Synthesis and PGE2 inhibitory activity of novel diarylheptanoids

Article abstract of DOI:10.1016/j.bmcl.2017.12.046

Prostaglandin E₂ (PGE₂) is a lipid mediator of inflammation and its inhibition has become a popular drug target due to its harmful physiological roles. Diarylheptanoids are one class of compounds that have shown successful inhibition of PGE₂. This paper reports the synthesis and PGE₂ inhibitory activity of a series of analogues of a naturally occurring diarylheptanoid. The most efficacious compounds were examined for dose-dependent PGE₂ inhibition. Among several promising compounds, the lead candidate exhibited an IC₅₀ value of 0.56 ng/μL or 1.7 μM with no detectable toxicity at the highest dose of 10 ng/μL.

Full text of DOI:10.1016/j.bmcl.2017.12.046

Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Synthesis and PGE2 inhibitory activity of novel diarylheptanoids
Richard D. McLane a, Léon Le Cozannet-Laidin b, Maxwell S. Boyle a, Lindsey Lanzillotta c,
Zachary L. Taylor c, Sarah R. Anthony c, Michael Tranter c, Amber J. Onorato a,
⇑
a Department of Chemistry, Northern Kentucky University, 1 Nunn Drive, Highland Heights, KY 41099, USA
b Départment de Measures Physiques, Institut Universitaire de Technologie de Lannion, BP 30219, Rue Edouard Branly, 22302 Lannion Cedex, France
c Division of Cardiovascular Health and Disease, University of Cincinnati College of Medicine, 231 Albert Sabin Way, Cincinnati, OH 45267, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Prostaglandin E2 (PGE2) is a lipid mediator of inflammation and its inhibition has become a popular drug
target due to its harmful physiological roles. Diarylheptanoids are one class of compounds that have
shown successful inhibition of PGE2. This paper reports the synthesis and PGE2 inhibitory activity of a ser-
ies of analogues of a naturally occurring diarylheptanoid. The most efficacious compounds were exam-
ined for dose-dependent PGE2 inhibition. Among several promising compounds, the lead candidate
exhibited an IC50 value of 0.56 ng/mL or 1.7 mM with no detectable toxicity at the highest dose of 10
ng/mL.
Received 29 October 2017
Revised 19 December 2017
Accepted 20 December 2017
Available online xxxx
Keywords:
Diarylheptanoids
Prostaglandin E2
Curcumin
Ó 2017 Elsevier Ltd. All rights reserved.
Inflammation
Cross metathesis
Inflammation is an adaptive, physiological response to harmful
stimuli that can include tissue damage and infection. Generally, the
inflammatory response is beneficial and protective to cells and tis-
sues. However, dysregulation of the inflammatory response can
result in adverse effects and contributes to the pathogenesis of
many diseases. The inflammatory response can be mediated
through the conversion of arachidonic acid to the cyclooxygenases
(COX-1 and COX-2), which produce thromboxanes (TX) and prosta-
glandins (PG) (Fig. 1).1 The prostaglandins, PGI2 and PGE2, induce
vasodilation and increase vascular permeability which can cause
long term swelling and tissue damage.2 Furthermore, PGE2 has
been shown to affect pain, tumorigenesis, neuronal functions,
female reproduction, gastric mucosal protection, and kidney func-
tion.2,3 Non-specific NSAIDs target both COX-1 and COX-2 with the
intent of down-regulating the inflammatory response entirely, but
also result in damage to the gastric mucosa.4 More recently, COX-2
inhibitors or coxibs have been developed to preferentially inhibit
COX-2 activity and avoid damaging the gastric mucosa. However,
selective COX-2 inhibition results in reduced PGI2 production by
the vascular endothelium and predisposes individuals to increased
vascular endothelial injury.4 For this reason, individuals with pre-
existing cardiovascular disease have an increased risk for cardiac
ischemia, heart failure, hypertension, and cardiac arrhythmia when
taking coxibs.5,6 In an effort to avoid these harmful side effects,
mPGES-1 (microsomal prostaglandin E2 synthase-1) inhibitors
have gained interest as a novel therapeutic target for the treatment
of inflammation associated with a variety of diseases.7 One class of
natural products, that have shown anti-inflammatory activity by
inhibition of mPGES-1, are diarylheptanoids.8
Diarylheptanoid natural products are produced as secondary
plant metabolites and are recognized by two aromatic rings linked
by a seven-carbon chain. They are primarily isolated from plants in
the genus Zingiber, Alnus, Alpinia, Curcuma, and Myrica. Diarylhep-
tanoids have become of increased interest due to their vast array of
biological activities including anti-inflammatory, antioxidant, anti-
tumor, antiosteoporotic, antibacterial, melanogenic, hepatoprotec-
tive, and neuroprotective properties.9,10
A
diarylheptanoid molecule (1-(400-methoxyphenyl)-7-(40-
hydroxyphenyl)-(E)-hept-2-ene, 1, Fig. 2) was isolated from Pleu-
ranthodium racemigerum, an Australian ginger plant.11 Upon isola-
tion,
1 was tested and showed moderate anti-inflammatory
activity through inhibition of PGE2 production.11 The goal of this
research was to synthesize and investigate the biological proper-
ties of next-generation diarylheptanoid molecules that possess
greater anti-inflammatory than 1 with enhanced on-target activity.
Herein, we describe the synthesis and anti-inflammatory activity
of 1 and seven novel analogues (Fig. 2). The molecules were syn-
thesized via a four or five step route with a cross metathesis reac-
tion using Grubbs CatalystTM 1st generation as the key step. Each
final compound was tested for its inhibition of PGE2 and the results
were compared against 1.
⇑
Corresponding author.
E-mail address: onoratoa1@nku.edu (A.J. Onorato).
https://doi.org/10.1016/j.bmcl.2017.12.046
0960-894X/Ó 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: McLane R.D., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.12.046

2
R.D. McLane et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
reagent.12 This shorter reaction time increased the yield of 11a to
Arachidonic Acid
89%. Compound 1 was then produced through the cross metathesis
reaction of the alkene 11a with 4-allylanisole and Grubbs CatalystTM
1st generation in anhydrous dichloromethane at 35 °C for 18 h. To
aid in the removal of the catalyst, 50 equivalents of DMSO (relative
to the catalyst) were added to the crude mixture, which was stirred
for an additional 18 h.14 While the mechanism remains unknown,
it has been proposed that DMSO acts as a mild oxidant and con-
verts the metal species into polar oxides that can be purified dur-
ing column chromatography.15 The final product was purified
using MPLC to provide 1 in an 18% yield. Based on NMR integration,
it was concluded that 1 was formed as a mixture of E- (85%) and Z-
isomers (15%). Chang was able to separate the mixture of E:Z cross
metathesis products by the use of AgNO3-doped silica gel.12 After
several efforts to separate the E:Z product mixture using this
method, we were unable to isolate either of the compounds and
tested 1 as a set of diastereomers with E being the major product.
The first novel analogue of 1 synthesized was compound 2,
which lacked the alkene on the seven-carbon linker. This analogue
was created in order to probe the importance of the alkene to its
biological activity. Molecule 1 underwent a hydrogenation reaction
with palladium on carbon to create 2 in a 60% yield.16
Our next novel analogues synthesized were compounds 3 and 4
(Scheme 1), which contain the phenol on the A ring in the ortho-
rather than the para-position as in compounds 1 and 2. Compound
3 was synthesized through the reduction of commercially available
3-(2-hydroxyphenyl)propionic acid using a borane-tetrahydrofu-
ran complex and trimethyl borate to afford the primary alcohol
9b in a 99% yield. While the reduction to provide alcohol 9b was
analogous to the alcohol 9a, the bromination reaction to prepare
10b gave much lower yields (33% at best). Based on the work of
Tongkate et al., the equivalents of CBr4 and PPh3 were lowered
and the reaction time was shortened to produce bromide 10b in
a 57% yield, which although low is almost double the initial yield.17
Compound 10b was further reacted with excess allylmagnesium
chloride for 18 h to provide 11b in an 83% yield. Molecule 3 was
then produced through the cross metathesis reaction of the alkene
11b with 4-allylanisole and Grubbs CatalystTM 1st generation under
the same reaction conditions that were used to produce compound
1. The final product was purified using MPLC to give 3 in a 25%
yield. Based on NMR integration, it was concluded that 3 was
formed as a mixture of E- (82%) and Z-isomers (18%), which was
used for biological testing without further purification. Compound
3 was then reacted with palladium on carbon under an atmosphere
of hydrogen to afford 4 in a 60% yield.16
PGG2
PGH2
PGI2
PGE2
PGD2
PGF2α
TXA2
Fig. 1. Prostanoid Biosynthetic Pathway.
Synthesis of diarylheptanoids
To date, we have synthesized the natural product 1 and seven
novel analogues to begin our investigation in developing a prelim-
inary structure-activity relationship (Fig. 2). The synthetic route
was adapted from Chang and involves a four- or five-step synthesis
with a cross metathesis reaction creating the seven-carbon linker
(Scheme 1).12
Our synthesis of 1 began with the reduction of commercially
available 3-(4-hydroxyphenyl)propionic acid using
a borane-
tetrahydrofuran complex and trimethyl borate to afford the pri-
mary alcohol 9a in a 99% yield. Initially, this reduction only pro-
duced a 41% yield, but based on the methodology described by
Zhou and co-workers, we found increasing the equivalents of tri-
methyl borate dramatically improved the yield.13 Compound 9a
was further reacted with carbon tetrabromide and triphenylphos-
phine in anhydrous dichloromethane to give the primary alkyl bro-
mide 10a in a 69% yield. The bromide 10a was refluxed in
anhydrous toluene for 48 h in the presence of excess allylmagne-
sium chloride to produce the Grignard product 11a in a 63% yield.
As an acidic proton is present in compound 10a, excess Grignard
reagent was used instead of adding a protecting group, as the
yields for the reaction were similar regardless of a protecting group
as reported by Chang in the original synthesis of 1.12 In addition,
being the second step of the synthesis, we looked into improving
the yield. The total reaction time for the Grignard reaction was
reduced to 18 h from the original 48 h, developed by Chang, to
decrease the amount of possible disproportionation of the Grignard
The next novel analogues synthesized were 5 and 6 (Scheme 1).
These derivatives have the phenol on the A ring in the para-posi-
HO
OCH3
OCH3
HO
2
4
1
3
OH
OCH3
OCH3
OH
OCH3
OCH3
HO
OCH3
OCH3
HO
OCH3
OCH3
6
8
5
7
OH
OCH3
OH
OCH3
Fig. 2. Synthesized Diarylheptanoids.
Please cite this article in press as: McLane R.D., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.12.046

R.D. McLane et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
3
O
a
c
b
OH
OH
Br
R1
R2
R1
R2
R1
R2
9a R1 = OH, R2 = H; 99%
9b R1 = H, R2 = OH; 99%
10a R1 = OH, R2 = H; 69%
10b R1 = H, R2 = OH; 57%
R3
e
R3
d
A
B
A
B
R1
R2
OCH3
R1
R2
R1
R2
OCH3
11a R1 = OH, R2 = H; 89%
11b R1 = H, R2 = OH; 83%
1 R1 = OH, R2 = H, R3 = H; 18%
3 R1 = H, R2 = OH, R3 = H; 26%
5 R1 = OH, R2 = H, R3 = OCH3; 9%
7 R1 = H, R2 = OH, R3 = OCH3; 5%
2 R1 = OH, R2 = H, R3 = H; 60%
4 R1 = H, R2 = OH, R3 = H; 60%
6 R1 = OH, R2 = H, R3 = OCH3; 31%
8 R1 = H, R2 = OH, R3 = OCH3; 87%
Scheme 1. Reagents and conditions: (a) B(OMe)3, BH3-THF complex, THF, 0 °C; (b) CBr4, PPh3, DCM, 0° to rt; (c) allylmagnesium chloride, toluene, 110 °C; (d) 4-allylanisole or
4-allyl-1,2-dimethoxybenzene, Grubbs CatalystTM 1st generation, DCM, 35 °C; (e) H2, Pd/C, EtOH.
tion and an additional methoxy group on the B ring. Compound 5
was synthesized by reacting alkene 11a with 4-allyl-1,2-
dimethoxybenzene in the presence of Grubbs CatalystTM 1st gener-
ation.12 After purification, 5 was isolated in a 10% yield. Based on
NMR integration, it was concluded that 5 was formed as a mixture
of E- (81%) and Z-isomers (19%), which was used for biological test-
ing without further purification. Compound 5 was then reacted
with palladium on carbon under an atmosphere of hydrogen to
afford 6 in a 31% yield.16
The last novel analogues synthesized were 7 and 8 (Scheme 1).
These derivatives have the phenol on the A ring in the ortho-posi-
tion and an additional methoxy group on the B ring. Compound 7
was synthesized by reacting alkene 11b with 4-allyl-1,2-
dimethoxybenzene in the presence of Grubbs CatalystTM 1st gener-
ation.12 After purification, 7 was isolated in a 5% yield and based on
NMR integration, it was concluded that 7 was formed as a mixture
of E- (89%) and Z-isomers (11%). This mixture of alkenes was used
for biological testing without further purification. Compound 7
was then reacted with palladium on carbon under an atmosphere
of hydrogen to afford 8 in an 87% yield.16
dinate more strongly with the catalyst. Despite favorable bypro-
duct formation, multiple purifications and low yields, we
obtained enough material for biological testing. These reactions
may be applied to create more final compounds as the synthetic
route is fairly short and does not require many expensive chemi-
cals, excluding the catalyst.
PGE2 inhibitory activity and biological properties of
diarylheptanoids
Diarylheptanoid compounds 1–8 (Fig. 2), 12, and 13 (Scheme 2)
were tested for their ability to inhibit the production of prostaglan-
din E2 (PGE2) in 3 T3 murine fibroblast cells. As shown in Fig. 3,
efficacy of the diarylheptanoid compounds at 1 ng/
ll and 10 ng/
ll ranged from no significant decrease (1 and 4) to significant
reduction in PGE2 (3, 5, 6, 8, and 12), as compared to vehicle con-
trol. The most efficacious inhibitor of PGE2 production was 6,
which reduced PGE2 levels to 33% of control at 10 ng/ll. Many of
the compounds (5, 6, 8, 12, and 13) also exhibited a dose-depen-
During the cross metathesis reactions to form 1 and 3, we
observed the formation of a major byproduct, 1,4-bis(4-methoxy-
phenyl)but-2-ene (12), which was formed by 4-allylanisole react-
ing with itself. This product was purified and tested alongside
the diarylheptanoid molecules, in addition to the hydrogenated
version (13). Compound 13 was formed, in a quantitative yield,
by reacting 12 with palladium on carbon under an atmosphere of
hydrogen (Scheme 2).
dent inhibition of PGE2 production, while others did not, suggest-
ing
a
difference in potency between the compounds.
Furthermore, we determined a more in-depth dose-response curve
for the four most efficacious compounds (5, 6, 8, and 12) and iden-
tified their IC50 values to be 0.56, 9.6, 5.5, and 3.2 ng/ll or 1.7, 29,
17 and 12 mM, respectively (Fig. 4). The steepness of the slope of 5
suggests that it has the highest potency and a maximum efficacy
comparable to 12.
As shown above, the purified quantities of the cross metathesis
reaction products are less than desirable. We found the low yields
were attributed to two major factors. Firstly, the major impurity
observed for each cross metathesis reaction was a byproduct
formed by either the 4-allylanisole or 4-allyl-1,2-dimethoxyben-
zene undergoing a cross metathesis reaction with itself. As the
desired product and byproduct share similarities in polarity and
solubility, isolating pure diarylheptanoid proved to be challenging.
Secondly, we found multiple purifications were necessary to obtain
pure product. This was especially true for compounds 5 and 7, as
they have an additional oxygen containing functionality and coor-
The diarylheptanoid compounds were also tested for inhibitory
activity of NF-
helps to regulate PGE2 activity. To test NF-
were transfected with an NF- B-luciferase reporter containing a
luciferase reporter gene under control of an NF- B specific gene
jB, a pro-inflammatory transcription factor that
jB inhibition, 3T3 cells
j
j
promoter. Curcumin, an extensively studied and structurally simi-
lar diarylheptanoid,18 has been shown to inhibit transcriptional
activation of NF-
assay.19 As expected, TNF-
significantly decreased by the addition of curcumin (10 ng/
j
B and was used as a positive control for the
-induced NF- B reporter activity was
l)
a
j
l
(Fig. 5). However, none of the tested diarylheptanoid compounds
OCH3
OCH3
e
H3CO
H3CO
13; 100%
12
Scheme 2. Reagents and conditions: (e) H2, Pd/C, EtOH.
Please cite this article in press as: McLane R.D., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.12.046

4
R.D. McLane et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Fig. 3. 3T3 cells were treated with compounds 1–8, 12, and 13 at doses of 1 and 10
ng/ml, and media was collected for determination of PGE2 secretion via ELISA. Many
Fig. 5. Curcumin, but not the tested diarylheptanoid compounds reduced TNF-a-
of the novel compounds show an inhibitory effect on PGE2 production (*P < .05, **
< .01, ***P < .001, ****P < .0001).
P
induced NF-kB transcriptional activity in 3T3 cells at 10 ng/
l
l. *P ꢀ .05 vs. TNF-
a
treated cells.
Fig. 4. 3 T3 cells were treated with compounds 5, 6, 8, and 12 at four different doses
to determine IC50 values for inhibition of PGE2 secretion. IC50 values were calculated
with SigmaPlot using four-parameter logistic fit.
Fig. 6. AlamarBlueÒ was used to assess 3T3 cell viability following six hours
treatment with compounds 5, 6, 8, and 12 at 10 ng/ l. While a six-hour treatment
(all at 10 ng/ll) exhibited any inhibitory activity toward NF-jB,
suggesting the possibility of enhanced on-target activity compared
to curcumin.
l
with H2O2 resulted in a dose-dependent decrease in cell viability, none of the
diarylheptanoid compounds induced any toxicity. *P ꢀ .05 vs. Vehicle control.
Cellular toxicity for the compounds was measured by ala-
marBlueÒ cell viability assay. The active ingredient of alamarBlueÒ,
resazurin, is reduced by viable cells to resorufin and can be used to
quantify the number of viable cells through fluorescence intensity
or absorbance.20 In both cases, we observed a dose-dependent
degree of cell death induced by H2O2 treatment, as expected. How-
ever, none of the tested diarylheptanoid compounds, administered
at a dose of 10 ng/ll, resulted in any decrease in cell viability. This
was consistent for both fluorescence and absorbance readouts of
the assay (Fig. 6).
phenol on the A ring shows preference for the para position. The
alkene on the seven-carbon linker does not appear to be crucial
for the observed inhibition of PGE2 as 6 was the most efficacious.
In contrast, the alkene on the linker does appear to be necessary
for the activity of 12 as compared to 13. The information gathered
from this preliminary structure-activity relationship will be used
to guide the synthesis of future diarylheptanoid analogues in order
to create a more potent lead compound. Further characterization of
the diarylheptanoid compounds is also necessary to help discern
their biological target(s). In addition, molecules with scaffolds sim-
ilar to compound 12 (1,4-bis(4-methoxyphenyl)but-2-ene) should
be further investigated.
In sum, we have described the synthesis of several novel diaryl-
heptanoid analogues of 1. Biological results demonstrate that com-
pounds 3, 5, 6, 8, and 12 inhibit PGE2 production with more
efficacy than 1. In addition, the lack of NF-jB inhibition of our
diarylheptanoids suggests possibly improved on target specificity
versus curcumin, which needs to be further investigated. Based
on the PGE2 inhibitory activity, the compounds that contain a
dimethoxy functional group on the B ring (5–8) show a greater
potency. When comparing the dimethoxy compounds (5–8), the
Acknowledgments
We gratefully acknowledge the NKU Department of Chemistry
and College of Arts and Sciences and the University of Cincinnati
Please cite this article in press as: McLane R.D., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.12.046

R.D. McLane et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
5. Caughey GE, Cleland LG, Penglis PS, Gamble JR, James MJ.
5
J
Immunol.
for their financial support. This work was also supported by grants
2001;167:2831–2838.
from NIGMS (grant # 8P20GM103436-14), the NKU Center for
Integrative Natural Science and Mathematics, and the Kentucky
Academy of Science Undergraduate Research Program. We thank
the School of Chemical Sciences Mass Spectrometry Laboratory at
the University of Illinois at Urbana-Champaign for obtaining HR-
MS data.
6. Funk CD, Fitzgerald GAJ. Cardiovasc Pharm. 2007;50:470–479.
7. Koeberle A, Laufer SA, Werz O. J Med Chem. 2016;59:5970–5986.
8. Koeberle A, Northoff H, Werz O. Mol Cancer Ther. 2009;8:2348–2355.
9. Ma X-N, Xie C-L, Miao Z, Yang Q, Yang X-W. RSC Adv. 2017;7:14114–14144.
10. Lv H, She G. Nat Prod Commun. 2010;5:1687–1708.
11. Wohlmuth H, Desco MA, Brushett DJ, et al. J Nat Prod. 2010;73:743–746.
12. Chang C-Y. J Chin Chem Soc. 2011;58:286–289.
13. Zhou Y, Gao G, Li H, Qu J. Tetrahedron Lett. 2008;49:3260–3263.
14. Ahn YM, Yang K, Georg GI. Org Lett. 2001;3:1411–1413.
15. Knochel P, Molander GA. Comprehensive organic synthesis. Amsterdam,
Netherlands: Elsevier; 2014.
A. Supplementary data
16. Suksamrarn A, Ponglikitmongkol M, Wongkrajang K, et al. Bioorg Med Chem.
2008;16:6891–6902.
Supplementary data associated with this article can be found, in
the online version, at https://doi.org/10.1016/j.bmcl.2017.12.046.
17. Tongkate P, Pluempanupat
2008;49:1146–1148.
W, Chavasiri W.
Tetrahedron Lett.
18. Nelson KM, Dahlin JL, Bisson J, Graham J, Pauli GF, Walters MA. J Med Chem.
2017;60:1620–1637.
References
19. Brennan P, O’Neill LAJ. Biochem Pharmacol. 1998;55:965–973.
20. alamarBlueÒ Cell Viability Assay Protocol https://www.thermofisher.com/us/
en/home/references/protocols/cell-and-tissue-analysis/cell-profilteration-
assay-protocols/cell-viability-with-alamarblue.html#4 [accessed 18.10.17].
1. Stitham J, Midgett C, Martin KA, Hwa J. Front Pharmacol. 2011;2:1–8.
2. Medzhitov R. Nature. 2008;454:428–435.
3. Samuelsson B, Morgenstern R, Jakobsson P-J. Pharmacol Rev. 2007;59:207–224.
4. Matsui H, Shimokawa O, Kaneko T, Nagano Y, Rai K, Hyodo IJ. Clin Biochem Nutr.
2011;48:107–111.
Please cite this article in press as: McLane R.D., et al. Bioorg. Med. Chem. Lett. (2017), https://doi.org/10.1016/j.bmcl.2017.12.046