**Antioxidant Compounds for the Inhibition of Enzymatic Browning by Polyphenol Oxidases in the Fruiting Body Extract of the Edible Mushroom** *Hericium erinaceus*

## **Seonghun Kim 1,2**


Received: 22 June 2020; Accepted: 16 July 2020; Published: 17 July 2020

**Abstract:** Mushrooms are attractive resources for novel enzymes and bioactive compounds. Nevertheless, mushrooms spontaneously form brown pigments during food processing as well as extraction procedures for functional compounds. In this study, the dark browning pigment in the extract derived from the edible mushroom *Hericium erinaceus* was determined to be caused by the oxidation of endogenous polyphenol compounds by the polyphenol oxidase (PPO) enzyme family. These oxidized pigment compounds were measured quantitatively using a fluorospectrophotometer and, through chelation deactivation and heat inactivation, were confirmed to be enzymatic browning products of reactions by a metalloprotein tyrosinase in the PPO family. Furthermore, a transcript analysis of the identified putative PPO-coding genes in the different growth phases showed that tyrosinase and laccase isoenzymes were highly expressed in the mushroom fruiting body, and these could be potential PPOs involved in the enzymatic browning reaction. A metabolite profiling analysis of two different growth phases also revealed a number of potential enzymatic browning substances that were grouped into amino acids and their derivatives, phenolic compounds, and purine and pyrimidine nucleobases. In addition, these analyses also demonstrated that the mushroom contained a relatively high amount of natural antioxidant compounds that can effectively decrease the browning reaction via PPO-inhibitory mechanisms that inhibit tyrosinase and scavenge free radicals in the fruiting body. Altogether, these results contribute to an understanding of the metabolites and PPO enzymes responsible for the enzymatic browning reaction of *H. erinaceus*.

**Keywords:** enzymatic browning; antioxidant compounds; *Hericium erinaceus*; mushroom metabolites; polyphenol oxidase; tyrosinase; laccase; natural inhibitor

## **1. Introduction**

Fungi, including yeasts, molds, and mushrooms, are attractive resources for biotechnological applications. Of these organisms, many mushroom species have historically been used for food or medicinal purposes. Approximately 140,000 mushroom species belonging to the phyla Ascomycetes and Basidomycetes have been identified [1]. Over 2000 mushroom species are edible, and at least 10% of these species have been used traditionally for food or medicine [1]. Moreover, mushrooms produce small bioactive molecules such as secondary metabolites and polysaccharides that can be used for their medicinal properties in antitumor, immunomodulatory, antibacterial, and antiviral treatments [1–4].

*Hericium erinaceus* (also called lion's mane mushroom, bearded tooth mushroom, satyr's beard, bearded hedgehog mushroom, pom pom mushroom, or bearded tooth fungus) is an edible and medicinal mushroom found in East Asia (China, Korea, and Japan) that belongs to the tooth fungus

group [5]. This mushroom species has a non-typical mushroom shape, with no cap and no stem. It is found on hardwoods and identifiable as a single clump within long dangling spines. The mushroom fruiting body is mainly white, although it becomes brown or yellow with age. In East Asia, *Hericium* species mushrooms are highly valued for their medicinal properties and have been used in traditional Chinese medicine [4]. In the past few decades, there has been considerable research interest in the bioactive properties of this species [5]. Several compounds isolated from *H. erinaceus* have been found to have biological activities, including antitumor, antimicrobial, antioxidant, and cytotoxic activities [5–8].

The mushroom fruiting body is usually white initially, but, when processed for food, the mushroom sometimes develops brown to black pigments. Indeed, in the initial purification process for both proteins and bioactive compounds, the solution extracted from the mushroom lysate is pale yellow, and the extract darkens over time during these processes [9]. The oxidation reaction of polyphenols and other oxidizable substances can potentially be prevented by antioxidant compounds in the mushroom fruiting body. However, oxidizable and aromatic-ring-containing substances in the mushroom fruiting-body extract can be exposed, resulting in their oxidization under aerobic conditions. Similarly to well-known plant substances [10], a number of aromatic compounds, including hericines, erinacines, alkalonides, lactones, and their derivatives [5], can presumably be converted to pigment-forming molecules. These mushroom metabolite substances are possibly oxidized products that are formed by polyphenol oxidases (PPOs) in the mushroom fruiting body. However, there are no reports on these pigment-forming compounds or the potential enzymes in *H. erinaceus*.

PPOs, including tyrosinase and laccase, are enzymes well-known for their activities related to post-harvest browning pigments in agricultural products [11]. Although these pigment compounds, such as certain secondary metabolites, might be useful for enhancing bioactive compounds in fermentation processes or for helping to preserve protein in forage crops, the browning reactions are generally viewed negatively in food processing or in fresh products [10–12]. The dark color reaction occurs via PPOs, which can be induced by wounding or pathogen attacks as part of the defense response in plants and fungi [10,12]. Indeed, it has been shown that PPOs play such a role in the white button mushroom (*Agaricus bisporus*) [13]. Nonetheless, the exact kinds of substances in the mushroom that are converted to brown products by some PPOs are unknown in mushroom species.

In our previous study, products of the dark browning reaction were observed as interfering substances during the purification process for a lectin protein from the fruiting-body lysate of *H. erinaceus*[9]; the chromatography fractions were dark brown with a gradient in the upper part. However, there is a lack of information available about this coloration event in the mushroom fruiting-body extract. Therefore, the aim of this study was to identify the cause of the dark brown pigmentation via the endogenous enzymatic reaction of the PPO family with oxidation activity, and to reduce the occurrence of this pigmentation. Furthermore, the putative PPO-coding genes and their transcripts in both the fruiting body and the mycelium were analyzed as putative proteins involved in the enzymatic browning reaction in the mushroom. In addition, a mushroom metabolite analysis was performed to identify natural compound families in the fruiting body and mycelium of *H. erinaceus* as potential antioxidant substances, or as the browning-reaction substrates of the PPOs.

#### **2. Results and Discussion**

#### *2.1. Observation of Dark Brown Pigment in the Fruiting-Body Lysate*

The fruiting body of the edible mushroom *H. erinaceus* is normally white. Its color becomes yellow or brown over time. In the preparation of the fruiting-body lysate for protein purification, the color of the mushroom crude lysate was initially pale yellow. However, over time, the color of the DEAE-sepharose column flow-through fractions collected during protein purification became dark brown (Figure 1). The dark brown color was observed in Fractions 3 (F3) and 4 (F4). Interestingly, the color pigments localized to the upper part of the test tubes, where the surface was exposed to air, which created a gradient from top to bottom. The yellow color remained at the bottom of the F3 and F4 tubes, but the dark pigmentation progressed over time. This dark brown pigmentation pattern implied that the pigmentation process involves oxygen and certain enzymes or proteins found in F3 and F4.

**Figure 1.** The fruiting body extract fractions (F1–F12) eluted from an ion-exchange column, DEAE-Sepharose, in the edible mushroom *Hericium erinaceus*, with dark brown pigments in F3 and F4.

## *2.2. Measurement of Dark Brown Pigment*

To measure the relative amount of the brown pigment, the absorbance of the fractions was scanned by ultraviolet-visible (UV-vis) spectroscopy in the wavelength range of 210–750 nm. However, the detection wavelength for the pigment was less than 210 nm (data not shown). Therefore, UV-vis spectrophotometry was not useful for measuring the pigmentation, because a mixture of proteins as well as pigment products were present in the fraction (data not shown). The proteins were interfering substances to detect the pigment in the mushroom extract at 210 nm.

An alternative method, fluorescence spectroscopy, was used to detect the brown pigmentation. The fractions with dark brown pigment were scanned. To determine the optimum emission and excitation wavelength for the pigment, the emission wavelength (*Em*) was fixed at 460 nm and the excitation (*Ex*) spectra ranged from 250 to 420 nm was scanned. The highest fluorescence was obtained at 380 nm as the optimal emission wavelength. The excitation wavelength was then fixed at 380 nm, and the emission spectra were scanned from 420 to 700 nm. The maximum intensity signal was detected at the emission wavelength at 444 nm (Figure 2a) setting the excitation wavelength at 380 nm. The emission intensity of each sample corresponded to the relative amount of brown pigmentation, which formed in a time-dependent manner (Figure 2b).

The fluorescence intensity of fraction pools collected was measured at *Ex*380nm~*Em*444nm. All fractions exhibited positive fluorescence intensity readings. The visible intensity of the dark brown pigmentation seemed to correlate with the measured fluorescence intensity for all fractions (Figure 3a). The maximum fluorescence intensity was observed in the pooled crude (CL) fraction (Figure 3b). The fluorescence intensity in the CL fraction was 1.4-fold higher than that in the pooled flow-through (FT) fraction; the CL fraction showed higher fluorescence intensity than the other fractions eluted from the DEAE-sepharose column. The fluorescence intensities of both the CL and FT fractions were 2.7–3.6 times higher than those in the NaCl elution fractions. Although the elution fraction (E250) with 250 mM NaCl was a light yellow color (Figure 3a), the fluorescence intensity of this fraction was lower than that of the CL or FT fractions. However, the E250 fraction showed an approximately 1.2–1.3-fold higher intensity than the other NaCl elution fractions (E50, E500, E1000, and E5000). All of the NaCl elution fractions, except for the E250 fraction, appeared clear, but a fluorescence signal from the dark pigmented compounds was still detected.

**Figure 2.** Fluorescence emission spectra of the dark brown pigment in the protein fraction. (**a**) Emission spectrum of the dark pigment fraction scanned by a fluorospectrophotometer with an excitation wavelength of 380 nm. (**b**) Fluorescence spectra of the pigment fractions for relative quantification of the brown pigment, which increased in a time-dependent manner under the same conditions. The fluorescence intensity of every fraction was measured with the following time points: initial 0 h, purple; 48 h, blue; 72 h, green; 96 h, yellow; 120 h, orange; 144 h, red.

**Figure 3.** Measurement of the browning pigment substances and tyrosinase activities in the fraction pools collected in the DEAE-Sepharose column. (**a**) The protein fraction pools eluted from the DEAE-Sepharose column: 1, crude extract fraction pool (CL); 2, flow-through fraction pool (FT); 2, wash fraction pool (W); 4–8, elution fraction pools with 50 mM (E50), 250 mM (E250), 500 mM (E500), 1000 mM (E1000), and 5000 mM NaCl (E5000), respectively, in 50 mM Tris-HCl buffer. (**b**) Fluorescence intensities (Arbitrary Unit, A.U.) of these fraction pools for the measurement of brown pigment under the optimum conditions with λEx380 nm~λEm460nm. (**c**) Tyrosinase-specific activities of these fraction pools under the enzyme assay conditions described in the material and method section.

The fluorescence spectra of the dark brown pigment suggested that the compounds responsible for the pigmentation might be aromatic or heterocyclic aromatic compounds [14]. Mushrooms contain a considerable number of phenolic compounds, such as flavonoids [5,7,8]. These compounds are antioxidant substances in mushrooms, but are also good substrates for polyphenol oxidases (PPOs) [15]. The mechanisms by which the dark pigment is formed in *H. erinaceus* are not clearly understood, but browning in cultivated button mushrooms (*Agaricus bisporus*), an extensively cultivated mushroom, is induced by bacterial infection and oxidation after harvesting [16]. It has reported that the brown pigments form as a defense mechanism when mushroom tissue becomes damaged or infected [16]. This pigmentation is mainly caused by the oxidation of endogenous phenolic substances, which are enzymatically catalyzed by PPOs, including tyrosinases and laccases.

### *2.3. Measurement of Tyrosinase Activity and its Deactivation in the Mushroom Extract*

To test whether PPOs were responsible for the pigmentation occurring in *H. erinaceus,* the enzymatic activity of the tyrosinase belonging to the PPO family was measured in the fractions eluted from the DEAE-sepharose column (Figure 3c). The CL fraction showed the highest tyrosinase activity per unit protein in all of the fractions tested. In addition, the FT fraction had higher enzyme activity than other fractions. However, tyrosinase activities in the NaCl elution fractions were much lower than those in the CL or FT fractions. The tyrosinase activity values correlated with the fluorescence intensity values in each fraction described above (Figure 3b). These results could suggest that the dark brown pigmentation that occurs in *H. erinaceus* could be due to enzymatic reactions catalyzed by tyrosinases. It seems the CL and FT fractions could contains tyrosinases with natural substrates such as phenolic compounds to convert the brown pigment. However, the NaCl eluted fractions contained low amount of the enzyme as well as less amount of the low molecule substances, which were already pass through the column. Therefore, the intensity of the brown pigment in the elution fractions would lower than the CL and FT fractions.

In addition, to confirm that the dark brown pigment was an enzymatic reaction product, the endogenous tyrosinase in the mushroom lysate was immediately deactivated or inactivated. The chelating agent ethylenediaminetetraacetic acid (EDTA) can eliminate metal ions in a copper-containing metalloprotein to deactivate tyrosinase [15,16]. EDTA was added in concentrations ranging from 1 to 25 mM into the mushroom lysate. At the initial time-point, the fluorescence intensities in EDTA-treated samples (black circles) and untreated samples were slightly different (Figure 4a). There was a slight trend towards decreasing fluorescence intensity, depending on the concentration of EDTA. After 14 days, the fluorescence intensity (white circles) did not increase significantly (less than 10%). With the chelation agent added to the lysate, tyrosinase activity was strongly inhibited (Figure 4b). Even low concentrations (1 mM) of EDTA abolished the tyrosinase activity by up to 90%.

For the inactivation of tyrosinase activity, the mushroom lysate was treated at 100 ◦C for 1 to 30 min. The fluorescence intensity and the tyrosinase activity in the heat-treated lysates were monitored. The fluorescence intensity (black squares) gradually increased in the heat-treated samples, depending on the incubation time (Figure 4c). The fluorescence intensity of the lysate heated for 30 min increased 1.3-fold compared with that of non-heated samples. The heat treatment may have acted to enhance the browning pigment responsible for the fluorescence intensity in these samples. Heat treatment reportedly changed the composition of polyphenolic compounds in shiitake mushroom extracts [17]. The polyphenolic content of shiitake extracts was enhanced with heating temperature and treatment time. However, after 14 days, the fluorescence intensity (white squares) of the heat-treated lysates had decreased by less than 10–13%. Moreover, heat treatment was effective with respect to tyrosinase activity, which decreased by 58.0% after 1 min, 75.2% after 5 min, and 76.8% after 10 min at 100 ◦C (Figure 4d). The enzymes in the samples treated for 30 min at 100 ◦C were almost completely deactivated, and the residual enzyme activity was 5.5%. However, the enzyme activity in the heat-treated samples was higher than that in EDTA-treated samples. The heat treatment did not completely inactivate the enzyme. In addition, the heat treatment could have influence on increasing spontaneously brown

coloring by the chemical reactions such as sugar-amine reactions in the endogenous substance of the mushroom extract. These results indicate that the brown coloring pigmentation in *H. erinaceus* extracts could be the product of an enzymatic reaction by a tyrosinase belonging to the PPO family of enzymes.

**Figure 4.** Measurement of fluorescence intensities and tyrosinase activities in the mushroom lysate treated with EDTA or heat. (**a**) Fluorescence intensities of the EDTA-treated lysates at the initial time (black circles) and 14 days later (white circles). (**b**) Total tyrosinase activities of EDTA-treated lysates. (**c**) Fluorescence intensities of the heat-treated lysates at the initial time (black squares) and 14 days later (white squares). (**d**) Total enzyme activities of the heat-treated lysates. Fluorescence intensities (Arbitrary Unit, A.U.) and tyrosinase activities of all fractions were measured under the optimum conditions with λE × 380 nm~λEm 460 nm and the enzyme assay conditions described, respectively.

## *2.4. Analysis of the Transcription Levels of PPO-Coding Genes in the Di*ff*erent Growth Phases of the Mushroom*

To identify the correlation of the PPO-coding cDNAs with the dark browning event, the expression levels of the tyrosinase and the laccase genes in the RNA-seq data of the fruiting body and the mycelia were analyzed (Figure 5). The isotranscript genes for tyrosinase and laccase were differentially expressed in the growth phases; among seven putative tyrosinase-coding genes identified from expressed transcripts in the mushroom genome, three of them, *ust*Q, *tyr*1, and *asq*I, were observed in both the fruiting body and mycelium. Interestingly, the expression of *ust*Q was approximately 20-fold higher in the fruiting body than in the mycelium, whereas the expression of the *tyr*1 gene was about 1.1-fold higher. However, the expression of another tyrosinase gene (*asq*I) was 1.4-fold lower in the fruiting body than in the mycelium. The *ust*Q-encoded tyrosinase could be a major PPO enzyme involved in the enzymatic browning reaction in the *H. erinaceus* fruiting body.

**Figure 5.** Transcription levels of tyrosinase- and laccase-coding genes involved in the enzymatic browning reaction in the fruiting body (dark gray) and the mycelium *(gray-white) of H. erinaceus*. (**a**) Log scale for transcription levels expressed as fragments per kilobase of transcript per million (FPKM). The log (FPKM) values of *lac2c* and *lac2d* were less than −0.1. (**b**) Normalized FPKM values relative to a suite of endogenous actin genes in each growth phase.

Thirteen putative laccase-coding genes were identified in the genome, but only 10 transcripts were expressed in the fruiting body and mycelium. These transcripts were differentially regulated in different growth phases (Figure 5). In the fruiting body, five laccase transcripts were upregulated, and four other genes were downregulated. Interestingly, the *lac1*, *lac2a*, and *lac2b* transcripts had significantly higher isotranscript expression in the fruiting body; the transcription levels of *lac1*, *lac2a*, and *lac2b* genes were upregulated 59-, 55-, and 38-fold in the fruiting body compared with the mycelium. Among these more highly expressed transcripts, *lac2* genes were expressed as four isotranscripts (*lac2a, lac3b, lac2c*, and *lac2d*), and the expression levels of all of these transcripts were higher in the fruiting body. On the other hand, five other laccase-coding transcript isoform genes were downregulated in the fruiting body, while these transcripts were highly expressed in the mycelium. Two transcripts, *lac4* and *lac5a*, among the five genes that were upregulated in the mycelium, had about 10-fold higher expression in the mycelium growth stage. The transcripts of *lac3*, *lac5b*, and *lac6* had 1.4-, 2.1-, and 7.7-fold higher expression in the mycelium.

These transcriptional analyses showed that the laccase transcripts were also highly expressed and regulated in the growth stages. Moreover, the laccase isoenzymes translated from the upregulated transcripts could be potential PPOs involved in the enzymatic browning reaction in the fruiting body.

#### *2.5. Analysis of the Main Metabolites in the Mushroom Fruiting Body and Mycelium*

To analyze the potential substances involved in the enzymatic browning reaction or compounds with protective roles in the mushroom fruiting body and mycelium, a metabolomic analysis was performed in two different growth phases using CE-TOF-MS and LC-TOF in two modes for cationic and anionic metabolites with molecular weights ranging from 50 to 1000 *m*/*z*. In the fruiting body, CE-TOF-MS detected 198 metabolites (138 metabolites in cation mode and 60 metabolites in anion mode), and LC-TOF-MS identified 63 metabolites (26 metabolites in positive mode and 37 metabolites in negative mode). On the other hand, in the mycelium, CE-TOF-MS detected 171 metabolites (107 metabolites in cation mode and 64 metabolites in anion mode), and LC-TOF-MS revealed 38 metabolites (12 metabolites in positive mode and 26 metabolites in negative mode).

Among these metabolites, well-known natural compounds with tyrosinase-inhibitory activity and antioxidant activity (Table 1) and low-molecular-weight substances potentially involved in the enzymatic browning reaction (Table 2) were identified, respectively, in the mushroom fruiting body and the mycelium. Interestingly, the antioxidant and enzymatic browning substances presented in a growth-stage-dependent manner. For protection against the endogenous browning reaction in the mushroom, 18 putative antioxidants substances were detected and categorized into tyrosinase inhibitors, antioxidants/free-radical scavengers, and polyamines (Table 1). A large number of metabolites with antioxidant activity were found in the fruiting body compared with in the mycelium (17 vs. 9), with an overlap of eight compounds. Of these antioxidant substances, ascorbic acid, caffeic acid, chlorogenic acid, curcumin, apyridoxal, and quinic acid—natural compounds with tyrosinase-inhibitor activity—were detected in the mushroom fruiting body, while vallinic acid was only found in the mycelium. However, terephalic acid was detected in both mushroom development stages. As tyrosinase inhibitors, the levels of ascorbic acid and quinic acid were significantly higher than those of other compounds in the fruiting body.


**Table 1.** Putative metabolites for the natural polyphenol oxidase inhibitors in *H. erinaceus* fruiting body and mycelium.

<sup>1</sup> Peak area × 100,000 divided by the internal standard peak area of metabolic compounds in the two growth stages (the mushroom fruiting body and mycelium) <sup>2</sup> MS scan range: *m*/*z* 50~1000 <sup>3</sup> MT (migration time) in CE/RT (retention time) in LC.

**Table 2.** Potential substrates in the enzymatic browning reaction in the *H. erinaceus* fruiting body and mycelium.



**Table 2.** *Cont*.

<sup>1</sup> Peak area × 100,000 divided by the internal standard peak area of metabolic compounds in the two growth stages (the mushroom fruiting body and mycelium) <sup>2</sup> MS scan range: *m*/*z* 50~1000 <sup>3</sup> MT (migration time) in CE/RT (retention time) in LC.

Moreover, adenosine, ergosterol, erothioneine, glutathione, and 5-hydroxytryptophan were commonly observed as other antioxidants/free-radical scavengers in both the fruiting body and the mycelium (Table 1). In terms of the differences among these compounds, the amount of erothioneine was 29 times higher, whereas the content of glutathione was 20 times lower in the fruiting body than in the mycelium. These data may indicate that different compounds in different mushroom growth stages are synthesized as antioxidant substances for the inhibition of the enzymatic browning reaction or for the scavenging of free radicals.

On the other hand, for the enzymatic browning products, free amino acids and their derivatives, phenolic compounds, and purine and pyrimidine nucleobases were identified as potential substrates (Table 2); these compounds were less abundant in the fruiting body than in the mycelium (18 vs. 29), with an overlap of 12 compounds. Interestingly, the compounds that were grouped as free amino acids and their derivatives were predominantly found in the fruiting body, while phenolic compounds and purine/pyrimidine nucleobases were mainly detected in the mycelium. Phenylalanine and tyrosine were the main substances in terms of the relative amounts of these compounds. These amino acids could be used as substrates for the enzymatic browning reaction by PPOs, including tyrosinase and laccase. Both phenylalanine and tyrosine are favorable substrates and initial precursors for the dark pigment in the melanin synthesis pathway [18–20]. In addition, purine and pyrimidine compounds that contain a pentose sugar—ribose or deoxyribose—with an amine group in their backbone structures are also known to be potential substrates for browning reactions via the Maillard reaction under high temperatures [21]. Adenosine nucleoside phosphate derivatives—AMP, ADP, and ATP—were the main compounds in the nucleobase family. They could be putative browning substances, since their levels were approximately 10 times higher in the fruiting body than in the mycelium. These metabolite analysis results showed that the fruiting body and the mycelium can biosynthesize endogenous antioxidant compounds to reduce the browning reaction, whereas synthesized metabolites such as aromatic amino acids, nucleobases, and their derivatives might be prospective substrates of PPOs for the enzymatic browning reaction.

In this study, to identify the browning reaction in the mushroom extract through the endogenous enzymatic reaction of the PPO-family enzymes with oxidation, a method was established to quantitatively measure dark brown pigmentation in mushroom fruiting-body lysate using a fluorospectrophotometer. Excitation and emission wavelengths of 380 nm and 444 nm, respectively, were found to be the most efficient for measuring this pigmentation. Compounds that fluoresce between these wavelengths generally have aromatic functional groups, aliphatic or alicyclic carbonyl structures, or highly conjugated double-bond structures with low energy-transition levels [22]. Although both the pH and solvent affected the fluorescence readings in these samples, the wavelengths used in this study quantitatively detected the dark brown pigments. Interestingly, the wavelengths used (λEx380nm ~ λEm460nm) were close to those used to detect cyclized *o*-quinone oxidation products (aminochrome, dopachrome, and furanoquinone), generated from the tyrosinase-catalyzed oxidation of dopamine, L-DOPA, and 3,4-dihydroxyphenylacetic acid [23]. These cyclized *o*-quinone oxidation products are eventually polymerized to melanin [23].

The pigment products were estimated to be oxidized polyphenol compounds similar to melanin, derived from PPOs in the mushroom lysate [24]. PPOs catalyze enzymatic browning by oxidizing phenolic compounds to their respective *o*-quinones, which subsequently undergo nonenzymatic oxidation to brown melanin pigments or participate in addition–polymerization reactions with protein functional groups to form cross-linked polymers [25]. This phenomenon is a major problem in the processing of both fruits and vegetables [26]. Moreover, both tyrosinase and laccase are well-known oxidase enzymes that cause brown pigmentation in mushroom species [15,16].

Tyrosinases (monophenol, *o*-diphenol:oxygen oxidoreductase, EC 1.14.18.1) are type-3 copper proteins that are involved in the initial step of melanin synthesis [15]. These enzymes catalyze both the *o*-hydroxylation of monophenols and the subsequent oxidation of the resulting *o*-diphenols into reactive *o*-quinones, which evolve spontaneously to produce intermediates that are associated with dark brown pigments [15]. In mushrooms, tyrosinases are endogenous biological factors that are associated with the formation and stability of spores in defense and virulence mechanisms, and in browning and pigmentations [27–29]. Although the catalytic mechanisms of these enzymes are still unknown, the deactivation and inactivation of the metalloprotein tyrosinases in the mushroom lysate seen in this study may be a way to prevent browning, because neither the EDTA-treated nor heat-treated lysate became pigmented.

The other enzyme type, laccase (EC 1.10.3.2), is a multicopper oxidase that is capable of oxidizing a wide range of aromatic compounds [30]. These enzymes also catalyze the oxidation of *o*- and *p*-diphenols into their corresponding quinones for the enzymatic browning reaction. Although the roles of plant laccase in the radical coupling of monolignols to form lignin and flavonoid polymerization in the cell wall [31] and in the protection of plants against environmental stresses [32,33] are well-understood, the in vivo biological functions of mushroom enzymes belonging to the PPO family are still unclear. Nevertheless, the broad substrate specificities of laccases make them enzymes with potential involvement in the pigment formation via oxidation of endogenous aromatic compounds in mushrooms [34].

According to the expression pattern in the fruiting body and mycelium growth phases, a transcription analysis was performed using RNA-seq sequencing, which showed that the number of laccase transcripts was more diverse than tyrosinase transcripts, and the expression levels of certain laccase genes, namely lac1, lac2a, and lac2b, were significantly increased in the fruiting body compared with levels in the mycelium (Figure 5). As PPOs, these laccases would be abundant in the tissue of the mushroom; furthermore, these enzymes could be the predominant enzymes that form the dark pigments in the fruiting-body lysate. Nevertheless, the color of the fresh mushroom fruiting body did not change in spite of these highly expressed proteins. The enzyme activities of PPOs, including tyrosinases and laccases, could be strongly inhibited by the endogenous antioxidant inhibitory compounds in the mushroom metabolites. However, in chromatography, PPO enzymes with high molecular weights in the mushroom lysate were separated into low-molecular-weight inhibitors

of metabolites when spontaneously exposed to oxygen; thus, these enzymes could be activated for the dark browning reaction.

The profile analysis of the mushroom metabolites in the different development phases revealed putative antioxidant compounds (Table 1) as well as enzymatic browning substrates (Table 2). Interestingly, some phenolic compounds, such as caffeic acid and chlorogenic acid, could be potential antioxidants as well as substrates of PPOs in the browning reaction in the mushroom fruiting-body extract. Adenine, a purine nucleobase, was also found to be a bilateral substrate in the mushroom. However, there were few data about how these compounds function as antioxidants or browning-reaction substances under certain conditions. Interestingly, contents of the endogenous tyrosinase inhibitors ascorbic acid and quinic acid were high only in the fruiting body (Table 1). Of these substances, ascorbic acid is already a well-known compound and additive that effectively prevents browning due to tyrosinase in *A. bisporus* [35]. Quinic acid is known to be an inducer of the antioxidant mechanism that increases urinary excretion of tryptophan and nicotinamide in the mammalian system [36], but it is also unknown whether this compound might play a similar role in mammals. Other antioxidant substances, such as ergosterol and ergothioneine, with the ability to scavenge free radicals and reactive oxygen species [37] as well as to chelate divalent metal ions [38], could inhibit the enzymatic browning reaction of PPOs in the mushroom fruiting body [39,40].

Indeed, these antioxidant compounds from mushrooms are already known for their significant bioactivity [19,20,41,42]. Antioxidant metabolites, mainly ascorbic acid, ergosterol, and ergothioneine, in the *H. erinaceus* fruiting body could prevent the oxidation of polyphenolic compounds and serve as anti-browning substances. Moreover, these functional compounds are potentially useful nutrient components themselves, as a food source to help the organism decrease oxidative stress [5,43]. Other potential applications of these active mushroom compounds include acting as natural ingredients for cosmetics because of their antioxidant, anti-tyrosinase, and anti-collagenase activities, among others [42,44]. Mushroom-based ingredients for skincare or anti-aging to fight against UV radiation and free radicals are already commercially available in cosmetic markets [42]. Mushroom metabolites identified as tyrosinase inhibitors, antioxidants/free-radical-scavenging compounds, and polyamines could be potential nutrient substances as well as cosmetic ingredients for nutricosmetics, combining foods, cosmetics, and pharmaceuticals [42–44].

## **3. Material and Methods**

## *3.1. Mushroom Strains and Culture Conditions*

*Hericium erinaceus* mushrooms were obtained from a local mushroom farm (Seojong, Korea). *H. erinaceus* mycelia isolated from the mushroom fruiting body were molecularly identified as NEU-2L and cultivated in potato dextrose broth (Difco, Detroid, MI, USA) at 20 ◦C for 20 days. The fruiting bodies and mycelia were collected, frozen immediately in liquid nitrogen, and then stored at −80 ◦C for total RNA extraction and preparation of the mushroom lysate.

### *3.2. Preparation of the Mushroom Fruiting-Body Lysates*

The fruiting bodies were ground into a fine powder with a mortar and pestle while being cooled by liquid nitrogen. The powder was then suspended in 10 mM Tris-HCl buffer (pH 8.0), which contained 1 mM phenylmethylsulfonyl fluoride (PMSF) and a cOmplete™ EDTA-free protease inhibitor cocktail (Roche, Basel, Switzerland), overnight for extraction. The homogenate was centrifuged at 10,000×*g* for 30 min, and the resultant supernatant was centrifuged again at 20,000× *g* for 60 min. The clarified solution was used for ion-exchange chromatography.

### *3.3. Fractionation of the Mushroom Fruiting-Body Lysate Using Ion-Exchange Chromatography*

The clarified crude lysate was applied to the purification columns. The crude protein solution was applied to a diethylaminoethanol (DEAE)-sepharose column (2.6 × 20 cm) (GE Healthcare, Chocago, IL, USA) equilibrated with 10 mM Tris-HCl buffer (pH 8.0) containing 1 mM PMSF and a protease-inhibitor cocktail. After the unbound proteins were washed with five column volumes of buffer, the protein fractions were eluted with three bed volumes of the buffer, followed by either a linear gradient of 0.0–1.0 M NaCl or a step gradient of 0.25 M, 0.5 M, 1.0 M, and 2.5 M NaCl. Fractions (7 mL) were collected at a flow rate of 1 mL/min. The fractions were stored at 4 ◦C.

## *3.4. Browning Pigment Analysis*

For ultraviolet-visible (UV-vis) spectroscopy analysis, the protein purification fractions (100 μL) collected in the DEAE-sepharose column were transferred to a 96 well polystyrene microtiter plate (Greiner Bio-one, Monroe, LA, USA). The microtiter plate was scanned at room temperature by a SynergyTM H1 Multi-Mode Reader (BioTek, Winooski, VT, USA) microplate reader, using wavelengths ranging from 210 to 750 nm.

For the fluorescence spectroscopy analysis, the protein purification fractions (100 μL) were transferred to a 96 well black microtiter plate (Greiner Bio-one, Monroe, LA, USA). The microtiter plate was also scanned by a SynergyTM H1 Multi-Mode Reader microplate reader. For identification of the optimum excitation wavelength, the emission wavelength (Em) was fixed at 460 nm, and then excitation wavelengths (Ex) ranging from 250 to 420 nm were scanned by the microplate reader. For identification of the optimum Em wavelength, the Ex wavelength was fixed at 380 nm, and then Em wavelengths ranging from 450 to 650 nm were scanned by the microplate reader. Finally, all of the fractions with dark color pigment were measured at the optimum Ex and Em wavelength conditions (Ex380nm~Em444nm).

## *3.5. RNA Sequencing of the Mushroom Fruiting Body*

RNA sequencing with total RNA extracted from the fruiting bodies and mycelia was performed as described previously [45]. The procedure is described here briefly. Total RNA were extracted from the mushroom mycelia and fruiting bodies using the RNeasy Plant Mini Kit (Qiagen, Hilden, Germany) according to the manufacturer's protocol, and then the extracted RNAs were further treated with DNase I. The DNase-I-treated RNA samples were purified using an RNeasy column (Qiagen, Hilden, Germany). The RNA quality and quantity were analyzed by UV spectrophotometry and gel electrophoresis. The cDNA library was constructed using a TruSeq RNA sample prep kit v2 (Illumina, San Diego, CA, USA), and RNA sequencing with an Illumine HiSeq 4000 sequencing system (Illumina, San Diego, CA, USA) was performed by MACROGEN (Seoul, Korea).

## *3.6. Tyrosinase Activity Assay*

Among polyphenol oxidases, tyrosinase activity with 4-*tert*-butylphenol (Sigma-Aldrich, St. Louis, MI, USA) as a substrate was determined by measuring the amount of *t*-butyl-quinone oxidized by the enzyme. The enzyme activity was assayed in a total volume of 100 μL with 1 mM substrate in a 96 well microtiter plate at 30 ◦C for 30 min. After the enzymatic reaction, the oxidized monophenol product was measured on a SynergyTM H1 Multi-Mode Reader microplate reader at 400 nm. One unit of tyrosinase was defined as the amount of the enzyme required to produce 1 μmol of *t*-butyl-quinone (ε400 nm = 1150 M−1cm−1) per min under the reaction conditions. The background oxidation of the substrate was deduced by using a reference sample with an identical composition to the reaction mixture without the enzyme.

## *3.7. Deactivation or Inactivation of the Polyphenol Oxidases' Activities in the Mushroom Lysate*

For deactivation of the tyrosinase activities, the mushroom lysate was treated by heating at 100 ◦C. After heat treatment, residual enzyme activities were assayed under standard conditions after incubating the mushroom lysate for the appropriate time.

For the inactivation of enzyme activities, the lysate was treated with a chelating agent, ethylenediaminetetraacetic acid (EDTA), at concentrations ranging from 1 to 25 mM. The appropriate

concentration of EDTA was added to the lysate, and then the sample was incubated at room temperature for 1 h. After incubation, residual enzyme activities were measured using the standard assay described above. All enzyme activities were determined in triplicate.

## *3.8. Metabolite Extracts and Analysis in the Mushroom Fruiting Body and Mycelium*

Metabolome measurements were carried out by Human Metabolome Technology (HMT) Inc. (Tsuruoka, Japan). For capillary electrophoresis time-of-flight mass spectrometry (CE-TOF-MS) analysis, the frozen hydrate sample (approximately 30 mg of each sample) was mixed with 500 μL of methanol containing internal standards (50 μM) and homogenized using a homogenizer (1500 rpm, 120 s × 2 times). Subsequently, 500 μL of ultrapure water was added to the homogenates, mixed thoroughly, and centrifuged at 2300× *g* for 5 min at 4 ◦C. The clarified solution was filtrated through a 5 kDa cut-off filter (ULTRAFREE-MC-PLHCC, HMT Inc., Yamagata, Japan) to remove macromolecules. The filtrate was centrifugally concentrated and resuspended in Milli-Q ultrapure water for CE-TOF-MS analysis at HMT. CE-TOF-MS was carried out using an Agilent CE Capillary Electrophoresis system (Agilent Technologies, Waldbronn, Germany) equipped with an pump an Agilent 6210 time of flight mass spectrometry, an Agilent 1100 isocratic HPLC pump, an Agilent G1603A CE-MS adapter kit, and an Agilent G1607A CE-ESI-MS spray kit. System were controlled using Agilent G2201AA ChemStation software version B.03.01 for CE. Metabolites were analyzed by using a fused silica capillary (50 μm × 80 cm), with commercial electrophoresis buffer (Solution ID: H3301-1001 for cation and I3302-1023 for anion analysis; HMT. Inc.) as the electrolyte. Samples were injected at 50 mbar for 10 s for cation analysis and 22 s for anion analysis. The spectra were scanned from *m*/*z* 50 to 1000.

For liquid chromatography time-of-flight/mass spectrometry (LC-TOF-MS) analysis, the frozen hydrate sample was mixed with 500 μL of 1% formic acid in acetonitrile (v/v) containing internal standards (10 μM) and homogenized using a homogenizer. The mixture was homogenized once more after adding 167 μL of Milli-Q ultrapure water, and then centrifuged at 2300× *g* for 5 min at 4 ◦C. After the supernatant was collected, an additional 500 μL of 1% (v/v) formic acid in acetonitrile and 167 μL of ultrapure water were added to the precipitation. The homogenization and centrifugation were performed as described above to extract the residual compounds in the precipitant. The supernatant was combined with the previously collected sample. The mixed supernatant was filtrated through a 3 kDa cut-off filter to remove proteins and filtrated through a solid-phase extraction column (Hybrid SPE phospholipid 55261-U Supeleo, Bellefonte, PA, USA) to remove phospholipids. The filtrate was desiccated and resuspended in 20 μL of 50% (v/v) isopropanol in ultrapure water for LC-TOF-MS at HMT, as described above. LC-TOF-MS was carried out using an Agilent 1200 series RRLc systems SL (Agilent Technologies, Waldbronn, Germany) coupled with an Agilent-6224 TOF mass spectrometer equipped with an electrospray interface. Chromatographic analysis was performed on an Agilent Zorbax ODS column (2 mm × 50 mm, 2 μm). The autosampler and column were maintained at 4 and 40 ◦C, respectively. The mobile phase was water with 0.1% acetic acid (A) and water/ isopropanol/ acetonitrile (5/65/30) with 0.1% acetic acid and 2 mM ammonium acetate (B). The elution gradient was optimized as follows: 0–0.5 min, 1% B; 0.5–13.5 min, 100% B; 13.5–20 min, 100%, which was delivered at 0.3 mL min−1. Re-equilibration duration was 7.5 min between individual runs. The optimized conditions of TOF/MS were: capillary voltage, 4.0 kV/−3.5 kV; drying gas (N2) temperature, 350 ◦C; drying gas (N2) flow rate, 10.0 L min−1; Oct RFV, 750 V; skimmer voltage, 65 V; fragmentor voltage, 175 V; nebulizer pressure, 40 psi; scan range, *m*/*z* 100–1700.

For analysis of scanned metabolites, peaks detected in CE-TOF-MS and LC-TOF-MS were extracted using the automatic integration software MasterHands software version 2.17.1.11 (Keio University, Tsuruoka, Japan). Signal peaks corresponding to isotopomers, adduct ions, and other product ions of known metabolites were excluded, and the remaining peaks were annotated with putative metabolites from the HMT metabolite database on the basis of their migration times (MTs) in CE-TOF-MS or retention times (RTs) in LC-TOF-MS and *m*/*z* values determined by TOF-MS. In addition, peak areas

were normalized against those of the internal standards, and the resultant relative area values were further normalized by the sample amount.

#### **4. Conclusions**

The dark browning pigment in the extract of the edible mushroom *H. erinaceus* was identified as potential byproducts of oxidization by endogenous polyphenol oxidases, including tyrosinases and laccases, measurable by fluorescence spectroscopy with 380 nm and 444 nm excitation and emission wavelengths, respectively.

Moreover, chelation deactivation and heat inactivation clearly showed that a tyrosinase—a metalloprotein as well as a PPO-family enzyme—affected the enzymatic browning reaction in the mushroom lysate. Furthermore, the identification of the putative PPO-coding genes and the analysis of their transcripts in the different growth phases suggested that the highly expressed tyrosinase and laccase isoenzymes in the mushroom fruiting body could be potential PPOs involved in the enzymatic browning reaction. The metabolite analysis of these growth statuses of the mushroom presented a number of prospective compounds as enzymatic browning substances, and they were grouped into amino acids and their derivatives, phenolic compounds, and purine and pyrimidine nucleobases. Nevertheless, the mushroom contained relatively high amounts of natural antioxidant compounds for the inhibition of tyrosinase and the scavenging of free radicals. These endogenous antioxidant substances could effectively decrease the browning reaction via PPO-inhibitory mechanisms in the mushroom fruiting body. With further evaluation of their health-promoting effects, these antioxidant metabolites and related compounds in *H. erinaceus* could be used to develop nutrient substances for healthcare and cosmetic ingredients for nutricosmetics.

**Funding:** This work was supported by Cooperative Research Program for Agriculture Science & Technology Development (PJ009783) from Rural Development Administration (RDA), the bilateral collaboration program (NRF-2018K1A3A1A39071061) between Korea and Slovakia through the National Research Foundation of Korea (NRF) and partially by KRIBB Research Initiative Program grant.

**Conflicts of Interest:** The author declares that there is no conflict of interests regarding the publication of this paper.

#### **References**


studies using cultured cancer cells, tumor-bearing animals, and clinical trials. *Appl. Microbiol. Biotechnol.* **2020**, *104*, 4675–4703. [CrossRef]


© 2020 by the author. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article* **Impact of Stability of Enriched Oil with Phenolic Extract from Olive Mill Wastewaters**

## **Rosa Romeo, Alessandra De Bruno, Valeria Imeneo, Amalia Piscopo \* and Marco Poiana**

Department of AGRARIA, University Mediterranea of Reggio Calabria, Vito, 89124 Reggio Calabria, Italy; rosa.romeo@unirc.it (R.R.); alessandra.debruno@unirc.it (A.D.B.); valeria.imeneo@unirc.it (V.I.); mpoiana@unirc.it (M.P.)

**\*** Correspondence: amalia.piscopo@unirc.it

Received: 9 June 2020; Accepted: 26 June 2020; Published: 30 June 2020

**Abstract:** The disposal of olive mill wastewaters is a considerable subject for the development of a sustainable olive oil industry considering their high content of pollutants. Nevertheless, the selective extraction of phenolic compounds from olive mill wastewaters represents a promising approach to obtain phenolics suitable for food enrichment. This work aimed to evaluate the efficiency of phenolic extract addition (50 mg L<sup>−</sup>1), used as natural antioxidant, in sunflower oil against oxidative deterioration; to this aim, XAD-7-HP resin was tested in the recovery of phenolic compounds from olive mill wastewaters. Ultra-high performance liquid chromatography was used to evaluate the single phenols contained in the extract; the most consistent amount was detected for hydroxytyrosol (834 mg 100 mL−1). The change in the oxidation state of fortified sunflower oil was studied by measuring physicochemical (refractive index, peroxide value and oxidative resistance to degradation) and antioxidant parameters (DPPH, ABTS and ORAC assays) during 90 days of storage. Results showed an enhancement of oxidative stability of 50% in the fortified oil compared to control.

**Keywords:** macroporous resin XAD-7HP; olive mill wastewaters; oxidative stability; polyphenols; sunflower oil

## **1. Introduction**

Olive oil production represents a very important sector for Mediterranean countries. Despite its economic importance, olive oil production is associated with some negative implications on the environment as contamination of soil, water body pollution, underground seepage and air emissions due to the large amount of waste generated [1,2]. Olive mass is composed of about 80% of olive pulp and stones, whereas the liquid and solid waste yield is greater than the oil after the production steps [3]. Specifically, the amount of produced olive mill wastewaters (OMWWs) varies from 0.3 to 1.1 m<sup>3</sup> for tons of processed olives, depending on the olive oil extraction system [4,5]. OMWW disposal is a serious problem for the development of a sustainable olive oil industry owing to the high content of pollutants as heavy metals, a considerable amount of suspended solid and organic compounds. As a result of their high water solubility, the polyphenol concentration in OMWW ranged from 5 to 25 g L−<sup>1</sup> [6,7]. These compounds show strong antioxidant activity, principally based on their ability to transfer the hydrogen atom of phenolic hydroxyl group to the free radicals. Phenolic compounds have potential beneficial effects with their anti-inflammatory and antimicrobial properties [8–10].

However, the selective recovery of phenolic substances from industrial wastes, such as OMWW, represents a valid approach for the reduction of their environmental toxicity and an opportunity to obtain high added value molecules [11,12]. Several methods for the recovery of OMWW polyphenols have been investigated, such as solvent extraction, ultrasound treatment, supercritical fluid extractions, membrane, chromatographic separations and Amberlite XAD resins, with the final aim to maximize the phenolic yield without impurities and toxic substances, and

after to use these in the food industry [8,13–15]. Literature is available on the application of solid extraction and purification methods with the use of polymeric resin. A variety of synthetic polymer adsorbents are nowadays available for OMWW treatment, such as polystyrenedivinylbenzene copolymers and divinylbenzeneethylvinylbenzene as acrylic ester-based Amberlite XAD7 and the styrenedivinylbenzene-based XAD16 resin copolymers [16–19]. The adsorption process is characterized by a simple design and operation, low operating costs, reduction of the amount of used solvent, reduction of the operation time and separation of large amounts of bioactive compounds [14,20,21]. Moreover, the relatively inexpensive resins are durable, chemically stable and safe. In this work, Amberlite-XAD7HP was used as mean for the obtainment of a natural phenolic extract (WE) suitable for oil enrichment. XAD7HP resin exhibits a suitable pore diameter, polarity and surface area to recovery low molecular weight phenols using solvents with low toxicity and safe for human use. In addition, the potential to be reused with a low loss of absorbing capacity over time makes it useful for industrial application.

Considering the OMWW dotation of phenols with recognised antioxidant activity and health benefits, in the last few years there was an increasing interest in their recovery to re-use these as natural antioxidants. The enrichment with natural phenolic extracts of foods, beverages, edible oils, etc. (that contain a low concentration of phenolic compounds), could be a viable alternative to obtain a healthy added-value product. Sunflower oil with higher amounts of unsaturated fatty acids, in particular polyunsaturated ones, is more exposed to oxidative rancidity or autoxidation. This process reduces the nutritional and sensorial qualities and also gives undesirable chemical products, such as organic acids, aldehydes and ketones, which are harmful for human health [22]. Chemical stability depends on the chemical constituents of the oil, like antioxidants and oxidizable components [23]. Sunflower oil is usually subjected to a refining process that involves the complete loss of valuable active components with interesting nutritional-, health- and antioxidant-related characteristics [24]. Although synthetic antioxidants initially have been used with the purpose of retarding the oxidative deterioration, their implications in some diseases, such as cancer and carcinogenesis, are still controversial [25]. Thus, the tendency to use natural antioxidants has been increasing in order to increase the food quality and satisfy the demand of consumers for healthy products [26].

The aim of this study was to evaluate the efficiency of the phenolic extract obtained from OMWW with a solid extraction method against sunflower oil oxidative deterioration.

## **2. Materials and Methods**

#### *2.1. Chemicals and Reagents*

Standards of phenolic compounds gallic acid (99%), vanillic acid (97%), tyrosol (97%), ferulic acid (99%), *p-*coumaric acid (98%) and *o*-cumaric acid (98%) were purchased from Fluka (Steinheim, Germany). Caffeic acid (98%), Apigenin (99%), luteolin (99%), and oleuropein (99%) were purchased from Extrasynthèse (Z. I. Lyon Nord, France). Hydroxytyrosol ((3,4-dihydroxyphenyl) ethanol) was acquired from TCI (Saitama, Japan). Verbascoside (99%) was procured from Sigma-Aldrich (St. Louis, MO, USA). The solvents used for chromatographic analysis (methanol, water, and acetonitrile) were ultra-high performance liquid chromatography (UHPLC)-MS grade (Carlo Erba, Milan, Italy). 2,2 -Azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS), 2,2-diphenyl-1-picrylhydrazyl (DPPH), Folin–Ciocalteu's phenol reagent and Trolox were purchased from Sigma Chemical Co. (St. Louis, MO, USA). AAPH (2,2 -azobis (2-amidino-propane) dihydrochloride) and fluorescein sodium were purchased from Acros Organics (New Jersey, USA) and Panreac (Barcelona, Spain), respectively. The reagents used for chemical investigation (ethanol absolute anhydrous, chloroform, isooctane, acetic acid, Diethyl ether, n-Hexane, sodium hydroxide solution, sodium thiosulfate solution, acetic acid glacial) were purchased from Carlo Erba (Milan, Italy); potassium iodide was acquired from Honeywell Fluka (Steinheim, Germany), Amberlite® XAD-7HP 20-60 mesh from Sigma-Aldrich (St. Louis, MO, USA) and Lecithin from Carlo Erba (Milan, Italy).

## *2.2. Sample Collection*

Olive Mill Wastewater (OMWW) is a secondary product of the olive oil extraction process containing soft tissues of the olive fruit and the water used in the various stages of the oil extraction treatment together with the water contained in the fruit. OMWW were obtained by mean of a three-phase centrifugation process from Ottobratica olive cultivar during the crop seasons 2019 and supplied by an olive oil mill located in the Calabrian region (Italy). Sunflower oil used for phenolic enrichment was purchased in a local market.

## *2.3. OMWW Extraction*

In order to obtain an extract enriched of phenolic compounds, OMWW were processed with Amberlite XAD-7-HP resin following the literature [27] with some modifications. Amberlite XAD-7-HP resin have a large surface with a macroreticular structure that allows to recover a mixture of different sizes of polyphenols. Moreover its design and operation are simple, operating costs low and the resin regeneration is easy. Before performing the extraction procedure, the adsorbent was pre-conditioned with NaOH 0.1 N for 2 h, rinsed with distilled water, immersed in HCl 0.1 N for 2 h and finally washed with distilled water.

For the extraction, 50 g of OMWW sample were mixed with 10 g of resin under stirring for 20 min. The adsorbent was washed three times with water; successively, it was eluted by a mean of three volumes of 50 mL of EtOH. The combined ethanol extract (WE) was desolvented in a rotary vacuum at 25 °C, then the dried residue was recovered with 100 mL of water, filtered using a 0.45-μm PTFE (Ø 15 mm) syringe filter and stored at 4 °C for the successive analytical determinations.

## *2.4. Production of Enriched Sunflower Oil*

Enriched sunflower oil (MBoil) was produced in a Food Technologies laboratory of the Mediterranean University of Reggio Calabria (Italy) following the literature [28] with some modifications. An aliquot of WE and lecithin were added to sunflower oil and mixed for five hours until complete homogenization, in order to obtain oil samples enriched with a final concentration of 50 mg L−<sup>1</sup> of hydroxytyrosol. Sunflower oil samples were used as control. Oil samples were kept in dark glass bottles (150 mL) at 10 and at 25 ◦C (three independent replicates for each thesis and time) and periodically analysed at different times (0, 15, 45 and 90 days of storage).

## Extraction of Antioxidant Compounds

Phenols of MBoil were obtained by liquid–liquid extraction using methanol and according to the method [29] opportunely modified. Five grams of oil were added with 2 mL of methanol:water (70:30) and 2 mL of hexane and mixed with a Vortex for 10 min. The hydro-alcoholic phase was separated from the oil phase in a refrigerated (NF 1200R) centrifuge apparatus (Nüve, Ankara, Turkey) ` at 5000 rpm, 4 ◦C for 10 min. Hydro-alcoholic extracts (WE) were recovered with a syringe, filtered through a 0.45-μm nylon filter, diameter 15 mm (Thermo Fischer Scientific, Waltham, MA, USA) and utilised for the phenolic compounds quantification and antioxidant activity.

## *2.5. Determination of Total Phenol Content and Evaluation of Antioxidant Activity*

Total Phenol content (TPC) of WE was determined in accordance to referred method [12] with minor modifications. Briefly, an aliquot of diluted WE was placed in a volumetric flask and mixed with deionized water (20 mL) and Folin–Ciocalteau reagent (0.625 mL). Then, 2.5 mL of saturated solution of Na2CO3 (20%) were then added after 3 min and made up to the 25 mL with deionized water. Thereafter, the mixture was left to react for 12 h in the dark and at room temperature. Sample absorbance was measured at 725 nm using a double-beam ultraviolet-visible spectrophotometer (8453 UV–Vis, Agilent, Waldbronn, Germany).

For the determination of TPC of MBoil, 0.05 mL of WE were mixed with 0.300 mL of Folin reagent and 0.25 mL of deionised water and, after 4 min, with 2.4 mL of an aqueous solution of Na2CO3 (5%). The mixture was maintained in a 40 ◦C water bath for 20 min and TPC was determined at 750 nm. Quantification was performed by mean of a calibration curve obtained at gallic acid concentrations from 1 to 10 mg L<sup>−</sup>1. The results were expressed as mg of gallic acid equivalent 100 mL−<sup>1</sup> of WE and mg gallic acid 100 g−<sup>1</sup> of oil.

The evaluation of antioxidant capacity of WE and MBoil was performed by DPPH and ABTS assays [30,31]. An aliquot of diluted WE (1:50) was mixed with DPPH solution (6 <sup>×</sup> 10−<sup>5</sup> mM) to the final volume of 3 mL and left in the dark for 30 min. The absorbance decrement was measured against methanol at 515 nm using a spectrophotometer (8453 UV–Vis, Agilent, Waldbronn, Germany) at 20 ◦C. The radical scavenging activity was expressed as mmol Trolox 100 mL−<sup>1</sup> of WE and μmol Trolox 100 g−<sup>1</sup> sample of oil. Then, 7 mM ABTS and 2.4 mM potassium persulphate (K2S2O8) solutions were mixed for the ABTS assay and placed at room temperature for 12 h in the dark for stabilization. The resulting ABTS·+ solution was diluted with ethanol to obtain a blue-green chromogen with an absorbance of 0.70 (±0.02) at 734 nm. Then, 10 μL of diluted sample were added to the radical solution up to 3 mL and after 6 min the absorbance was measured. The quenching of initial absorbance was plotted against Trolox concentration (from 1.5 to 24 μM) and the results were expressed as TEAC values (mmol Trolox 100 mL−<sup>1</sup> of WE and μmol Trolox 100 g−<sup>1</sup> of oil).

In addition, the antioxidant activity of MBoil samples was also analysed by ORAC assay according to our previous study [32]. The ORAC assay was carried out on VICTOR X2 2030 Multilabel Plate Readers (PerkinElmer, Boston, Massachusetts, USA) in 96-well black microplate (PerkinElmer, Boston, Massachusetts, USA) using a fluorescence filter with an excitation wavelength of 485 nm and emission wavelength of 520 nm. The mix for reaction consisted of 130 μL of fluorescein solution, 50 μL of AAPH solution and 20 μL of phenolic extract. The fluorescence was measured at 37 ◦C immediately after the addition of fluorescein (time 0) and measurements of fluorescence kinetic were taken every minute for 30 times until the relative fluorescence intensity was less than 5% of the initial value. Results were expressed as μmol Trolox 100g−<sup>1</sup> of oil.

## Identification and Quantification of Phenolic Compounds

Identification and determination of the principal bioactive phenolic compounds of WE and MBoil were performed by UHPLC in accordance with [32]. The UHPLC system consisted of an UHPLC PLATINblue (Knauer, Berlin, Germany) equipped with a binary pump system using a Knauer blue orchid column C18 (1.8 μm, 100 × 2 mm) coupled with a PDA–1 (Photo Diode Array Detector) PLATINblue (Knauer, Berlin, Germany). The Clarity 6.2 software was used. Before the injection, phenolic compounds of MBoil were extracted using a variation of method [33]. One millilitre of oil was extracted with 1 mL of a methanol:water (80:20, *v*:*v*) in 2 mL Eppendorf reaction tubes. The mixture was shaken vigorously for 1 min using a vortex and then centrifuged (Micro Centrifuge Model 1K15 SIGMA, Laborzentrifugen, Osterode am Harz, Germany) at 13,000 rpm for 10 min at 10 ◦C. The methanolic phase was filtered with 0.22-μm nylon syringe filters, diameter 13 mm (Thermo Fischer Scientific, Waltham, MA, USA). The flow rate was 0.4 mL min−<sup>1</sup> and the injection volume 5 μL. Acidified water (pH 3.10) (A) and acetonitrile (B) were the mobile phases and the applied gradient was the following: 95% A and 5% B (0–3 min), 95–60% A and 5–40% B (3–15 min); 60–0% A and 40–100% B (15–15.5 min). Quantification was performed by external standards (1–100 mg L−1) and results expressed as mg kg−<sup>1</sup> of sample.

## *2.6. Measurement of Chemical and Physical Properties of Oils*

Free acidity (% oleic acid), peroxide value (mEq O2 kg−1) analyses and extinctions parameters K232 and K270 were performed according to Official and standard methods [34–36].

The moisture of samples was tested in an Electronic Moisture Analyser MA37 (Sartorius, Goettingen, Germany). The analysis was performed using 5 g of sample at 105 ◦C. The results were expressed as percentages.

## Sunflower Oil Oxidative Stability in Accelerated Storage Test

OXITEST Oxidation Test Reactor (VELP Scientifica, Usmate Velate, MB, Italy) was used in order to evaluate the opposition to fat oxidation. This method is recognized by AOCS International Standard Procedure (Cd 12c–16) for the determination of oxidation stability of food, fats and oils [37]. The analysis consists of monitoring the oxygen uptake of the reactive constituent of food samples to determine the oxidative stability under conditions of accelerated oxidation. Briefly, 5 g of oil sample were distributed homogenously in a hermetically sealed titanium chamber; oxygen was purged into chamber up to a pressure of 6 bar. The reactor temperature was set at 90 °C. These reaction working conditions allow obtaining the sample Induction Period (IP) within a short time. The OXITEST allows to measure the modification of absolute pressure inside the two chambers and, through the OXISoft™ Software (Version 10002948 Usmate Velate, MB, Italy), automatically generates the IP expressed as hours by the graphical method.

## *2.7. Statistical Analysis*

Results of the present study were expressed as mean ± *SD* of three measurements (*n* = 3). Multivariate and One-way analysis of variance with Tukey's *post hoc* test at *p* < 0.05 were performed by SPSS Software (Version 15.0, SPSS Inc., Chicago, IL, USA).

## **3. Results and Discussion**

## *3.1. Water Extract Characterization*

As is well known, OMWW represent a complex medium in which more than 50% of the total phenolic components of the olive drupes are present [38]. UHPLC analysis provided identification of individual phenols in WE, as illustrated in Figure 1, but only the principal compounds were quantified (Table 1). The principal constituents of the Amberlite-desorption fraction were phenyl acids (caffeic, chlorogenic, *p*-coumaric and vanillic acids), phenyl alcohols (hydroxytyrosol and tyrosol), secoridoids (oleuropein), flavonoids (apigenin and luteolin), derivatives of hydroxycinnamic acid (verbascoside). WE showed a high content of hydroxytyrosol (834.51 mg 100 mL−<sup>1</sup> of sample) and tyrosol (147.55 mg 100 mL−<sup>1</sup> of sample) in agreement with other authors [39]. The TPC of WE was instead about 788.96 mg 100 mL−<sup>1</sup> of sample. The amount of recovered phenolic compounds is related to interactions between the adsorbates and adsorbent as well as chemical structure of the compounds themselves. Non-polar resins, or weakly polar, such as XAD7HP, allow the recovery of low molecular weight phenols, especially when ethanol is used as desorbing solvent [15]. On the other hand, a high concentration of ethanol promotes the solubilisation of alcohol-soluble impurities resulting in a drop-in desorption capacity [40]. Considering the high initial phenolic concentration and the amount of soluble impurities in OMWW, an increased amount of adsorbed molecules could occur per unit mass of absorbent, leading to saturation and reduction of desorption yield. This could explain why the TPC value results lower than the sum of individual compounds quantified by UHPLC.

The antioxidant activity of WE was measured by mean of DPPH and ABTS•<sup>+</sup> assays. Results obtained with the ABTS assay were higher as mmol TE 100 mL−<sup>1</sup> than those obtained by DPPH, according to our previous study [32]. As reported in the literature, antioxidant activity is due to the synergism between the various phenolic compounds and the assay responses are affected by the functional group's reactivity and characteristics, reaction time and complexity of the reaction kinetics [41,42].

(6)

*p*-coumaric

 acid; (7)

verbascoside;

 (8) luteolin; (9) oleuropein;

 (10) apigenin.


**Table 1.** Phenolic characterisation and antioxidant activity of wastewaters extract (WE). Data are expressed as mg 100 mL−<sup>1</sup> for phenols and mmol TE 100 mL−<sup>1</sup> for ABTS and DPPH assays.

## *3.2. Evaluation of E*ff*ect of WE on Sunflower Oil Stability*

#### 3.2.1. Qualitative Parameters

In sunflower oil enrichment, lecithin was used in order to promote the dispersion of WE into the lipid matrix. Lecithin stabilizes the added phenolic compounds in the oil matrix due to its amphiphilic behaviour, producing reverse micelles that include the extract [28]. Firstly, the effectiveness of enrichment was evaluated in terms of qualitative parameters and oxidative stability. The characteristics of sunflower oil used in this work are shown in Table 2. The oil showed a low free acidity (0.05%) while 6.1 mEq O2 kg−<sup>1</sup> of PV, according to the data about the commercial sunflower oil [43]. Spectrophotometric indices at 232 and 270 nm evidence the presence of dienes and trienes and the detected values (2.45 and 1.24) were characteristic of refined oils. Regarding enriched sunflower oil (Mboil), it is important to observe as the addition of extract involved the inclusion of water in the oil, as result of an increase in moisture content from 0.3% to 1% (Table 3). Free acidity of Mboil showed a rising trend during the storage: from 0.28 after enrichment to 0.32 at 10 ◦C and 0.35 at 25 ◦C after 90 days of storage, the formation of peroxides was reduced to about 49% in sunflower oil enriched with WE comparing with control sample at the beginning of storage. It could be linked to the phospholipids (the main constituents of soy lecithin) conferring to the oil oxidative stability [44]. According to the data in the literature [45–47], enriched samples, stored at different temperatures, showed a significant change (*p* < 0.01) of PV during storage. It can be noticed that the value of the PV was fluctuating progressively in time, reaching its maximum value on the 45th day (5.51 ± 0.21 at 10 ◦C and 5.47 ± 0.18 at 25 ◦C). The assessment of conjugated diene (K232) and conjugated triene (K270) is a reliable parameter for the measurement of oxidative deterioration of oils and, thus, the effectiveness of antioxidants in oils. The enriched samples had a little higher K270 values than the control; moreover, the conjugated triene content in the oil samples could be linked to the secondary oxidation compounds and conjugated trienes in the used commercial lecithin [48]. ANOVA data elaboration showed that no significant variations (*p* > 0.05) were observed during the storage at different temperatures. Likewise, other authors [46] have investigated the antioxidant efficacy of OMWW for the stabilization of lipid matrix obtaining the same our results.

**Table 2.** Qualitative parameters of sunflower oil.


**Table 3.** Qualitative parameters of Mboil during storage at 10 and 25 ◦C.


Different letters show significant differences among mean values by Tukey's *post hoc* test. \*\* Significance at *p* < 0 01; ns: not significant.

## 3.2.2. Oxidative Stability

To evaluate the resistance of fat oxidation, the oil samples were subjected to a high-oxidative stress environment using OXITEST reactor that shows a curve of oxidation characterized by an Induction Period (IP). It is the time necessary to reach an end point of oxidation that corresponds to a detectable rancidity or a rapid change in the oxidation rate. Oil stability was measured on control and Mboil just after the addition of WE and during storage, to evaluate the effect in protection from oxidation. In Figure 2, two examples of oxidation curves of oils stored at different temperatures were reported. The addition of the WE significantly involved an increase of the oxidative stability of the oils: Mboil samples had an average rise of oxidative stability of 50% (IP of 1022 min) with respect to the control that showed an Induction Period of about 540 min. The resistance of oxidation did not show a significant variation over time regardless of the storage temperature, from 0 to 90 days as reported in the tables included in Figure 2. The higher value of oxidative stability observed in enriched oils can be linked to the incorporation of phenolic compounds that are able to donate a hydrogen atom to the radical formed during the propagation phase of lipid oxidation [49]. Moreover, sunflower oil added with only lecithin was also analysed in order to evaluate the effect of the addition of lecithin on the sample stability against oxidation. From the Rancimat analysis of oil plus lecithin, a high Induction Period was observed compared with oil without emulsifier [50]. In our investigation, sunflower oil added with lecithin showed a lower value than control (420 min, Figure S1, Supplementary Materials). This confirms that the oxidative stability of enriched sample was related to the added phenolic extract. Previous studies also showed that the antioxidant protection of lecithin, attributed to phospholipids, was not effective for sunflower oil [51].

## 3.2.3. Phenolic Composition and Antioxidant Activity

The evolution of individual phenolic compounds added to oil was analysed during storage by UHPLC and the determination was repeated at different times (0, 45 and 90 days of storage). The analysis of the samples stored at different temperatures showed a similar phenolic composition (Table 4). According to the literature [52], samples were preferentially enriched with 50 mg L−<sup>1</sup> of hydroxytyrosol. After the enrichment process, the chromatographic analysis of the samples showed (at time 0) a higher content of hydroxytyrosol and tyrosol, and a lower amount of caffeic acid, luteolin, oleuropein and verbascoside was also detected. In general, a significant decrease (*p* < 0.01) of phenolic compounds was observed during the storage, particularly after 45 days: about 59% of hydroxytyrosol and 32% of tyrosol were lost. Not significant variations were detected at the 90th day compared to the loss detected at 45th day, except for tyrosol, which showed another loss of 40% at the end of storage. In this regard, it is important to point out that authors ascribe to the amphiphilic character of lecithin the reduction of extraction yield of phenolic compounds due to the development of stable emulsion between lecithin and phenols [48]. This also could explain the low amount of phenolic compounds quantified compared to those added. In addition, we speculated that the formation of small drops of WE affected the reaction with Folin-Ciocalteau, underestimating results (Table 5).

**Table 4.** Phenolic compounds (mg kg−<sup>1</sup> of sample) in Mboil during the storage (days) at 10 and 25 ◦C. letters and \*\* see Table 3.


The antioxidant activity of Mboil samples was analysed during storage using three different methods: ABTS, DPPH and ORAC assays. No single method is enough to determine the food antioxidant property, since different methods can give widely different results [53]. Moreover, the application of a single method can yield only a limited suggestion of the antioxidant activity of the samples under investigation [54]. The antioxidant activity of sample assays showed a significant variation (*p* < 0.01) over the time of storage of the enriched oils. A decrease of antioxidant activity was detected by ABTS assay according to the trend observed for TPC (*r* > 0.9). In contrast, a negative correlation was detected between TPC, DPPH and ORAC values. In addition, a non-linear trend was observed for DPPH and ORAC results with a minimum value detected at the 45th day. It should be considered that the added compounds have hydrophilic nature, thus their distribution in lipid phase is linked to their partition coefficient that may determinate also the distribution speed. It is conceivable that these compounds could be at first aggregated and only afterwards a suitable period distributed to the matrix. Multivariate analysis revealed that different temperatures did not significantly influence the total phenol content and the antioxidant activity measured by different assays while time seemed to affect them. Considering the oxidative stability of samples over time, the decrease of antioxidant

activity is probably linked to the use of lecithin that led to the formation of lamellar structures in which hydrophilic antioxidants may be entrapped [48]. Despite this, from the start to the end of storage, at both temperatures, only about 3% of reduction was observed for the ORAC test: this is in line with the slow decrease of TPC in enriched oils and the observed oxidative stability of samples, so it confirms the robustness and the validity of ORAC method for the determination of the radical scavenging activity of the lipid matrix [50].

**Table 5.** TPC (mgGA 100 g<sup>−</sup>1) and total antioxidant activity by ABTS, DPPH and ORAC assays (μmol TE 100 g<sup>−</sup>1) of the samples stored at different temperatures.


Means within a row with different letters are significantly different by Tukey's post hoc test. \*\* Significance at *p* < 0.01.

The obtained results confirm that the use of a low concentration of WE improve the nutritional quality of refined oil, whereas in previous studies the effectiveness of retarding lipid oxidation of oil is correlated to addition of high concentration of phenolic extract [9]. It is possible to hypothesize the use of WE for the fortification of different kinds of food products.

## **4. Conclusions**

In conclusion, the addition of a phenolic extract obtained by OMWW in sunflower oil permitted the production of an enriched oil with a higher content of polyphenols and antioxidant properties for up to 90 days of storage. The two tested temperatures did not affect these results, so it can be considered an initial step to the enriched sunflower oil production, with increased antioxidant characteristics for a more prolonged time. The successful results are also linked to the valorisation of olive industry by-product, such as OMWW, converted from waste to resource, through an efficient methodology of extraction.

**Supplementary Materials:** The following are available online at http://www.mdpi.com/2304-8158/9/7/856/s1. Figure S1. Oxidation curves: red (sunflower oil), blue (sunflower oil plus lecithin), yellow, violet, orange and green (Mboil during the storage).

**Author Contributions:** Sampling and laboratory analysis R.R. and V.I.; literature search R.R.; Conceptualization, R.R. and A.P.; writing original draft, data provision R.R. and A.D.B.; review, editing and approval of final text, A.P. and M.P.; supervision and approval M.P. All authors have read and agreed to the published version of the manuscript.

**Funding:** This research was funded by AGER 2 PROJECT, grant number 2016-0105".

**Conflicts of Interest:** The authors declare no conflict of interest.

#### **References**


© 2020 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (http://creativecommons.org/licenses/by/4.0/).

## *Article*
