In vivo pharmacokinetics of hesperidin are affected by treatment with glucosidase-like BglA protein isolated from yeasts

Article abstract of DOI:10.1021/jf800105c

Hesperidin is an abundant flavanone glycoside in citrus fruits and has been reported to possess a wide range of biological activities. However, hesperidin has poor bioavailability. Here, we tested the hypothesis that hesperetin found in chenpi will have a better bioavailability than hesperidin and that treatment of hesperidin with the glucosidase-like yeast Bg1A protein will increase its bioavailability. The results indicate that hesperidin in pure or extract form is hydrolyzed by BglA protein extracted from Sporobolomyces singularis or expressed in Escherichia coli BL21 (DE3). This biotransformation affected the plasma pharmacokinetics of total hesperetin in rats, in that the plasma T _max was significantly shorter after administration of BglA protein-treated hesperidin than after administration of hesperidin extract. In addition, the area under the curve values for total hesperetin after administration of Bg1A-treated hesperidin were ～4-fold higher by oral administration and 3-fold higher by intravenous administration, respectively. In contrast, the plasma clearance value and volume of distribution after administration of Bg1A-treated hesperidin extract or pure hesperetin were significantly smaller than after administration of untreated hesperidin extract or pure hesperidin. This is the first study that systemically determines the absolute bioavailability of hesperidin and hesperetin simultaneously, shows clearly that hesperetin is more bioavailable than hesperidin regardless of the route of administration, and shows that prior transformation of hesperidin to hesperetin via fermentation should significantly increase its bioavailability because of the action of the yeast glycosidase-like protein BglA.

Full text of DOI:10.1021/jf800105c

5550 J. Agric. Food Chem. 2008, 56, 5550–5557
In Vivo Pharmacokinetics of Hesperidin Are Affected
by Treatment with Glucosidase-like BglA Protein
Isolated from Yeasts
YONG-MEI LI,† XIAO-MIAN LI,† GUANG-MING LI,† WEN-CAI DU,† JING ZHANG,†
WEI-XIA LI,† JIANSHE XU,† MING HU,‡ AND ZE ZHU*,†
Department of Medical Microbiology, Tianjin Medical University, Tianjin 300070,
People’s Republic of China, and Department of Pharmacological and Pharmaceutical Sciences,
College of Pharmacy, University of Houston, Houston, Texas 77030
Hesperidin is an abundant flavanone glycoside in citrus fruits and has been reported to possess a
wide range of biological activities. However, hesperidin has poor bioavailability. Here, we tested the
hypothesis that hesperetin found in chenpi will have a better bioavailability than hesperidin and that
treatment of hesperidin with the glucosidase-like yeast Bg1A protein will increase its bioavailability.
The results indicate that hesperidin in pure or extract form is hydrolyzed by BglA protein extracted
from Sporobolomyces singularis or expressed in Escherichia coli BL21 (DE3). This biotransformation
affected the plasma pharmacokinetics of total hesperetin in rats, in that the plasma Tmax was
significantly shorter after administration of BglA protein-treated hesperidin than after administration
of hesperidin extract. In addition, the area under the curve values for total hesperetin after
administration of Bg1A-treated hesperidin were ∼4-fold higher by oral administration and 3-fold higher
by intravenous administration, respectively. In contrast, the plasma clearance value and volume of
distribution after administration of Bg1A-treated hesperidin extract or pure hesperetin were significantly
smaller than after administration of untreated hesperidin extract or pure hesperidin. This is the first
study that systemically determines the absolute bioavailability of hesperidin and hesperetin
simultaneously, shows clearly that hesperetin is more bioavailable than hesperidin regardless of the
route of administration, and shows that prior transformation of hesperidin to hesperetin via fermentation
should significantly increase its bioavailability because of the action of the yeast glycosidase-like
protein BglA.
KEYWORDS: Hesperidin; chenpi; biotransformation; glycosidase; pharmacokinetics; bioavailability
INTRODUCTION
in human intestinal microflora that are capable of hydrolyzing
hesperidin (19, 20).
Hesperidin is an abundant flavanone glycoside found in citrus
fruits and is often consumed in the form of orange juice (1).
Hesperidin has been reported to possess anti-inflammatory,
antioxidant, and regulation activities of hepatic cholesterol
synthesis (2–4). Other reported activities of these flavonoids
include lowering cancer risk and inhibiting bone resorption (5–7).
However, hesperidin, like many other flavonoids, has a poor
bioavailability (8–12). The reasons for this poor bioavailability
include the low water solubility of hesperidin and disposition
via phase II enzymes (13, 14). Additionally, glycosides of
hesperetin have been shown to be effluxed in Caco-2 cells (15),
consistent with observations that many flavonoid glycosides are
effluxed by enterocytes (16–18). The bioavailability is also
highly variable, perhaps because of interindividual differences
In Western countries, hesperidin represents a major flavonoid
consumed daily, and some estimates suggest that it could even
be responsible for up to 50% of the daily dietary intake of
flavonoids in a few developed countries (5). Hesperidin is a
major active ingredient in the Chinese traditional medicinal herb
called chenpi, which is made from tangerine peel and has been
used for more than 10 centuries. Chenpi also is used in the
traditional medicines of other Asian countries such as Japan,
where it is called chinpi (13). Chenpi has expectorant activities
and a broncho-dilative effect and is usually used to treat upper
respiratory tract infections using traditional Chinese medicinal
formulae. Chenpi is prescribed as an herbal medicine for
inflammation, allergies, and hepatopathy in Japan (13). Chenpi
is prepared using a well-documented and ancient nine-step
fermentation procedure because chenpi so produced appears to
have a greater bioactivity than air- or sun-dried tangerine peel.
In preliminary studies, we isolated a yeast strain from chenpi
* Corresponding author. Tel.: (86)-22-23542649; e-mail: zhuze@
tijmu.edu.cn.
† Tianjin Medical University.
‡ University of Houston.
10.1021/jf800105c CCC: $40.75  2008 American Chemical Society
Published on Web 06/21/2008

Pharmacokinetics of Hesperidin
J. Agric. Food Chem., Vol. 56, No. 14, 2008 5551
cell extracts and then centrifuged (10 000g, 30 min). The pellet was
resuspended in 50 mM potassium phosphate buffer (pH 7.0) and loaded
onto a DEAE sepharose CL-6B column (50 mm × 200 mm,
Pharmacia). Further purification was performed by gel filtration on a
sephacryl S-300 HR column (15 mm × 900 mm, Pharmacia). The
eluent was 50 mM sodium phosphate buffer containing 0.1 M NaCl
(pH 7.2). The ꢀ-glucosidase activity of the eluted fractions was
measured as described later.
Figure 1. Structure of hesperidin and proposed site of hydrolysis by the
BglA protein.
Cloning, Expression, and Purification of BglA Gene in E. coli
BL21 (DE3). S. singularis was harvested, and the total RNA was
extracted with TRIzol reagent (Invitrogen). PCR was performed using
the following procedure: initial denaturation, 94 °C for 2 min; 30 cycles
at 94 °C for 30 s, 58 °C for 30 s, and 72 °C for 2 min; and final
extension at 72 °C for 5 min. PCR products were resolved by 1.5%
agarose gel electrophoresis, stained with ethidium bromide, and
visualized with UV light. A pair of primers (5′-CGC GGATCC
ATGATGCTGCATGCGGCAC-3′ (forward primer) and 5′-CCC
AAGCTT TCAGAGGTGGTTGCGACC-3’′ (backward primer)) were
designed and synthesized with BamH I and Hind III restriction sites
added to the primers (underlined). The PCR product was digested with
BamH I and Hind III. The digested fragments were purified by gel
extraction and then ligated to the pQE30-Xa vector (Qiagen) to generate
the recombinant expression vector, pQE30-Xa-BglA. The ligated
sequence was then confirmed by sequencing (not shown).
Subsequently, pQE30-Xa-BglA was used to transfect the E. coli BL21
(DE3) strain. Briefly, a single transfected colony was picked from a
selective plate and inoculated into LB medium supplemented with
ampicillin (200 µg/mL). When the OD600 of the cultured media reached
0.6, isopropyl-beta-D-thiogalactopyranoside (IPTG) was added to the
media at a final media concentration of 1 mM. After an additional 4 h
of incubation, a portion of the bacteria was suspended in 200 µL of
the sample buffer and boiled for 5 min at 90 °C. The expression of
BglA was detected by SDS-PAGE, carried out according to the method
of Laemmli (25). Proteins were stained with the Bio-Safe Coomassie
G250 stain (Amresco) according to manufacturer’s instructions. The
remaining bacteria were harvested and washed twice with 50 mM
potassium phosphate buffer (pH 7.0) to pellet the bacteria using
centrifugation (4000g, 20 min).
To purify the BglA protein form the bacteria, each gram of cell pellet
was resuspended in 3 mL of TE buffer. Following the addition of
lysozyme to a final concentration of 0.2 mg/mL, the cell suspension
was incubated for 20 min at 37 °C with gentle shaking. Then cell
extracts were obtained by centrifugation (12 000g, 30 min), and the
ꢀ-glucosidase activity of the crude BglA protein was measured as
described next. An additional purification step, as described previously
using various columns, was needed to further purify the BglA protein
from bacteria to higher specific activities.
Assay for ꢀ-Glucosidase Activity of BglA Protein. The ꢀ-glucosi-
dase activity was measured using PNP-Glc as the substrate (24). The
incubation mixture (total volume of 1 mL) contained 1.5 mM PNP-
Glc, 50 mM phosphate buffer (pH 6.0), and a 0.1 mL solution of crude
or purified enzyme preparation. The reaction was carried out at 37 °C
for 4 h and then terminated by adding 4 mL of 0.25 M Na2CO3. The
absorbance at 420 nm, subtracting the blank, was used as a measure of
the amount of p-nitrophenol (PNP) released. One unit of ꢀ-glucosidase
activity was defined as the amount of enzyme that hydrolyzes 1 µmol
of PNP-Glc per minute under the assay conditions.
and identified it to be a yeast strain closely related to Sporobo-
lomyces singularis. This strain has glycosidase activities that
are similar to other yeast strains identified previously to produce
a glycosidase-like BglA protein (21, 22).
We were interested in determining the presence of BglA in
S. singularis because this glucosidase-like protein, an enzyme
isoform belonging to family 1 glycosyl hydrolases, can hydro-
lyze hesperidin abundant in citrus peels to hesperetin. Since
hesperetin is absorbed much more quickly than hesperidin (15),
we hypothesized that ancient Chinese fermentation methods
increase the bioavailability of chenpi by removing the glycosides
and thereby increasing the bioavailability of hesperidin (Figure
1). Testing this hypothesis allowed us to provide a mechanistic
explanation as to why fermentation-prepared chenpi may have
higher activities, which until now has been unclear. To test this
hypothesis, we biotransformed hesperidin extract with BglA
protein extracted from S. singularis or expressed in Escherichia
coli BL21 (DE3) in vitro. We also determined how the BglA
protein influenced the plasma pharmacokinetics of total hes-
peretin in rats after administration of BglA protein-treated or
untreated hesperidin extracts. For comparison purposes, phar-
macokinetic studies also were conducted using pure hesperetin
and its glycoside hesperidin.
MATERIALS AND METHODS
Materials. Chenpi was obtained from a local drugstore. Standard
hesperidin was purchased from the National Institute for the Control
of Pharmaceutical and Biological Products (Beijing, China). Standard
hesperetin, ꢀ-glucuronidase type VII-A, H5,ꢀ-glucosidase, and p-ni-
trophenyl-ꢀ-D-glucopyranoside (PNP-Glc) were purchased from Sigma
Chemical Co. (St. Louis, MO). All other chemicals were of analytical
or HPLC grade. S. singularis (ATCC 24193) was obtained from ATCC
(Manassas, VA) and cultured in YM medium containing 3 g of yeast
extract, 3 g of malt extract, 5 g of peptone, 10 g of dextrose, and 20 g
of agar in 1.0 L of distilled water (pH 5.0) and was maintained
aerobically at 24 °C.
Extraction of Hesperidin from Chenpi. The hesperidin was
extracted from chenpi with an ultrasonic method adapted from a
published report (23). Briefly, 4 g of chenpi was ground to 0.5 to ∼1
mm, blended with 200 mL of methanol, and placed in an ultrasonic
machine (KQ-100A, Kunming, China) running at 60 mHz for 60 min
at 40 °C. After initial extraction, the solvent was recovered and then
replaced with fresh methanol, and the process was then repeated. At
the end of the second extraction procedure, methanol was then separated
from the chenpi and then combined with the first methanol extract
solution. The combined extract solution was filtered, and the filtrate
was evaporated to dryness. The residues were again dissolved in
medium at a maximum of 5% (v/v) propylene glycol with a concentra-
tion of 10 mg/mL. Then, the solutions were acidified with 0.1 mL of
0.01 M oxalic acid and quantified by HPLC.
In Vitro Biotransformation of Hesperidin by BglA Protein. Each
reaction mixture (total of 1 mL) containing 0.5 µM hesperidin, 50 mM
phosphate citrate buffer (pH 6.0), and 0.01 mg of enzyme equivalent
of BglA protein was either extracted from S. singularis or expressed
in E. coli BL21 (DE3). The mixture without enzyme was preincubated
at 40 °C for 5 min, and the reaction was started by the addition of
enzyme. After incubation at 40 °C for 30 min, the reaction was stopped
by boiling for 5 min. Another two mixtures were incubated under the
same conditions, which were the positive control of ꢀ-glucosidase and
the mock control. Hhesperidin and hesperetin in the reaction mixture
were analyzed by HPLC. Samples removed from the reaction mixture
for HPLC application were centrifuged (3000g, 20 min) and extracted
with 500 µL of methanol from the supernatant. The organic layer was
transferred to a microcentrifuge tube and dried under vacuum. The
Extraction and Purification of BglA Protein from S. singularis.
The S. singularis yeast cells in midexponential phase were harvested
and centrifuged (2000g, 5 min). The pellet was washed twice with 50
mM sodium phosphate buffer (pH 7.0), and cells were resuspended in
50 mM phosphate citrate buffer (pH 4.0) and treated with 1% (w/v)
Usukizyme (Kyowa, Japan) for 2 h at 37 °C. The soluble fraction of
lysate was mixed with the supernatant prepared by homogenization to
give a crude extract. The purification steps were carried out according
to previous reports (24). Briefly, ammonium sulfate was added to the

5552 J. Agric. Food Chem., Vol. 56, No. 14, 2008
Li et al.
Figure 2. Metabolism of hesperidin by BglA protein and ꢀ-glucosidase. The experiments were performed using 0.5 µmol/L of hesperidin and 0.01 mg
of enzyme equivalent prepared from yeast or expressed in E. coli (DE3) using procedures described in the Materials and Methods.
Figure 4. SDS-PAGE of BglA protein extracted from E. coli BL21 (DE3)
and undergoing various steps of purification. Lane a: bacteria cell extract;
lane b: molecular mass standards; lane c: (NH4)2SO4 precipitate (55-60%
saturation); lane d: pooled DEAE sepharose fractions; and lane e: pooled
sephacryl fractions. The narrow band indicates the BglA protein.
Table 1. Purification of BglA Protein Expressed in E. coli BL21 (DE3)
total activity total protein specific activity purification recovery
(U/mg protein)a (-fold)
step
(U)
(mg)
(%)
crude extract
216.0
49.2
25.6
16.2
772.0
19.6
8.5
0.3
2.5
3.0
5.8
1.0
8.4
10.0
19.3
100.0
22.8
11.9
7.5
ammonium sulfate
DEAE sepharose
sephacryl
2.8
a One unit of ꢀ-glucosidase activity was defined as the amount of enzyme that
produced 1 µmol of PNP-Glc per min under assay conditions.
system, Shimadzu LC-6 with variable wavelength detector; column,
YWG-C18 (4.6 mm × 150 mm, 5 µm, Shimadzu); column temper-
ature, 30 °C; injection volume, 20 µL; detector wavelength, 283
nm; flow rate, 1 mL/min; and mobile phase, 40:60 (v/v) mixture of
methanol and 0.04 M KH2PO4 (pH adjusted to 4.0 using H3PO4).
Quantification of hesperidin and hesperetin peaks from all matrices
was based on the standard curves. For the analysis of hesperetin
and hesperidin, standard solutions containing 0.00, 0.002, 0.005,
0.02, 0.05, 0.2, 0.5, and 1 µmol of added hesperetin/liter or
hesperidin/liter were prepared, respectively. The standard curve was
obtained by plotting the peak height of standards versus hesperetin
and hesperidin concentrations. Day-to-day variation (CV %) for
hesperetin and hesperidin from all matrices was 11%, and within-
day variation for them was 7%. The limits of detection for hesperetin
and hesperidin from plasma were 1.5 and 3.5 nmol/L, respectively.
The linear response ranges of the method for hesperetin and
hesperetin were 1.5 to ∼1000 and 3.5 to ∼1000 nmol/mL,
respectively. The equations for linear responses, described by the
relationship between peak height (y) and concentration (x), were y
Figure 3. Representative HPLC chromatograms of (A) standards of
hesperetin and hesperidin, (B) blank rat plasma, and (C) 0.02 µmol/L of
hesperetin and 0.02 µmol/L of hesperidin in rat plasma.
residue was reconstituted with 100 µL of methanol and centrifuged
(3000g, 10 min). Then, the supernatant was transferred for HPLC
analysis. Experiments were carried out in triplicate for each
experiment.
HPLC Assays of Hesperidin and Hesperetin. The concentrations
of hesperidin and hesperetin were analyzed by HPLC as follows:

Pharmacokinetics of Hesperidin
J. Agric. Food Chem., Vol. 56, No. 14, 2008 5553
centrifugation (3000g, 10 min). The supernatant layer was transferred
to a microcentrifuge tube and vacuum-dried. The residue was recon-
stituted with 100 µL of methanol, vortexed, and then centrifuged (3000g,
10 min) to produce samples for HPLC analysis.
Calculation of Pharmacokinetic Indexes and Statistical Analysis.
Pharmacokinetic analysis of the data was performed using WinNonlin
(Pharsight) (version 5.2 Build 200701231637). The pharmacokinetic
parameters were derived using noncompartmental analysis and a linear
trapezoidal (Linear Interpolation) method, and other pharmacokinetic
parameters were determined using standard equalizers. Each rat plasma
concentration versus time course was analyzed individually. The
statistical analysis of the pharmacokinetic parameters was achieved
using one-way ANOVA with post hoc analysis. A statistically
significant difference was set at 5% or p > 0.05.
RESULTS
Hydrolysis Hesperidin by BglA Protein in Vitro. BglA
protein, whether isolated from S. singularis or expressed in E.
coli BL21 (DE3), was capable of hydrolyzing hesperidin
exclusively to hesperetin (Figure 2), as hesperetin was not
detected in a mock mixture free of it. In BglA protein-treated
hesperidin extracts, the hesperetin concentration was nearly
similar to that treated with commercial ꢀ-glucosidase (p > 0.05)
(not shown). In addition, the source of the BglA protein did
not affect their activities as two different sources of BglA protein
had nearly identical activities (not shown). Lastly, the identity
of hesperidin was verified by being compared to an authentic
standard using HPLC (as shown in Figure 3) since HPLC
clearly separated hesperetin (tR ) 10.77 min) from hesperidin
(tR ) 20.38 min).
Identification of BglA expressed in E. coli. SDS-PAGE
analysis of the cell extract of E. coli BL21 (DE3) transfected
with the BglA gene revealed a predominant protein band of
∼66 kDa (Figure 4), which was in close agreement to the
expected molecular mass of 65.67 kDa deduced from the amino
acid sequence of the BglA gene. This predominant protein band
was not detected in the cell extract from E. coli BL21 (DE3)
not transfected with BglA plasmid. Lastly, the BglA protein
activity in transfected E. coli BL21 (DE3) was approximately
100-fold higher than that in cell extracts from S. singularis (0.3
U mg of protein-1 vs 0.0027 U mg of protein-1).
Figure 5. Changes in plasma concentration of total hesperetin in rats
following IV administration. The rats were divided into five groups, and
each rat was administered intravenously a 31 µmol/kg equivalent amount
of hesperidin. Plasma samples were treated with ꢀ-glucuronidase, and
therefore, the plasma levels represent the total hesperetin equivalent.
Values were presented as mean ( SEM (n ) 5).
) 0.1676x - 0.0092 (r2 ) 0.998) for hesperidin and y ) 0.1083x
- 0.0065 (r2 ) 0.998) for hesperetin, respectively. Recoveries of
these compounds by this method were <95%.
Animals and Study Design. Male Wistar rats (6 weeks old, 180 to
∼220 g, Beijing Administration Office of Laboratory Animal, Beijing,
China; certification number SCXK-Military-2002) were housed in an
air-conditioned room (23 ( 2 °C) under a 12 h light-dark cycle for a
1 week period of acclimatization. Rats fasted overnight before the
experiments. The animals were handled according to the Guidelines
for the Care and Use of Laboratory Animals, and the experiments were
conducted at Tianjin Medical University.
Before the experiment started, rats were weighed and divided by
weight in 10 groups as follows: in test group 1 (test 1), rats were
gavaged with purified hesperidin; in test 2, rats were gavaged with
purified hesperetin; in test 3, rats were gavaged with the untreated
hesperidin extracts; in test 4, rats were gavaged with hesperidin extract
treated with expressed BglA protein; and in test 5, rats were gavaged
with hesperidin treated with BglA protein isolated from S. singularis.
The other five tests (5 rats per test) were carried out using hesperidin
or hesperetin administered intravenously with the chemicals in the same
order as stated previously. For both oral and IV administration, 18.9
mg/kg of hesperidin or 9.4 mg/kg of hesperetin was administered as
an aqueous isotonic solution at a maximum of 5% (v/v) propylene
glycol. The dose of flavonoids given was based on recommended
amounts of chenpi given to people by the Chinese Pharmacopoeia. In
cases of oral administration, blood samples were withdrawn from the
jugular artery cannula before and at 0, 0.5, 1, 2, 3, 4, 6, 8, 10, 12, 14,
and 24 h after flavonoid administration. Before the flavonoid solution
can be used intravenously, the solution is processed to minimize
potential hazards to the rats. First, the flavonoid solution was carefully
chilled (4 °C), filtered, and centrifuged (10 000g, 2 min) to remove
any particulate matter. After the solution was shown to be nonhemolytic,
it was autoclaved at 121 °C, cooled, and then used for intravenous
administration. Similar to oral administration, the plasma samples were
collected before intravenous administration and at 0, 0.5, 1, 2, 4, 6, 8,
12, and 24 h after IV administration. Plasma samples were prepared
by centrifugation (10 000g, 2 min), and the plasma was removed and
stored at -20 °C within 30 min after a blood sample was taken.
The plasma samples after administration were used for the quanti-
fication of total hesperetin, after the metabolites were released by
hydrolysis with ꢀ-glucuronidase. Plasma samples (50 µL) were
incubated with appropriate amounts of ꢀ-glucuronidase type H-5 (500
units/mL) in 0.1 mM sodium acetate buffer (pH 5.0) at 37 °C for 90
min. To stop the reaction of enzyme, 0.01 M oxalic acid (100 µL) was
added to the reaction mixture. Then, the mixture was extracted with
500 µL of methanol by vortexing the mixture for 2 min and
The BglA protein was enriched as described previously (in the
Materials and Methods), and a summary of the purification
procedure of the BglA protein from transfected E. coli BL21 (DE3)
along with each step’s yield is presented in Table 1. The
heterologously expressed BglA protein was purified by ammonium
sulfate fractionation and column chromatography with an (overall)
19.3-fold enrichment, and the recovery yield was 7.5%.
Effects of BglA Protein Treatment on in Vivo Pharma-
cokinetics of Total Hesperetin in Rats after IV Administra-
tion. We first conducted parallel IV pharmacokinetic studies
of hesperetin and hesperidin because it had not been performed
in the same laboratory. Previously, pharmacokinetic studies of
hesperetin or hesperidin were conducted separately, and the
major chemical species found in human plasma were glucu-
ronide conjugates, while the sulfates or sulfate and glucuronide
conjugates were only detected in minor quantities (8, 11). This
was confirmed by another study in which free hesperetin or
hesperidin was not detected in the rat plasma after administration
of hesperidin not treated with ꢀ-glucuronidase (13). On the basis
of this literature and the fact that hesperetin glucuronides are
not commercially available, plasma samples from the present
study also were treated with ꢀ-glucuronidase to release aglycone
hesperetin. Our experiment could not quantify the hesperetin
conjugates due to lack of standards; therefore, we termed

5554 J. Agric. Food Chem., Vol. 56, No. 14, 2008
Li et al.
Table 2. Pharmacokinetic Parameters of Total Hesperetin after Single IV Administration of Hesperidin or Hesperetin or Hesperidin Treated with Different
BlgA Protein Treatments
hesperidin extract
untreated
hesperidin extract +
BlgA from E. coli
hesperidin extract +
BlgA from yeast
parameters
pure hesperidin
pure hesperetin
AUCobsd av (nmol h/mL) SD
30.09
12.88
6.28
61.47a
22.20
6.52
18.01
10.86
6.25
59.40b
14.68
6.05
57.90b
11.73
6.15
half-life av (h) SD
Vz av (mL) SD
0.28
0.61
0.81
1.01
0.32
10.73
4.34
1.17
5.40a
2.42
19.27
7.44
2.10
4.70b
0.47
4.88b
0.82
CL av (mL/h) SD
0.56a
0.22
0.54b
0.11
0.55b
0.09
0.43
0.81
a There are statistically significant differences between AUCobsd for pure hesperidin and pure hesperetin (p < 0.05). b There are statistically significant differences
between AUCobsd, Vz, and CL for hesperidin with different BlgA protein treatment and untreated hesperidin extract (p > 0.05).
Effects of BglA Protein Treatment on in Vivo Pharma-
cokinetics of Total Hesperetin in Rats after Oral Adminis-
tration. Hesperetin or hesperidin was absorbed by rats from
all treatment groups. The corresponding times to reach maxi-
mum plasma concentration (Tmax) in rats were significantly
shorter after administration of BglA protein-treated hesperidin
and the standard hesperetin than those in the other two groups
(Figure 6). After dosing with the standard hesperidin and
hesperidin extract, total hesperetin was detected at 2 h and
reached a maximum (∼0.35 and 0.20 µM) at 6 h (Table 3). On
the other hand, after the administration of BglA protein-treated
hesperidin and standard hesperetin, total hesperetin was detected
as early as 30 min after administration and reached a maximum
concentration (∼0.82, 0.74, and 0.95 µM) at 4 h (Table 3).
The results show that plasma Cmax and Tmax levels were similar
in hesperetin and BglA-treated hesperidin groups since the latter
Figure 6. Changes in plasma concentration of total hesperetin in rats
following oral administration. The rats were divided into five groups, and
each rat was administered orally 31 µmol/kg equivalent amount of
hesperidin. Plasma samples were treated with ꢀ-glucuronidase, and
therefore, the plasma levels represent the total hesperetin equivalent.
Values were presented as mean ( SEM (n ) 5).
rapidly hydrolyzed hesperidin to hesperetin.
Whereas untreated hesperidin not only was absorbed more
slowly, it also had much smaller AUC values following IV
administration (2-fold). In oral administration, the difference
between pure hesperetin and hesperidin was more than 3-fold,
whereas the difference between untreated hesperidin and BglA
protein-treated hesperidin was even more at 4-fold or higher.
Again, this is similar to IV administration where the extract
had a larger difference since the extract may contain additional
phytochemicals that may interact with the absorption and/or
disposition process of hesperidin.
hesperetin released with ꢀ-glucuronidase from all species of
hesperetin conjugates as total hesperetin. We quantified the
plasma concentration of total hesperetin as well.
Analysis of plasma concentration-time curves of total hesperetin
(shown in Figure 5) following IV administration indicated that
pure hesperetin had a much larger area under curve (AUC) value
than pure hesperidin and that the relative bioavailability of
hesperetin doubled that of hesperidin (61.5 nmol h/mL vs 30.1
nmol h/mL) (Table 2). This large AUC value appears to be
consistent with a slower clearance of hesperetin when compared
to hesperidin (1.17 mL/h vs 0.56 mL/h). However, we also see
that hesperetin has a much smaller volume of distribution, perhaps
due to its tight binding to the plasma protein when compared to
hesperidin. Several other important pharmacokinetic parameters of
pure compounds are shown in Table 2.
In addition to pure compounds, we also injected purified
hesperidin extracts with or without BlgA protein treatment,
which hydrolyzed hesperidin to hesperetin. As expected, the
hesperidin extract without BglA treatment behaved just like
hesperidin shown earlier, whereas that treated with BglA
protein behaved almost like pure hesperetin. The only
exception was that the untreated hesperidin extract had an
even smaller AUC value, faster clearance, and larger volume
of distribution than pure hesperidin, suggesting that the herbal
mixture contained additional phytochemicals that may interact
with hesperidin to moderately change its pharmacokinetic
parameters.
DISCUSSION
Hesperidin has been reported to have multiple biological
activities but has a poor bioavailability. Chenpi is the common
source of hesperidin, and it has been found to have a higher
bioactivity after a nine-step fermentation process. Moreover, a
yeast strain closely related to S. singularis and containing a
ꢀ-glycosidase-like BglA protein was isolated from chenpi.
Therefore, the present hypothesis is that better herbal materials
may result from fermentation processes because they produce
more bioavailable active species.
Our study provides strong evidence to show that the hydroly-
sis of hesperidin could be enabled in vitro by using yeast
enzymes present in the chenpi fermentation process. For
example, yeast BglA proteins extracted from S. singularis or
expressed in E. coli BL21 (DE3) (Figure 4) are active against
hesperidin (Figure 2). In the presence of this BglA protein
(Figure 2), which hydrolyzed hesperidin to hesperetin, the total
plasma hesperetin pharmacokinetics was significantly altered
following IV and oral administration (Figures 5 and 6). For
example, the plasma Tmax value was significantly shorter, and
the plasma concentrations of total hesperetin were significantly

Pharmacokinetics of Hesperidin
J. Agric. Food Chem., Vol. 56, No. 14, 2008 5555
Table 3. Pharmacokinetic Parameters of Total Hesperetin after Single Oral Administration of Hesperidin or Hesperetin or Hesperidin Treated with Different
BlgA Protein Treatments
hesperidin extract
untreated
hesperidin extract +
BlgA from E. coli
hesperidin extract +
BlgA from yeast
parameters
pure hesperidin
pure hesperetin
AUCobsd av (nmol h/mL) SD
2.63
1.48
0.35
0.10
6
7.91a
4.15
0.95
0.37
4
1.50
0.79
0.20
0.10
6
6.82b
2.53
0.82
0.27
4
5.93b
2.26
0.74
0.19
4
Cmax av (nmol/mL) SD
Tmax av (h) SD
Vz av (mL) SD
0
0
0
0
0
9.86
5.05
56.49
27.42
8.53
2.30
5.65a
5.28
19.15a
6.34
13.26
4.68
13.14
1.73
114.06
32.24
10.36
8.31
2.75b
1.55
24.48b
5.99
11.39
3.02
4.29b
2.09
26.95b
10.32
10.03
2.35
CL av (mL/h) SD
F (%) av SD
a There are statistically significant differences between AUCobsd for pure hesperidin and pure hesperetin (p > 0.05). b There are statistically significant differences
between AUCobsd, Vz, and CL for hesperidin with different BlgA protein treatment and untreated hesperidin extract (p > 0.05).
Figure 7. Possible pathways for hesperidin and hesperidin disposition in the gut. The image shows hydrolysis of hesperidin by the BglA protein
in vitro, by microflora in lumen in vivo, transport pathways of hesperidin and hesperetin, and disposition pathways of hesperidin and hesperetin.
higher after oral administration of BglA protein-treated hespe-
ridin (Figure 6). Similarly, AUC values of hesperetin or treated
hesperidin also were significantly higher than hesperidin or
untreated hesperidin extracts (Tables 2 and 3). Therefore, the
fermentation process utilizing yeasts capable of secreting the
BglA protein likely will result in a higher bioavailability of
the flavonoids present in chenpi.
The higher bioavailability is likely the result of increased
uptake of hesperetin after administration of BglA protein-treated
hesperidin (containing mostly hesperetin) as compared to
untreated hesperidin extract (containing mostly hesperidin).
Faster absorption of hesperetin as compared to hesperidin was
shown previously in the Caco-2 cell culture model and various
rat and human pharmacokinetic studies (13, 26). In our own
studies, the data appear to suggest that efflux contributed to
slower absorption for hesperidin (glycoside) when compared
to hesperetin (aglycone) since Tmax was longer and Cmax was
smaller for the glycoside (Figure 6).
following both oral and IV administration. Since we conducted
the first side-by-side study of this kind, our data provide for
the first time strong evidence that hesperidin (a glycoside) also
was eliminated much more quickly than hesperetin (aglycone).
This is an important finding since it complements the current
theory that glycosides have a poor bioavailability because they
are poorly permeable or effluxed by transporters such as
MPR2 (15, 17, 18, 27). Another reason is the extensive
metabolism of aglycone (hesperetin) after absorption (8, 13,
26).
Although hesperidin is poorly absorbed and rapidly
eliminated, it has a reasonable half-life (6 h). This reasonable
half-life came first from a prolonged absorption phase, as
shown by a longer Tmax (Table 3). The prolonged absorption
phase came from its poor absorption, which forced a large
portion of glycosides (e.g., hesperidin) to the ileum and the
colon, where bacteria ꢀ-glucosidases capable of cleaving the
flavonoid glycosides are abundantly present and impact
bioavailability (28). Because of the action of intestinal
Another important reason as to why hesperetin or BglA-
treated hesperidin extract has a higher bioavailability is the faster
clearance of the glycosides. In fact, clearance values were higher

5556 J. Agric. Food Chem., Vol. 56, No. 14, 2008
Li et al.
marrow culture in vitro. Mol. Cell. Biochem. 2007, 303 (1-2),
83–8.
(8) Erlund, I.; Meririnne, E.; Alfthan, G.; Aro, A. Plasma kinetics
and urinary excretion of the flavanones naringenin and hesperetin
in humans after ingestion of orange juice and grapefruit juice. J.
Nutr. 2001, 131 (2), 235–41.
microflora, significant absorption of hesperetin would occur
after hesperidin gradually is hydrolyzed in the terminal ileum
and colon.
Another reason for hesperetin to have a reasonable plasma
half-life, as evidenced by the maintenance of the total
hesperetin concentration at a reasonable level even at 24 h,
is that hesperetin participates in enteric and enterohepatic
dual recycling (29). This is because hesperetin has been
shown to be extensively metabolized into phase II conjugates
such as glucuronides and sulfates. These two metabolites are
formed and eliminated by enterocytes and hepatocytes with
a substantial portion of these metabolites excreted into the
intestinal lumen (Figure 7). At the lower small intestine and
colon, these metabolites may be hydrolyzed by bacteria
ꢀ-glucuronidases and sulfatases to release the absorbable
aglycones, which may be reabsorbed into the plasma (Figure
7).
In conclusion, this is the first comprehensive and comparative
study of IV and oral pharmacokinetics of both hesperidin and
hesperetin in the same laboratory. The results indicate for the
first time that rapid plasma clearance is partially responsible
for the poor bioavailability of flavonoid glycosides, adding to
our knowledge that rapid clearance together with poor perme-
ation, low solubility, and/or extensive intestinal efflux are
responsible for the poor bioavailability. Because hesperetin, the
aglycone form of hesperidin, is much better absorbed with a
slower clearance and therefore higher bioavailability, a classical
fermentation process utilizing BglA protein-containing yeasts
should produce chenpi that have more bioavailable hesperetin
than dried citrus peels, thereby increasing their biological
activity.
(9) Gil-Izquierdo, A.; Gil, M. I.; Ferreres, F.; Tomas-Barberan,
F. A. In vitro availability of flavonoids and other phenolics in
orange juice. J. Agric. Food Chem. 2001, 49 (2), 1035–41.
(10) Kanaze, F. I.; Bounartzi, M. I.; Georgarakis, M.; Niopas, I.
Pharmacokinetics of the citrus flavanone aglycones hesperetin and
naringenin after single oral administration in human subjects. Eur.
J. Clin. Nutr. 2007, 61 (4), 472–7.
(11) Manach, C.; Morand, C.; Gil-Izquierdo, A.; Bouteloup-De-
mange, C.; Remesy, C. Bioavailability in humans of the
flavanones hesperidin and narirutin after the ingestion of two
doses of orange juice. Eur. J. Clin. Nutr. 2003, 57 (2), 235–42.
(12) Ross, J. A.; Kasum, C. M. Dietary flavonoids: Bioavailability,
metabolic effects, and safety. Annu. ReV. Nutr. 2002, 22, 19–34.
(13) Yamada, M.; Tanabe, F.; Arai, N.; Mitsuzumi, H.; Miwa, Y.;
Kubota, M.; Chaen, H.; Kibata, M. Bioavailability of glucosyl
hesperidin in rats. Biosci., Biotechnol., Biochem. 2006, 70 (6),
1386–94.
(14) Jeong, E. J.; Liu, X.; Jia, X.; Chen, J.; Hu, M. Coupling of
conjugating enzymes and efflux transporters: Impact on bioavail-
ability and drug interactions. Curr. Drug Metab. 2005, 6 (5), 455–
68.
(15) Kobayashi, S.; Tanabe, S.; Sugiyama, M.; Konishi, Y. Transepi-
thelial transport of hesperetin and hesperidin in intestinal Caco-2
cell monolayers. Biochim. Biophys. Acta 2007.
(16) Liu, Y.; Liu, Y.; Dai, Y.; Xun, L.; Hu, M. Enteric disposition
and recycling of flavonoids and ginkgo flavonoids. J. Altern.
Complement. Med. 2003, 9 (5), 631–40.
(17) Walgren, R. A.; Walle, U. K.; Walle, T. Transport of quercetin
and its glucosides across human intestinal epithelial Caco-2 cells.
Biochem. Pharmacol. 1998, 55 (10), 1721–7.
(18) Walle, U. K.; French, K. L.; Walgren, R. A.; Walle, T. Transport
of genistein-7-glucoside by human intestinal CACO-2 cells:
Potential role for MRP2. Res. Commun. Mol. Pathol. Pharmacol.
1999, 103 (1), 45–56.
(19) Kim, D. H.; Jung, E. A.; Sohng, I. S.; Han, J. A.; Kim, T. H.;
Han, M. J. Intestinal bacterial metabolism of flavonoids and its
relation to some biological activities. Arch Pharm Res 1998, 21
(1), 17–23.
ACKNOWLEDGMENT
Dr. Lin Wen is thanked for assistance with parameter
calculations using the WinNonlin software in the laboratory of
Prof. Tony Kong (Department of Pharmaceutics, Ernest-Mario
School of Pharmacy, Rutgers, the State University of New
Jersey, Piscataway, NJ).
LITERATURE CITED
(20) Lee, N. K.; Choi, S. H.; Park, S. H.; Park, E. K.; Kim, D. H.
Antiallergic activity of hesperidin is activated by intestinal
microflora. Pharmacology 2004, 71 (4), 174–80.
(21) van Rooyen, R.; Hahn-Hagerdal, B.; La Grange, D. C.; van Zyl,
W. H. Construction of cellobiose-growing and fermenting Sac-
charomyces cereVisiae strains. J. Biotechnol. 2005, 120 (3), 284–
95.
(22) Li, X. L.; Ljungdahl, L. G.; Ximenes, E. A.; Chen, H.; Felix,
C. R.; Cotta, M. A.; Dien, B. S. Properties of a recombinant beta-
glucosidase from polycentric anaerobic fungus Orpinomyces PC-2
and its application for cellulose hydrolysis. Appl. Biochem.
Biotechnol. 2004, 113-116, 233–50.
(23) Ma, Y. Y.; Xingqian; Yunbin, H.; Guoneng, X.; Guihua, X.;
Donghong, L. Ultrasound-assisted extraction of hesperidin from
Penggan (Citrus reticulata) peel. Ultrason. Sonochem. 2008, 15
(3), 227–32.
(24) Ishikawa, E.; Sakai, T.; Ikemura, H.; Matsumoto, K.; Abe, H.
Identification, cloning, and characterization of a Sporobolomyces
singularis beta-galactosidase-like enzyme involved in galacto-
oligosaccharide production. J. Biosci. Bioeng. 2005, 99 (4), 331–
9.
(25) Laemmli, U. K. Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature (London, U.K.)
1970, 227 (5259), 680–5.
(1) Mouly, P.; Gaydou, E. M.; Auffray, A. Simultaneous separation
of flavanone glycosides and polymethoxylated flavones in citrus
juices using liquid chromatography. J. Chromatogr., A 1998, 800
(2), 171–9.
(2) Yeh, C. C.; Kao, S. J.; Lin, C. C.; Wang, S. D.; Liu, C. J.; Kao,
S. T. The immunomodulation of endotoxin-induced acute lung
injury by hesperidin in vivo and in vitro. Life Sci. 2007, 80 (20),
1821–31.
(3) Kim, J. Y.; Jung, K. J.; Choi, J. S.; Chung, H. Y. Hesperetin: A
potent antioxidant against peroxynitrite. Free Radical Res. 2004,
38 (7), 761–9.
(4) Wilcox, L. J.; Borradaile, N. M.; de Dreu, L. E.; Huff, M. W.
Secretion of hepatocyte apoB is inhibited by the flavonoids,
naringenin and hesperetin, via reduced activity and expression of
ACAT2 and MTP. J. Lipid Res. 2001, 42 (5), 725–34.
(5) Knekt, P.; Kumpulainen, J.; Jarvinen, R.; Rissanen, H.; Heliovaara,
M.; Reunanen, A.; Hakulinen, T.; Aromaa, A. Flavonoid intake
and risk of chronic diseases. Am. J. Clin. Nutr. 2002, 76 (3), 560–
8.
(6) Tanaka, T.; Kohno, H.; Mori, H. Chemoprevention of colon
carcinogenesis by dietary non-nutritive compounds. Asian Pac.
J. Cancer PreV. 2001, 2 (3), 165–177.
(7) Yamaguchi, M.; Hamamoto, R.; Uchiyama, S.; Ishiyama, K.
Effects of flavonoid on calcium content in femoral tissue culture
and parathyroid hormone-stimulated osteoclastogenesis in bone

Pharmacokinetics of Hesperidin
J. Agric. Food Chem., Vol. 56, No. 14, 2008 5557
(26) Nielsen, I. L.; Chee, W. S.; Poulsen, L.; Offord-Cavin, E.;
Rasmussen, S. E.; Frederiksen, H.; Enslen, M.; Barron, D.;
Horcajada, M. N.; Williamson, G. Bioavailability is improved by
enzymatic modification of the citrus flavonoid hesperidin in
humans: A randomized, double-blind, crossover trial. J. Nutr.
2006, 136 (2), 404–8.
(27) Liu, Y.; Hu, M. Absorption and metabolism of flavonoids in the
caco-2 cell culture model and a perused rat intestinal model. Drug
Metab. Dispos. 2002, 30 (4), 370–7.
(28) Silberberg, M.; Morand, C.; Mathevon, T.; Besson, C.; Manach,
C.; Scalbert, A.; Remesy, C. The bioavailability of polyphenols
is highly governed by the capacity of the intestine and of the liver
to secrete conjugated metabolites. Eur. J. Nutr. 2006, 45 (2), 88–
96.
(29) Liu, Z.; Hu, M. Natural polyphenol disposition via coupled
metabolic pathways. Expert Opin. Drug Metab. Toxicol. 2007, 3
(3), 389–406.
Received for review January 11, 2008. Revised manuscript received
March 26, 2008. Accepted April 29, 2008. This study was supported
by The Scientific Research Foundation for the Returned Overseas
Chinese Scholars, State Education Ministry of People’s Republic of
China, Tianjin Municipal Science and Technology Commission
(05YFJMJC03900 and 08JCZDJC23400), and The National Natural
Science Foundation of China (30772483) to all authors except M.H. M.H.
is supported by a grant from The National Institutes of Health
(GM070737).
JF800105C