Sensomics Analysis of Key Bitter Compounds in the Hard Resin of Hops (Humulus lupulus L.) and Their Contribution to the Bitter Profile of Pilsner-Type Beer

Article abstract of DOI:10.1021/acs.jafc.5b00239

Recent brewing trials indicated the occurrence of valuable bitter compounds in the hard resin fraction of hop. Aiming at the discovery of these compounds, hop's ε-resin was separated by means of a sensory guided fractionation approach and the key taste mo

Full text of DOI:10.1021/acs.jafc.5b00239

Article
pubs.acs.org/JAFC
Sensomics Analysis of Key Bitter Compounds in the Hard Resin of
Hops (Humulus lupulus L.) and Their Contribution to the Bitter Profile
of Pilsner-Type Beer
†
†
,†,‡
Michael Dresel, Andreas Dunkel, and Thomas Hofmann*
†
̈
̈
Chair of Food Chemistry and Molecular and Sensory Science, Technische Universitat Munchen, Lise-Meitner-Straße 34, D-84354
Freising, Germany
‡
Bavarian Center for Biomolecular Mass Spectrometry, Gregor-Mendel-Straße 4, 85354 Freising, Germany
S Supporting Information
*
ABSTRACT: Recent brewing trials indicated the occurrence of valuable bitter compounds in the hard resin fraction of hop.
Aiming at the discovery of these compounds, hop’s ε-resin was separated by means of a sensory guided fractionation approach
and the key taste molecules were identified by means of UV/vis, LC-TOF-MS, and 1D/2D-NMR studies as well as synthetic
experiments. Besides a series of literature known xanthohumol derivatives, multifidol glucosides, flavon-3-on glycosides, and p-
coumaric acid esters, a total of 11 bitter tastants are reported for the first time, namely, 1″,2″-dihydroxanthohumol F, 4′-
hydroxytunicatachalcone, isoxantholupon, 1-methoxy-4-prenylphloroglucinol, dihydrocyclohumulohydrochinone, xanthohumols
M, N, and P, and isoxanthohumols M, N, and P, respectively. Human sensory analysis revealed low bitter recognition threshold
concentrations ranging from 5 (co-multifidol glucopyranoside) to 198 μmol/L (trans-p-coumaric acid ethyl ester) depending on
their chemical structure. For the first time, LC-MS/MS quantitation of these taste compounds in Pilsner-type beer, followed by
taste re-engineering experiments, revealed the additive contribution of iso-α-acids and the identified hard resin components to be
truly necessary and sufficient for constructing the authentic bitter percept of beer. Finally, brewing trails using the ε-resin as the
only hop source impressively demonstrated the possibility to produce beverages strongly enriched with prenylated hop
flavonoids.
KEYWORDS: hops, Humulus lupulus L., xanthohumol, bitter, taste dilution analysis, sensomics
2
6
INTRODUCTION
afford beverages with a strong and pleasant bitter character,
■
thus indicating the occurrence of additional valuable bitter
compounds.
Due to its characteristic aroma and taste, beer is widely
appreciated by consumers all over the world. A vast number of
studies have been focused in the past on the volatile aroma
components in hops and beers, as well as on the nonvolatile
terpenoids contributing to bitter taste, foam stability, and
The objective of the present investigation was, therefore, to
locate bitter compounds in hop’s hard resin fraction using an
activity-guided fractionation approach, to determine their
chemical structure by means of LC-MS and NMR experiments,
to characterize their sensory activity on the basis of human
threshold concentrations, and to verify their taste contribution by
means of taste re-engineering experiments. Furthermore, an LC-
MS/MS method should be developed for the quantitation of the
comprehensive set of hop bitter compounds and applied to the
analysis of hop products/extracts and beer samples produced by
using hard resin extract.
1
−8
9
−11
antimicrobial activity of the final beer.
The bitter principles of hops can be divided into two classes:
first, the α-acids (1, Figure 1) and β-acids (2), both present in the
so-called soft resin, are prone to isomerization into cis- (3) and
trans-iso-α-acids (4) during wort boiling and are considered the
major contributors to the beers’ bitter taste,
series of β-acid and trans-iso-α-acid transformation products
12−14
followed by a
15−17
adding to bitterness perception only to a minor extent.
In
addition to these soft resin derived compounds, the major hard
resin component xanthohumol (5) and its isomerization product
isoxanthohumol (6) as well as desmethylxanthohumol (7) and
its isomerization products 8-prenylnaringenin (8) and 6-
MATERIALS AND METHODS
■
Chemicals. The following compounds were obtained commercially:
formic acid, sulfuric acid, ethanol, hexane (Merck KGaA, Darmstadt,
Germany); hydrochloric acid, 0.1 M potassium hydroxide (Riedel-de-
Haen, Seelze, Germany); acetonitrile, methanol, hydrobromic acid
solution (33% in acetic acid), Amberlyst 15, scandium trifluorometha-
nesulfonate, trans-p-coumaric acid (Sigma-Aldrich, Steinheim, Ger-
18−20
prenylnaringenin (9),
respectively, have been proposed to
contribute to the bitter taste of beer, primarily upon coactivation
of the human bitter taste receptors hTAS2R1, hTAS2R14, and
2
1
̈
many); D O, CD OD, and DMSO-d (Euriso-Top, Saarbrucken,
hTAS2R40. Very recently, a series of α-acid derived
2
3
6
degradation products were isolated from the hard resin fraction
2
2
of stored hops. It is, however, not clear whether and, if so,
which additional hop components are required to fully construct
Received: January 14, 2015
Revised: March 18, 2015
Accepted: March 18, 2015
23−25
the unique perception of a beer’s bitter profile.
Interestingly,
brewing of beer using the hop’s hard resin fraction was found to
©
XXXX American Chemical Society
A
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Figure 1. Chemical structures of key taste compounds identified in the hop hard resin fraction: cohumulon (1a), n-humulon (1b), adhumulon (1c),
colupulon (2a), n-lupulon (2b), adlupulon (2c), cis-isocohumulon (3a), cis-isohumulon (3b), cis-isoadhumulon (3c), trans-isocohumulon (4a), trans-
isohumulon (4b), trans-isoadhumulon (4c), xanthohumol (5), isoxanthohumol (6), desmethylxanthohumol (7), 8-prenylnaringenin (8), 6-
prenylnaringenin (9), 4′-hydroxytunicatachalcone (10), isoxantholupon (11), 1″,2″-dihydroxanthohumol C (12), 1″,2″-dihydroisoxanthohumol C
(
(
13), 1″,2″-dihydroxanthohumol K (14), xanthohumol P (15), isoxanthohumol P (16), 5′-prenylxanthohumol (17), 1″,2″-dihydroxanthohumol F
18), xanthohumol D (19), xanthohumol B (20), xanthohumol C (21), xanthohumol H (22), isoxanthohumol H (23), xanthohumol N (24), 2″-
hydroxy-xanthohumol M (25), xanthohumol I (26), xanthohumol O (27), xanthohumol L (28), xanthohumol M (29), isoxanthohumol M (30), 2″,3″-
dehydrocyclohumulohydrochinon (31), 1-methoxy-4-prenylphloroglucinol (32), cis-/trans-p-coumaric acid (33a/b), cis-/trans-p-coumaric acid methyl
ester (34a/b), cis-/trans-p-coumaric acid ethyl ester (35a/b), quercetin (36), quercetin-3-O-β-D-glucopyranoside (37), kaempferol-3-O-β-D-
glucopyranoside (38), kaempferol-3-O-β-D-(6″-malonyl)glucopyranoside (39), 1-O-β-D-(2-methylpropanoyl)phloroglucinol glucopyranoside (co-
multifidol glucopyranoside) (40a), 1-O-β-D-(2-methylbutyryl)phloroglucinol glucopyranoside (ad-multifidolglucopyranoside) (40c), phloroisovaler-
ophenon-3,5-di-C-β-D-glucopyranoside (n-multifidol-di-C-glucopyranoside) (41b), and trans-N-feruloyltyramine (42).
Germany). Water for high-performance liquid chromatography
HPLC) and medium-pressure liquid chromatography (MPLC)
Hop pellets type 90 and spent hop pellets, which were obtained after
supercritical carbon dioxide extraction of hops, were delivered from Joh.
Barth & Sohn GmbH & Co. KG (Nuremberg, Germany); XanthoFlav
(
separation was purified by means of a Milli-Q water advantage A 10
water system (Millipore, Molsheim, France), and bottled water was used
for sensory studies (Evian, Danone, Wiesbaden, Germany). Solvents for
HPLC were of HPLC grade and for LC-MS were of MS grade
1
5, XanthoFlav 75, hop tannin extracts (from both aroma and bitter
hops), and resin extracts (from both aroma and bitter hops) were
obtained from Hopsteiner (Simon H. Steiner GmbH, Mainburg,
Germany). A novel ε-extract was produced in cooperation with Joh.
(
J.T.Baker, purchased through Sigma-Aldrich).
Barth & Sohn GmbH & Co. KG and NateCO GmbH & Co. KG
2
B
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Figure 2. RP-MPLC chromatogram (left) and taste dilution chromatogram (right) of the ε-resin fraction isolated from spent hops.
(
Wolnzach, Germany) by means of a supercritical CO extraction using
resin), whereas lyophilization of the residual material afforded the ε-
resin (79.8% of total hard resin).
Fractionation of the ε-Resin by Medium-Pressure Liquid
Chromatography (MPLC). Aliquots of the ε-resin (0.5 g) were
dissolved in methanol/water (40/60, v/v; 20 mL) and separated on a
2
27
ethanol as a modifier. Reference substances for individual iso-α-acids
were purified from an iso-α-acid extract (30%; Hopsteiner), and purified
α- and β-acids were isolated from an ethanolic hop extract (Hopsteiner,
Simon H. Steiner GmbH, Mainburg, Germany) following the protocol
reported recently. Xanthohumol was purified from a commercial
xanthohumol extract, and isoxanthohumol was prepared from
xanthohumol by alkaline cyclization as reported recently. 8-
Prenylnaringenin was prepared from isoxanthohumol by demethylation
using scandium trifluoromethanesulfonate. Experimental brewing
trails were conducted in the Research Brewery St. Johann St. Johann
GmbH & CO. KG (Train-St. Johann, Germany). Unhopped “zero” beer
was obtained from the Chair of Brewing and Beverage Technology,
28
4
4
60 × 16 mm glass column (Bu
0 μm LiChroprep RP18 material (Merck KGaA, Darmstadt, Germany)
chi Sepacore, Flawil, Switzerland) consisting
̈
chi, Flawil, Switzerland) filled with 25−
2
8
using an MPLC system (Bu
̈
of two C-605 pumps, a C-620 control unit, a C-660 fraction collector, a
6-way-injection valve, and a C-635 UV detector monitoring the effluent
at 272 nm. Data acquisition was performed by means of Sepacore
Control Chromatography Software, Version 1.0. Operating with a flow
rate of 30.0 mL/min and using 0.1% aqueous formic acid as solvent A
and methanol containing 0.1% formic acid as solvent B, chromatography
was performed starting with 40% solvent B for 1 min and, then,
increasing solvent B to 60% within 6 min, to 70% within 30 min and,
finally, to 100% within an additional 20 min, followed by isocratic
elution for 20 min. After 86.0 min, solvent B decreased again to 40% and
was kept for 10 min prior to the next injection. A total of 11 fractions
29
Technische Universitat
̈
̈
Munchen (Freising, Germany).
Preparative Isolation of Soft Resin and δ- and ε-Hard Resin.
30
According to the literature with some modifications, spent hops (100
g) were exhaustively extracted with methanol/diethyl ether (1/5, v/v;
9
(
00 mL) for 1 h while stirring. After the addition of hydrochloric acid
300 mL; 0.1 mol/L), the mixture was stirred for additional 10 min and
then filtered (320 mm, Macherey-Nagel, Duren, Germany), and the
(
e1−e11) were collected, separated from solvent in vacuum, and then
̈
33−35
used for a taste dilution analysis
to locate fractions e2 (yield: 3.4%),
organic layer was extracted with hydrochloric acid (3 × 100 mL; 0.1
mol/L) before the solvent was separated in vacuum. To remove waxes,
the residue was solubilized in methanol (400 mL), and, after incubation
at −20 °C overnight and filtration (round filter paper, 320 mm, Roth,
Karlsruhe, Germany), the solution was applied onto 55 μm, 70A, Strata
C18-E Giga Tube cartridges (10 g, 60 mL; Phenomenex, Aschaffenburg,
Germany), conditioned with methanol. Chlorophylls were separated by
flushing the cartridges with methanol (150 mL), and the effluent was
collected and separated from solvent in vacuum. The resin obtained was
e3 (yield: 3.0%), e4 (yield: 3.2%), e5 (yield: 3.6), e6 (yield: 3.2%), e7
yield: 2.6%), e8 (yield: 2.7%), e9 (yield: 18.4%), e10 (yield: 31.3%),
(
and e11 (yield: 22.9%) with the highest bitter impact (Figure 2).
Isolation of Key Bitter Molecules from MPLC Fractions of ε-
Resin. MPLC fractions were separated by means of preparative HPLC
on a 250 × 21.2 mm, 5 μm, Luna Phenyl-Hexyl column (Phenomenex,
Aschaffenburg, Germany) operated with a flow rate of 15 mL/min and
using 0.1% aqueous formic acid and acetonitrile containing 0.1% formic
acid as solvents. Separation of MPLC fractions e2 to e11 gave
subfractions e2-1 to e2-14, e3-1 to e3-13, e4-1 to e4-5, e5-1 to e5-5, e6-1
to e6-11, e7-1 to e7-16, e8-1 to e8-8, e9-1 to e9-9, e10-13 to e10-13, and
e11-1 to e11-15, all of which were collected, separated from solvent in
vacuum, and subjected to a taste dilution analysis (Figure 3).
Structure Determination of Bitter Compounds in ε-Resin
Subfractions. The individual ε-resin subfractions isolated above were
directly analyzed by means of LC-TOF-MS and 1D/2D-NMR
spectroscopy and, if needed, further purified by means of rechromatog-
raphy using a semipreparative 250 × 10.0 mm Hypersil RP-18 column
(Thermo Hypersil GmbH, Kleinostheim, Germany) using a solvent
̈
suspended in hexane and filtered (185 mm, Macherey-Nagel, Duren,
Germany). The filtrate was collected and separated from solvent in
vacuum to give the soft resin (91.9% yield), whereas the residue was
extracted with hexane (50 mL) to give the nonsoluble hard resin (8.1%
31,32
yield). Following a literature protocol for ε-resin isolation,
resin was suspended in water at 40 °C during ultrasonification. After
filtration (185 mm, Macherey-Nagel, Duren, Germany), the aqueous
phase was collected, whereas the residue was again suspended in water
40 °C). After this process was repeated three times, the combined
aqueous layer was freeze-dried to obtain the δ-resin (20.2% of total hard
the hard
̈
(
C
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Figure 3. RP-HPLC-TDA chromatogram of the ε-resin subfractions e2−e11 isolated from spent hops. * Fraction with an astringent taste.
−
gradient of aqueous 0.1% formic acid and methanol containing 0.1%
(12, [M − H] ), 119 (100), 93 (3), 217 (2), 245 (1). LC-TOF-MS
−
−
formic acid (flow rate: 4.7 mL/min). Compounds 5−7, 9−12, 14, 17−
(ESI ): m/z 421.2014 ([M − H] , measured), m/z 421.2015 ([M −
− − 1
2
2, 25−29, 31−34, 36, 37, 40a, and 42−44 were isolated from the
H] , calcd for C H O ). H NMR (500 MHz, MeOD; COSY): δ
26 29 5
subfractions as summarized in Table 1 in a purity of more than 98% and,
after the solvent was separated in vacuum and freeze-dried twice, were
used for structure determination by means of UV/vis, LC-MS/MS,
UPLC-TOF-MS, and 1D/2D NMR, as well as for sensory experiments.
Based on the comparison of spectroscopic (UV/vis, LC-TOF-MS,
H/ C NMR) and chromatographic data with those recorded for
references, previously reported compounds were confirmed as
(ppm) 1.56 [s, 4 × 3H, H−C(4″, 4″a, 5″, 5″a)], 2.58 [m, 3H, H−C(1″,
1″a)], 3.99 [s, 3H, H−C(1‴)], 4.84 [m, 2H, H−C(2″, 2″a)], 5.56 [s,
1H, H−C(6)], 6.85 [m, 2H, J = 8.9 Hz, H−C(3′, 5′)], 7.54 [m, 2H, J =
8.9 Hz, H−C(2′, 6′)], 7.61 [d, 1H, J = 15.7 Hz, H−C(3)], 7.88 [d, 1H, J
= 15.7 Hz, H−C(2)]. C NMR (125 MHz, MeOH; HSQC, HMBC): δ
(ppm) 17.9 [C(5″, 5″a)], 25.8 [C(4″, 4″a)], 39.5 [C(1″, 1″a)], 57.1
[C(1‴)], 62.6 [C(8)], 98.9 [C(6)], 106.1 [C(10)], 116.6 [C(3′, 5′)],
118.8 [C(2″, 2″a)], 119.1 [C(3)], 128.1 [C(1′)], 131.8 [C(2′, 6′)],
136.2 [C(3″, 3″a)], 145.9 [C(2)], 161.9 [C(4′)], 174.3 [C(5)], 180.3
[C(4)], 199.8 [C(7)], 205.9 [C(9)].
13
1
13
2
4
24
xanthohumol (5), isoxanthohumol (6), desmethylxanthohumol
2
5
24
(
(
7), 6-prenylnaringenin (9), 1″,2″-dihydroxanthohumol C
23,36,37 36,37
12),
1″,2″-dihydroxanthohumol K (14),
5′-prenylxantho-
24,25
24,25
25
humol (17),
xanthohumol D (19), xanthohumol B (20),
Isoxantholupon, 11 (Figure 4). UV/vis (ACN/water; 0.1% formic
2
4
23,36,37
−
−
xanthohumol C (21), xanthohumol H (22),
2″-hydroxyxan-
acid): λmax = 308 nm. LC-MS (ESI ): m/z (%) 475.2 (100, [M − H] ),
38
23,36,37
39
−
thohumol M (25), xanthohumol I (26),
xanthohumol O (27),
243 (95), 119 (55), 363 (29), 286 (26), 257 (15). LC-TOF-MS (ESI ):
3
9
29
−
−
xanthohumol L (28), cis-/trans-p-coumaric acid (33a/b), cis-/trans-
p-coumaric acid methyl ester (34a/b), quercetin (36), quercetin-3-
O-β-D-glucopyranoside (37),
m/z 475.2492 ([M − H] , measured), m/z 475.2484 ([M − H] , calcd
− 1
29
40
for C30H O ). H NMR (500 MHz, MeOD; COSY): δ (ppm) 1.51/
35 5
4
1
1-O-β-D-(2-methylpropanoyl)-
1.53/1.56/1.59 [s, 4 × 3H, H−C(4‴, 4‴a, 5‴, 5‴a)], 1.55 [s, 3H, H−
C(4″)], 1.61 [s, 3H, H−C(5″)], 2.60/2.62 [d, 2 × 2H, J = 6.75 Hz, J =
7.35 Hz, H−C(1‴, 1‴a)], 2.96 [dd, 1H, J = 7.0 Hz, J = 18.2 Hz, H−
C(3eq)], 3.00 [d, 2H, J = 7.0 Hz, H−C(1″)], 3.16 [dd, 1H, J = 9.8 Hz, J =
18.2 Hz, H−C(3ax)], 4.71/4.82 [t, 2 × 2H, J = 6.75 Hz, J = 7.35 Hz, H−
C(2‴, 2‴a)]; 4.89 [t, 2H, J = 7.0 Hz, H−C(2″)]; 5.3 [dd, 1H, J = 4.0 Hz,
J = 9.8 Hz, H−C(2)], 6.82 [m, 2H, J = 8.9, H−C(3′, 5′)], 7.28 [m, 2H, J
phloroglucinol glucopyranoside (co-multifidol glucopyranoside)
4
3
23,36,37
13
(
40a), trans-N-feruloyltyramine (42),
cohumulon (41a), n-
13
13
13
humulon (41b), adhumulon (41c), colupulon (44a), n-lupulon
1
3
13
(
44b), and adlupulon (44c). The chemical structures of compounds
1
0, 11, 18, 29, 31, and 32 have been determined by means of UV/vis,
LC-MS, and 1D/2D NMR experiments and, to the best of our
knowledge, have not been earlier reported (Figure 4).
13
= 8.9, H−C(2′, 6′)]. C NMR (125 MHz, MeOD; HSQC, HMBC): δ
(ppm) 17.9/18.1 [C(4″, 4‴, 4‴a)], 22.2 [C(1″)], 25.9/26.0 [C(5″, 5‴,
5‴a)], 38.5 [C(3)], 39.1/39.7 [C(1‴, 1‴a)], 61.9 [C(6)], 78.3 [C(2)],
4
′-Hydroxytunicatachalcone, 10 (Figure 4). UV/vis (ACN/
−
water; 0.1% formic acid): λ = 406 nm. LC-MS (ESI ): m/z (%) 421
max
D
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Table 1. Orosensory Quality and Human Recognition Threshold Concentration of Taste Compounds Identified in Hops’ Hard
Resin and Hop Extracts
d
d
sensory
b
TC
sensory
b
TC
a
c
a
c
compound (no.)
quality
fraction no.
[μmol/L]
compound (no.)
quality
fraction no.
e7-11
[μmol/L]
cis-p-coumaric acid (33a)
as
e2-4
e2-4
22
2″,3″-
dehydrocylcohumulohydrochinon
31)
bi
22
trans-p-coumaric acid (33b)
as
22
(
e
f
quercetin-3-O-β-D-glcp (37)
bi/as
bi/as
bi/as
e2-5; −
28/2
29/1
35/5
desmethylxanthohumol (7)
″,2″-dihydroxanthohumol F (18)
bi
bi
bi
bi
bi
bi
bi
e7-12
16
12
14
19
10
8
f
kaempferol-3-O-β-D-glcp (38)
−
1
e7-14
f
kaempferol-3-O-β-D-(6″-malonyl)-
glcp (39)
−
xanthohumol M (29)
isoxanthohumol M (30)
6-prenylnaringenin (9)
8-prenylnaringenin (8)
xanthohumol (5)
e8-3; e9-4
g
f
−
co-multifidol glcp (40a)
ad-multifidol glcp (40c)
n-multifidol di-C-glcp (41b)
trans-N-feruloyltyramine (42)
xanthohumol L (28)
bi
bi
bi
bi
bi
bi
as
bi
e2-7; −
5
f
e8-5
−
10
37
10
8
g
f
−
−
e8-7; e9-6;
e10-2
10
e2-11
e2-13
e3-4
1
″,2″-dihydroxanthohumol C (12)
bi
bi
bi
e10-4; e11-4
e10-5; e11-5
8
6
6
xanthohumol I (26)
29
15
44
xanthohumol C (21)
quercetin (36)
e3-5
g
1
″,2″-dihydroisoxanthohumol C
13)
−
1
-methoxy-4-prenylphloroglucinol
22)
e3-7
(
(
4
′-hydroxytunicatachalcone (10)
bi
bi
bi
bi
bi
bi
bi
bi
bi
bi
bi
e10-6
7
cis-p-coumaric acid methyl ester
34a)
bi
bi
e3-7
184
189
n-humulon (1b)
e10-8; e11-8
e10-9; e11-9
e10-10; e11-8
e10-12
21
21
7
(
adhumulon (1c)
trans-p-coumaric acid methyl ester
34b)
(
5′-prenylxanthohumol (17)
isoxantholupon (11)
colupulon (2a)
g
cis-p-coumaric acid ethyl ester (35a)
bi
bi
−
195
198
6
g
trans-p-coumaric acid ethyl ester
−
e11-7
17
35
37
13
20
24
(
35b)
xanthohumol O (27)
″,2″-dihydroxanthohumol K (14)
n-lupulon (2b)
e11-11
bi
bi
e3-9
9
adlupulon (2c)
e11-13
1
e3-11; e4-4;
e7-3
16
g
xanthohumol N (24)
xanthohumol P (15)
isoxanthohumol P (16)
−
−
−
g
isoxanthohumol (6)
bi
e4-2; e5-3; e6-
; e9-2
16
g
5
a
b
xanthohumol H (22)
isoxanthohumol H (23)
xanthohumol B (20)
bi
bi
bi
bi
bi
e6-7; e7-4
25
25
15
9
Chemical structures are given in Figure 1. Sensory quality:
g
c
−
astringent (as), bitter (bi). Number of HPLC fraction referring to
Figure 3. Taste recognition threshold concentrations (TC) were
determined in aqueous ethanolic solution (5%, v/v; pH 4.4). glcp:
d
e7-7
e7-8
e7-9
e
2″-hydroxyxanthohumol M (25)
f
glucopyranoside. Compound isolated from hop tannin extract.
xanthohumol D (19)
24
g
Compound synthesized.
1
2
1
1
05.3 [C(10)], 114.8 [C(8)], 116.5 [C(3′, 5′)], 118.9/119.0 [C(2‴,
‴a)], 123.5 [C(2″)], 129.1 [C(2′, 6′)], 130.3 [C(1′)], 131.9 [C(3″)],
36.1/136.2 [C(3‴, 3‴a)], 159.2 [C(4′)], 165.7 [C(9)], 185.7 [C(4)],
98.6 [C(7)].
NMR (125 MHz, MeOD; HSQC, HMBC): δ (ppm) 18.6 [C(1″)], 26.4
[C(4″, 5″)], 39.4 [C(2″)], 50.6 [C(6″)], 57.1 [C(1‴)], 77.5 [C(3″)],
92.5 [C(6)], 110.8 [C(8)], 117.8 [C(3′, 5′)], 126.7 [C(3)], 129.3
[C(1′)], 132.1 [C(2′, 6′)], 144.3 [C(2)], 161.9 [C(4′)], 163.3 [C(5)],
1
″,2″-Dihydroxanthohumol F, 18 (Figure 4). UV/vis (ACN/water;
1
64.7 [C(9)], 167.1 [C(7)], 195.0 [C(4)].
−
0
.1% formic acid): λ = 326 nm. LC-MS (ESI ): m/z (%) 421 (100,
max
2″,3″-Dehydrocyclohumulonhydrochinon, 31 (Figure 4). UV/vis
−
−
[
(
M − H] ), 119 (98), 365 (10). LC-TOF-MS (ESI ): m/z 421.2008
−
(
(
ACN/water; 0.1% formic acid): λ = 280 nm. LC-MS (ESI ): m/z
max
−
−
[M − H] , measured), m/z 421.2015 ([M − H] , calcd for
−
−
−
%) 291 (100, [M − H] ), 289 (95, [M − 3H] ), 207 (25, [M − H] ),
−
1
C H O ). H NMR (500 MHz, MeOD; COSY): δ (ppm) 1.22 [s,
−
−
2
6
29
5
235 (21), 136 (20). LC-TOF-MS (ESI ): m/z 291.1222 ([M − H] ,
2
× 3H, H−C(4″, 5″)], 1.68 [s, 3H, H−C(4‴)], 1.76 [s, 3H, H−
−
−
1
measured), m/z 291.1238 ([M − H] , calcd for C H O ). H NMR
16
19
5
C(5⁗)], 1.79 [t, 2H, J = 6.8 Hz, H−C(2″)], 2.65 [t, 2H, J = 6.8 Hz, H−
C(1″)], 3.30 [2H, H−C(1⁗)], 3.61 [s, 3H, H−C(1‴)], 5.20 [m, 1H,
H−C(2⁗)], 6.80 [m, 2H, H−C(3′, 5′)], 6.82 [d, 1H, J = 15.9 Hz, H−
C(2)], 7.25 [d, 1H, J = 15.9 Hz, H−C(3)], 7.43 [m, 2H, H−C(2′, 6′)].
(
3
2
500 MHz, MeOD; COSY): δ (ppm) 1.41 [s, 6H, H−C(1′, 2′)], 1.64 [s,
H, H−C(4″)], 1.76 [s, 3H, H−C(5″)], 2.67 [s, 2H, H−C(3)], 3.20 [d,
13
H, J = 7.3 Hz, H−C(1″)], 5.15 [m, 1H, J = 7.3 Hz, H−C(2″)].
C
NMR (125 MHz, MeOD; HSQC, HMBC): δ (ppm) 16.6 [C(5″)], 21.2
1
3
C NMR (125 MHz, MeOH; HSQC, HMBC): δ (ppm) 17.98
[
[
1
C(1″)], 24.5 [C(4″)], 25.5 [C(1′, 2′)], 49.5 [C(3)], 78.0 [C(2)], 101.0
C(10)], 107.8 [C(8)], 122.8 [C(2″)], 129.9 [C(3″)], 147.6 [C(5)],
[
C(5⁗)], 18.20 [C(1″)], 23.53 [C(1⁗)], 25.91 [C(4⁗)], 26.79 [C(4″,
5
1
1
1
″)], 33.12 [C(2″)], 63.09 [C(1‴)], 75.38 [C(3″)], 106.57 [C(8)],
15.06 [C(6)], 116.43 [C(10)], 116.97 [C(3′, 5′)], 124.76 [C(2⁗)],
27.03 [C(1′)], 127.48 [C(3)], 131.48 [C(2′, 6′)], 131.92 [C(3⁗)],
47.57 [C(2)], 151.71 [C(9)], 156.33 [C(5)], 156.44 [C(7)], 161.50
51.7 [C(9)], 154.1 [C(7)], 197.3 [C(4)].
1
-Methoxy-4-prenylphloroglucinol, 32 (Figure 4). UV/vis (ACN/
−
water; 0.1% formic acid): λ = 292 nm. LC-MS (ESI ): m/z (%) 207
max
−
(
(
100, [M − H] ), 137 (25), 121 (14), 109 (10), 149 (9). LC-TOF-MS
[
C(4′)], 198.33 [C(4)].
−
−
ESI ): m/z 207.1035 ([M − H] , measured), m/z 207.1027 ([M −
Xanthohumol M, 29 (Figure 4). UV/vis (ACN/water; 0.1% formic
−
−
1
−
−
H] , calcd for C H O ). H NMR (500 MHz, MeOD; COSY): δ
acid): λmax = 368 nm. LC-MS (ESI ): m/z (%) 385 (100, [M − H] ),
12 15 3
−
(ppm) 1.65 [s, 3H, H−C(4′)], 1.75 [s, 3H, H−C(5′)], 3.2 [d, 2H, J =
7.3 Hz, H−C(1′)], 3.83 [s, 3H, H−C(1″)], 5.21 [m, 1H, J = 7.3 Hz, H−
1
19 (87), 265 (2), 163 (1). LC-TOF-MS (ESI ): m/z 385.1664 ([M −
− − − 1
H] , measured), m/z 353.1657 ([M − H] , calcd for C H O ). H
22
25
6
13
C(2′)], 5.42 [s, 1H, H−C(2)], 6.04 [s, 1H, H−C(6)]. C NMR (125
MHz, MeOD; HSQC, HMBC): δ (ppm) 17.8 [C(5′)], 22.0 [C(1′)],
25.9 [C(4′)], 56.2 [C(1″)], 93.4 [C(6)], 99.8 [C(2)], 105.9 [C(4)],
123.2 [C(2″)], 132.4 [C(3′)], 159.6 [C(1)], 167.9 [C(5)], 173.3
[C(3)].
NMR (500 MHz, MeOD; COSY): δ (ppm) 1.23 [s, 6H, H−C(4″, 5″)],
.65 [m, 2H, H−C(2″)], 2.57 [m, 2H, H−C(1″)], 3.28 [s, 3H, H−
1
C(6″)], 3.91 [s, 3H, H−C(1‴)], 6.04 [s, 1H, H−C(6)], 6.83 [m, 2H, J =
8.5 Hz, H−C(3′, 5′)], 7.50 [m, 2H, J = 8.5 Hz, H−C(2′, 6′)], 7.67 [d,
1H, J = 15.6 Hz, H−C(3)], 7.80 [d, 1H, J = 15.6 Hz, H−C(2)].
13
C
E
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Figure 4. Chemical structures of compounds 10, 11, 15, 17, 18, 24, and 29−32 with heteronuclear key correlations (HMBC experiment) highlighted.
Figure 5. RP-HPLC/UV chromatogram of a methanolic solution of xanthohumol (5) after treatment (45 min, rt) with hydrobromic acid solution (33%
in acetic acid). Numbering of compounds refer to structures given in Figure 1
Isolation of Compounds from Hop Tannin Extract. Aliquots of
the hop tannin extract (200 μL), diluted with water (500 μL), were
separated by means of HPLC on a 250 × 21.2 mm, 5 μm, Luna Phenyl-
Hexyl column (Phenomenex, Aschaffenburg, Germany) operated with a
flow rate of 15 mL/min and using 0.1% aqueous formic acid and
acetonitrile containing 0.1% formic acid as solvents. Chromatography
was performed by starting with 5% solvent B for 0.5 min, then increasing
solvent B to 15% within 0.5 min, and, after keeping at 15% for 34.0 min,
solvent B was increased to 25% within additional 7 min, then increased
to 100% within 1 min, followed by isocratic elution for 3.0 min. After 46
min, solvent B decreased again to 5% within 1.0 min and was kept for 3.0
min prior to the next injection. The compounds eluting at 24.5 (37),
from solvent in vacuum and freeze-dried twice. For preparative isolation
of additional compounds, the following gradient was used: 5% solvent B
for 0.5 min and then increasing solvent B to 15% within 0.5 min. Solvent
B was kept for 41.0 min at 15% and, finally, increased to 100% within an
additional minute, followed by isocratic elution for 3.0 min. After 46
min, solvent B decreased again to 5% within 1.0 min and was kept for 3.0
min prior to the next injection. Compounds eluting after 31.5 (40a),
32.3 (38), 33.0 (40c), and 33.3 min (41b) were collected, separated
from solvent under vacuum, and freeze-dried twice. Compounds 37−
41b (>98% purity) were analyzed by means of UV/vis, LC-MS/MS,
UPLC-TOF-MS, and 1D/2D NMR. Based on the comparison of
spectroscopic and chromatographic data with those recorded for
references, previously reported compounds were confirmed as
2
7.0 (40a), 30.0 (38), and 32.0 min (39) were collected and separated
F
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
4
1
quercetin-3-O-β-D-glucopyranoside (37), kaempferol-3-O-β-D-gluco-
acid. The solution was stirred under reflux for 24 h. Afterward the
reaction mixture was subjected to a solid phase extraction using ethanol
as an eluent. A HPLC separation with two injections of the crude extract
after 0.0 and 3.0 min, using a gradient as follows, was applied to achieve
fast separation of the cis- and trans-isomers: 41% solvent B for 20 min
and, then, increasing solvent B to 100% within 0.5 min, followed by an
isocratic elution for 1 min. After 21.5 min, solvent B decreased again to
41% within 0.1 min and was kept for 3.0 min prior to the next injection.
After removing the solvent in vacuum and freeze-drying twice,
compounds 35a and 35b were isolated (>98% purity) and were used
for sensory experiments as well as UV/vis, TOF-MS, and 1D/2D NMR.
Based on the comparison of spectroscopic and chromatographic data
with those recorded for references, compounds reported earlier in the
literature were identified as cis-/trans-p-coumaric acid ethyl ester (35a/
41
pyranoside (38), kaempferol-3-O-β-D-(6″-malonyl)glucopyranoside
42
(
39), 1-O-β-D-(2-methylpropanoyl)phloroglucinol glucopyranoside
43
(
co-multifidol glucopyranoside, 40a), 1-O-β-D-(2-methylbutyryl)-
phloroglucinol glucopyranoside (ad-multifidolglucopyranoside,
43
4
0c), and phloroisovalerophenon-3,5-di-C-β-D-glucopyranoside (n-
43
multifidol-di-C-glucopyranoside) (41b).
Synthetic Preparation of Xanthohumol Derivatives 12−14,
2
2−24, 29, and 30. An aliquot (50 mg) of xanthohumol (5) was solved
in a mixture of hydrobromic acid solution (33% in acetic acid; 5 mL) and
methanol (5 mL) and stirred for 45 min at room temperature. The
reaction mixture was separated by semipreparative HPLC on a 250 ×
1
0.0 mm Hypersil RP-18 column (Thermo Hypersil GmbH,
Kleinostheim, Germany) using water (0.1% formic acid) as solvent A
and methanol (0.1% formic acid) as solvent B at a flow rate of 4.7 mL/
min. Chromatography was performed using the following gradient: 62%
solvent B for 4.5 min, then increasing solvent B to 64% within 5.5 min.
After 3.0 min solvent B was increased to 70% within 10 s, then to 72%
within 5.3 min, to 80% within 0.5 min, kept for 6.0 min and, finally, to
00% within an additional 1.5 min, followed by isocratic elution for 2.5
min. After 29.0 min, solvent B decreased again to 62% within 1 min and
was kept for 3.0 min prior to the next injection. Besides xanthohumol
5) and isoxanthohumol (6), the effluent of the peaks of compounds
2−14, 22−24, 29, and 30 (Figure 5) was collected and, after the
solvent was removed in vacuum and freeze-dried twice, the individual
compounds (>98% purity) were used for sensory and identification
experiments (UV/vis, LC-MS/MS, UPLC-TOF-MS, 1D/2D NMR).
Based on the comparison of spectroscopic (UV/vis, LC-TOF-MS,
2
9
b).
Synthetic Preparation of Xanthohumol P (15). An aliquot (25
mg) of xanthohumol (5) and Amberlyst 15 (1 g) was stirred in ethanol
(10 mL) for 4 h at 80 °C. HPLC purification of 15 was achieved using a
gradient as follows: 30% solvent B for 2.0 min and, then, increasing
solvent B to 80% within 12.5 min and, finally, increasing solvent B to
100% within 1.0 min, followed by isocratic elution for 2.5 min. After 16.0
min, solvent B decreased again to 30% within 1.0 min and was kept for
3.0 min prior to the next injection. The compound eluting after 15.3 min
was isolated (>98% purity) and, after the solvent was removed in
vacuum and freeze-dried twice, was used for sensory experiments.
Structure verification by means of UV−vis, LC-MS/MS, UPLC-TOF-
MS, and 1D/2D NMR undoubtedly identified the compound as
xanthohumol P, 15 (Figure 4), which has to the best of our knowledge
not been described yet in the literature. In addition, compounds 6 and
12−14 were isolated from the reaction mixture.
1
(
1
1
13
H/ C NMR) and chromatographic data (retention time) with those
recorded for reference substances, compounds reported earlier in the
2
3,36,37
literature were identified as 1″,2″-dihydroxanthohumol C (12),
Xanthohumol P, 15 (Figure 1). UV/vis (ACN/water; 0.1% formic
4
5
−
−
1
″,2″-dihydroisoxanthohumol C (13), 1″,2″-dihydroxanthohumol K
acid): λmax = 367 nm. LC-MS (ESI ): m/z (%) 399 (57, [M − H] ), 119
3
6,37
23,36,37
39
−
(14),
xanthohumol H (22),
and isoxanthohumol H (23).
(100), 353 (64), 232 (20), 174 (13). LC-TOF-MS (ESI ): m/z
−
−
Furthermore, compound 29 was identified as xanthohumol M based on
the comparison with the compound isolated above form the ε-resin,
whereas compounds 24 and 30 have not been previously reported
399.1802 ([M − H] , measured), m/z 399.1808 ([M − H] , calcd for
− 1
C H O ). H NMR (400 MHz, MeOD; COSY): δ (ppm) 1.18 [t,
23 27 6
3H, H−C(7″)], 1.23 [s, 6H, H−C(4″, 5″)], 1.65 [m, 2H, J = 7.1, J = 8.5,
H−C(2″)], 2.58 [m, 2H, J = 7.1, J = 8.5), H−C(1″)], 3.53 [dd, 2H, H−
C(6″)], 3.90 [s, 3H, H−C(1‴)], 6.03 [s, 1H, H−C(6)], 6.87 [m, 2H, J =
(
Figure 4).
Xanthohumol N, 24 (Figure 4). UV/vis (ACN/water; 0.1% formic
acid): λmax = 368 nm. LC-MS (ESI ): m/z (%) 353 (100, [M − H] ),
19 (35), 163 (15), 233 (12), 283 (8). LC-TOF-MS (ESI ): m/z
53.1394 ([M − H] , measured), m/z 353.1394 ([M − H] , calcd for
C H O ). H NMR (500 MHz, MeOD; COSY): δ (ppm) 1.79 [s,
−
−
8.7 Hz, H−C(3′, 5′)], 7.50 [m, 2H, J = 8.7 Hz, H−C(2′, 6′)], 7.67 [d,
−
13
1
3
1H, J = 15.5 Hz, H−C(3)], 7.80 [d, 1H, J = 15.5 Hz, H−C(2)].
C
−
−
NMR (100 MHz, MeOD; HSQC, HMBC): δ (ppm) 16.3 [C(7″)], 17.8
[C(1″)], 26.3 [C(4″, 5″)], 39.0 [C(2″)], 56.2 [C(1‴)], 57.5 [C(6″)],
76.4 [C(3″)], 91.7 [C(6)], 106.5 [C(10)], 110.0 [C(8)], 116.9 [C(3′,
5′)], 125.9 [C(3)], 128.5 [C(1′)], 131.3 [C(2′, 6′)], 143.4 [C(2)],
161.1 [C(4′)], 162.4 [C(5)], 163.8[C(9)], 166.2 [C(7)], 194.1 [C(4)].
Alkaline Isomerization of Compounds 9, 12, 15, and 29. An
aliquot (10 mg) of the compound 9, 12, 15, and 29, respectively, was
dissolved in 5 mL of methanol and, after adjusting the pH to a value of 14
with potassium hydroxide (0.1 mol/L), stirred for 30 min in a closed
glass vial at 60 °C. The reaction mixture was diluted with water and
subjected to a solid phase extraction using 55 μm, 70A, Strata C18-E
Giga Tube cartridges (200 mg, 3 mL; Phenomenex, Aschaffenburg,
Germany) and methanol as an eluent, prior to HPLC separation on a
250 × 21.2 mm, 5 μm, Luna Phenyl-Hexyl column (Phenomenex,
Aschaffenburg, Germany). The HPLC system was operated with a flow
rate of 15 mL/min, using 0.1% aqueous formic acid and acetonitrile
containing 0.1% formic acid as solvents, and using a gradient as follows:
20% solvent B for 2.5 min and then increasing solvent B to 100% within
20.0 min followed by isocratic elution for 2.5 min. After 25.0 min,
solvent B decreased again to 20% within 2.0 min and was kept for 3.0
min prior to the next injection. After the solvent was removed in vacuum
and freeze-dried twice, compounds 7, 8, 13, 16, and 30 (>98% purity)
were used for sensory experiments and structure determination. Based
−
1
2
1
21
5
3
H, H−C(5″)], 2.18 [m, 2H, J = 8.1, H−C(2″)], 2.69 [m, 2H, J = 8.1,
H−C(1″)], 3.28 [s, 3H, H−C(6″)], 3.90 [s, 3H, H−C(1‴)], 4.66 [s, 2H
H−C(4″)], 6.03 [s, 1H, H−C(6)], 6.83 [m, 2H, J = 8.7 Hz, H−C(3′,
5
′)], 7.50 [m, 2H, J = 8.7 Hz, H−C(2′, 6′)], 7.67 [d, 1H, J = 15.5 Hz, H−
13
C(3)], 7.80 [d, 1H, J = 15.5 Hz, H−C(2)]. C NMR (125 MHz,
MeOD; HSQC, HMBC): δ (ppm) 22.1 [C(1″)], 22.6 [C(5″)], 37.9
[
1
[
C(2″)], 56.2 [C(1‴)], 91.6 [C(6)], 106.5 [C(10)], 109.7 [C(8)],
10.1 [C(4″)], 116.9 [C(3′, 5′)], 125.9 [C(3)], 128.5 [C(1′)], 131.2
C(2′, 6′)], 143.3 [C(2)], 161.0 [C(4′)], 162.5 [C(5)], 163.9 [C(7)],
1
66.2 [C(9)], 194.1 [C(4)].
Isoxanthohumol M, 30 (Figure 4). UV/vis (ACN/water; 0.1%
−
formic acid): λmax = 289 nm. LC-MS (ESI ): m/z (%) 385 (8, [M − H]-
−
)
, 119 (100), 265 (19), 165 (6), 163 (5), 197(4). LC-TOF-MS (ESI ):
−
−
m/z 385.1664 ([M − H] , measured), m/z 353.1657 ([M − H] , calcd
for C H O ). H NMR (400 MHz, MeOD; COSY): δ (ppm) 1.13 [s,
−
1
22
25
6
6
2
1
H, H−C(4″, 5″)], 1.60 [m, 2H, H−C(2″)], 2.53 [m, 2H, H−C(1″)],
.66 [dd, 1H, J = 3.0, J = 16.9, H−C(3eq)], 3.00 [dd, 1H, J = 12.9, J =
6.9, H−C(3ax)], 3.06 [s, 3H, H−C(6″)], 3.80 [s, 3H, H−C(1‴)], 5.30
[
dd, 1H, J = 3.0, J = 12.9, H−C(2)], 6.13 [s, 1H, H−C(6)], 6.81 [m, 2H,
13
J = 8.7 Hz, H−C(3′, 5′)], 7.33 [m, 2H, J = 8.7 Hz, H−C(2′, 6′)].
C
NMR (100 MHz, MeOD; HSQC, HMBC): δ (ppm) 18.1 [C(1″)],
1
13
2
5.6/27.2 [C(4″, 5″)], 38.7 [C(2″)], 46.2 [C(3)], 49.3 [C(6″)], 56.0
on the comparison of spectroscopic (UV/vis, LC-TOF-MS, H/ C
NMR) and chromatographic data (retention time) with those recorded
for reference substances, the following compounds were identified:
[
[
[
C(1‴)], 76.4 [C(3″)], 80.1 [C(2)], 93.5 [C(6)], 105.9 [C(10)], 110.5
C(8)], 116.2 [C(3′, 5′)], 129.0 [C(2′, 6′)], 131.4 [C(1′)], 158.9
C(4′)], 161.9 [C(5)], 163.9 [C(9)], 164.4 [C(7)], 192.9 [C(4)].
Synthetic Preparation of cis/trans-p-Coumaric Acid Ethyl
2
5
25
desmethylxanthohumol (7), 8-prenylnaringenin (8), 1″,2″-dihy-
4
5
droisoxanthohumol C (13), and isoxanthohumol M (30). To the best
of our knowledge, isoxanthohumol P, 16 (Figure 4), has not been earlier
reported in the literature.
Ester (35a/b). An aliquot (10 mL) of an ethanolic solution of trans-p-
coumaric acid (33b; 100 mmol/L) was acidified with 0.1 mL of sulfuric
G
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Table 2. Concentrations of Taste-Active Soft and Hard Resin Components in Commercial Hop Products
concn (μmol/100 g)
tannin extract
bitter
hard resin
pellets
Xantho Flav extract
resin extract
aroma
soft resin
Taurus
a
tastant
aroma hops
hops
Perle
Taurus
Herkules
15%
75%
hops
bitter hops
Perle
ε-extract
1
1
1
2
2
2
3
3
3
4
4
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
a
b
c
a
b
c
a
b
c
a
b
c
799.5
798.2
378.8
730.4
644.6
279.1
8.91
559.0
818.3
363.9
650.7
441.1
233.9
681.4
857.3
362.0
679.7
453.7
212.6
694.9
833.2
450.9
785.6
724.8
428.8
15987
15964
7573
14605
12892
5584
178.1
137.9
216.6
19.7
590.1
731.6
342.6
815.6
825.9
419.4
578.3
639.2
307.3
644.4
606.5
268.7
8727
12139
5403
8216
11665
4898
14720
9396
3783
7.13
5820
9533
3904
12603
7971
4259
5786
12327
4083
251.7
423.8
173.9
614.2
456.5
188.9
0.70
14808
15763
5101
19220
11818
6364
4.54
2.89
2.64
1.47
3.04
14.09
14.56
5.28
6.91
5.88
13.54
0.34
16.05
19.14
1.47
17.40
14.03
0.66
2.76
6.33
4.93
5.00
57.21
129.45
6.18
41.93
96.96
6.57
15.85
28.90
2.86
21.09
16.85
3.86
1.07
10.83
0.98
1.60
0.21
0.69
0.46
<0.01
<0.01
101.3
30.83
<0.01
11.41
55.26
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
5.00
<0.01
<0.01
100.5
45.44
17.14
18.05
49.97
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
33.88
<0.01
7.33
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
4993
265.7
57.25
57.84
139.1
34.56
1575
3.56
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
1082
112.8
<0.01
32.23
73.35
4.84
<0.01
<0.01
736.1
98.08
5.50
<0.01
<0.01
35.18
1.57
<0.01
<0.01
629.6
126.5
<0.01
34.80
52.50
<0.01
<0.01
8601
7671
21.06
658.9
1154
<0.01
6879
66.13
<0.01
37.74
453.6
762.1
7155
87.09
438.5
259.7
1206
50.85
41.41
245.5
<0.01
681.3
<0.01
569.3
15.52
38.88
<0.01
2086
<0.01
691.0
183.5
<0.01
<0.01
<0.01
<0.01
3408
124.08
<0.01
<0.01
62811
7202
93.68
33031
4817
15.61
653.6
1067
35953
13916
<0.01
207646
9754
670.1
<0.01
1.18
3449
6532
1319
2524
3478
25.26
57.15
36.83
<0.01
1.08
13873
3.01
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
16383
241.4
<0.01
156.1
8142
140.3
140.2
2.18
1193
<0.01
20347
733.3
7540
919.8
341.2
4107
511.6
1651
1017
370.5
<0.01
70.73
<0.01
717.5
397.4
752.1
504.7
<0.01
<0.01
0.98
<0.01
2.26
<0.01
41.16
<0.01
<0.01
<0.01
13.67
4571
218.5
21723
<0.01
6.55
317.5
112.9
601.9
896.0
1149
<0.01
2306
1004
983.8
<0.01
<0.01
1562
<0.01
2854
757.9
1676
23.96
1429
<0.01
1568
<0.01
276.8
<0.01
<0.01
2063
82522
351.8
1864
1178
348.1
380.7
<0.01
2.01
<0.01
<0.01
11.43
2.82
<0.01
2.38
363.4
271.4
180.3
32.09
464.0
<0.01
758.4
133.2
303.7
8.9
11.26
5.95
0.55
<0.01
20.48
<0.01
<0.01
<0.01
11.90
<0.01
<0.01
0.21
20.94
<0.01
<0.01
<0.01
14.43
<0.01
7.58
20445
58.27
<0.01
14.41
20.24
42.05
<0.01
<0.01
16.89
<0.01
17.57
<0.01
27.61
1.02
90.45
<0.01
12.02
8.45
1320
1967
1139
<0.01
114.2
<0.01
74.18
<0.01
<0.01
6.56
1797
745.7
570.1
815.1
5165
1032
31.31
<0.01
<0.01
12.26
<0.01
7.08
<0.01
<0.01
3.02
7331
1715
1316
2797
1171
466.8
<0.01
166.2
116.0
34.27
97.36
67.85
<0.01
261.1
<0.01
68.38
227.2
<0.01
<0.01
26.12
5622
31869
250.0
4639
1415
263.2
100.6
98.55
330.5
<0.01
<0.01
<0.01
<0.01
<0.01
27.97
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
9775
1238
2491
<0.01
<0.01
<0.01
41.93
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
42.55
127.5
<0.01
<0.01
<0. 01
<0.01
0.47
<0.01
<0.01
<0.01
<0.01
15.94
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
293.7
401.5
<0.01
129.9
146.6
<0.01
<0.01
289.2
1459
983.5
20354
827.3
601.9
4548
850.4
11656
626.3
2138
1344
110.0
1924
1022
2357
3801
<0.01
507.8
4189
5447
1318
2906
1820
252.0
1637
21.19
17.65
<0.01
3.73
22.40
95.80
<0.01
0.56
<0.01
<0.01
<0.01
<0.01
<0.01
2.59
<0.01
<0.01
<0.01
<0.01
156.4
<0.01
<0.01
27.16
28.98
197.5
12.02
20.42
19.94
8.95
<0.01
<0.01
<0.01
<0.01
<0.01
<0.01
55.21
15.06
<0.01
79.87
21.80
<0.01
7.27
1418
2997
<0.01
229.7
1239
661.8
45105
3a/b
4a/b
5a/b
6
1065
1777
<0.01
1041
831.4
3436
254.88
<0.01
3.32
7
5927
4623
813.2
8
14541
4280
<0.01
<0.01
3.85
9
124.4
2968
0a
0c
1b
2
1363
4802
1235
1038
297.88
219.35
<0.01
585.1
428.7
<0.001
800.6
<0.01
1264
1.31
<0.01
<0.01
<0.01
2999
17.19
a
Numbering of molecules is referring to structures given in Figure 1.
Isoxanthohumol P, 16 (Figure 1). UV/vis (ACN/water; 0.1% formic
2.67 [dd, 1H, J = 3.2, J = 16.9, H−C(3eq)], 3.02 [dd, 1H, J = 12.8, J =
16.9, H−C(3ax)], 3.34 [dd, 2H, H−C(6″)], 3.79 [s, 3H, H−C(1‴)],
5.32 [dd, 1H, J = 3.2, J = 12.8, H−C(2)], 6.12 [s, 1H, H−C(6)], 6.81 [m,
2H, J = 8.8 Hz, H−C(3′, 5′)], 7.33 [m, 2H, J = 8.8 Hz, H−C(2′, 6′)]. C
NMR (100 MHz, MeOD; HSQC, HMBC): δ (ppm) 15.8 [C(7″)], 17.5
[C(1″)], 25.3/27.1 [C(4″)/C(5″)], 38.0 [C(2″)], 45.9 [C(3)], 55.3
−
−
acid): λmax = 287 nm. LC-MS (ESI ): m/z (%) 399 (10, [M − H] ), 119
−
−
(
100), 265 (21). LC-TOF-MS (ESI ): m/z 399.1804 ([M − H] ,
−
−
1
13
measured), m/z 399.1808 ([M − H] , calcd for C H O ). H NMR
23
27
6
(
[
400 MHz, MeOD; COSY): δ (ppm) 1.12 [s, 6H, H−C(4″, 5″)], 1.12
t, 3H, H−C(7″)], 1.60 [m, 2H, H−C(2″)], 2.54 [m, 2H, H−C(1″)],
H
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Table 3. Concentration of Hop-Derived Taste Components in a Commercial Pilsner-Type Beer and Beer Samples I−IV from a
Brewing Trial Using ε-Extract and Spent Hops, Respectively
c
c
concn (μmol/L)
concn (μmol/L)
a
b
a
b
tastant
Pilsner beer
beer I
beer II
beer III
beer IV
tastant
Pilsner beer
beer I
beer II
beer III
beer IV
1
1
1
2
2
2
3
3
3
4
4
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
a
b
c
a
b
c
a
b
c
a
b
c
1.30
0.99
3.34
0.68
0.16
1.45
0.04
0.02
17.63
26.17
7.16
5.23
7.44
<0.01
0.31
<0.01
0.08
<0.01
0.09
0.59
0.39
0.05
0.18
<0.01
0.01
0.03
0.22
<0.01
<0.01
0.04
0.28
0.05
2.27
23.56
28.14
4.53
0.95
0.37
0.10
0.96
1.65
0.55
0.83
0.69
<0.01
56.12
4.02
0.10
<0.01
3.25
0.08
<0.01
0.06
10.44
1.16
1.61
0.87
4.12
<0.01
3.69
4.00
3.84
0.94
78.97
11.63
6.85
1.10
0.44
0.08
<0.01
6.34
9.09
1.93
3.86
3.86
<0.01
4.16
0.46
0.21
0.02
0.50
0.11
<0.01
0.04
3.35
0.91
0.30
9.84
0.47
0.07
0.76
0.88
0.98
0.65
96.74
7.50
4.55
0.83
0.34
0.10
0.02
2.07
3.31
0.83
1.10
1.24
<0.01
2.86
0.92
0.07
0.01
0.80
0.17
<0.01
0.02
2.55
1.25
0.12
9.68
0.11
<0.01
0.67
1.02
0.74
1.20
110.8
24
25
26
27
28
29
30
31
32
<0.01
<0.01
0.03
0.02
<0.01
0.02
<0.01
<0.01
0.03
0.22
0.20
0.22
34.83
0.03
<0.01
0.49
1.42
4.50
0.12
0.69
0.12
0.16
1.02
0.12
<0.01
<0.01
0.18
0.03
<0.01
<0.01
0.87
0.08
11.43
<0.01
0.29
0.17
0.08
<0.01
0.05
0.05
8.62
32.37
0.01
9.33
0.01
<0.01
0.01
0.01
0.01
16.49
24.32
9.71
8.87
12.71
<0.01
7.07
13.15
0.96
<0.01
0.10
4.02
8.29
6.17
5.10
33a/b
34a/b
35a/b
36
0.01
48.32
0.104
1.83
69.36
0.12
46.24
0.09
6.24
<0.01
0.01
<0.01
0.04
0.51
2.06
0.01
0.947
4.64
15.14
566.4
663.3
24.08
261.7
142.0
111.1
23.71
0.79
5.54
37
3.54
46.82
43.74
2.03
0.01
38
12.03
0.38
9.42
0.05
39
0.364
20.85
8.52
0.06
40a
40c
41b
42
3.93
45.36
17.93
6.67
0
1
2
3
4
5
6
7
8
9
0
1
2
3
<0.01
<0.01
<0.01
0.03
0.90
0.13
0.717
0.48
7.59
7.57
a
b
Chemical structures are given in Figure 1. Commercial Pilsner-type
c
0.01
beer. Beers I−IV were produced by the Research Brewery St. Johann
(Germany) and their International Bitterness Units (IBU) adjusted as
follows: beer I (reference) was produced using an Isohop extract
<0.01
0.04
(
30%) added to the beer to adjust an IBU value of 23. Beers II−IV
<0.01
<0.01
<0.01
0.01
were produced using the same wort as for beer I just differing in the
hop products. For beer II (27 IBU) and beer IV (27 IBU), an ε-extract
was added to the wort postboiling and at the beginning of the kettle
boiling, respectively. Beer III (51 IBU) was made by adding spent
hops, obtained by exhaustive carbon dioxide extraction, to the wort
prior to boiling.
0.01
0.01
0.02
[
[
[
C(6″)], 56.4 [C(1‴)], 75.6 [C(3″)], 79.8 [C(2)], 93.8 [C(6)], 105.3
C(10)], 109.9 [C(8)], 115.8 [C(3′, 5′)], 128.4 [C(2′, 6′)], 130.6
C(1′)], 158.2 [C(4′)], 161.2[C(5)], 163.3 [C(9)], 163.2 [C(7)],
corresponding reference compounds. The ion spray voltage was set to
−3
−4500 V, and dwell time for each mass transition was 2.5 × 10 s.
−5
Nitrogen was used as the collision gas (4 × 10 Torr). Using unhopped
beer as a blank matrix, 8-point external matrix calibration curves were
recorded for the individual analytes 1−42 in low (2.5 nmol/L to 200
μmol/L) and high concentrations (0.2 mmol/L to 20 mmol/L) by
means of UPLC-MS/MS. Quantitation of the co- and ad-congeners of
α-acids (1), β-acids (2), and cis-/trans-iso-α-acids (3, 4) was performed
by using the calibration curve of the n-congeners. Calibration functions
were forced through zero, thus leading to correlation coefficients of
>0.99 for all the reference compounds in unhopped beer. Concen-
trations of cis-/trans-p-coumaric acid (33a/b) were corrected for
amounts detected in the nonhopped beer (control). The limit of
detection was determined to be 0.3 nmol/L (S/N ratio: 3) for all
compounds (1−42), and the limit of quantification was accordingly
1
92.4 [C(4)].
Quantitation of Bitter Compounds in Hop Extracts and Beer
by LC-MS/MS. For quantitation of bitter compounds, aliquots of hop
pellets (1.5 g), hop extracts (100 mg), and samples (40 mg) of the
advanced hop extracts XanthoFlav 15, XanthoFlav 75, and polyphenol
extracts, respectively, were exhaustively extracted for 1 h with methanol
(
4 × 25 mL) during ultrasonification. After filtration, the extracts were
diluted (1:100) with methanol and, after membrane filtration (0.45 μm,
Sartorius, Gottingen, Germany), samples (5 μL) were directly injected
̈
into a Dionex UltiMate 3000 series UHPLC system consisting of two
pumps, a degasser, a column compartment, and an autosampler
(
Dionex, Idstein, Germany) connected to an API 4000 Q-TRAP mass
4
6
spectrometer (SCIEX, Darmstadt, Germany) which was equipped with
an electrospray ionization (ESI) source and operated in negative
ionization mode.
estimated to be 0.001 μmol/L (S/N ratio: 10). Data processing and
integration was performed by means of Analyst software version 1.5 (AB
Sciex Instruments, Darmstadt, Germany). As stationary phase, a 150 ×
2.0 mm, 4 μm, Synergi Hydro-RP column (Phenomenex, Aschaffen-
burg, Germany) was used. The mobile phase consisted of acetonitrile
(0.1% formic acid) as solvent A and water (0.1% formic acid) as solvent
B. Using a flow rate of 0.6 mL/min, chromatographic separation was
achieved (Figure 5) by gradient elution increasing solvent A from 10 to
40% within 15 min and further increased to 60% in 13 min, to 70% in 20
min, and finally to 100% within 2 min. It was maintained at 100% for 1
min, following by readjustment to 10% within 5 min and re-equilibrated
for 4 min prior to the next injection.
For the analysis of beer samples, the beverage was degassed by
̈
ultrasonification and membrane filtered (0.45 μm, Sartorius, Gottingen,
Germany), and aliquots (5 μL) were directly injected into the SCIEX
LC-MS system. The temperature of the autosampler was set to 5 °C and
of the column compartment to 40 °C. The declustering potential (DP),
the cell exit potential (CXP), and the collision energy (CE) of the mass
spectrometer were optimized for each substance to induce fragmenta-
−
tion of the pseudo molecular ion [M − H] to the corresponding target
product ions after collision-induced dissociation. Quantitative analysis
was performed by means of multiple reaction monitoring (MRM) mode
using optimized fragmentation parameters and retention times of the
Sensory Analysis. General Conditions and Panel Training.
Twelve subjects (7 women and 5 men, ages 23−29 years), who gave
I
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
consent to participate in the sensory tests of the present investigation
and had no history of known taste disorders, participated for at least two
years in weekly training sessions and were recruited from the Chair of
Food Chemistry and Molecular Sensory Science (Freising, Germany).
To familiarize the subjects with the bitter taste of hop-derived
compounds and to get them trained in recognizing bitterness by using
Liquid Chromatography/Mass Spectrometry (LC-MS/MS). For
compound identification, mass and product ion spectra were acquired
on an API 4000 Q-Trap triple quadrupole/linear ion trap mass
spectrometer (SCIEX, Darmstadt, Germany). Isolated fractions were
dissolved in methanol/water (50/50, v/v) and directly introduced into
the mass spectrometer by flow infusion using a syringe pump. Data were
acquired in full-scan mode with negative electrospray ionization (−4500
V). Both quadrupoles operated at unit mass resolution. Nitrogen served
as curtain gas (30 psi) and as turbo gas (450 °C). Fragmentation of the
33
the sip-and-spit method, xanthohumol (5) and trans-isocohumulone
4a) were used as references. The sensory experiments were performed
(
at 20−22 °C. To prevent cross-modal interactions with olfactory or
visual inputs, the panelists used nose clips and yellow light or whenever
necessary red light, respectively, was used.
−
pseudo molecular ions [M − H] into specific product ions was induced
−5
by collision with nitrogen (4.5 × 10 Torr). Data acquisition and
instrumental control were performed with Analyst 1.5 software
Pretreatment of Fractions and Pure Compounds. Prior to sensory
analysis, the fractions or isolated compounds were suspended in water,
and, after removal of the volatiles under high vacuum (<5 hPa), were
(
SCIEX).
UPLC/Time-of-Flight Mass Spectrometry (UPLC/TOF-MS).
1
Mass spectra of the compounds were measured on a Synapt G2
HDMS mass spectrometer (Waters, Manchester, U.K.) coupled to an
Acquity UPLC core system consisting of a binary solvent manager,
sample manager, and column oven. The compounds were dissolved in
acetonitrile (1 mL), and aliquots (1−5 μL) were injected into the
UPLC-TOF-MS system equipped with a 2 × 150 mm, 1.7 μm, BEH C18
column (Waters). Operated with a flow rate of 0.3 mL/min at 40 °C, the
following gradient was used for chromatography: starting with a mixture
freeze-dried twice. H NMR spectroscopy, GC/MS, and high-
performance ion chromatographic analysis of an aliquot revealed that
fractions treated by that procedure are essentially free of solvents and
buffer compounds used. Formic acid, which is GRAS listed as a flavoring
agent for food and feed applications, was used to adjust the pH value of
solutions to be sensorially evaluated, because trace amounts of this acid
do not influence the sensory profile of the test solution.
Sensory Analysis of Isolated Hop Resins. The individual resin
fractions isolated from hops as described above were evaluated in their
bitter intensity. To achieve this, each resin sample was taken up in 5%
aqueous ethanol adjusted to pH 4.4 with trace amounts of 10% aqueous
formic acid. The bitter intensity of these solutions was judged on a scale
from 0 (not detectable) to 5 (strongly detectable).
(
5/95, v/v) of acetonitrile and aqueous formic acid (0.1%, pH 2.5), the
acetonitrile content was increased to 100% within 5 min and held at
00% for 1 min. Measurements were performed using negative ESI and
1
the resolution modus consisting of the following parameters: capillary
voltage −3.0 kV, sampling cone 30, extraction cone 4.0, source
temperature 150 °C, desolvation temperature 450 °C, cone gas 30 L/h,
and desolvation gas 850 L/h. The instrument was calibrated over an m/z
range of 100−1200 using a solution of sodium formate (0.5 mM) in a 2-
propanol/water mixture (9/1, v/v). All data were lock mass corrected
3
3−35
Taste Dilution Analysis (TDA)..
An aliquot of each fraction
isolated from hops was freeze-dried twice and dissolved in water
containing 5% ethanol (pH 4.4) in “natural” concentration ratios
matching the amounts present in 2.0 g of hops hard resin. After each
fraction had been sequentially diluted 1 + 1 with water (pH 4.4), the
serial dilutions of each of these fractions were presented to the sensory
panel in order of ascending concentration. Using a duo test, panelists
were asked to determine the dilution step at which a difference between
sample and blank could be detected. This so-called taste dilution (TD)
−
using leucine enkephalin as the reference (m/z 554.2615, [M − H] ).
Data processing was performed by using MassLynx 4.1 (Waters) and the
tool “elemental composition”.
1
Nuclear Magnetic Resonance Spectroscopy (NMR). 1D/2D- H
13
and C NMR spectra were acquired on a 400 MHz DRX and a 500
MHz Avance III spectrometer (Bruker, Rheinstetten, Germany),
33
factor determined by the 12 sensory assessors was averaged and did
not differ by more than plus/minus one dilution step between the
trained individuals.
respectively. CD OD, D O, or DMSO-d was used as solvent, and
3
2
6
chemical shifts are reported in parts per million relative to the solvent
signal. For signal assignment, 2D-NMR experiments (COSY, HMQC,
HMBC) were carried out using the pulse sequences taken from the
Bruker software library. Data processing was performed by using XWin-
NMR software (version 3.5; Bruker, Rheinstetten, Germany) as well as
MestReNova 5.1.0-2940 (Mestrelab Research S.L., Santiago de
Compostela, Spain).
Bitter Recognition Threshold Concentrations. Bitter taste threshold
concentrations of the purified compounds were determined by 12
panelists using a duo test. The geometric mean of the last and the second
to last concentrations was calculated and taken as the individual
recognition threshold. The threshold value of the sensory group was
approximated by averaging the threshold values of the individuals in
three independent sessions and did not differ by more than plus/minus
one dilution step. Based on the three independent repetitions, the
confidence interval (α = 0.05) was calculated for each compound.
Taste Re-Engineering Experiments. To investigate the impact of the
hop constituents on the bitter perception of beer, four taste
recombinants were produced by spiking unhopped “zero” beer with
bitter taste molecules in concentrations as determined in a standard
Pilsner-type beer (Table 3) by means of LC-MS/MS. Recombinant A:
RESULTS AND DISCUSSION
■
As the use of the hop’s hard resin fraction for beer manufacturing
was reported to reveal beers with an intense and pleasant bitter
26
profile, the sensory impact of the hard resin fractions should be
evaluated. To achieve this, the hard resin fraction was isolated
30
from hops, chlorophylls and hop waxes were separated by
means of solid phase extraction, and the resin was separated to
give the water-soluble δ-resin and the nonsoluble ε-resin fraction
“
zero” beer was spiked with iso-α-acids (3a−c, 4a−c). Recombinant B:
recombinant A was spiked with xanthohumol (5) and isoxanthohumol
6). Recombinant C: recombinant A was spiked with prenylated
31,32
(
by extraction with water at 40 °C.
Aliquots of the δ-resin, ε-
flavonoids 5−32. Recombinant D: recombinant A was spiked with
compounds 5−42. Using a linear 5-point scale, the bitterness of these
recombinants A−D was then compared with that of a standard Pilsner-
type beer (reference).
resin, hard resin, soft resin, and total hop resin, respectively, were
solubilized in 5% ethanolic solution in their “natural”
concentration ratio (0.2/0.8/1/9/10; w/w), and the pH was
adjusted to 4.4 to match the pH value of a standard beer. The
trained sensory panelists, who were asked to rate the bitter
intensity on a linear scale from 0 (not detectable) to 5.0 (very
intense), judged the bitterness of the soft resin solution with the
highest score of 4.25 (±0.25), followed by the ε-resin and the
hard resin with 0.51 (±0.08) and 0.5 (±0.25), whereas bitterness
was almost not detectable (0.08 ± 0.04) in the solution
containing the δ-resin (data not shown). These data confirm the
previously reported major bitter contribution of the soft resin
Sensomics Heatmapping. Data analysis was performed within the
4
7
programming and visualization environment R (version 2.13.2). The
sensomics heatmap was calculated using the heatmap.2 function of R
based on the concentration data after scaling the sum of all
sensometabolites to an equal value for each beer sample. The
dendrograms were constructed by means of an agglomerative linkage
algorithm specifying the distance between two clusters as the increase in
the error sum of squares after fusing two clusters into a single cluster and
48
seeking for a minimum distance at each clustering step.
J
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
components and indicated the ε-resin to account for the
bitterness of the hop’s hard resin fraction. The ε-resin was,
therefore, further separated to locate and identify the key bitter
molecules in that hydrophobic fraction.
xanthohumol H (22), thus suggesting a similar chemical
structure. However, an additional proton signal for H−C(6″)
resonating at 3.28 ppm with an integral of three protons and an
13
additional C resonance signal for C(6″) at 50.6 ppm indicated
compound 29 as the 3″-methoxy adduct of xanthohumol H (22).
This was confirmed by the heteronuclear coupling observed
between carbon atom C(3″) and the protons of the methyl group
H−C(6″) in the HMBC spectrum (Figure 4). To confirm the
natural occurrence of 29 and to exclude any artifact formation
during workup, hop was extracted with ethanol instead of
methanol. LC-MS/MS analysis of this ethanolic extract
unequivocally identified xanthohumol M (29) in hops.
Identification of Key Bitter Compounds in the ε-Resin.
Fractionation of the ε-resin by means of MPLC on RP18 material
revealed 11 fractions (e1−e11), which were dissolved in 5%
ethanolic solution in their natural concentration ratio and, then,
evaluated in their bitter impact by means of a taste dilution
3
3−35
analysis (TDA).
Fractions e5, e6, and e8−e11 were judged
with the highest TD factors of 64 and 256, respectively, followed
by fractions e2−e4 judged with TD factors between 6 and 32
1
13
(Figure 2). Aimed at locating the key contributors to the bitter
Compound 18 showed similar H and C NMR shifts for the
naringenin moiety and the cyclic prenylic side chain as found for
1″,2″-dihydroxanthohumol K (14) and 1″,2″-dihydroxanthohu-
taste of fractions e2−e11, each MPLC fraction was separated by
means of preparative RP-HPLC and the subfractions were again
analyzed by means of TDA to locate most intense bitter tastants
1
mol C (12), respectively. But in comparison to 12 and 14, any H
(
Figure 3).
NMR signal for H−C(6) was lacking, whereas additional signals
were observed for a second prenyl side chain as known for 5′-
prenylxanthohumol (17). The HMBC spectrum indicated a
heteronuclear coupling between carbon atom C(6) and the
protons at C(1⁗) and confirmed the attachment of a prenyl side
chain to carbon C(6) (Figure 4). The hypsochromic shift of the
UV maximum of xanthohumol (5) and 1″,2″-dihydroxanthohu-
mol C (12), respectively, from ∼370 to 326 nm was well in line
with the intramolecular cyclization of the prenyl side chain
attached to carbon C(8) with the hydroxyl group bound to
Comparison of spectroscopic data (UV/vis, LC-MS/MS,
1
13
H/ C NMR) of xanthohumol (5) in subfractions e8-7, e9-6,
and e10-2, isoxanthohumol (6) in subfractions e4-2, e5-3, e6-5,
and e9-2, desmethylxanthohumol (7) in subfraction e7-12, 8-
prenyl- (8) and 6-prenylnaringenin (9) in subfraction e8-5, 5′-
prenylxanthohumol (17) in subfractions e10-10 and e11-8,
xanthohumol D (19) in e7-9, xanthohumol B (20) in e7-7, and
xanthohumol C (21) in e10-5 and e11-5 with the literature
2
4,25
data
confirmed the identity of these compounds in the ε-
50
resin (Table 1). In addition, 1″,2″-dihydroxanthohumol C (12)
was identified in subfractions e10-4 and e11-4, 1″,2″-
dihydroxanthohumol K (14) in e4-2, e5-3, and e7-3,
xanthohumol H (22) in e6-7 and e7-4, xanthohumol I (26) in
e3-4, and 2″-hydroxyxanthohumol M (25) in e7-8 (Table 1), the
spectroscopic data of which confirmed previously published
carbon C(9). In consequence, the structure of this previously
unknown compound was assigned as 1″,2″-dihydroxanthohumol
F, 18 (Figure 4).
The spectroscopic data of compound 10, showing major
differences compared to those reported for other hop
constituents in the past, indicated the presence of two prenyl
side chains, a p-coumaroyl moiety, and a methoxy group in the
HSQC spectrum. The HMBC spectrum revealed both prenyl
side chains to be attached to carbon C(8) and to show
connectivity to two keto functions, thus indicating a cyclo-
hexenedione structure (Figure 4). In contrast to the keto
function at C(4) as found in xanthohumol derivatives, carbon
2
3,36−38,45
data.
Another polar bitter compound in fraction e2 was
43
identified as co-multifidol glucopyranoside 40a in subfraction
e2-7 (Table 1). Some trace amounts of α-acids (1a/b/c) were
identified in e10-8/9 and e11-8/9, and β-acids (2a/b/c) in e11-7,
e11-11, and e11-13, thus indicating some residual amounts of
soft resin components in the hard resin fraction. Next to these
bitter compounds, some astringent molecules were identified as
cis/trans-p-coumaric acid (33a/b) in subfraction e2-4,
quercetin (36) in e3-5, and quercetin-3-O-β-D-glucopyrano-
side (37) in e2-5 (Table 1). The bitter molecules ad-multifidol
glucopyranoside (40c) and n-multifidol-di-C-glucopyranoside
13
C(4) showed a C chemical shift as expected for a hydroxylated
carbon. However, the proton signal of H−C(2) showed an
extraordinarily strong low-field shift as found for xanthohumol
4
9
4
0
4
1
51
due to the peri-effect caused by the keto function at C(9).
43
Taking all spectroscopic data into consideration, the target bitter
compound was identified as the previously not reported 4′-
hydroxytunicatachalcone, 10 (Figure 4), which is the 4′-hydroxy
4
4
(40b) as well as the astringent molecules trans-N-feruloyltyr-
23,36,37 41
amine (42),
kaempferol-3-O-β-D-glucopyranoside (38),
42
52
and kaempferol-3-O-β-D-(6″-malonyl)glucopyranoside (39)
derivative of tunicatachalcone identified earlier.
1
13
were isolated from a hop polyphenol extract and were also
identified as hard resin constituents (Table 2). For unequivocal
structure elucidation by means of NMR spectroscopy, glycosides
The H and C NMR spectrum of compound 11 showed very
similar chemical shifts as found for 5′-prenylxanthohumol (17)
but additional resonance signals indicating the presence of
another prenylic side chain and a keto function. Furthermore, an
HSQC experiment revealed compound 11 to be a flavonoid
showing the characteristic proton resonances at 2.96 and 3.16
ppm for the diastereotopic protons H−C(3). The HMBC
spectrum showed a heteronuclear coupling between carbon C(9)
and one prenylic side chain, and between carbonyl carbon C(7),
resonating at 198.6 ppm, to three prenylic side chains (Figure 4).
Taking all spectroscopic data into consideration, the structure of
compound 11 was determined to be 5-hydroxy-2-(4-hydrox-
yphenyl)-6,6,8-triprenyl-6H-chroman-4,7-dione, coined isoxan-
tholupon, 11 (Figure 4), which to the best of our knowledge has
not previously been reported.
3
7−41 were isolated in larger quantities from the hop tannin
extract.
In contrast to the above-mentioned taste molecules,
compounds 10, 11, 18, 23, 27−29, 31, and 32 isolated from
the ε-resin have not yet been identified as natural phytochemicals
in hops and beer, respectively. Although not yet reported as hop
components, xanthohumol O (27), xanthohumol L (28), and
isoxanthohumol H (23) were found earlier as metabolites in
feces of xanthohumol-fed rats. The chemical structures of the
previously not reported hop bitter constituents 10, 11, 18, 29, 31,
and 32 were determined by means of LC-MS and 1D/2D-NMR
experiments.
39
NMR analysis of compound 29, isolated from subfractions e8-
and e9-4, showed similar H and C shifts as found for
Compared to xanthohumol (5), compound 31 was missing the
typical H shifts of the p-coumaric acid moiety. Furthermore, the
1
13
1
3
K
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Figure 6. RP-HPLC-MS/MS mass chromatogram of taste compounds in hop extract. Numbering of compounds refers to chemical structures given in
Figure 1.
spectra also showed that a substitution had occurred at carbon
atom C(6) due to the lacking proton signal at this carbon atom.
But interestingly, compound 31 exhibited typical H shifts of an
additional singlet at 5.4 ppm and lacked the typical aromatic
proton shifts of the p-coumaric acid moiety. Heteronuclear
couplings in the HMBC spectrum indicated that a prenylic side
chain is attached at carbon C(4) and that a methoxy group is
connected to carbon C(1). Taking all spectroscopic data into
account, the target compound was identified as the previously
not reported 1-methoxy-4-prenylphloroglucinol, 32 (Figure 4).
It is proposed in the literature that xanthohumol is biosynthe-
sized from coumaroyl-CoA and three malonyl-CoA to form
chalconaringenine, followed by a prenylation reaction to form
first desmethylxanthohumol (7) and, thereafter, xanthohumol
(5) by additional methylation. Interestingly, the occurrence of
compound 32 in hops provides evidence that the prenylation and
methoxylation of the phloroglucin moiety during the biosyn-
thesis of 5 can occur prior to its reaction with p-coumaroyl-CoA.
Synthetic Preparation of Xanthohumol Derivatives
12−16, 22−24, 29, and 30. In order to screen the hop
fractions for additional xanthohumol derivatives, xanthohumol P
1
intact prenylic side chain as well as signals of two chemically
1
identical methyl groups which showed no J-coupling with any
other proton. A long-range coupling of carbon C(9) and the
methyl protons H−C (1′, 2′) in the HMBC spectrum revealed
the structure of compound 31 to be 2,2-dimethyl-5,6,7-
trihydroxy-8-(3-methylbut-2-enyl)chroman-4-one, which is
coined 2″,3″-dehydrocyclohumulohydrochinone, 31 (Figure
53
4), due to its structural similarity with humulohydrochinon.
54
As LC-MS-MS analysis undoubtedly confirmed the presence of
this compound in hops and excluded its formation as workup
artifact (data not shown), compound 31 is considered a
previously not reported hop constituent.
1
The H NMR data of compound 32, showing a molecular
composition of C H O (TOF-MS), revealed large similarities
12
16
3
with those recorded for xanthohumol (5), but showed an
L
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
Figure 7. Sensomics heatmapping of concentrations of taste compounds in hop products/extracts taken from Table 3. The dendrogram is based on an
48
agglomerative linkage algorithm. Numbering of compounds refers to structures given in Figure 1
(
15) was synthesized by heating an ethanolic solution of
prenylic side chain, while an intramolecular cyclization (12, 14,
20, 28) seemed not to influence the bitter threshold levels. More
lipophilic molecules like compound 12 that elute later on RP-18
material seem to have a lower human threshold concentration
than structurally related earlier eluting substances like compound
14. Interestingly, the polar O-β-D-glycosides of the multifidol
congeners 40a and 40c feature surprisingly low human threshold
concentrations of 5 and 10 μmol/L, respectively (Table 1). As
the α/β-acids, iso-α-acids, xanthohumol, and isoxanthohumol
activate three bitter taste receptors, namely, hTAS2R1,
xanthohumol (5) for 4 h at 80 °C in the presence of the acidic
cation exchanger Amberlyst 15, while isoxanthohumol P (16)
and isoxanthohumol M (30) were obtained after alkaline
isomerization of xanthohumol P (15) and xanthohumol M
(29), respectively. Reaction of xanthohumol (5) with hydro-
bromic acid in methanol/acetic acid revealed, besides com-
pounds 6, 15, 16, 29, and 30, additional major transformation
products, namely, 12, 14, and 22−24 (Figure 5). These
compounds were identified as 1″,2″-dihydroxanthohumol C
21
(12) and K (14), and xanthohumol H (22) by LC-MS and 1D/
hTAS2R14, and hTAS2R40 receptor, these data suggest the
involvement of another bitter taste receptor responding to the
bitter β-glycosides 40a and 40c. One candidate receptor would
be the hTAS2R16, which has been reported to be activated by β-
2
D-NMR experiments as well as by cochromatography with the
corresponding reference substance (Figure 4). Isoxanthohumol
H (23) and 1″,2″-dihydroisoxanthohumol C (13) were isolated
and identified by comparing the LC-MS and NMR data to those
55
glycosides like salicin.
3
9,45
reported in literature.
Reaction product 24 exhibited similar
Quantitation and Sensomics Heatmapping of Hop-
Derived Bitter Compounds in Hop Products and Beer. To
gain insight into the occurrence and concentrations of the
individual hop-derived bitter compounds in beer, compounds 1−
42 (Figure 1) were quantitated in hop extracts as well as a
Pilsner-type beer by means of HPLC-MS/MS in a single
chromatographic run. Operating the mass spectrometer in the
multiple reaction monitoring (MRM) mode, a highly selective
and sensitive quantitation of the analytes could be accomplished
(Figure 6).
As expected, xanthohumol (5) was the most predominant
compound (>30 mmol/100 g) found in the hard resin (Table 2).
With the exception of the very lipophilic isoxantholupon (11)
and cis-/trans-p-coumaric acid ethyl ester (35a/b), all other
compounds identified above were found in the hard resin as well.
1-Methoxy-4-prenyl-phloroglucinol (32) and 2″,3″-dehydrocy-
clohumulohydrochinon (31) were only detectable in the hard
resin isolated from the bitter hop variety Taurus (Table 2). As
expected, trace amounts of most of the prenylated flavonoid were
found in the soft resins too. Only the nonpolar 4′-
hydroxytunicatachalcone (10) and 5′-prenylxanthohumol (17),
respectively, were observed in considerable amounts of 1193 and
1320 μmol/100 g in the soft resin of the bitter hop variety
Taurus. Analysis of the commercially available pure resin extracts
contained high amounts of >700 μmol/100 g of xanthohumol
(5), whereas all other compounds (6−42) were present in
comparatively low amounts (Table 2).
1
13
H and C shifts as found for the naringenin substructure in
xanthohumol (5) but with a different proton signal pattern
recorded for the prenylic side chain. Instead of two nonaromatic
signals integrating for three protons, only one signal at 1.79 ppm
with an integral of three protons was observed. In addition,
resonance signals integrating for two protons were observed at
2
.18 and 4.66 ppm, respectively, suggesting that the double bond
in 24 is located between C(3″) and C(4″). The heteronuclear
coupling between C(3″) and H−C(4″) observed in the HMBC
spectrum (Figure 4) confirmed the structure of 24 as
xanthohumol N (Figure 1), which to the best of our knowledge
has not yet been reported earlier.
Sensory Evaluation of Purified Hop Components. Prior
to sensory analysis, the purity of each compound was confirmed
by HPLC-MS as well as H NMR spectroscopy to be more than
9
bitter sensation induced by the hop derivatives identified,
aqueous solutions of target compounds were evaluated in serial
dilutions by means of a duo test (Table 1). The bitter threshold
concentrations determined for the α-acids (1), β-acids (2),
xanthohumol (5), and isoxanthohumol (6) by means of the duo
test were comparable to those recorded recently by means of a
1
8%. To determine the human threshold concentrations for the
1
5−17
triangular test,
setup.
thus confirming the suitability of the duo test
Except for the p-coumaric acid esters (34a/b, 35a/b), the
human bitter threshold concentrations were relatively low and
ranged from 5 to 44 μmol/L (Table 1). Comparably high
threshold concentrations (>20 μmol/L) were found for
compounds 15, 16, 19, 22, 23, 26, and 30 exhibiting a modified
Quantitative analysis of the XanthoFlav 75% extract revealed,
next to xanthohumol (5), high concentrations of isoxanthohu-
mol (6, 9754 μmol/100 g), desmethylxanthohumol (7, 13873
M
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
μmol/100 g), xanthohumol L (28, 11656 μmol/100 g), and
kaempferol-3-O-β-D-glycopyranoside (38, 45105 μmol/100 g).
With the exception of xanthohumol (1), the XanthoFlav 15%
extract did not differ much in its composition from the
xanthohumol enriched extract XanthoFlav 75%. The only
significant difference was the presence of much higher amounts
are distinctive for their content of specific hard resin derived
compounds and are a useful resource to adjust levels of desired
compounds in final hop-containing products like beer.
The quantitative analysis of a traditional Pilsner beer (Table 3)
confirmed that the iso-α-acids (3a/b/c, 4a/b/c) are the
12−14
quantitatively predominant bitter tasting compounds
with
56
(
8142 vs 3.87 μmol/100 g) of isoxantholupon (11) in the extract
concentrations between 5.0 and 24.5 μmol/L. Due to the
isomerization reaction during wort boiling, significant amounts
(5.5 μmol/L) of isoxanthohumol (6) are present in the beer
sample. Next to the iso-α-acids (3a/b/c, 4a/b/c) and
isoxanthohumol (6), the glycosides (37−41b) are the third
largest group with concentrations between 0.9 and 4.0 μmol/L.
All other xanthohumol and isoxanthohumol derivates (5, 8−30)
are present in concentrations below 0.6 μmol/L. With the
exception of the iso-α-acids (3a/b/c, 4a/b/c), none of the other
compounds were detected in beer above their taste threshold
concentration (Table 1). In order to investigate whether the
subthreshold prenylated flavonoids and glycosides contribute
additively to the beer’s bitterness, taste re-engineering experi-
ments were performed in the following.
27
XanthoFlav 15%. The ε-extract contained quite the same
compounds as the XanthoFlav extracts, however, xanthohumol
(5) and isoxanthohumol (6) were present in comparably low
amounts of 8601 and 7671 μmol/100 g, respectively, and the
polar compounds 10, 34−39, 41b, and 42 were not detectable at
all (Table 2).
Commercially available tannin extracts revealed very high
amounts of the flavonol glycosides 37−39 (813.2−14541 μmol/
100 g) and the intensely bitter tasting multifidol glycosides 40a,
4
0b, and 41b (428.7−4802 μmol/100 g). Interestingly, the
extract prepared from an aroma hop variety has higher amounts
of the flavononol glycosides 37−39 (18.60 vs 11.36 mmol/100
g), whereas the extract made from a bitter hop variety showed
higher levels of the multifidol glycosides 40a, 40b, and 41b (2.38
vs 7.08 mmol/100 g). Both extracts contained only minor
amounts (<105 μmol/100 g) of xanthohumol (5) and its
derivatives.
Taste Re-Engineering Experiments. Samples of unhop-
ped “zero” beer were spiked with natural concentrations (Table
3) of the individual purified iso-α-acids 3 and 4 to give
recombinant A, the iso-α-acids (3a−c, 4a−c), xanthohumol (5),
and isoxanthohumol (6) to give recombinant B, the iso-α-acids
(3a−c, 4a−c) and the prenylated flavonoids 5−32 to give
recombinant C, and the iso-α-acids (3a−c, 4a−c) plus all the
identified compounds 5−42 to give recombinant D. Using a
linear 5-point scale, the bitterness of these recombinants A−D
was then compared with that of a standard Pilsner-type beer as
reference (Figure 8). Without any further addition, the bitterness
To examine the multivariate distances between the individual
sensometabolites throughout the set of different hop products,
the concentrations determined for hard resin-derived com-
pounds (5−42) were scaled, followed by a hierarchical cluster
analysis to arrange the taste compounds into six clusters labeled
1−6 as visualized in a sensomics heatmap (Figure 7). Cluster 1
consisted of the glycosides 37−41b present in high concen-
trations in the tannin extracts and in hop pellets, thus
demonstrating the quantitative dominance of glycosides over
the xanthohumol derivatives. Cluster 2 comprised xanthohumol
(
(
5), desmethylxanthohumol (7), its isomerization products 8-
8) and 6-prenylnaringenin (9), xanthohumol I (26), cis-/trans-
p-coumaric acid ethyl ester (35a/b), and quercetin (36) being
the characteristic attribute of the XanthoFlav extract 75%. In
comparison, the large cluster 3 consisted of isoxanthohumol (6),
isoxantholupon (11), isoxanthohumol P (16), and 1-methoxy-4-
prenylphloroglucinol (32) separating in subcluster 3a and 1″,2″-
dihydroisoxanthohumol C (13), xanthohumol P (15), xantho-
humol D (19), xanthohumol B (20), xanthohumol N (24),
xanthohumol O (27), isoxanthohumol M (30), and N-trans-
feruloyltyramine (42) in subcluster 3b. Whereas higher
concentrations of compounds in subcluster 3a seem to be
characteristic for the ε-extract and the XanthoFlav 75% extract,
the extract XanthoFlav 15% and the hard resins showed high
concentrations of compounds of subcluster 3b. Interestingly, the
hard resin fraction prepared from the hop varieties Taurus and
Perle showed significant differences in clusters 4 and 5.
Compounds 4′-hydroxytunicatachalcone (10), 1″,2″-dihydrox-
anthohumol C (14), xanthohumol H (22), isoxanthohumol H
Figure 8. Bitter taste intensity of (R) standard Pilsner-type (reference),
(0) unhopped “zero” beer and “zero” beer spiked with (A) iso-α-acids
(
3a/b/c, 4a/b/c), (B) iso-α-acids (3a/b/c, 4a/b/c), xanthohumol (5),
and isoxanthohumol (6), (C) iso-α-acids (3a/b/c, 4a/b/c) and
compounds 5−35 and 42, and (D) iso-α-acids (3a/b/c, 4a/b/c) and
all hard resin-derived compounds 5−42 (Table 1). Confidence interval
(
α = 5%).
of the “zero” beer was rated with a low score of 1.03 ± 0.33
compared to the reference Pilsner-type beer (3.02 ± 0.29), thus
demonstrating that a majority of the beer’s bitterness is due to
hop-derived components while a marginal portion of remaining
bitterness comes from other sources such as, e.g., yeast and malt.
Recombinant A containing the iso-α-acids showed an increased
intensity of 2.20 ± 0.25 for a scratchy bitterness which was
perceived with an instantaneous onset but a short offset. Addition
of xanthohumol (5) and isoxanthohumol (6) to recombinant A
induced a marginally decreased bitter intensity of recombinant B
(1.93 ± 0.40) which was reported by the panelists to exhibit a
more long lasting and less sharp bitter profile when compared to
(23), and xanthohumol M (29) are grouped in cluster 4 and are
representative for the Perle hard resin, whereas 1″,2″-
dihydroxanthohumol K (14), 5′-prenylxanthohumol (17),
xanthohumol C (21), 2″-hydroxyxanthohumol M (25), 2″,3″-
dehydrocyclohumulohydrochinon (31), and cis-/trans-p-couma-
ric acid methyl ester (34a/b), classified in cluster 5, seem to be
typical for the Taurus hard resin. As expected, the α-acids (1a/b/
c) and β-acids (2a/b/c) are characteristic for the total resin
extract and the soft resin and are therefore grouping together in
cluster 6. These data clearly show that the various hop products
N
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
recombinant A. Recombinant C containing iso-α-acids plus
compounds 5−32 showed a typical beer-like, well-rounded
bitterness with a medicinal-herbal-like off-taste, and the bitter
intensity increased and was determined to be 2.64 ± 0.25.
Sensory evaluation of the total taste recombinant D revealed a
bitterness of 2.95 ± 0.26, which did not differ significantly from
the bitter intensity of the reference beer (3.02 ± 0.29) and was
described as beer-like, long lasting, and well-rounded without any
specific off-taste. Taking all these sensory data into account, iso-
α-acids (3a−c, 4a−c) and the identified compounds 5−42 were
demonstrated for the first time to be truly sufficient for
constructing the authentic bitter percept of beer. The knowledge
on the structure and concentration of these bitter compounds
paves the way for the targeted manufacturing of tailored taste
profiles of beer and hop-containing foods and beverages.
ingenin (9), 8-prenylnaringenin (8), isoxanthohumol H (23),
and isoxanthohumol M (30). Interestingly, high concentrations
of isoxanthohumol P (16) and cis-/trans-p-coumaric ethyl ester
(35a/b) were found in beer IV, which might be explained by
postfermentation reactions with ethanol produced by yeast.
The concentrations of most compounds (5−35) found in beer
III, which was produced with spent hops only, did not
significantly differ from those found in beer IV, thus confirming
that all target compounds are already present in the spent
material and are extracted and/or isomerized during kettle
boiling. Interestingly, trans-N-feruloyltyramine (42), quercetin
(36), and the bitter glycosides 37−41b were also found in beer
III in up to 10-fold higher concentrations than in beer IV. With
the exception of the multifidol glycosides 40a and 40c, these very
polar compounds were not present in the ε-extract and are
extracted during wort boiling from spent hops. As beer III and
beer IV showed quite similar concentrations of prenylated
flavonoids (Table 3), but do differ drastically in the content of
glycosides 37−41, it is assumed that the overall harsh and
unpleasant bitter taste sensation perceived for beer IV is due to
the rather high levels of these glycosides.
Sensory and Quantitative Analysis of Hop-Derived
Taste Compounds in Beer Manufactured with ε-Extract
and Spent Hops, Respectively. It has been recently reported
that the use of hop’s hard resin instead of the soft resin fraction
for beer production revealed a beer exhibiting an intense and
26
pleasant bitter character. As the above-mentioned ε-extract was
found to be enriched in bitter tasting prenylated flavonoids, it was
interesting to quantitate the bitter molecules and to sensorially
evaluate the bitter taste of a beer manufactured by using this ε-
extract as the only hop source. To achieve this, four different
beers (beers I−IV) were produced and their International
Bitterness Units (IBU) were determined by a trained sensory
With the exception of isoxanthohumol (6) and 8-prenylnar-
ingenin (8), the concentrations of compounds 5−42 were much
higher in beer IV than in the standard Pilsner-type beer, e.g., beer
IV contained 68, 240, and 5500 times higher levels of
xanthohumol, isoxanthohumol P, and isoxanthohumol H when
compared to the standard Pilsner beer (Table 3). Since
prenylated flavonoids have been shown to possess health
5
7
panel. Beer I was produced as a bitter reference using an Isohop
extract (30%) added to the final, fermented beer to adjust an IBU
value of 23. Beers II−IV were produced using the same wort as
for beer 1 just differing hop products. For manufacturing of beer
II (27 IBU) and beer IV (27 IBU), the ε-extract was added to the
wort postboiling and at the beginning of the kettle boiling,
respectively. Beer III (51 IBU) was made by adding spent hops,
which were obtained by exhaustive carbon dioxide extraction, to
the wort prior to boiling. Descriptive sensory analysis revealed
beer I (23 IBU) and beer IV (27 IBU) as well-rounded with a
pleasant and typical Pilsner-type bitter profile. In comparison,
beer III produced from spent hops (51 IBU) was described by
the trained panelists as strongly bitter with a harsh bitter and
irritating character perceived at the back of the tongue, while beer
II (27 IBU) was reported to exhibit a medicinal or herbal bitter
taste profile.
2
3,45,59,60
beneficial effects in vitro
but a 60-fold enrichment of
xanthumol derivatives is necessary to benefit from the proposed
61
activities, these results impressively demonstrate the possibility
to produce beverages strongly enriched with prenylated hop
flavonoids.
ASSOCIATED CONTENT
Supporting Information
■
*
S
Chromatographic parameters for HPLC separation of MPLC
subfractions. Spectroscopic data of all hard resin derived
compounds. Comprehensive spectroscopic data of isolated hop
compounds. Table S1: Optimized mass spectrometric parame-
ters for the quantitative analysis of taste-active hop-derived
compounds by means of LC-MS/MS using negative electrospray
−
ionization (ESI ). This material is available free of charge via the
To gain some more insight into the concentrations of the
individual hop-derived bitter compounds in beer, compounds 1−
Internet at http://pubs.acs.org.
4
2 were quantitated in beers I−IV and a standard Pilsner-type
AUTHOR INFORMATION
Corresponding Author
Phone: +49-8161/71-2902. Fax: +49-8161/71-2949. E-mail:
■
*
beer by means of HPLC-MS/MS in a single chromatographic
run (Table 3). Analysis of beer I revealed about 60 μmol/L iso-α-
acids (3, 4) as well as small amounts of 4′-hydroxytunicatachal-
thomas.hofmann@tum.de.
58
cone (10) and isoxantholupon (11). Van Opstaele already
described that iso-α-acids extracts also contain some other hop-
derived compounds which are accordingly introduced into a beer
spiked with an iso-α-acids extract. All other compounds of the
hard resin (5−9, 12−42) were present just in trace amounts or
were not detectable. In comparison, beer II contained significant
amounts of the nonisomerized xanthohumol derivatives 5, 12,
Funding
We are grateful to the Barth-Haas-Group GmbH & Co. KG,
Nuremberg, Germany, for their generous financial support.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
We are grateful to Christina Scho
1
4, 15, 17, 19−22, 24, and 28, besides α-acids (1) and β-acids
■
(2), being well in line with manufacturing the beer by using a
̈
nberger (Barth-Haas-Group)
xanthohumol derivative enriched extract after wort boiling.
Adding the ε-extract at the beginning of the wort boiling
procedure (beer IV) resulted in rather low amounts of
xanthohumols but significantly increased amounts of isoxantho-
humol derivatives, namely, isoxanthohumol (6), 6-prenylnar-
and Martin Biendl (Hopsteiner; Simon H. Steiner GmbH,
Mainburg, Germany) for providing hop samples, to Andreas
Gahr for providing the beer samples produced at the Research
Brewery St. Johann (Train St. Johann, Germany), and to Cynthia
Almaguer and Martina Gastl (Chair of Brewing and Beverage
O
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Technology, Technische Universitat Munchen, Germany) for
providing unhopped “zero” beer. The authors wish to thank
SCIEX Darmstadt, Germany, for providing technical and
application support.
Article
̈
̈
(20) Heller, W. and Forkmann, G. Biosynthesis of flavonoids. In The
flavonoids: Advances in research since 1986; Harborne, J. B., Ed.;
Chapman and Hall: London, 1994; pp 499−536.
(
21) Intelmann, D.; Batram, C.; Kuhn, C.; Haseleu, G.; Meyerhof, W.;
Hofmann, T. Three TAS2R bitter taste receptors mediate the
psychophysical responses to bitter compounds of hops (Humulus
lupulus L.) and beer. Chemosens. Percept. 2009, 2, 1936−5802.
REFERENCES
■
(
22) Taniguchi, Y.; Taniguchi, H.; Yamada, M.; Matsukura, Y.;
(
1) Fritsch, H. T.; Schieberle, P. Identification based on quantitative
measurements and aroma recombination of the character impact
Koizumi, H.; Furihata, K.; Shindo, K. Analysis of the components of
hard resin in hops (Humulus lupulus L.) and structural elucidation of
their transformation products formed during the brewing process. J.
Agric. Food Chem. 2014, 62, 11602−11612.
23) Chadwick, L. R.; Nikolic, D.; Burdette, J. E.; Overk, C. R.; Bolton,
J. L.; van Breemen, R. B.; Frohlich, R.; Fong, H. H. S.; Farnsworth, N. R.;
Pauli, G. F. Estrogens and congeners from spent hops. J. Nat. Prod. 2004,
7, 2024−2032.
24) Stevens, J. F.; Ivancic, M.; Hsu, V. L.; Deinzer, M. L.
Prenylflavonoids from Humulus lupulus. Phytochemistry 1997, 44,
575−1585.
25) Stevens, J. F.; Taylor, A. J.; Nickerson, G. B.; Ivancic, M.; Henning,
odorants in a Bavarian Pilsner-type beer. J. Agric. Food Chem. 2005, 53,
7
(
(
544−7551.
2) Kaltner, D.; Steinhaus, M.; Mitter, W.; Biendl, M.; Schieberle, P.
R)-Linalool as key flavor for hop aroma in beer and its behavior during
beer staling. Monatsschr. Brauwiss. 2003, 56, 192−196.
3) Palamand, S. R.; Aldenhoff, J. M. Bitter tasting compounds of beer.
(
̈
(
6
(
Chemistry and taste properties of some hop resin compounds. J. Agric.
Food Chem. 1973, 21, 535−543.
(
4) Schieberle, P. Primary odorants of pale lager beer. Differences to
1
(
other beers and changes during storage. Z. Lebensm.-Unters. Forsch.
991, 193, 558−565.
5) Collin, S.; Nizet, S.; Gros, J. Le houblonnage a
speciales belges est bien plus qu’une simple dissolution des composes
aromatiques du houblon. Cerevisia 2012, 36, 119−124.
6) Praet, T.; Van Opstaele, F.; Jaskula-Goiris, B.; Aerts, G.; De
1
J.; Haunold, A.; Deinzer, M. L. Prenylflavonoid variation in Humulus
lupulus: Distribution and taxonomic significance of xanthogalenol and
(
̀ ̀
cru des bieres
́
́
4
′-O-Methylxanthohumol. Phytochemistry 2000, 53, 759−775.
(26) Almaguer, C.; Gastl, M.; Dresel, M.; Hofmann, T.; Becker, T.
Comparison of the quality parameters of hop hard resins. Poster EBC
(
Cooman, L. Biotransformation of hop-derived aroma compounds by
Congress, Glasgow, Scotland, 2011.
saccharomyces cerevisiae upon fermentation. Cerevisia 2012, 36, 125−
(27) Schulmeyr, J.; Dresel, M.; Hofmann, T. Hop resin-based
substance. PCT Int. Appl. WO2012100953 (A1), 2012.
1
32.
(
7) Gros, J.; Nizet, S.; Collin, S. Occurrence of odorant polyfunctional
(28) Intelmann, D.; Haseleu, G.; Hofmann, T. LC-MS/MS
quantitation of hop-derived bitter compounds in beer using the
ECHO technique. J. Agric. Food Chem. 2009, 57, 1172−1182.
(29) Wilhelm, H.; Wessjohann, L. A. An efficient synthesis of the
phytoestrogen 8-prenylnaringenin from xanthohumol by a novel
demethylation process. Tetrahedron 2006, 62, 6961−6966.
thiols in the super alpha Tomahawk hop cultivar. Comparison with the
thiol-rich Nelson Sauvin bitter variety. J. Agric. Food Chem. 2011, 59,
8
(
853−8865.
8) Dunkel, A.; Steinhaus, M.; Kotthoff, M.; Nowak, B.; Krautwurst,
D.; Schieberle, P.; Hofmann, T. Nature’s chemical signatures in human
olfaction: A foodborne perspective for future biotechnology. Angew.
Chem., Int. Ed. 2014, 53, 7124−7143.
̈
(30) Drawert, F. Resin fractionation: modified Wollmer method (in
German). In Brautechnische Analysenmethoden: Band I+II (MEBAK);
Drawert, F., Ed.; MEBAK: Freising-Weihenstephan: 1979; pp 212−233.
(
9) Jaskula, B.; Kafarski, P.; Aerts, G.; De Cooman, L. A kinetic study
on the isomerization of hop α-acids. J. Agric. Food Chem. 2008, 56,
408−6415.
10) Verzele, M.; De Keukeleire, D. Chemistry and analysis of hop and
beer bitter acids; Elsevier: Amsterdam, London, New York, Tokyo, 1991.
11) De Keukeleire, D. Fundamentals of beer and hop chemistry.
Quim. Nova 2000, 23, 108−112.
12) Kowaka, M.; Kokubo, E. Composition of bitter substances of hops
and characteristics of beer bitterness. J. Am. Soc. Brew. Chem. 1976, 35,
6−21.
13) Verzele, M.; Jansen, H. E.; Ferdinandus, A. Organoleptic trials
with hop bitter substances. J. Inst. Brew. 1970, 76, 25−28.
14) Fritsch, A.; Shellhammer, T. H. Alpha-acids do not contribute
bitterness to lager beer. J. Am. Soc. Brew. Chem. 2007, 65, 26−28.
15) Haseleu, G.; Intelmann, D.; Hofmann, T. Identification and RP-
(31) Stevens, R. The chemistry of hop constituents. Chem. Rev. 1967,
1
(
9−71.
32) Walker, T. K.; Zakomorny, M.; Blakebrough, N. Delta resin, a
water-soluble, bitter, bacteriostatic portion of resin of hops. J. Inst. Brew.
952, 58, 439−442.
33) Frank, O.; Ottinger, H.; Hofmann, T. Characterization of an
6
(
1
(
(
intense bitter-tasting 1H,4H-quinolizinium-7-olate by application of the
taste dilution analysis, a novel bioassay for the screening and
identification of taste-active compounds in foods. J. Agric. Food Chem.
(
1
(
2
(
001, 49, 231−238.
34) Brock, A.; Hofmann, T. Identification of the key astringent
compounds in spinach (Spinacia oleracea) by means of the taste dilution
analysis. Chem. Percept. 2008, 1, 268−281.
(
(35) Frank, O.; Blumberg, S.; Kunert, C.; Zehentbauer, G.; Hofmann,
(
T. Structure determination and sensory analysis of bitter-tasting 4-
vinylcatechol oligomers and their identification in roasted coffee by
means of LC-MS/MS. J. Agric. Food Chem. 2007, 55, 1945−1954.
HPLC-ESI-MS/MS quantitation of bitter-tasting β-acid transformation
products in beer. J. Agric. Food Chem. 2009, 57, 7480−7489.
(
16) Intelmann, D.; Kummerlowe, G.; Haseleu, G.; Desmer, N.;
(
36) Chadwick, L. R.; Pauli, G. F.; Farnsworth, N. R. The
Schulze, K.; Frohlich, R.; Frank, O.; Luy, B.; Hofmann, T. Structures of
storage-induced transformation products of the beer’s bitter principles,
revealed by sophisticated NMR spectroscopic and LC-MS techniques.
Chem.Eur. J. 2009, 15, 13047−13058.
pharmacognosy of Humulus lupulus L. (Hops) with an emphasis on
estrogenic properties. Phytomed. 2004, 13, 119−31.
(37) Chadwick, L. R.; Fong, H. H. S.; Farnsworth, N. R.; Pauli, G. CCC
sample cutting for isolation of prenylated phenolics from hops. J. Liq.
Chromatogr. Relat. Technol. 2005, 28, 1959−1969.
(
17) Haseleu, G.; Intelmann, D.; Hofmann, T. Structure determination
and sensory evaluation of novel bitter compounds formed from β-acids
of hop (Humulus lupulus L.) upon wort boiling. Food Chem. 2009, 116,
(38) Zhao, F.; Watanabe, Y.; Nozawa, H.; Deikonnya, A.; Kondo, K.;
Kitanaka, S. Prenylflavonoids and phloroglucinol derivatives from hops
Humulus lupulus). J. Nat. Prod. 2005, 68, 43−49.
39) Nookandeh, A.; Frank, N.; Steiner, F.; Ellinger, R.; Schneider, B.;
7
1−81.
18) De Keukeleire, D.; Verzele, M. The absolute configuration of the
isohumulones and the humulinic acids. Tetrahedron. 1971, 27, 4939−
945.
19) Han
and cyclization to give flavanones (in German). Arch. Pharm. 1988, 321,
7−40.
(
(
(
Gerhauser, C.; Becker, H. Xanthohumol metabolites in faeces of rats.
Phytochemistry 2004, 65, 561−570.
(40) Wackerbauer, K.; Kramer, P.; Siepert, J. Phenolic aromatic
substances in beer. Phenolic carboxylic acid and phenols in the raw
material and beer preparation. Brauwelt 1982, 122, 618−626.
4
(
̈
sel, R.; Schulz, J. Demethylxanthohumol: isolation from hops
3
P
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX

Journal of Agricultural and Food Chemistry
Article
(
41) Vancraenenbroeck, R.; Vanclef, A.; Lontie, R. Characterization of
flavonols and glucoside of phloroisobutyrophenone in hops. Proc.
Congr.Eur. Brew. Conv. 1965, 360−371.
(
42) Schwarz, B.; Hofmann, T. Sensory-guided decomposition of red
currant juice (Ribes rubrum) and structure determination of key
astringent compounds. J. Agric. Food Chem. 2007, 55, 1394−1404.
(
̈
43) Bohr, G.; Gerhauser, C.; Knauft, J.; Zapp, J.; Becker, H. Anti-
inflammatory acylphloroglucinol derivatives from hops (Humulus
lupulus). J. Nat. Prod. 2005, 68, 1545−1548.
(
44) Honma, D.; Tagashira, M.; Kanda, T. 2-Acylphloroglucinol-4,6-
di-C-β-D-glucopyranoside having antioxidant action. U.S. Pat. Appl.
Publ. JP 2008 174458, 2008.
(
̈
45) Gerhauser, C.; Alt, A. P.; Klimo, K.; Knauft, J.; Frank, N.; Becker,
H. Isolation and potential cancer chemopreventive activities of phenolic
compounds of beer. Phytochem. Rev. 2003, 1, 369−377.
(
46) Kane, M. A.; Folias, A. E.; Napoli, J. L. HPLC/UV quantitation of
retinal, retinol, and retinyl esters in serum and tissues. Anal. Biochem.
008, 378, 71−79.
47) R Development Core Team. R: A language and environment for
2
(
statistical computing; R Foundation for Statistical Computing: Vienna,
Austria, 2008; http://www.R-project.org.
(
48) Ward, J. H. Hierarchical Grouping to optimize an objective
function. J. Am. Stat. Assoc. 1963, 58, 236−244.
49) McMurrough, I.; Cleary, K.; Murray, F. Applications of high-
(
performance liquid chromatography in the control of beer bitterness. J.
Am. Soc. Brew. Chem. 1986, 44, 101−108.
(
50) Chadwick, L. R. Estrogens and congeners from hops. Dissertation,
University of Illinois, Chicago, IL, 2004.
51) Kuznetsov, A. G.; Bush, A. A.; Rybakov, V. B.; Babaev, E. V. An
(
improved synthesis of some 5-substituted indolizines using regiospecific
lithiation. Molecules 2005, 10, 1074−1083.
(
52) Andrei, C. C.; Ferreira, D. T.; Faccione, M.; De Moraes, L. A. B.;
Carvalho, M. G.; Braz-Filho, R. C-Prenylflavonoids from roots of
Tephrosia tunicata. Phytochemistry 2000, 55, 799−804.
(
53) Riedl, W.; Nickl, J. C,C-Hydrogenolysis of hop bitter compounds
(
in German). Chem. Ber. 1956, 89 (8), 1838−1849.
(
54) Nagel, J.; Culley, L. K.; Lu, Y.; Liu, E.; Matthews, P. D.; Stevens, J.
F.; Page, J. E. EST Analysis of hop glandular trichomes identifies an O-
methyltransferase that catalyzes the biosynthesis of xanthohumol. Plant
Cell 2008, 20, 186−200.
(
55) Bufe, B.; Hofmann, T.; Krautwurst, D.; Raguse, J.-D.; Meyerhof,
W. The human TAS2R16 receptor mediates bitter taste in response to β-
glucopyranosides. Nat. Genet. 2002, 32, 397−401.
(
56) De Clippeleer, J.; De Cooman, L.; Aerts, G. Beer’s bitter
compounds − A detailed review on iso-alpha-acids: Current knowledge
of the mechanisms for their formation and degradation. Brew. Sci. 2014,
6
(
7, 167−182.
57) Research Brewery St. Johann and Barth-Haas-Group, Internal
report, 2012.
58) Van Opstaele, F. Hoppy aroma of beer: Characterisation of
(
sensorially differentiated hop aromas and evaluation in brewing practice.
Dissertation, Katolieke Universiteit Leuven, Heverlee, Belgium, 2011.
(
59) Fraga, C. G. Plant phenolics and human health - Biochemistry,
nutrition, and pharmacology; John Wiley & Sons: Hoboken, NJ, 2010.
60) Piendl, A.; Biendl, M. On the physiological impact of polyphenols
and hop bitter compounds in beer (in German). Brauwelt 2000, 140,
26−537.
61) Hieronymus, S. For the love of hops. The practical guide to aroma,
bitterness and the culture of hops; Brewers Publications: Boulder, CO,
012.
(
5
(
2
Q
DOI: 10.1021/acs.jafc.5b00239
J. Agric. Food Chem. XXXX, XXX, XXX−XXX