**4. Discussion**

### *4.1. Total Phenols in Hop Cones*

In the four factorial experiments, a significant interaction was found between two or three factors of each experiment for several traits related to total phenols in the cones. This means that the effect of a factor on a given variable was dependent on the other(s) factor(s) under study, and the year was the factor with greatest influence. The accumulation of total phenols in cones in plants of different vigour, in those subject to different foliar treatments and grown in different plots, and between different cultivars was dependent on the year. Abram et al. [29] also reported that the year influenced the phenolic content of hop cones of different cultivars and of hop plants grown in different locations (Slovenia, Austria, Czech Republic). The year effect results from the combination of important environmental variables, such as precipitation, temperature, solar radiation, etc., which are able to influence physiological and biochemical processes in plants and also the efficiency of foliar nutrition [30]. The year had a marked effect on total phenol content. Total phenols showed lower values in 2018 in most experiments in comparison with 2017.

During important phases of the growing season, such as flowering and initial cone development (June, July), the temperature was lower in 2018 than in 2016 and 2017, and precipitation was higher (Figure 1). This region is at a low latitude, compared to Europe's major hop producing regions. In lower temperature years, plant growth conditions are closer to those observed at higher latitudes, where hops have better general growing conditions [2,8]. In several studies, it has been shown that the growing region, in general, has a grea<sup>t</sup> influence on the performance of hop plants [11,13,29,31–34]. It is also known that environmental variables can affect the secondary metabolism of plants and, therefore, the accumulation of phenolic compounds [35,36]. Although plant vigor had a marked effect on tissue nutrient concentration [24], its effect on total phenols in hop coves was reduced.

The average content of total phenols in hop cones of the Nugget cultivar did not vary significantly between the plots of different plant vigour. The stress affecting plant growth and yield in the low vigour plots did not influence total phenols in the cones. A previous study analysing these plots [21] has shown that the plants appeared with excessive levels of Fe and Mn in the leaves, which may indicate poor soil aeration, probably caused by a deficient spatial water distribution along the rows by the flooding irrigation system. The soil texture in these plots did not seem to be different enough to create a gradient effect. Phenols significantly decreased with liming treatment. Likewise, Zu et al. [37] found a decrease in the flavonoid content of *Panax notoginseng* with calcium and lime application under cadmium stress. Although calcium seems to have an inhibitory effect on important enzymes in the phenolic pathway, it seems that the greater amount of cadmium in the roots inhibited the absorption of calcium and influenced flavonoid content. Unfortunately, with the data collected, it was not possible to identify the stress factors that caused the reduction in the content of phenols in the limed plots.

The foliar sprays did not influence significantly the content of total phenols in hop cones. To the best of our knowledge, results from hop cones have not ye<sup>t</sup> been reported from experiments using foliar sprays. Foliar sprays, including those containing seaweed extracts, usually tend to increase the content of total phenols in plant tissues [17,38–40]. However, some studies have also reported an absence of a significant response to the application of this kind of products [41,42]. Of the cultivars, Nugget showed lower average values of total phenols in comparison to Cascade or Columbus if the two years were taken into account. From the samples selected for phenolic characterization, Nugget presented slightly higher values of total phenolic compounds but, in this case, just the samples with higher phenol content from the first year were characterized.

Previous studies have also shown significant differences in total phenols when different hop cultivars were compared [29,43,44]. The phenols content seems to depend on the cultivar and, in general, low molecular weight phenols are found in greater amounts in aroma cultivars, as the increase in alpha acid content seems to be achieved at the expense of the phenol content [4]. This seems to be true for Cascade, which showed significantly lower levels of alpha acid content, but not for Colombus, which was similar to Nugget, both presenting significantly higher levels of alpha acid content in comparison to Cascade [22]. Overall, the year average values found in this study ranged from 11.9 to 21.2 mg g<sup>−</sup><sup>1</sup> and were of similar magnitude to those reported by Kowalczyk et al. [45], varying between 16.2 and 25.5 mg g<sup>−</sup><sup>1</sup> (water extraction, followed by the Folin–Ciocalteau method). Lower values of 7.12 ± 0.09 mg GAE g<sup>−</sup><sup>1</sup> were reported by Keskin et al. [46] (methanol extraction, followed by the Folin–Ciocalteau method). These results emphasize the potential of the region to grow the cultivars Cascade and Columbus, along with the well-established Nugget.

### *4.2. PCA and Correlation Analysis*

PCA and correlation analysis indicate a significant and negative association between total phenols and Zn concentrations in the cones. The results also indicate a negative influence of Cu, N and Fe in the accumulation of total phenols in the cones. Hop is a species particularly sensitive to Zn deficiency, affecting plant growth and cone production [47]. In this case, an association of Zn with plant vigour was not found, but higher concentrations of Zn, Cu and Fe were previously reported for these plots, and the result associated with poor soil aeration [21].

Enhanced absorption of Zn and Cu was also noticed in industrial hemp (*Cannabis sativa subsp. Sativa*) with higher irrigation level, with Zn showing higher mobility to aerial tissues [48]. The results of correlation analysis also showed significant and positive correlation between cone Cu and Fe, and cone Zn and Cu. Regarding Fe, the high levels previously reported in soil and plants [21], may have contributed to lowering total phenol concentrations. Zn, Fe and Cu do not seem to be important nutrients in phenolic biosynthesis, and they may interfere negatively with other nutrients that provide co-factors for many enzymes of the flavonoid pathway [35].

Regarding N, its supplementation has been negatively associated with the phenolic composition of plant tissues in several crops [49,50], and associated with plant growth particularly in sensitive species to soil N availability [51]. In accordance with the protein competition model (PCM), since phenols and proteins compete for a common precursor, conditions that increase plant growth may reduce the concentration of total phenols [51]. Phenols are secondary metabolites synthesized through the shikimate pathway in which the amino acid phenylalanine is released, and this amino acid is a common precursor of phenylpropanoids and protein synthesis [35,51].

### *4.3. Phenolic Compounds Identification and Quantification*

The phenolic compounds identified were mainly flavonols (quercetin and kaempferol) and phenolic carboxylic acids (*p*-coumaric and caffeic acids), which represent a minor fraction of the polyphenols that can be found in hop cones [5,7]. The result might be due to the in-water extraction method, which while suitable for many applications, is less efficient than the hydroalcoholic extraction method, particularly on hop prenylated flavonoids detection, which are lipophilic compounds [45]. The phenolic profile of *H. lupulus* is in accordance with those previously reported for bracts [7], leaves [52] and cones [53,54] and also for leaves, stem and roots of *H. japonicus* Siebold and Zucc [55]. The identification of peaks 8 ([M-H]− at *m/z* 609), 9 ([M-H]− at *m/z* 463), 11 ([M-H]− at *m/z* 593), and 12 ([M-H]− at *m/z* 447), quercetin-3- *O*-rutinoside, quercetin-3- *O*-hexoside, kaempferol-3- *O*-rutinoside, kaempferol-3- *O*-glucoside, respectively, was performed by comparison of their retention time, UV spectra, and mass fragmentation patterns with the available commercial standards. Three caffeoylquinic acid derivatives were tentatively identified regarding the phenolic acid groups, peaks 1, 4, and 5 (3- *O*-, 4- *O*-, and 5- *O*-caffeoylquinic acids, respectively).

According to Clifford et al. [56,57], peaks 1 and 5 present a major ion MS<sup>2</sup> fragment at *m/z* 191, whereas peak 4 presents at *m/z* 173 an abundance of 100%, indicating the connection 4- *O*- position in the molecule. The organization of the three peaks, besides the major abundant fragments, was performed according to the hierarchical keys developed by Clifford et al. [56,57]. The two 3-*p*-coumaroylquinic acids found (peaks 2 and 3, *cis* and *trans*, respectively) were also assigned using the same hierarchical keys developed by Clifford et al. [56,57] the base peak at *m/z* 163 is for 3-*p*-coumaroylquinic acids. Since both peaks presented the same chromatographic characteristics, they were assigned as *cis* and *trans* isomers. Tanaka et al. [7] have also reported the same phenolic acids in the bracts of hop plants and Choi et al. [55] in the leaves, stem and roots of *Humulus japonicus* Siebold and Zucc. Finally, two *O*-glycosylated quercetin derivatives and two *O*-glycosylated kaempferol derivatives were also tentatively identified in the hop cones, peaks 6 and 10, and peaks 7 and 13, respectively. The tentative identification of these four peaks was performed based on those previously described in *H. lupulus* samples [7,52].

The hop cones of the less vigorous plants of the Nugget cultivar were higher in quercetin and kaempferol, whereas in the hops from the more vigorous plants, the kaempferol flavonoids and caffeic acids were found in small concentrations or were not even detectable. Environmental variables such as light exposure and temperature can significantly influence the accumulation of quercetin and kaempferol compounds in plant tissues [35]. Galieni et al. [58] have also found an increase in caffeic acid and other phenolic compounds in *Latuca sativa* L. grown under drought stress and an increase in cell wall lignification as a tolerance response.

In these experiments, the increased levels of phenolic compounds in less vigorous plants are probably a response to the environmental stress affecting plants' growth. The plants treated with foliar sprays presented slightly lower values of phenolic compounds. Similarly, Xu and Leskovar [42] did not find any effect of applying a seaweed extract on flavonoid content in spinach. Hop cones of plants on limed soil presented a significantly higher concentration of kaempferol-3- *O*-(2-rhamnosyl)-rutinoside and 4- *O*-caffeoylquinic

acid though not significantly. Likewise, Ngadze et al. [59] found an increase in caffeic acid content in potato (*Solanum tuberosum* L.) as a response to Ca applications.

As far as we know, no studies have reported the phenolic composition of hop cones after liming. Cascade stood out from the other cultivars, showing higher concentrations in quercetin and kaempferol compounds and lower in *p*-coumaric acids. Similarly, Almeida et al. [60] reported isoquercitrin followed by quercetin as the major phenolic compounds found in extracts of Cascade hops grown in Brazil. Santagostini et al. [61] identified quercetin-3-*O*-malonylglucoside and kaempferol-3-*O*-malonylglucoside compounds for the first time in Cascade hop. These compounds were also identified for the cultivars used in this study. In agreemen<sup>t</sup> with the present results, other studies [29,43] also showed significant differences in phenolic composition of different cultivars of hop, which probably was due to the potential influence of genetic factors on agronomic and biochemical traits [62].
