Analytical determination of antioxidant capacity of hop-derived compounds in beer using specific rapid assays (ORAC, FRAP) and ESR-spectroscopy

There is a relationship between antioxidant activity and ageing stability of beer. The high-throughput antioxidant capacity assays ORAC and FRAP, which rely on specific reaction mechanisms, are validated as an addition to the established ESR-ST method with a focus on hop-derived antioxidants. Beers were brewed with systematic variations in hop variety, hop product, and the hopping regime (late hopping) to achieve sample beers varying widely in the concentrations of hop-derived antioxidants (α- and iso-α-acids, phenolic compounds). A significant positive correlation between phenolic compounds and the ORAC- and FRAP-values (p < 0.01) was found. The effect of individual resinous substances and free phenolic acids and flavonoids on the antioxidant capacity of beer (ORAC, FRAP, and ESR-ST) was proven in spiking trials. We found a correlation between the occurrence of the o-di-OH-group and the ORAC- and FRAP-values of phenolic compounds. Phenolic compounds did not react as anti- or prooxidants in the radical generation (ESR-ST). Higher concentrations of unisomerized α-acids significantly reduced ESR-signal intensity but had no significant effect on ORAC- or FRAP-values. Beers brewed with late hop addition, which yielded higher concentrations of unisomerized α-acids and phenolic compounds, had higher ORAC, FRAP-values and a reduced ESR-signal intensity. These three methods rely on different reaction principles, and therefore, different groups of hop-derived compounds act as antioxidants in these assays. A combination of the two high-throughput methods (ORAC, FRAP) and ESR-ST is advantageous for the evaluation of the antioxidant capacity of beers varying in hop-derived compounds.


Introduction
Antioxidants are a field of active research in food science due to their potential health benefits and their effects on food stability [1]. Beer stability is defined as ageing stability (sensory changes occurring during ageing) and as haze stability (appearance of turbidity). Turbidities in beer are caused by the interaction of proteins with polyphenolic compounds. The sensory changes during storage time are caused by a variety of oxidative and non-oxidative reactions. Oxidative beer staling is caused by reactive oxygen species (ROS), which are generated from stable triplet oxygen ( 3 O 2 ) in a series of metal-induced reduction reactions [2]. The triplet state is the ground state of oxygen; both π* orbitals are occupied by two electrons with parallel spin. This ground state is the reason for oxygen's paramagnetic and diradical characteristics. Single-state oxygen ( 1 O 2 ) is formed from stable 3 O 2 in photochemical reactions. The π* orbitals are occupied by two electrons with antiparallel spin, making single-state oxygen diamagnetic. 1 O 2 is highly reactive and a strong oxidizing agent. The higher the reduction status of the oxygen species, the higher is its reactivity, ultimately resulting in •OH-radicals, which react unselectively with the organic compounds in beer in almost diffusion-controlled reactions [2,3]. The reductive oxygen activation is promoted by compounds which are able to reduce metal ions, thus recycling the prooxidant reduced metal ion. However, antioxidants are Contribution for the Special Issue: The chemistry behind malt and beer production -from raw material to product quality. able to inhibit the effects of oxygen by capturing ROS or free radicals or by complexing transition metal ions. [2] Antioxidants act by different reaction principles: as reducing agents, radical scavengers, inhibitors of reactive oxygen species or by complexing prooxidant metal ions [4]. A positive influence of hop-derived antioxidants on beer ageing stability is discussed [5][6][7][8][9].
As a yeast-fermented beverage made from malted barley (Hordeum vulgare L.) and hops (Humulus lupulus L.), beer contains a number of antioxidant compounds. Barley malt contributes reducing Maillard reaction compounds, antioxidant amino acids, and phenolic substances to the antioxidant potential of beer [10,11]. Valuable hop compounds are generally classified as hard and soft resinous compounds, polyphenols, and hop essential oils [12]. Hop hard resin contains prenylated chalcones like xanthohumol, which have been intensely researched due to their potential health benefits [13]. Phenolic acids and flavonoids are the predominant classes of low-molecular-weight phenolic substances in beer [11]. Among flavonoids, the most common compounds in beer are monomeric flavanols (e.g., catechin, epicatechin) and their condensation products, the oligomeric flavanols or proanthocyanidins (e.g., procyanidin B3). Phenolic acids, and monomeric and oligomeric flavanols are potent antioxidants [11,[14][15][16].
Electron spin resonance spin trapping (ESR-ST) is the method of choice to assess antioxidant activity in beer quality control [17][18][19][20][21][22]. Radical generation in beer samples during oxidative forced ageing follows a sigmoidal curve. After a lag-time, in which no radicals are detected, the concentration of radicals in the sample increases exponentially and then reaches a steady level. [3,18] The length of the lag phase was shown to be linearly dependent on the SO 2 content of beer [21][22][23]. Hop-derived substances impact radical formation in beer. Unisomerized α-acids act as antioxidants by complexing and precipitating prooxidant metal ions and thus reduced ESR-signal intensity in a number of studies, which resulted in a better ageing stability [24][25][26][27]. However, polyphenols had only a negligible effect on ESRsignal intensity [19,26,28]. Recently, Marques et al. have proposed a good correlation between sensory data and the area under the radical generation curve [29].
Antioxidant activity of a substance is defined as its ability to reduce prooxidative or reactive species of pathological significance [30]. Therefore, antioxidant activity deals with the kinetics of the reaction between the antioxidant and the prooxidant. The term antioxidant capacity is used for the thermodynamic efficiency with which a prooxidant probe is conversed upon reaction with an antioxidant [31]. The antioxidant assays used in this work measure the antioxidant capacity of an antioxidant in the reaction with different prooxidants. Antioxidant capacity assays are classified by their reaction mechanisms into electron transfer (ET) and hydrogen atom transfer (HAT) reactions. Methods based on an ET-mechanism (e.g., Ferric Reducing Antioxidant Power, FRAP) use the reduction of a transition metal ion complex [34,35]. HAT-based reactions measure the radical scavenging or radical chain breaking ability of antioxidants by hydrogen donation. The Oxygen Radical Absorbance Capacity (ORAC)-assay as a classical HAT method measures radical chain breaking ability of antioxidants against peroxyl radicals and uses a fluorescent probe as the oxidizable substrate [32,36]. The aim of this work was to test the validity of ORAC and FRAP for the prediction of the antioxidant capacity of beer as an addition to ESR-ST. In targeted brewing trials, we systematically varied the hop product and the hopping regime to achieve sample beers, which varied widely in the content of hop-derived antioxidants: resinous (iso-α-acids and unisomerized α-acids) and phenolic compounds. These sample beers were used as a tool to evaluate the explanatory power of the antioxidant capacity assays ORAC and FRAP for hop-derived antioxidants alongside the established ESR-ST method. In complex food samples like beer, it is difficult to ascertain the influence of specific substances or substance groups on the antioxidant capacity of the whole beer sample. Therefore, we used spiking trials to verify that specific groups of hop-derived compounds reacted in the ORAC-and FRAP-assays and had an influence on radical formation in ESR-ST. Apart from that, the antioxidant capacity of individual phenolic acids and flavonoids was measured by ORAC and FRAP to investigate the effect of structural motives on antioxidant potential as well as to estimate the contribution of specific substances to antioxidant capacity.

Brewing trials
This work is based on brewing trials, which aimed at generating beers with a wide variation in the concentrations of hop-derived antioxidants. The proportion of the different groups of hop substances in the final beer depends on the type of hop product (hop extracts or pelletized hops), the hop variety, and the hopping technology. All three of these factors were systematically varied in the brewing trials, to achieve wide variations in the concentrations of hop-derived antioxidants. Bitter hop varieties with a high content of α-acids usually have lower polyphenol contents [12]. Whereas supercritical CO 2 is specific for the solubilization of soft resins (α-and β-acids) and hop essential oils (CO 2 -extract, CEX), ethanol also extracts hard resinous compounds like prenylflavanoids (ethanol extract, EEX) [7].

3
The aqueous extract mainly contains glycosidically bound hop phenolic compounds (tannin extract, TEX) [7]. A late hop addition yields higher concentrations of hop polyphenols as longer hop boiling times lead to a greater depletion of polyphenols [8]. Hop dosage time also influences the isomerization of α-acids, and a late hop addition results in beers with a higher concentration of unisomerized α-acids [25].
All beers were brewed from the same unhopped wort (brewed from pilsner malt), so the differences between the beers are solely due to the variations in the hopping regime (see Table 1). As hop varieties Hallertauer Herkules (1), content of α-acids: 12-17% and Hallertauer Tradition (2), content of α-acids: 4-7%, were chosen. Four different hop products CEX, EEX, TEX, and hop pellets type 90 (Pe) were used at the beginning of wort boiling to study the influence of the hop product. The hopping regime was varied by the time of hop addition: at the beginning of wort boiling with and without a second hop addition in the whirlpool. The second hop addition during whirlpool rest consisted of 2.5 g/L Pe.
All sample beers were brewed in duplicate. Hop addition at the beginning of wort boiling was calculated to achieve 25 bitter units (measured by HPLC as iso-α-acids, see "Standard analyses"). Worts were boiled for 60 min. For some sample beers, 2.5 g/L hop pellets were added during the whirlpool rest. After a whirlpool rest of 10 min, the worts were cooled.
Yeast (Saflager W-34/70 by Fermentis, rehydrated and allowed to revive in first wort) was added and worts were fermented for 5 days at 12 °C until their extract value dropped below 3.5%-w/w. Green beer was matured at 16 °C until they reached a total diacetyl value below 0.1 mg/L (analyzed according to MEBAK 2.21.5.1 [17]). The beers were stored at 0 °C for 14 days and filtered (3 filter layers, K150, Pall Corporation). All sample beers were filled under CO 2 in 0.33 L longneck bottles and stored at 4 °C.

Spiking trials
For the spiking trials, a fresh commercial bright lager beer was obtained from a local brewery. It contained 5% ethanol, 17.5 mg/L total iso-α-acids, and 0.7 mg/L total α-acids (measured according to MEBAK 2.17.3 [17]). Individual phenolic compounds (pHB, prot, van, syr, p-cou, fer, sin, cin, cat, and rut) were determined in triplicate by HPLC following a solid-phase extraction clean-up procedure as published in Wannenmacher et al. (2019) [37]. All spiked samples [three different substance groups: (1) phenolic compounds (see Table 4), (2) iso-α-acids, (3) α-acids] were prepared on 2 consecutive days (technological replicates: n = 2). ESR-measurements were made from the freshly spiked sample, FRAP-and ORAC-values were determined from sample aliquots kept at − 18 °C until analysis. Phenolic compounds dissolved in ethanol were added to the degassed sample (20 min iced ultrasonic bath) in the following steps:

Antioxidant capacity analyses
High-throughput assays: For the determination of the antioxidant capacity in the FRAP-and ORAC-assays, 96-well plates and a synergy H4 (Biotek, Bad Friedrichshall, Germany) micro-plate reader were used. All samples were stored at − 18 °C before analysis. Both the FRAP-and ORAC-values were calculated using a trolox standard calibration. For the FRAP-assay according to Jimenez-Alvarez et al. 2008 [35], samples were diluted in distilled water prior to analysis. Reducing activity was measured in quadruples. 25 µL of the samples were pipetted into a 96-well plate and 250 µL of FRAP-reagent [acetate buffer (300 mM, pH = 3.6)/ TPTZ (10 mM in 400 mM HCl)/iron(III) chloride (20 mM, aqueous), 10/1/1, v/v/v] was added. The plate was agitated and then incubated at 25 °C for 8 min prior to measurement of the absorbance at 593 nm. For the determination of the ORAC-value according to Spreng et al. 2018 [10], samples were diluted in phosphate buffer (PB, 10 mM, pH = 7.4) and measured in quadruples. 25 µL of the samples were pipetted into the 96-well plate and 150 µL of fluorescein solution (10 nM, in PB) was added, and the plate was agitated and then incubated at 37 °C for 30 min. Fluorescence was then measured in three cycles (f 0 , every 90 s, excitation: 485 nm, emission: 520 nm) prior to addition of AAPH (240 mM, in PB). After the addition of AAPH, fluorescence was measured in 57 cycles (f i , every 90 s, excitation: 485 nm, emission: 520 nm). Antioxidant capacity as trolox equivalents was determined by calculating the area under the fluorescence decay curve (AUC) relative to the trolox standard calibration according to Ou et al. 2001 [36] ESR-spectroscopy: Radical generation in the 60 °C forcing test was determined according to MEBAK 2.15.3 [17] with modifications. Briefly, sample beers were degassed in an iced ultrasonic bath for 20 min. Ethanol content of the sample beers (5.0 ± 0.3%Vol) was adjusted to 6.15 ± 0.3%Vol by adding 150 µL ethanol to 12 mL sample in the ESR-Vial. All vials were cleaned with Edisonite ® (Merz Consumer Care GmbH, Frankfurt, Germany) and distilled water prior to measurement to reduce interferences. A 10 µM aqueous solution of the stable organic radical TEM-POL was used as a standard. All ESR-measurements were carried out using an e-scan beer analyzer (Bruker BioSpin, Rheinstetten, Germany) equipped with an autosampler with a heating block (Techne Driblock DB-3, Labtech International, Burkhardtsdorf, Germany) and a peristaltic pump (Gilson, Limburg-Offheim, Germany). The e-scan settings were optimized using a 10 µM TEMPOL standard solution: center field: 3480 G, sweep width: 17 G, attenuation: 7 dB, power: 3.13 mW, receiver gain: 2.24 * 10 2 , modulation amplitude: 1.96 G, modulation frequency: 86 kHz, time constant: 5.12 ms, conversion time: 40.96 ms, resolution: 512 and scans: 8. The spin-trap POBN was dissolved in water to a final concentration of 865 mM, and the spin-trap solution was added to the beer samples to achieve a concentration of 3.5 mM POBN in the sample directly before inserting the vial into the heating block (set to 60 ± 0.5 °C) of the autosampler. All ESR-measurements were carried out in triplicates. Radical generation in the 60 °C forcing test follows a sigmoidal curve over the forcing test time. It is described by the lag-time during which no radicals are formed, by different time points in the radical generation curve (T150: signal intensity after 150 min and T400: signal intensity after 400 min) and by the area under the radical generation curve (AUC). From the radical generation curves, T150 and T400-values (radical generation after a forcing test time of 150 and 400 min, respectively) were calculated from a logistic fit with T(t): signal intensity at time t, D, A, C, B: constants. The sigmoidal curve was fitted to the experimental values using the solver add-in of excel. Signal intensities are given as relative values using the 10 µM TEMPOL solution as a standard. Area under the signal intensity curves (AUC) was calculated in Origin 2017 from a logistic fit [see (3)]. Relative signal intensity values were calculated by dividing the signal intensity of the sample at a specific time point by the respective signal intensity of the TEMPOL standard solution. The logistic curve of relative intensity values over time was integrated between t 0 = 0 min and t 1 = 400 min.

Statistical data analysis
Statistical evaluations were performed using the SAS JMP ® 12.2.0 (64 bit) Software, SAS Institute, USA. Analytical values of the sample beers are presented as mean ± standard deviation of the two technological replicates and multiple analytical determinations. Before multivariate statistical evaluations, data was normalized with n: normalized value, x: sample mean, x, min: minimum sample mean, and x, max: maximum sample mean. Pearson correlation coefficients between beer composition data and antioxidant capacity values were calculated. A p value below 0.05 was considered significant. Data were checked for normal distribution by the Shapiro-Wilk test. Differences between groups of samples were analyzed by ANOVA and Tukey HSD, and a p value below 0.05 was considered significant. For the analysis of differences between groups of sample beers, all technological replicates (n = 2 for each variant of the hopping regime) were taken into account. (3)

Targeted brewing trials-correlations between ORAC-, FRAP-, and ESR-ST-values and hop-derived compounds
The targeted brewing trials conducted in this study aimed at generating lager beers with a wide variation in the content of different groups of hop-derived compounds: resinous (iso-α-acids and unisomerized α-acids) and phenolic compounds (see Table 2). At the same time, the sample beers were comparable in standard quality parameters: alcohol content (5.0-5.3%Vol), extract (1.98-2.17 g/100 mL), color (5.3-6.5), and pH (4.1-4.5).
The beer samples were evaluated for their antioxidant capacity by the ORAC-and FRAP-assays and by ESR-ST (see Table 3) to validate these assays for beers with variations in hop-derived antioxidants.
The relationship between the concentrations of hopderived beer ingredients (phenolic hop compounds, unisomerized α-acids, iso-α-acids) and antioxidant capacity was evaluated by multivariate linear correlations (see Table 4). All beer samples were evaluated for their antioxidant capacity by ESR-ST. As expected, all values calculated from the radical generation curve (T150, T400, AUC) correlated significantly positively with each other (p < 0.01). A significant (p < 0.05) negative relationship between the concentration of iso-α-acids and bitter units and the T400value was found. In a study by Wietstock et al., unisomerized hop α-and β-acids reduced the ESR-signal intensity in experiments with hopped worts [26]. There was no significant relationship between the concentration of phenolic compounds and ESR-ST values. An insignificant effect of phenolic compounds on radical generation as measured by ESR-ST has been described by other authors [19,38].
Reducing capacity (FRAP) and radical scavenging capacity (ORAC) of the sample beers correlated significantly positively with each other and with total polyphenol content (p < 0.01). Both correlated positively with total anthocyanogen content (FRAP: p < 0.01, ORAC: p < 0.05). FRAP-values correlated significantly positively with beer color. Beer color is influenced by melanoidins, which act as reducing agents due to their enediol groups [2,39]. Total polyphenol and anthocyanogen content correlated significantly positively with FRAP-values (p < 0.01). Phenolic compounds play a role for beer color and especially for color changes during beer storage [40,41].

Main variations in the hopping regime and antioxidant capacity (ORAC; FRAP, ESR)
The sample beers were grouped according to the main variations in the hopping regime: Hop variety (variety 1: Hallertauer Herkules, variety 2: Hallertauer Tradition) and time of hop addition (WP: second hop addition in the whirlpool, nWP: without a second hop addition in the whirlpool). Antioxidant capacity of the grouped sample beers was determined by ORAC and FRAP and by ESRspectroscopy, and phenolic compounds were determined by the total polyphenols assay (EBC) and the total anthocyanogen assay (see Table 3 and Fig. 1).
These results confirm that hop polyphenols react as antioxidants in the ORAC-and FRAP-assays. Our results showed that when hop is added in higher quantities and during whirlpool rest, higher concentrations of hop phenolic compounds in the wort survive into the final beer. Beers with a second hop addition in the whirlpool had significantly higher total polyphenol values than beer without a second hop addition (p < 0.05). These beers had a significantly higher reducing power (FRAP) and higher peroxyl radical scavenging activity (ORAC). The beers hopped with hop variety 2 and a second hop addition had the highest antioxidant capacity (ORAC and FRAP) and the highest polyphenol content.
The sample beers in this study did not exhibit a lagtime; therefore, ESR-ST measurement data are discussed by T150, T400, and AUC-values (see Table 3 and Fig. 1B). The results from ESR-ST show the inhibiting effect of higher concentrations of hop resinous compounds (iso-α-and unisomerized α-acids) on radical generation, the sample beers hopped with a second hop addition and the beers hopped with hop variety 1 had higher concentration ranges of hop resinous compounds and lower ESR-ST-values. A positive impact of unisomerized α-acids on oxidative beer stability due to the complexation and precipitation of prooxidative metal ions has been described by other authors [24,25,27]. There were no significant differences in antioxidant capacity between the beers hopped with different hop products at the beginning of wort boiling (results not shown).

Spiking trials: verification of the antioxidant capacity of specific compounds using ORAC, FRAP, and ESR-ST
The validity of the high-throughput assays ORAC and FRAP was tested alongside the established ESR-spectroscopic method in spiking trials. All spiking trials were carried out with the same commercial lager beer sample. The antioxidant capacity of unisomerized α-acids in ESR-spectroscopic experiments has been reported by other authors [24][25][26][27]. However, the antioxidant capacity of unisomerized α-acids in beer has not yet been tested in the high-throughput assays ORAC and FRAP. Spiking of unisomerized α-acids had no significant influence on the ORAC-values of the samples (see Table 6). However, a significant reduction in the radical generation by ESR-ST was observed for AUC and T400 (see Table 5 and Fig. 2) for the samples spiked with the highest concentration of α-acids compared to the samples without addition and the lowest level of addition of unisomerized α-acids. Results of previous studies also showed a marked decrease in T600 (ESR-signal intensity after 600 min forcing test time) with higher concentrations of unisomerized α-acids in hopped worts [25,26]. There were significant differences in the FRAP-values; however, no linear relationship between α-acid content and reducing capacity was observed.
Iso-α-acids were spiked in five steps (see Table 5). No significant differences were observed for ORAC-and FRAPvalues. There were no significant differences in the values calculated from the radical generation curve during the oxidative forcing test as measured by ESR. This is in accordance with studies on hopped worts, where T600 ESR-signal intensity was shown to increase with higher degrees of α-acid isomerization [26]. In a study using the DPPH-assay, the antioxidant capacity of iso-α-acids was markedly lower than that of unisomerized α-acids or phenolic compounds [26,42].
The phenolic acids p-hydroxybenzoic, protocatechuic, vanillic, syringic, p-coumaric, ferulic, and sinapic acids are found in beer in significant concentrations as free phenolic acids and these substances are potent antioxidants [11,[14][15][16]37]. These substances also differ in the number and nature of substituents of the phenolic ring. Therefore, those phenolic acids were chosen for our spiking experiments as well as the flavonoid compounds rutin and catechin. A mix of the phenolic standard compounds was spiked to the lager beer in the concentration ratio previously determined (see Table 6) in four concentration steps. The concentrations of these phenolic acids and flavonoids were shown to be significantly higher in whirlpool-hopped sample beers by Wannenmacher et al. [37]. Antioxidant capacity of the spiked beer samples was determined in the FRAP-and ORAC-assays and ESR-ST (see Table 5). Whereas no significant differences were observed by ESR-ST and the ORAC-assay, the FRAP-values of spiked samples were significantly higher. A linear relationship between the content of spiked phenolic substances and the FRAP-value was observed. All of the phenolic compounds used in the spiking trials were evaluated for their individual antioxidant capacity in the ORACand FRAP-assays. For both assays, the capacity values presented here are in good accordance with the literature values; however, variations due to differences in experimental procedures were observed (see Table 6). The reducing capacity in the FRAP-assay of both benzoic and cinnamic acid derivatives depended on the number and nature of functional groups. In the case of benzoic acids, reducing activity in the FRAP-assay decreased in the following order: gallic (3, 4, 5-tri-OH) > syringic (3, 5-di-OMe, 4-OH) > protocatechuic (3, 4-di-OH) > vanillic (3-OMe, 4-OH) > p-hydroxybenzoic acid (4-OH). This result is in accordance with other studies, where gallic acid had the highest activity, while p-hydroxybenzoic acid had no detectable activity in the FRAP-assay [43]. As described by other authors, phenolic OH-groups have a higher antioxidant capacity than phenolic OMe-groups [44]. The reducing capacity in the FRAP-assay of cinnamic acid derivatives increased with the number of functional groups. FRAP-values of the isomers catechin and epicatechin were similar and in the same range as the value of protocatechuic acid, which also bears an o-di-OH-group. ORAC-values were especially high for phenolic compounds with an o-di-OH-group: catechin, epicatechin, protocatechuic acid, and rutin. This is in accordance with other studies [45][46][47][48]. However, phenolic acids with a single OH-group like p-hydroxybenzoic and p-coumaric acid showed high ORAC-values (see Table 6).

Discussion and conclusions
Antioxidants in beer have been a field of active research due to their potential to enhance the storage stability of beer [10,[24][25][26][27]37]. There are different approaches to measuring antioxidant capacity of beers to predict the ageing stability: the established ESR-ST method and high-throughput assays, which rely on specific reactions under controlled conditions [18,33,49]. The aim of this research was to evaluate the specific rapid antioxidant capacity assays ORAC and FRAP Table 4 Correlation coefficients (Pearson r) between antioxidant capacity and polyphenols/hop bitter substances The significant correlations are highlighted in bold (**p ≤ 0.01, *p ≤ 0.05, n In these experiments, a significant positive correlation between the concentration of total polyphenols and the ORAC-and FRAP-values was found. The antioxidant capacity of individual phenolic acids (p-hydroxybenzoic, protocatechuic, gallic, vanillic, syringic, p-coumaric, ferulic, and sinapic acid) and the monomeric flavanols catechin and epicatechin as well as the glycosylated flavonol rutin in the ORAC-and FRAP-assays was determined. The occurrence of ortho-di-OH-groups and the antioxidant capacity in both assays correlated. This is in accordance with the results of other authors [45][46][47]. The ability of an antioxidant to reduce a radical species depends on the stability of the antioxidant radical A-O• [50]. The antioxidant radical is stabilized by electronic delocalization and conjugation, which depends on the planar structure of the molecule. In phenolic compounds with an o-di-OH-group, the radical A-O• is stabilized by a neighboring O-H-group by hydrogen-bond formation [50]. However, phenolic acids with a single OH-group like p-hydroxybenzoic and p-coumaric acid also showed especially high ORAC-values (5 and 6 TE). In brewing quality control, phenolic compounds are mainly determined as total polyphenols by the EBC-method, which is based on the formation of colored complexes between phenolic compounds with an ortho-di-OH-group and iron (III) in alkaline solution [51][52][53][54][55]. However, compounds with a single OH-group do not react in the total polyphenols assay. As these compounds have high peroxyl scavenging activities, it is necessary to directly measure antioxidant capacity to arrive at a meaningful result with regards to the antioxidant status of the beer. The specific reaction mechanisms of ORAC and FRAP allow the differentiation between compounds acting as electron and hydrogen atom donors. The DPPH-assay measures the reduction of a stable organic nitrogen radical, which is based mainly on an ET reaction, the HAT-pathway being a negligible side reaction [33]. Due to the similar reaction mechanism, compounds reducing the DPPH-radical also react in the FRAP-assay. A combination of FRAP and ORAC can replace the DPPH-assay, since FRAP specifically measures the reduction of an iron complex in an ET reaction, whereas ORAC specifically measures radical chain breaking ability of antioxidants in an HAT reaction [33,49]. A combination of these two assays is advantageous as it allows the differentiation between the two reaction mechanisms.
As the ability to reduce transition metal ions like iron is essential for the prooxidant potential of reducing compounds, the FRAP-assay can give insight into the prooxidative behavior of phenolic compounds [2,32,34,35]. However, we found no relationship between FRAP-values and radical generation in the oxidative forcing test measured by ESR-ST. Spiking of free phenolic acids, flavanols, and glycosidically bound flavonol resulted in significantly higher FRAP-values, but did not have a significant accelerating effect on radical generation (ESR-ST). A prooxidant effect of the spiked phenolic compounds on radical generation could not be observed. No effect of phenolic substances on radical formation was observed by Andersen et al., but an accelerating effect of the reducing compound ascorbate on radical formation was shown [19]. The phenolic compounds tested here do not act as a reducing prooxidant in the oxidative forcing test, even though these compounds act as reducing compounds against the ferric complex of the FRAP-reagent. Therefore, high reducing activity values in the FRAP-assay are not necessarily linked to a high prooxidant potential of those reducing compounds.
Results from the brewing trials and the spiking trials have shown that higher levels of unisomerized α-acids lead to lower ESR-signals and thus lower radical concentrations in the oxidative forcing test. Apart from the elimination of prooxidant metal ions [25][26][27], unisomerized α-acids could  The high-throughput antioxidant capacity assays ORAC and FRAP are advantageous for quality control as they offer fast results and allow the simultaneous measurement of a high number of samples. A significant positive correlation between beer phenolic compounds and the ORACand FRAP-values was found. Phenolic compounds did not act as pro-or antioxidants in the radical generation under the conditions of the oxidative forcing test by ESR-ST. However, a significant effect of higher concentrations of unisomerized α-acids on ESR-signal intensity was found. Using only ESR-ST for antioxidant determination in beer can lead to an underestimation of the antioxidant contribution of phenolic compounds. Since ESR-ST, ORAC, and FRAP are based on different reaction principles, these assays allow a differentiated evaluation of the antioxidant capacity of beer samples. A combination of the two high-throughput methods and ESR-ST is advantageous for the evaluation of the antioxidant capacity of sample beers, especially with regard to hop-derived compounds.
Funding Open Access funding enabled and organized by Projekt DEAL.

Conflict of interest The authors declare no conflicts of interest.
Ethical approval This article does not contain any studies with human participants or animals performed by any of the authors.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.