See

discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/44620467

Quantification of Quercetin Glycosides in 6

Onion Cultivars and Comparisons of
Hydrolysis-HPLC and Spectrophotometric
Methods in Measuring Total Quercetin
Concentrations

Article in Journal of Food Science March 2010

DOI: 10.1111/j.1750-3841.2009.01469.x Source: PubMed

CITATIONS READS

14 142

3 authors:

Kil Sun Yoo Eun Jin Lee

Texas A&M University Seoul National University
98 PUBLICATIONS 1,186 CITATIONS 45 PUBLICATIONS 208 CITATIONS

SEE PROFILE SEE PROFILE

Bhimanagouda S Patil
Texas A&M University
297 PUBLICATIONS 4,412 CITATIONS

SEE PROFILE

All in-text references underlined in blue are linked to publications on ResearchGate, Available from: Eun Jin Lee
letting you access and read them immediately. Retrieved on: 18 August 2016
 JFS C: Food Chemistry

Quantification of Quercetin Glycosides in 6 Onion

Cultivars and Comparisons of Hydrolysis-HPLC
C: Food Chemistry

and Spectrophotometric Methods in Measuring

Total Quercetin Concentrations
KIL SUN YOO, EUN JIN LEE, AND BHIMANAGOUDA S. PATIL

ABSTRACT: This study was performed to purify and quantify quercetin glycosides (QG) and aglycone (free)
quercetin (Q) in 6 selected onion cultivars and to compare analytical approaches based on high-performance liq-
uid chromatography (HPLC) and spectrophotometry for the quantification of total quercetin (TQ) concentrations.
Individual mono- and di-glycoside Q compounds were purified using a semipreparative HPLC and identified by
comparing spectral data and by confirming corresponding peaks of QG and Q after incomplete enzyme-hydrolysis.
Purified QG were quantified as Q by enzyme-hydrolysis/HPLC. TQ concentrations obtained from 20 onion
bulbs with enzyme-hydrolysis/HPLC, no-hydrolysis/HPLC, and a spectrophotometric method without prior hy-
drolysis were significantly correlated (r 2 = 0.99) and were about 15% higher, identical, or 10% less than those con-
centrations by a standard acid-hydrolysis/HPLC method, respectively. During enzyme-hydrolysis of onion extracts,
progressive reduction of the QG and formation of the corresponding mono-glycosides and Q were monitored us-
ing an analytical HPLC. TQ ranged from 83 to 330 g/g F.W. in 6 selected cultivars of long-day or short-day onions.
Q3,4 G and Q4 G were the 2 major compounds and comprised approximately between 94% and 97% of TQ in onions.
Keywords: Allium cepa L., analysis, flavonoids, glycosides, HPLC

Introduction and others 2009), and spectrophotometry (Trammell and Peterson

uercetin (3, 3 , 4 , 5, 7-pentahydroxyflavone; Q) is a major
Q flavonoid found in onions, where it occurs primarily in the
form of quercetin glycoside (QG) and minor in aglycone (free)
quercetin (Q) (Bilyk and Sapers 1985, 1986; Price and Rhodes 1997;
Wach and others 2007; Grzelak and others 2009). In a recent sur-
vey of 28 vegetables and 9 fruits, the onion ranked highest in Q
compounds (Hertog and others 1992; Galdon and others 2008).
Q compounds are related to skin color and disease resistance
(Brandwein 1965; Trammell and Peterson 1976) in plant. The struc-
tural genes regulating Q and anthocyanin compounds in various
colored onions were characterized by our group (Kim and others
2004).
There is increasing interest in Q compounds because of possi-
ble human health benefits including inhibition of growth of several
kinds of cancer cells (Hosokawa and others 1990; Hertog and others
1992; Chu and others 2000), Qs action as an antioxidant, chelating
agent (Torel and others 1986), and free radical scavenger (Hemeda
and Klein 1990; Murota and Terao 2003; Clifton 2004). The agly-
cone Q has greater pharmacological activity than the QG and the
fecalases produced in the human gut are capable of hydrolyzing QG
to yield the Q (Leighton and others 1992).
Various analytical methods, such as thin layer chromatography
and high-performance liquid chromatography (HPLC) (Koeppen
and Van Der Spuy 1961; Bilyk and Sapers 1986; Leighton and others
1992; Bonaccorsi and others 2008; Galdon and others 2008; Grzelak
MS 20090779 Submitted 8/12/2009, Accepted 11/3/2009. Authors are with
Vegetable & Fruit Improvement Center, Dept. of Horticultural Science, Texas
A&M Univ., College Station, TX 77843, U.S.A. Direct inquiries to author Yoo
(E-mail: k-yoo@tamu.edu).
1976; Leighton and others 1992) have been utilized for the quantifi-
cation of Q compounds.
Although the majority of Q compounds in onions are in the
mono- and di-glycoside forms, individual QG have not been quan-
tified due to the lack of appropriate commercially available stan-
dard compounds. Consequently, most previous studies reported
only total quercetin (TQ) concentrations after acid-hydrolysis of
QG. Some degradation of the QG occurs during acid-hydrolysis and
the recovery rate varies from 70% to 110% based on the conditions
of hydrolysis (Hertog and others 1992).
Previously, we have surveyed concentrations of Q and TQ, but
not of individual QG in different colored onions (Patil and others
1995b). Measurement of TQ is considered valid for most purposes,
but the quantity of individual, naturally-occurring QG is also of
interest.
In this study, we purified each QG from onion extracts and
established an HPLC method to quantify them. We have mon-
itored HPLC peak profile changes of individual QG during
enzyme-hydrolysis and also compared acid-hydrolysis/HPLC with
enzyme-hydrolysis/HPLC, no-hydrolysis/HPLC, and spectropho-
tometric methods in measuring TQ concentrations. Using the pu-
rified QG and the established HPLC method, we have measured
individual QG concentrations from 6 selected long- and short-day
types of onion cultivars.
Materials and Methods
Chemical reagents
All chemicals and solvents were purchased from Sigma-Aldrich
(St. Louis, Mo., U.S.A.) or Fisher Scientific (Pittsburg, Pa., U.S.A.).

C160 JOURNAL OF FOOD SCIENCEVol. 75, Nr. 2, 2010 

C 2010 Institute of Food Technologists
R

doi: 10.1111/j.1750-3841.2009.01469.x
Further reproduction without permission is prohibited
Onion quercetin quantitative measurement . . .

Aglycone Q (3, 3 , 4 , 5, 7-pentahydroxyflavone) standard and - 18 min. Total of 200 mL of extract were injected and detection was
glycosidase enzyme were purchased from Sigma-Aldrich. Amberite monitored at 365 nm.
IRC-50 (H+ ) ion exchange resin was obtained from Supelco (Belle- Analytical HPLC. The analytical HPLC system was consisted of
fonte, Pa., U.S.A.). a Perkin Elmer Model 200 pump, an autosampler, an Econosil C-18
analytical column (4.6 250 mm, Alltec Associates) with a guard
Plant materials cartridge, and LC-295 UV/Vis detector. Solvents A and B were H2 O
A red (TX 90977), a white (Texas Early White), and five yel- and MeOH, respectively, containing 0.5% H3 PO4 . The solvent gra-

C: Food Chemistry
low (TG 1015Y, TX 50307, Sweet Sandwich, Fortress, and dient was from 40% to 90% solvent B in 18 min and was maintained
Vega) onion cultivars were used. The short-day type TX 90977, at 90% solvent B for 5 min at a flow rate of 1 mL/min. A 20 L sam-
TG 1015Y, TX 50307, and Texas Early White onions were ob- ple was injected and detection was carried out at 374 nm. Q com-
tained from our breeding lines, harvested in April in Texas, and pounds dissolved in MeOH ranging between 0 and 68 g/mL were
stored at 5 C until October. The long-day type Sweet Sandwich, used as an external standard to quantify the samples. In the anal-
Fortress, and Vega cultivars were from Seminis Seeds (St. Louis, ysis using no-hydrolysis/HPLC, Q concentrations of individual QG
Mo., U.S.A.) and were harvested in October in Michigan. were summed to calculate the TQ concentrations.

Purification of QG Acid- and enzyme-hydrolysis

About 1 kg of TG 1015Y onion tissue was peeled and the neck
and basal part were removed. The tissue was homogenized for
5 min in 2 L of 80% EtOH using a home blender. The homogenate
was filtered through Whatman nr 1 filter paper and the extract was
concentrated to 200 mL in a rotary evaporator at 40 C. The QG
were collected by passing the concentrated extract through a 5
40 cm column of Amberite IRC-50 (H+ ) ion exchange resin
(Brandwein 1965; Trammell and Peterson 1976; Calvo and others
1997). The column was washed with H2 O, and the QG were eluted
with 2 L of 95% EtOH and concentrated to 50 mL in the rotary
evaporator.
Individual QG were purified using a semipreparative HPLC de-
scribed subsequently in the HPLC section. Peak fractions of column
eluate were repeatedly collected, combined, and concentrated to
dryness under vacuum. Approximately 10 to 100 mg of QG were col-
lected. Spectral data of each peak were also collected for the iden-
tification purpose. Purity of each QG was estimated by calculating
% of the peak area from the total peak areas in the analytical HPLC
listed below in the HPLC section.
Identification and quantification of QG
Purified QG were spotted and separated on Whatman nr 1 fil-
ter paper (2.5 45 cm) using the solvent H2 O : CH3 COOH (85 : 15,
vol/vol). The retention factor (Rf ) value of each spot was deter-
mined and confirmed by comparison to previous data (Brandwein
1965; Mabry and others 1970). Individual QG were also dissolved
in MeOH and analyzed by the analytical HPLC method and ab-
sorption spectra were collected and compared with previously
mentioned data. The purified Q3,4 G was hydrolyzed incompletely
to confirm their corresponding mono-glycosides, as described in
incomplete-hydrolysis section subsequently.
For quantification of QG, paired series of purified QG solutions
in MeOH and 1 mL of enzyme-hydrolyzed QG, as described in
enzyme-hydrolysis subsequently, were analyzed by the analytical
HPLC. The peak areas of the QG standards could be calibrated to
Q concentrations by measuring the Q concentrations in the hy-
drolyzed samples. For quantification of Q concentrations, the ex-
ternal standards ranging between 0 and 68 g/mL were used.
Semi-preparative and analytical HPLC conditions
Semi-preparative HPLC. Q compounds in onion extracts were
isolated using a semi-preparative HPLC system, which was con-
sisted of a pump (Model 200, Perkin-Elmer, Norwalk, Conn., U.S.A.)
and an LC-235 diode array detector (wavelength range 195 to 365
nm), a manual injector, and a preparative column (Econosil C-18
column, 10 250 mm, Alltech Associates, Deerfield, Ill., U.S.A.) at
a flow rate of 4 mL/min with a gradient of 30% to 80% MeOH in
Acid-hydrolysis. One milliliter of EtOH extract of onions was
mixed with 1 mL 4N HCl and incubated at 90 C for 60 min. Total
of 200 L of concentrated NH4 OH was added and the hydrolysate
was dried under vacuum. The dried sample was then dissolved in
1 mL of MeOH, and analyzed by the analytical HPLC system.
Enzyme-hydrolysis. One milliliter of purified Q compounds or
EtOH extract of onions was dried under vacuum at 40 C. One
milliliter of -glycosidase solution (2 mg/mL) was added and the
solution was kept at 40 C for 3 h to achieve complete hydrolysis.
The hydrolysate was dried under vacuum at 40 C, suspended in
1 mL of 80% MeOH, and analyzed by the analytical HPLC system.
Completion of hydrolysis was confirmed by disappearance of QG
peaks in the chromatogram.
Incomplete enzyme-hydrolysis. The purified Q3,4 G was dis-
solved in MeOH, dried, mixed with the enzyme, and kept at 40 C
for 1 h to achieve incomplete hydrolysis.
Monitoring enzyme-hydrolysis profile over time. Changes in
the levels of individual QG and Q in the samples during enzyme-
hydrolysis were monitored over time. Total of 100 mL of EtOH ex-
tract from TG 1015Y onions was concentrated to 20 mL by the
rotary evaporator. One milliliter of extract was mixed with 1 mL of
-glycosidase solution and incubated at 40 C for 1, 3, 5, 10, 20, 30,
40, 50, and 60 min. After the incubation samples were placed on ice
to stop the enzyme activity. Samples were then freeze-dried, dis-
solved in MeOH, and analyzed by the analytical HPLC.
Peak areas of each compound were plotted to monitor the hy-
drolysis profile over time. In this study, the progress of hydrolysis
of onion extract was of interest and actual quantification was not
attempted.
Estimation of TQ by spectrophotometry
Absorbance of the crude EtOH extracts of onions was measured
at 374 nm. Samples with absorbance readings over 1.5 were di-
luted with 80% EtOH. The concentration of TQ in samples was
calculated using the standard curve of Q in 80% MeOH, ranging be-
tween 0 and 17 g/mL.
Comparison of TQ concentration
by analytical methods
Eighteen bulbs of TG 1015Y (yellow) and 2 bulbs of Texas
Early White (white) were used. A 2-cm thick cross section was cut
from the equator of the bulbs and a 20 g wedge of tissue was re-
moved. The sample was placed in a plastic bag and frozen at 20 C
until analysis. The frozen sample was homogenized in 80 mL of
80% EtOH for 2 min and filtered through P8 filter paper (Fisher
Scientific).

Vol. 75, Nr. 2, 2010JOURNAL OF FOOD SCIENCE C161

 Onion quercetin quantitative measurement . . .
TQ concentrations of the onion extracts were estimated
by 4 methods: (i) acid-hydrolysis followed by HPLC (acid-
hydrolysis/HPLC); (ii) enzyme-hydrolysis followed by HPLC
(enzyme-hydrolysis/HPLC); (iii) HPLC without prior hydrolysis
(no-hydrolysis/HPLC); and (iv) spectrophotometry without prior
hydrolysis (spectrophotometry).
Regression analysis between TQ concentrations measured by
C: Food Chemistry
acid-hydrolysis/HPLC method (i), and the other 3 methods (ii, iii,
and iv) was performed by the SAS general linear model (Cary, N.C.,
U.S.A.).
Quantification of Q compounds in 6 onion cultivars
A red (TX 90977) and 5 yellow (TG 1015Y, TX 50307, Sweet
Sandwich, Fortress, and Vega) onion cultivars were used. A
Figure 1 --- The absorption spectra of each purified
quercetin compound from a yellow onion (TG 1015Y)
by a semi-preparative HPLC. The spectra are in the same
scale and are stacked to show the different patterns. Ab-
breviations for each compound are indicated in the text.
C162 JOURNAL OF FOOD SCIENCEVol. 75, Nr. 2, 2010
20 g of EtOH extract was prepared as the previous section and the
homogenate was filtered through a 0.5 m nylon filter and analyzed
by the analytical HPLC condition. Total of 5 replicate bulbs per cul-
tivar were analyzed.
The concentrations of QG in samples were calculated using the
enzyme-hydrolyzed QG standards and expressed as Q concentra-
tions. Means, SD, relative portion (%) of each QG, and coefficient
of variation for the TQ concentrations were calculated for each
cultivar.
Results and Discussion
Purification and identification of individual QG
We could successfully obtain 7 purified Q compounds from a yel-
low onion extract (TG 1015Y) using ion exchange column chro-
matography and a semipreparative HPLC column. The absorption
spectra of each purified Q compound in our study are presented in
Figure 1.
Three major (peaks 2, 4, and 5) and 3 minor (peaks 1, 3, and
6) Q compounds were separated from onion EtOH extracts by
an analytical HPLC (Figure 2A). After comparison of the Rf val-
ues and absorption spectra of the each purified compound, peaks
identified were peak 1: quercetin-4 ,7-diglycoside (Q4 ,7G); peak 2:
quercetin-3,4 -di-glycoside (Q3,4 G); peak 3: quercetin-3-glycoside
(Q3G); peak 4: quercetin-4 -glycoside (Q4 G); peak 5: isorhamnetin-
4 -glycoside (IR-G); and peak 6: Q.
Chromatograms of two major purified compounds, Q3,4 G (peak
2) and Q4G (peak 4), are presented in Figure 2B and 2C, re-
spectively. The purity of all these isolated compounds was shown
to be over 95% by peak areas. Identification of each compound
was also supported by incomplete enzyme-hydrolysis of Q3,4 G
(peak 2), which produced Q3G (peak 3), Q4 G (peak 4) as in-
termediate and Q (peak 6) as final products of the hydrolysis
(Figure 2D).
The IR-G (peak 5) has been misidentified as kaempferol in some
early reports (Bilyk and others 1984; Bilyk and Sapers 1985) but later
identified as IR-G (Park and Lee 1996). We have confirmed the IR
peak by comparing retention time and absorption spectra of puri-
fied and hydrolyzed samples (Figure 1).
Figure 2 --- Chromatograms of (A) 6
quercetin compounds in EtOH extract
of a yellow onion (TG 1015Y); (B)
purified Q3,4 G; (C) purified Q4 G; and
(D) incompletely hydrolyzed Q3,4 G by
-glycosidase and its hydrolysis
products. Abbreviations for each
compound are indicated in the text.
Onion quercetin quantitative measurement . . .

Changes in Q compounds profiles
during enzyme-hydrolysis
Figure 3 shows the changes of HPLC chromatograms of Q com-
pounds in onion EtOH extracts during enzyme-hydrolysis at 0 (be-
fore hydrolysis), 30, and 60 min. Reduction in peak sizes of Q3,4 G
(peak 2) and Q3G (peak 3), as accompanied by increases of Q (peak
6) and Q7G (peak 7), was clearly observed. Q7G (peak 7) and IR
pounds, Q3,4 G, Q4 G, and Q, are shown in Figure 4A, while the mi-
nor compounds are presented in Figure 4B in a 1/12 scale.
The general pattern was that the di-glycoside compounds,
Q3,4 G and Q4 7G, decreased continuously during enzyme-
hydrolysis. However, the mono-glycoside compounds, Q3G and
Q4 G, initially increased for about 10 min and decreased there-
after. The Q7G peaks showed a slowly increasing trend. Peak

(peak 8), not commonly seen in the diluted samples, were observed
in this study where the concentrated onion extracts were used.
Dynamic changes in the levels of individual Q compounds dur-
ing enzyme-hydrolysis are presented in Figure 4. The major com-
C: Food Chemistry
areas of IR-G decreased linearly, while the IR continuously in-
creased. The concentration of Q continuously increased during the
60 min enzyme-hydrolysis. These results confirmed the hypothe-
sis that QG are initially hydrolyzed to their 2 mono-glycosides and,
eventually, to Q. The hydrolysis reaction was not complete after
60 min, as indicated by the Q4 G and Q3G remaining; therefore, we
allowed 3 h for complete hydrolysis in the analysis of TQ content.

Comparison of analytical methods in

Figure 3 --- Chromatograms for monitoring the changes of
quercetin compound peak profiles in concentrated EtOH
extract of a yellow onion (TG 1015Y) during enzyme-
hydrolysis: (A) 6 quercetin compounds of onion before
hydrolysis; (B) enzyme-hydrolysis of the sample after
30 min; and (C) after 60 min. Abbreviations for each com-
pound are indicated in the text.
measuring TQ concentrations
Figure 5 shows the correlation analysis among TQ concen-
trations in onion EtOH extracts determined by the 4 analytical
methods. The acid-hydrolysis/HPLC method was regarded as a
standard method (x-axis) in this study, because it has been
widely used. The enzyme-hydrolysis/HPLC method generated
concentration values that were about 15% higher than the acid-
hydrolysis/HPLC method. The method employing HPLC with-
out prior hydrolysis produced TQ concentrations similar to those
obtained using the acid-hydrolysis/HPLC method. TQ concentra-
tions measured by spectrophotometric method without hydrol-
ysis were lower by about 10% than those generated using the
acid-hydrolysis/HPLC method.
Since the hydrolysis reaction is done with strong acid (about 1
to 2 N HCl) and high temperature (90 C), some degree of break-
down of the Q compounds was thought to be possible, as the
recovery rates varied by the conditions of hydrolysis (Hertog and
others 1992). On the other hand, the enzyme-hydrolysis reaction
takes place under conditions of moderate temperature (40 C) and
neutral pH and generated the highest responses. The underestima-
tion by the spectrophotometric method was probably the result of
measuring the absorbance of the onion EtOH extract at 374 nm,
the apex of Q, which differs by 17 nm from the apex of the QG at
357 nm (Mabry and others 1970).
We evaluated that all method tested produced highly corre-
lated responses (r 2 = 0.99) to the standard acid-hydrolysis/HPLC
method and were acceptable to measure the TQ concentrations
accurately. The acid-hydrolysis/HPLC method was simple and
commonly used but the recovery rate must be tested. The enzyme-
hydrolysis/HPLC measured the TQ with highest levels but needs
extra process to react with the enzyme. No-hydrolysis/HPLC

12000 1000 Figure 4 --- Changes in HPLC peak

A B areas of (A) major and (B) minor
quercetin compounds in EtOH extract
10000 800 of yellow onion (TG 1015Y) during
HPLC peak area

enzyme-hydrolysis. Abbreviations for

8000 each compound are indicated in the
600 Q4',7G
Q7G text.
6000 Q3G
Q3,4'G IR-G
400
4000 Q4'G IR
Q
2000 200

0 0
0 10 20 30 40 50 60 0 10 20 30 40 50 60
Enzyme-hydrolysis time (min) Enzyme-hydrolysis time (min)

Vol. 75, Nr. 2, 2010JOURNAL OF FOOD SCIENCE C163

 Onion quercetin quantitative measurement . . .

Table 1 --- Means, standard deviation, relative portions of quercetin glycosides and aglycone (free) quercetin, and
coefficient of variation (C.V.) in EtOH extracts of 6 selected short- and long-day onion cultivars by an analytical
HPLC.
Mean concentration (g/g F.W.)
Cultivar (color) Q4 ,7Gz Q3,4 G Q3G Q4 G Q Total C.V.x (%)
Short-day
TX 90977 (red) 0.7 0.2y 46.1 6.5 2.5 0.2 49.8 7.5 2.2 0.6 101.2 13.4 13.2
C: Food Chemistry

TG 1015Y (yellow) 0.6 0.6 50.9 38.2 2.6 0.9 37.2 29.7 1.9 0.7 93.2 69.7 74.8
TX 50307 (yellow) 0.4 0.2 40.4 21.4 2.5 0.7 38.1 19.2 1.8 0.3 83.2 40.5 48.7
Long-day
Sweet Sandwich (yellow) 1.1 0.4 93.2 40.6 1.5 0.2 44.0 15.4 2.4 0.6 142.2 55.5 39.0
Fortress (yellow) 2.8 1.2 186.9 77.1 2.2 0.1 137.7 65.2 3.1 0.7 332.8 140.8 42.3
Vega (yellow) 1.7 0.7 79.6 35.6 2.9 1.1 87.6 65.2 5.8 1.4 177.6 140.8 34.4
Relative portion (%) 0.7 0.1 52.6 7.9 2.0 1.0 42.8 7.0 2.1 0.8 100
z
Abbreviations for each compound are indicated in the text.
y
Standard deviation (n = 5).
x
C.V.s (coefficient of variation) were calculated from the total concentrations.

method has an advantage to measure the individual QG but the
standard QG compounds need to be prepared. The spectropho-
tometric method is very simple but may underestimate the TQ
concentrations. If rapid screening of many samples is required
in an onion breeding program, it might be useful with proper
calibration.
Quantification of Q compounds in 6 onion cultivars
TQ concentrations ranged between approximately 80
and 330 g/g F.W. in 6 onion cultivars tested (Table 1). The
Q3,4 G and Q4 G were the 2 major compounds and comprised
approximately between 94% and 97% of TQ in onions tested. The
long-day type onions contained higher concentrations of TQ than
the short-day type onions in this study.
250
Enzyme-hydrolysis/ HPLC
No-hydrolysis/ HPLC
TQ concentration (g/g F.W.)
TQ concentrations in this study were in the lower range of
concentrations reported previously for long-day onions, which
contained TQ over 1000 mg/kg F.W. (Trammell and Peterson 1976;
Price and Rhodes 1997; Vagen and Slimestad 2008) and were similar
to those of the short-day onions (Leighton and others 1992). Bonac-
corsi and others (2008) reported that both Q3,4 G (288 to 368 mg/kg
F.W.) and Q4 G (221 to 312 mg/kg F.W.) were the major compounds
in red and gold onions, while the white onion contained no Q
compounds.
Trammell and Peterson (1976) reported that lighter yellow bulb
color was inversely correlated with TQ concentration. Thus, a
higher concentration of TQ in the long-day onions was thought
to be related to their brown color, in contrast to the light yellow
color of short-day onion bulbs in this study. Various colored onions
were known to contain different levels of Q compounds. Red
onions have greater amounts of Q compounds than yellow; yellow
onions possess greater amounts than white bulbs (Trammell and
Peterson 1976; Bilyk and others 1984; Patil and others 1995b; Lee
and others 2008; Vagen and Slimestad 2008).
There was also wide variation in TQ concentrations within cul-

200 Spectrophotometry tivars, as indicated by coefficient of variation higher than 40%.

Great variation in TQ among the short-day onion cultivars has been
previously reported (Patil and others 1995a, 1995b). High varia-
150 tion within cultivars is not surprising because there have been no
known efforts to breed onions with uniform TQ concentrations
and such variation would give an opportunity to develop for new
100 onion cultivars with high levels of Q compounds for more health
benefits.

50 Conclusions

0
0 50 100
TQ concentration (g/g F.W.) by
acid-hydrolysis/ HPLC
Figure 5 --- Regression analysis of total quercetin con-
centrations in EtOH extracts of onions between acid-
hydrolysis/HPLC (x-axis) and enzyme-hydrolysis/HPLC
(), no-hydrolysis/HPLC ( ), and spectrophotometry ().
Regression different for enzyme-hydrolysis/HPLC, y =
1.14x + 0.64, r 2 = 0.99 , no-hydrolysis/HPLC, y = 0.99x +
0.17, r 2 = 0.99 , and spectrophotometer, y = 0.89x + 0.77,
r 2 = 0.99 . N = 20 (18 bulbs TG 1015Y yellow and 2
bulbs Texas Early White red onion cultivars). indicates
significant at P 0.01.
W e have purified, identified individual Q compounds from yel-
low onions, and quantified each Q compound in 6 selected
onion cultivars. We confirmed that Q3,4 G and Q4 G were the major
150 200 250 Q compounds in the selected onion cultivars and that high C.V. ex-
isted in TQ in onion cultivars, indicating that there were great vari-
ations among individual bulbs within cultivars.
We also compared TQ concentrations measured using 3 HPLC-
based methods with hydrolysis or without hydrolysis and a spec-
trophotometric method and all methods have been proved to
measure TQ accurately. The enzyme-hydrolysis/HPLC method
yielded the highest TQ concentrations but required extra process-
ing. The HPLC analysis using EtOH extracts accurately measured
the concentration of individual QG and the TQ content. However,
preparing the individual QG standard is considered as too cumber-
some in many laboratories. The spectrophotometric method has

C164 JOURNAL OF FOOD SCIENCEVol. 75, Nr. 2, 2010

Onion quercetin quantitative measurement . . .
limitation but could be useful in a special application requiring fast
processing of many samples, such as onion breeding programs.
We demonstrated changes in the levels of individual Q com-
pounds during enzyme-hydrolysis. Degradation of mono- and di-
glycoside Q compounds and accumulation of Q were monitored.
Our study will serve as a useful resource for selecting the appro-
priate method for analysis of Q compounds, with an emphasis on
either accuracy or speed, in onions and other crops.
Acknowledgments
This study was based upon research supported by the Coopera-
tive State Research, Education, and Extension Service, U.S. Dept.
of Agriculture, under agreement nr 2006-34402-17121, Designing
Foods for Health through the Vegetable & Fruit Improvement Cen-
ter, Texas AgriLife Research.
References
Bilyk A, Sapers GM. 1985. Distribution of quercetin and kaempferol in lettuce, kale,
chive, garlic chive, leek, horseradish, red radish, and red cabbage tissues. J Agric
Food Chem 33:2268.
Bilyk A, Sapers GM. 1986. Varietal differences in the quercetin, kaempferol, and
myricetin contents of highbush blueberry, cranberry, and thornless blackberry
fruits. J Agric Food Chem 34:5858.
Bilyk A, Cooper PL, Sapers GM. 1984. Varietal differences in distribution of quercetin
and kaempferol in onion (Allium cepa L.) tissue. J Agric Food Chem 32:2746.
Bonaccorsi P, Caristi C, Gargiulli C, Leuzzi U. 2008. Flavonol glucosides in Allium
species: a comparative study by means of HPLC-DAD-ESI-MS-MS. Food Chem
107:166873.
Brandwein B. 1965. The pigments in three cultivars of the common onion (Allium
cepa). J Food Sci 30:6805.
Calvo MI, San Julian A, Fernandez M. 1997. Identification of the major compounds in
extracts of Verbena officinalis L. (Verbenaceae) by HPLC with post-column deriva-
tization. Chromatographia 46:2414.
Chu YH, Chang CI, Hsu HF. 2000. Flavonoids content of several vegetables and their
antioxidant activity. J Sci Food Agric 80:5616.
Clifton PM. 2004. Effect of grape seed extract and quercetin on cardiovascular and
endothelial parameters in high-risk subjects. J Biomed Biotechnol 5:2728.
Galdon BR, Rodrguez EMR, Romero CD. 2008. Flavonoids in onion cultivars (Allium
cepa L.). J Food Sci 73:599605.
Grzelak K, Milala J, Krol B, Adamicki F, Badelek E. 2009. Content of quercetin glyco-
sides and fructooligosaccharides in onion stored in a cold room. Euro Food Res
Technol 228:10017.
Hemeda HM, Klein BP. 1990. Effect of naturally occurring antioxidants on peroxidase
activity of vegetable extracts. J Agric Food Chem 55:1845.
Hertog MGL, Hollman PCH, Katan MB. 1992. Content of potentially anticarcinogenic
flavonoids of 28 and 9 fruit commonly consumed in the Netherlands. J Agric Food
Chem 40:237883.
Hosokawa N, Hosokawa Y, Sakai T, Yoshida M, Marui N, Nishino H. 1990. Inhibitory
effect of quercetin on the synthesis of possibly cell-cycle related 27-k Da protein in
C: Food Chemistry
human colon cancer cells. Int J Cancer 45:111924.
Koeppen BH, Van Der Spuy JE. 1961. Microbial hydrolysis of quercetin glycosides from
the inner scales of the onion (Allium cepa L.). S Afr J Agric Sci 4:55763.
Lee SU, Lee JH, Choi SH, Lee JS, Ohnisi-Kameyama M, Kozukue N, Levin CE, Fried-
man M. 2008. Flavonoid content in fresh, home-processed, and light-exposed
onions and in dehydrated commercial onion products. J Agric Food Chem
56:85418.
Leighton T, Ginther C, Fluss L, Harter WK, Cansado J, Notario V. 1992. Molecular char-
acterization of quercetin and quercetin glycosides in Allium vegetables. ACS Sym
Series 507:22038.
Kim S, Binzel ML, Park SH, Yoo KS, Pike LM. 2004. Inactivation of DFR (Dihy-
droflavonol 4-reductase) gene transcription results in blockage of anthocyanin pro-
duction in yellow onions (Allium cepa). Molecular Breeding 14:25363.
Mabry TJ, Markham KR, Thomas MB. 1970. The systemic identification of flavonoids.
New York: Springer-Verlag.
Murota K, Terao J. 2003. Atioxidant flavonoids quercetin: implication of its intestinal
absorption and metabolism. Arch Biochem Biophys 417:127.
Park Y, Lee CY. 1996. Identification of isorhamnetin 4 -glucoside in onions. J Agric
Food Chem 44:346.
Patil BS, Pike LM, Hamilton BK. 1995a. Changes in quercetin concentration in
onion (Allium cepa L.) owing to location, growth stage and soil type. New Phytol
130:34955.
Patil BS, Pike LM, Yoo KS. 1995b. Variation in the quercetin content in different col-
ored onions (Allium cepa L.). J Am Soc Hort Sci 120:90913.
Price KR, Rhodes MJC. 1997. Analysis of the major flavonol glycosides present in four
varieties of onion (Allium cepa) and changes in composition resulting from autoly-
sis. J Sci Food Agric 74:3319.
Torel T, Cillard J, Cillard P. 1986. Antioxidant activity of flavonoids and reactivity with
peroxyradical. Phytochemistry 25:3835.
Trammell KW, Peterson CE. 1976. Quantitative differences in the flavonol content of
yellow onion, Allium cepa L. J Am Soc Hort Sci 101:2057.
Vagen IM, Slimestad R. 2008. Amount of characteristic compounds in 15 cultivars of
onion (Allium cepa L.) in controlled field trials. J Sci Food Agric 88:40411.
Wach A, Pyrzynska K, Biesaga M. 2007. Quercetin content in some food and herbal
samples. Food Chem 100:69970.
Vol. 75, Nr. 2, 2010JOURNAL OF FOOD SCIENCE C165