Structure-activity relationships of trans-substituted-propenoic acid derivatives on the nicotinic acid receptor HCA2 (GPR109A)

Article abstract of DOI:10.1016/j.bmcl.2010.11.091

Nicotinic acid (niacin) has been used for decades as an antidyslipidemic drug in man. Its main target is the hydroxy-carboxylic acid receptor HCA2 (GPR109A), a G protein-coupled receptor. Other acids and esters such as methyl fumarate also interact with the receptor, which constituted the basis for the current study. We synthesized a novel series of substituted propenoic acids, such as fumaric acid esters, fumaric acid amides and cinnamic acid derivatives, and determined their affinities for the HCA2 receptor. We observed a rather restricted binding pocket on the receptor with trans-cinnamic acid being the largest planar ligand in our series with appreciable affinity for the receptor. Molecular modeling and analysis of the structure-activity relationships in the series suggest a planar trans-propenoic acid pharmacophore with a maximum length of 8 ? and out-of-plane orientation of the larger substituents.

Full text of DOI:10.1016/j.bmcl.2010.11.091

Bioorganic & Medicinal Chemistry Letters 21 (2011) 2736–2739
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Structure–activity relationships of trans-substituted-propenoic acid derivatives
on the nicotinic acid receptor HCA2 (GPR109A)
J. P. D. van Veldhoven, C. C. Blad, C. M. Artsen, C. Klopman, D. R. Wolfram, M. J. Abdelkadir,
⇑
J. R. Lane, J. Brussee, A. P. IJzerman
Division of Medicinal Chemistry, Leiden/Amsterdam Center for Drug Research, Leiden University, PO Box 9502, 2300 RA Leiden, The Netherlands
a r t i c l e i n f o
a b s t r a c t
Article history:
Nicotinic acid (niacin) has been used for decades as an antidyslipidemic drug in man. Its main target is
the hydroxy-carboxylic acid receptor HCA2 (GPR109A), a G protein-coupled receptor. Other acids and
esters such as methyl fumarate also interact with the receptor, which constituted the basis for the current
study. We synthesized a novel series of substituted propenoic acids, such as fumaric acid esters, fumaric
acid amides and cinnamic acid derivatives, and determined their affinities for the HCA2 receptor. We
observed a rather restricted binding pocket on the receptor with trans-cinnamic acid being the largest
planar ligand in our series with appreciable affinity for the receptor. Molecular modeling and analysis
of the structure–activity relationships in the series suggest a planar trans-propenoic acid pharmacophore
with a maximum length of 8 Å and out-of-plane orientation of the larger substituents.
Received 29 October 2010
Accepted 19 November 2010
Available online 25 November 2010
Keywords:
Nicotinic acid
GPR109A
Fumaric acid
Cinnamic acid
G protein-coupled receptor
Ó 2010 Elsevier Ltd. All rights reserved.
Since the 1950’s nicotinic acid (niacin) has been used as an anti-
dyslipidemic drug in man. Even today nicotinic acid is the most
efficacious drug to raise the levels of HDL, the ‘good’ cholesterol.1
In 2003 different groups identified that the lipid-lowering actions
of nicotinic acid are mediated by the G protein-coupled receptor
HCA2. HCA2 is also known as GPR109A, HM74A, NIACR1 or, in
mice, as PUMA-G. It is a member of a G protein-coupled receptor
subfamily involved in metabolism, with HCA3 (GPR109B) and
HCA1 (GPR81) as closely related members.2–5 The HCA2 receptor
is primarily expressed in adipocytes, spleen tissue, retinal pigment
epithelium,6 intestinal epithelium7 and various immune cells such
as monocytes and macrophages.8 Unfortunately, the HCA2 expres-
sion in a type of epidermal macrophages known as Langerhans
cells is the cause of flushing of the skin, a harmless but unpleasant
side effect of nicotinic acid which undermines treatment
compliance.3,9
Due to the discovery of the HCA2 receptor, industrial and aca-
demic groups have now started or intensified synthetic research
lines to improve on the poor safety and pharmacokinetic proper-
ties of nicotinic acid. The majority of promising novel agonists,
such as derivatives of acifran, anthranilic acids, anthranilic acid
bioisosteres, xanthines, barbituric acid and pyrazole-3-carboxylic
acids, was published and/or patented by GSK, Merck, Arena,
Schering-Plough, Roche, Incyte, and our group.10 Recently, ‘simple’
acids such as trans-cinnamic acid and 4-hydroxy-cinnamic acid
have been described as modestly active HCA2 receptor agonists
with potencies in the higher micromolar range. Cinnamic acid
derivatives had been described before as anti-inflammatory com-
pounds and as suppressors of elevated blood lipid levels in athero-
sclerosis.11–13
Some other simple acid derivatives, that is, methyl fumarate
and ethyl fumarate, were also reported as potent agonists for the
HCA2 receptor.14 These fumarates have long been known as anti-
psoriasis compounds.15 They are micromolar affinity agonists for
the HCA2 receptor, but have not been extensively explored in a
synthetic structure–activity approach. Therefore, we decided to
investigate the medicinal chemistry of such fumaric and cinnamic
acid derivatives in more detail. These constrained propenoic acid
derivatives appeared also useful in a pharmacophore analysis that
we performed.
The fumaric acid esters 2 and 3 (Table 1) were commercially
available. Compounds 4–24 (Table 1 and Supplementary data)
were synthesized according to two methods; (A) starting from fu-
maric acid (1), the appropriate alcohol and EDC dissolved in DMF.16
(B) starting from a mixture of fumaric acid (1) and the suitable
alcohol dissolved in DMF under microwave conditions.17 Method
A resulted in a mixture of both trans (4, 9–11, 13) and cis isomers
of the desired esters, even if the reaction was carried out at 0 °C.
Due to the difficult separation of the two isomers this method
was eventually not preferred. According to method B, described
by Aver’yanov,17 an equimolar mixture of 1 and the appropriate
alcohol in DMF was heated in a sealed tube in the microwave at
180 °C. This method resulted solely in the desired trans substituted
fumarates (5–8, 12, 14–24). With both methods also a substantial
amount of the disubstituted fumarates was formed.
⇑
Corresponding author. Tel.: +31 71 5274651; fax: +31 71 5274565.
E-mail address: ijzerman@lacdr.leidenuniv.nl (A.P. IJzerman).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.11.091

J. P. D. van Veldhoven et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2736–2739
2737
Table 1
Table 2
Affinities of substituted fumaric acid esters 1–24 in radioligand binding assays of the
Affinities of trans-substituted-propenoic acids 25–58 in radioligand binding assays on
human HCA2 receptor
the human HCA2 receptor
OH R1
OH
O
O
R
O
R
O
Compound
R
R1
Ki (l
M) or % disp.a
M) or % disp.a
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
48
49
50
51
52
53
54
55
56
57
58
Me
Et
Me
Phenyl
H
H
Me
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Me
Phenyl
NHCOMe
H
H
H
H
H
H
H
H
H
H
H
6%
19%
0%
4.9 1.8
4%
0%
0%
0%
6%
0%
14
2%
17%
0%
2%
0%
4%
0%
7%
1%
0%
0%
3%
7%
Compound
R
Ki (
l
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
H
Me
Et
Pr
iPr
Bu
Pe
10%
0.18 0.03
0.41 0.02
1.0 0.1
4.2 0.9
0.76 0.19
0.70 0.05
2.5 0.03
17%
10%
3.5 0.2
5.7 0.1
10
26%
0%
2-OH phenyl
2-Me phenyl
3-OH phenyl
3-Me phenyl
3-Cl phenyl
3-NO2 phenyl
4-OH phenyl
4-Me phenyl
4-Cl phenyl
4-OMe phenyl
4-NH2 phenyl
4-N(CH3)2 phenyl
3,4-di-OH phenyl
3-OMe, 4-OH phenyl
3,4-OCH2O– phenyl
Phenyl
Phenyl
Phenyl
Pyridin-3-yl
Pyrrol-2-yl
Furan-3-yl
Furan-2-yl
5-Br-furan-2-yl
5-Me-furan-2-yl
5-Et-furan-2-yl
5-(4-Cl-Ph)-furan-2-yl
Thiophen-2-yl
3-Br-thiophen-2-yl
4-Br-thiophen-2-yl
Hex
cHex
Phenyl
Benzyl
Benzyl-
Phenyl ethyl
Phenyl propyl
2-Br benzyl
2-OMe benzyl
3-Br benzyl
3-Cl benzyl
3-F benzyl
3-OMe benzyl
4-Br benzyl
4-Cl benzyl
4-Me benzyl
4-OMe benzyl
2
a-methyl rac.
1
5%
9.8 0.6
8.9 1.5
2.4 0.6
4%
21%
14%
30%
7%
14 2.5
8.1 0.8
9%
7%
14%
0%
5.5 0.3
6%
a
Ki SEM (n = 3), % displacement at 10 lM (average of n = 2, with less than 10%
difference between the two values). Ki values were determined in full displacement
studies on membranes from HEK293T cells stably expressing HCA2 (GPR109A),
using [3H]-nicotinic acid as the radioligand. Single point displacement assays were
carried out using 10 l
M of the test compound and 20 nM [3H]-nicotinic acid.
6%
a
See footnote Table 1.
The propenoic acids with aromatic rings (49, 52–54, 57 and 58)
that were not commercially available were prepared in a 32–74%
yield (Table 2 and Supplementary data), catalyzed by piperidine
via the Knoevenagel condensation of the commercially available
aromatic aldehydes and malonic acid (47).18
displacement at 10 lM, respectively. Subsequently, various
additional substitutions of the benzylic ring system were explored.
The binding pocket of the HCA2 receptor was not able to accommo-
date the ortho substituted compounds 15 and 16 at a concentration
All compounds listed in Tables 1 and 2 were tested at 10 lM in
radioligand binding assays for displacement of [3H]-nicotinic acid
(20 nM) from the human HCA2 receptor stably expressed in
HEK293 cells. Homologous displacement with unlabeled nicotinic
acid yielded a Ki value of 64 nM for nicotinic acid (data not shown).
Both the methyl and ethyl fumarates (2 and 3) also displayed submi-
cromolar affinities for the HCA2 receptor (Table 1), comparable to
data reported by Tang.14 In comparison with the reference agonist
nicotinic acid, only a three-fold (methyl derivative) or seven-fold
(ethyl derivative) lower affinity was obtained. The unsubstituted fu-
maric acid (1) did not display any appreciable affinity towards the
receptor, suggesting that the intact ester is crucial for receptor activ-
ity. In a series of aliphatic fumarate esters, increasing size did not
substantially affect the receptor affinity. The propyl, butyl and pen-
tyl substituents ( 4, 6 and 7) showed affinities between 0.7 and
of 10 lM. meta Substitution, on the other hand, was better tolerated
(17–20). The 3-bromo- and 3-chloro-benzyl derivatives (17 and 18)
showed a slight decrease in affinity and the smaller 3-fluoro-benzyl
compound(19)aslightincreaseinaffinitywithrespecttotheunsub-
stituted benzyl fumarate. On the contrary the 3-methoxy-benzyl
derivative 20 showed no affinity for the receptor. Furthermore,
introduction of para substituents such as halogen, methyl or meth-
oxy (21–24) resulted in a reduced affinity.
Since the ester moiety in methyl and ethyl fumarate can be
hydrolyzed in vivo,19 we investigated the non-hydrolysable amide
linker as an alternative. However, these trans amide isosteres of
compounds 2, 3, 10 and 11 were not able to bind to the receptor
at 10 lM (data not shown).
To further explore the SAR, a number of cis analogs of the active
trans fumaric acid derivatives were synthesized and tested, namely
the maleic acid esters and maleic acid amides. None of these cis
compounds interacted with the receptor (data not shown).
Next, a series of trans-substituted-propenoic acids ( 25–46,
48–58) were tested for their affinities (Table 2). The small aliphatic
compounds 25–27 were without effect, while the phenyl deriva-
tive 28 (cinnamic acid) showed micromolar affinity. In our assay,
1.0
l
M, which is in the same range as the ethyl derivative. The larger
M.
hexyl substituent (8) resulted in a slightly poorer Ki value of 2.5
l
Branched aliphatic compounds were also less tolerated, for example,
derivatives 5 and 9. A phenyl substituent (10) was not well tolerated
either, but introduction of a spacer between the fumarate moiety
and the aromatic system resulted in a gain of affinity. The methylene
spacer, as in 11 (Ki = 3.5
ylbenzyl (12), phenylethyl (13) and phenylpropyl (14) substituents
resulted in Ki values of 5.7 M and 10 M, and 26% of radioligand
lM), appeared to be optimal since a-meth-
compound 28 showed a higher affinity (Ki = 4.9
lM) compared to
l
l
the Ki value of 36
l
M reported by Ren et al.20 To explore this lead,

2738
J. P. D. van Veldhoven et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2736–2739
8 Å
7.3 Å
Figure 1. Aligned pharmacophore model (left—nicotinic acid in plane, right—nicotinic acid 90° rotated) constructed of the active compounds nicotinic acid, 6, 11 and 28 and
the inactive compound 10. Red: nicotinic acid, green: cinnamic acid (28), gray: phenyl fumarate (10), blue: butyl fumarate (6), yellow: benzyl fumarate (11).
commercially available aromatic substituted trans propenoic acids
were tested for their affinities (29–43). Only the 4-hydroxy deriv-
ative 35 was able to bind with an affinity of 14 lM. Related substit-
In conclusion, methyl fumarate, ethyl fumarate and cinnamic
acid have been published as agonists for the HCA2 receptor.14,20
Our synthetic program confirmed the affinity of these compounds
for the HCA2 receptor and further explored the structure–activity
relationships for a series of derivatives. Molecular modeling stud-
ies and the analysis of the structure–activity relationships in the
series suggest a planar trans-propenoic acid pharmacophore with
a maximum length of 8 Å and out-of-plane orientation of the larger
substituents
uents such as 4-methoxy (38) and 4-amino (39) decreased the
affinity dramatically. In general, except for the 4-hydroxy group,
aromatic substitution is not tolerated on the ortho, meta or para po-
sition. Also substituents at the b-position of cinnamic acid (44–46)
resulted in a dramatic decrease in binding. Replacement of the
phenyl moiety in cinnamic acid by aromatic isosteres such as 3-
pyridinyl (48) and 2-pyrrole (49) resulted in a significant loss of
affinity. On the contrary, 2-furanyl (51) and 2-thiophenyl (56) were
accommodated like the phenyl compound. 3-Furanyl substitution
(50) resulted in a two-fold decrease compared to the 2-furanyl
derivative 51. As in the cinnamic acid series, the 5-substituted 2-
furanyl derivatives (52–55) and both the 3-bromo and 4-bromo-
substituted 2-thiophenyl derivatives (57, 58) were devoid of affin-
ity for the receptor.
Acknowledgments
This work was performed within the framework of the Dutch
Top Institute Pharma project D1-105.
Supplementary data
To visualize the SAR, a pharmacophore model was generated by
manually superimposing the minimized structures of: nicotinic
acid, cinnamic acid 28, fumaric acid esters 6 and 11, and the inac-
tive phenylfumaric acid ester 10 (Fig. 1). The alignment of the two
sp2 carbons of the propenoic fragment, which all the compounds
have in common, resulted in a planar and constrained pharmaco-
phore. The carbonyl oxygen of the ester function of compounds
6, 10 and 11 and the nitrogen of nicotinic acid overlay smoothly
as a hydrogen acceptor region. This might explain the improved
binding characteristics of the fumaric acid esters compared to cin-
namic acid and also why the trans configuration is superior over
the cis substituted propenoic acids. Molecular modeling and anal-
ysis of the structure–activity relationships in the series suggest a
planar trans-propenoic acid pharmacophore with a maximum
length of 8 Å, because this is the size of the largest planar ligand
(28) in our series with appreciable affinity for the receptor. Larger
compounds need an out-of-plane orientation as in the case of the
fumaric acid ester series (2–24).
Supplementary data (reaction schemes and experimental data)
associated with this article can be found, in the online version, at
doi:10.1016/j.bmcl.2010.11.091.
References and notes
1. Mack, W. J.; Selzer, R. H.; Hodis, H. N.; Erickson, J. K.; Liu, C. R.; Liu, C. H.;
Crawford, D. W.; Blankenhorn, D. H. Stroke 1993, 24, 1779.
2. Soga, T.; Kamohara, M.; Takasaki, J.; Matsumoto, S.; Saito, T.; Ohishi, T.;
Hiyama, H.; Matsuo, A.; Matsushime, H.; Furuichi, K. Biochem. Biophys. Res.
Commun. 2003, 303, 364.
3. Tunaru, S.; Kero, J.; Schaub, A.; Wufka, C.; Blaukat, A.; Pfeffer, K.; Offermanns, S.
Nat. Med. 2003, 9, 352.
4. Wise, A.; Foord, S. M.; Fraser, N. J.; Barnes, A. A.; Elshourbagy, N.; Eilert, M.;
Ignar, D. M.; Murdock, P. R.; Steplewski, K.; Green, A.; Brown, A. J.; Dowell, S. J.;
Szekeres, P. G.; Hassall, D. G.; Marshall, F. H.; Wilson, S.; Pike, N. B. J. Biol. Chem.
2003, 278, 9869.
5. Offermanns, S.; Colleti, S. L.; Lovenberg, T. W.; Semple, G.; Wise, A.; IJzerman, A.
P. Pharmacol. Rev., in press.
6. Martin, P. M.; Ananth, S.; Cresci, G.; Roon, P.; Smith, S.; Ganapathy, V. Mol. Vis.
2009, 15, 362.
7. Thangaraju, M.; Cresci, G. A.; Liu, K.; Ananth, S.; Gnanaprakasam, J. P.;
Browning, D. D.; Mellinger, J. D.; Smith, S. B.; Digby, G. J.; Lambert, N. A.;
Prasad, P. D.; Ganapathy, V. Cancer Res. 2009, 69, 2826.
8. Gille, A.; Bodor, E. T.; Ahmed, K.; Offermanns, S. Annu. Rev. Pharmacol. Toxicol.
2008, 48, 79.
Molecular modeling studies of the Merck Research group based
on anthranilic acid derivatives confirmed the importance of the pla-
nar orientation of the carboxylic acid function and the nearby a,b
sp2 carbon atoms.21,22 Full saturation of the phenyl ring in anthra-
nilic acid resulted in inactive compounds. If the double bond in
9. Benyo, Z.; Gille, A.; Bennett, C. L.; Clausen, B. E.; Offermanns, S. Mol. Pharmacol.
2006, 70, 1844.
10. Shen, H. C.; Colletti, S. L. Expert Opin. Ther. Pat. 2009, 19, 957.
11. Simonyan, A. V. Pharm. Chem. J. 1993, 27, 92.
the
a,b-position was maintained, as in tetrahydro-anthranilic acids,
the planar orientation and also the affinity was regained however.22

J. P. D. van Veldhoven et al. / Bioorg. Med. Chem. Lett. 21 (2011) 2736–2739
2739
12. Watanabe, T.; Hayashi, K.; Yoshimatsu, S.; Sakai, K.; Takeyama, S.; Takashima,
K. J. Med. Chem. 1980, 23, 50.
13. Lee, M. K.; Park, Y. B.; Moon, S. S.; Bok, S. H.; Kim, D. J.; Ha, T. Y.; Jeong, T. S.;
Jeong, K. S.; Choi, M. S. Chem. Biol. Interact. 2007, 170, 9.
14. Tang, H.; Lu, J. Y. L.; Zheng, X. M.; Yang, Y. H.; Reagan, J. D. Biochem. Biophys. Res.
Commun. 2008, 375, 562.
15. Schweckendiek, W. Med. Monatsschr. 1959, 13, 103.
18. Shiotani, S.; Morita, H. J. Heterocycl. Chem. 1982, 19, 1207.
19. Litjens, N.; van Strijen, E.; van Gulpen, C.; Mattie, H.; van Dissel, J.; Thio, H. B.;
Nibbering, P. BMC Pharmacol. 2004, 4, 22.
20. Ren, N.; Kaplan, R.; Hernandez, M.; Cheng, K.; Jin, L.; Taggart, A. K.; Zhu, A. Y.;
Gan, X.; Wright, S. D.; Cai, T. Q. J. Lipid Res. 2009, 50, 908.
21. Deng, Q. L.; Frie, J. L.; Marley, D. M.; Beresis, R. T.; Ren, N.; Cai, T. Q.; Taggart, A.
K. P.; Cheng, K.; Carballo-Jane, E.; Wang, J.; Tong, X. C.; Waters, M. G.; Tata, J. R.;
Colletti, S. L. Bioorg. Med. Chem. Lett. 2008, 18, 4963.
16. Ekici, Ö. D.; Götz, M. G.; James, K. E.; Li, Z. Z.; Rukamp, B. J.; Asgian, J. L.; Caffrey,
ˇ
C. R.; Hansell, E.; Dvorák, J.; McKerrow, J. H.; Potempa, J.; Travis, J.; Mikolajczyk,
22. Raghavan, S.; Tria, G. S.; Shen, H. C.; Ding, F. X.; Taggart, A. K.; Ren, N.;
Wilsie, L. C.; Krsmanovic, M. L.; Holt, T. G.; Wolff, M. S.; Waters, M. G.;
Hammond, M. L.; Tata, J. R.; Colletti, S. L. Bioorg. Med. Chem. Lett. 2008,
18, 3163.
J.; Salvesen, G. S.; Powers, J. C. J. Med. Chem. 2004, 47, 1889.
17. Aver’yanov, D. M.; Batrakova, A. V.; Samuilov, Y. D.; Spiridonova, R. R.; Kochnev,
A. M.; Galibeev, S. S.; Gnezdilov, O. I. Russ. J. Gen. Chem. 2008, 78, 1920.