Synthesis and evaluation of butein derivatives for in vitro and in vivo inflammatory response suppression in lymphedema

Article abstract of DOI:10.1016/j.ejmech.2020.112280

Herein, we demonstrate that butein (1) can prevent swelling in a murine lymphedema model by suppressing tumor necrosis factor α (TNF-α) production. Butein derivatives were synthesized and evaluated to identify compounds with in vitro anti-inflammatory activity. Among them, 20 μM of compounds 7j, 7m, and 14a showed 50percent suppression of TNF-α production in mouse peritoneal macrophages after lipopolysaccharide stimulation. Compound 14a, exhibited the strongest potency with an in vitro IC₅₀ of 14.6 μM and suppressed limb volume by 70percent in a murine lymphedema model. The prodrug strategy enabled a six-fold increase in kinetic solubility of compound 1 and five-fold higher levels of active metabolite in the blood for compound 14a via oral administration in the pharmacokinetics study. We suggest that the compound 14a could be developed as a potential therapeutic agent targeting anti-inflammatory activity to alleviate lymphedema progression.

Full text of DOI:10.1016/j.ejmech.2020.112280

Journal Pre-proof
Synthesis and evaluation of butein derivatives for in vitro and in vivo inflammatory
response suppression in lymphedema
Kangsan Roh, Jung-hun Lee, Hee Kang, Kyewon Park, Youngju Song, Sukchan Lee,
Jin-Mo Ku
PII:
S0223-5234(20)30249-X
DOI:
https://doi.org/10.1016/j.ejmech.2020.112280
EJMECH 112280
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 31 December 2019
Revised Date: 26 March 2020
Accepted Date: 26 March 2020
Please cite this article as: K. Roh, J.-h. Lee, H. Kang, K. Park, Y. Song, S. Lee, J.-M. Ku, Synthesis
and evaluation of butein derivatives for in vitro and in vivo inflammatory response suppression
in lymphedema, European Journal of Medicinal Chemistry (2020), doi: https://doi.org/10.1016/
j.ejmech.2020.112280.
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 Masson SAS.

GRAPHICAL ABSTRACT

Synthesis and evaluation of butein derivatives for
in vitro and in vivo inflammatory response
suppression in lymphedema
2†
1†
3†
4
5
2
Kangsan Roh , Jung-hun Lee , Hee Kang , Kyewon Park , Youngju Song , Sukchan Lee *,
1
Jin-Mo Ku *
1
Bio-Center, Gyeonggido Business & Science Accelerator, 147 Gwanggyo-ro, Suwon
16229, Republic of Korea
2
Department of Integrative Biotechnology, Sungkyunkwan University, Suwon 16419,
Republic of Korea
3
Humanitas College, Kyung Hee University, Yongin 17104, Republic of Korea
4
Department of Food Science and Biotechnology, Sungkyunkwan University, Suwon 16419,
Republic of Korea
5
Department of Biomedical Science and Technology, Graduate School, Kyung Hee
University, Seoul 02447, Republic of Korea
*
To whom correspondence should be addressed.
1

For J.-M. Ku: phone, 82-31-888-6987; fax, 82-31-888-6979; e-mail, medichem@gbsa.or.kr
For S. Lee: phone, 82-31-290-7866; e-mail, cell4u@skku.edu
*
Authors for co-correspondence
†
These authors contributed equally to this work and should be considered co-first authors.
KEYWORDS: Acquired lymphedema, inflammation, chalcones, therapeutic agent
2

GRAPHICAL ABSTRACT
ABSTRACT:
Herein, we demonstrate that butein (1) can prevent swelling in a murine lymphedema model
by suppressing tumor necrosis factor α (TNF-α) production. Butein derivatives were
synthesized and evaluated to identify compounds with in vitro anti-inflammatory activity.
Among them, 20 µM of compounds 7j, 7m, and 14a showed 50% suppression of TNF-α
production in mouse peritoneal macrophages after lipopolysaccharide stimulation. Compound
1
4a, exhibited the strongest potency with an in vitro IC of 14.6 μM and suppressed limb
50
volume by 70% in a murine lymphedema model. The prodrug strategy enabled a six-fold
increase in kinetic solubility of compound 1 and five-fold higher levels of active metabolite
in the blood for compound 14a via oral administration in the pharmacokinetics study. We
3

suggest that the compound 14a could be developed as a potential therapeutic agent targeting
anti-inflammatory activity to alleviate lymphedema progression.
INTRODUCTION
Lymphedema occurs when an abnormality or injury in the lymphatic system blocks adequate
lymph drainage from the arm or leg. Lymphedema is classified into congenital lymphedema
1
and acquired lymphedema. Acquired lymphedema is often caused by cancer treatments,
2
such as surgery, radiation, and chemotherapy, or after physical trauma or infections. The
manifestations of lymphedema include swelling, reduced functionality of the limbs, and
subsequent secondary infections, lead to psychological distress and detrimental effects on
3
social and work life. Currently, there is no therapeutic agent that can inhibit lymphedema
progression or alleviate its symptoms. Complete decongestive therapy (CDT), a
physiotherapeutic approach, is commonly accepted as the primary treatment for lymphedema
4
as it aims to decrease fluid accumulation.
Although many efforts have been underway to find new targets for systemically controlling
lymphedema, its molecular mechanisms are not clearly defined. It is known that the
5
accumulation of interstitial fluid leads to inflammation, fat deposition, and fibrosis. Many
reports have indicated that immune responses related to inflammatory cell infiltration play a
6
critical role in the pathophysiology of lymphedema. Upregulation of inflammatory genes
7
,8
have been observed in both murine models and patients with lymphedema.
Based on
previous studies, inflammatory responses were necessary for lymphedema formation during
9
,10
adipose tissue inflammation in obese mouse models.
4
Swapna et al. also revealed that the

proinflammatory responses caused by macrophage infiltration can be significantly affected in
1
1
response to lymphedema. Patients using CDT showed reduced expression of tumor necrosis
1
2
factor(TNF)-α and other proinflammatory mediators.
These data suggest that anti-
inflammatory agents can be potential targets for alleviating lymphedema progression.
Several anti-inflammatory compounds have been tried in experimental lymphedema
models and have shown promising effects. Ketoprofen, a commonly used nonsteroidal anti-
inflammatory drug (NSAID), was subcutaneously applied to a murine tail and it ameliorated
1
3
swelling and other pathologic symptoms. Tacrolimus, an anti-T-cell agent used for chronic
cutaneous inflammation and fibrosis, was successful in preventing and improving
1
4
lymphedema in the above-mentioned lymphedema model.
Although ketoprofen and
tacrolimus are currently being assessed as novel clinical lymphedema therapies, their
potential toxicities may limit their long-term use.
In our investigation into finding therapeutic candidates, we previously established a new
lymphedema mouse model that mimics lymphedema in cancer patients by removing a
superficial inguinal lymph node, a popliteal lymph node, a deep inguinal lymph node, and the
femoral lymphatic vessel. Using the model, we found that Rhus verniciflua Stokes (RVS)
1
5
ameliorated hind leg edema. RVS extract contains several anti-inflammatory polyphenols,
including sulfuretin and compound 1. Compound 1 is more potent than other polyphenols,
e.g., sulfuretin, in suppressing lipopolysaccharide (LPS)-induced nitrite and prostaglandin E2
1
6
production in macrophages. In addition to its anti-inflammatory activity, compound 1 can
1
7,18
inhibit adipogenesis and fibrosis.
These properties prompted us to explore the systemic
effect of compound 1 on lymphedema. At the same time, we attempted to optimize the
chemical structure of compound 1 in order to improve in vitro and in vivo efficacy through
structure-activity relationship (SAR) analysis. The goal was to produce a prodrug of
5

compound 1 to enhance the potency, solubility, permeability, and pharmacokinetic
parameters, including half-life and pharmaceutically active levels in the blood.
RESULTS AND DISCUSSION
In Vitro and In Vivo Lead Compound Validation
As an initiative to find therapeutic candidates, our strategy is to choose a lead compound that
can suppress TNF-α, the master inflammatory cytokine that is mainly produced by activated
macrophages and recruit circulating neutrophils to injured tissue. Compound 1 was tested for
its ability to inhibit the production of TNF-α in lipopolysaccharide (LPS)-stimulated
macrophages. Compound 1 decreased TNF-α production by 10% at 20 μM and had a dose-
dependent IC of 43.3 μM (supporting information 1A).
5
0
We verified that swelling of the affected limb decreased with oral administration of
compound in a murine lymphedema model. The limb volume was reduced with compound 1
treatment on days three and seven post-surgery (supporting information 1B). Oral
administration of compound 1 (200 mg/kg/d) caused dramatic suppression of lymphedema,
2
reducing the volume by 41.6% (41.5 mm ) compared with that of the untreated mouse (99
2
mm ). H&E staining of the affected lower limb supported this finding (supporting
information 1C). Thereby, we determined that compound 1 is a promising anti-inflammatory
lead compound through in vitro and in vivo work, suggesting the need for further lead
6

optimization of compound 1 to improve the potency and efficacy via structure-activity
relationship (SAR) studies.
Lead Optimization Through Chemical Modification and Anti-inflammatory
Evaluation
Compound 1 exhibits poor water solubility (79 μM, 21.0 μg/mL), with a partition coefficient,
(
logP) of 0.42. In addition, it has low potency (IC , 43.3 μM) and insufficient efficacy, with
50
only 21% inhibition at 20 μM in vitro, indicating that it is not a suitable candidate for
evaluation in vivo. To address these deficiencies, we performed structure-activity-guided
modifications of compound 1 to improve its physicochemical properties (solubility, logP) and
pharmacokinetic parameters, including maximum concentration of a drug (Cmax),
bioavailability (BA), half-life (T ), and total concentration of the drug (area under curve,
1
/2
AUC) after oral administration.
Compound 1 is one of natural product which is structurally similar to isoliquiritigenin (2),
echinatin (3), and licochalcone B (4), all of which have a chalcone backbone with phenols. .
(
Figure 1A). We compared the inhibitory effects of compounds 1 and 3 and ketoprofen at 20
μM on LPS-stimulated TNF-α production in macrophages and found a similar activity
Supporting information 2A). Ketoprofen had no effect in this model. Subsequently, we
(
planned to modify the structure of compound 1 via structure-activity relationship studies by
functional group interconversion to inhibit TNF-α production in macrophages.
Functional group modifications on each aromatic ring in sections A and B of compound 1
helped identify pharmacophores for biological activity (Figure 2). Catechol and quinone
derivatives are used in cancer studies but they are susceptible to oxidation and reduction and
7

1
9
show low chemical stability, making them unsuitable in drug discovery. Therefore,
substitutions of the catechol in section B with other functional groups, such as F or OCH3,
were pursued.
The compounds with newly introduced functional groups on each aromatic ring (Figure 2,
section A and B) and the cyclized compounds at 20 μM were evaluated for their potential to
inhibit TNF-α production in LPS-stimulated macrophages (Figure 3). First, we assessed the
2
activity of cyclized compound 9, because the 2-hydroxy group at R readily attacks an
intramolecular Michael acceptor to generate the cyclic form, a flavanone, which could be
synthesized during flavonoid biosynthesis. However, compound 9 was a weak inhibitor, so
we continued to focus on functional group modifications of the open-chain structure, the
chalcones. The replacement of catechol was performed to generate stable and relevant
bioisosteres because catechol is susceptible to oxidation and reduction and is labile to
metabolism in vivo (Figure 3). Both hydroxyl groups at the catechol in section B were
substituted with OCH and F (compounds 7a-e), which resulted in the loss of TNF-α
3
4
inhibition. Although the substitution of the hydroxyl group with the fluoride at R generated
compound 7d (22% inhibition) with inhibitory activity similar to compound 1 (20%
inhibition), compound 7d was readily decomposed at room temperature in the presence of
oxygen.
2
Next, the replacement of hydroxyl group R was performed to enhance the chemical
stability and inhibitory activity of TNF-α production as the hydroxyl group is a nucleophile
and is prone to intramolecular nucleophilic attacks, which lead to the formation of a cyclized
compound. Substitutions of 2-hydroxyl by 2-fluoro (7g, 32% inhibition) and 2-methyl (7h,
2
4
1% inhibition) at the R on the A ring showed more efficacy for TNF-α inhibition compared
8

to compound 1 (20% inhibition). Also, no cyclized compounds were observed during
synthesis of the desired compounds.
2
Compounds 7i and 7j replaced hydroxyl groups with alkoxy groups at C and significantly
suppressed TNF-α production. The more sterically hindered alkoxy group presented with
greater inhibition of TNF-α production (OMe < OiPr). Compound 7j, bearing the more
sterically bulky isopropoxyl group, showed the strongest activity at 64% inhibition. However,
unfortunately, 7J decomposed at ambient temperature. Also, mono-substituted compounds 7l
6
(61% inhibition) with a 2-chloro and compound 7m (62% inhibition) with a 2-methyl at R
showed similar efficacy to compound 7j (64% inhibition). Furthermore, we also used a
prodrug strategy to modify the physicochemical properties of compound 1 due to its low
kinetic solubility (21 μg/mL) and low permeability (-5.35) in a parallel artificial membrane
permeability assay (PAMPA) (Table 2, entry 1).
2
To design prodrugs of compound 1, we modified the hydroxyl group at R to prevent
cyclization, which could occur during flavonoid biosynthesis reactions. Compounds 14a-d,
8
a, and 8b contained various acetyl and aryl esters that are typically cleaved by ubiquitous
2
0,21
hydrolases in the plasma and during first-pass metabolism.
These compounds were also
synthesized and evaluated in the TNF-α production inhibition assay at a single concentration
2
(20 μM) (Figure 4). Compound 14a (51% inhibition) containing a benzoyl ester at R
strongly reduced TNF-α production and globally acetylated compound 8a (73% inhibition) at
5
R on the A and B rings from compound 1, exhibited highly reproducible inhibitory activity.
To confirm the anti-inflammatory activities, we examined the dose-dependent inhibition of
TNF-α production by LPS-stimulated to determine the half-maximal inhibitory concentration
2
(
IC ) (Table 1). Substitution of R with an electron-donating group, such as methyl or
5
0
9

methoxy, on the A section of the chalcone generated compounds with moderate IC . The
5
0
inhibitory activity of compounds 7h and 7i was similar to that of compound 1 (IC , 43.3
5
0
μM). Interestingly, compounds 7l (IC , 18.6 μM) and 7m (IC , 17.3 μM), with the weak
5
0
50
electron donating/withdrawing-group instead of the hydroxyl group, were two-fold more
active than compound 1. However, compound 7m was not a suitable lead due to low
solubility (28 μg/mL). Phase I metabolic stability using liver microsomes from mice, rats, and
humans was evaluated by examining the percent of compound remaining 30 minutes after the
compound was reacted. Compound 7m exhibited relatively low metabolic stability in Phase I
metabolism in the various species (mouse, 7%; rat, 21%; and human, 38%) (Table 2, entry 2;
Table 3, entry 2). Interestingly, compound 14s has a relatively strong inhibition of TNF-α
production at a single dose (20 μM), which correlated with potency in a dose-dependent
manner (Figure 4 and Table 1, entries 6-9). In a test using compounds 14b (IC , 18.9 μM),
5
0
1
4c (IC , 20.8 μM), and 14d (IC , 17.9 μM), small donating/withdrawing groups, including
50 50
methyl, methoxy, and chloro at the para-position of the benzoyl ester, exhibited two-fold
more or less inhibition of LPS-induced TNF-α production than that of compound 1, results
which were similar to that of compound 14a (IC , 14.6 μM) (Table 1, entries 6-9).
5
0
5
Compound 8a, which is globally acetylated at R on the A and B rings, exhibited the
strongest potency (IC 12.8 μM) (Table 1, entry 10). However, we decided not to test this
5
0
compound in the lymphedema animal model because it was not miscible with water at high
concentrations and tended to form droplets in aqueous solutions.
Importantly, we found that compounds 14a-d were relatively strong inhibitors of TNF-α
production in the single-dose test and the dose-dependency assay. We assumed that
compound 1 was the pharmacologically active component for TNF-α production inhibition
and investigated whether the promoiety of the prodrug could be readily cleaved by enzymatic
1
0

hydrolysis. Based on the activity profile and the physicochemical properties, we chose
compound 14a for an animal experiment using oral administration and compared its efficacy
and blood concentration with those of compound 1. The water solubility of compound 14a
(136 μg/mL) was approximately five-fold higher than that of compound 1 (21 μg/mL) (Table
2
, entry 1, 3). The blood exposure levels of the compounds during systemic circulation were
compared by integrating the area under the curve (AUC) after oral administration of 100
mg/kg of compound 1 and 14a. We found that compound 14a was readily metabolized into
compound 1 by ubiquitous hydrolytic cleavage activities in the blood plasma, even after
intravenous administration. Compound 14a was not detectable above the limit of quantitation
(
LOQ, < 1 ng/mL). We decided that the oral administration of compound 14a could be
estimated by monitoring compound 1 as the pharmaceutically active ingredient as it is the
main metabolite derived from compound 14a. In the pharmacokinetic study using oral
administration, it was more efficient to use compound 14a (AUC, 1224 h × ng/mL) for the
delivery of compound 1 into the systemic circulation than using unmodified compound 1
(AUC, 251 h × ng/mL).
Preventive Effect of Orally Administered Compound 14a on Lymphedema Formation
Since oral administration of compound 1 decreased swelling in the murine lymphedema
model, we hypothesized that the inhibitory effects of compounds 14a and 1 may be similar,
but the altered pharmacokinetic properties of compound 14a, such as drug delivery through
oral dose, could be an advantage. To test this hypothesis, we pretreated mice with compounds
1
and 14a via oral administration for seven days before surgical induction of lymphedema.
The representative dorsal images from treated mice unequivocally indicated the suppression
of lymphedema swelling in the right hind limb (Figure 5A). At day seven post-surgery, the
lymphedema volume was decreased by 52%, 53%, and 70% with treatment using compounds
1
1

1
, 7m, and 14a, respectively, compared with the vehicle treated lymphedema group (Figure
5
).
Histology and Molecular Analysis
We investigated H&E images from the limbs of mice in the normal group, vehicle-treated
lymphedema group, and the lymphedema groups treated with compounds 1 or 14a (Figure
6
A). As a result, compound 14a reduced neutrophil infiltration and ameliorated dermal layer
lymphedematous tissue. TNF-α mRNA expression was assessed in the lymphedematous
tissue and revealed decreased TNF-α expression with compound 1 and 14a treatment. The
group treated by compound 14a had the lowest level of TNF-α mRNA expression, although
there was no statistical difference (P-value= 0.613) in expression between the compound 1
and 14a treatment groups. In addition, we examined the regulatory effects of compound 1 and
compound 14a on adipogenesis observed in lymphedema. The representative histologic
images indicated that both compounds prevented the development of adipose tissue in the
dermal layer (Figure 7A). Peroxisome proliferator-activated receptor γ (PPARγ) is the master
regulator of adipogenesis. Immunofluorescence staining showed that the distribution of
PPARγ protein was suppressed in the dermal layer of lymphedematous tissue following
treatment with compounds 1 and 14a (Figure 7B). Western blot analysis revealed that these
compounds decreased PPARγ and its target protein, fatty acid-binding protein 4 (Fabp4) in
the affected limbs (Figure 7C). Consistently, the levels of PPARγ and Fabp4 mRNA were
reduced in both treatment groups, with compound 14a being more potent than compound 1
(Figure 7D). These molecular analyses showed that compounds 1 and 14a displayed a similar
1
2

anti-inflammatory activity but compound 14a was more efficient in ameliorating
adipogenesis during lymphedema development.
Synthesis of Compounds 7s and 14s
Compound 1 is structurally classified as a chalcone, which can be referred to as an open-
chain flavonoid. Two aromatic rings are linked by a double bond and a carbonyl, which are
fully conjugated. Several reports on the synthesis of chalcones have described various
methods, with the most common being the base-catalyzed Claisen-Schmidt reaction in which
the condensation of ketones with aldehydes is performed in the presence of an aqueous base
2
2,23,24,25,26
solution.
However, this method has limitations leading to the decomposition of the
phenol/catechol involving acidic protonation of the substrate, which affects the synthesis
yield and the diversity of substances.
Syntheses of the various compounds 7s are described in Scheme 1. The Claisen
o
condensation of compounds 5a-j and 6a-e at 25 C under basic conditions afforded
compounds 1 and 7a-m (method A). Compound 1 was acylated with acetic anhydride and
benzoyl chloride to produce compounds 8a and 8b, respectively. Because possible
degradation of catechol/phenol in compounds 5b and 5f were observed, we alternatively
performed Cu(II)-mediated catalytic aldol condensation by oxidation/reduction of Cu(II) in a
2
7
tandem process to produce compounds 7h and 7i (method B). The synthesis of 2-acylated
compounds 14a-d bearing a para-substituted benzoyl ester as a prodrug is presented in
Scheme 2. The ester bond readily enables cleavage by ubiquitous hydrolases in the
gastrointestinal tract and blood plasma. The selective acylation at the C2-hydroxyl group of
compounds 14a-d required global hydroxyl protection of acetophenone 10, because the
1
3

hydroxyl group at the para-position of acetophenone 10 was predominantly reacted with acyl
chlorides. Compound 5a was reacted with methoxymethyl chloride (MOM-Cl) to generate
compound 10, which was selectively protected at the ortho hydroxyl group. Compound 11
was obtained by reacting compound 6a with MOM-Cl. Claisen condensation of compounds
1
0 and 11 for 6 h in the presence of 60% KOH (aq) was carried out to produce compound 12.
Esterification of compound 12 with various acyl chloride compounds in dimethylformamide
DMF) for 3 h was followed by global MOM deprotection of compounds 13a-d to produce
(
the desired products (14a-d).
CONCLUSION
Our findings indicate that compounds having anti-inflammatory activities are potential
pharmaceutical agents against the pathogenesis of acquired lymphedema. By screening
natural products, we identified compound 1 as being able to suppress LPS-induced TNF-α
production in inflammation. Dose-dependent inhibitory activity was confirmed in a murine
peritoneal macrophage model, in vitro. For pharmaceutical properties, we modified its
structure by functional group conversion to enhance the potency, solubility, and permeability.
Furthermore, we designed a prodrug which could be readily converted to a pharmaceutically
active ingredient by ubiquitous hydrolase in vivo activity. As a result, we determined that
compound 7m showed a three-fold higher potency (17.3 μM) for anti-inflammatory effects in
vitro than compound 1 (43.3 μM). Despite its insufficient solubility (28 μg/mL), compound
7
m presented with enhanced permeability and could suppress limb swelling (53%) with oral
administration (100 mg/kg/day). Compound 14a was the most potent (14.6 μM) and had five-
fold higher solubility (136 μg/mL). As a prodrug, compound 14a was immediately converted
into compound 1 by the liver microsomes from three species, including mice, rats, and
1
4

humans. Compound 14a also had the advantage of being at a higher concentration in the
blood (AUC, 1224 hr × ng/mL) than compound 1 (AUC, 251 hr × ng/mL) when administered
orally. Correspondingly, compound 14a suppressed limb swelling by 70% with oral
administration (100 mg/kg/day). Compound 14a showed anti-inflammatory activity by
decreasing TNF-α mRNA expression and neutrophil infiltration, and ameliorated the
lymphedema in the dermal layer, as seen by H&E staining. Conclusively, the orally
administered compound 14a contributed to pharmacological suppression of inflammation and
can be a potential therapy for the treatment of lymphedema.
EXPERIMENTAL SECTION
Cell Culture
For peritoneal macrophage isolation, male Balb/c mice were injected intraperitoneally with 2
mL of 3.5% sterile thioglycollate (BD, Sparks, MD, USA) four days before sacrifice. Mice
were sacrificed via cervical dislocation and the peritoneal exudate cells were aseptically
isolated via peritoneal lavage with cold DMEM (HyClone, Logan, UT, USA) containing 10%
fetal bovine serum (FBS; HyClone) and 1% penicillin-streptomycin. After centrifugation, the
cells were resuspended and counted using a TC20 Automated Cell Counter (Bio-Rad
Laboratories, Hercules, CA, USA). Peritoneal exudate cells were plated in 24-well plates
overnight at 37 ºC, and the non-adherent cells were removed. The cells were then stimulated
with 100 ng/mL LPS for 24 h. The supernatant was collected for cytokine analysis. The
levels of TNF-α in the supernatants were determined using BD OptEIA mouse ELISA sets
(BD Biosciences, San Diego, CA, USA) according to the manufacturer’s protocol.
Animals
1
5

All animal studies were approved by the Institutional Animal Care and Use Committee
IACUC) of Sungkyunkwan University School of Medicine (Approval Number:
(
SKKUIACUC-20150037). All methods were conducted in accordance with the approved
guidelines. The murine model for acquired lymphedema was surgically induced in the lower
2
0
limb of male ICR mice, using a protocol that has been previously established. Briefly, the
major lymphatic components, including superficial inguinal lymph node, popliteal lymph
node, deep inguinal lymph node, and a femoral lymphatic vessel were removed and limitedly
cauterized in the right lower limb of mice that were under anesthesia.
Lymphedema Volume Quantitation
Lymphedema size measurements were conducted by observers blinded to the treatment status
of the mice. Limb volumes were calculated on postoperative days three and seven using a
2
0, 21
caliper as previously described.
Histology and Immunofluorescence
Tissue samples were harvested seven days post-surgery 15–25 mm distal from the right lower
limb. The collected tissues were fixed in 10% formalin solution. The specimens were cross-
sectioned at a thickness of 5 μm. The sections were deparaffinized in 100% xylene,
rehydrated in a graded ethanol series (100%, 95%, 90%, 80%, and 70%), and rinsed in water.
Sections were stained with a hematoxylin solution at RT. After washing in PBS, sections
were then stained in eosin at RT. Images were captured with a Pannoramic Viewer
microscope (3DHISTECH Ltd., Hungary). Prepared slides were incubated with anti-mouse
PPARγ (Santa Cruz Biotechnology, USA), Alexa Fluor 488, and DAPI (Life Technologies,
1
6

USA). Samples were analyzed under a Nikon ECLIPSE 80i microscope (Nikon Instruments
Inc., Japan).
Quantitative Real-Time Polymerase Chain Reaction (RT-PCR)
Total RNA was prepared from affected tissues using TRI reagent (Invitrogen, Carlsbad, CA).
Complementary DNA was synthesized from 500 ng of total RNA using M-MLV Reverse
Transcriptase (Promega, USA). Quantitative RT-PCR was performed using a SYBR Mix
(
TOYOBO, Osaka, Japan). The gene-specific primer sets for PCR were as follows:
peroxisome proliferator-activated receptor (Pparγ) forward, 5ʹ-
CCATTCTGGCCCACCAAC-3ʹ and reverse, 5ʹ-GCACCTGCACCAGGGC-3ʹ;
CCAAT/enhancer-binding protein α (C/ebpα) forward, 5ʹ-GCGGGCAAAGCCAAGAA-3ʹ
and reverse, 5ʹ-GCGTTCCCGCCGTACC-3ʹ; Fabp4 forward, 5ʹ-
CACCGCAGACGACAGGAAG-3ʹ and reverse, 5ʹ-GCACCTGCACCAGGGC-3ʹ; CD36
forward, 5ʹ-GGCCAAGCTATTGCGACAT-3ʹ and reverse, 5ʹ-
γ
CAGATCCGAACACAGCGTAGA-3ʹ.
Western Blot Analysis
2
Western blotting was performed as previously described . Briefly, proteins were extracted
from tissue samples using 500 μl of PRO-PREP protein extraction solution (Intron
Biotechnology, Seoul, Korea). Proteins were incubated with Pparγ (1:1000, Santa Cruz
Biotechnology, USA) and Fabp4 (1:1000, Santa Cruz Biotechnology, USA).
1
7

Statistical Analysis
The statistical significance of the differences between treatments was considered using an
unpaired one-tailed Student’s t-test and a one-way ANOVA, followed by the Bonferroni
post-hoc test, where appropriate. All data are presented as mean ± standard deviation. P-
values were presented to show the level of significance.
Pharmacokinetic Analysis of Compounds 1 and 14a
Male CD-1 mice (5 weeks) were purchased from Charles River Laboratories (Wilmington,
MA). Each mouse was administered with a test compound via tail vein at 1 mg/kg or oral
administration at 20 mg/kg or 100 mg/kg. Three animals were used for each independent
serial sampling experiment. Blood samples were collected at 5, 15, 30, 60, 120 (or 180), 360,
and 1440 min from each animal via the saphenous vein; 20-30 μL was drawn at each time
point and 15 μL of blood was used for quantitative bioanalysis by HPLC-MS/MS. The
fundamental pharmacokinetic parameters (half-life, clearance, volume of distribution, and
AUC) were obtained from non-compartmental analysis using Win-Non-lin. This experiment
performed simultaneous analysis of the analytes via the multiple reaction mirroring (MRM)
mode using TQ5000 LC/MS/MS (AB/SCIEX) combined with HTC-PAL.
General Methods
All substances were purchased from Sigma-Aldrich (St. Louis, MO) and used as received.
Purchased anhydrous methanol was used without further purification. Analytical thin-layer
chromatography was conducted on E. Merck TLC plates (silica gel 60 F , aluminum back).
2
54
1
8

Silica gel 60 (230–400 mesh) for column chromatography was purchased from E. Merck.
Melting points were measured with a Kofler block or a Büchi melting point apparatus, model
1
13
B-545. The H and C NMR spectroscopic data were recorded in CDCl , DMSO-d and
3
6,
CD OD solutions at ambient temperature with Avance II 400 (Bruker BioSpin, Germany)
3
1
13
(
400 MHz for H NMR and 100 MHz for C NMR) and Ascend III 700 (Bruker BioSpin,
1
13
Germany) (700 MHz for H NMR and 175 MHz for C NMR). Chemical shifts were
recorded as δ values in parts per million (ppm) and were indirectly referenced to
tetramethylsilane by the solvent signal. Coupling constants (J) were reported in Hertz.
Infrared spectra were recorded on a Nicolet 6700 FTIR spectrophotometer. Low-resolution
mass spectra were measured on a Micromass Quattro Micro API (Waters, USA), whereas
high-resolution mass spectra were recorded on an LTQ Orbitrap XL (Thermo Scientific,
USA) spectrometer. Characterization data for known compounds were consistent with
literature values.
Representative Compounds and Their Synthetic Procedures
All compounds were synthesized as shown in Scheme 1 (compounds 1, 7a-m, 8a-b, and 9) or
Scheme 2 (compounds 14a-d), except the commercially available compounds (compounds 2,
3
).
General Procedure for Butein Derivatives
Method A
Syntheses of (E)-1-(2, 4-dihydroxyphenyl)-3-phenylprop-2-en-1-one (1)
1
9

3
,4-Dihydroxybenzaldehyde 6a (1.0 g, 7.2 mmol) and 1-(2,4-dihydroxyphenyl)ethan-1-one
a (1.1 g, 7.2 mmol) were dissolved in ethanol (10 mL). An aqueous solution of 60% KOH
5
(1 mL) was added and the mixture was incubated at room temperature for 6 h. Solvents were
removed under reduced pressure and the residue was extracted three times with 30 mL of
diethyl ether. The diethyl ether layer was washed three times with 30 mL of water and dried
over anhydrous MgSO . After solvent removal, the product was purified by silica gel column
4
chromatography (silica gel, n-hexane-ethyl acetate, 1:1) to produce the indicated product.
(E)-1-(2,4-dihydroxyphenyl)-3-(3,4-dihydroxyphenyl)prop-2-en-1-one (1) as a yellow solid
1
(
42%); H NMR (MeOD, 400 MHz) δ 7.95 (1H, d, J = 8.8 Hz), 7.73 (1H, d, J = 15.6 Hz),
7
.54 (1H, d, J = 15.6 Hz), 7.20 (1H, d, J = 2.0 Hz), 7.12 (1H, dd, J = 8.4, 2.4 Hz), 6.83 (1H,
1
3
d, J = 8.4 Hz), 6.43 (1H, dd, J = 8.8, 2.4 Hz), 6.31 (1H, d, J = 2.4 Hz; C NMR (100 MHz,
MeOD) δ 193.50, 167.52, 166.37, 149.95, 146.86, 146.11, 133.30, 128.45, 123.64, 118.31,
+
1
16.63, 115.84, 114.74, 109.17, 103.85; LRMS (ESI) calcd. for C H O [M+H] : 273.08
1
5
13
5
+
found : 273.10; HRMS (ESI) calcd. for C H O [M+H] : 273.0685, found : 273.0754; m.p.
1
5
13
5
=
214.7 °C.
Method B
1
-(4-Hydroxy-2-methylphenyl)ethan-1-one
5b
(150
mg,
1.0
mmol),
2,4-
dihydroxybenzaldehyde 6a (138 mg, 1.0 mmol), and CuBr (22 mg, 0.1 mmol) in a pressured
2
tube were dissolved in 5 mL ethyl acetate at room temperature. The reaction mixture was
stirred at 80 °C. After 12 h the mixture was cooled to room temperature and filtered through
Celite. The organic layer was concentrated in vacuo and the product, 7h (66%), was a yellow
solid that was purified by silica gel column chromatography (silica gel, n-hexane-ethyl
acetate, 1:1).
2
0

(E)-1-(2,4-Dihydroxyphenyl)-3-(4-hydroxy-3-methoxyphenyl)prop-2-en-1-one (7a)
1
The title compound via method A was isolated as a yellow solid (46%); H NMR (MeOD,
4
00 MHz) δ 7.93 (1H, d, J = 8.8 Hz), 7.69 (1H, d, J = 15.2 Hz), 7.55 (1H, d, J = 15.2 Hz),
.27 (1H, d, J = 2.0 Hz), 7.12 (1H, dd, J = 8.4, 2.0 Hz), 6.75 (1H, d, J = 8.0 Hz), 6.31 (1H,
7
1
3
dd, J = 8.8, 2.4 Hz), 6.19 (1H, d, J = 2.4 Hz) 3.84 (3H, s); C NMR (100 MHz, MeOD) δ
1
1
93.4, 167.5, 166.4, 151.7, 148.0, 145.5, 133.4, 129.5, 123.5, 119.2, 115.2, 114.7, 112.5,
+
09.2, 103.8, 56.56; LRMS (ESI) calcd. for C H O [M+H] : 287.09, found : 287.10;
16
15
5
+
HRMS (ESI) calcd. for C H O [M+H] : 287.0914, found : 287.0908; m.p. = 198.1 °C.
1
6
15
5
(E)-1-(2,4-Dihydroxyphenyl)-3-(3-hydroxy-4-methoxyphenyl)prop-2-en-1-one (7b)
1
The title compound via method A was isolated as a yellow solid (48%); H NMR (400 MHz,
MeOD) δ 7.96 (1H, d, J = 8.0 Hz), 7.74 (1H, d, J = 15.6 Hz), 7.59 (1H, d, J = 15.2 Hz), 7.25
(1H, d, J = 2.4 Hz), 7.20 (1H, dd, J = 8.4, 2.0 Hz), 6.98 (1H, d, J = 8.4 Hz), 6.43 (1H, dd, J =
1
3
8
1
1
.8, 2.4 Hz), 6.31 (1H, d, J = 2.4 Hz) 3.92 (3H, s); C NMR (100 MHz, MeOD) δ 193.4,
67.5, 166.4, 151.7, 148.0, 145.5, 133.4, 129.5, 123.5, 119.2, 115.2, 114.7, 112.5, 109.2,
+
03.8, 56.39; LRMS (ESI) calcd. for C H O [M+H] : 287.09, found : 287.10; HRMS (ESI)
1
6
15
5
+
calcd. for C H O [M+H] : 287.0914, found : 287.0908; m.p. = 206.4 °C.
1
6
15
5
(E)-1-(2,4-Dihydroxyphenyl)-3-(3-fluoro-4-hydroxyphenyl)prop-2-en-1-one (7c)
1
The title compound via method A was isolated as a yellow solid (53%); H NMR (400 MHz,
MeOD δ 7.99 (1H, d, J = 8.8 Hz), 7.75 (1H, d, J = 15.2 Hz), 7.65 (1H, d, J = 15.6 Hz), 7.57
(1H, dd, J = 12.4, 2.0 Hz), 7.39 (1H, dd, J = 8.4, 1.6 Hz), 6.97 (1H, t, J = 8.8 Hz), 6.43 (1H,
1
3
dd, J = 8.8, 2.4 Hz), 6.30 (1H, d, J = 2.4 Hz); C NMR (100 MHz, MeOD) δ 193.2, 167.5,
1
66.5, 154.2, 151.9, 149.1, 144.1, 133.5, 127.4, 119.9, 118.9, 116.5, 114.6, 109.2, 103.8;
2
1

+
LRMS (ESI) calcd. for C H O F [M+H] : 275.07, found : 275.10; HRMS (ESI) calcd. for
15
12
4
+
C H O F [M+H] : 275.0714, found : 275.0709; m.p. = 210.5 °C.
15
12
4
(E)-1-(2,4-Dihydroxyphenyl)-3-(4-fluoro-3-hydroxyphenyl)prop-2-en-1-one (7d)
1
The title compound via method A was isolated as a yellow solid (52%); H NMR (400 MHz,
MeOD) δ 7.98 (1H, d, J = 8.8 Hz), 7.74 (1H, d, J = 15.6 Hz), 7.67 (1H, d, J = 15.6 Hz), 7.32
(1H, dd, J = 8.4, 2.0 Hz), 7.25 - 7.21 (1H, m), 7.15 - 7.10 (1H, m), 6.44 (1H, dd, J = 8.8, 2.4
1
3
Hz), 6.31 (1H, d, J = 2.4 Hz); C NMR (100 MHz, MeOD) δ 193.2, 167.6, 166.7, 155.7,
1
46.7, 144.3, 133.5, 133.2, 122.0, 121.4, 118.5, 117.4, 114.6, 109.3, 103.8; LRMS (ESI)
+
calcd. for C H O F [M+H] : 275.07, found : 275.10; HRMS (ESI) calcd. for C H O F
1
5
12
4
15 12
4
+
[
M+H] : 275.0714, found : 275.0709; m.p. = 252.4 °C.
(E)-1-(2,4-Dihydroxyphenyl)-3-(4-hydroxyphenyl)prop-2-en-1-one (7e)
1
The title compound via method A was isolated as a yellow solid (43%); H NMR (400 MHz,
MeOD) δ 7.98 (1H, d, J = 9.2 Hz), 7.80 (1H, d, J = 15.2 Hz), 7.64 - 7.60 (3H, m), 6.86 (2H,
1
3
d, J = 8.0 Hz), 6.43 (1H, dd, J = 8.8, 2.4 Hz), 6.31 (1H, d, J = 2.4 Hz); C NMR (100 MHz,
MeOD) δ 193.5, 167.5, 166.3, 161.5, 145.6, 133.3, 131.8, 127.8, 118.5, 116.9, 114.7, 109.1,
+
1
03.8; LRMS (ESI) calcd. for C H O [M+H] : 257.08, found : 257.10; HRMS (ESI) calcd.
1
5
13
4
+
for C H O [M+H] : 257.0801, found : 257.0801; m.p. = 205.7 °C.
1
5
13
4
(E)-1-(2,4-Dihydroxyphenyl)-3-(3-hydroxyphenyl)prop-2-en-1-one (7f)
1
The title compound via method A was isolated as a yellow solid (52%); H NMR (400 MHz,
MeOD) δ 8.00 (2H, d, J = 8.8 Hz), 7.66 (1H, d, J = 15.6 Hz), 7.51 (1H, d, J = 15.2 Hz), 7.19
2
2

(1H, d, J = 2.4 Hz), 7.11 (1H, dd, J = 8.0, 2.0 Hz), 6.91 (2H, d, J = 8.8 Hz), 6.83 (1H, d, J =
1
3
8
1
2
.0 Hz); C NMR (100 MHz, MeOD) δ 191.0, 163.7, 149.8, 146.2, 132.2, 131.2, 128.4,
+
23.4, 119.5, 116.6, 116.4, 115.6; LRMS (ESI) calcd. for C H O [M+H] : 257.08, found :
1
5
13
4
+
57.10; HRMS (ESI) calcd. for C H O [M+H] : 257.0808, found : 257.0803; m.p. = 208.0
1
5
13
4
°
C.
(E)-3-(3,4-Dihydroxyphenyl)-1-(2-fluoro-4-hydroxyphenyl)prop-2-en-1-one (7g)
1
The title compound via method A was isolated as a yellow solid (53%); H NMR (400 MHz,
MeOD) δ 7.75 (1H, t, J = 8.8 Hz), 7.62 (1H, dd, J = 15.6, 2.0 Hz), 7.28 (1H, dd, J = 15.2, 2.4
Hz), 7.14 (1H, d, J = 2.0 Hz), 7.19 (1H, d, J = 2.4 Hz), 7.04 (1H, dd, J = 8.4, 2.0 Hz), 6.82
1
3
(
1H, d, J = 8.0 Hz), 6.72 (1H, dd, J = 8.8, 2.4 Hz), 6.61 (1H, dd, J = 13.2, 2.4 Hz); C NMR
(100 MHz, MeOD) δ 189.5, 165.0, 150.0, 146.8, 133.6, 128.2, 123.6, 123.2, 119.6, 116.6,
+
1
2
2
15.4, 113.1, 104.2, 103.9; LRMS (ESI) calcd. for C H O F [M+H] : 275.08, found :
1
5
12
4
+
75.10; HRMS (ESI) calcd. for C H O F [M+H] : 275.0714, found : 275.0708; m.p. =
1
5
12
4
07.8 °C.
(E)-3-(3,4-Dihydroxyphenyl)-1-(4-hydroxy-2-methylphenyl)prop-2-en-1-one (7h)
1
The title compound via method B was isolated as a yellow solid (66%); H NMR (400 MHz,
MeOD) δ 7.54 (1H, d, J = 9.2 Hz), 7.41 (1H, d, J = 15.6 Hz), 7.12 (1H, d, J = 2.4 Hz), 7.09
(1H, d, J = 16.0 Hz), 7.02 (1H, dd, J = 8.0, 2.0 Hz), 6.81 (1H, d, J = 8.4 Hz), 6.73 - 6.70 (2H,
13
m), 2.42 (3H, s); C NMR (100 MHz, MeOD) δ 197.0, 161.3, 149.9, 147.3, 146.8, 141.7,
1
32.5, 131.6, 128.1, 124.0, 123.4, 119.2, 116.5, 115.4, 113.3, 21.26; LRMS (ESI) calcd. for
2
3

+
+
C H O [M+H] : 271.10, found : 271.10; HRMS (ESI) calcd. for C H O [M+H] :
16
15
4
16 15
4
2
71.0965, found : 271.0958; m.p. = 208.2 °C.
(E)-3-(3,4-Dihydroxyphenyl)-1-(4-hydroxy-2-methoxyphenyl)prop-2-en-1-one (7i)
1
The title compound via method B was isolated as a yellow solid (46%); H NMR (400 MHz,
MeOD) δ 7.59 (1H, d, J = 8.8 Hz), 7.51 (1H, d, J = 15.6 Hz), 7.38 (1H, d, J = 15.2 Hz), 7.12
(1H, d, J = 2.0 Hz), 7.00 (1H, dd, J = 8.4, 2.0 Hz), 6.80 (1H, d, J = 8.0 Hz), 6.53 (1H, d, J =
1
3
2
1
1
.0 Hz), 6.47 (1H, dd, J = 8.4, 2.0 Hz), 3.91 (3H, s); C NMR (100 MHz, MeOD) δ 193.0,
64.5, 162.5, 149.5, 146.8, 144.5, 133.7, 128.6, 125.1, 123.3, 121.7, 116.5, 115.2, 108.9,
+
00.1, 56.15; LRMS (ESI) calcd. for C H O [M+H] : 287.09, found : 287.10; HRMS (ESI)
1
6
15
5
+
calcd. for C H O [M+H] : 287.0914, found : 287.0906; m.p. = 200.4 °C.
1
6
15
5
(E)-3-(3,4-Dihydroxyphenyl)-1-(4-hydroxy-2-isopropoxyphenyl)prop-2-en-1-one (7j)
1
The title compound via method A was isolated as a yellow solid (53%); H NMR (400 MHz,
MeOD) δ 7.58 (1H, d, J = 8.8 Hz), 7.47 (2H, d, J = 2.8 Hz), 7.11 (1H, d, J = 2.0 Hz), 7.00
(1H, dd, J = 8.4, 2.4 Hz), 6.81 (1H, d, J = 8.0 Hz), 6.51 (1H, d, J = 2.4 Hz), 6.45 (1H, dd, J =
1
3
8
.8, 2.4 Hz), 4.69 (1H, hept, 6.0 Hz), 1.40 (6H, d, J = 6.0 Hz); C NMR (100 MHz, MeOD)
δ 193.4, 160.6, 146.8, 143.9, 133.7, 128.7, 125.5, 123.1, 122.9, 116.5, 115.2, 109.1, 102.2,
+
7
2.21, 22.43; LRMS (ESI) calcd. for C H O [M+H] : 315.12, found : 315.20; HRMS (ESI)
1
8
19
5
+
calcd. for C H O [M+H] : 315.1227, found : 315.1219; m.p. = 80.4 °C.
1
8
19
5
(E)-3-(3,4-Dihydroxyphenyl)-1-(2,4-dimethoxyphenyl)prop-2-en-1-one (7k)
2
4

1
The title compound via method A was isolated as a yellow solid (63%); H NMR (400 MHz,
MeOD) δ 7.65 (1H, d, J = 8.4 Hz), 7.51 (1H, d, J = 15.6 Hz), 7.34 (1H, d, J = 15.6 Hz), 7.12
(1H, d, J = 2.0 Hz), 7.00 (1H, dd, J = 8.4, 2.0 Hz), 6.81 (1H, d, J = 8.4 Hz), 6.65 - 6.61 (2H,
1
3
m), 3.93 (3H, s), 3.89 (3H, s); C NMR (100 MHz, MeOD) δ 193.3, 166.0, 162.0, 149.7,
1
46.8, 145.0, 133.4, 128.5, 124.9, 123.4, 116.5, 115.2, 106.7, 99.53, 56.20; LRMS (ESI)
+
calcd. for C H O [M+H] : 301.10, found : 301.10; HRMS (ESI) calcd. for C H O
1
7
17
5
17 17
5
+
[
M+H] : 301.1071, found : 301.1065; m.p. = 162.9 °C.
(E)-3-(3,4-Dihydroxyphenyl)-1-(p-tolyl)prop-2-en-1-one (7l)
1
The title compound via method A was isolated as a yellow solid (66%); H NMR (400 MHz,
MeOD) δ 7.96 (2H, d, J = 8.0 Hz), 7.68 (1H, d, J = 15.6 Hz), 7.50 (1H, d, J = 15.2 Hz), 7.36
(2H, d, J = 8.0 Hz), 7.20 (1H, d, J = 2.0 Hz), 7.11 (1H, dd, J = 8.4, 2.4 Hz), 6.83 (1H, d, J =
1
3
8
.4 Hz), 2.44 (3H, s); C NMR (100 MHz, MeOD) δ 192.2, 150.0, 147.0, 146.8, 145.0,
1
37.1, 130.4, 129.6, 128.3, 123.6, 119.6, 116.6, 115.7, 21.64; LRMS (ESI) calcd. for
+
+
C H O [M+H] : 255.10, found : 255.10; HRMS (ESI) calcd. for C H O [M+H] :
16
15
3
16 15
3
2
55.1016, found : 255.1014; m.p. = 190.9 °C.
(E)-1-(4-Chlorophenyl)-3-(3,4-dihydroxyphenyl)prop-2-en-1-one (7m)
1
The title compound via method A was isolated as a yellow solid (63%); H NMR (400 MHz,
MeOD) δ 8.06 (2H, d, J = 8.8 Hz), 7.71 (1H, d, J = 15.6 Hz), 7.56 (2H, d, J = 8.8 Hz), 7.49
(1H, d, J = 15.6 Hz), 7.21 (1H, d, J = 2.0 Hz), 7.14 (1H, dd, J = 8.4, 2.4 Hz), 6.84 (1H, d, J =
1
3
8
.4 Hz); C NMR (100 MHz, MeOD) δ 191.1, 147.8, 146.9, 140.0, 138.2, 131.1, 129.9,
+
1
28.1, 123.8, 119.1, 116.6, 115.8; LRMS (ESI) calcd. for C H O Cl [M+H] : 275.05, found
1
5
12
3
2
5

+
:
275.10; HRMS (ESI) calcd. for C H O Cl [M+H] : 275.0469, found : 275.0465; m.p. =
15 12 3
2
05.4 °C.
Syntheses
of
(E)-4-(3-(2,4-diacetoxyphenyl)-3-oxoprop-1-en-1-yl)-1,2-phenylene
diacetate (8a)
A solution of (E)-1-(2,4-dihydroxyphenyl)-3-(3,4-dihydroxyphenyl)prop-2-en-1-one 1 (500
mg, 1.8 mmol) in Ac O (8 mL) was heated to 120-130 °C and stirred for 5 h at this
2
temperature. After cooling to r.t., the resulting mixture was concentrated in vacuo. The
residue was purified by column chromatography (silica gel, n-hexane-ethylacetate, 1:2)
afforded compound. (E)-4-(3-(2,4-Diacetoxyphenyl)-3-oxoprop-1-en-1-yl)-1,2-phenylene
1
diacetate (8a) as a white solid (397 mg, 90%); H NMR (400 MHz, MeOD) δ 7.86 (2H, d, J =
8
.4 Hz), 7.63 - 7.61 (2H, m), 7.58 (1H, d, J = 16.0 Hz), 7.35 (1H, d, J = 12.4 Hz), 7.31 (1H,
d, J = 5.2 Hz), 7.22 (1H, dd, J = 8.4, 2.4 Hz), 7.10 (1H, d, J = 2.4 Hz), 2.33 (3H, s), 2.31 (6H,
1
3
d, J = 3.2 Hz), 2.23 (3H, s); C NMR (100 MHz, MeOD) δ 188.8, 168.6, 168.0, 153.4,
1
2
49.5, 143.7, 142.4, 142.3, 133.1, 131.2, 128.9, 127.5, 125.4, 124.1, 123.3, 119.6, 117.6,
+
0.76, 20.32; LRMS (ESI) calcd. for C H O [M+H] : 441.12, found : 441.10; HRMS (ESI)
2
3
21
9
+
calcd. for C H O [M+H] : 441.1180, found : 441.1171; m.p. = 131.3 °C.
2
3
21
9
Syntheses of (E)-4-(3-(2,4-bis(benzoyloxy)phenyl)-3-oxoprop-1-en-1-yl)-1,2-phenylene
dibenzoate (8b)
(E)-1-(2,4-Dihydroxyphenyl)-3-(3,4-dihydroxyphenyl)prop-2-en-1-one 1 (500 mg, 1.8 mmol)
was dissolved in anhydrous pyridine (5 mL), treated with benzoylchloride (2.1 ml, 18.4
mmol), refluxed for 4 h, left overnight at 22-24 °C, and diluted on the following day with
2
6

cold H O. The residue was extracted three times with ethyl acetate 30 ml. The ethyl acetate
2
layer was washed three times with water (30 ml x 3) and dried over anhydrous MgSO . After
4
removal of the solvents, the product was purified by silica gel column chromatography (silica
gel, n-hexane-ethylacetate, 1:2) as the eluent to give the indicated product. (E)-4-(3-(2,4-
bis(benzoyloxy)phenyl)-3-oxoprop-1-en-1-yl)-1,2-phenylene dibenzoate (8b) as a white solid
1
(
413 mg, 60%); H NMR (400 MHz, MeOD) δ 8.24 - 8.21 (2H, m), 8.16 - 8.13 (2H, m), 8.05
1
3
-
7.99 (6H, m), 7.65 - 7.57 (7H, m), 7.51 - 7.37 (11H, m); C NMR (100 MHz, MeOD) δ
1
1
88.6, 164.2, 163.4, 153.7, 149.7, 143.7, 142.4, 134.2, 133.4, 132.7, 131.1, 129.7, 129.6,
29.4, 129.2, 128.9, 128.4, 128.1, 127.8, 125.7, 124.1, 123.2, 120.1, 118.0; LRMS (ESI)
+
calcd. for C H O [M+H] : 689.18, found : 689.20; HRMS (ESI) calcd. for C H O
4
3
29
9
43 29
9
+
[
M+H] : 689.1812, found : 689.1794; m.p. = 36.7 °C.
Syntheses of 1-(2-hydroxy-4-(methoxymethoxy)phenyl)ethanone (10)
-(2,4-Dihydroxyphenyl)ethanone 5a (1.5 g, 10 mmol) in 12 mL acetone was added K CO3
1
2
(1.4 g, 10 mmol) followed by slow addition of chloro(methoxy)methane (885 mg, 11 mmol)
with stirring under Ar atmosphere at rt. The reaction mixture was stirred at r.t for 4 h and
filtered. Solid was washed with EtOAc (80 mL) and the combined filtrate was washed with
NaH PO (sat. 50 mL) and water (80 mL) and dried over MgSO and concentrated. The crude
2
4
4
product was purified with silica chromatography (n-hexane-ethylacetate, 1:4) to afford 1-(2-
1
hydroxy-4-(methoxymethoxy)phenyl)-ethanone 10 as a off-white solid (1.2 g, 61%); H
NMR (400 MHz, DMSO-d ) δ 12.49 (1H, s), 7.85 (1H, d, J = 8.8 Hz) 6.60 (1H, dd, J = 8.8,
6
1
3
2
.4 Hz), 5.53 (1H, d, J = 2.4 Hz), 5.27 (2H, s), 3.39 (3H, s), 2.57 (3H, s); C NMR (100
MHz, DMSO-d ) δ 203.1, 163.5, 133.2, 114.5, 108.0, 102.9, 93.62, 55.90, 26.71; HRMS
6
2
7

+
(
ESI) calcd. for C H O [M+H] : 197.08, found : 197.10; HRMS (ESI) calcd. for
10 12 4
+
C H O [M+H] : 197.0825, found : 197.0815; m.p. = 60.2 °C.
10
12
4
Syntheses of 3,4-bis(methoxymethoxy)benzaldehyde (11)
Potassium carbonate (4.0 g, 19 mmol) and methoxymethyl chloride (2.3 g, 2.2 mL, 19 mmol)
were added to a solution of 3,4-dihydroxybenzaldehyde 6a (1 g, 7.2 mmol) in acetone (40
mL). The mixture was refluxed for 2.5 h. The precipitate was filtered and dispatched. The
solvent was removed under reduced pressure. The residue was dissolved in CH Cl (30 mL),
2
2
washed with water (80 mL) and dried over MgSO and concentrated. The crude product was
4
purified with silica chromatography (n-hexane-ethylacetate, 1:4) to afford 3,4-
1
bis(methoxymethoxy)benzaldehyde (11) as white solid (1.1 g, 70 %); H NMR (400 MHz,
DMSO-d ) δ 9.85 (1H, s), 7.60 - 7.58 (2H, m), 7.31 (1H, d, J = 8.0 Hz), 5.30 (4H, d, J = 23.6
6
13
Hz), 3.42 (6H, d, J = 2.0 Hz); C NMR (100 MHz, DMSO-d ) δ 191.3, 152.1, 146.9, 130.5,
6
+
1
2
2
26.1, 115.6, 115.3, 94.67, 94.34, 55.97, 55.79; HRMS (ESI) calcd. for C H O [M+H] :
11
14
5
+
27.09, found : 227.10; HRMS (ESI) calcd. for C H O [M+H] : 227.0935, found :
11
14
5
27.0935; m.p. = 44.6 °C.
Syntheses of (E)-3-(3,4-bis(methoxymethoxy)phenyl)-1-(2-hydroxy-4-(methoxymethoxy)
-phenyl)-prop-2-en-1-one (12)
1
-(2-Hydroxy-4-(methoxymethoxy)phenyl)-ethanone 10 (1.0 g, 5.09 mmol) and 3,4-
bis(methoxymethoxy)benzaldehyde 11 (1.2 g, 5.09 mmol) were dissolved in ethanol (10 mL).
0% KOH (aq) (2 mL) was added, and the mixture was at r.t for 6 h. Solvents were removed
6
2
8