A new synthesis of 4′-resveratrol esters and evaluation of the potential for anti-depressant activity

Article abstract of DOI:10.1016/j.bmcl.2013.03.046

The 4′-ester analog of the disease preventative resveratrol 1 (RV), 4′-acetyl-RV 2 along with 4′-pivaloate 13 and benzoate 14 RV were synthesized. The previously developed palladium catalyzed decarbonylative Heck coupling was used to assemble the stilbene core together with 3,5-dibenzyl protected phenol intermediates that allowed for efficient coupling and deprotection using boron trifluoride etherate. Studies with Long-Evans rats were performed to establish safety, toxicity, and behavioral parameters. In addition, the Porsalt forced-swim test was used to demonstrate anti-depressant activity.

Full text of DOI:10.1016/j.bmcl.2013.03.046

Bioorganic & Medicinal Chemistry Letters 23 (2013) 2941–2944
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
A new synthesis of 40-resveratrol esters and evaluation
of the potential for anti-depressant activity
Mark J. Acerson b, Kimberly M. Fabick a, Yong Wong b, Crystal Blake a, Edwin D. Lephart a,
Merritt B. Andrus b,
⇑
a Department of Physiology/The Neuroscience Center, Brigham Young University, Provo, UT 84602, USA
b Department of Chemistry, Brigham Young University, Provo, UT 84602, USA
a r t i c l e i n f o
a b s t r a c t
The 40-ester analog of the disease preventative resveratrol 1 (RV), 40-acetyl-RV 2 along with 40-pivaloate
13 and benzoate 14 RV were synthesized. The previously developed palladium catalyzed decarbonylative
Heck coupling was used to assemble the stilbene core together with 3,5-dibenzyl protected phenol inter-
mediates that allowed for efficient coupling and deprotection using boron trifluoride etherate. Studies
with Long-Evans rats were performed to establish safety, toxicity, and behavioral parameters. In addition,
the Porsalt forced-swim test was used to demonstrate anti-depressant activity.
Article history:
Received 22 January 2013
Revised 4 March 2013
Accepted 12 March 2013
Available online 21 March 2013
Keywords:
Resveratrol
Ó 2013 Elsevier Ltd. All rights reserved.
Polyphenol
Palladium catalysis
Heck coupling
Anti-depressant
Resveratrol (RV) is a polyphenolic stilbene phytoalexin pro-
duced by several plants, including Japanese knotweed, when chal-
lenged by bacterial or fungal pathogens or under stress conditions
including heat or drought.1 Previously we reported the synthesis of
various 3, 5, and 40-acetate ester analogs and fluoro variants of res-
veratrol together with their anticancer activity using HL-60 cell
and melanoma assays and longevity studies using yeast.2 40-Acetyl
RV 2 (4AR, Scheme 1) was found to be slightly most potent (ED50
(3,5-dihydroxybenzoic acid).2a,b On large scale, problems were
encountered with deprotection of the previously employed 3,5-
di-MOM (methoxymethyl) groups using TMSI (trimethylsilyl
iodide),5 as reported for 40-acetyl–RV 2 (Scheme 1).2b On scales
above 1 g, the quenching agent sodium thiosulfate reacted with ex-
cess TMSI giving sulfur by-products leading to greatly lowered
yields (10–20%). Other quenching agents, including HFꢀpyr., were
also ineffective. A new, more efficient and economical route
amenable to large scale is now reported together with two new
40-ester variations using this approach, 40-pivaloate and benzoate
RV, 3 (R = t-Bu, Ph). In addition, studies with Long-Evans rats are
reported that demonstrate safety, non-toxicity, bioavailability,
and behavioral parameters together with the Porsalt forced-swim
test to assess potential antidepressant activity.
Following the reported procedure, the 3,5-dibenzyl protected
benzoyl chloride coupling partner 6 was produced in high overall
yield (Scheme 2).6 Inexpensive resorcylic acid (3,5-dihydroxyben-
zoic acid) 4 was esterified and benzylated to give 5, which was
treated with base to access the protected benzoic acid followed
by thionyl chloride to give the acid chloride 6 (60% overall yield).
Intermediate 6 was then used in crude form to couple with para-
acetoxystyrene 7 using the previously established decarbonylative
Heck conditions.7,2a,2b Palladium(II) acetate (1 mol %) catalyst was
formed in situ with bis-adamantyl 8 (1.5 mol %, N,N-di-(1-adaman-
tyl)-4,5-dihydro-1H-imidazolium tetrafluoroborate)8 as precursor
to the NHC (N-heterocyclic carbene) ligand SIAd generated in the
17 lM) compared to RV (22 l
M) using HL-60 assays2b and more
effective than RV using DM445 and DM738 cells.2a 40-Acetyl RV 2
was also found to be more resistant to decomposition, remaining
as a stable, pure white solid over extended time.2b Longevity stud-
ies with 40-acetyl RV 2 also demonstrated enhanced stability and
activity compared to RV by significantly extending the lifespan of
yeast with reduced dosing.2c As with most dietary polyphenols,
resveratrol has a relatively short 30 min in vivo half life and is rap-
idly cleared.3 Recent studies indicate that multi-drug resistance
transport proteins, MRP2 and BCRP, are a contributing factor for
the low bioavailability of RV.4 Further improved pro-drug analogs
of this dietary agent offer the potential for new therapeutic appli-
cations with enhanced stability and bioavailability.2
The previous route to RV esters featured an efficient decarbony-
lative Heck route starting with inexpensive resorcylic acid
⇑
Corresponding author. Tel.: +1 801 422 8171; fax: +1 801 422 0153.
E-mail address: mbandrus@chem.byu.edu (M.B. Andrus).
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.03.046

2942
M. J. Acerson et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2941–2944
Scheme 1. Resveratrol and 40-ester analogs.
Scheme 3. Synthesis of pivaloate and benzoate RV.
with these intermediates. Use of excess BF3ꢀOEt2 and Me2S was
again shown to reliably provide the new 40-pivaloyl 13 and the
benzoate 14 RV esters.
After initial evaluation via gene expression studies and in vitro
extracellular matrix protein assays, the 40-acetyl-RV 2 was selected
totestinvivo using young adult intact/cycling femaleLong-Evans rats.
This was done to establish safety/toxicity characteristics along with
quantifying regulatory behavioral parameters such as food/water in-
tact, bodyweightgain, whiteadiposetissue(WAT)/deposition/weight
and sequence/pattern of the estrous cycle for each animal. Also,
in vitro studies quantified hypothalamic aromatase enzyme activity,
which is a biomarker for reproductive function were performed. Final-
ly, behavioral studies using the Porsolt forced-swim test (PFST) were
used to index/quantify depressive-like behaviors.
Scheme 2. Decarbonylative Heck route to acetate 2.
Thirty-six female Long-Evans rats for two independent studies
(21 and 15 animals) of approximately 55 days of age were initially
placed on a phytoestrogen low diet (10 ppm/kg)13 to eliminate
other potential polyphenolic influences upon the measured param-
eters. The rats were allowed to habituate for two weeks then di-
vided into three treatment groups (age and body weight
matched) with a total of twelve animals in each group. The rats
were placed in the following groups: (1) DSMO control, (2)
20.0 mg /kg of 4AR 2 in DMSO and (3) 90.0 mg/kg of 4AR in DMSO.
At the end of a 2 week habituation period, they were given 0.1 mL
subcutaneous injections by treatment at the nape of the neck for
21 consecutive days. This was done since large oral doses of natural
resveratrol are needed to be effective in vivo and we wanted to
examine the maximal influence of 4AR 2 under conditions to test
the safety/toxicity, physiological and behavioral parameters.14
During the experimental treatment interval the rats were weighed
daily and food and water intake was measured twice (48 h inter-
vals). Before and during the treatment interval estrous cycle pat-
terns were monitored for eight days via cellular vaginal
histological examination.15 At approximately 90 days of age (the
last day of the treatments) all experimental groups were tested
in the PFST to quantify depressive-like behaviors, as reported pre-
viously.15 This is a standard anti-depressant test where in a typical
test, if a rat is forced to swim vigorously, over time some animals
become immobile which indicates behavioral despair. On the other
hand, animals that display increased levels of mobility times indi-
cate low levels of despair or depression. Therefore, the total time
an animal is immobile is an index of depression. The following
day all animals were sacrificed where WAT weight was recorded
and hypothalamic tissue was dissected to quantify aromatase en-
zyme activity levels.16
presence of excess amine base. Use of N-ethyl morpholine
(1.2 equiv) was also critical to the success of this key step. After
3 h at 120 °C, the coupling product, benzyl protected stilbene 9
was obtained in 88% yield. Use of other NHC ligands, N,N-di-t-butyl
(SItBu) and bis-2,6-diisopropyl (SIPr), gave lower yields of 9, 80%
and 72% for this step. At this point a variety of benzyl deprotection
conditions were explored that would maintain the sensitive alkene
and acetate ester functionality. Boron tribromide (CH2Cl2, ꢁ78 °C)
removed the benzyl ethers from 9,9 however a large amount of
by-product, close in Rf (tlc) to the product was also produced.
Use of aluminum chloride with N,N-dimethylaniline produced
numerous products with 9.10 Various palladium catalyzed selective
debenzylation conditions, with 1,4-cyclohexadiene or formate as
hydrogen sources, were also attempted without success.11 TMSI,
used previously with the corresponding di-MOM intermediate,
was again found not to be effective in this case with 9. Finally,
use of boron trifluoride etherate (15 equiv) with dimethylsulfide
(20 equiv)12 was found to provide the desired product, 40-acetyl-
RV 2, in 85% yield. The process was also found to be robust and
amenable to scale up to reproducibly access 2 in highly pure form.
On 50 g scale with 9 using BF3ꢀOEt2 (8 equiv) and Me2S (12 equiv)
with silica gel filtration and recrystallization (hexanes) pure prod-
uct 2 was obtained in 70% yield.
The new coupling route also allowed for the production of new
40-pivaloate and benzoate ester analogs in three additional steps
(Scheme 3). Hydrolysis of acetate 9 provided the protected 40-phe-
nol 10 in 90% yield. Acylation with pivaloyl chloride gave ester 11
in 83% yield and use of benzoyl chloride gave 12 in 87% yield. The
versatility of the benzyl ether removal conditions was also evident

M. J. Acerson et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2941–2944
2943
Table 1
Food/water intake
90
80
Treatment group
Food (g)
Water (mL)
70
Days 8 and 9 (48 h)
33.1 1.5
DMSO
53.3 2.7
20 4AR 2a
90 4AR 2b
34.3 1.4
35.2 1.9
Days 20 and 21 (48 h)
33.8 1.8
52.1 2.6
56.7 3.2
60
50
40
30
20
10
0
*
*
DMSO
50.7 3.5
50.9 2.9
55.8 4.1
20 4AR 2a
90 4AR 2b
35.2 2.1
36.9 4.5
a
20 4AR 2 group given 20 mg/kg 2.
90 4AR 2 group given 90 mg/kg 2.
b
Table 1 displays the food/water intake at 8–9 days and at 20–
DMSO
20 4AR
90 4AR
21 days of the treatments. Also, WAT weights and WAT weights ex-
pressed as a ratio to body weight (per 100 g of body weight) were
also measured. Food intake and water intake were normal (ranging
from 32 to 38 g/2 days and 49 to 59 mL/2 days, respectively. While
WAT weights among the treatment groups ranged from 6.6 to 8.2 g
and the ratios of WAT/body weight were also within a normal
range (2.4–3.0). Finally, Figure 1 displays the body weight gain
over the experiment interval for each treatment group and all gain
patterns appear normal.
Prior to the injection interval all rats in each treatment group
displayed a normal 4 day estrous cycle pattern.15 The estrous cycle
patterns during the treatment interval were normal however, 4 out
of the 12 rats in each treatment group displayed abnormal estrous
cycles most likely due to the stress of the daily injections. Thus,
there were no significant changes in estrous cycle patterns among
the treatments. When hypothalamic aromatase enzyme activity
was quantified, normal levels were observed among the treatment
groups (DMSO = 60 6.1; 20 mg/kg 4AR 2 = 71 7.4 and 90 mg/kg
4AR 2 = 73 6.9 pmol/h incubation/mg protein, respectively), sug-
gesting no significant alterations in hypothalamic aromatase levels
among the treated rats.
Figure 2. Time (seconds) immobile in the 6 min Porsolt forced-swim tests.
4AR 2 treatments significantly decrease depressive-like behaviors.
Finally, there were no significant differences in average swim speed
(meters/second), dives (for escape), inverts (partial dives) or number
of boli excreted (indexing emotionally) among the treatment
groups. In this study, food/water intake, body weight gain, WAT
deposition, pattern of the estrous cycles and hypothalamic aroma-
tase enzyme activity were not altered by the 4AR 2 treatments vs.
control values, suggesting that these 4AR 2 doses are safe and non-
toxic. Previous studies with higher oral doses suggest that resvera-
trol 1 can decrease weight gain and WAT deposition,17 whereas,
the lack of alterations in brain aromatase activity in the present
study are comparable with other polyphenolic treatments.16 Addi-
tionally, in human clinical studies the safety and health benefits of
resveratrol have been reported.18
When depressive-like behaviors were examined in the Porsolt
forced swim test (PFST) near the end of the treatment interval,
the 4AR 2-treated animals displayed the least amount of depres-
sion compared to control values. Speculation that 4AR 2 binds to
nuclear estrogen receptor beta (ERb) is supported by this data since
ERb expression is high in the frontal cortex, amygdala, hippocam-
pus and hypothalamic regions of the brain that influence depres-
sion and anxiety.19 Further support for this perspective is seen in
ERb-deficient mice that express increased anxiety/depressive-like
behaviors, whereas, when ERb agonists were administered in these
animals it decreased anxiety and depression.20 This may imply that
neurotransmitter synthesis, release and reuptake may be improved
by the 4AR 2 treatments such as that seen for serotonin that has
dramatic influences on mood and depression.15 This notion is sup-
ported by the findings of Xu et al., where anti-depressive effects of
resveratrol 1 were mediated via serotonin mechanisms.21 How-
ever, more recent evidence suggests that resveratrol can protect
against homocysteine induced oxidative stress and cognitive
impairment.22 In any event, further study for the potential of
4AR ester0s antidepressant actions or influences is warranted.
The results shown in Figure 2 indicate that rats treated with 4AR
2 at 20 mg/kg or 90 mg/kg displayed significantly less depressive-
like behaviors (immobile times/seconds) compared to the DMSO-
treated control animals. Since the mobile index is an inverse of the
immobility intervals they are not displayed graphically but was de-
scribed here: DMSO = 286 8.7, 4AR at 20 mg/kg = 312 4.9 and
4AR at 90 mg/kg = 319 5.3 in seconds, again suggesting that the
275
265
255
245
DMSO
Acknowledgments
20 4AR
90 4AR
We are grateful for support provided by the Brigham Young
University Cancer Center to M.J.A. and Y.W. and a TTO/LS Coll Grant
19-22156- to E.D.L. for funding.
235
225
Supplementary data
Treatment Interval
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2013.03.046.
These data include MOL files and InChiKeys of the most important
compounds described in this article.
Figure 1. Daily body weight profiles of the treatment groups. DMSO = control
group, 20 4AR = 20 mg/Kg of 4AR 2 and 90 4AR = 90 mg/kg of AR 2 administered (in
DMSO) daily; n = 12 animals per treatment group.

2944
M. J. Acerson et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2941–2944
8. Dinger, M. B.; Nieczypor, P.; Mol, J. C. Organometallics 2003, 22, 5291.
References and notes
9. Punna, S.; Meunier, S.; Finn, M. G. Org. Lett. 2004, 6, 2777.
10. Akiyama, T.; Hirofuji, H.; Ozaki, S. Tetrahedron Lett. 1991, 32, 1321.
11. Nishihama, Y.; Ishikawa, Y.; Nishiyama, S. Tetrahedron Lett. 2009, 50, 2801.
12. Fuji, K.; Kawabata, T.; Fujita, E. Chem. Pharm. Bull. J. 1980, 28, 3662.
13. Legg, R. L.; Tolman, J. R.; Lovinger, C. T.; Lephart, E. D.; Setchell, K. D. R.;
Christensen, M. J. Reprod. Biol. Endocrinol. 2008, 6, 57.
14. Andres-Lacueva, C.; Macarulla, M. T.; Rotches-Ribalta, M.; Boto-Ordonez, M.;
Urpi-Sarda, M.; Rodriguez, M.; Portillo, M. P. J. Agric. Food Chem. 2012, 60, 4833.
15. Blake, C.; Fabick, K. M.; Setchell, K. D. R.; Lund, T. D.; Lephart, E. D. BMC
Neurosci. 2011, 12, 28.
1. (a) Lancon, A.; Kaminski, J.; Tili, E.; Michaille, J.-J.; Latruffe, N. J. Agri. Food Chem.
2012, 60, 8783–8789; (b) Gescher, A. J. Planta Med. 2008, 74, 1651–1655; (c)
Signorelli, P.; Ghidoni, R. J. Nut. Biochem. 2005, 16, 449; (d) Garesveratrolin, S.;
Ollinger, K.; Dabrosin, C. Can. Lett. 2006, 231, 113.
2. (a) Wong, Y.; Osmond, G.; Brewer, K. I.; Tyler, D. S.; Andrus, M. B. Bioorg. Med.
Chem. Lett. 2010, 20, 1198; (b) Andrus, M. B.; Liu, J. Tetrahedron Lett. 2006, 47,
5811; (c) Yang, H.; Baur, J. A.; Chen, A.; Miller, C.; Sinclair, D. A. Aging Cell 2007,
6, 35.
3. (a) Walle, T. Ann. N.Y. Acad. Sci. 2011, 1215, 9; Kuhnle, G.; Spencer, J. P.;
Chowrimootoo, G.; Schroeter, H.; Debnam, E. S.; Srai, S. K. Biochem. Biophys. Res.
Commun. 2000, 272, 212; (b) Piver, B.; Fer, M.; Vitrac, X.; Merillon, J.-M.;
Dreano, Y.; Berthou, F.; Lucas, D. Biochem. Pharmacol. 2004, 68, 773.
4. Planas, J. M.; Alfaras, I.; Colon, H.; Juan, M. E. Arch. Biochem. Biophys. 2012, 527,
67.
5. Cuevas, C.; Perez, M.; Martin, M. J.; Chicharro, J. L.; Fernandez-Rivas, C.; Flores,
M.; Francesch, A.; Gallego, P.; Zarzuelo, M.; de la Calle, F.; Garcia, J.; Polanco, C.;
Rodriguez, I.; Manzanares, I. Org. Lett. 2000, 2, 2545.
16. Lephart, E. D.; Thompson, J. M.; Setchell, K. D. R.; Adlercreutz, H.; Weber, K. S.
Brain Res. 2000, 859, 123.
17. (a) Lai, C.-Y.; Yang, J.-Y.; Rayalam, S.; Della-Fera, M. A.; Ambati, S.; Lewis, R. D.;
Hamrick, M. W.; Hartzell, D. L.; Baile, C. A. J. Med. Food 2011, 14, 1352; (b) Kim,
S.; Jin, Y.; Choi, Y.; Park, T. Biochem. Pharmacol. 2011, 81, 1343.
18. Smoliga, J. M.; Baur, J. A.; Hausenblas, H. A. Mol. Nutr. Food Res. 2011, 55, 1129.
19. (a) Lephart, E. D.; Hedges, D. W. In Focus on Alzheimer’s Disease Research; Welsh,
E. M., Ed.; Nova Biomed. Books, 2003; pp 51–68; (b) Walf, A. A.; Frye, C. A.
Neuropsychopharmacology 2006, 31, 1097.
6. Szirmai, M.; Odqvist, H.; Halldin, M. M. J. Labelled Compd. Radiopharm. 1996, 38,
309.
20. Walf, A. A.; Frye, C. A. Steroids 2008, 73, 997.
21. Xu, Y.; Wang, Z.; You, W.; Zhang, X.; Li, S.; Barish, P. A.; Vemon, M. M.; Du, X.;
Li, G.; Pan, J.; Ogle, W. O. Eur. Neuropsychopharm. 2010, 20, 405.
22. Koz, S. T.; Etem, E. O.; Baydas, G.; Yuce, H.; Ozercan, H. I.; Kulogu, T.; Koz, S.;
Etem, A.; Demir, N. Brain Res. 2012, 1484, 29.
7. (a) Andrus, M. B.; Liu, J.; Meredith, E. L.; Nartey, E. Tetrahedron Lett. 2003, 44,
4819; (b) Blaser, H.-U.; Spencer, A. J. Organomet. Chem. 1982, 233, 267. Other
coupling routes to RV; (c) Guiso, M.; Marra, C.; Farina, A. Tetrahedron Lett. 2002,
43, 597; (d) Jeffery, T.; Ferber, B. Tetrahedron Lett. 2003, 44, 193; (e) Iijima, T.;
Makabe, H. Biosci., Biotechnol., Biochem. 2009, 73, 2547.