Natural Isoflavones and Semisynthetic Derivatives as Pancreatic Lipase Inhibitors

Article abstract of DOI:10.1021/acs.jnatprod.0c01387

Obesity, now widespread all over the world, is frequently associated with some chronic diseases. Thus, there is a growing interest in the prevention and treatment of obesity. To date, the only antiobesity drug is orlistat, a natural product-derived pancreatic lipase (PL) inhibitor with some undesired side effects. In the last decades, many natural compounds or derivatives have been evaluated as potential PL inhibitors, and natural polyphenols are among the most promising for possible exploitation as antiobesity agents. However, few studies have been devoted to isoflavones. In this work, we report a study on the PL inhibitory properties of a small library of semisynthetic isoflavone derivatives together with the natural leads daidzein (1), genistein (2), and formononetin (3). In vitro lipase inhibition assay showed that 2 is the most promising PL inhibitor. Among synthetic isoflavones, the hydroxylated and brominated derivatives were more potent than their natural leads. Detailed studies through fluorescence measurements and kinetics of lipase inhibition showed that 2 and the bromoderivatives 10 and 11 have the greatest affinity for PL. Docking studies corroborated these findings highlighting the interactions between isoflavones and the enzyme, confirming that hydroxylation and bromination are useful modifications.

Full text of DOI:10.1021/acs.jnatprod.0c01387

pubs.acs.org/jnp
Article
Natural Isoflavones and Semisynthetic Derivatives as Pancreatic
Lipase Inhibitors
Nunzio Cardullo,* Vera Muccilli, Luana Pulvirenti, and Corrado Tringali*
Cite This: J. Nat. Prod. 2021, 84, 654−665
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Obesity, now widespread all over the world, is
frequently associated with some chronic diseases. Thus, there is a
growing interest in the prevention and treatment of obesity. To
date, the only antiobesity drug is orlistat, a natural product-derived
pancreatic lipase (PL) inhibitor with some undesired side effects.
In the last decades, many natural compounds or derivatives have
been evaluated as potential PL inhibitors, and natural polyphenols
are among the most promising for possible exploitation as
antiobesity agents. However, few studies have been devoted to
isoflavones. In this work, we report a study on the PL inhibitory
properties of a small library of semisynthetic isoflavone derivatives
together with the natural leads daidzein (1), genistein (2), and
formononetin (3). In vitro lipase inhibition assay showed that 2 is the most promising PL inhibitor. Among synthetic isoflavones, the
hydroxylated and brominated derivatives were more potent than their natural leads. Detailed studies through fluorescence
measurements and kinetics of lipase inhibition showed that 2 and the bromoderivatives 10 and 11 have the greatest affinity for PL.
Docking studies corroborated these findings highlighting the interactions between isoflavones and the enzyme, confirming that
hydroxylation and bromination are useful modifications.
ccording to the World Health Organization (WHO)
Adefinition, overweight and obesity are defined as
abnormal or excessive fat accumulation, and they present a
risk to health. Although, in the past, obesity was frequently
observed only in high income countries, now it is dramatically
spreading also in low- and middle-income countries,
particularly in urban settings. In 2016, 39% of both women
and men aged 18 and over were overweight.1 Leaving aside
the aesthetic problems, which can cause psychological or
social problems, obesity is a significant risk factor for a series
of pathological conditions including cardiovascular diseases,
hypertension, hyperlipidemia, diabetes, mental disorders, and
some forms of cancer.2,3 Thus, prevention and treatment of
obesity have attracted the attention of many individual
researchers and health organizations. Of course, physical
exercise and a carefully balanced diet are the first prerequisites
for fighting obesity. However, they often do not solve the
problem, and therefore, it is necessary to employ specific
therapies. Because liposuction is a surgical treatment (not
without risk) giving satisfactory results only in some cases,2
the main recently adopted therapies have been based
essentially on three drugs, namely, orlistat (Xenical), a natural
product-derived inhibitor of pancreatic lipase (PL, triacylgly-
cerol acyl hydrolase), sibutramine (Reductil), and rimonabant
(Acomplia), the latter two are both appetite suppressants,
acting with different mechanisms.4 Nevertheless, sibutramine
and rimonabant have been withdrawn from the market due to
major adverse cardiovascular events and significant psychiatric
undesired effects, respectively.5,6 Thus, orlistat remains the
only drug on the market to treat obesity, although it shows
unpleasant gastrointestinal side effects.6,7 This situation
prompted many research groups to search for new, effective,
and safe drugs to treat obesity, and several potential
antiobesity drugs are currently in clinical trials. In this
scenario, PL inhibitors play a major role: PL normally
hydrolyzes 50−70% of total dietary fats;4 the interaction of
inhibitors with the enzyme causes the blocking of its catalytic
activity, which is the hydrolysis of triglycerides, thus reducing
the intestinal fat absorption, and the consequent accumulation
of adipose tissue. PL inhibitors are normally excreted with the
enzyme, causing no long-term effects, and are generally
regarded as relatively safe.2
In the search for new PL inhibitors, a special focus has been
devoted to natural compounds or derivatives, in consideration
of the number and the variety of plants and microorganisms
producing metabolites with potential antiobesity activity.4,5,8
Special Issue: Special Issue in Honor of A. Douglas
Kinghorn
Received: December 28, 2020
Published: March 1, 2021
© 2021 American Chemical Society and
American Society of Pharmacognosy
https://dx.doi.org/10.1021/acs.jnatprod.0c01387
J. Nat. Prod. 2021, 84, 654−665
654

Journal of Natural Products
pubs.acs.org/jnp
Article
Moreover, many natural products show low absorption rates,
and this generally reduces systemic adverse effects.9 These
natural inhibitors may also be considered as lead compounds
to obtain, through chemical modification, optimized products
with enhanced properties, but they maintain low or absent
toxicity. Orlistat itself is a semisynthetic compound derived
from the natural product lipstatin, an unusual β-lactone
originally isolated from the actinomycete Streptomyces
toxytricini.10−12
Previously reported natural inhibitors belong to various
chemical classes, including carbohydrates, saponins, terpenes,
and polyphenols; also, polyphenol-rich extracts are reported
for PL inhibition,4 and some phenolics have showed
multifunctional antiobesity properties.8 Thus, polyphenols
are among the most promising natural products for possible
exploitation as antiobesity agents. A significant number of
polyphenolic compounds with PL inhibitory activity have
been reported, including flavonoids, phenolic acids, lignans,
the stilbenoid resveratrol, and proanthocianydins.3,7,13 Never-
theless, less attention has been devoted to some polyphenol
subfamilies, such as isoflavones. These natural products are
well-known for other beneficial properties, mainly as
phytoestrogenic,14 anti-inflammatory,15 cancer chemopreven-
tive,16 antioxidative, antidiabetic,17 and neuroprotective
agents.18 The PL inhibitory activity of the isoflavones daidzein
(1) and genistein (2) has been reported,19 but no data are
available for their mechanism of action nor for isoflavone
derivatives.
Furthermore, it is noteworthy here that, in recent years, the
beneficial effect of soy products in preventing metabolic
disorders, including hyperlipidemia, have been reported.
These properties have been attributed to the high levels of
isoflavones in soy (as well as of proteins, fibers, and
unsaturated fats).20 In some clinical trials, the effect of soy
or soy isoflavones consumption on weight and other obesity-
related pathologies has been examined, concluding that soy is
a suitable food for antiobesity effects.21 More in detail, the
antiobesity action of isoflavones is imputable, on one hand, to
inhibition of lipogenesis and to the increase of fatty acid β-
oxidation22 and, on the other hand, to regulation of food
intake and energy expenditure, as well as in prevention of fat
accumulation in adipose tissue.21
On this basis, as a continuation of our study on the search
of bioinspired polyphenols capable of inhibiting enzymes
involved in metabolic diseases,23−25 we report herein the PL
inhibitory properties of a small library of natural isoflavones
and semisynthetic derivatives.
RESULTS AND DISCUSSION
■
As the first step of this study, we planned to obtain a small
library of natural and semisynthetic isoflavones from daidzein
(1), genistein (2), and formononetin (3) and to evaluate
these compounds as PL inhibitors. Unfortunately, attempts to
obtain genistein derivatives were unsuccessful (experimental
details are listed in the Supporting Information). Further steps
were devoted to in vitro lipase inhibition assays, fluorescence
spectra measurements, kinetics of lipase inhibition, and
docking studies, as detailed in the following.
Synthesis. In the present work, a metal-free synthetic
methodology has been adopted to afford isoflavone analogues
with an ortho-dihydroxylated aromatic ring (catechol group).
This approach consists of the preparation of deoxybenzoin
from respective substituted phenol and phenyl acetic acid
precursors by Friedel−Crafts acylation and the use of
Vilsmeier reagent [N,N′-dimethyl(chloromethylene)-ammo-
nium chloride] formed from methyl sulfonyl chloride and
DMF. This route is very effective for converting deoxy-
benzoins into corresponding isoflavones with mild reaction
conditions, short reaction times, and excellent yields.26,27
More specifically, in a two-step procedure, a deoxybenzoin
intermediate was obtained by reaction of resorcinol (4) or
orcinol (5) with 3,4-dihydroxyphenylacetic acid (6), in the
presence of BF3·Et2O. The intermediates were subsequently
treated with the Vilsmeier reagent in the presence of BF3·Et2O
(Scheme 1a), thus affording, respectively, 7 and 8. In other
experiments, we also carried out a one-pot synthesis, directly
treating the intermediates with Vilsmeier reagent to afford 7
and 8 (Scheme 1b).
Scheme 1. (a) Two-Step and (b) One-Pot Synthesis of Isoflavones 7 and 8
655
https://dx.doi.org/10.1021/acs.jnatprod.0c01387
J. Nat. Prod. 2021, 84, 654−665

Journal of Natural Products
pubs.acs.org/jnp
Article
Table 1 reports the reaction conditions for both two-step
(entries 1−3) and one-pot (entries 4 and 5) syntheses with
the respective reaction yields.
δ 7.13 (bs), 7.02 (d, J = 8.5 Hz), and 6.97 (d, J = 8.5 Hz)
have been assigned respectively to H-2′, H-6′, and H-5′ of
ring B. The broad singlets at δ 6.67 and 6.70 are attributable
to the protons of ring A. HSQC correlations allowed the
assignments of CH sp2 carbons, as reported in Table 2. The
carbon signals at δ 116.9 and 100.0 couple with the proton
signals at δ 6.67 and 6.70, respectively; however, their
unambiguous assignment as CH-6 and CH-8 was only
possible after the analysis of the HMBC correlations. Namely,
the signal at δ 6.70 (H-8) showed key correlations with the
carbons at δ 161.0 (C-7) and 158.8 (C-8a). Analogously, the
signal at δ 6.67 (H-6) was long-range correlated with the
bridged carbon at δ 114.7 (C-4a) and the methyl carbon a δ
22.9. This latter correlation was pivotal to establish the C-5
position for the methyl group in isoflavone 8. The phenolic
OH signal at δ 10.62 was positioned at C-7 on the basis of key
correlations with the deshielded nonprotonated carbon at δ
161.0 (C-7) as well as with the carbon signals of C-8 and C-6.
To obtain further isoflavone analogues, we resorted to
simple reactions, namely, ortho-hydroxylation and bromination
of the natural isoflavones daidzein (1) and formononetin (3).
The corresponding catechol derivatives were prepared
employing the hypervalent iodine reagent 2-iodoxybenzoic
acid (IBX). The optimization of this reaction (Table S1,
Supporting Information) suggested the use of different
conditions for each substrate. According to literature findings
on similar compounds,29 the ortho-hydroxylation of daidzein
should have afforded products with a catechol group on ring
A, ring B, or both. Nevertheless, in the reported conditions
(Scheme 2a), only one major compound was obtained,
Table 1. Reaction Conditions for the Synthesis of 7 and 8
a
entry
R
BF3·OEt2
T (°C)
overall yield (%)
Two-Step
b
b
b
1
2
3
H
12 equiv
8 equiv
8 equiv
110
80
42
57
16
c
H
110
c
80
Me
110
c
80
d
One-Pot
4
5
H
7.5 equiv
i. 110
ii. 80
i. 110
ii. 80
94
37
Me
7.5 equiv
a
The yield was determined after precipitation or liquid chromatog-
b
raphy purification. Temperature of the first step of the reaction.
c
d
Temperature of the second step of the reaction. See Scheme 1b for
details.
The one-pot method allows obtaining higher yields and
requires lower equivalents of BF3·Et2O; thus, the one-pot
method was preferred for preparative syntheses. Of note, this
method afforded the natural isoflavone 7 with 94% yield, far
higher than previously reported in the literature.28
The previously unreported isoflavone 8 was subjected to
spectroscopic characterization by HRMS and 1D and 2D
1
NMR spectroscopy. The assignments of H and 13C NMR
Scheme 2. Selective Hydroxylation of (a) 1 and (b) 3
signals are listed in Table 2, and HRMS data are reported in
1
Table 2. H (500 MHz) 13C (125 MHz) and HMBC NMR
Data of 8 in DMSO-d6
a
position
δC, type
δH, multiplet (mult.) (J in Hz)
HMBC
2
3
4
4a
5
6
7
8
8a
1′
2′
3′
4′
5′
6′
5-Me
151.9, CH
124.8, C
176.4, C
114.7, C
142.2, C
116.9, CH
161.0, C
100.0, CH
158.8, C
124.4, C
115.4, CH
147.6, C
147.9, C
112.7, CH
121.0, CH
22.9, CH3
8.19, s
4, 8a, 1′
6.67, bs
10.62, s (OH)
6.70, bs
4a, 5-Me
6, 7, 8
7, 8a
subsequently identified as the isoflavone 7. The ortho-
hydroxylation of formononetin afforded the natural isoflavone
retusin (9)26 with 43% yield (Scheme 2b).
7.13, bs
3, 4′, 6′
Also the preparation of the brominated analogues required
an optimization step, as summarized in Table 3. The
experiments were performed mainly on formononetin (3),
and the results suggested the conditions to be employed for
daidzein (1). Two brominating reagents, namely, NaBr in the
presence of oxone or H2O2 and Br2, were employed in
different solvents and at various temperatures.
8.31, s (OH)
6.97, d (8.5)
7.02, d (8.5)
2.70, s
1′, 3′
3, 2′, 4′
4a, 5, 6
a
HMBC correlations are from proton(s) stated to the indicated
carbon.
The isoflavones 1 and 3 reacted appreciably only with
bromine (entries 3−5). The syntheses were scaled-up
according to the best reaction conditions, as reported in
Scheme 3. Although a diluted Br2 solution was gradually
added to daidzein and the reaction was maintained at 0 °C
(Scheme 3a), 1 gave exclusively 8,3′,5′-tribromodaidzein (10)
with 70% yield. By treating 3 under the same conditions, only
8-bromoformononetin (11) was recovered with 80% yield
the Experimental Section. The presence in the 1H NMR
spectrum of the distinctive singlet at δ 8.19 (H-2) is indicative
of the formation of ring C and therefore of an isoflavone. The
key HMBC correlation of H-2 with the signals at δ 176.5 (C-
4), 158.8 (C-8a), and 124.4 (C-1′) confirmed the isoflavone
skeleton and allowed the reported assignments. The signals at
656
https://dx.doi.org/10.1021/acs.jnatprod.0c01387
J. Nat. Prod. 2021, 84, 654−665

Journal of Natural Products
pubs.acs.org/jnp
Article
Table 3. Reaction Conditions for the Synthesis of 10−12
a
entry
isoflavone
solvent
brominating agent (1.2 equiv)
T (°C)
time
6 h
6 h
40 min
1 h
3 h
yield (%)
1
2
3
4
5
3
3
3
3
1
CH3COOH
CH3COOH
CH3COOH
CHCl3, MeOH
CHCl3, MeOH
NaBr, H2O2
rt
rt
rt
0
3
8
92
80
67
NaBr, oxone
Br2 in CH3CN
Br2 in CHCl3
Br2 in CHCl3
b
b
0
a
After the completion of reaction, a saturated solution of Na2S2O3 was added to reduce the excess of halogenating agent. The yield of the product
b
was determined after isolation by liquid−liquid partition, or when necessary, after purification by liquid chromatography. Employed in a small
amount to solubilize the isoflavone, see Experimental Section.
The new brominated isoflavones 11 and 12 were subjected
Scheme 3. Synthesis of Brominated Isoflavones 10−12
to spectroscopic characterization by HRMS spectrometry and
1
1D and 2D NMR spectroscopy. The assignments of H and
13C NMR signals are listed in Table 4, and HRMS data are
reported in the Experimental Section. The brominated
isoflavone 10 has been obtained very recently from daidzein
as a biotransformation product by Actinomadura sp. RB99; it
has been named maduraktermol H.30
To establish if isoflavones without free phenolic groups
could inhibit PL, the isoflavone 7 was subjected to
methylation and acetylation to obtain, respectively, its
permethylated derivative, the natural product cabreuvine
(13),31 and its peracetylated derivative 14 (Scheme 4).
Although 14 has been previously reported,32 its NMR data
have not been tabulated; therefore, they are listed in Table 4.
In Vitro Lipase Inhibitory Activity. The natural
isoflavones 1−3, together with the synthesized products 7−
14, were studied as potential pancreatic lipase (PL) inhibitors.
A spectrophotometric methodology employing 4-nitrophenyl
butyrate as a substrate for the PL was applied.33 The results of
the in vitro assay are reported in Table 5 as IC50 values; the
commercial antiobesity drug, orlistat, was employed as a
positive control.
(Scheme 3b). When 3 was dissolved in acetic acid and a
bromine solution was slowly added at room temperature, 8,3′-
dibromoformononetin (12) was obtained with 92% yield
(Scheme 3c).
1
Table 4. H (500 MHz) and 13C (125 MHz) NMR Data of 11, 12, and 14
a
b
b
11
12
14
position
δC, type
δH, mult. (J in Hz)
δC, type
δH, mult. (J in Hz)
δC, type
δH, mult. (J in Hz)
2
3
4
4a
5
6
7
8
8a
1′
2′
3′
4′
5′
6′
153.9, CH
123.7, C
174.8, C
117.9, C
126.3, CH
115.0, CH
160.2, C
97.1, C
154.5, C
124.2, C
130.6, CH
114.1, CH
159.5, C
114.1, CH
130.6, CH
55.6, CH3
8.50, s
152.8, CH
123.6, C
175.9, C
118.0, C
126.2, CH
114.3, CH
159.7, C
97.0, C
155.3, C
125.1, C
133.6, CH
111.6, C
156.0, C
111.9, CH
129.6, CH
56.2, CH3
7.98, s
153.4, CH
123.9, C
175.2, C
7.98, s
122.2, C
7.97, d (8.8)
7.12, d (8.8)
8.01, d (8.8)
6.96, d (8,8)
8.30, bs (OH)
127.8, CH
119.6, CH
154.0, C
110.9 CH
156.5, C
130.3, C
124.0, CH
141.0, C
8.26, d (8.8)
7.13, d (8.8)
7.26, bs
7.41, bs
7.52, d (8.6)
6.99, d (8.6)
7.68, d (2.2)
142.0, C
123.4, CH
126.8, CH
7.52, d (8.6)
6.99, d (8.6)
3.78, s
6.92, d (8.5)
7.42, dd (2.2, 8.5)
3.86, s
7.21, d (8.5)
7.39, d (8.5)
4′-OMe
3′/4′−OCO
7−OCO
3′/4′−OCOMe
7−OCOMe
168.1, C
168.4, C
20.6, CH3
21.1, CH3
2.25, s
2.30, s
a
b
NMR spectra were run in DMSO-d6. NMR spectra were run in chloroform-d.
657
https://dx.doi.org/10.1021/acs.jnatprod.0c01387
J. Nat. Prod. 2021, 84, 654−665

Journal of Natural Products
pubs.acs.org/jnp
Article
comparing 3 with 1, and more generally, the lack of free OH
groups is highly detrimental for the activity, as shown by the
data on the permethylated 13 and peracetylated 14, with
respect to the parent compound 7. Also bromination proved
to be an effective modification for improving the PL inhibitory
activity. In particular, the most potent synthetic compound
obtained in this study was 8,3′,5′-tribromodaidzein (10),
significantly more active than its natural lead 1. Similarly, 8-
bromoformonetin (11) and 8,3′-dibromoformonetin (12)
were more active than 3. Within brominated isoflavones, we
could not find a clear structure−activity relationship between
number and position of the bromine atoms and their
inhibitory capacity. The structure−activity relationships
described above are graphically summarized in Figure 1.
These encouraging results prompted us to carry out further
experiments on selected isoflavones, as detailed in the
following sections.
Scheme 4. Synthesis of Cabreuvine (13) and of Isoflavone
14
Table 5. PL Inhibitory Activity (IC50) of Isoflavones 1−3
and 7−14
a
compound
IC50 SD (μM)
Fluorescence Spectra Measurements. According to
literature data, there are 7 Trp, 25 Phe, and 16 Tyr residues in
pancreatic lipase that may be responsible for the fluorescence
emission of this protein.9 Changes in the fluorescence
intensity of PL are attributed to the interaction with ligands
which can modify the polarity of the environment surrounding
the Trp residues, although changes in the environment of Phe
and Tyr residues also give a minor contribution to
fluorescence quenching. Intrinsic fluorescence experiments
were performed to evaluate possible PL−isoflavone inter-
actions. Thus, fluorescence spectra of PL were recorded upon
the addition of different aliquots of the natural isoflavones 1−
3, as well as of the synthetic products 10 and 11, showing
promising inhibitory activity. The measurements were carried
out at three different temperatures, namely, 27, 32, and 37 °C,
and showed that the PL fluorescence intensity decreases with
increasing concentrations of all compounds under study
(Figures S31−S35). A slight bathochromic shift of the
maximum fluorescence intensity (from 350 to 362 nm) was
observed in all the titrations. The decrease in fluorescence
intensity accompanied by a bathochromic shift has been
reported to be related with an increase in the polarity
environment surrounding the Trp residues, probably due to a
conformational transition of the protein caused by the
interaction between these isoflavones and PL.9,34
1
2
3
7
8
9
83
60
120
70
92
93 14
1
5
6
3
2
10
11
12
13
14
orlistat
63
87
4
8
100 10
360 10
122 10
0.6 0.1
a
Results are reported as mean SD (n = 3).
Daidzein (1) and genistein (2) showed promising PL
inhibitory activity when compared with the natural for-
mononetin (3); of note, compounds 7 and 9, corresponding
to the hydroxylated derivatives of daidzein and formononetin,
respectively, proved to be more active inhibitors than the
natural leads. More in detail, the position and the number of
hydroxy groups is determinant for the PL inhibitory activity.
Namely, the insertion of an OH either into ring A (as in 9) or
into ring B (as in 7), in ortho position to phenolic moiety,
increases the activity, and the loss of the OH at C-5 (as in 2
with respect to 1) diminishes the efficacy. In compound 8 the
insertion of a methyl group at C-5 was detrimental for the
activity, as highlighted by comparison with 7. The OH at C-4′
seems to be important for PL inhibition, as can be observed
The data obtained from fluorescence titrations were
analyzed according to the Stern−Volmer equation (eq 2),
and the results were plotted as depicted in Figures 2 and 3 to
gain some information on the mechanism of quenching. The
Figure 1. Structural features affecting the PL inhibitory activity of isoflavones.
658
https://dx.doi.org/10.1021/acs.jnatprod.0c01387
J. Nat. Prod. 2021, 84, 654−665

Journal of Natural Products
pubs.acs.org/jnp
Article
Figure 2. Stern−Volmer plots for the interaction of PL with (A) daidzein (1), (C) genistein (2), and (E) formononetin (3). Double log Stern−
Volmer plots of PL with (B) 1, (D) 2, and (F) 3. Each line was obtained with data of three independent experiments.
linearity observed by plotting F0/F and [Q] (Figures 2 and 3)
allowed determination of the existence of a single quenching
process, static or dynamic. Table 6 summarizes the Ksv and Kq
values for the PL−isoflavone interactions. As shown in Table
6, Ksv values decrease as the temperature increases for
daidzein (1), genistein (2), and the bromoderivative 11
suggesting that fluorescence quenching occurs with a static
mechanism.35 Indeed, the rate constant values (Kq) obtained
are greater than the typical values reported for collisional
diffusion processes (2.0 × 1010), confirming the occurrence of
static quenching.36 The data obtained for formononetin (3)
and tribromodaidzein (10) suggest a different behavior, and in
particular, the increase of KSV and rate constant values with
increasing temperature is indicative of a dynamic quenching
process.
Further parameters useful to study the interactions of
isoflavones with PL, namely the binding constant (Ka) and the
number of binding sites per enzyme molecule (n) were
obtained plotting eq 3 (Figures 2 and 3). These values are
reported in Table 6. The Ka value is inversely correlated with
the temperature for isoflavones 1, 2, and 11, corroborating the
assumption of a static quenching based on KSV variation,
whereas the Ka value of 3 and 10 increases with temperature,
as expected for a dynamic quenching. The highest Ka values at
37 °C were obtained for the bromoderivatives 10 and 11,
suggesting a greater affinity for PL for these isoflavones.
Kinetics of Lipase Inhibition. The inhibition of
pancreatic lipase by selected isoflavones was also evaluated
through an enzyme kinetics study. As detailed in the
Experimental Section, the mode of inhibition by compounds
1−3, 10, and 11 on PL was determined through the
Lineweaver−Burk graphs, by plotting the reciprocal of initial
velocity (v0) versus the reciprocal of substrate concentration
(Figure 4).
These plots showed that all of the tested isoflavones, except
genistein (2), act as mixed-type inhibitors of PL, exhibiting
both competitive and noncompetitive mechanisms. Namely, 1,
3, 10, and 11 compete with the employed substrate for
binding to PL, but they could also bind to the PL−substrate
complex. Differently, 2 inhibits the PL activity in a
competitive manner. The secondary plots of the y-intercept
and slope, against the isoflavones concentration, were linearly
fitted, indicating a single class of inhibition sites or a single
inhibition site between PL and the tested isoflavones. The Ki
659
https://dx.doi.org/10.1021/acs.jnatprod.0c01387
J. Nat. Prod. 2021, 84, 654−665

Journal of Natural Products
pubs.acs.org/jnp
Article
Figure 3. Stern−Volmer plots for the interaction of PL with (A) 10 and (C) 11. Double log Stern−Volmer plots of PL with (B) 10 and (D) 11.
Each line was obtained with data of three independent experiments.
Table 6. KSV, Kq, Ka, and n Values of the Interaction Between PL and Selected Isoflavones
T (°C)
KSV (×104 L/mol)
Kq (×1013 L/mol s)
R2
Ka (×104 L/mol)
n
R2
1
27
32
37
27
32
37
27
32
37
27
32
37
27
32
37
8.95 0.32
4.94 0.41
4.08 0.62
1.88 0.05
1.50 0.11
1.07 0.07
0.72 0.02
1.02 0.03
1.06 0.08
0.58 0.02
0.96 0.04
1.11 0.06
4.69 0.17
3.34 0.08
2.29 0.11
5.62 0.32
3.11 0.41
2.57 0.62
1.18 0.05
0.94 0.11
0.67 0.07
0.45 0.02
0.64 0.03
0.66 0.08
0.36 0.02
0.60 0.04
0.70 0.06
2.95 0.17
2.10 0.08
1.43 0.11
0.9921
0.9968
0.9992
0.9938
0.9907
0.9971
0.9935
0.9976
0.9937
0.9937
0.9903
0.9955
0.9931
0.9967
0.9924
5.30 0.67
1.20 0.12
1.02 0.17
1.24 0.52
0.87 0.21
0.44 0.07
0.16 0.04
0.22 0.04
0.27 0.01
0.32 0.08
1.99 0.14
3.21 0.72
12.4 0.71
8.47 0.56
1.47 0.26
1.66
1.36
1.43
1.20
1.16
1.03
0.62
0.59
0.55
0.87
1.05
1.12
1.14
1.09
0.91
0.9986
0.9981
0.9914
0.9971
0.9977
0.9922
0.9908
0.9913
0.9929
0.9945
0.9900
0.9968
0.9920
0.9957
0.9962
2
3
10
11
values (related with the competitive mechanism) determined
for 1, 3, 10, and 11, were lower than K′i values (related with
the noncompetitive mechanism), thus implying that these
isoflavones could bind much tighter to the PL catalytic site
than to the PL−substrate complex. Moreover, a very good
correlation between Ki and IC50 values was observed (R2 =
0.9963), corroborating the results about a promising PL
inhibitory activity of isoflavones and, particularly, of genistein
(2) and tribromodaidzein 10.
Molecular Docking Study. A molecular docking study
was carried out to determine the affinity of the isoflavones 1−
3 and 7−14 for the pancreatic lipase catalytic site and their
preferred orientation. This could also allow a better
understanding of the inhibition mechanism. The binding site
of PL includes the following residues: Ser152, Phe215,
Arg256, His263, Leu264, Asp176, and Tyr114. The computa-
tional experiments were acquired employing Autodock4 and
Autodock Vina (see the Experimental Section for more
details).37−39
The docking outcomes showed that all of the ligands are
well accommodated into the binding pocket, occupying almost
an equivalent spatial portion of the cavity. The calculated
binding energy (kcal/mol), listed in Table S2, suggests a
promising affinity of this class of inhibitors, also confirmed by
visual inspection of docked conformations; a list of molecular
interactions for each analyzed compound is reported in the
Supporting Information (Table S3).
The observation of docking poses showed that the
isoflavone skeleton accommodates into the binding site and
also highlighted that the compounds under study are able to
establish noncovalent interactions with His263, His151,
Ser152, Asp79, Pro180, Gly76, and Tyr114 of catalytic
660
https://dx.doi.org/10.1021/acs.jnatprod.0c01387
J. Nat. Prod. 2021, 84, 654−665

Journal of Natural Products
pubs.acs.org/jnp
Article
Figure 4. Lineweaver−Burk plots of isoflavones 1−3, 10, and 11 with PL. The secondary inset plots represent slope versus [I] and y-intercept
versus [I].
Figure 5. (a) Daidzein (1), (b) formononetin (3), and (c) isoflavone 7 into the binding site of PL; the residues of PL structure were represented
using stick structures. The labeled dashes lines (yellow) represent hydrogen-bonding interactions.
pocket. Among the natural isoflavones, daidzein (1) and
genistein (2) showed the best fitting in terms of
intermolecular interactions with PL, if compared with
formononetin (3). In particular, the pose of 1 (Figure 5a)
into the cavity shows that the OH at C-7 interacts with both
Arg256 and Asp79; the oxygen of the ketone function is
involved in two hydrogen bonds with His263 and Phe215,
and the OH at C-4′ establishes a hydrogen bond with Tyr114.
In genistein (2), the additional hydroxy group at C-5 gives a
further hydrogen bond with His151. These data are in perfect
661
https://dx.doi.org/10.1021/acs.jnatprod.0c01387
J. Nat. Prod. 2021, 84, 654−665

Journal of Natural Products
pubs.acs.org/jnp
Article
High-resolution mass spectra (HRMS) were acquired with an
Orbitrap Fusion Tribrid (Q-OT-qIT) mass spectrometer (Thermo
Fisher Scientific) equipped with an ESI ion source. Mass data were
acquired on samples directly infused and volatilized using the
optimized parameters previously described.40
The lipase inhibition assay was performed on a 96-well microplate,
and the UV−vis measurements were done on a Synergy H1
microplate reader (BioTek) at 405 nm. Kinetic measurements were
carried out on a Synergy H1 microplate reader (BioTek) under the
same conditions reported for PL inhibition; optical density
measurements were acquired every minute for 30 min. Intrinsic
fluorescence experiments were performed on an Agilent Cary Eclipse
spectrometer. Fluorescence spectra were acquired in the range 315−
500 nm, setting the following parameters: λEXC at 285 nm and
excitation and emission slits of 10 nm. Experiments were carried out
at 27, 32, and 37 °C.
agreement with the higher inhibitory activity of 1 and
especially 2 with respect to 3 (see Tables 5 and 6). The
compounds 7 (Figure 5c) and 9, hydroxylated derivatives of
daidzein (1) and formononetin (3), respectively, showed
better affinity for PL than the parent compounds thanks to the
new OH groups. In fact, the OH at C-3′ in 7 establishes a
further hydrogen bond with Phe215; analogously, the OH at
C-8 in 9 forms two hydrogen bonds with Phe77 and His151.
The observed interactions highlight the importance of the
phenolic groups for PL affinity. Furthermore, it is worth
noting that the catechol group in compound 7 improves its
overall interaction with the protein. On the other hand, the
OMe group at C-4′ of 3 (Figure 5b) and 9 does not establish
any interaction with the amino acid residues of catalytic
pocket.
Unless otherwise stated, all solvents and reagents were purchased
from commercial sources; they were analytically pure and for
spectroscopy use. Daidzein, genistein, and formononetin were
purchased from TCI Europe N. V. p-Nitrophenyl butyrate, porcine
pancreatic lipase (PL; Type II), and orlistat were purchased from
Merck. IBX was prepared before use, as previously reported.21
Synthesis of Isoflavones 7 and 8. Typical Procedure for the
Two-Step Synthesis of the Isoflavones 7 and 8. For step 1,
Resorcinol (4; 102.2 mg; 0.93 mmol) or orcinol (5; 115.1 mg; 0.93
mmol) was treated with 3,4-dihydroxy-phenylacetic acid (6; 153.5
mg; 0.93 mmol) and a 46.5% BF3·Et2O solution (2.1 mL; 7.5 mmol).
The mixture was heated at 110 °C under N2 for 1.5 h. The crude was
cooled to rt and treated with cold H2O (20 mL). The deoxybenzoin
4a was recovered by precipitation, whereas the intermediate 5a was
recovered by partitioning the aqueous phase with Et2O (3 × 10 mL).
The combined organic layer was successively partitioned with
saturated NaHCO3 (2 × 20 mL), washed with H2O (30 mL),
dried over anhydrous Na2SO4, and concentrated in vacuo.
For step 2, the deoxybenzoin (4a or 5a; 0.12 mmol) was dissolved
in dry DMF (0.2 mL), and a solution of MeSO2Cl (0.12 mL in 0.6
mL of dry DMF; Vilsmeier reagent), prepared separately, was added.
The mixtures were stirred at 80 °C under N2 atmosphere for 1.5 h
and then at 40 °C overnight. In another experiment, the
deoxybenzoin 4a (0.12 mmol) was treated as described above, but
an aliquot of 46.5% BF3·Et2O solution (0.15 mL; 0.52 mmol) was
added before the addition of Vilsmeier reagent. The mixtures were
diluted with a 12% AcONa solution (20 mL), and they were
partitioned with EtOAc (3 × 10 mL). The combined organic layer
was washed with water, dried over anhydrous Na2SO4, and
evaporated until dry.
Typical Procedure for the One-Pot Synthesis of the Isoflavones
7 and 8. Resorcinol (4; 500.1 mg; 4.54 mmol) or orcinol (5; 563.4
mg; 4.54 mmol) was mixed with 3,4-dihydroxyphenylacetic acid (6;
763.4 mg; 4.54 mmol), and a 46.5% BF3·Et2O solution (9.7 mL; 34.3
mmol) was added. The mixture was heated at 110 °C under N2
atmosphere for 2 h. Separately, a solution of MeSO2Cl (4.5 mL in 22
mL of dry DMF; Vilsmeier reagent) was prepared and stirred at rt for
30 min. When the mixture containing the deoxybenzoin was cooled
down, the Vilsmeier reagent was added in three times. Then the
mixture was heated at 80 °C for 1.5 h, and finally, it was stirred at rt
overnight. The mixtures were diluted with a 12% AcONa solution
(100 mL), and they were partitioned with EtOAc (3 × 40 mL). The
combined organic layer was washed with H2O (100 mL), dried over
anhydrous Na2SO4, and evaporated until dry.
In general, the brominated isoflavones 10 and 12 (Figure
S36) present almost the same interactions observed for the
natural leads, forming a more stable complex for the
polarization effect caused by bromine groups.
As a confirmation that the presence of OH groups in
isoflavones improves the interaction with PL, the docking
study showed that the permethylated (13) and peracetylated
(14) derivatives of isoflavone 7 have a lower affinity with the
PL catalytic site than the parent compound, due to the lack of
free OH groups. Also these results are in perfect agreement
with the very low inhibitory activity observed for 13 and 14.
It is worth noting that the residues Ser152 and His263 are
part of the catalytic triad, and in particular, Ser152 is believed
to play a pivotal role in the catalytic activity of PL.33 Hence,
on the basis of docking data, the isoflavones under study
could have a role as competitive inhibitors. These findings are
not in contrast with the above-discussed kinetic data. In fact,
notwithstanding, Lineweaver−Burk plots indicated that
genistein (2) is the only competitive inhibitor, and 1, 3, 10,
and 11 resulted in mixed inhibitors; the latter showed an
affinity for the PL catalytic site (similarly to a competitive
inhibitor) greater than for the PL−substrate complex.
In conclusion, in this work, a detailed study on inhibition of
the pancreatic lipase activity by natural isoflavones and
semisynthetic derivatives has been carried out for the first
time. The above-reported results highlight the subclass of
isoflavones as promising polyphenols for the development of
potential antiobesity agents. An in vitro lipase inhibition assay
showed that the natural isoflavone 2 is the most potent PL
inhibitor. However, hydroxylation and bromination resulted in
advantageous modifications for improving the PL inhibitory
activity of isoflavones, as confirmed by intrinsic fluorescence
and kinetic measurements. The docking study pointed out the
molecular interactions involved between the isoflavones and
the PL catalytic site, thus supporting our findings.
EXPERIMENTAL SECTION
■
General Experimental Procedures. UV−vis spectra were
acquired with a Jasco V630 spectrophotometer using a 1 cm quartz
cuvette at 27 °C and in the range 200−600 nm.
The two isoflavones were recovered from the organic crude after
column chromatography on silica gel was eluted with CH2Cl2:MeOH
(100:0 → 85:15).
3-(3,4-Dihydroxyphenyl)-7-hydroxy-4H-chromen-4-one (7).
There is a 42−57% yield from two-step synthesis and 94% yield
from one-pot reaction. Rf (TLC) = 0.30 (93:7 CH2Cl2:MeOH). UV
(50:50 CH3OH:H2O) λmax(logε) = 254 (4.32), 403 (1.65) nm. The
spectroscopic data were in agreement with those reported in the
literature.28
NMR spectra were run on a Varian Unity Inova spectrometer
operating at 499.86 (1H) and 125.70 MHz (13C). All NMR
experiments, including 2D spectra, g-COSY, g-HSQCAD, and g-
HMBCAD, were acquired at constant temperature (27 °C). g-
HMBCAD experiments were optimized for a long-range 13C−1H
coupling constant of 8.0 Hz. Chemical shifts are indirectly referred to
tetramethylsilane using residual solvent signals at 2.50 ppm for
DMSO-d6 and 7.26 ppm for chloroform-d.
662
https://dx.doi.org/10.1021/acs.jnatprod.0c01387
J. Nat. Prod. 2021, 84, 654−665

Journal of Natural Products
pubs.acs.org/jnp
Article
3-(3,4-Dihydroxyphenyl)-7-hydroxy-5-methyl-4H-chromen-4-
Synthesis of 8-Bromo-7-hydroxy-3-(4-methoxyphenyl)-4H-chro-
men-4-one (11). Formononetin (3; 200 mg; 0.74 mmol) was
solubilized with CHCl3:MeOH 90:10 (9.5 mL), and the solution was
kept in an ice bath. A freshly prepared Br2 solution (40 μL; 0.74
mmol; in 32 mL of CHCl3) was poured dropwise to 3. The mixture
was stirred at 0 °C, and after 4.5 h, a saturated solution of Na2S2O3
(20 mL) was added. The two phases were separated, and the
aqueous phase was partitioned with CHCl3 (3 × 20 mL). The
organic layer was washed with H2O and dried over anhydrous
Na2SO4, and the solvent was evaporated until dry. The isoflavone 11
was obtained with 80% yield without further purification. Rf (TLC) =
0.64 (95:5 CH2Cl2:MeOH). UV (50:50 MeOH:H2O) λmax(log ε) =
261 (4.26), 340 (3.75) nm. 1H and 13C NMR, Table 4. HRMS
(ESI−) m/z 344.9781 [M − H]− (100) (calcd for C16H1079BrO4,
344.9762) and 346.9762 [M − H]− (97) (calcd for C16H1081BrO4,
346.9742).
one (8). There is a 16% yield from two-steps synthesis and 37% yield
1
from one-pot reaction. Rf (TLC) = 0.72 (93:7 CH2Cl2:MeOH). H,
13C, and HMBC NMR (Table 2). HRMS (ESI−) m/z 283.0628 [M
− H]− (calcd for C16H11O5, 283.0606).
Hydroxylation of Isoflavones 1 and 3. The preliminary
experiments are reported in the Supporting Information.
Synthesis of 7. Daidzein (15 mg; 0.06 mmol) was dissolved in
DMSO (0.6 mL) and treated with IBX (20.2 mg; 0.07 mmol) at 50
°C for 2 h. Then, the mixture was quenched by adding a saturated
Na2S2O4 solution (0.6 mL), and the mixture was stirred at rt for 5
min. The mixture was diluted to 5 mL, and it was partitioned with
EtOAc (3 × 5 mL). The organic layer was washed with H2O, dried
over anhydrous Na2SO4, and taken to dryness. The column
chromatography on silica gel eluting with CH2Cl2:MeOH (from
100:0 → 90:10) afforded 7 with 28% yield. Spectroscopic data were
in agreement with those reported in the literature.28
Synthesis of Retusin (9). Formononetin (100.2 mg, 0.37 mmol)
was dissolved in DMSO (3.5 mL), and IBX (124.7 mg; 0.44 mmol)
was added. Then, the reaction was treated with 4 mL of a saturated
Na2S2O4 solution at rt for 5 min. The mixture was diluted to 15 mL,
and it was partitioned with EtOAc (3 × 15 mL). The organic layer
was washed with H2O and dried over anhydrous Na2SO4, and the
solvent was evaporated until dryness. The column chromatography
on silica gel eluting with n-hexane:acetone (from 100:0 → 50:50)
afforded 9 with 43% yield. Rf (TLC) = 0.42 (60:40 n-
hexane:acetone). UV (50:50 MeOH:H2O) λmax(log ε) = 259
(4.53), 305 (3.92) nm. Spectroscopic data were in agreement with
those reported in the literature.26
Synthesis of 8-Bromo-3-(3-bromo-4-methoxyphenyl)-7-hy-
droxy-4H-chromen-4-one (12). Formononetin (3; 100 mg; 0.37
mmol) was solubilized with glacial acetic acid (1.5 mL). A freshly
prepared Br2 solution (20 μL; 0.44 mmol; in 16 mL of CH3CN) was
added dropwise to 3. The mixture was stirred at rt, and after 40 min,
the reaction was quenched by addition of a saturated solution of
Na2S2O3 (10 mL). The crude was partitioned with CHCl3 (3 × 20
mL). The organic layer was washed with water and dried over
anhydrous Na2SO4, and the solvent was evaporated until dry. The
isoflavone 12 was obtained with 92% yield without further
purification. Rf (TLC) = 0.50 (97:3 CH2Cl2:MeOH). UV (50:50
1
MeOH:H2O) λmax(log ε) = 255 (4.21), 344 (3.69) nm. H and 13C
NMR, Table 4. HRMS (ESI−) (1:2:1 ratio) m/z 422.8895 [M −
H]− (calcd for C16H979Br2O4, 422.8867), 424.8873 [M − H]− (100)
(calcd for C16H979Br81BrO4, 424.8847), 426.8852 [M − H]− (50)
(calcd for C16H981Br2O4, 426.8827).
Bromination of Isoflavones 1 and 3. For bromination of 1 and
3, some preliminary screenings were performed as follows:
(1) Compound 3 (4.0 mg; 0.015 mmol) was stirred with
CH3COOH (75 μL), NaBr (1.5 mg; 0.015 mmol), and
30% H2O2 (50 μL; 0.044 mmol) at rt for 6 h;
(2) Compound 3 (4.0 mg; 0.015 mmol) was stirred with
CH3COOH (75 μL), NaBr (1.5 mg; 0.015 mmol), and
oxone (11.2 mg; 0.017 mmol, previously dissolved in 50 μL of
water) at rt for 6 h;
(3) Compound 3 (4.0 mg; 0.015 mmol) was dissolved with
CH3COOH (75 μL), and a Br2 solution (1 μL; 0.017 mmol;
in 850 μL of CH3CN) was dropwise added; the mixture was
mixed at rt for 40 min;
(4) Compound 3 (3 mg; 0.011 mmol) was dissolved in CHCl3
(0.115 mL) and MeOH (20 μL); then, a Br2 solution (0.6 μL;
0.011 mmol in 450 μL of CHCl3) was dropwise added, and
the mixture was kept at 0 °C for 1 h;
Synthesis of Derivatives of 7. Synthesis of Cabreuvine (13).
The isoflavone 7 (40.2 mg; 0.14 mmol) was dissolved in dry acetone
(3 mL), and it was stirred with anhydrous K2CO3 (174.7 mg;1.23
mmol) at rt for 10 min. Then, CH3I (85 μL; 1.23 mmol) was added
to the reaction flask, and the mixture was refluxed for 24 h. After 5 h,
another aliquot of CH3I (20 μL, 0.32 mmol) was added. The
expected product was recovered after filtration and distillation of the
solvent by rotavapor with 52% yield. Rf (TLC) = 0.80 (96:4
CH2Cl2:MeOH). UV (50:50 MeOH:H2O) λmax(log ε) = 250 (4.52),
304 (4.02) nm. NMR spectroscopic data of 13 were in agreement
with those reported in the literature for cabreuvin.31
Synthesis of 4-(7-Acetoxy-4-oxo-4H-chromen-3-yl)-1,2-phenyl-
ene diacetate (14). The isoflavone 7 (40.9 mg; 0.14 mmol) was
solubilized with CHCl3 (200 μL), and it was mixed with anhydrous
K2CO3 (76.7 mg; 0.56 mmol) and acetic anhydride (55 μL; 0.6
mmol) at rt for 19 h. The mixture was partitioned with EtOAc (3 ×
20 mL), and the combined organic layer was washed with H2O and
dried over anhydrous Na2SO4; the solvent was evaporated until dry.
Column chromatography on diol silica gel eluted with n-
hexane:CH2Cl2 (20:80 → 0:100) furnished the acetylated isoflavone
14 with 51.0% yield. Rf(TLC) = 0.37 (98:2 CH2Cl2/MeOH). UV
(50:50 MeOH:H2O) λmax(log ε) = 248 (4.67), 303 (4.08) nm. The
1H and 13C NMR data are listed in Table 4. HRMS (ESI+) m/z
(5) Compound 1 (20 mg; 0.08 mmol) was dissolved in CHCl3
(1.2 mL) and MeOH (0.1 mL); then, a Br2 solution (4.0 μL;
0.08 mmol; in 3.5 mL of CHCl3) was dropwise added, and
the mixture was kept at 0 °C for 3 h.
The reactions were monitored by TLC, and they were quenched
by addition of a saturated Na2S2O3 solution (1 mL). The crude was
partitioned with CHCl3 (3 × 1 mL), and the organic layer was
washed with H2O, dried over anhydrous Na2SO4, and taken to
dryness.
419.0780 [M + Na]+ (calcd for C21H16NaO8, 419.0743).
Measurements of Lipase Inhibition. The PL inhibition assay
was performed adopting the conditions previously reported.33,41
Briefly, in a 96-well microplate, 150 μL of phosphate buffer (50 mM,
pH = 7.2), the PL solution (300 U/mL in phosphate buffer; 15 μL),
and different aliquots (2, 4, 6, 8, 10, and 15 μL) of tested compounds
(stock solutions ranging from 1 mM to 5 mM were prepared in
MeOH or MeOH:DMSO 95:5) or of orlistat (6.7 μM in buffer)
were mixed. The reactions were incubated at 37 °C for 10 min.
Then, the substrate p-nitrophenyl butyrate (3.2 mM in H2O:DMF
70:30, 10 μL) was added, and the microplate was incubated at 37 °C
for 30 min under moderate shaking. The plate measurements were
acquired at 405 nm. The assays were performed in triplicate with five
different concentrations for each compound. The amount of MeOH
or DMSO used in the experiment did not affect the lipase inhibitory
Synthesis of 8-Bromo-3-(3,5-dibromo-4-hydroxyphenyl)-7-hy-
droxy-4H-chromen-4-one (10). Daidzein (1; 100 mg; 0.39 mmol)
was solubilized with CHCl3: MeOH 90:10 (6 mL) and stirred in an
ice bath. A Br2 solution (20 μL; 0.39 mmol) in CHCl3 (17 mL),
freshly prepared, was added dropwise to 1. After 3 h, the reaction
was quenched by addition of a saturated Na2S2O3 solution (15 mL).
The crude was partitioned with CHCl3 (3 × 15 mL); the total
organic layer was washed with H2O (30 mL), and dried over
anhydrous Na2SO4. The solvent was evaporated until dryness. The
isoflavone 10 was recovered with 70% yield after column
chromatography on silica gel and eluting with n-hexane:acetone
(90:10 → 60:40). Rf(TLC) = 0.46 (60:40 n-hexane/acetone).
Spectroscopic data were in agreement with those reported in the
literature.30
663
https://dx.doi.org/10.1021/acs.jnatprod.0c01387
J. Nat. Prod. 2021, 84, 654−665

Journal of Natural Products
pubs.acs.org/jnp
Article
activity. The inhibition percentage was calculated by the following
equation:
Molecular Docking Analysis. The .sdf files of orlistat, genistein,
daidzein, and formononetin were downloaded from Pubchem
(https://pubchem.ncbi.nlm.nih.gov/, ID codes, respectively, are
3034010, 5280961, 5281708, and 5280378). Meanwhile, the other
ligands, namely, the isoflavones 7−14 were drawn by Chemdraw and
saved in .sdf files. The 3D models were geometrically minimized with
optimized potential liquid simulation (OPLS3) force fields44
(Acontrol − Asample
)
inhibition % =
× 100
Acontrol
(1)
where Acontrol is the absorbance measured for the mixture enzyme/
substrate (without tested compounds) and Asample is the absorbance
measured in the same conditions and in the presence of the tested
compounds. The concentration required to inhibit the 50% activity
of the enzyme (IC50) was calculated by regression analysis.
considering the protonation state at pH of 7.0
1 and processed
̈
using LigPrep interfaced with Maestro (Version 11) of Schroedinger
suite.45 The minimized geometries were converted in .pdbqt files by
Autodock Tools (1.5.6).
Fluorescence Spectra Measurements. The interaction be-
tween the isoflavones and PL was performed by fluorimetric titration.
For each tested compound, the experiments (each in triplicate) were
performed at 27, 32, and 37 °C. The PL (1.1 μM in 0.1 M phosphate
buffer containing 0.1 M NaCl, pH 7.4; 2 mL) was titrated by
successive additions (1 or 2 μL) of the tested compounds. The
fluorescence spectrum was acquired 1 min after each addition from
310 to 500 nm. The concentration of starting solution of the
isoflavones were chosen on the basis of the IC50 values. The
fluorescence at the maximum intensity was employed to obtain the
Stern−Volmer plots according eqs 2 and 3:35
The 3D structure of pancreatic lipase (PDB ID: 1LPB) was
downloaded from Protein Data Bank.46 The lipase is cocrystallized
with colipase in the presence of methoxy undecyl phosphonic
(MUP) as inhibitor and β-octylglucoside (BOG) as surfactant.
Protein was prepared with Protein wizard, MUP, and BOG, and H2O
molecules were removed; then, the .pdb file obtained was processed
with Autodock Tools 1.5.6 and converted in .pdbqt file, merging
nonpolar hydrogens and adding Gasteiger charges. The molecular
docking studies were performed using Autodock4 (AD4)37 and
Autodock Vina software 1.5.6.38,39 Autogrid4 (4.2.6 version) was
used to generate the gridmaps. The grid box was centered in the
binding site of protein, with grid coordinates of 50 × 40 × 50 Å3 for
x, y, and z, respectively. The spacing between the grid coordinates
was 0.708 Å. The grid center was set to 7.500, 26.042, and 47.696 Å.
For the docking experiments carried out with AD4, The Lamarckian
Genetic Algorithm (LGA) was chosen to search for the best
conformers. During the docking process, a maximum of 10
conformers was considered for each ligand. AD4 GA parameters
are set as default: ga_pop_size = 150, ga_num_evals = 2 500 000,
ga_num_generations = 27 000, ga_elitism = 1, ga_mutation_rate =
0.02, ga_crossover_rate = 0.8, ga_cauchy_alpha = 0.0, ga_cauchy_-
beta = 1.0. The docking carried out with autodock Vina was set with
exhaustiveness and energy difference as default (8 and 3 respectively)
and saving 9 conformations as the maximum number of binding
mode. In Docking calculation, ligands were treated as flexible, and
protein were treated as rigid. AD4 and autodock Vina were compiled
and run in OSX Yosemite (10.10.5) environment. The analysis of
docking outcomes were carried out by Autodock Tool (1.5.6), and
figures of 3D models were generated by Pymol (2.3.5).
F
F
o
= 1 + KSV[Q ] = 1 + Kqτ0[Q ]
(2)
F − F
F
0
log
= log Ka + nlog [Q ]
(3)
F0 and F are the fluorescence intensities of enzyme before and after
the addition of quencher, respectively. [Q] represents the
concentration of the isoflavones studied. τ0 is the average life of
protein, and it was reported to be 1.59 × 10−9s.42 KSV and Kq
represent the quenching constant and the quenching rate constant,
respectively. Ka is the binding constant, and n is the number of
binding sites.
Kinetics of Lipase Inhibition. The mode of inhibition of PL in
the presence of tested isoflavones was determined by elaboration of
UV−vis spectroscopic data with Lineweaver−Burk plots. The
experiments were performed in 96-well plates employing the enzyme
(120 U/mL in phosphate buffer; 10 μL), the selected isoflavones,
and the increasing concentrations of the substrate p-nitrophenyl
butyrate (from 0.3 to 1.9 mM) in a final volume of 200 μL. Optimal
concentration of the tested isoflavones were chosen on the basis of
the IC50 values. The absorbance was read at 405 nm every 1 min for
30 min at 37 °C.
ASSOCIATED CONTENT
■
sı
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.jnatprod.0c01387.
The inhibition constants were calculated from the equations
vmax
S
I
Experimental details for preliminary hydroxylation
reactions of 1−3; preliminary experiments of bromina-
tion of 2; HRMS, 1H, 13C, and 2D NMR spectra of new
compounds 8, 11, 12, and 14; fluorescence spectra of
PL in presence of 1−3, 10, and 11; and list of binding
energy and noncovalent interactions between isofla-
vones 1−3, 7−14, and PL (PDF)
v0 =
v0 =
Km 1 +
+ S
(
)
Ki
(4)
(5)
vmax
S
I
I
K′i
Km 1 +
+ S 1 +
(
)
(
)
Ki
for competitive and mixed-type inhibitors, respectively, where ν0 is
the initial velocity in the absence and presence of the inhibitor, S and
I are the concentrations of substrate and inhibitor, respectively, νmax
is the maximum velocity, Km is the Michaelis−Menten constant, Ki is
the competitive inhibition constant, and K’i is the uncompetitive
inhibition constant.
The graphs of slope and y-intercept of Lineweaver−Burk plots
versus the inhibitor concentration gave a straight line, whose
intercept corresponds to Ki and K’i values, respectively.43
Data Analysis. Results of the above-reported experiments were
obtained by plotting the experimental measurements on OriginPro
2018 or Microsoft Excel 2016 and were expressed as means SD.
Statistical analysis was performed using one-way analysis of variance
(ANOVA), and differences were designated as statistically significant
when p < 0.05.
AUTHOR INFORMATION
■
Corresponding Authors
Corrado Tringali − Dipartimento di Scienze Chimiche,
̀
Universita degli Studi di Catania, 95125 Catania, Italy;
orcid.org/0000-0003-4268-0115; Phone: +39
0957385025; Email: ctringali@unict.it; Fax: +39
095580138
Nunzio Cardullo − Dipartimento di Scienze Chimiche,
̀
Universita degli Studi di Catania, 95125 Catania, Italy;
orcid.org/0000-0003-1480-0861; Email: ncardullo@
unict.it
664
https://dx.doi.org/10.1021/acs.jnatprod.0c01387
J. Nat. Prod. 2021, 84, 654−665

Journal of Natural Products
pubs.acs.org/jnp
Article
(20) Mu, Y.; Kou, T.; Wei, B.; Lu, X.; Liu, J.; Tian, H.; Zhang, W.;
Liu, B.; Li, H.; Cui, W.; Wang, Q. Nutrients 2019, 11, 2790.
(21) Akhlaghi, M.; Zare, M.; Nouripour, F. Adv. Nutr. 2017, 8,
705−717.
(22) Ørgaard, A.; Jensen, L. Exp. Biol. Med. 2008, 233, 1066−1080.
(23) Cardullo, N.; Spatafora, C.; Musso, N.; Barresi, V.; Condorelli,
D.; Tringali, C. J. Nat. Prod. 2015, 78, 2675−2683.
(24) Pulvirenti, L.; Muccilli, V.; Cardullo, N.; Spatafora, C.;
Tringali, C. J. Nat. Prod. 2017, 80, 1648−1657.
(25) Cardullo, N.; Catinella, C.; Floresta, G.; Muccilli, V.; Rosselli,
S.; Rescifina, A.; Bruno, M.; Tringali, C. J. Nat. Prod. 2019, 82, 573−
582.
Authors
̀
Vera Muccilli − Dipartimento di Scienze Chimiche, Universita
degli Studi di Catania, 95125 Catania, Italy; orcid.org/
0000-0002-6212-5154
Luana Pulvirenti − Dipartimento di Scienze Chimiche,
Universita degli Studi di Catania, 95125 Catania, Italy
̀
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.jnatprod.0c01387
Notes
The authors declare no competing financial interest.
(26) Yadav, S. K. Int. J. Org. Chem. 2014, 4, 236−246.
(27) Balasubramanian, S.; Nair, M. G. Synth. Commun. 2000, 30,
469−484.
ACKNOWLEDGMENTS
■
(28) Goto, H.; Terao, Y.; Akai, S. Chem. Pharm. Bull. 2009, 57,
346−360.
This research was funded by the by MIUR ITALY PRIN 2017
(Project No. 2017A95NCJ). The authors acknowledge the
Bio-Nanotech Research and Innovation Tower (BRIT) of
University of Catania for the availability of the Synergy H1
microplate reader.
(29) Barontini, M.; Bernini, R.; Crisante, F.; Fabrizi, G. Tetrahedron
2010, 66, 6047−6053.
(30) Lee, S. R.; Schalk, F.; Schwitalla, J. W.; Benndorf, R.; Vollmers,
J.; Kaster, A.-K.; de Beer, Z. W.; Park, M.; Ahn, M.-J.; Jung, W. H.;
Beemelmanns, C.; Kim, K. H. J. Nat. Prod. 2020, 83, 3102−3110.
(31) Ndemangou, B.; Sielinou, V. T.; Vardamides, J. C.; Ali, M. S.;
Lateef, M.; Iqbal, L.; Afza, N.; Nkengfack, A. E. J. Enzyme Inhib. Med.
Chem. 2013, 28, 1156−1161.
DEDICATION
■
Dedicated to Dr. A. Douglas Kinghorn, The Ohio State
University, for his pioneering work on bioactive natural
products.
(32) Funayama, S.; Anraku, Y.; Mita, A.; Komiyama, K.; Omura, S.
J. Antibiot. 1989, 42, 1350−1355.
(33) Roh, C.; Jung, U. Int. J. Mol. Sci. 2012, 13, 1710−1719.
(34) van der Weert, M.; Stella, L. J. Mol. Struct. 2011, 998, 144−
150.
(35) Cardullo, N.; Muccilli, V.; Pulvirenti, L.; Cornu, A.; Pouysegu,
L.; Deffieux, D.; Quideau, S.; Tringali, C. Food Chem. 2020, 313,
126099.
REFERENCES
■
(1) World Health Organization. Fact Sheet Obesity and Over-
Weight. https://www.who.int/news-room/fact-sheets/detail/obesity-
and-overweight/ (accessed October 2020).
(2) Liu, T. T.; Liu, X. T.; Chen, Q. X.; Shi, Y. Biomed.
Pharmacother. 2020, 128, 110314.
(36) Li, Y. Q.; Zhou, F. C.; Gao, F.; Bian, J. S.; Shan, F. J. Agric.
Food Chem. 2009, 57, 11463−11468.
(3) Sergent, T.; Vanderstraeten, J.; Winand, J.; Beguin, P.;
Schneider, Y.-J. Food Chem. 2012, 135, 68−73.
(37) Mohapatra, S.; Prasad, A.; Haque, F.; Ray, S.; De, B.; Ray, S. S.
J. Appl. Pharm. Sci. 2015, 5, 042−047.
(4) Birari, R. B.; Bhutani, K. K. Drug Discovery Today 2007, 12,
879−889.
(38) Vu, H. M. T.; Vu, D. D.; Truong, V. D.; Nguyen, T. V. P.;
Tran, T. D. MedPharmRes. 2017, 1, 26−36.
(5) Seyedan, A.; Alshawsh, M. A.; Alshagga, M. A.; Koosha, S.;
Mohamed, Z. Evidence-Based Complementary and Alternative Medicine
2015, 2015, 1−13.
(39) Tang, W.-Z.; Liu, J.-T.; Hu, Q.; He, R.-J.; Guan, X.-Q.; Ge, G.-
B.; Han, H.; Yang, F.; Lin, H.-W. J. Nat. Prod. 2020, 83, 2287−2293.
(40) Cardullo, N.; Barresi, V.; Muccilli, V.; Spampinato, G.;
D’Amico, M.; Condorelli, D. F.; Tringali, C. Molecules 2020, 25, 733.
(41) Slanc, P.; Doljak, B.; Kreft, S.; Lunder, M.; Janes, D.; Strukelj,
B. Phytother. Res. 2009, 23, 874−877.
(6) Yung Cheung, B. M.; Tsang Cheung, T.; Samaranayake, N. R.
Ther. Adv. Drug Saf. 2013, 4, 171−181.
(7) Buchholz, T.; Melzig, M. F. Planta Med. 2015, 81, 771−783.
(8) Yun, J. W. Phytochemistry 2010, 71, 1625−1641.
(9) Martinez-Gonzalez, A. I.; Alvarez-Parrilla, E.; Diaz-Sanchez, A.
G.; de la Rosa, L. A.; Nunez-Gastelum, J. A.; Vazquez-Flores, A. A.;
Gonzalez-Aguilar, G. A. Food Technol. Biotechnol. 2017, 55, 519−530.
(10) Weibel, E. K.; Hadvary, P.; Hochuli, E.; Kupfer, E.; Lengsfeld,
H. J. Antibiot. 1987, 40, 1081−1085.
(42) Nasri, R.; Bidel, L. P. R.; Rugani, N.; Perrier, V.; Carriere, F.;
Dubreucq, E.; Jay-Allemand, C. Molecules 2019, 24, 2888.
(43) Meiering, S.; Inhoff, O.; Mies, J.; Vincek, A.; Garcia, G.;
Kramer, B.; Dormeyer, M.; Krauth-Siegel, R. L. J. Med. Chem. 2005,
48, 4793−4802.
(44) Harder, E.; Damm, W.; Maple, J.; Wu, C.; Reboul, M.; Xiang,
J. Y.; Wang, L.; Lupyan, D.; Dahlgren, M. K.; Knight, J. L.; Kaus, J.
W.; Cerutti, D. S.; Krilov, G.; Jorgensen, W.; Abel, R.; Friesner, R. A.
J. Chem. Theory Comput. 2016, 12, 281−296.
(45) Schrodinger Release 2017-1, LigPrep; Schrodinger, LLC: New
York, NY, 2017.
(11) Hochuli, E.; Kupfer, E.; Maurer, R.; Meister, W.; Mercadal, Y.;
Schmidt, K. J. Antibiot. 1987, 40, 1086−1091.
(12) Laura de la Garza, A.; Milagro, F. I.; Boque, N.; Campion, J.;
Alfredo Martinez, J. Planta Med. 2011, 77, 773−785.
(13) Goncalves, R.; Mateus, N.; De Freitas, V. J. Agric. Food Chem.
2010, 58, 11901−11906.
(46) RCSB. Protein Data Bank. https://www.rcsb.org/structure/
1lpb (accessed April 2020).
(14) Vitale, D. C.; Piazza, C.; Melilli, B.; Drago, F.; Salomone, S.
Eur. J. Drug Metab. Pharmacokinet. 2013, 38, 15−25.
(15) Yu, J.; Bi, X. J.; Yu, B.; Chen, D. W. Nutrients 2016, 8, 361.
(16) Andres, S.; Abraham, K.; Appel, K. E.; Lampen, A. Crit. Rev.
Toxicol. 2011, 41, 463−506.
(17) Umeno, A.; Horie, M.; Murotomi, K.; Nakajima, Y.; Yoshida,
Y. Molecules 2016, 21, 708.
(18) Wang, D. M.; Hu, M.; Li, X. P.; Zhang, D.; Chen, C. J.; Fu, J.
M.; Shao, S.; Shi, G. N.; Zhou, Y.; Wu, S.; Zhang, T. T. Eur. J. Med.
Chem. 2019, 168, 207−220.
(19) Shimura, S.; Itoh, Y.; Yamashita, A.; Kitano, A.; Hatano, T.;
Yoshida, T.; Okuda, T. Nippon Shokuhin Kogyo Gakkaishi 1994, 41,
847−850.
665
https://dx.doi.org/10.1021/acs.jnatprod.0c01387
J. Nat. Prod. 2021, 84, 654−665