Development of a radioimmunoassay for the quantitative determination of 8-prenylnaringenin in biological matrices

Article abstract of DOI:10.1021/jf047897u

Seven carboxylic acid haptens of 8-prenylnaringenin (8-PN) were synthesized, coupled to cationized bovine serum albumin, and employed to raise specific antisera in rabbits. Two linkers of different lengths (C ₃H₆COOH and C₆

Full text of DOI:10.1021/jf047897u

J. Agric. Food Chem. 2005, 53, 2881−2889 2881
Development of a Radioimmunoassay for the Quantitative
Determination of 8-Prenylnaringenin in Biological Matrices
OLAF SCHAEFER,† ROLF BOHLMANN,† WOLF-DIETER SCHLEUNING,‡
KAI SCHULZE-FORSTER,§ AND MICHAEL HU¨ MPEL*,†
Research Laboratories Schering AG, 13342 Berlin, Germany, NOXXON Pharma AG,
Max-Dohrn-Strasse 8-10, D-10589 Berlin, Germany, and CellTrend GmbH, Im Biotechnologiepark,
14943 Luckenwalde, Germany
Seven carboxylic acid haptens of 8-prenylnaringenin (8-PN) were synthesized, coupled to cationized
bovine serum albumin, and employed to raise specific antisera in rabbits. Two linkers of different
lengths (C3H6COOH and C6H12COOH) were coupled to the C7-OH group and separated into their
respective enantiomers yielding the first four haptens. Racemic derivatives with C4′-OH coupled linkers
C5H10COOH and C9H18COOH were synthesized carrying a methylated C7-OH. Another racemic C4′-
OH hapten (CH2COOH) was prepared starting from naringenin. The haptens elicited variable antibody
titers dependent on linker lengths, with short linkers giving the best results. Three antisera were
characterized in detail: anti-C7-carboxy-propyloxy-2S-(-)-8-PN (anti-H-11), anti-C7-carboxy-propy-
loxy-2R-(+)-8-PN (anti-H-10), and anti-C4′-carboxy-methoxy-rac-8-PN (anti-H-25). anti-H-10 and anti-
H-11 showed about 9% enantiomeric cross-reactivity, and anti-H-11 did not discriminate between
isoxanthohumol (IX) and 8-PN (84% cross-reactivity). For anti-H-10, cross-reactivities in the range
of 2-5% were found for xanthohumol, IX, and 6-prenylnaringenin. Respective numbers for anti-H-25
were 0.02, 0.1, and 0.2%. Tritiated 8-PN was synthesized yielding a 3H-tracer of high specific
radioactivity (2.22 GBq/mg). A radioimmunoassay using anti-H-25 and 3H-8-PN was established and
used for the quantitative determination of 8-PN in various beer brands and in the urine of six men
after the consumption of three different brands of beer. Furthermore, the dose-dependent excretion
of 8-PN was tested after the consumption of a higher volume of a single beer brand with and without
spiking with 8-PN and a small oral dose of authentic 8-PN, respectively. Conflicting results led to a
pilot test on the in vivo conversion (demethylation) of IX into 8-PN in two men. Conversion rates of
1.9 and 4.4% were estimated. Thus, the total 8-PN dose in beer brands spiced with natural hop or
hop products seems to be the sum of the 8-PN amount in a consumed volume and the amount
arising from the conversion of IX.
KEYWORDS: Phytoestrogens; 8-prenylnaringenin; RIA; beer; isoxanthohumol
INTRODUCTION
tential negative effects have been suggested as well (2).
However, the estrogenic potency of most of these compounds
is very low as compared to endogenous 17â-estradiol. The
prenylated flavanone 8-prenylnaringenin (8-PN) was recently
isolated from the wood of the tropical tree Anaxagorea
luzonensis (3) and also identified as the estrogenic component
in hops (Humulus lupulus) and beer (4, 5). As 8-PN was
identified as the most potent phytoestrogen, it has become the
subject of various research activities in chemistry (6), receptor
binding and transactivation (7), plant biochemistry (8, 9), in
vitro pharmacology (3, 10-13), and analytical methods (14-
17).
Phytoestrogens are polyphenolic, secondary metabolites found
in leafy plants and fungi that exhibit estrogenic activity both in
vivo and in vitro. Initially discovered by their negative impact
on the fertility of domestic sheep feeding on Trifolium subter-
raneum (red clover) (1), phytoestrogens are subdivided by their
chemical structures into the classes of isoflavones, flavanones,
coumestans, and resorcylic acid lactones. Epidemiological
evidence suggests that phytoestrogens ingested with foodstuffs
may have beneficial effects on the prevention of breast cancer,
perimenopausal symptoms, and atherosclerosis, although po-
Because broad strata of the population are widely exposed
to 8-PN by consuming beer, which is known to contain different
amounts of this compound depending on the brand (from 0 to
100 µg/L) (4, 14), it is of considerable interest for public health
* To whom correspondence should be addressed. Tel: +49(0)30-8010-
9257. Fax: +49(0)30-8010-9256. E-mail: michael.huempel@schering.de.
† Research Laboratories Schering AG.
‡ NOXXON Pharma AG.
§ CellTrend GmbH.
10.1021/jf047897u CCC: $30.25
© 2005 American Chemical Society
Published on Web 03/24/2005

2882 J. Agric. Food Chem., Vol. 53, No. 8, 2005
Schaefer et al.
evaluation, and to POD for enzymatic quantification (ELISA). Car-
boxylic acid haptens were activated in dry DMF using NHS and EDAC
and subsequently coupled to the respective protein dissolved in
phosphate-buffered saline (PBS, pH 7.5).
to have a sensitive, specific, cheap, and rapid analytical method
at hand to detect different exposures in consumers and to analyze
8-PN concentrations in different beer brands and production
batches. Immunological assays normally meet these criteria
although specificity needs to be demonstrated for each matrix.
Earlier attempts to raise polyclonal antisera for the detection of
the isoflavonoids formononetin, daidzein, biochanin-A, and
genistein were only partly successful, and separation steps were
necessary prior to enzyme-linked immunosorbent assay (ELISA)
quantification of one specific compound (18). Two main criteria
had to be met in the case of 8-PN: First, the antiserum needed
to discriminate between 8-PN and structurally similar isoxan-
thohumol (IX) because the latter is present in beer at concentra-
tions exceeding those of 8-PN about 100-fold. Second, after
absorption and metabolism of 8-PN in man, the antiserum
needed to discriminate between parent compound and metabo-
lites, which are not yet known but most probably contain,
besides phase I metabolites, conjugates with glucuronic and/or
sulfuric acid.
Two sites of the molecule (C7-OH and C4′-OH) were targeted
for derivative syntheses in order to obtain suitable haptens (for
structures see Figure 1). Linker lengths were varied (C3 and C6
at C7-OH and C6 and C9 at C4′-OH) to achieve maximum
immunogenicity. Linkers at C4′-OH could not be directly
introduced into 8-PN without covalent modification (i.e.,
methylation) of the C7-OH group since various attempts to
reversibly protect the C7-OH group within the 8-PN molecule
failed. Then, C7-OH haptens were synthesized and purified as
both 2S-(-)- and 2R-(+)-enantiomers. Another approach was
followed by starting the hapten synthesis from naringenin
resulting in a racemic C1-linked C4′-OH 8-PN with a free C7-
OH group. Because the peroxidase (POD) conjugates generated
from the haptens did not show sufficient chemical stability, a
tritiated tracer was subsequently synthesized to extend the use
of antisera to a radioimmunoassay (RIA). Possible applications
for the most specific antiserum will be reported in the context
of beer analyses and human exposure of 8-PN by beer
consumption.
OVA conjugates were purified via Sephadex G-25 gel filtration
whereas dialysis against PBS (pH 7.5) was used for purification of
cBSA conjugates. The coupling efficiency was determined by spec-
trophotometric analysis. Stock solutions of cBSA and OVA conjugates
were stored at -20 °C. POD conjugates were purified by Sephadex
G-25 gel filtration and stored at +4 °C.
Synthesis of Tritiated Tracer. The synthesis of the 3H-labeled tracer
of 8-PN is shown in Figure 4. First, the 3′,5′-dibromo derivative 36 of
8-PN was prepared starting from racemic naringenin. Both bromine
atoms were exchanged to 3H in a final, palladium-catalyzed step leading
to the tritiated tracer 37.
Immunization of Rabbits. Two rabbits were immunized for each
hapten-cBSA conjugate. All animals were ear-sampled for preimmune
serum testing prior to immunization. One milligram of hapten-cBSA
conjugate in 0.5 mL of PBS (pH 7.5), emulsified in 1.0 mL of Freund’s
adjuvant (Biogenes, Berlin), was applied to each animal by intramus-
cular injection. Subsequent injections were performed at different body
areas according to the following schedule: 1 and 2, injection in 1 week
intervals; 3, injection after a 2 week interval; 4 and 5, injection in 4
week intervals; 6, injection after a 6 week interval; and 7, injection
after a 12 week interval. Blood samples were drawn at various intervals
to monitor serum titer development. Animals were sacrificed 1 week
after the last injection, and 50 mL of blood was collected from each
rabbit. Serum was obtained after blood clotting at +4 °C for 24 h and
centrifugation at 3000g for 10 min. IgG fractions were isolated from
serum aliquots of 10 mL by protein A sepharose affinity chromatog-
raphy. For storage, IgG fractions were dialyzed against PBS (pH 7.5),
lyophilized, and stored at -20 °C.
Titer Development. Interim and final bleedings of animals were
checked for immune response. Serial dilutions of antisera were
incubated with the respective OVA conjugates and coated onto 96 well
microtiter plates. Bound antibodies were detected with an anti-rabbit-
IgG-POD conjugate. The POD-IgG loading ratio was estimated by
spectrometry at 450 nm.
RIA Procedures. Assay volumes were 1.0 mL consisting of 0.1
mL of standards (75-10000 pg/mL) plus 0.7 mL of buffer or 0.8 mL
of extracts of unknowns, 0.1 mL of tracer (10000 dpm/vial), and 0.1
mL of antiserum solution (final vial dilution 1:100000). The mixture
was incubated at +4 °C overnight. Adhesive separation of unbound
antigen was achieved by admixture of 0.2 mL of 1.25% dextran-coated
charcoal suspension at 0 °C for 15 min, subsequent low speed
centrifugation (2300g), and decantation of the supernatant. The
radioactivity was counted in a liquid scintillation counter after admixture
of 4 mL of Atomlight Plus to the supernatant for 4 min.
Evaluation. TECAN easyWIN fitting Version 6.1 was used to fit
the standard curve and to read the concentration of unknowns. Cross-
reactivity was calculated as a percentage of that concentration of test
substance able to displace the tracer by 50% in relation to the respective
8-PN concentration.
Sample Preparation. Aliquots of urine samples of each individual
were diluted with physiological saline to give 0.5 mL and then extracted
with 2.5 mL of MTBE for 50 min (Heidolph-Rotor). After the aqueous
phase was frozen at -20 °C, the organic layer was separated into a
new vial and evaporated to dryness under nitrogen flow. Extract residues
were redissolved in 0.8 mL of assay buffer, and the solution was directly
used for RIA analysis.
Cleavage of conjugates was carried out using â-glucuronidase/
arylsulfatase as described earlier (23). After cleavage, the samples were
extracted as described before and analyzed by RIA.
Cross-Reactivity. Cross-reactivity was tested for 2S-(-)-6-prenyl-
naringenin (6-PN), 2R-(+)-6-PN, xanthohumol (XH), and IX, all
compounds that are also found in hops. In addition, selected endogenous
sex steroids (estradiol, estrone, estriol, testosterone, and progesterone)
were tested for cross-reactivity. Ten-fold dilutions of 10 or 100 µg/
mL solutions of compounds were tested.
EXPERIMENTAL PROCEDURES
Chemicals. Nonradioactive syntheses were performed using com-
mercially available chemicals from Sigma-Aldrich (Germany), Roth
(Karlsruhe, Germany), and Merck (Darmstadt, Germany). Naringenin
was purchased from Senn-Chemicals (Dielsdorf, Switzerland). Detailed
syntheses, 1H NMR, and mass spectra data of new compounds are
available electronically as Supporting Information. Cationized bovine
serum albumin (cBSA) and immunization adjuvant were prepared by
Biogenes (Berlin); ovalbumin (OVA) and POD (EIA grade) were
obtained from Roche; dimethyl formamide (DMF) was obtained from
Fluka; and 1-ethyl-3-(3-dimethyl aminopropyl) carbodiimide (EDAC)
and N-hydroxysuccinimide (NHS) were purchased from Sigma-Aldrich.
Instruments. Extinction (ELISA) was measured at 450 nm in a
OpsysMR (Dynex Technologies) microplate reader, and radioactivity
(RIA) was counted for 4 min in a scintillation counter (Packard 1900
TR TriCarbo). Extraction of samples with methyl tert-butyl ether
(MTBE) was performed with a head-to-head rotor (Heidolph) and low
speed centrifugation (Heraeus-Christ).
Hapten Syntheses. The syntheses of haptens 9-12, 15, and 17 are
shown in Figure 2. The starting compound in these reactions was
racemic 8-PN 4 that was obtained by a synthesis described before (6).
The synthesis of hapten 25, starting from racemic naringenin, is shown
in Figure 3.
Preparation of Conjugates. Hapten-protein conjugates were
synthesized using the carbodiimide/NHS (EDAC/NHS) method adapted
from Bauminger et al. (19). The haptens 9, 11, 15, 17, and 25 were
coupled to cBSA for immunization of rabbits, to OVA for serum titer
After enzymatic cleavage of conjugates, ether extracts were separated
by high-performance liquid chromatography (HPLC) and collected

8-Prenylnaringenin in Biological Matrices
J. Agric. Food Chem., Vol. 53, No. 8, 2005 2883
Figure 1. Chemical structures of 8-PN, XH, and IX and synthesized haptens 9−12, 15, 17, and 25.
Figure 2. Scheme of synthesis of haptens 9−12, 15, and 17. (i) Acetone, K2CO3, RBr, reflux. (ii) LiI, 2,4,6-collidine, reflux. (iii) MeI, K2CO3, acetone,
reflux. (iv) P(Ph)3, DIAD, THF, ROH.
fractions were subsequently analyzed by RIA in order to verify
specificity of 8-PN analysis in urine collected within the beer drinking
and IX conversion tests. The HPLC system consisted of a Kromasil
C18 5 µm (4.6 mm × 250 mm) column operated at room temperature
and a steady flow of 1 mL/min. Elution was achieved by 20%
acetonitrile in 0.2% acetic acid for 1 min followed by a linear gradient
to 80% acetonitrile in 0.2% acetic acid within 19 min. Thereafter, the
column was reconstituted by a linear gradient (4 min) and isocratic
elution with 20% acetonitrile in acetic acid (1 min). The retention time
of 8-PN was 18.2 min. Residues of 0.5 min fractions were quantified
by RIA.
Beer Consumption. Six healthy men (39 ( 12 years, 83 ( 6 kg,
and 185 ( 7 cm) participated in this experimental setting and were
informed about the aims of the test and signed a written protocol. They
were also informed about the outcome of a clinical single dose study,
in which 50, 250, and 750 mg of 8-PN was given to each of six

2884 J. Agric. Food Chem., Vol. 53, No. 8, 2005
Schaefer et al.
Figure 3. Scheme of the synthesis of hapten 25. (i) PhCH2Br, acetonitrile, K2CO3, reflux. (ii) P(Ph)3, DIAD, THF, HOCH2CO2Me. (iii) EtOH, THF, Pd/C,
H2. (iv) Pyridine, (CH3CO)2O. (v) P(Ph)3, DEAD, (CH3)2CHdCHCH2OH, THF. (vi) Decaline, reflux. (vii) K2CO3, MeOH, H2O. (viii) LiI, 2,4,6-collidine,
reflux.
Figure 4. Scheme of the synthesis of 3H-tracer 37. (i) MOMCl, K2CO3, acetone, reflux. (ii) p-OH-3,5-dibromobenzaldehyde, EtOH, THF, KOH. (iii) HCl,
MeOH, reflux. (iv) NaOAc, MeOH, reflux. (v) (CH3CO)2O, pyridine. (vi) P(Ph)3, DIAD, (CH3)2CHdCHCH2OH, THF. (vii) CHCl3, Eu[FOD]3, reflux. (viii)
K2CO3, MeOH, H2O. (ix) Acetic acid ethyl.
menopausal women without any drug-related side effects. This informa-
tion was provided since each participant had to take 0.25 mg of
authentic substance on test days (TDs) 5 and 6. The spiking dose
corresponded to 0.03% of the highest tolerated dose of the clinical study.
The authentic compound was taken from a GMP production batch.
The test was in six sessions, and urine was to be collected prior to
(blank sample) and for 24 h after each consumption. After thorough
mixing, the total volume was recorded by participants and an aliquot
was kept at -16 °C in a plastic screw vial until analysis. Three days
before each test, beer consumption but not other alcohol was prohibited.
No other restrictions were made for daily life activities or diet habits.
Three beer brands were used, and bottles of identical production batches
were distributed to the participants. Beer A was spiced with CO2-
extracted bitter acids and thus did not contain 8-PN or IX. Analysis
indeed revealed a concentration of <1.6 µg 8-PN/L. Beer B had a
medium concentration of 8-PN (about 15 µg/L), and beer C had a higher
concentration (33 µg/L).
In sessions 1-3, participants consumed 0.5 L of beers A-C within
50 min; in session 4, they consumed 1 L of beer C within 90 min; and
in session 5, they consumed 1 L of beer C, which was spiked with
0.25 mg of 8-PN as aqueous/ethanolic (1:1) solution. In session 6,
participants ingested only 0.25 mg of 8-PN as a 5% (v/v) alcoholic

8-Prenylnaringenin in Biological Matrices
J. Agric. Food Chem., Vol. 53, No. 8, 2005 2885
solution without beer. All participants finished all test consumptions
in the order 1-6 except participant MH, who followed the order 1, 2,
3, 6, 4, and 5.
Table 1. Cross-Reactivity of Three Antisera against 8-PN Enantiomers
and Structurally Similar Hop Constituents
RIA Analysis of Beer Brands. Sixteen beer brands (Table 6) were
analyzed. Beer aliquots were degassed at +20 °C and 50 mbar, of which
0.1 mL was mixed with 0.4 mL of physiological saline and then
extracted with ether as described above. RIA analysis was carried out
as duplicates.
Test for IX Conversion. Two of the six subjects participating in
the beer drinking test (MH and OS) ingested 10 mg of IX mixed into
a small volume of a high percentage alcoholic beverage and then
collected urine at 12 h intervals for 3-4 days. Urine volumes were
recorded as before, and small samples were kept deep frozen until
analysis. Both participants did not drink any beer at least 3 days before
ingestion of IX and during the urine collection period. IX normally is
present in beer at a concentration of 2 mg/L. Thus, both volunteers
were administrated an IX dose equivalent to that present in 5 L of
beer. IX was purchased from Phytochem (Ichenhausen, purity >95%).
cross-reactivity (%)
analyte
2S-( )-8-PN
2R-( )-8-PN
XH
IX
anti-H-11
anti-H-10
anti-H-25
−
+
100
8.6
0.6
84
9.1
100
2.2
2.2
4.6
100
98
0.02
0.10
0.15
2S-(−)-6-PN
0.2
side chain was introduced as described above. Double bromine
to tritium exchange of 36 gave the 3H tracer 37 with good yield
and very high stability of the tritium lable. The specific activity
was 2.22 GBq/mg.
Conjugation Tests. The conjugates obtained were tested for
hapten loading by spectrophotometric methods. According to
chemical instability of conjugates, ELISA development was
abandoned in favor of the RIA (see the Supporting Information).
RESULTS AND DISCUSSION
Synthetic Chemistry. Four carboxylic acid haptens of 8-PN
were synthesized with a linker at the C7-O position (9-12) and
three haptens bearing a spacer arm at the C4′-O position (15,
17, and 25). A tritiated tracer (37) was furthermore synthesized
by halogen to tritium substitution of a dibromo derivative of
8-PN (36).
Series of Haptens with Linker on C7-O. Starting with
racemic 8-PN, which was obtained employing the route of
synthesis described by Gester et al. (6), the respective methyl
esters 5 and 7 were obtained by classic William’s ether synthesis.
Subsequent saponification to the free racemic acids 6 and 8 was
carried out with good yield using lithium iodine in collidine to
avoid either basic or acidic aqueous conditions to which 8-PN
responds with elimination/ring fission, described previously as
Wesley-Moser rearrangement of flavonoids (20). Optically pure
haptens 9-12 were isolated by chiral HPLC (for details, see
the Supporting Information).
Haptens with Linker on C4′-O. The syntheses of a first set
of C4′-O haptens started from reacemic 8-PN as well. The highly
nucleophilic C7-OH group had to be protected by methylation
in order to obtain the respective methyl esters 14 and 16 via
Mitsunobu ether synthesis. Common protection group chemistry
for phenols failed due to the above-mentioned pH sensibility
of the molecule. Free acids 15 and 17 had therefore a
C7-methoxy group instead of a phenol. Signficantly reduced
immunogenicity of these haptens led to the choice of a second
synthetic approach starting with naringenin. The nucleophilic
C7-OH was initially benzyl protected (18) and the linker was
coupled at the free C4′-OH via Mitsunobu (19). The prenyl
group was then introduced as described before (6), following
deprotection of C7-OH. Methyl ester cleavage under the same
conditions as mentioned before gave hapten 25.
Tritiated Tracer. Pd-catalyzed halogen to tritium substition
is a suitable method for radiolabeling of small organic com-
pounds. Treatment of 8-PN with either N-bromo- or N-iodo-
succinimide (NBS/NIS) in THF resulted in the respective bromo/
iodo derivative at C6. Still, the subsequently introduced tritium
lable at this position instantly exchanged protons with protic
solvents. Excess equivalents of either NBS or NIS led to
halogenation of the aliphatic prenyl side chain instead of the
aromatic B ring. We therefore pursued the approach to introduce
bromine residues to the B ring by de novo synthesis. Methoxy
methyl protected 2,4,6-tridydroxyacetophenone (28) and 3,5-
dibromo-4-hydroxybenzyaldehyde were used for an aldol con-
densation and subsequent catalytic ring closure to give the 3′,5′-
dibromo-naringenin derivative 31. From there on, the prenyl
Immunization of Rabbits. Immunogenicity and titer devel-
opments were clearly different for the antigenes investigated.
Gernerally, long linkers such as anti-H-15 (C9 linker) or anti-
H-17 (C5 linker) caused no or very little immune response. Both
of these antigenes were methylated at C7-OH, which most likely
further reduced immunogenicity. anti-H-9 and anti-H-10 (both
carrying C6 linkers at C7-O) showed a high immune response,
which was sufficient for detailed evaluation of the antisera. A
very high immunogenicity was obtained with short linkers such
as anti-H-11/12 (C3 linker at position C7-O) and anti-H-25 (C1
linker at position C′4-O). Long linkers presumably hid the
antigen within the protein matrix or did not provide a strong
antigen presentation and therefore caused the loss of immuno-
genicity.
Cross-Reactivity. Three antisera representing different link-
ing strategies were tested in detail for cross-reactivities (Table
1). The stereoselectivity of anti-H-10 and anti-H-11 was
pronounced, but the cross-reactivity with the opposite enanti-
omer was still about 9%. Thus, the stereoselectivity of both
antisera was not sufficient for quantitative determination of a
single enantiomer. In addition, anti-H-11 showed an extensive
cross-reaction with IX while anti-H-10 did not sufficiently
discriminate 6-PN. The most specific antiserum was anti-H-
25, showing a low cross-reactivity with XH, IX, and 6-PN. This
antiserum was also tested in the RIA set up for cross-reaction
with endogenous sex steroids. As expected, no cross-reactivity
(<0.001%) was observed for estradiol, estrone, estriol, test-
osterone, and progesterone. Hapten 11, linked at position C7-O
via a short C3 residue, led to an antigen, which was unable to
present the free C5-OH as a determinant while the open ring
structure of XH was well-discriminated by the respective
antiserum. On the other hand, hapten 10 linked at the same
position with a long (C6) residue and gave an antigen that was
able to present the respective C5-OH group more efficiently.
However, the improvement in specificity against IX was reached
on the expense of lower specificities against 6-PN and XH.
Thus, anti-H-10 presumably did not only show a somewhat
lower immunogenicity as compared to anti-H-11 but also might
contain a higher variety of antibodies due to the higher flexibility
of presented hapten. The higher immunogenicity and specificity
of the short-linked anti-H-11 and anti-H-25 found in the present
experiments is in agreement with respective data found for
isoflavonoid haptens (18, 21, 22). Stereoselectivity of antisera
was found to be insufficient for both antisera tested. Because
racemization during coupling and immunization can be ne-

2886 J. Agric. Food Chem., Vol. 53, No. 8, 2005
Schaefer et al.
glected, it is true that stereoisomers were important but not
exclusively determinant for the (flexible) antigens presented.
On the basis of specificity data, anti-H-25 was selected and
developed as RIA for the quantitative determination of racemic
8-PN in biological matrices. It must be emphasized at this point,
however, that also anti-H-11 represents a valuable tool for the
quantitative determination of 8-PN when racemic 8-PN is used
for standardization and the presence of XH and IX in samples
can be excluded. This is the case when the pharmacokinetics
and metabolic fate of 8-PN as an authentic compound need to
be followed.
A high cross-reactivity to metabolites had to be taken into
account for assay development, according to the linking position
of the antigen leading to anti-H-25 (C4′) and the anticipated
occurrence of C4′-glucuronides or -sulfates of 8-PN in animal
species and humans. This problem was solved by determination
of ether extracts instead of direct quantification in biological
fluids.
Table 2. Time Intervals between Testing Days and Concentrations of
Free 8-PN in Blank Urine Samples of Six Individualsa
MHb
AS
OS
DS
FW
JF
TD
A
B
A
B
A
B
A
B
A
B
A
B
1
2
3
4
5
6
<min
<min
<min
755
138
<min
<min
128
647
930
933
190
<min
<min
139
218
1014
322
104
<min
102
485
170
<min
<min
<min
<min
<min
<min
<min
<min
1988
1
1
1
4
1
0
0
0
1
1
0
8
0
0
0
1
2
0
1
2
0
0
0
0
7
9
0
11 <min
0
0
171
636
1551
a A
pg/mL; and min
the follow-up of TDs (1
)
time interval between TDs in days; B
detection limit of 100 pg/mL (300
3, 6, 4, 5).
)
blank urine concentration in
µ
L extracted). b MH changed
)
−
Table 3. Time Intervals between TDs and Concentrations of Free Plus
Conjugated 8-PN in Blank Urine Samples of Six Individualsa
MHb
AS
OS
DS
FW
JF
RIA Tests. Sample Preparation. Aliquots of urine samples
(untreated or deconjugated blank or pooled samples) were
diluted with physiological saline to give 0.5 mL and then
extracted with 2.5 mL of MTBE for 50 min (Heidolph-Rotor).
After the aqueous phase was frozen, the organic layer was
transferred into a new vial and evaporated to dryness under
nitrogen. Extract residues were redissolved in 0.8 mL of assay
buffer, and the solution was directly used for RIA analysis.
Duplicate analysis was carried out for all tests. Cleavage of
conjugates in serum and urine samples was carried out using
â-glucuronidase/arylsulfatase as described earlier (23).
RecoVery by Ether Extraction. Extraction of 8-PN either by
means of diethyl ether or MTBE was complete, as shown by
the fact that extracted and not extracted standard curves were
superimposable (Figure 4). The linearity of extraction was
shown at two concentrations (0.3 and 3.0 ng/mL) of spiked blank
urine samples. No matrix interference was seen up to an
extracted volume of 300 µL.
Robustness and Stability of Agents. Stock solutions (in assay
buffer) of standards, tracer, and antiserum were stable for more
than 2 months. No degradation of the tracer in ethanolic solution
was found over a period of 6 months when kept at -20 °C. In
principle, extractions of urine samples can be carried out with
either diethyl ether or MTBE. However, the assay is sensitive
(elevated blanks) against peroxides known to form in diethyl
ether; therefore, extraction was performed with MTBE.
Precision (Inter- and Intraassay Variation). Blank urine
samples were spiked with 8-PN to give nominal concentrations
of 0.3 and 1 ng/mL. Samples were analyzed 7-fold in one assay
and at least in duplicate in eight different assays.
The intraassay coefficient of variation (CV) was 8.8 and 7.0%
for 0.3 or 1.0 ng/mL urine, respectively. The interassay CV was
26.8% for 0.3 ng/mL urine and 17.6% for 1.0 ng/mL sample.
Thus, for both concentrations, the intraassay precision was
within the accepted range for analytical methods (CV < 20%)
while the interassay precision was met only for the higher
concentration.
Accuracy. Blank urine samples were spiked with 1 and 3 ng
8-PN/mL and analyzed 5-fold. Found concentrations were 1.36
( 0.14 and 3.29 ( 0.24 ng/mL, indicating an overestimation
of roughly 30% in these experiments.
SensitiVity. The RIA system described has a sensitivity limit
of 0.3 ng 8-PN/mL urine using 0.1 mL of sample. In case of
free analyte in urine, 0.3 mL were extracted and the sensitivity
was increased to 0.1 ng/mL. Thus, the lower limit of quantifica-
tion was set to 0.1 ng/mL.
TD
A
B
A
B
A
B
A
B
A
B
A
B
1
2
3
4
5
6
0.78
0.55
2.1
5.06
1.47
3.25
0.39
0.65
0.93
3.34
6.34
3.97
0.72
0.37
1.02
1.93
4.68
5.59
1.25
0.98
1.31
3.85
3.44
9.10
0.31
<min
0.30
<min
0.96
1.53
0.58
0.72
1.81
1.98
1.43
20.8
1
1
1
4
1
0
0
0
1
1
0
8
0
0
0
1
2
0
1
2
0
0
11
0
0
0
0
7
9
0
a A
ng/mL; min
(1−3, 6, 4, 5).
)
time interval between TDs in days; B
)
) blank urine concentration in
detection limit of 0.3 ng/mL. b MH changed the follow-up of TDs
In summary, with the anti-H-25 antiserum and tritiated 8-PN,
we developed a RIA suitable for the quantitative determination
of free and conjugated (after enzymatic cleavage of conjugates)
8-PN in urine. Almost identical quality criteria for the assay
were found for the determination in human serum. Data will
be described elsewhere (24).
Renal Excretion in Beer Consumption Tests. Urine samples
collected during the beer drinking tests were analyzed both prior
and after enzymatic cleavage of conjugates. Results were
calculated as excretion per 24 h.
Because the experimental design did not demand a specific
time interval between testing days, this question was handled
differently by individuals (Table 2). Concentrations of free 8-PN
before the first TD were below the detection limit of 0.1 ng/
mL. However, concentrations in urine samples collected directly
before beer consumption increased remarkably and reached
values of 1-2 ng/mL in individual samples. This can be
explained by increasing doses of 8-PN at testing days 1-6 and
an overlap of incomplete excretion and new 8-PN dose. Thus,
it turned out that excretion of 8-PN is slower than expected
and an interval of several days between TDs would have been
necessary to avoid accumulation of 8-PN during the study.
Following enzymatic cleavage of conjugates, 0.3-1.25 ng/mL
were found in blank urines (Table 3). Although it cannot be
excluded that beer consumption 4 days before TD 1 might have
led to elevated blank values, there were two participants who
hardly ever consume beer. Thus, it is likely that there are other
sources of 8-PN (or IX, see below) in regular diet. Precon-
sumption concentrations in urine markedly increased with TDs
and short intervals between TDs. The lowest values were found
for participant FW at TD 2 and TD 4 (after an interval of 11
days) and the highest (20.8 ng/mL) for JF at TD 6 (no interval
to TD 5). On the basis of these data, it is clear that there was
a varying overlap of 8-PN excretion from the foregoing TDs.

8-Prenylnaringenin in Biological Matrices
J. Agric. Food Chem., Vol. 53, No. 8, 2005 2887
Table 4. 24 h Renal Excretion of Free Plus Conjugated 8-PN at TD
1−6 (in µg)
TD
MHa
AS
OS
DS
FW
JF
mean
SD
1
2
3
4
5
6
2.56
9.14
12.9
49.9
68.2
66.1
2.23
2.53
3.55
18.0
40.3
23.6
1.55
2.21
3.62
3.88
11.4
15.5
1.39
1.25
2.78
14.4
63.8
43.3
0.57
1.47
2.88
2.26
6.31
35.4
38.9
1.63
3.60
4.28
18.5
40.5
33.0
0.70
3.20
4.36
18.5
22.2
18.8
0.59
24.0
20.6
a MH changed the follow-up of TDs (1
−3, 6, 4, 5).
Figure 5. RIA using anti-H-25 (total IgG fraction) and 3H-8-PN as tracers;
shown are nonextracted and ether-extracted standard curves.
Table 5. Renal Excretion of RIA Reactive Material Following a Single
Oral Intake of 10 mg of IX in Two Individuals; Data Are Given as
µ
g
8-PN Per 12 or 24 ha
MH
OS
time interval (h)
12 (prevalue)
12
µ
g/0.5 days
µ
g/day
1.41
µ
g/0.5 days
µ
g/day
−
0.703
0.615
1.23
5.10
5.49
3.49
0−
4.25 NT
9.66 DT
9.07 NT
3.17 DT
2.97 NT
1.12 DT
2.02 NT
1.17 DT
1.87 NT
3.68 DT
1.42 NT
2.37 DT
3.12 NT
2.53 DT
0.96 NT
1.26 DT
12−
24−
36−
48−
60−
72−
84−
96−
24
13.9
12.4
4.09
3.19
36
48
60
72
84
96
108
Figure 6. HPLC separation of blank urine extracts following cleavage of
conjugates and subsequent quantification of residues per eluent fraction
by RIA. Amounts per fraction were converted into % of amounts found in
SUM
t1 2 (h)
/
35.3
22.3
33.6
31 (4 dp)
15.3
nc
14.0
30 (2 dp)
total chromatography. 8-PN elutes in fractions 36
blank sample before TD 4 (5).
−37. MH (AS) means
a nc
)
not calculable; DT ) daytime; NT ) nighttime; and dp ) data points.
To validate specificity of determined 8-PN in enzyme treated
urine samples, extracts of two urine samples collected directly
before beer consumption were separated by HPLC and eluates
were collected in fractions. Residues were then analyzed by RIA.
As shown in Figure 6, only negligible interference was observed
and the assay can therefore be regarded as highly specific.
doses of TDs 5 and 6 (283 and 250 µg) was much lower with
values of 16 and 13%, respectively.
Adding up all 8-PN excreted per participant for TD 1-TD
6, the mean amount was 99 ( 62 µg (range: 38-209 µg). This
corresponds to 17% of total 8-PN dose (590 µg).
Renal Excretion in IX Conversion Test. Two participants
of the beer consumption test (MH and OS) volunteered for an
oral intake of 10 mg of IX. Twelve hour urine fractions were
collected for 4.5 (MH) and 3.5 (OS) days. 8-PN was quantified
in urine after cleavage of conjugates (Table 5).
Again, there were positive pretest 8-PN signals in the RIA,
suggesting a daily excretion of 1.2-1.4 µg of 8-PN not related
to beer consumption. In one subject, 35 µg of 8-PN was found
in 4.5 days of urine and roughly the half (15 µg) in the 3.5
days urine of the other subject. Generally, after the first day, a
higher proportion of the daily excreted amount was found in
urine collected overnight. On the basis of 12 h data, the excretion
half-life for MH was determined to 22.3 h, and a somewhat
slower elimination was estimated using daily excreted amounts
of both subjects.
Because it is known that the RIA has a low but measurable
cross-reactivity to IX (0.15%), the specificity of analysis was
verified as described above. As shown in Figure 7, there were
minor elevations in fractions 21 and 28 (OS) or 26 and 30 (MH),
which probably reflect large amounts of IX in the urine.
However, about 90% of the RIA detection was due to 8-PN
eluting in fraction 36. Thus, although specificity of RIA
measurements for 8-PN in urine after intake of IX was not
100%, data can be used for further considerations and clearly
demonstrate an in vivo conversion of IX into 8-PN.
The mean (( SD) value of excretion of free 8-PN was 0.08
( 0.06 µg for TD 1 (no 8-PN dose) and then increased to 0.14
( 0.04, 0.54 ( 0.38, and 0.85 ( 0.53 µg for TD 2-4 with
nominal 8-PN doses of 7.5, 16.5, and 33 µg, respectively.
Excreted amounts correlate with increasing 8-PN doses for TD
2-4. Following consumption of the spiked beer (TD 5) and
authentic 8-PN (TD 6), about 3.5 µg was excreted, which is
less than half of the amount expected by linear extrapolation
of dose increases (ca. 13 µg).
8-PN conjugates are the primary renal excrements. Mean
values of 8-PN conjugates plus free 8-PN excretion were 10-
20-fold higher than for free compound alone (Table 4). Again,
there was a detectable excretion of 8-PN at TD 1. On TD 1, an
average 1.6 ( 0.7 µg was found although no 8-PN or IX had
been administered. This mean value corresponds well to
preconsumption excretion of 1.4 or 1.2 µg of 8-PN in the IX
conversion test (Table 5). It therefore appears most likely that,
beside beer, there is one or more hitherto unknown dietary
source of 8-PN or IX, which itself can be converted into 8-PN.
Following TDs 2-4, an increase in excreted amounts was
observed with increasing doses. However, interindividual varia-
tions were high leading to a CV of means of above 100% (Table
4). This points to a nonnormal distribution of values or to
outliers. Indeed, if values of participant MH were not included
into mean value calculations, CVs ranged from 50 to 60%. On
TDs 2-4, the mean 8-PN excretion accounted for 41, 26, and
50% of nominal 8-PN dose, respectively. Excretion of nominal
Conclusions on In Vivo Tests. The availability of a highly
specific and sensitive RIA for 8-PN detection was the basis for
the present two pilot studies in which uptake of different 8-PN

2888 J. Agric. Food Chem., Vol. 53, No. 8, 2005
Schaefer et al.
Table 6. Concentrations of 8-PN in Various Beer Brands and Their
Spicing Methodsa
code (origin)
type
8-PN (
µ
g/L)
spicing
A (D)
B (D)
C (D)
D (D)
E (D)
F (D)
G (D)
H (D)
I (D)
K (D)
L (CZ)
M (D)
N (D)
O (IR)
P (IR)
Q (FIN)
nonalcoholic
pilsner
pilsner
pilsner
pilsner
pilsner
pilsner
cryst. wheat
cryst. wheat
diet
pilsner
stout
stout
stout
stout
< 1.6
15.1
33.0
7.3
16.0
16.6
8.1
12.0
10.3
7.1
16.3
24.3
98.4
46.2
138.5
41.9
I
I
I
I
III
II
I
III
III
I +
nr
III
II
I
nr
+
+
+
II
II
III
+ III
+
+
III
III
Figure 7. HPLC separation of ether extracts of two urine samples after
cleavage of conjugates and subsequent RIA analysis of residues of HPLC
fractions; sample MH (OS) means 12−24 h (0−12 h) collection period
after oral intake of 10 mg of IX; 8-PN elutes in fraction 36; data are
shown as a percentage per fraction of total HPLC run.
III
+ III
+
III
porter
I +
III
doses in beer and the conversion of IX into 8-PN were tested.
Excretion data can be interpreted within the context of renal
excretion of free and conjugated 8-PN elaborated within a
clinical phase Ia study (24). The study revealed that only 0.23%
of a 50 mg oral dose was renally excreted as free 8-PN and
about 8% as conjugates. In contrast to these figures, markedly
higher mean values of excreted free and conjugated 8-PN were
estimated after beer consumption. On TDs 2-6, free 8-PN
excretion accounted for 1.8, 3.2, 2.5, 1.3, and 1.4% of nominal
8-PN dose. A respective calculation on the basis of excreted
conjugates (Table 4) reveals similar high values, i.e., 41, 26,
50, 16, and 13% of the nominal dose. Thus, both data sets clearly
show much higher than expected excretion on the basis of phase
Ia study data. Possible explanations include different pharma-
cokinetics of 8-PN after a high (50 mg) acute oral dose and
low (µg range) doses in beer. However, it is rather improbable
that routes of elimination have changed with different doses,
because complete absorption of the high dose was found in the
clinical study. The degree of metabolism could have changed
with different doses, but in this scenario, the low dose (beer)
would have been metabolized to a higher extent than the high
dose. Therefore, differences in pharmacokinetics at different
8-PN doses do not offer a plausible explanation for the present
results.
There is, however, another possible source for 8-PN in beer.
Beer spiced with hop products contains about 2 mg IX/L.
Because concentrations of 8-PN (ca. 20 µg/L) and IX in beer
differ by a factor of 100, a low in vivo conversion of IX into
8-PN (by oxidative O-demethylation) could significantly in-
crease the total 8-PN dose. Very recently, it was shown that
human liver microsomes indeed convert IX into 8-PN in vitro
(25). The pilot test in two individuals ingesting 10 mg of IX
showed that this conversion indeed has occurred. Estimated
conversion rates were 1.9 (OS) and 4.4% (MH) of the IX dose
and generated 8-PN was excreted with a half-life of about 1
day, whichsat a daily consumption of beerswould lead to an
1.6-fold accumulation of 8-PN. Obviously, the amount of 8-PN
in beer is not the only and main source of estrogenicity but
also the conversion of IX needs to be considered.
a Explanation: Origin D, CZ, IRL, FIN
Finland; nr no response from brewery; spicing methods:
ethanolic extract, and III hop or hop pellets.
)
Germany, Czech Republic, Ireland,
CO2 extract, II
)
I
)
)
)
burden of 8-PN can be approximated to 1 mg/day. More studies
on steady state 8-PN burden and variations in IX conversion
need to be done in order to get more information and a proper
basis for estimating various consumer scenarios. The developed
RIA has been proven to constitue an excellent tool for respective
future investigations.
As a result of the high sensitivity of the RIA, we found
elevated blank 8-PN values and a daily “normal” excretion of
about 1.5 µg of 8-PN. We conclude that beer is not the only
source of 8-PN in the diet. The present data show that there are
at least two possible sources for 8-PN in the diet or in beverages,
i.e., 8-PN itself and IX. This conclusion is in conflict with the
literature, which negate other sources for 8-PN except hop.
However, even low concentrations in vegetarian foodstuffs
escaping the used detection methods or in meat of animals fed
with hop tresters from CO2 extracts might have been sufficient
to raise pretreatment blank values. The RIA and the cleavage
of conjugates represent a very powerful system for detection
of low 8-PN concentrations in urine. It is estimated that a daily
dose of less than 1 µg of 8-PN can be detected by our system.
8-PN Concentrations in Beer Brands. RIA results (Table
6) show that 8-PN concentrations vary widely between different
brands. Most beers tested contained less than 40 µg/L, but in
some stouts, 8-PN concentrations were close to or even above
100 µg/L. Beer brands A-C were used in the drinking test
described above.
Principal techniques of beer spicing employ CO2 extracts of
hop lacking both 8-PN and IX, ethanolic hop extracts containing
both compounds or pellets of dried hop. Most beers listed in
Table 6 were spiced by combined methods and often by a
combination of CO2 extract and dried hop. Respective 8-PN
concentrations (beers D, G, K, O, and Q) were either below 10
or above 40 µg/L, reflecting different amounts of used hop
products. In agreement with the soley applied CO2 spicing, no
8-PN was found in beer A. However, no clear correlation was
seen between the method of beer spicing and the found 8-PN
concentrations because too many other variables were not
known.
Whether consumption of high amounts of beer over a long
period of time causes estrogenic effects and might feminize men
has been the subject of numerous discussions. Clearly, with the
presented limited data, this question cannot be answered but
rather helps us to focus the ongoing debate by developing a
“worst case scenario” for the exposure to 8-PN. With an
assumed individual IX conversion rate of 5%, a daily consump-
tion of 3 L of beer (50 µg 8-PN/L, 2 mg IX/L) and calculating
with an accumulation factor of 2, the resulting steady state body
ABBREVIATIONS USED
8-PN, 8-prenylnaringenin; 6-PN, 6-prenylnaringenin; IX,
isoxanthohumol; XH, xanthohumol; cBSA, cationized bovine

8-Prenylnaringenin in Biological Matrices
J. Agric. Food Chem., Vol. 53, No. 8, 2005 2889
(11) Henderson, M. C.; Miranda, C. L.; Stevens, J. F.; Deinzer, M.
L.; Buhler, D. R. In vitro inhibition of human P450 enzymes by
prenylated flavonoids from hops, Humulus lupulus. Xenobiotica
2000, 30 (3), 235-251.
serum albumin; POD, peroxidase; OVA, ovalbumin; ELISA,
enzyme-linked immunosorbent assay; RIA, radioimmunoassay;
MTBE, methyl tert-butyl ether.
(12) Shirataki, Y.; Motohashi, N.; Tani, S.; Sakagami, H.; Satoh, K.;
Nakashima, H.; Mahapatra, S. K.; Ganguly, K.; Dastidar, S. G.;
Chakrabarty, A. N. In vitro biological actvity of prenylfla-
vanones. Anticancer Res. 2001, 21, 275-280.
(13) Pepper, M. S.; Hazel, S. J.; Hu¨mpel, M.; Schleuning, W.-D.
8-Prenylnaringenin, a novel phytoestrogen, inhibits angiogenesis
in vitro and in vivo. J. Cell. Physiol. 2004, 199, 98-107.
(14) Rong, H.; Zhao, Y.; Lazou, K.; De Keukeleire, D.; Milligan, S.
R.; Sandra, P. Quantitation of 8-prenylnaringenin, a novel
phytoestrogen in hops (Humulus lupulus L.), hop products, and
beers, by benchtop HPLC-MS using electrospray ionization.
Chromatographia 2000, 51 (9/10), 545-552.
ACKNOWLEDGMENT
We thank Dr. J. Gay and Dr. C. Gu¨nther (both Schering AG)
for their manifold help in isotope chemistry and pharmacoki-
netics, M. Kirchner (Biogenes) for raising the antisera and for
ELISA development, and Dr. M. Biendl (Hallertauer Hopfen-
veredlungs GmbH) for support of selected beer brands and
expert advice.
Supporting Information Available: Details on syntheses and
spectrometric data of reported new compounds; coupling
efficiency of haptens to cBSA, OVA, and POD; and titer
development of antigenes of 8-PN in rabbits. This material is
available free of charge via the Internet at http://pubs.acs.org.
(15) Stevens, J. F.; Taylor, A. W.; Deinzer, M. L. Quantitative analysis
of xanthohumol and related prenylflavonoids in hops and beer
by liquid chromatography-tandem mass spectrometry. J. Chro-
matogr. 1999, 832, 97-107.
(16) Tekel, J.; De Keukeleire, D.; Rong, H.; Daeseleire, E.; van
Peteghem, C. Determination of the hop-derived phytoestrogen,
8-prenylnaringenin, in beer by gas chromatography/mass spec-
trometry. J. Agric. Food Chem. 1999, 47, 5059-5063.
(17) Vanhoenacker, G.; Dermaux, A.; De Keukeleire, D.; Sandra, P.
Single-run capillar electrochromatographic analysis of hop acids
and prenylated hop flavonoids. J. Sep. Sci. 2001, 24, 55-58.
(18) Benneteau-Pelissero, C.; Le Houe´rou, C.; Lamothe, V.; Le Menn,
F.; Babin, P.; Benneteau, B. Synthesis of haptens and conjugates
for ELISAs of phytoestrogens. Development of the immunologi-
cal tests. J. Agric. Food Chem. 2000, 48, 305-311.
(19) Bauminger, S.; Wilchek, M. The use of carbodiimides in the
preparation of immunizing conjugates. Methods Enzymol. 1980,
70, 151-159.
(20) Mukerjee, S. K.; Seshadri, T. K. A new isomeric change in the
flavonoids. Chem. Ind. 1955, 271-275.
(21) Lapcik, O. R.; Hampl, R.; al-Maharik, A.; Salaka, A.; Wahala,
K.; Adlercreutz, H. A novel radio immunoassay for daidzein.
Steroids 1997, 62, 315-320.
(22) Lapcik, O. R.; Hampl, R.; Hill, M.; Wahala, K.; Maharik, N.
A.; Adlercreutz, H. Radio immunoassay of free genistein in
human serum. J. Steroid Biochem. Mol. Biol. 1998, 64, 261-
268.
(23) Hu¨mpel, M.; Nieuweboer, B.; Milius, W.; Wendt, H.; Kurowski,
M. The pharmacokinetics and biotransformation of a new
benzodiazepine (lormetazepam) in humans. I. Absorption, dis-
tribution, elimination and metabolism of lormetazepam 5-14C.
Eur. J. Drug Metab. Pharmacokinet. 1979, 4, 237-243.
(24) Rad, M.; Hu¨mpel, M.; Schaefer, O.; Schoemaker, R.; Schleuning,
W.-D.; Burggraaf, K. Safety, tolerance, and pharmacokinetics
following single oral doses of the phytoestrogen 8-prenylnarin-
genin in postmenopausal women and first signs of systemic
efficacy. Manuscript in preparation.
(25) Nikolic, D.; Li, Y.; Chadwick, L. R.; Pauli, G. F.; van Breemen,
R. B. Metabolism of xanthohumol and isoxanthohumol, pren-
ylated flavonoids from hops (Humulus lupulus L.), by human
liver microsomes. J. Mass Spectrom. 2005, 40, 289-299.
LITERATURE CITED
(1) Bennetts, H. W.; Underwood, E. J.; Shier, F. L. A specific
breeding problem of sheep on subterranean clover pastures in
western Australia. Aust. Vet. J. 1946, 22, 2-12.
(2) Russell, L.; Hicks, G. S.; Low, A. K.; Shepherd, J. M.; Brown,
C. A. Phytoestrogens: A viable option? Am. J. Med. Sci. 2002,
324, 185-188.
(3) Kitaoka, M.; Kadokawa, H.; Sugano, M.; Ichikawa, K.; Taki,
M.; Takaishi, S.; Iijima, Y.; Tsutsumi, S.; Boriboon, M.;
Akiyama, T. Prenylflavonoids: A new class of nonsteroidal
phytoestrogen (part 1). Isolation of 8-isopentenylnaringenin and
an initial study on its structure-activity relationship. Planta Med.
1998, 64, 511-515.
(4) Milligan, S. R.; Kalita, J. C.; Heyerick, A.; Rong, H.; De
Cooman, L.; De Keukeleire, D. Identification of a potent
phytoestrogen in hops (Humulus lupulus L.) and beer. J. Clin.
Endocrinol. Metab. 1999, 83 (6), 2249-2252.
(5) Stevens, J. F.; Taylor, A. W.; Clawson, J. E.; Deinzer, M. L.
Fate of xanthohumol and related prenylflavonoids from hops to
beer. J. Agric. Food Chem. 1999, 47, 2421-2428.
(6) Gester, S.; Metz, P.; Zierau, O.; Vollmer, G. An efficient
synthesis of the potent phytoestrogens 8-prenylnaringenin and
6-(1,1-dimethylallyl)naringenin by europium(III)-catalyzed Clais-
en rearrangement. Tetrahedron 2001, 57, 1015-1018.
(7) Schaefer, O.; Hu¨mpel, M.; Fritzemeier, K.-H.; Bohlmann, R.;
Schleuning, W.-D. 8-Prenyl naringenin is a potent ERR selective
phytoestrogen in hops and beer. J. Steroid Biochem. Mol. Biol.
2003, 84, 359-360.
(8) Stevens, J. F.; Taylor, A. W.; Nickerson, G. B.; Ivancic, M.;
Henning, J.; Haunold, A.; Deinzer, M. L. Prenylflavonoid
variation in Humulus lupulus: distribution and taxonomic
significance of xanthogalenol and 4′-O-methylxanthohumol.
Phytochemistry 2000, 53, 759-775.
(9) Stevens, J. F.; Page, J. E. Xanthohumol and related prenylfla-
vonoids from hops and beer: To your good health! Phytochem-
istry 2004, 65, 1317-1330.
(10) Depypere, H.; Rong, H.; Boterberg, T.; Serreyn, R.; Van
Slambrouck, S.; De Keukeleire, D.; Stove, C.; Mareel, M.;
Bracke M. 8-Prenylnaringenin, the phytoestrogen in hops and
beer, upregulates the function of the E-cadherin/catenin complex
in human mammary carcinoma cells. Breast 2001, 10 (Suppl.
1), p S22.
Received for review December 14, 2004. Revised manuscript received
February 28, 2005. Accepted March 1, 2005. The work was in part
supported by a grant of the “Arbeitsgemeinschaft Industrieller Fors-
chungsgemeinschaften Otto von Guericke (AiF)”.
JF047897U