Analysis of Fluorescence Spectra of Citrus Polymethoxylated Flavones and Their Incorporation into Mammalian Cells

Article abstract of DOI:10.1021/acs.jafc.8b02052

Citrus polymethoxylated flavones (PMFs) influence biochemical cascades in human diseases, yet little is known about how these compounds interact with cells and how these associations influence the actions of these compounds. An innate attribute of PMFs is their ultraviolet-light-induced fluorescence, and the fluorescence spectra of 14 PMFs and 7 PMF metabolites were measured in methanol. These spectra were shown to be strongly influenced by the compounds' hydroxy and methoxy substituents. For a subset of these compounds, the fluorescence spectra were measured when bound to human carcinoma Huh7.5 cells. Emission-wavelength maxima of PMF metabolites with free hydroxyl substituents exhibited 70-80 nm red shifts when bound to the Huh7.5 cells. Notable solvent effects of water were observed for nearly all these compounds, and these influences likely reflect the effects of localized microenvironments on the resonance structures of these compounds when bound to human cells.

Full text of DOI:10.1021/acs.jafc.8b02052

Article
pubs.acs.org/JAFC
Cite This: J. Agric. Food Chem. XXXX, XXX, XXX−XXX
Analysis of Fluorescence Spectra of Citrus Polymethoxylated
Flavones and Their Incorporation into Mammalian Cells
Danielle R. Gonca
and Thais B. Cesar†
̧
lves,† John A. Manthey,*,‡ Paulo I. da Costa,§ Marilia C. M. Rodrigues,†
†Department of Food and Nutrition, Laboratory of Nutrition, Faculty of Pharmaceutical Sciences, Sao
(UNESP), Araraquara 01049-010, Brazil
̃
Paulo State University
‡U.S. Horticultural Research Laboratory, ARS, United States Department of Agriculture, 2001 South Rock Road, Fort Pierce, FL
34945, United States
§Clinical Analysis Department, School of Pharmaceutical Sciences, Sao
Brazil
̃
Paulo State University (UNESP), Araraquara 01049-010,
ABSTRACT: Citrus polymethoxylated flavones (PMFs) influence biochemical cascades in human diseases, yet little is known
about how these compounds interact with cells and how these associations influence the actions of these compounds. An innate
attribute of PMFs is their ultraviolet-light-induced fluorescence, and the fluorescence spectra of 14 PMFs and 7 PMF
metabolites were measured in methanol. These spectra were shown to be strongly influenced by the compounds’ hydroxy and
methoxy substituents. For a subset of these compounds, the fluorescence spectra were measured when bound to human
carcinoma Huh7.5 cells. Emission-wavelength maxima of PMF metabolites with free hydroxyl substituents exhibited 70−80 nm
red shifts when bound to the Huh7.5 cells. Notable solvent effects of water were observed for nearly all these compounds, and
these influences likely reflect the effects of localized microenvironments on the resonance structures of these compounds when
bound to human cells.
KEYWORDS: tangeretin, nobiletin, sinensetin, metabolites, citrus, Rutaceae, solvent effects, resonance, pyrylium, Huh7.5 cells
tridesmethyl nobiletin. In our study, fluorescence spectra of the
PMFs naturally occurring in orange juice as well as seven
metabolites of these compounds were measured in methanol.
The solvent effects of water on the fluorescence spectra of
these PMFs and their metabolites were also investigated.
Solvent effects of water are likely due to hydrogen bonding by
water molecules with partial-charge-transfer centers within
PMF resonance structures. An observation of the binding of
PMFs to human Huh7.5 carcinoma cells was made by the
detection of emissions from the 405 nm laser illumination of
cells preincubated with a number of these compounds for 24 h.
These spectra may possibly reflect features of the many
chemical microenvironments involved in the binding of these
compounds to human cells.
INTRODUCTION
■
Polymethoxylated flavones (PMFs) are a group of flavonoids
characteristically found in the genus Citrus (Rutaceae),
particularly in sweet and sour oranges, grapefruits, and
tangerines.1−3 They possess a typical flavonoid C6−C3−C6
carbon skeleton with methoxy groups (OCH3) distributed on
the A- and B-rings and rarely at the 3-position of the middle
pyrone ring (Figure 1). This polymethoxylation facilitates
cellular absorption of these compounds4−6 and influences the
levels of activities of these compounds against oxidative
damage, cancer, inflammation, erythrocyte aggregation, and
lipid biosynthesis.7−15 Recent studies have now shown that
certain desmethylated and glucuronidated PMF metabolites,
once thought to be less active than the original compounds,
have in vitro actions that exceed those of the original
compounds16−18 and thus are also likely to be active in vivo.
The substitution patterns of the PMF A- and B-rings
influence the molecular electronic structures of these
compounds,19 and these influences are clearly observed in
their UV−visible absorbance and fluorescence spectra. Like-
wise, large differences often occur in the in vitro biochemical
effectiveness of individual PMFs and their metabolites. Such
differences can arise from variations in membrane perme-
abilities as well as from differences in the binding of these
compounds to catalytic enzymes, transport proteins, and
nuclear receptors.20−23 A recent report described the detection
within cells of a fluorescent conjugate of 5,3′,4′-tridesmethyl
nobiletin and 2-aminoethyl diphenyl borate.24 Much weaker
fluorescence was detected in cells incubated with only 5,3′,4′-
MATERIALS AND METHODS
■
Material. Distillation residues of cold-pressed orange oil with
approximately 10% PMF content were obtained from a local Florida
flavor company. The residues were mixed with methanol (1/1, v/v)
and stirred overnight at room temperature, and the PMF-containing
methanol solution was allowed to separate from the denser insoluble-
residue mixture; it was then decanted and saved. The remaining
residue was again mixed and stirred with another aliquot of methanol,
and the first and second methanol solutions were combined and
evaporated to an oil containing approximately 50% PMF. This
Received: April 20, 2018
Revised: June 20, 2018
Accepted: June 21, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/acs.jafc.8b02052
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
PMF administration, no rats were lost, but rather all showed good
tolerance to the 200 mg kg−1 daily dose, maintained typical feed and
water intake, and showed the weight gain normal for this species.
Extraction of PMF Metabolites in Rat Urine. The pooled rat
urine from each group was loaded onto a methanol- and water-
prewashed SNAP RP-C18-HS (60 g) column (Biotage). The
compounds retained on the column were eluted with methanol and
dried under vacuum. This dried urine residue was dissolved in 200
mL of methanol and added to 800 mL of aqueous 0.5% formic acid.
The metabolites were extracted three times with 500 mL of ethyl
acetate. The combined ethyl acetate extracts were evaporated, and the
recovered residue was kept at −20 °C. The urine of the control group
contained no metabolites and was not processed further.
Purification of PMF Metabolites. The initial step in the
isolation of the PMF metabolites involved counter-current chroma-
tography using fast-centrifugal-partition chromatography (FCPC).
The FCPC conditions were optimized by choosing a biphasic-solvent
system with a suitable partition coefficient.25 For this application, the
biphasic-solvent system was composed of ethyl acetate/acetonitrile/
water with 0.5% formic acid (1/1/2, v/v/v), and the mobile phase
consisted of the organic fraction (upper phase), whereas the
stationary phase consisted of the aqueous fraction (lower phase).
Prior to injection, the sample was diluted in the two-phase system (v/
v) and passed through a 0.45 μm filter. Chromatographic separations
were achieved by using the Kromaton A-1000 rotor, operated at 1200
rpm, with the mobile phase delivered by a Waters Delta 600 HPLC
pump at 20 mL min−1 with 20 mL collected fractions. Desmethyl
aglycone metabolites were recovered in the early-eluting FCPC
fractions, and the glucuronidated metabolites eluted later (data not
shown). Detection of these compounds was achieved by HPLC-MS.
Further purification of the metabolites was achieved by preparative
thin-layer chromatography using silica-gel TLC plates (20 × 20 cm,
Uniplate Analtech, Inc.) and toluene/acetone (3/1) as the developing
solvent.
Figure 1. Sites of methoxylation and identities of the PMFs.
Final purifications of the PMF glucuronide metabolites were
accomplished with preparative HPLC using Varian ProStar 210/215
pumps (Varian) equipped with a Varian model 335 UV detector and
the Star Chromatography Data System. Compound separations were
achieved with either a Waters XBridge C8 (19 × 100 mm, 5 μm) or a
Waters Atlantis dC18 (19 × 100 mm, 5 μm) reverse-phase
semipreparative column, in which compound separations involved
isocratic chromatographic runs with acetonitrile/0.5% formic acid in
water with ratios of 20/80 to 30/70 (v/v), depending on the elution
characteristics of each metabolite. All samples were diluted in DMSO
prior to injection.
HPLC-ESI-MS Analyses. Analyses of the PMFs and PMF
metabolites were done with a Waters 2695 Alliance HPLC (Waters)
connected in parallel with a Waters 996 PDA detector and a Waters/
Micromass ZQ single-quadrupole mass spectrometer equipped with
an electrospray-ionization (ESI) source. Compound separations and
MS detections were achieved as previously described.26
material was dissolved in toluene and loaded onto 340 g HP-Sil silica-
gel columns (Biotage). Column chromatography was run using linear
gradients of hexane and ethyl acetate to achieve partial separations of
tangeretin (TAN), nobiletin (NOB), sinensetin (SIN), and
3′,4′,3,5,6,7,8-heptamethoxyflavone (HMF). The purifications of
these compounds were completed by reverse-phase liquid chromatog-
raphy using a 130 g RediSep C18 Reverse Phase column (Teledyne/
Isco) with linear gradients of methanol/aqueous 0.5% formic acid.
Final purification was achieved by recrystallization in hexane,
methanol, or mixtures of solvents at −20 °C and was followed by
washes with 4 °C hexane.
Production and Collection of PMF Metabolites from Rats.
Forty male Wistar rats weighing 180 5 g from the Animal Center of
Sao Paulo State University (UNESP) were separated into 4 groups,
̃
with 10 animals in each group, to receive doses of (1) TAN, (2)
NOB/(SIN) (70/30 w/w), (3) HMF, or (4) vehicle (the control
group). They were placed in individual metabolic cages and
Fluorescence and UV Spectra. Fluorescence spectra were
recorded with a PerkinElmer LS55 scanning fluorescence spectrom-
eter. Spectra of the pure compounds were recorded in methanol at
appropriate concentrations of either 0.01 or 0.001 mg mL−1. The exit-
and emission-slit widths were 5 nm with a scan speed of 200 nm/min.
UV spectra were recorded with a UV-2401PC Shimadzu UV−vis
recording spectrophotometer.
Molecular-Structure Determination of PMF Metabolites.
Structure investigations of the PMF metabolites involved UV−visible
spectrophotometry,27 mass spectrometry,28,29 and nuclear magnetic
resonance (NMR),30 as well as chemical modification using acid
hydrolysis and alkaline degradation. PMF fragment ions formed
during ESI-MS by retro-Diels−Alder fissions are reliable predictors of
the substitution patterns of the flavone A- and B-rings.28,30−33 This
information was used along with the results of the alkaline
degradation to give preliminary structural analyses of these
compounds. These details were further verified by 1H NMR and
comparisons with known standards.
maintained under controlled conditions: room temperature (23
°C), 55 5% relative humidity, and a 12−12 h day−night cycle.
Initially, the rats were fed the normal lab chow diet (Purina Evialis do
Brasil Nutrica Animal Ltd.a.) and water ad libitum for 7 days to
1
̧ o
̃
́
precondition the rats to their environment. Plain yogurt (Nestle) was
used as the vehicle for the PMF animal feeding via gavage. PMFs were
separately incorporated into 1 mL of plain yogurt per day, using a
homogenizer (Metabo, Marconi) operating at 27 000 rpm for 10 min
at a controlled temperature of 25 °C.
After the 1 week adaptation, the rats received 200 mg of a single
PMF per kilogram of body weight by gavage once a day at 5:30 p.m.
over 15 days. The rats were weighed every other day to calculate the
correct daily doses. The control group received 1 mL of the vehicle
per day. The animal care and study protocols were approved by the
Ethics Committee for Animal Use at the School of Pharmaceutical
Sciences, UNESP (No. 46/2015). Collection of urine was performed
twice a day (8 a.m. and 5 p.m.). The entire daily volume of urine was
homogenized and stored at −80 °C for 15 days. During the period of
B
DOI: 10.1021/acs.jafc.8b02052
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Alkaline Degradation. PMF metabolites were individually mixed
(∼1.0 mg of dry weght) with 1 mL of 25% KOH in a 15 mL screw-
top test tube and heated to 70 °C with a heating block for 6−24 h.
Some compounds were initially dissolved in 200 μL of methanol.
Upon completion of the heating, the KOH solution was combined
with ∼1 mL of glacial acetic acid to adjust the pH to 3−6. The
degradation products were analyzed by HPLC-ESI-MS. Alkaline-
degradation products of SIN, TAN, NOB, and tetramethylscutellarein
(TMS) were similarly analyzed for method validation.
Acid Hydrolysis. PMF metabolites were mixed (∼1.0 mg) with 1
mL of 0.1 M HCl and heated to 100 °C for 2 h. Upon cooling, the
hydrolysis solution was neutralized with 0.2 M KOH and then
analyzed by HPLC-ESI-MS.
Table 1. PMF Metabolites 1−7: ESI-MS of Protonated
Molecular-Mass Ions, HPLC-Elution Times, and UV-
Wavelength Maxima
[M + H]+1
ESI-MS+
RT
(min)
UV spectrum
(nm)
compound
proposed compound
1
535/359
18.74 230, 271, 320 TAN-4′-O-
glucuronide
2
521/345
16.70 243, 264, 329 didesmethyl SIN-
glucuronide
3
4
5
6
7
359
359
565/389
389
25.22
268, 325
4′-desmethyl TAN
23.40 242, 265, 333 monodesmethyl SIN
19.40 249, 267, 336 NOB-glucuronide
25.21 249, 271, 332 4′-desmethyl NOB
60 MHz NMR. PMF metabolites were also analyzed with a 60 MHz
1
Nanalysis 60e permanent-magnet H NMR. Dried metabolite (1−5
581/405
23.38 254, 270, 346 3′,4′-didesmethyl
mg) was dissolved in either 0.7 mL of deuterated dimethyl sulfoxide
(DMSO-d6) or deuterated methanol. Spectra were recorded at 30 °C.
Typically, spectra were collected as a sum of 100 to 4064 Fourier-
transform scans depending on the amount of compound (5 to 1 mg)
available for measurements. Some compounds were also analyzed by
400 MHz high-resolution NMR at 25 °C, and the compounds were
dissolved in either DMSO-d6 or deuterated methanol.
HMF
Table 2. Fluorescence Parameters of PMF Standards in
Orange-Peel-Oil-Distillation Residue
a
compound λEm max (nm) FIU
structure
DMSO
5DMN
5DMHMF
5DMS
5DMTMS
5DMT
ISOSIN
SIN
QHME
TMS
GTME
ISOTMS
TAN
none
none
none
none
none
none
435
443
452
500
none
Cell Culture and Viability. The Huh7.5 human hepatoma cell
line was kindly provided by Dr. Paula Rahal (Laboratory of Genomic
Studies of the Biology Department of the Institute of Biosciences,
Letters and Exact Sciences, IBILCE, UNESP, Sao Jose do Rio Preto,
SP, Brazil). Huh7.5 cells are derived from the cell line Huh7, a lineage
of hepatocytes from a carcinoma that were originally taken from a
Japanese man, aged 57 years, in 1982. Cells were cultured in DMEM
(Dulbecco’s modified Eagle’s medium, Sigma D1152) supplemented
with 10% fetal bovine serum, 1% penicillin/streptomycin, and 1%
nonessential amino acids (DMEM-C) in a 5% CO2-humidified
incubator at 37 °C. The doses used to evaluate the incorporation of
the PMF metabolites into mammalian cells were determined by cell-
viability assays using the MTT colorimetric assay.34
Confocal-Microscopy Analysis. The incorporation of pure
PMFs and metabolites into the Huh7.5 cells was observed by
confocal microscopy. The cells were plated at a density of 1 × 105
cells mL−1, prepared in DMEM-C, and distributed in 2 mL in a plate
suitable for confocal microscopy (35 × 10 mm and adhered glass
cover sheet, 22 × 22 mm). The plate was incubated at 37 °C in a
humidified chamber with 5% CO2 for 12 h. Next, PMFs were added at
50 μM in DMEM-C, whereas the metabolites were added at 25 μM,
and the cells were incubated over 24 h. These concentrations were
selected from the cytotoxicity data to ensure high doses with cell
viability greater than 80%. After this period, cells were washed twice
with phosphate-buffer solution (PBS), and 1 mL of PBS was added to
each plate to perform confocal-microscopy analysis.
none 5-desmethyl nobiletin
none 5-desmethyl heptamethoxyflavone
none 5-desmethyl sinensetin
none 5-desmethyl tetramethylscutellarein
none 5-desmethyl tangeretin
45
190
343
96
3′,4′,5,7,8-pentamethoxyflavone
3′,4′,5,6,7- pentamethoxyflavone
3′,4′,3,5,6,7-hexamethoxyflavone
4′,5,6,7-tetramethoxyflavone
3′,4′,3,5,7,8-hexamethoxyflavone
4′,5,7,8-tetramethoxyflavone
520/450(sh)
438
25
62
535
trace 4′,5,6,7,8-pentamethoxyflavone
NOB
450
42
3′,4′,5,6,7,8-hexamethoxyflavone
HMF
452
43
3′,4′,3,5,6,7,8-
heptamethoxylflavone
a
Measurements recorded in methanol with excitation wavelengths (λ)
at 285 and 330 nm. Compound concentrations were 0.01 mg mL−1
except ISOSIN, SIN, and QHME, which were 0.001 mg mL−1. For
micromolar concentrations see Table 3. Fluorescence-intensity units
(FIU) are measures against the maximum fluorescence intensity
defined as 1000 units by the PerkinElmer LS-55 fluorescence
spectrophotometer.
The images were obtained with a Confocal Zeiss 780 Microscope
̃ ̃
at the Physics Institute of Sao Carlos, University of Sao Paulo. The
fluorescence spectra were obtained using a 405 nm laser, filters from
413 to 692 nm, a 98 μm pinhole, and a 40× objective. The images
were obtained with a 405 nm laser, also with the 40× objective, and
an emission-filter range between 415 and 550 nm. The negative
control of the test consisted of cells treated with only DMEM-C
medium.
RESULTS
■
PMF Metabolites from Rat Urine. Seven metabolites, 1−
7, were isolated from PMF-dosed animals, and their main ESI-
MS fragment ions, HPLC elution times, and UV-absorbance-
wavelength maxima are listed in Table 1. The results of initial
structural determinations of 1−7 are discussed below. MS
fragmentation provided partial information about the sub-
stituents attached to the A- and B-rings,28,30,32 and alkaline
hydrolysis provided further A- and B-ring-structure valida-
tion.35,36 There were not sufficient amounts of purified 2, 4,
and 5 to achieve complete structure determinations.
Figure 2. Fluorescence spectra of QHME and SIN at 0.001 mg mL−1
and of NOB and HMF at 0.01 mg mL−1 in methanol. The Y-axis is
presented in fluorescence-intensity units (FIU).
Metabolite 1. Tangeretin-4′-O-glucuronide. UV λmax 271,
320 nm. ESI+ 535 m/z [M + H]+1, 345, 211, 183, 121, 119.
C
DOI: 10.1021/acs.jafc.8b02052
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
and 153 m/z are evidence of a trimethoxy A-ring, and fragment
ions 135 and 137 m/z are indicative of a dihydroxy B-ring.
Acid hydrolysis yielded a flavone with ESI+ 345 m/z [M +
H]+1 and fragment ions 195, 181, 153, 135, 137, and 125,
which are identical to 2. It is suggested that hydrolyzed 2 is
3′,4′-didesmethyl-sinensetin. Sinensetin was the only PMF
administered to the rats with a trimethoxy A-ring.
Metabolite 3. 4′-Hydroxy-5,6,7,8-tetramethoxyflavone (4′-
desmethyl tangeretin). UV λmax 268, 325 nm. ESI+ 389 m/z
[M + H]+1, 211, 183, 121, 119. Fragment ions 211 and 183 m/
z are evidence of a tetramethoxy A-ring, and fragment ions 119
and 121 m/z are evidence of a monohydroxy B-ring. High-
1
resolution H NMR of 3 is identical to that for 4′-desmethyl
tangeretin.37
Metabolite 4. 3′-Desmethyl sinensetin or 4′-desmethyl
sinensetin. UV λmax 242, 265, 333 nm. ESI+ 359 m/z [M +
H]+1, 181, 153, 151, 149. Fragment ions 181 and 153 m/z are
evidence of a trimethoxy A-ring, and fragment ions 149 and
151 m/z are indicative of a monohydroxy-monomethoxy B-
Figure 3. Fluorescence spectra of SIN at 0.001 mg mL−1 versus TMS
and NOB at 0.01 mg mL−1 versus TAN at 0.01 mg mL−1 in methanol.
The Y-axis is presented in fluorescence-intensity units (FIU).
1
ring. H NMR (60 MHz, CD3OD): H2′, H6′ (d, 2H, 7.48,
7.57), H8 (s, 7.17), H5′ (d, Jortho = 8.8 Hz, 1H, 6.80, 6.95), H3
(s, 1H, 6.68); four OCH3 at 3.92, 3.86, 3.77, 3.74 ppm. This
metabolite was not isolated in sufficient amounts to allow
complete structural elucidation relating to the position of the
unsubstituted hydroxyl on the B-ring nor for accurate
calculations of the coupling constants for H2′ and H6′ using
60 MHz NMR.
Metabolite 5. Nobiletin-O-glucuronide. UV λmax 249, 267,
336 nm. ESI+ 565 m/z [M + H]+1, 389, 359, 197, 169, 165,
163. Fragment ions 197 and 169 m/z are evidence of a
monohydroxy-trimethoxy A-ring, and fragment ions 165 and
163 m/z are indicative of a dimethoxy B-ring. Acid hydrolysis
yielded a flavone aglycone with UV λmax 250, 279, and 336 nm
and MS-fragmentation ions identical to those of 5, with the
exception of the glucuronide-conjugate ion at 565 m/z.
Alkaline hydrolysis of 5 produced further evidence of a
1
monohydroxy-trimethoxy A-ring. Hydrolyzed 5 exhibited H
NMR (60 MHz, CD3OD) as the following: H2′, H6′ (d, Jortho
= 9.3 Hz; 2H, 7.52, 7.68), H5′ (d, Jortho = 8.3 Hz; 1H, 7.05,
1
7.19), H3 (s, 1H, 6.78). H NMR (60 MHz, CD3OD) of 5
exhibited nearly identical aromatic proton signals as those
detected for hydrolyzed 5 as well as five OCH3 at 4.09, 4.01,
1
3.90 (×2), and 3.87 ppm. High-resolution H NMR (400
MHz, DMSO-d6) of hydrolyzed 5 exhibited the following: H2′,
H6′ (d, 2H, 7.64, 7.65; 7.63, 7.62; 7.54, 7.53), H5′ (d, 1H,
7.17, 7.15), H3 (s, 1H, 6.78); five OCH3 at 3.96 (×2), 3.88,
3.85, 3.73. 13C NMR provides additional evidence of the
occurrence of adjacent 3′- and 4′-methoxy substituents and no
C7-methoxy signal (data not shown). Sufficient amounts were
not available to achieve a complete structure identification of 5,
but the results above support a flavone with a 3′,4′-dimethoxy
B-ring and a monohydroxy-trimethoxy A-ring. Positive
identifications of 7-desmethyl nobiletin and 6-desmethyl
nobiletin have been reported in nobiletin metabolites
produced by human liver microsomes,38 and we propose that
5 occurs as one of these metabolites. 7-Desmethyl nobiletin in
citrus peels has also been previously reported.31,39
Metabolite 6. 3′-Methoxy-4′-hydroxy-5,6,7,8-tetramethox-
yflavone (4′-desmethyl nobiletin). UV λmax 249, 271, 332 nm.
ESI+ 389 m/z [M + H]+1, 211, 183, 149, 151. Fragment ions
211 and 183 m/z are evidence of a tetramethoxy A-ring, and
fragment ions 149 and 151 m/z are indicative of a
monohydroxy-monomethoxy B-ring. 1H NMR (60 MHz,
Figure 4. (A) Fluorescence spectra of NOB(m) and HMF(m) at 0.01
mg mL−1 and SIN(m) at 0.001 mg mL−1 in methanol and of
NOB(aq), HMF(aq), and SIN(aq) in 80/20 (v/v) water/methanol.
The Y-axis is presented in fluorescence-intensity units (FIU). (B)
Fluorescence spectra of TAN in methanol at 0.01 mg mL−1 with
decreased percentages of methanol following dilution with water. The
Y-axis is presented in fluorescence-intensity units (FIU).
Fragment ions 211 and 183 m/z are evidence of a
tetramethoxy A-ring, and fragment ions 119 and 121 m/z are
indicative of a monohydroxy B-ring. High-resolution 1H NMR
of the aromatic protons of 1 is identical to that of 4′-hydroxy
tangeretin.37 Acid hydrolysis of 1 yielded 3 (4′-desmethyl
tangeretin).
Metabolite 2. 3′,4′-Dihydroxy-5,6,7-trimethoxyflavone-4′-
glucuronide or 3′,4′-dihydroxy-5,6,7-trimethoxyflavone-3′-glu-
curonide. UV λmax 243, 264, 329 nm. ESI+ 521 m/z [M +
H]+1, 345, 195, 181, 153, 135, 137, 125. Fragment ions 181
D
DOI: 10.1021/acs.jafc.8b02052
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Table 3. Fluorescence Parameters of PMFs and PMF Metabolites Measured in Methanol and in Huh7.5 Cells
a
b
b
c
compound (μM)
λEm max (nm, cell bound)
λEm max (nm, methanol)
FUI (cell bound)
FUI (methanol)
solvent effects (ΔFIU/ΔλEm max)
1 (19)
2 (19)
3 (26)
4 (2.8)
5 (17)
6 (2.5)
7 (2.5)
487
470
487
513
470
522
522
435
435

425
508
412
450
446
456
494
450
452
450
443
430
535
496
434
14
27
12
22
9
48
48
23
74



25


14
451
31
trace
not measured
+24/+28
−1036/+10 nm
+640/+29 nm
−125/+14
−539/−29 nm
+475/+20 nm
+211/+26 nm
−69/+15 nm
+573/+13 nm
−2/+17 nm
60
120
153
608
65
NOB (25)
HMF (23)
QHME (2.5)
SIN (2.7)
ISOSIN (2.7)
72
351
255
45
10
201
181


d
TAN (27)
435

−7535 nm/+54419 nm
−146496 nm/+170415 nm
−39/+15 nm
d
TMS (29)
ISOTMS (29)

a
Structures include 1, tangeretin-4′-glucA; 2, dihydroxysinensetin-glucA; 3, tangeretin-4′-OH; 4, 3′- or 4′-hydroxysinensetin; 5, nobiletin-glucA; 6,
b
3′- or 4′-hydroxynobiletin; 7, 3′,4′-dihydroxyheptamethoxyflavone. For TAN, NOB, HMF, and SIN, see Table 2. Measurements recorded in
methanol with excitation wavelengths (λ) at 330 nm and measurements recorded in aqueous solutions (cell bound) with excitation wavelengths (λ)
at 405 nm using confocal microscopy. Concentrations are reported as micromolar (μM). Concentrations between 2.8 to 2.5 μM represent solutions
c
of 0.001 mg mL−1, and higher values between 17 to 29 μM represent solutions of 0.01 mg mL−1. Water-solvent effects are shown as ΔFIU/ΔλEm
max. The value ΔFIU represents the gains or losses in fluorescence-intensity units (FIU) from the peak in methanol to the peak in water/methanol
(80/20, v/v). The change in the wavelength of the emission maximum recorded in methanol versus that recorded in water/methanol (80/20, v/v)
d
is represented as ΔλEm max. TAN and TMS exhibited unusual shifts in their emission-wavelength maxima. The loss of FIU at the original λEm max
is listed first, followed by the gain in FIU in the aqueous λEm max.
Fluorescence Spectra of PMFs and PMF Metabolites
in Methanol. The fluorescence spectra of 14 PMFs were
measured at room temperature in methanol using excitation
wavelengths at 320 and 280 nm. These excitation wavelengths
occur within typical PMF absorbance bands I and II,
respectively.19 As shown in Table 2, no fluorescence occurred
for any of the 5-desmethyl PMFs, including 5DMN (5-
desmethyl nobiletin), 5DMHMF (5-desmethyl heptamethox-
flavone), 5DMTMS (5-desmethyl tetramethylscutellarein),
5DMT (5-desmethyl tangeretin), and 5-DMS (5-desmethyl
sinensetin). The absence of fluorescence in 5-hydroxyflavones
has been attributed to the formation of strong intramolecular
hydrogen bonds between the 5-hydroxyl of the A-ring and the
pyrone-ring carbonyl and the losses of planarity of the flavone
molecules.40 Both attributes disrupt the fluorescence emissions
arising from extended electron resonances in flavones. Among
the PMFs exhibiting UV-induced fluorescence, most had
similarly shaped emission spectra with wavelength maxima
between 440 and 460 nm (Figure 2). No qualitative differences
occurred in the spectra when either excitation wavelengths
(320 and 280 nm) was used (data not shown). In contrast to
the spectra in Figure 2, tangeretin (TAN) and tetramethylscu-
tellarien (TMS) showed emission-wavelength maxima at 535
and 498 nm, respectively, with TAN exhibiting only trace
fluorescence intensity compared with the other compounds
(Figure 3). The influence of having adjacent methoxy
substituents at the 3′- and 4′-positions of the B-ring is
illustrated by the contrasting fluorescence spectra of TAN
versus nobiletin (NOB) and TMS versus sinensetin (SIN),
respectively, in Figure 3.
The fluorescence spectra of many of the PMFs were notably
affected by the addition of water to the PMF/methanol
solutions. Although not all compounds in Table 1 were
evaluated for this effect, the fluorescence spectra of NOB, SIN,
quercetagetin hexamethyl ether (QHME), TMS, heptamethox-
yflavone (HMF), and TAN all showed solvent effects with the
Figure 5. Fluorescence spectra of TMS(aq) in water/methanol (80/
20, v/v) versus TMS(m) in methanol at 0.01 mg mL−1 and similarly
for isoTMS(aq) and isoTMS(m). The Y-axis is presented in
fluorescence-intensity units (FIU).
CD3OD): H2′, H6′ (d, 2H, 7.45, 7.17), H5′ (d, Jortho = 6.5 Hz,
1H, 7.01, 6.90), H3 (s, 1H, 6.70); four OCH3 at 3.98, 3.94,
3.81, 3.75. Alkaline hydrolysis of 6 produced vanillic acid, thus
providing evidence of a 4′-hydroxy-3′-methoxy B-ring,
consistent with 4′-desmethyl nobiletin.
Metabolite 7. 3′,4′-Dihydroxy-3,5,6,7,8-pentamethoxyfla-
vone (3′,4′-didesmethyl HMF). UV λmax 254, 270, 346 nm.
ESI+ 405 m/z [M + H]+1, 211, 183, 137, 135. Fragment ions
211 and 183 m/z are evidence of a tetramethoxylated A-ring,
and fragment ions 135 and 137 m/z are evidence of a
1
dihydroxy B-ring. H NMR (60 MHz, CD3OD): 5 methoxy
groups at 4.07, 3.99, 3.90, and 3.78 ppm; H2′, H6′ (d, Jortho
=
5.9, 2H, 7.69/7.55, 7.52), H5′ (d, 1H, Jortho = 7.6 Hz, 6.97,
6.84). The signal for the C-3 proton was absent, as expected
for an HMF metabolite.
E
DOI: 10.1021/acs.jafc.8b02052
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
and 3, originally from TAN, showed only trace fluorescence,
whereas 2 and 4−7, derived from SIN (2 and 4), NOB (5 and
6), and HMF (7), exhibited moderate to strong fluorescence,
with emission-wavelength maxima between 425 to 508 nm.
These compounds showed a variety of different solvent effects
with water. Because of the weak fluorescence-emission
intensities of 1 and 3, these solvent effects were difficult to
define, although there appeared to be decreased fluorescence
intensity throughout the spectrum of 1 with increased water
content, and for 3, the fluorescence intensities between 500 to
520 nm weakly increased, and there was no change at 420 nm
(data not shown). Metabolites 2, 4, 6, and 7, with free
hydroxyl substituents, showed dramatic losses in fluorescence
intensities when measured in aqueous methanol (Table 3),
whereas the fluorescence spectrum of 5 (lacking free hydroxyl
substituents), showed a 7.5× increased intensity in 80% water
compared with in 100% methanol.
PMF and PMF Metabolite Binding to Intact Huh7.5
Cells. UV-induced fluorescence was detected by confocal
microscopy in human carcinoma Huh7.5 cells preincubated 24
h with PMFs and PMF metabolites 1−7 using 405 nm laser-
induced excitation. Untreated control cells exhibited trace
background fluorescence, which contrasted with the intense
fluorescence exhibited by the PMF- and metabolite-treated
cells. Figure 6 shows the fluorescence images of illuminated
Huh7.5 cells incubated separately with metabolites 1−7.
Similar images were observed with the cells preincubated
with the original PMFs: TAN, NOB, and HMF (data not
shown). Despite the negligible levels of fluorescence of 1, 3,
and TAN in methanol, intense fluorescence was observed for
the Huh7.5 cells incubated with these compounds. Incubations
of the Huh7.5 cells were performed separately with 50 μM
PMFs and with 25 μM 1−7. These concentrations were
determined to ensure cell viability of >80%, confirmed by
MTT cell-viability measurements.34 MTT cytotoxicity assays
showed IC50 values for 1−7 between 139 and 219 μM, with an
average of 160 30 μM (data not shown).
Measurements of the emission spectra of the PMF-incubated
Huh7.5 cells were made to more fully characterize the
fluorescence of these compound-treated cells (Figure 7). Not
all of the PMFs were analyzed in this manner. The
fluorescence-emission spectra of the cells incubated with
NOB and HMF showed wavelength maxima at 461 and 460
nm, respectively, and a second maximum between 435 and 440
nm. The longer-wavelength maxima at 454 and 461 nm for
NOB and HMF are close to the wavelength maxima measured
for these compounds in methanol. TAN, which exhibited an
emission-wavelength maximum in methanol at 535 nm but at
434 nm in 80% water/methanol, exhibited an emission-
wavelength maximum of 454 nm when bound to the cells. The
spectrum obtained with cell-bound 2 shared some similarity
with the spectra of cell-bound TAN, NOB, and HMF. The
Huh7.5 cells treated with 4, 6, and 7 showed fluorescence-
emission spectra with wavelength maxima near 525 nm,
contrasting sharply with the spectra of Huh7.5-cell-bound
NOB, HMF, and TAN. The weak emission spectra obtained
from the Huh7.5 cells incubated with 1, 3, and 5 lacked
sufficient intensities to make clear spectroscopic determina-
tions, but in the cases of 1 and 5, the differences in the spectra
versus the control spectrum revealed weak emission maxima
between 440 and 450 nm and again at approximately 480 nm
(data not shown).
Figure 6. Confocal-microscope image of Huh7.5 after 24 h of
incubation with PMFs, metabolites, and a control. The scale is 10 μm.
addition of water. The intensities of the HMF, SIN, and NOB
fluorescence emissions in water/methanol (80/20, v/v)
solutions were sharply higher than the fluorescence intensities
of these same compounds in methanol (Figure 4A). In water/
methanol (80/20, v/v) solutions, the emission-wavelength
maxima of HMF, SIN, and NOB shifted to 480, 457, and 470
nm, respectively, compared with 452, 443, and 450 nm in
methanol (Table 3), each representing a 15 to 30 nm red-shift
to lower-energy transitions. Notably different solvent effects
were observed with the addition of water to methanol solutions
of TAN (Figure 4B). The new fluorescence-emission-wave-
length maximum of TAN in water/methanol (80/20, v/v)
occurred at 420 nm, compared with the previous value of 535
nm in methanol. Similar changes occurred in the spectrum of
4′,5,6,7-tetramethoxyflavone (TMS, Figure 5), yet the isomer
iso-tetramethylscutellarein (ISOTMS, 4′,5,7,8-tetramethoxyfla-
vone) did not show these same solvent effects. The emission-
wavelength maximum of ISOTMS in methanol occurred at
438 nm, red-shifting to 450 nm in water/methanol (80/20, v/
v) and decreasing by 18% in emission intensity (Figure 5).
The parameters of the fluorescence spectra of metabolites
1−7 measured in methanol are listed in Table 3. Metabolites 1
F
DOI: 10.1021/acs.jafc.8b02052
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Figure 7. Fluorescence spectra of PMFs and metabolites 1−7 in Huh7.5 cell culture. Metabolite 1, TAN-4′-glucA; metabolite 2, di-OH-SIN-glucA;
metabolite 3, TAN-4′-OH; metabolite 4, 3′- or 4′-hydroxy-SIN; metabolite 5, NOB-glucA; metabolite 6, 4′-desmethyl NOB; metabolite 7, 3′,4′-
didesmethyl HMF. The Y-axis is presented in fluorescence-intensity units (FIU).
stabilized by hydrogen bonding with solvent molecules as well
as by electron donation by the phenyl-ring substituents. As a
result of this, most flavones exist as mixtures of pyrone and
pyrylium resonance forms,65 both of which are differently
influenced by chemical environments.
DISCUSSION
■
The PMFs obtained from citrus peels constitute an extremely
small subset of the flavonoids in human diets, yet because of
their potential health benefits to humans, these compounds
have been extensively studied.16,41,42 Research has shown that
these compounds, along with many other highly substituted
flavones, exert beneficial effects by interacting with enzymes
involved in the progression of human diseases. Such targeted
enzymes include, in part, phosphodiesterases,43−45 xanthine
oxidase,46−48 adenosine receptors,21,49,50 cell-signaling
kinases,51−53 DNA topoisomerases,54−56 lipid-synthesis and
glucose-metabolism enzymes,11,57 and transport proteins.23,58
To investigate interactions such as these in human cells,
fluorescence spectroscopy may be useful because of the
nonintrusive nature of fluorescence measurements and the
sensitivity of these measurements to chemical microenviron-
ments.59
The fluorescence spectra of flavonoids reflect to a large
degree the extensive delocalization of electrons in the
molecules’ ground and excited states, both of which are
controlled by their chemical structures.60−62 A commonly
observed structural influence is the long-wavelength green-
colored fluorescence of flavonols (3-hydroxyflavones) pro-
duced by excited state keto−enol tautamers,62−64 and although
flavonols often exhibit this long-wavelength fluorescence, most
other flavones exhibit shorter-wavelength blue fluorescence. An
important aspect for these electron delocalizations is the
concurrent electron donation by the oxygen at the pyrone 1-
position (Figure 1) and the electron acceptance by the C4-
carbonyl oxygen to create charge-transfer zwitterionic reso-
nance structures. These π-electron delocalizations are often
In this study, the fluorescence spectra of 14 PMFs dissolved
in methanol and 7 additional PMF metabolites, as well as the
solvent effects of water on these spectra, were measured. An
increase in solvent polarity generally produces red-shifts in the
fluorescence-emission-wavelength maxima and lowers the
emission intensities for many fluorescent compounds.60
However, our study showed that for most of the original
unmodified PMF compounds (e.g., QHME, SIN, NOB, and
HMF), the fluorescence-emission intensities sharply increased
with the addition of water to the methanol solutions, but then
the wavelength maxima of these original unmodified PMFs
red-shifted by 15−30 nm, as seen in many other fluorescent
compounds. These solvent effects with water appear to be self-
contradictory and seem not to fit the pattern normally
observed with other fluorescent compounds.
A structural feature common to the above group of
compounds (QHME, SIN, NOB, and HMF) is the adjacent
methoxyl substituents on the B-ring 3′- and 4′-positions.
However, in contrast to these compounds, two other PMFs,
tetramethylscutellarein (TMS, 4′-hydroxy-5,6,7-trimethoxyfla-
vone) and tangeretin (TAN, 4′-hydroxy-5,6,7,8-trimethoxy-
flavone), lack the 3′-methoxyl substituent adjacent to the 4′-
position, and these compounds exhibited notably different
water-solvent effects. The addition of water to methanol
solutions of TMS and TAN created new intense fluorescence
emissions near 425 nm excitation, and this occurred regardless
G
DOI: 10.1021/acs.jafc.8b02052
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
suggest that compounds with methoxylation at the C8 position
(GHME, ISOTMS, TAN, NOB, and HMF) exhibit lower
fluorescence intensities than those without a substituent (SIN,
QHME, and TMS) at the C8 position. The research findings
obtained with metabolites 1−7 further support this tendency,
as well as the study of Wolfbeis,61 in which 5-methoxyflavone,
6-methoxyflavone, and 7-methoxyflavone were shown to
exhibit substantially greater fluorescence intensities than 8-
methoxyflavone.
The isolation and subsequent investigation of the
fluorescence spectra of PMF metabolites 1−7 expanded the
range of structures analyzed in this study and, in several cases,
provided valuable comparisons between closely related
compounds. The low fluorescence intensity of 1 is likely a
result of the closely matching A- and B-ring substitutions
shared with TAN, although the glucuronide substituent of 1
appears to cause differences in the solvent effects of water
(nearly completely absent in 1, see Table 3, but prominent for
TAN, see Figure 4), as well as differences in the fluorescence-
wavelength maxima of the two compounds (535 nm for TAN
and 425 nm for 1) in methanol (Table 3). In contrast, the
structurally similar NOB and 5 show strong similarities in their
fluorescence spectra both in methanol and in 80% water/
methanol, but the fluorescence spectra of these compounds
contrasted sharply when bound to the Huh7.5 cells.
Compounds 2−4, 6, and 7, which possess free hydroxyl
groups, showed dramatic drops in fluorescence emission in the
presence of water, and as discussed above, these decreases are
in sharp contrast to the changes of the original unmodified
PMF compounds, which showed large emission increases with
water. The solvent effect of water on the desmethylated
metabolites (compounds 2−4, 6, and 7) is similar to that of
quercetin (3′,4′,3,5,7-pentahydroxyflavone) and apigenin
(4′,5,7-trihydroxyflavone) with which sharp decreases in
fluorescence intensities occur in aqueous methanol and
acetonitrile solutions.69 These sharp decreases in fluorescence
emission can be readily attributed to intermolecular hydrogen
bonding, which contorts the planarity of either the ground or
excited state’s geometry or lowers the extent of π-electron
delocalization in the excited electronic state.
The images of the Huh7.5 cells preincubated with the PMFs
and the PMF metabolites (Figure 6) show bright fluorescence
associated with internal structures of the cells, presumably
cellular organelles. Nearly circular structures putatively
identified as oil droplets appear not to show any PMF
sequestration, and as a consequence, it is proposed that most
of the cellular associations of these compounds involve protein
binding. Research into the fluorescence emission of flavones
has led to earlier detections of these compounds in cultured
cells,70−74 including direct detection of kaempferol and other
flavonols in nuclei of Hepa-1c1c7 cells.73 An alternative
method of detecting flavones involved the detection of the
fluorescence of spontaneously formed flavonoid-2-amino-
ethoxydiphenyl borate (DPBA) conjugates in animal74 and
plant cells.75−77 The nobiletin metabolite 5,3′,4′-tridesmethyl-
nobiletin (TDN) was recently detected as the DPBA-TDN
conjugate in mouse cells.74 Fluorescence emissions of many
dietary plant flavonols are also strongly intensified by their
binding to target proteins, typically producing intense green-
colored fluorescence.25 This accentuated fluorescence can also
be used with confocal microscopy to detect flavonol
incorporation into cells, as previously shown for the Hepa-
1c1c7 cells.73
Figure 8. Resonance spectrum of PMFs with oxygen-containing
substituents on the A-ring C6 positions. (A) Example of
demethylation of the C6 methoxy of zapotin as discussed by
Dryer.66 (B) Proposed resonance structure of ISOTMS with a free C6
phenyl A-ring position. The free C6 position allows additional
resonance structures to further stabilize the ground and excited states
of compounds like ISOTMS, thus differentiating ISTMS from TMS
and TAN with methoxy substituents at the C6 position.
of the A-ring substituents at 5,6,7,8 in TAN or 5,6,7 in TMS.
However, an isomer of TMS, isotetramethylscutellarein
(ISOTMS, 4′-hydroxy-5,7,8-trimethoxyflavone) did not show
this same solvent effect, although it has the same 4′-methoxy
B-ring substitution. An analysis of these structures showed that
the main difference between the combined structural features
of TAN and TMS versus ISOTMS is the absence of an oxygen-
containing substituent at the A-ring C6 position in ISOTMS.
The manner by which the A-ring C6 position influences the
stabilities and electronic resonance structures of PMF
compounds is not certain, but it is known from other research
that mass fragmentation involving demethylation of A-ring
methoxyl substituents proceeds most readily at the C6 position
relative to any of the other A-ring positions. As shown by
Dreyer,66 a major factor contributing to this preference for
demethylation at the C6 position is the formation of a highly
stabilized pyrylium inner-C-ring resonance structure, such as
that illustrated for the compound zapotin in Figure 8A.
Without demethylation, the electron-donating properties of
methoxyl substituents at the C6 position in TAN and TMS
work against the formation of this type of pyrylium resonance
structure. However, this is not the case for ISOTMS with no
substituent at the C6 position (Figure 8B). Previous studies
have shown examples of pyrulium-like resonance stabilization
and enhanced fluorescence intensities accompanying the
binding of flavonols (3-hydroxyflavones) to human and rat
serum albumins.67,68
Further analysis of the fluorescence spectra of PMFs
measured in this study suggests that intensities of fluorescence
emissions are strongly influenced by the presence of a methoxy
substituent at the C8 position. Results in Tables 2 and 3
H
DOI: 10.1021/acs.jafc.8b02052
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
(6) Manach, C.; Morand, C.; Gil-Izquierdo, A.; Bouteloup-
The results in this study suggest that the interactions
between cells and the citrus PMFs and their metabolites are
complex, and at this time, details concerning solvent factors
controlling their fluorescence remain poorly understood.
Although the fully methylated PMFs and glucuronidated
metabolites without free phenolic hydroxyls show small shifts
in their emission-wavelength maxima when bound to the
Huh7.5 cells, these differences are by no means similar to the
large differences that occur in these spectra when measured in
methanol versus in aqueous methanol solutions. However, this
is not the case with metabolites with free hydroxyls, for which
there are large red shifts in their emission-wavelength maxima
when bound to the Huh7.5 cells. Curiously, these same
compounds exhibit significant losses in their fluorescence
intensities in aqueous solutions, but there are no obvious losses
in the emission intensities of these compounds when bound to
the Huh7.5 cells. These complexities are understandable in
light of the many chemical microenvironments within cells,
and the many factors yet to be considered in studying the
fluorescence of PMFs, including pH at the binding site, the
ionic interactions between the binding sites and resonance
structures of the PMFs, and the viscosity effects of compound
binding to proteins and membranes of intracellular organelles.
́
́
Demange, 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, 235−242.
(7) Robbins, R. C. Effect of methoxylated flavones on erythrocyte
aggregation and sedimentation in blood or normal subjects: evidence
of a dietary role for flavonoids. Int. J. Vitamin Nutr. Res. 1973, 43,
494−503.
(8) Robbins, R. C. Action of flavonoids on blood cells: trimodal
action of flavonoids elucidates their inconsistent physiological effects.
Int. J. Vitamin Nutr. Res. 1974, 44, 203−216.
(9) Manthey, J. A.; Guthrie, N. Antiproliferative activities of citrus
flavonoids against six human cancer cell lines. J. Agric. Food Chem.
2002, 50, 5837−5843.
(10) Manthey, J. A.; Bendele, P. Anti-inflammatory activity of an
orange peel polymethoxylated flavone, 3′,4′,3,5,6,7,8-heptamethoxy-
flavone, in the rat carrageenan/paw edema and mouse lip-
opolysaccharide-challenge assays. J. Agric. Food Chem. 2008, 56,
9399−9403.
(11) Kurowska, E. A.; Manthey, J. A.; Casaschi, A.; Theriault, A. G.
Modulation of HepG2 cell net apolipoprotein B secretion by the
citrus polymethoxyflavone, tangeretin. Lipids 2004, 39, 143−151.
(12) Gosslau, A.; Chen, K. Y.; Ho, C. T.; Li, S. Anti-inflammatory
effects of characterized orange peel extracts enriched with bioactive
polymethoxyflavones. Food Sci. and Human Wellness 2014, 3, 26−35.
(13) Huang, H.; Li, L.; Shi, W.; Liu, H.; Yang, J.; Yuan, X.; Wu, L.
The multifunctional effects of nobiletin and its metabolites in vivo and
in vitro. Evid. Based Complement. Alternat. Med. 2016, 2016, 2918798.
(14) Kim, Y. J.; Choi, M. S.; Woo, J. T.; Jeong, M. J.; Kim, S. R.;
Jung, U. J. Long-term dietary supplementation with low-dose
nobiletin ameliorates hepatic steatosis, insulin resistance, and
inflammation without altering fat mass in diet-induced obesity. Mol.
Nutr. Food Res. 2017, 61, 1600889.
(15) Wu, X.; Song, M.; Gao, Z.; Sun, Y.; Wang, M.; Li, F.; Zheng, J.;
Xiao, H. Nobiletin and its colonic metabolites suppress colitis-
associated colon carcinogenesis by down-regulating iNOS, inducing
antioxidative enzymes and arresting cell cycle progression. J. Nutr.
Biochem. 2017, 42, 17−25.
(16) Li, S.; Wang, H.; Guo, L.; Zhao, H.; Ho, C. T. Chemistry and
bioactivity of nobiletin and its metabolites. J. Funct. Foods 2014, 6, 2−
10.
(17) Zheng, J.; Song, M.; Dong, P.; Qiu, P.; Guo, S.; Zhong, Z.; Li,
S.; Ho, C. T.; Xiao, H. Identification of novel bioactive metabolites of
5-demethylnobiletin in mice. Mol. Nutr. Food Res. 2013, 57, 1999−
2007.
(18) Xiao, H.; Yang, C. S.; Li, S.; Jin, H.; Ho, C. T.; Patel, T.
Monodemethylated polymethoxyflavones from sweet orange (Citrus
sinensis) peel inhibit growth of human lung cancer cells by apoptosis.
Mol. Nutr. Food Res. 2009, 53, 398−406.
(19) Jurd, L. Spectral properties of flavonoid compounds. In The
Chemistry of Flavonoid Compounds; Geissman, T. A., Ed.; Pergamon
Press: New York, 1962.
(20) Conseil, G.; Baubichon-Cortay, H.; Dayan, G.; Jault, J.-M.;
Barron, D.; Di Pietro, A. Biochemistry Flavonoids: A class of
modulators with bifunctional interactions at vicinal ATP- and steroid-
binding sites on mouse P-glycoprotein. Proc. Natl. Acad. Sci. U. S. A.
1998, 95, 9831−9836.
(21) Jacobson, K. A.; Moro, S.; Manthey, J. A.; West, P. L.; Ji, X.-D.
Interactions of flavones and other phytochemicals with adenosine
receptors. Adv. Exp. Med. Biol. 2002, 505, 163−171.
(22) Williams, R. J.; Spencer, J. P. E.; Rice-Evans, C. Flavonoids:
antioxidants or signaling molecules? Free Radical Biol. Med. 2004, 36,
838−849.
AUTHOR INFORMATION
■
Corresponding Author
*Tel.: (772) (462) 5930. E-mail: john.manthey@ars.usda.gov.
ORCID
Danielle R. Gonca̧ lves: 0000-0003-4924-6696
John A. Manthey: 0000-0001-9122-0460
Notes
Disclaimer: This article is a U.S. Government work and is in
the public domain in the United States. Mention of a
trademark or proprietary product is for identification only
and does not imply a guarantee or warranty of the product by
the U.S. Department of Agriculture. The U.S. Department of
Agriculture prohibits discrimination in all its programs and
activities on the basis of race, color, national origin, gender,
religion, age, disability, political beliefs, sexual orientation, and
marital or family status.
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
We thank Veronica Cook for critical reading of the manuscript,
assistance in figure and legend preparations, and expertise in
conducting the fluorescence measurements and analyses by
HPLC-MS.
REFERENCES
■
(1) Swift, L. J. TLC-spectrophotometric analysis for neutral fraction
in orange peel juice. J. Agric. Food Chem. 1967, 15, 99−101.
(2) Horowitz, R. M.; Gentili, B. Flavonoid constituents of Citrus. In
Citrus Science and Technology; Nagy, S., Shaw, P.E., Veldhuis, M. K.,
Eds.; AVI Publishing Company: Westport, CT, 1977; pp 397−426.
(3) Dugo, P.; McHale, D. The oxygen heterocyclic compounds of
citrus essential oils. In Citrus. The Genus Citrus; Dugo, G., Di
Giacomo, A., Eds.; CRC Press: Boca Raton, FL, 2002; pp 355−390.
(4) Wen, X.; Walle, T. Methylated flavonoids have greatly improved
intestinal absorption and metabolic stability. Drug Metab. Dispos.
2006, 34, 1786−1792.
(23) Yue, Y.; Zhang, Y.; Li, Y.; Zhu, J.; Qin, J.; Chen, X. Interaction
of nobiletin with human serum albumin studied using optical
spectroscopy and molecular modeling methods. J. Lumin. 2008,
128, 513−520.
(24) Chen, J.; Song, M.; Wu, X.; Zheng, J.; He, L.; McClements, D.
J.; Decker, E.; Xiao, H. Direct fluorescent detection of a
(5) Wen, X.; Walle, T. Methylation protects dietary flavonoids from
rapid hepatic metabolism. Xenobiotica 2006, 36, 387−397.
I
DOI: 10.1021/acs.jafc.8b02052
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
polymethoxyflavone in cell culture and mouse tissue. J. Agric. Food
Chem. 2015, 63, 10620−10627.
(25) Ito, Y. Golden rules and pitfalls in selecting optimum
conditions for high-speed counter-current chromatography. J.
Chromatog. A 2005, 1065, 145−168.
(45) Manthey, J. A.; Grohmann, K.; Montanari, A.; Ash, K.;
Manthey, C. L. Polymethoxylated flavones derived from citrus
suppress tumor necrosis factor-α expression by human monocytes.
J. Nat. Prod. 1999, 62, 441−444.
́
(46) de Araujo, M. E. M. B.; Franco, Y. E. M.; Alberto, T. G.;
(26) Silveira, J. Q.; Cesar, T. B.; Manthey, J. A.; Baldwin, E. A.; Bai,
J.; Raithore, S. Pharmacokinetics of flavanone glycosides after
ingestion of single doses of fresh-squeezed orange juice versus
commercially processed orange juice in healthy humans. J. Agric. Food
Chem. 2014, 62, 12576−12584.
Messias, M. C. F.; Leme, C. W.; Sawaya, A. C. H. F.; Carvalho, P. O.
Kinetic study on the inhibition of xanthine oxidase by acylated
derivatives of flavonoids synthesised enzymatically. J. Enzyme Inhib.
Med. Chem. 2017, 32, 978−985.
(47) Nagao, A.; Seki, M.; Kobayashi, H. Inhibition of xanthene
oxidase by flavonoids. Biosci., Biotechnol., Biochem. 1999, 63, 1787−
1790.
(27) Mabry, T. J.; Markham, K. R.; Thomas, M. B. The Systematic
Identification of Flavonoids; Springer-Verlag: New York, 1970.
(28) Fabre, N.; Rustan, I.; de Hoffman, E.; Quetin-Leclercq, J.
Determination of flavone, flavonol, and flavanone aglycones by
negative ion liquid chromatography electrospray ion trap mass
spectrometry. J. Am. Soc. Mass Spectrom. 2001, 12, 707−715.
(29) Xing, T. T.; Zhao, X. J.; Zhang, Y. D.; Li, Y. F. Fast separation
and sensitive quantitation of polymethoxylated flavonoids in the peels
of Citrus using UPLC-Q-TOF-MS. J. Agric. Food Chem. 2017, 65,
2615−2627.
(48) Cos, P.; Ying, L.; Calomme, M.; Hu, J. P.; Cimanga, K.; Van
Poel, B.; Pieters, L.; Vlietinck, A. J.; Berghe, D. V. Structure-activity
relationship and classification of flavonoids as inhibitors of xanthine
oxidase and superoxide scavengers. J. Nat. Prod. 1998, 61, 71−76.
(49) Alexander, S. P. H. Flavonoids as antagonists at A1 adenosine
receptors. Phytother. Res. 2006, 20, 1009−1012.
(50) Avior, Y.; Bomze, D.; Ramon, O.; Nahmias, Y. Flavonoids as
dietary regulators of nuclear receptor activity. Food Funct. 2013, 4,
831−844.
(51) Sato, T.; Koike, L.; Miyata, Y.; Hirata, M.; Mimaki, Y.; Sashida,
Y.; Yano, M.; Ito, A. Inhibition of activator protein-1 binding activity
and phosphatidylinositol 3-kinase pathway by nobiletin, a poly-
methoxy flavonoid, results in augmentation of tissue inhibitor of
metalloproteinases-1 production and suppression of production of
matrix metalloproteinases-1 and −9 in human fibrosarcoma HT-1080
Cells. Cancer Res. 2002, 62, 1025−1029.
(52) Onozuka, H.; Nakajima, A.; Matsuzaki, K.; Shin, R. W.; Ogino,
K.; Saigusa, D.; Tetsu, N.; Yokosuka, A.; Sashida, Y.; Mimaki, Y.;
Yamakuni, T.; Ohizumi, Y. Nobiletin, a citrus flavonoid, improves
memory impairment and AB pathology in a transgenic mouse model
of Alzheirmer’s disease. J. Pharmacol. Exp. Ther. 2008, 326, 739−744.
(53) Baby, B.; Antony, P.; Vijayan, R. Interactions of quercetin with
receptor tyrosine kinases associated with human lung carcinoma. Nat.
Prod. Res. 2017, 1.
(54) Yamashita, Y.; Kawada, S. Z.; Nakano, H. Induction of
mammalian topoisomerase II dependent DNA cleavage by non-
intercalative flavonoids, genistein and orobol. Biochem. Pharmacol.
1990, 39, 737−744.
(55) Constantinou, A.; Mehta, R.; Runyan, C.; Rao, K.; Vaughan, A.;
Moon, R. Flavonoids as DNA topoisomerase antagonists and poisons:
structure-activity relationships. J. Nat. Prod. 1995, 58, 217−225.
(56) Lopez-Lazaro, M.; Willmore, E.; Austin, C. A. The dietary
flavonoids myricetin and fisetin act as dual inhibitors of DNA
topoisomerase I and II in cells. Mutat. Res., Genet. Toxicol. Environ.
Mutagen. 2010, 696, 41−47.
(57) Tsutsumi, R.; Yoshida, T.; Nii, Y.; Okahisa, N.; Iwata, S.;
Tsukayama, Ml.; Hashimoto, R.; Taniguchi, Y.; Sakaue, H.; Hosaka,
T.; Shuto, E.; Sakai, T. Sudachitin, a polymethoxylated flavone,
improves glucose and lipid metabolism by increasing mitochondrial
biogenesis in skeletal muscle. Nutrition and Metabolism 2014, 11, 32.
(30) Li, S.; Lo, C.-Y.; Ho, C.-T. Hydroxylated polymethoxyflavones
and methylated flavonoids in sweet orange (Citrus sinensis) peel. J.
Agric. Food Chem. 2006, 54, 4176−4185.
(31) Chen, J.; Montanari, A. M.; Widmer, W. W. Polymethoxylated
flavones, a class of compounds with potential anticancer activity,
isolated from cold pressed Dancy tangerine peel oil solids. J. Agric.
Food Chem. 1997, 45, 364−368.
(32) Rizzi, G. P.; Boeing, S. S. Mass spectral analysis of some
naturally occurring polymethoxyflavones. J. Agric. Food Chem. 1984,
32, 551−555.
(33) Berahia, T.; Gaydou, E. M.; Cerrati, C.; Wallet, J.-C. Mass
spectrometry of polymethoxylated flavones. J. Agric. Food Chem. 1994,
42, 1697−1700.
(34) Ferrari, M.; Fornasiero, M. C.; Isetta, A. M. MTT colorimetric
assay for testing macrophage cytotoxic activity in vitro. J. Immunol.
Methods 1990, 131, 165−172.
́
(35) Quijano, L.; Calderon, J. S.; Gomez G, F.; Escobar, E.; Rios, T.
Octasubstituted flavones from Ageratum houstonianum. 1985.
Phytochemistry 1985, 24, 1085−1088.
(36) Quijano, L.; Calderon, J. S.; Garibay, F. G.; Escobar, E.; Rios,
T. Further polysubstituted flavones from. Phytochemistry 1987, 26,
2075−2078.
(37) Nielsen, S. E.; Breinholt, V.; Cornett, C.; Dragsted, L. O.
Biotransformation of citrus flavone tangeretin in rats. Identification of
metabolites with intact flavane nucleus. Food Chem. Toxicol. 2000, 38,
739−746.
(38) Koga, N.; Ohta, C.; Kato, Y.; Haraguchi, K.; Endo, T.; Ogawa,
K.; Ohta, H.; Yano, M. In vitro metabolism of nobiletin, a
polymethoxy-flavonoid. Xenobiotica 2011, 41, 927−933.
(39) Sugiyama, S.; Umehara, K.; Kuroyanagi, M.; Ueno, A.; Taki, T.
Studies on the differentiation inducers of myeloid leukemic cells from
Citrus species. Chem. Pharm. Bull. 1993, 41, 714−719.
(40) Tyukavkina, N. A.; Pogodaeva, N. N.; Brodskaya, E. I.;
Sapozhnikov, Y. M. Ultraviolet absorption of flavonoids. V. The
structure of 3- and 5-hydroxyflavones. Chem. Nat. Compd. 1975, 11,
613−616.
(41) Li, S.; Pan, M. H.; Lo, C. Y.; Tan, D.; Wang, Y.; Shahidi, F.; Ho,
C. T. Chemistry and health effects of polymethoxyflavones and
hydroxylated polymethoxyflavones. J. Funct. Foods 2009, 1, 2−12.
(42) Ho, C. T.; Pan, M. H.; Lai, C. S.; Li, S. Polymethoxyflavones as
food factors for the management of inflammatory diseases. J. Food
Drug Anal 2012, 20, 337−341.
(43) Petkov, E.; Nikolov, N.; Uzunov, P. Inhibitory effect of some
flavonoids and flavonoid mixtures on cyclic AMP phosphodiesterase
activity of rat heart. Planta Med. 1981, 43, 183−186.
(44) Nikaido, T.; Ohmoto, T.; Sankawa, U.; Tomimori, T.;
Miyaichi, Y.; Imoto, Y. Inhibition of adenosine 3′,5′-cyclic mono-
phosphate phosphodiesterase by flavonoids. II. Chem. Pharm. Bull.
1988, 36, 654−661.
́
́
(58) Kameníkova, M.; Furtmuller, P. G.; Klacsova, M.; Lopez-
̈
́
̌
Guzman, A.; Toca-Herrera, J. L.; Vitkovska, A.; Devínsky, F.; Mucaji,
P.; Nagy, M. Influence of quercetin on the interaction of gliclazide
with human serum albumin - spectroscopic and docking approaches.
Luminescence 2017, 32, 1203−1211.
(59) Azuma, K.; Suzuki, S.; Uchiyama, S.; Kajiro, T.; Santa, T.; Imai,
K. A study of the relationship between the chemical structures and the
fluorescence quantum yields of coumarins, quinoxalinones and
benzoxazinones for the development of sensitive fluorescent
derivatization reagents. Photochem. Photobiol. Sci. 2003, 2, 443−449.
(60) Kumar, S.; Jain, S. K.; Rastogi, R. An experimental and
theoretical study of excited-state dipole moments of some flavones
using an efficient solvatochromic method based on the solvent
polarity parameter, ETN. Spectrochim. Acta, Part A 2001, 57, 291−298.
(61) Wolfbeis, O. S. W.; Begum, M.; Geiger, H. Fluorescence
properties of hydroxy- and methoxyflavones and the effect of shift
reagents. Z. Naturforsch., B: J. Chem. Sci. 1984, 39b, 231−237.
J
DOI: 10.1021/acs.jafc.8b02052
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
(62) Sengupta, P. K.; Kasha, M. Excited state proton-transfer
spectroscopy of 3-hydroxyflavone and quercetin. Chem. Phys. Lett.
1979, 68, 382−385.
(63) Schipfer, R.; Wolfbeis, O. S.; Knierzinger, A. pH-Dependent
fluorescence spectroscopy. Part 12. Flavone, 7-hydroxyflavone, and 7-
methoxyflavone. J. Chem. Soc., Perkin Trans. 2 1981, 1443−1448.
(64) Chaudhuri, S.; Sengupta, B.; Taylor, J.; Pahari, B. P.; Sengupta,
P. K. Interactions of dietary flavonoids with proteins: Insights from
fluorescence spectroscopy and other related biophysical studies. Curr.
Drug Metab. 2013, 14, 491−503.
(65) Deriglazov, N. M.; Kalistratova, E. F.; Tyukavkina, N. A.
Polarographic study of flavonoids. Flavone, flavonol, and 3-
chloroflavone. Chem. Heterocycl. Compd. 1978, 14, 821−824.
(66) Dreyer, D. L. Chemotaxonomy of the Rutaceae. IV.
Constituents of Murraya paniculata (Linn.) Jack. J. Org. Chem.
1968, 33, 3574−3576.
(67) Dangles, O.; Dufour, C.; Manach, C.; Morand, C.; Remesy, C.
Binding of flavonoids to plasma proteins. Methods Enzymol. 2001,
335, 319−333.
(68) Dangles, O.; Dufour, C.; Bret, S. Flavonol-serum albumin
complexation. Two-electron oxidation of flavonols and their
complexes with serum albumin. J. Chem. Soc., Perkin Trans. 2 1999,
2, 737−744.
(69) Park, H. R.; Daun, Y.; Park, J. K.; Bark, K.-M. Spectroscopic
Properties of Flavonoids in Various Aqueous-Organic Solvent
Mixtures. Bull. Korean Chem. Soc. 2013, 34, 211−220.
(70) Tyukavkina, N. A.; Pogodaeva, N. N.; Brodskaya, E. I.;
Sapozhnikov, Y. M. Ultraviolet absorption of flavonoids. V. The
structure of 3- and 5-hydroxyflavones. Chem. Nat. Compd. 1975, 11,
613−616.
(71) Vakarelska-Popovska, M. H.; Velkov, Z. A. Structure of flavones
and flavonols. Part II: Role of position on the O-H bond dissociation
enthalpy. Computational Chem. 2014, 2, 1−5.
(72) Gutzeit, H. O.; Tokalov, S. V.; Ludwig-Muller, J.; Rusak, G.
̈
Monitoring flavonoid metabolism in human cells by exploiting
fluorescence elicited upon quercetin/protein interactions. Croatica
Chemica Acta. 2005, 78, 337−342.
(73) Baran, I.; Ganea, C.; Ursu, I.; Baran, V.; Calinescu, O.; Iftime,
A.; Ungureanu, R.; Tofolean, I. T. Fluorescence properties of
quercetin in human leukemia Jurkat T-cells. Rom. J. Phys. 2011, 56,
388−398.
(74) Mukai, R.; Terao, J.; Shirai, Y.; Saito, N.; Ashida, H.
Determination of subcellular localization of flavonol in cultured
cells by laser scanning. In Laser Scanning, Theory and Applications;
Wang, C.-C., Ed.; IntechOpen: London, 2011. Available at http://
www.intechopen.com/books/laser-scanning-theory-and-applications/
determination-of-subcellularlocalization-of-flavonol-in-cultured-cells-
by-laser-scanning.
(75) Chen, J.; Song, M.; Wu, X.; Zheng, J.; He, L.; McClements, D.
J.; Decker, E.; Xiao, H. Direct fluorescence detection of a
polymethoxyflavone in cell culture and mouse tissue. J. Agric. Food
Chem. 2015, 63, 10620−10627.
(76) Buer, C. S.; Muday, G. K.; Djordjevic, M. A. Flavonoids are
differentially taken up and transported long distances in. Plant Physiol.
2007, 145, 478−490.
(77) Lee, J. H.; Kim, Y.; Hoang, M. H.; Jun, H. J.; Lee, S. J. Rapid
quantification of cellular flavonoid levels using quercetin and a
fluorescent diphenylboric acid 2-amino ethyl ester probe. Food Sci.
Biotechnol. 2014, 23, 75−79.
K
DOI: 10.1021/acs.jafc.8b02052
J. Agric. Food Chem. XXXX, XXX, XXX−XXX