babich, O.; Prosekov, A.; Asyakina, L.; Krasavin, M. Molecules 2019,
24, 4165.
25.
26.
Shoichet, B. K. Nat. Chem. 2013, 5, 9.
Kuranov, S. O.; Tsypysheva, I. P.; Khvostov, M. V.; Zainullina, Liana F.; Borisevich, S.
S.; Vakhitova, Yu. V.; Luzina, O. A.; Salakhutdinov, N. F. Bioorg. Med. Chem. 2018, 26, 4402.
27.
28.
29.
30.
Parchinsky, V.; Shumsky, A.; Krasavin, M. Tetrahedron Lett. 2011, 52, 7161.
Ertl, P.; Roggo, S.; Schuffenhauer, A. J. Chem. Inf. Model. 2008, 48, 68.
Robbins, R. J.; Schmidt, W. F. J. Label. Compd. Radiopharm. 2004, 47, 797.
Armarego, W.L.F.; Chai, C. Purification of Laboratory Chemicals, 6th ed., Butterworth-
Heinemann, 2009.

Exploring bulky natural and natural-like periphery in the design of p-
(benzyloxy)phenylpropionic acid agonists of free fatty acid receptor 1
(GPR40)
Sergey O. Kuranov, Olga A. Luzina, Oleksandra Onopchenko, Irina Pishel, Sergey Zozulya,
Maxim Gureev, Nariman F. Salakhutdinov, and Mikhail Krasavin*
COOH
O
H
N
FFA1 EC50 = 0.75 M
Full agonist
Dose-dependent efficacy in oral glucose tolerance test

 FFA1 is a clinically validated antidiabetic target with no approved drugs on the market.
 3-Phenyl propionic acid is a known mimetic of endogenous ligands of FFA1.
 Natural and natural-like amines have been employed as periphery motifs.
 Bornyl- and cytisine-bearing compounds were most active in vitro as FFA1 agonists.
 In vivo efficacy testing (oGTT) showed the bornyl compound to be the lead compound.