Materials Science and Engineering C 68 (2016) 594–602

Contents lists available at ScienceDirect

Materials Science and Engineering C

journal homepage: www.elsevier.com/locate/msec

Nanoencapsulation of dietary flavonoid fisetin: Formulation and in vitro

antioxidant and α-glucosidase inhibition activities
Mario Sechi a,b, Deeba N. Syed c, Nicolino Pala a, Alberto Mariani a, Salvatore Marceddu d, Antonio Brunetti e,
Hasan Mukhtar c, Vanna Sanna a,b,⁎
a
Department of Chemistry and Pharmacy, University of Sassari, Via Vienna 2, 07100 Sassari, Italy
b
Laboratory of Nanomedicine, Department of Chemistry and Pharmacy, University of Sassari, c/o Porto Conte Ricerche, Tramariglio, 07041 Alghero, Italy
c
Department of Dermatology, School of Medicine and Public Health, University of Wisconsin, 1300 University Avenue, Madison, USA
d
CNR – Istituto Scienze delle Produzioni Alimentari, Traversa La Crucca, 3 – Località Baldinca, 07040 Li Punti, Sassari, Italy
e
POLCOMING Department, Section of Information Engineering, University of Sassari, via Piandanna 4, 07100 Sassari, Italy

a r t i c l e i n f o a b s t r a c t

Article history: The bioactive flavonoid fisetin (FS) is a diet-derived antioxidant that is being increasingly investigated for its
Received 13 April 2016 health-promoting effects. Unfortunately, the poor physicochemical and pharmacokinetic properties affect and
Received in revised form 31 May 2016 limit the clinical application.
Accepted 12 June 2016
In this study, novel polymeric nanoparticles (NPs), based on Poly-(ε-caprolactone) (PCL) and PLGA-PEG-COOH,
Available online 14 June 2016
encapsulating FS were formulated as suitable oral controlled release systems. Results showed NPs having a mean
Keywords:
diameter of 140–200 nm, and a percent loading of FS ranging from 70 to 82%. In vitro release studies revealed that
Fisetin NPs are able to protect and preserve the release of FS in gastric simulated conditions, also controlling the release
Nanoparticles in the intestinal medium. Moreover, the DPPH and ABTS scavenging capacity of FS, as well as α-glucosidase inhi-
Controlled release bition activity, that resulted about 20-fold higher than commercial Acarbose, were retained during
Antioxidant activity nanoencapsulation process.
α-glucosidase inhibition In summary, our developed NPs can be proposed as an attractive delivery system to control the release of antiox-
idant and anti-hyperglycemic FS for nutraceutical and/or therapeutic application.
© 2016 Elsevier B.V. All rights reserved.

1. Introduction gluconeogenesis in vitro [10–12], mainly attributed to its antioxidant

Flavonoids are a broadly distributed class of plant pigments that are
regularly consumed in the human diet. Due to their abundance in die-
tary products and the potential pharmacological and nutritional effects,
the flavonoids have attracted considerable interest for drug develop-
ment as well as nutraceutical application [1].
One such flavonoid, fisetin (FS; 3,3′,4′,7-tetrahydroxyflavone) (Fig.
1), is found in various fruits and vegetables, such as strawberry, apple,
persimmon, grape, onion, and cucumber [2].
It has been demonstrated that FS possesses multiple pharmacologi-
cal activities including anti-inflammatory [3,4], antihyperlipidemic [5],
neuroprotective [6], antineoplastic, and chemopreventive properties
[7,8].
More recently, FS supplementation has been reported to decrease
cardiovascular risks [9] and downregulate both glycogenolysis and
⁎ Corresponding author at: Department of Chemistry and Pharmacy, University of
Sassari, Via Vienna 2, 07100 Sassari, Italy.
E-mail address: vsanna@uniss.it (V. Sanna).
capacities.
In spite of the beneficial effects of FS in preventing and treating dis-
eases, its low water solubility, chemical instability, poor absorption, and
extensive and rapid metabolism, dramatically contribute to the low oral
bioavailability, thus restraining its clinical application [13,14].
In this context, nanoencapsulation is one of the drug/nutraceutical
delivery processes that can help in overcoming these physicochemical
and pharmacokinetic limitations.
To date, very few reports are available in the literature on nanosys-
tems for the encapsulation of FS. For example, nanoemulsion and lipo-
somal formulations have been developed by Chabot's group to
improve the FS bioavailability and the antitumor efficacy in Lewis lung
carcinoma bearing mice [15–17]. It has also been reported that the en-
capsulation of FS into monomethyl poly (ethylene glycol)-poly (ε-
caprolactone) copolymer micelles improved the therapeutic effect in
colon cancer [18].
In this scenario, the present study was aimed to formulate FS into
novel polymeric nanoparticles (NPs), based on poly (ε-caprolactone)
(PCL) and poly(D,L-lactic-co-glycolic acid)-block-poly(ethylene glycol)
carboxylic acid (PLGA-PEG-COOH), suitable as oral controlled release

http://dx.doi.org/10.1016/j.msec.2016.06.042
0928-4931/© 2016 Elsevier B.V. All rights reserved.
 M. Sechi et al. / Materials Science and Engineering C 68 (2016) 594–602
Fig. 1. Chemical structure of fisetin (FS).
systems. NPs were characterized for morphology, size, FS loading, FTIR
and XRD analyses, and in vitro release studies in gastrointestinal simu-
lated fluids. In order to evaluate the retention of FS activity after encap-
sulation, the DPPH and ABTS radical scavenging activities of
encapsulated FS were assessed in comparison to those of free FS. Fur-
thermore, the α-glucosidase inhibition capacity of FS and FS-loaded
NPs was evaluated and compared to commercial drug Acarbose.
2. Materials and methods
2.1. Materials
Poly-(ε-caprolactone) (PCL), 2,2-diphenyl-1-picryl hydrazyl
(DPPH), 2,20-azinobis(3-ethylbenzothiazoline-6-sulphonic acid-)
diammonium salt (ABTS), sodium persulfate, α-glucosidase from Sac-
charomyces cerevisiae (type I; 10 U/mg protein), p-nitrophenyl α-D-
glucopyranoside (pNPG), and trolox (6-hydroxy-2,5,7,8-
tetramethylchromane-2-carboxylic acid) were obtained from Sigma-
Aldrich (Steinheim, Germany). Fisetin (FS), from leaves of Rhus succeda-
nea L. (N 98% purity), was purchased from Shaanxi Taiji Huaqing Tech-
nology Co., Ltd. (Shaanxi, People's Republic of China). Acarbose was
obtained from Carbosynth Limited (Berkshire, UK). The PLGA-PEG-
COOH block copolymer was prepared and characterized as previously
reported [19]. All solvents and other chemicals were obtained from
Sigma-Aldrich, were analytical grade and were used without further
purification.
2.2. Preparation of nanoparticles
Fisetin from leaves of Rhus succedanea L. was loaded into NPs by
modification of a nanoprecipitation method previously described [19].
Briefly, polymers at different weight ratio and FS were dissolved in
3 mL of acetonitrile. The organic solution was then added dropwise
under magnetic stirring to an aqueous solution of Pluronic F-127
(0.1%, wt.%), giving a final polymer concentration of 5.0 mg/mL.
Table 1 reports the total amount of solid materials used for the prep-
aration of FS-loaded NPs.
The resulting suspension was stirred at room temperature to remove
the organic solvent, and NPs were centrifuged at 5000 rpm for 5 min,
washed three times with water and concentrated. Unloaded NPs
batches (F01, F02, and F03) were produced in a similar manner and
Table 1
Theoretical composition (wt.%) of prepared loaded NPs.
Batch PCL
F1 95.0
F2 66.5
F3 47.5
595
used as comparison. The obtained NPs suspension was used immediate-
ly for assay or lyophilized for storage at −50 °C.
2.3. Characterization of nanoparticles
2.3.1. Measurement of particle size (PS) and Polydispersity Index (PI)
PS and PI of the NPs were measured by using photon correlation
spectroscopy with a Zetasizer Nano ZS (Malvern Instruments,
Worcestershire, UK) at 25 °C, and a scattering angle of 90°. The samples
were diluted with deionized water and sonicated for 5 min before mea-
surement. Each analysis was carried out in triplicate.
2.3.2. Scanning Electron Microscopy (SEM) analysis
Morphological examination of NPs was performed by Scanning Elec-
tron Microscopy (SEM) (model DSM 962; Carl Zeiss Inc., Germany). A
drop of NPs aqueous suspension was placed on aluminum stub and
dried under vacuum for 12 h. The samples were then analyzed at
20 kV acceleration voltage after gold sputtering, under an argon
atmosphere.
2.3.3. FS loading (DLC%), entrapment efficiency (EE%), and yield of NPs
(YP%)
The amount of FS encapsulated was determined by dissolving an al-
iquot of NPs (1.5 mg) in 1 mL of acetonitrile: EtOH mixture (7:3, v/v).
The solution was analyzed by UV–vis spectroscopy (Cary 3, Varian) at
350 nm, and calculated by referring to the calibration curve (standard
solutions in the range of 1.25–50 μg/mL; R2 = 0.999). The DLC%, EE%,
and YP% were presented by the following equations, respectively:
DLC% ¼ ðweight of drug in NPs=weight of NPsÞ  100
EE% ¼ ðactual RSV content=theoretical RSV contentÞ  100
YP% ¼ ðweight of NPs recovered=weight of polymer and RSV fed initiallyÞ
 100
2.3.4. Fourier transform infrared spectroscopy (FTIR)
The chemical composition of FS, unloaded, and FS-loaded NPs was
analyzed by FTIR spectroscopy using a Vertex 70 Bruker spectropho-
tometer equipped with a RT-DTGS detector and a KBr beam splitter.
The spectra were recorded in the 400–4000 cm−1 range with a resolu-
tion of 4 cm−1.
2.3.5. X-ray diffraction (XRD)
The physical state of the pure FS, PCL and PLGA-PEG-COOH poly-
mers, empty and loaded NPs, as well as the FS-polymers physical mix-
tures was analyzed by XRD (Bruker — Mod. D2 Phaser), using CuK α-
radiation (λ = 1.5418 Å) and operating at 30 kV and 10 mA. The X-
ray scans were performed in open angle 2θ in the range from 10 to
100°, with the scanning rate at 1°/min.
2.4. In vitro release studies
FS-loaded NPs were characterized by in vitro release kinetic at differ-
ent pH conditions: at pH 1.2 for 2 h, followed by pH 7.4, to simulate gas-
tric and intestinal fluids, respectively. About 2.0 mg of FS-loaded NPs
were placed into dialysis bags and suspended in 15 mL of release
PLGA-PEG-COOH FS
– 5.0
28.5 5.0
47.5 5.0
596 M. Sechi et al. / Materials Science and Engineering C 68 (2016) 594–602
media, then incubated at 37 °C and stirred at 200 rpm. At predetermined
time intervals, 1 mL of samples were withdrawn and replaced with
equal volume of the corresponding fresh media to maintain a constant
volume. Samples were filtered, and EtOH (4%, v/v) was added in order
to ensure the complete solubilization of FS released.
FS concentration was detected spectrophotometrically at 350 nm,
thus calculated by referring to the respective calibration curve with
standard solutions in the range of 1.25–50 μg/mL, prepared by using
0.1 N HCl:EtOH (96:4, v/v) for pH 1.2 conditions (R2 = 0.9992), and
PBS:EtOH (96:4, v/v), for pH 7.4 conditions (R2 = 0.9991). Each exper-
iment was performed in triplicate.
2.5. Determination of antioxidant activity
In order to evaluate the antioxidant activity of FS and its retention
after encapsulation, the DPPH and ABTS radical scavenging activities
were investigated on free FS and on FS after extraction from NPs (FS-
NPs) by dissolving a known amount of loaded-NPs in a mixture of
ACN:EtOH (4:1, v/v), and diluting them to give different final concentra-
tions of FS.
2.5.1. DPPH radical scavenging activity
DPPH scavenging activity of FS before and after encapsulation was
determined according to the methods previously reported, with minor
changes [20–22]. A solution of DPPH (0.1 mM, 1.4 mL) in 80% aqueous
methanol and 0.1 mL of sample (free FS and FS-NPs) at different concen-
trations (1.9–40 μg/mL) were mixed in a cell. The absorbance was mea-
sured at 517 nm, and then the solution has been allowed to stand in the
dark for 30 min. Trolox, a water-soluble derivative of vitamin E, in the
concentration range of 5–200 μg/mL, was tested as a positive control,
and distilled water as negative control. Additionally, the corresponding
empty NPs were evaluated considering the same concentration of the
polymers as in FS-loaded NPs.
The scavenging percentage was calculated using the following for-
mula:
DPPH scavenging effect ð%Þ ¼ ½ðAc −As Þ=Ac   100
where Ac and As are absorption of control (distilled water) and sample
applied DPPH at 517 nm, respectively.
2.5.2. ABTS radical cation scavenging activity
The radical scavenging activity of samples against the ABTS radical
cation was measured using the previously reported methods with
some modifications [23,24]. A sodium persulfate stock solution
(6.89 × 10− 3 M, 1.0 mL) was added to ABTS stock solution
(5.00 × 10−4 M, 99.0 mL) and stored in the dark for 16 h before use.
Fifty microlitres of samples at different concentrations (2.5–30 μg/mL)
was mixed with 1.45 mL of ABTS solution. After reaction in darkness
at 37 °C for 10 min, the absorbance at 734 nm was measured.
Trolox (3.1–50 μg/mL) and distilled water were used as a positive
and negative control, respectively. Furthermore, empty NPs were tested
considering the same concentration of the polymers as in FS-NPs.
The scavenging activity was calculated as follows:
ABTS scavenging effect ð%Þ ¼ ½ðAc −As Þ=Ac   100
where Ac is absorbance of a control (distilled water) lacking any radical
scavenger, and As is absorbance of the remaining ABTS•+ in the pres-
ence of a scavenger [25].
2.6. α-Glucosidase inhibition assay
The effect of the free and FS-NPs on α-glucosidase activity was de-
termined according to the previously described procedure [26]. The
substrate solution p-nitropheynyl-α-D-glucopyranoside (pNPG)
(3.0 mM) was prepared in 20 mM phosphate buffer, pH 6.9. One hun-
dred μL of α-glucosidase (0.1 U/mL) was mixed with 50 μL of the differ-
ent concentrations of the free and extracted FS (15.6–250 μg/mL), and
the mixture was incubated at 37 °C for 10 min. After pre-incubation,
50 μL of pNPG solution was added to start the reaction. The reaction
mixture was incubated at 37 °C for additional 10 min and stopped by
adding 2 mL of 0.1 M Na2CO3. The α-glucosidase activity was deter-
mined by measuring the yellow colored para-nitrophenol released
from pNPG at 405 nm. The α-glucosidase inhibition activity was calcu-
lated as follow:
Glucosidase inhibition ð%Þ ¼ ½ðAc −As Þ=Ac   100
where Ac is the absorbance of control and As is the absorbance of sam-
ples, respectively.
Commercial inhibitor Acarbose (1.25–20 mg/mL) was used as a pos-
itive control and distilled water as a negative control. All samples were
assayed in triplicate. As a measure of potency of the inhibitors tested,
IC50 values were calculated from the enzyme activity data and defined
as the concentration of an inhibitor required for reducing 50% of the en-
zyme activity obtained from activity vs concentration plot.
2.7. Statistical analysis
All data were subjected to one-way analysis of variance (ANOVA)
(GraphPadPrism, version 5.03). Individual differences were evaluated
using a nonparametric post hoc test (Tukey's test) and considered sta-
tistically significant at P b 0.05.
3. Results and discussion
3.1. Formulation and characterization of nanoparticles
In this study, FS-loaded NPs were prepared by a nanoprecipitation
technique using PCL and PLGA-PEG-COOH as carriers. These biodegrad-
able and biocompatible polymers have been commonly used to encap-
sulate hydrophobic drugs, which often display poor oral bioavailability
[27].
In our previous studies, NPs based of PCL alone or in combination
with other polymers (PLGA-PEG-COOH, Alginate) have been developed
as carriers for the encapsulation of resveratrol for prostate cancer treat-
ment [19], and white tea extract for nutraceutical application [22].
As shown in Table 2, the mean hydrodynamic diameters range from
140 to 200 nm, depending on the composition of the NPs. The larger
particle size observed for batches containing PLGA-PEG-COOH could
be related to the presence on NPs surface of hydrophilic PEG chains
that orient themselves towards the external aqueous phase, thus deter-
mining an increased of hydrodynamic diameter [28]. Furthermore, all
NPs were characterized by PI values b0.3, indicating a narrow and
unimodal distribution of particles, as also depicted in Fig. 2.
The SEM images (Fig. 3) revealed that the NPs are fine dispersed
with a well-defined spherical shape. Additionally, not significant differ-
ences were observed for the size and morphology of blank NPs (F01-
F03) with respect to corresponding FS-loaded batches (data not
shown).
The encapsulation efficiencies values (Table 2) were found to be 82%,
75%, and 70%, for F1, F2, and F3, respectively.
These findings indicated that composition significantly influences
the loading capacity of FS into NPs (4.1%, 3.74%, and 3.49%, for F1, F2,
and F3, respectively). In particular, the PCL ensures a better encapsula-
tion due to high affinity of hydrophobic FS, which decreased with a pro-
gressive increasing of more hydrophilic PLGA-PEG-COOH and the
reduction of PCL content in the formulations.
Furthermore, the nanoprecipitation technique provided good yields
of production that resulted of about 70%, without significant differences
within different NPs batches.
 M. Sechi et al. / Materials Science and Engineering C 68 (2016) 594–602 597

Table 2
Particle size (PS), Polydispersity Index (PI), encapsulation efficiencies (EE), Drug (FS) loading content (DLC), and yields of production (YP) of formulated NPs. *Significant differences (P b
0.05) between loaded NPs. Data are means ± SD, n = 3.

Batch PS (nm) PI EE (%) DLC (%) YP (%)

F01 143.2 ± 4.3 0.142 ± 0.03 – – 73.32 ± 3.2

F1 146.2 ± 2.3* 0.123 ± 0.05 81.96 ± 3.8* 4.10 ± 0.2* 71.10 ± 2.0
F02 203.21 ± 5.2 0.123 ± 0.04 – – 76.44 ± 4.5
F2 198.7 ± 6.0* 0.158 ± 0.02 74.78 ± 1.9* 3.74 ± 0.1 74.76 ± 2.4
F03 170.22 ± 3.9 0.112 ± 0.01 – – 70.32 ± 3.8
F3 165.4 ± 3.3* 0.148 ± 0.02 69.76 ± 2.8* 3.49 ± 0.1* 72.49 ± 2.7

The qualitative composition of NPs was investigated by FT-IR spec- C–C stretching (ring C), C–O stretching (ring A, B), C–O–H and C–C–H
troscopy (Fig. 4). bending modes (ring A, B) have been observed, and correlated to the
According to the literature [29], the most intense bands of the FS frequency 1163 cm−1. Finally, the bands at 1110 and 1017 cm−1 are
spectrum have been found at 1610 cm−1 region, attributed to combina- assigned to C–C–H, C–O–H, and C–C–C bending vibrations of the ring B.
tion of C_O (ring C), C–C (ring A, B), and O–C (ring B) stretching modes,
and at 1575 cm− 1, due to C–C (ring A, C), C2_C3 (ring C), and C_O
(ring C) stretching vibration. The bands at 1514 and 1448 cm−1 involve
O–C and C–C stretching of ring B and ring A, respectively. Medium in-
tensity bands at 1385 cm−1 are assigned to C–C–O and C–O–H bending
vibrations of the ring C. The signals centered at 1326 cm−1 are attribut-
able to C–C stretching, C3′–OH, and C4′–OH bending vibrations of the
ring B. Again, the band at 1265 cm−1 corresponds to C–C and C–O
stretching, while band at 1206 cm−1 overally involve C–O stretching
(ring A, C) and C–O–H bending of the ring A (7-OH). Furthermore, the

Fig. 2. Particle size distributions of F1 (A), F2 (B), and F3 (C) NPs. Fig. 3. SEM photographs of F1 (A), F2 (B), and F3 (C) NPs.
598 M. Sechi et al. / Materials Science and Engineering C 68 (2016) 594–602
For F01 NPs, containing only PCL, it was possible to detect strong
bands such as the C_O stretching mode near 1725 cm−1, and bands
at 2949–2865, 1293, 1240, and 1190 cm−1 corresponding to CH2, C–C,
asymmetric C–O–C, and O–C–O stretching vibrations, respectively
[30]. The spectrum of F03 NPs, containing PCL and PLGA-PEG-COOH
mixture, in addition to F0 spectrum, shows the strong signal at
1757 cm−1, assigned to C_O stretching vibrations. As previously re-
ported [19], the FTIR spectrum of PLGA-PEG-COOH shows an absorption
band at 3300–3500 cm−1, due to the terminal OH groups, and bands
around 3000 cm−1 and 2860 cm−1, due to C–H stretching of CH3 and
CH2, respectively. The strong signal observed at 1757 cm−1 is assigned
to C–O stretching, the absorption peaks at 1640 and 1556 cm−1 are as-
sociated to C–O and C–N of amide bond, while the band centered at
1190 cm−1 is due to C–O stretch (data not shown).
In F1- and F3-loaded NPs spectra, most of the absorption peaks from
F01 and F03-unloaded NPs and bands at 1650–1580 cm−1 of pure FS o-
verlapped, suggesting that no strong chemical interaction occurred be-
tween FS and polymers.
To investigate the crystal transformation of the nanoparticle system
and to get more information about the interactions between FS and
polymers, a series of powder XRD analyses were performed. Fig. 5A dis-
play the XRD patterns of pure PCL, pure PLGA-PEG-COOH, pure FS, NPs
(F01, F1 and F3), and those of the physical mixtures of PCL-FS and PCL-
PLGA-PEG-COOH-FS.
The diagram of PCL shows two distinctive peaks at angles of 21.81
and 24.17, and two small additional peaks at 16.08 and 30.37° degrees
2θ [31].
The characteristic peaks of pure FS appeared at a diffraction angle of
2θ 11.54°, 12.89°, 14.40°, 15.75°, 16.70°, 17.72°, 18.44°, 19.38°, 20.73°,
21.49°, 22.44°, 23.05°, 23.90°, 24.26°, 24.78°, 26.53°, and 28.71°, thus
suggesting that the FS is present as a highly crystalline structure [32].
Fig. 4. Comparison of FT-IR spectra of pure FS and F01, F1, F03, and F3 NPs, chosen as
examples.
The XRD pattern of PLGA-PEG-COOH copolymer displays two peaks
at angles of 19.06 and 23.01, and two diffuse broad bands in the 2θ re-
gion at 29.87° and 39.95°.
The diagram of empty NPs (F01) resulted similar to the pure PCL,
confirming that the crystalline structure of PCL is retained and not
changed into amorphous phase during the preparation of NPs. On the
other hand, disappearance of characteristic FS peaks in the XRD patterns
of the FS-loaded NPs (F1 and F3) further suggest that FS crystalline na-
ture is converted into amorphous or disordered crystalline form, be-
cause of the formation of an amorphous complex with the
intermolecular interaction between FS and polymers (Fig. 5A and B).
The amorphous state of FS in the polymer matrix is more soluble than
crystalline form, due to free energies involved in the dissolution process,
leading to improved absorption of FS. Furthermore, the absence of crys-
talline peak in loaded NPs further confirms that NPs are free from sur-
face adsorbed FS [33].
Conversely, as also depicted in Fig. 5B, the patterns of physical mix-
tures of PCL-FS (mix 1) and PCL-PLGA-PEG-COOH-FS (mix 2) revealed a
superimposition of the patterns of raw materials with both sharp crys-
talline peaks of FS and peaks of PCL, thus indicating no change in FS crys-
talline state during the mixing process.
3.2. In vitro release studies
Conventional oral administration of flavonoids appears to be ineffi-
cient, based on various issues such as low dissolution rate from solid
dosage forms, partial degradation in the harsh pH conditions of the gas-
tric environment, poor permeability, and extensive first pass metabo-
lism before reaching systemic circulation [34].
Several evidences have revealed that polymeric NPs are suitable car-
riers for the oral administration of bioactive flavonoids because improve
the solubilization and physico-chemical stability, enhance the bioavail-
ability by increasing the absorption from enterocytes, and maintain the
therapeutic levels in blood and plasma, with a significant improvement
in mean residence time as well as bioavailability [35,36].
To evaluate the potential of the prepared NPs for oral administration
of FS, we performed the in vitro release test in simulated gastro-intesti-
nal conditions. As reported in Fig. 6, during the first two hours at acidic
pH the FS released from NPs was b 15% in all cases. This suggests that
most of the FS in the nanoprototypes would be available to be absorbed
within the intestinal tract, and protected from the harsh gastric fluid. No
difference was found in the FS release from all NPs. However, at pH 7.4, a
slight and gradual increase in the FS release was observed in all formu-
lations. In particular, F2 and F3, containing PLGA-PEG-COOH, showed a
similar behavior, with about 70% of FS released after 7 h, and an almost
complete release after 24 h. On the other hand, in the F1 formulation,
based on PCL alone, a lower percentage of FS released was found, with
54% dissolved after 7 h incubation, and about 70% after 24 h. The high
release from F2 and F3 batches can be attributed to the presence of
more hydrophilic PLGA-PEG-COOH, which ease the diffusion of the re-
lease medium into the NPs, compared to F1, containing only hydropho-
bic PCL.
3.3. Antioxidant activity
The diet-derived antioxidant FS is now being increasingly investigat-
ed for its health-promoting effects [2]. Due to specific structural fea-
tures, FS exhibits strong antioxidant properties, mainly due to its
hydroxyl groups at C-3, C-3′, C-4′ and C-7 positions, with the influence
of the carbonyl group at C-4. The presence of double bond between C-2
and C-3 conjugated with the 4-oxo group also facilitates higher electron
delocalization [37].
As far as the radical-scavenging activity of flavonoids is concerned, it
is related to the presence of phenolic hydrogens, which are potentially
able to quench free radicals by forming resonance-stabilized phenoxyl
radicals [38].
 M. Sechi et al. / Materials Science and Engineering C 68 (2016) 594–602 599

Fig. 5. Comparison of XRD diagrams of pure PCL, pure PLGA-PEG-COOH, pure FS, empty NPs (F01), loaded NPs (F1 and F3), physical mixtures of PCL-FS (mix 1) and PCL-PLGA-PEG-COOH-
FS (mix 2) (A). Magnified XRD patterns of pure PCL, pure FS, F1 NPs, and mix 1, chosen as examples, to show detail peaks from 10° to 20° of 2θ (B).

In particular, the DPPH molecules are characterized as a stable free
radicals due to the delocalisation of the spare electron over the molecule
as a whole, that also gives rise to the deep violet colour. Mixing DPPH so-
lution with a substance that can donate a hydrogen atom (like flavo-
noids) leading to the reduced form, with the loss of violet colour [39].
Similarly, the decolorization of the blue-green ABTS radical cation,
generated by reacting a strong oxidizing potassium persulfate with
the ABTS salt, reflects the capacity of an antioxidant species to donate
electrons or hydrogen atoms, which are needed to inactivate this radical
species [40].
The radical scavenging activity of FS before and after encapsulation
was evaluated by DPPH and ABTS free radicals by using F1 NPs, chosen
as model. The comparison on the DPPH scavenging ability of native FS
and FS extracted from NPs (FS-NPs), from 40 to 2.5 μg/mL concentra-
tions, is presented in Fig. 7.
The results clearly show that FS scavenged DPPH radicals in a dose-
dependent manner, also sharing a good linear correlation between the
DPPH scavenging activity and the FS concentration. It is worth noting
that no statistical significance between free FS and FS-NPs was observed
at all tested concentrations, suggesting that FS retained the same DPPH
free radical scavenging efficiency after encapsulation into NPs. Percent-
age inhibition of DPPH ranged from 65% at maximum FS concentration
(40 μg/mL) to about 5% at a lower concentration (2.5 μg/mL), with sim-
ilar IC50 values (30.4 ± 0.35 and 31.2 ± 0.52 μg/mL, P N 0.05) for free FS
and FS-NPs, respectively.
The DPPH• scavenging effect of positive control Trolox at different
concentrations (5–200 μg/mL ) is reported in Fig. 8. The percentage in-
hibition of DPPH ranged from 90%, detected at higher doses of Trolox
to 3.5% (the lower doses), with IC50 values of 112.08 ± 3.4 μg/mL,
which is approximately 3.5 times lower than that of free and
nanoencapsulated FS.
The data derived from the comparison on the ABTS•+ scavenging
ability for free FS and FS-NPs, detected from 30 to 2.5 μg/mL concentra-
tions, are shown in Fig. 9. Similarly to DPPH inhibition, the ABTS scav-
enging effect resulted concentration dependent, with a percentage
inhibition of about 75% at 30 μg/mL FS concentration that linearly de-
creased to 7% for lowest concentration, and without significant differ-
ences between free and encapsulated FS. The calculated IC50 values
were found to be 18.6 ± 0.54 and 20.1 ± 0.24 μg/mL, P N 0.05) for
free FS and FS-NPs, respectively.

Fig. 6. In vitro release profiles of F1, F2, and F3 NPs performed in 0.1 N HCl (pH 1.2) for 2 h, Fig. 7. Comparison of DPPH• scavenging effect of FS before (free FS) and after
followed by PBS (pH 7.4). Data are means ± SD, n = 3. encapsulation (FS-NPs) at different concentrations. Data are means ± SD, n = 3.
600 M. Sechi et al. / Materials Science and Engineering C 68 (2016) 594–602

As presented in Fig. 10, the standard Trolox shows a maximum ABTS

inhibition percentage of about 70%, and a minimum inhibition of 4.5% at
50 and 3.1 μg/mL, respectively.
The calculated IC50 values resulted 36.80 ± 1.6 μg/mL, significantly
lower (about 2-fold) with respect to that of the FS, clearly confirming
that FS is a potent scavenger of free radicals in vitro.
Furthermore, in order to evaluate the influence of polymers on the
DPPH and ABTS assays, further tests were performed using empty NPs
(F01, F02, and F03) at the same concentration of the polymers present
in FS-loaded NPs. At all tested concentrations the absorbance values ob-
served for empty NPs are identical to those of the negative control (dis-
tilled water), indicating that the polymeric materials used did not cause
interference with the antioxidant tests.

3.4. α-Glucosidase inhibition

The α-glucosidase inhibitor Acarbose is widely used in the treatment

of patients with type 2 diabetes. This drug inhibits the upper gastrointes-
tinal glucosidases enzymes that convert complex polysaccharide carbo-
hydrates into monosaccharides in a dose-dependent fashion, resulting
in a delayed glucose absorption and a lowering of postprandial hypergly-
cemia. However, gastrointestinal side effects, including mainly flatulence
and sometimes soft stools or abdominal discomfort, have been reported
as a limiting factor for patients treatment [41,42].
Herein, the α-glucosidase inhibitory activity of free and encapsulat-
ed FS was compared to that of Acarbose, used as positive control.
As reported in Fig. 11A, the highest concentration of Acarbose
(20 mg/mL) showed a maximum inhibition percentage of 95%, while
the lowest concentration of 1.0 mg/mL showed a minimum inhibition
of 46.7%. Furthermore, similar to literature data [43], the half maximal
inhibitory concentration (IC50) resulted of nearly 1.25 mg/mL equiva-
lent to 1.94 mM. On the other hand, FS exerted higher α-glucosidase in-
hibitory activity than Acarbose, with strong inhibition percentages
ranging from 93% to about 40% for the tested concentrations of 250–
15.6 μg/mL (Fig. 11B). Additionally, the IC50 values of FS resulted
0.030 mg/mL, equivalent to 0.1 mM, significantly lower (about 20
fold) with respect to commercial Acarbose. These findings suggest the
potential application of FS as dietary means for postprandial glycaemic
control in type-2 diabetics.
Concerning the mechanism of action of FS in inhibiting α-glucosi-
dase, we assumed that it would behave similarly to other flavonoids
(i.e. quercetin, myricetin), previously identified as putative inhibitors.
The structural determinants involved on the interaction between flavo-
noids and the enzyme have been previously explored using the compu-
tational molecular docking approach [44]. Briefly, from the literature
analysis it emerges that the 3′,4′-dihydroxyl groups of B ring in
Fig. 9. Comparison of ABTS•+ scavenging effect of FS before (free FS) and after
encapsulation (FS-NPs) at different concentrations. Data are means ± SD, n = 3.
flavonoids, structurally related to FS, are crucial in engaging direct bind-
ing within the active-site residues. Additionally, the 3-OH of C ring is
predicted to play a role in maintaining the proper binding orientation
of flavonoid molecules, whereas the hydroxyl groups located on B ring
would tightly interact in a hydrophobic pocket of the enzyme [44].
4. Conclusions
In summary, in this work, the natural flavonoid FS was efficiently en-
capsulated into polymeric NPs by a nanoprecipitation method. NPs pro-
vide a high FS loading capacity and are able to control the FS release in
simulated gastro-intestinal conditions, thus offering the possibility of
the oral administration of nanosystems. Moreover, the results demon-
strated that the scavenging capacity of FS as well as α-glucosidase inhi-
bition activity could be preserved during nanoencapsulation processing.
Furthermore, FS showed a better α-glucosidase inhibition capacity than
Acarbose, supporting the potential of FS for postprandial glycaemic con-
trol. These findings indicate that the developed NPs can be proposed as
an attractive delivery system to control the release of antioxidant and
anti-hyperglycemic FS, with the goal to improve its pharmacological
properties also for prevention and treatment of various diseases.
Acknowledgments
The authors gratefully acknowledge the Regione Autonoma della
Sardegna for the financial support of Grant CRP 25920, awarded to
M.S. within the frame of “Legge regionale n. 7/2007 - Annualità 2010”.

Fig. 8. DPPH• scavenging effect of standard Trolox at different concentrations. Data are Fig. 10. ABTS•+ scavenging effect of Trolox at different concentrations. Data are means ±
means ± SD, n = 3. SD, n = 3.
 M. Sechi et al. / Materials Science and Engineering C 68 (2016) 594–602
Fig. 11. Comparison of α-glucosidase inhibition percentage of Acarbose (A) and FS (B)
before (free FS) and after encapsulation (FS-NPs) at different concentrations. Data are
means ± SD, n = 3.
References
[1] G.R. Beecher, Overview of dietary flavonoids: nomenclature, occurrence and intake,
J. Nutr. 133 (2003) 3248S–3254S.
[2] N. Khan, D.N. Syed, N. Ahmad, H. Mukhtar, Fisetin: a dietary antioxidant for health
promotion, Antioxid. Redox Signal. 19 (2013) 151–162.
[3] H.H. Park, S. Lee, J.M. Oh, M.S. Lee, K.H. Yoon, B.H. Park, J.W. Kim, H. Song, S.H. Kim,
Anti-inflammatory activity of fisetin in human mast cells (HMC-1), Pharmacol. Res.
55 (2007) 31–37.
[4] O.L. Woodman, E.C. Chan, Vascular and anti-oxidant actions of flavonols and fla-
vones, Clin. Exp. Pharmacol. Physiol. 31 (2004) 786–790.
[5] I. Myara, I. Pico, B. Vedie, N. Moatti, A method to screen for the antioxidant effect of
compounds on low-density lipoprotein (LDL): illustration with flavonoids, J.
Pharmacol. Toxicol. Methods 30 (1993) 69–73.
[6] P. Maher, T. Akaishi, K. Abe, Flavonoid fisetin promotes ERK-dependent long-term
potentiation and enhances memory, Proc. Natl. Acad. Sci. U. S. A. 103 (2006)
16568–16573.
[7] N. Khan, M. Asim, F. Afaq, Z.M. Abu, H. Mukhtar, A novel dietary flavonoid fisetin in-
hibits androgen receptor signaling and tumor growth in athymic nude mice, Cancer
Res. 68 (2008) 8555–8563.
[8] I. Murtaza, V.M. Adhami, B.B. Hafeez, M. Saleem, H. Mukhtar, Fisetin, a natural
flavonoid, targets chemoresistant human pancreatic cancer AsPC-1 cells
through DR3-mediated inhibition of NF-kappa B, Int. J. Cancer 125 (2009)
2465–2473.
[9] Y. Cho, J.H. Chung, H.J. Do, H.J. Jeon, T. Jin, M.J. Shin, Effects of fisetin supplementa-
tion on hepatic lipogenesis and glucose metabolism in Sprague-Dawley rats fed on
a high fat diet, Food Chem. 139 (2013) 20–727.
[10] R.P. Constantin, J. Constantin, C.L. Pagadigorria, E.L. Ishii-Iwamoto, A. Bracht, K. Ono
Mde, N.S. Yamamoto, The actions of fisetin on glucose metabolism in the rat liver,
Cell Biochem. Funct. 28 (2010) 149–158.
[11] G.S. Prasath, S.P. Subramanian, Modulatory effects of fisetin, a bioflavonoid, on hy-
perglycemia by attenuating the key enzymes of carbohydrate metabolism in hepatic
and renal tissues in streptozotocin-induced diabetic rats, Eur. J. Pharmacol. 668
(2011) 492–496.
601
[12] G.S. Prasath, S.I. Pillai, S.P. Subramanian, Fisetin improves glucose homeostasis
through the inhibition of gluconeogenic enzymes in hepatic tissues of
streptozotocin induced diabetic rats, Eur. J. Pharmacol. 740 (2014) 248–254.
[13] C.S. Shia, S.Y. Tsai, S.C. Kuo, Y.C. Hou, P.D. Chao, Metabolism and pharmacokinetics of
3,3′,4′,7-tetrahydroxyflavone (fisetin), 5-hydroxyflavone, and 7-hydroxyflavone
and antihemolysis effects of fisetin and its serum metabolites, J. Agric. Food Chem.
57 (2009) 83–89.
[14] N. Mignet, J. Seguin, G.G. Chabot, Bioavailability of polyphenol liposomes: a chal-
lenge ahead, Pharmaceutics 5 (2013) 457–471.
[15] H. Ragelle, S. Crauste-Manciet, J. Seguin, D. Brossard, D. Scherman, P. Arnaud, G.G.
Chabot, Nanoemulsion formulation of fisetin improves bioavailability and
antitumour activity in mice, Int. J. Pharm. 427 (2012) 452–459.
[16] N. Mignet, J. Seguin, M. Ramos Romano, L. Brullé, Y.S. Touil, D. Scherman, M.
Bessodes, G.G. Chabot, Development of a liposomal formulation of the natural flavo-
noid fisetin, Int. J. Pharm. 423 (2012) 69–76.
[17] J. Seguin, L. Brullé, R. Boyer, Y.M. Lu, M. Ramos Romano, Y.S. Touil, D. Scherman, M.
Bessodes, N. Mignet, G.G. Chabot, Liposomal encapsulation of the natural flavonoid
fisetin improves bioavailability and antitumor efficacy, Int. J. Pharm. 444 (2013)
146–154.
[18] Y. Chen, Q. Wu, L. Song, T. He, Y. Li, L. Li, W. Su, L. Liu, Z. Qian, C. Gong, Polymeric mi-
celles encapsulating fisetin improve the therapeutic effect in colon cancer, ACS Appl.
Mater. Interfaces 7 (2015) 534–542.
[19] V. Sanna, I.A. Siddiqui, M. Sechi, H. Mukhtar, Resveratrol-loaded nanoparticles
based on poly(epsilon-caprolactone) and poly(D , L -lactic-co-glycolic acid)-
poly(ethylene glycol) blend for prostate cancer treatment, Mol. Pharm. 10
(2013) 3871–3881.
[20] I. Gülçin, Antioxidant properties of resveratrol: a structure-activity insight, Innova-
tive Food Sci. Emerg. Technol. 11 (2010) 210–218.
[21] V. Sanna, N. Pala, G. Dessì, P. Manconi, A. Mariani, S. Dedola, M. Rassu, C. Crosio, C.
Iaccarino, M. Sechi, Single-step green synthesis and characterization of gold-conju-
gated polyphenol nanoparticles with antioxidant and biological activities, Int. J.
Nanomedicine 9 (2014) 4935–4951.
[22] V. Sanna, G. Lubinu, P. Madau, N. Pala, S. Nurra, A. Mariani, M. Sechi, Polymeric nano-
particles encapsulating white tea extract for nutraceutical application, J. Agric. Food
Chem. 63 (2015) 2026–2032.
[23] H. Zhao, W. Chen, J. Lu, M. Zhao, Phenolic profiles and antioxidant activities of com-
mercial beers, Food Chem. 119 (2010) 1150–1158.
[24] V. Sanna, L. Pretti, Effect of wine barrel ageing or sapa addition on total polyphenol
content and antioxidant activities of some Italian craft beers, Int. J. Food Sci. Technol.
50 (2015) 700–707.
[25] I. Gülçin, R. Elias, A. Gepdiremen, L. Boyer, Antioxidant activity of lignans from
fringe tree (Chionanthus virginicus L.), Eur. Food Res. Technol. 223 (2006)
759–767.
[26] R. Subramanian, M.Z. Asmawi, A. Sadikun, In vitro alpha-glucosidase and alpha-am-
ylase enzyme inhibitory effects of Andrographis paniculata extract and
andrographolide, Acta Biochim. Pol. 55 (2008) 391–398.
[27] E. Marin, M.I. Briceño, C. Caballero-George, Critical evaluation of biodegradable poly-
mers used in nanodrugs, Int. J. Nanomedicine 8 (2013) 3071–3091.
[28] J.V. Jokerst, T. Lobovkina, R.N. Zare, S.S. Gambhir, Nanoparticle PEGylation for imag-
ing and therapy, Nanomedicine (London) 6 (2011) 715–728.
[29] J.M. Dimitrić Marković, Z.S. Marković, D. Milenković, S. Jeremić, Application of com-
parative vibrational spectroscopic and mechanistic studies in analysis of fisetin
structure, Spectrochim. Acta A Mol. Biomol. Spectrosc. 83 (2011) 120–129.
[30] L. Azouz, F. Dahmoune, F. Rezgui, C. G'Sell, Full factorial design optimization of anti-
inflammatory drug release by PCL-PEG-PCL microspheres, Mater. Sci. Eng. C. Mater.
Biol. Appl. 58 (2016) 412–419.
[31] H. Ge, Y. Hu, S. Yang, X. Jiang, C. Yang, Preparation, characterization, and drug release
behaviors of drug-loaded ε-caprolactone/L-lactide copolymer nanoparticles, J. Appl.
Polym. Sci. 75 (2000) 874–882.
[32] J.Q. Zhang, K.M. Jiang, K. An, S.H. Ren, X.G. Xie, Y. Jin, J. Lin, Novel water-soluble
fisetin/cyclodextrins inclusion complexes: preparation, characterization, molecular
docking and bioavailability, Carbohydr. Res. 418 (2015) 20–28.
[33] J.B. Mendes, M.K. Riekes, V.M. de Oliveira, M.D. Michel, H.K. Stulzer, N.M. Khalil, S.F.
Zawadzki, R.M. Mainardes, P.V. Farago, PHBV/PCL microparticles for controlled re-
lease of resveratrol: physicochemical characterization, antioxidant potential, and ef-
fect on hemolysis of human erythrocytes, Sci. World J. 542937 (2012).
[34] D.V. Ratnam, D.D. Ankola, V. Bhardwaj, D.K. Sahana, M.N.V. Ravi Kumar, Role of an-
tioxidants in prophylaxis and therapy: a pharmaceutical perspective, J. Control. Re-
lease 113 (2006) 189–207.
[35] G. Leonarduzzi, G. Testa, B. Sottero, P. Gamba, G. Poli, Design and development of
nanovehicle-based delivery systems for preventive or therapeutic supplementation
with flavonoids, Curr. Med. Chem. 17 (2010) 74–95.
[36] A. Bilia, B. Isacchi, C. Righeschi, C. Guccione, M. Bergonzi, Flavonoids loaded in
nanocarriers: an opportunity to increase oral bioavailability and bioefficacy, Food
Nutr. Sci. 5 (2014) 1212–1327.
[37] Z.S. Marković, S.V. Mentus, J.M. Dimitrić Marković, Electrochemical and density
functional theory study on the reactivity of fisetin and its radicals: implications on
in vitro antioxidant activity, J. Phys. Chem. A 113 (2009) 14170–14179.
[38] S. Dudonné, X. Vitrac, P. Coutière, M. Woillez, J.M. Mérillon, Comparative study of
antioxidant properties and total phenolic content of 30 plant extracts of industrial
interest using DPPH, ABTS, FRAP, SOD, and ORAC assays, J. Agric. Food Chem. 57
(2009) 1768–1774.
[39] C.A. Rice-Evans, N.J. Miller, G. Paganga, Structure-antioxidant activity relationships
of flavonoids and phenolic acids, Free Radic. Biol. Med. 20 (1996) 933–956.
[40] M.N. Alam, N.J. Bristi, M. Rafiquzzaman, Review on in vivo and in vitro methods
evaluation of antioxidant activity, Saudi Pharm. J. 21 (2013) 143–152.
602 M. Sechi et al. / Materials Science and Engineering C 68 (2016) 594–602

[41] H. Laube, Acarbose, Clin. Drug Investig. 22 (2002) 141–156.
[42] C. Rosak, G. Mertes, Critical evaluation of the role of acarbose in the treatment of di-
abetes: patient considerations, Diabetes Metab. Syndr. Obes. 5 (2012) 357–367.
[43] E.A. Mohamed, M.J. Siddiqui, L.F. Ang, A. Sadikun, S.H. Chan, S.C. Tan, M.Z. Asmawi,
M.F. Yam, Potent α-glucosidase and α-amylase inhibitory activities of standardized
50% ethanolic extracts and sinensetin from Orthosiphon stamineus Benth as anti-di-
abetic mechanism, BMC Complement. Altern. Med. 12 (2012) 176.
[44] H. Xu, Inhibition kinetics of flavonoids on yeast α-glucosidase merged with docking
simulations, Protein Pept. Lett. 17 (2010) 1270–1279.