Metabolomic Profiling and Biological Activities of Pleurotus columbinus Quél. Cultivated on Different Agri-Food Byproducts

The genus Pleurotus (Fr.) P. Kumm (Pleurotaceae, Basidiomycota) comprises a cosmopolitan group of mushrooms highly appreciated for their nutritional value and health-promoting benefits. Despite there being many studies about the phytochemical composition of Pleurotus spp., there are very few reports dealing with the phytochemistry, antioxidant and antimicrobial activities of P. columbinus Quél. In this study, a mass spectrometry ultra-performance liquid chromatography mass spectrometry (UHPLC)-QTOF method, coupled with principal component analysis (PCA), was applied to the P. columbinus metabolome in order to investigate the influence of different agri-food residues as growth substrates for P. columbinus cultivation, on the bioactive chemical profile of fruiting bodies and evaluated their potential as antioxidants and antimicrobials. Additionally, a quantitative HPLC-DAD-MS analysis was conducted on phenolic and flavonoid compounds, that could explain, albeit partially, the observed biological effects of P. columbinus extracts. The qualitative metabolic profile identified 97 metabolites, whereas the quantitative HPLC-DAD-MS analysis confirmed the presence of phenolic and flavonoids, in the mushroom extracts, which also showed intrinsic scavenging/reducing and antimicrobial effects. The antibacterial effects were particularly evident against Escherichia coli, whereas Tricophyton and Aspergillus were the dermatophytes more sensitive to the mushroom extracts. The present study supports more in-depth investigations, aimed at evaluating the influence of growth substrate on P. columbinus antimicrobial and antioxidant properties. The extracts from P. columbinus revealed valuable sources of primary and secondary metabolites, thus suggesting potential applications in the formulation of food supplements with biological properties, above all in terms of antioxidant and antimicrobial properties.

The names P. columbinus Quél. has been applied to blue-greenish blue variant of P. ostreatus. Hilber [31] considers P. columbinus a variety of P. ostreatus (P. ostreatus var. columbinus Quél., Enchir. Fung.: 148. 1886) and demonstrated a high degree of intercompatibility (>85%) between var. columbinus and var. ostreatus. The investigation of Zervakis and Labarère [32], based on isoenzymes from 23 Pleurotus spp. isolates examined by polyacrylamide gel electrophoresis isoelectric and protein blotting, however, demonstrated that P. columbinus (Figure 1) could be regarded as a separate taxon. Despite there being many studies about the phytochemical composition of Pleurotus spp., also highlighted by recent reviews [33][34][35] there are very few studies dealing with the phytochemistry, antioxidant and antimicrobial activities of P. columbinus Quél [36,37]. Metabolomics is a new discipline which is defined as the monitoring of metabolite concentration in fungi, bacteria and plants. Nowadays, the metabolomics-based approach has been gradually applied in the field of edible and medicinal mushrooms to gain insight into the chemical compositions of biological processes and the understanding response of mushrooms to certain environmental conditions [38].
In this study, a mass spectrometry ultra-performance liquid chromatography mass spectrometry (UHPLC)-QTOF method, coupled with principal component analysis (PCA), was applied to the P. columbinus metabolome in order to investigate the influence of different agri-food residues as growth substrates for P. columbinus cultivation, on the bioactive chemical profile of fruiting bodies and evaluated their potential as antioxidants and antimicrobials. Additionally, a quantitative HPLC-DAD-MS analysis was conducted on phenolic and flavonoid compounds, possibly involved in the observed biological effects of P. columbinus extracts.

Mushroom Identification
The morphological characteristics of Pleurotus columbinus fruiting body correspond to those reported by Bas et al. [39]. Despite there being many studies about the phytochemical composition of Pleurotus spp., also highlighted by recent reviews [33][34][35] there are very few studies dealing with the phytochemistry, antioxidant and antimicrobial activities of P. columbinus Quél [36,37]. Metabolomics is a new discipline which is defined as the monitoring of metabolite concentration in fungi, bacteria and plants. Nowadays, the metabolomics-based approach has been gradually applied in the field of edible and medicinal mushrooms to gain insight into the chemical compositions of biological processes and the understanding response of mushrooms to certain environmental conditions [38].
In this study, a mass spectrometry ultra-performance liquid chromatography mass spectrometry (UHPLC)-QTOF method, coupled with principal component analysis (PCA), was applied to the P. columbinus metabolome in order to investigate the influence of different agri-food residues as growth substrates for P. columbinus cultivation, on the bioactive chemical profile of fruiting bodies and evaluated their potential as antioxidants and antimicrobials. Additionally, a quantitative HPLC-DAD-MS analysis was conducted on phenolic and flavonoid compounds, possibly involved in the observed biological effects of P. columbinus extracts.

Mushroom Identification
The morphological characteristics of Pleurotus columbinus fruiting body correspond to those reported by Bas et al. [39].
Considering the influence of cultivation conditions on morphological and physiological features, the DNA barcoding is requested for the identification of Pleurotus species. The exact characterization and identification of medicinal mushrooms is fundamental for exploiting their full potential in food and pharmaceutical industries [29].
The taxonomic affiliation of the mushroom strain was performed via targeting the internal transcribed region of the ribosomal DNA. The ITS sequence of sample PeruMyc 2474 was consistent with the species P. columbinus; accordingly, a phylogenetic tree with Hypsizygus marmoreus as an outgroup ( Figure 1).
The PeruMyc 2474 strain clearly clustered with other P. columbinus strains and were related to P. pulmonarius, P. ostreatus and P. eryngii, as well.

Untargeted LC-MS/MS-Based Metabolomics
In this study, the chemical profile of P. columbinus was evaluated through mass spectrometry ultra-performance liquid chromatography mass spectrometry (UHPLC)-QTOF method, coupled with principal component analysis (PCA). The full list of metabolites annotated using the mummichog algorithm is included in Table S1 (Supplementary Material). The data matrix showing the annotation and peak areas for each sample was subjected to statistical analysis. The ANOVA performed with a p-value cut-off of 0.001 found 97 significant and 136 non-significant metabolites. In Table S2 (Supplementary Material), it is reported the list of significance using post hoc analysis (Fisher's LSD), whereas Figure 2 shows the heatmap using the 50 most significant metabolites from the ANOVA test. Specifically, the fungi grown in the substrate D showed higher levels of carbohydrates, such as sucrose and mannose, compared to the reference substrate A, whereas the substrate B is related to higher amounts of aminoacids, such as L-glutamine and L-proline, in the fruiting bodies. This is also consistent with the functional metanalysis carried out to contextualize the metabolomics profile ( Figure 3). Comparisons were made between the metabolic profiles of the fungus grown in substrates B-D with respect to substrate A taken as reference. The most evident thing is that the metabolic pathways are strongly influenced by the chemical composition of the growth substrate. Some differences can be tentatively explained. For example, mannose degradation is greater for the fungus grown on substrate C. Indeed, this substrate contains, among other things, coffee grounds which are rich in mannose. A similar trend is noted for the metabolic pathway of leucine biosynthesis which is increased for the fungus grown on substrate C. In this substrate, free leucine is scarce compared to substrates B and D where the presence of soybeans constitutes an immediate source of free leucine.

Phenolic and Flavonoid Determination via HPLC-DAD-MS
An HPLC-DAD-MS analysis was also carried out in order to measure the levels of selected phenolic and flavonoid compounds, namely, gallic acid, hydroxytyrosol, catechin, chlorogenic acid, epicatechin and benzoic acid, that play a major role in the antioxidant/antimicrobial response following mushroom extract administration [40,41]. Specifically, Table 1 shows that the level of gallic acid is higher in extracts A and C, whereas extract B do not show a relevant amount of this compound. In the extracts B-D, the catechin fraction was present at higher concentrations, compared to extract A. This last extract also had the highest epicatechin content, whereas extract A showed higher benzoic acid concentration. According to the quantitative analysis conducted, extract C was the richest, in terms of qualitative and quantitative composition in phenolic compounds.

Antimicrobial and Antioxidant Effects
The antimicrobial activity of the extracts A-D are shown in Table 2, also in comparison with reference antimicrobial drugs ciprofloxacin, fluconazole and griseofulvin. All extracts from mushroom displayed antimicrobial activity in the concentration range of 6.25 to 200 µg/mL. Regarding the yeasts, C. parapsilosis (YEPGA 6551) was the most sensitive strain to the extracts, with MIC ranges of 31.49->200 µg mL −1 , while C. albicans (YEPGA 6379) showed the least sensitivity to the mushroom extracts. The results of the growth inhibition of yeast strains evidenced a major activity of the extract A, underlining the role of growth substrate on P. columbinus extract properties. With reference to bacteria, the strongest inhibition was observed for the extracts B and C [MIC 6.25-12.5 µg/mL against E. coli (ATCC 10536) and PeryMycA 2]. Collectively, Gram− bacterial strains (PeruMyc 2, 3, 5 and 7) were less sensitive to mushroom extracts than that of Gram+ ones. Intriguingly, the B. cereus strain PeruMycA 4 showed the lowest MIC values. All results from the tested extracts showed active inhibition of dermatophytes growth. Regarding A. currey (CCF 5207), it was the most sensitive fungal species to mushroom extracts, with MIC range between 9.92 and 79.37 µg/mL. Values of MIC < 100 µg/mL was considered as an index of high antimicrobial activity (Dogan et al. 2013). The highest antimicrobial activity of riseofulvin was against T. tonsurans (CCF 4834) (MIC: 0.125-0.25 µg/mL). On the other hand, the present data did not permit to whether the isolates were resistant to the griseofulvin as no breakpoints have yet been established. * MIC values are reported as geometric means of three independent replicates (n = 3). MIC range concentrations are reported within brackets.
Regarding the antioxidant activity, experimental data were normalized and expressed as EC 50 values (µg/mL) for each mushroom extract and Trolox, which was used as reference antioxidant compound. The results were given in Table 3. Values for DPPH radical scavenging activity varied between 2.25 and 4.98; extract B was the most active. Decidedly low was the activity of the extract A, with a mean value of 4.98, whilst the extract C had a medium-low mean value (3.81). Values for ABTS radical scavenging activity varied between 4.34 and 6.16, and the best activity was shown by the extracts B, with a mean value of 4.34 referred to the Trolox. The values for β-carotene-linoleic acid assay varied between 7.47 and 11.65; the higher activity was showed by the extract B, with a mean value of 7.47 referred to the Trolox, while lower antioxidant effects were detected for the other extracts. Although the content of phenolic compounds was previously related to the scavenging/reducing and antimicrobial properties of extracts, especially in the case of polar extracts [42,43], in the present study the intrinsic antioxidant and antimicrobial abilities of the extracts seemed to be dependent, at least in part, from the content of the gallic acid present in the extracts (their correlation coefficients with gallic acid were <0.43). Matrix analysis (Table S3) revealed a strong positive correlation (correlation coefficients > 0.89) of quantitative presence of gallic acid in the extracts and the antioxidant activities. Furthermore, among the three tests used for antioxidant activity a largely positive correlation was observed as expressed by coefficients falling in the range 0.88-0.99. On the other hand, antioxidant properties of the extracts were less affected by the presence of other detected flavonoids and phenolics. This reflected the complexity of the analyzed biological matrices. Regarding the antimicrobial properties, the results from the present study did not provide the optimal substrate for the cultivation of fungi with antimicrobial properties; however, the effect of the substrate was present and should be deeply considered in view of the production of antimicrobial extracts from Pleurotus species. For instance, the differences in the metabolic pathway activation induced by the substrates seemed to affect the sensitivity of Candida species to the mycostatic effects by the extracts. Future studies need to unravel the mechanism underlying this effect.

Mushroom Material
The fruiting bodies of the strain of P. columbinus For the isolation of mycelia, portions (less than 5 mm) of context were excised aseptically from inside the basidiome, transferred into Petri dishes containing Rose Bengal Chloramphenicol agar (Sigma-Aldrich, Milan, Italy) and incubated for 7 d at 24 • C [45,46]. The mycelial strains are deposited in the DCBB culture collection and subcultured on MEA medium every three months.

Molecular Identification
Angelini et al. [47] method was used to extract the total genomic DNA from ten days mycelium grown in MEA. The detailed protocol is included in our recent paper [2]. Whereas the phylogenetic analysis was conducted according to literature [48][49][50].

Spawn Production
P. columbinus spawn was produced following the standard method with the use of barley grains. Details about the protocol followed are reported in a previous paper of ours [2].

Mushroom Cultivation Substrates
The P. columbinus strains were cultivated on four substrates consisting of (A) wheat straw in 1:1 w/w ratio with beech sawdust (as control), (B) wheat straw in 4:2:1 w/w/w ratio with oak sawdust and soya beans, (C) wheat straw in 4:2:1 w/w/w ratio with oak sawdust and coffee grounds and (D) wheat straw in 3:2:1 w/w/w ratio with beech sawdust and soya beans. Wheat straw, sawdust, soya beans and coffee ground were obtained from the "Soc. Coop. Umbria Verde farm" (Perugia, Central Italy).
Soya beans make certain free amino acids such as Glutamate, Aspartate, Leucine, Arginine, Serine, Lysine and Proline readily available. Coffee grounds are particularly rich in sugars such as mannose, galactose and arabinose. The detailed protocol is included in our recent paper [2].

Preparation of Mushrooms Methanol Extracts
The fruiting bodies of P. columbinus grown on substrates A-D were manually collected and selected discarding parts presenting non-healthy aspect, of physical damage. Fresh material was grossly divided in slices and immediately macerated in methanol for seven days at 20 • C (1:10 w:v). Extracts were then centrifuged (5000× g for 10 min), the residue was removed and the liquid phase was directly used for phytochemical investigation or taken to dryness in a rotary evaporator (bath temperature below 50 • C) and calculated the loss of drying. The extraction yield resulted 41.03, 28.92, 43.19 and 30.90 mg of dry extract from 1 g of fresh fruiting body, respectively, from substrates A-D. Extraction of samples deriving from each substrate (A-D) were performed in two independent extractions and the liquid phase from both extractions combined together and used as single sample.
As for the biological tests, an accurately weighted aliquot of extract was mixed in a defined volume of distilled water and solubilized in a sonicating bath at room temperature for one hour.

Untargeted LC-MS/MS-Based Metabolomics and Statistical Analysis
Untargeted metabolomics was carried out by using ultra-performance liquid chromatography mass spectrometry (UHPLC)-QTOF employing a 1260 ultra-high-performance liquid chromatograph and a G6530A QTOF mass spectrometer (Agilent Technologies, Santa Clara, CA, USA). The chromatographic conditions are fully described in our recent paper [2].
Compound annotation was made using mummichog algorithm [51] implemented in "Functional analysis" module of MetaboAnalyst 5.0 [52] using 5 ppm of tolerance for both polarities. Heatmap, ANOVA and Functional Meta Analysis were also performed with MetaboAnalyst. For statistical analysis, samples were normalized by median, followed by pareto scaling.

Phenolic and Flavonoid Determination: HPLC-DAD-MS Analyses
The identification and quantification of selected phenolic compounds, namely gallic acid, benzoic acid, catechin, hydroxytyrosol, chlorogenic acid and epicatechin was carried out through HPLC-DA-MS analysis. The detailed protocol is included in a recent paper of ours [53].

ABTS Assay
The ABTS radical cation scavenging activity was performed according to Re et al. [57] and Ozturk et al. [55].

Antifungal Activity
Susceptibility testing against yeasts and filamentous fungi was performed according to the CLSI M27-A3 and M38-A2 protocols, respectively.
For the mushroom extracts, the MIC end-points were defined as the lowest concentration that showed total growth inhibition.

Statistical Analysis
The results were expressed as mean ± standard deviation and analyzed via Student test. The statistical significance was set a p < 0.05 and analysis was conducted through GraphPad Prism 5.01 version (GraphPad Software, San Diego, CA, USA).

Conclusions
The present findings support more in-depth investigations aimed at evaluating the influence of growth substrate on P. columbinus antimicrobial and antioxidant properties. The extracts from P. columbinus revealed valuable sources of primary and secondary metabolites, thus suggesting potential applications in the formulation of food supplements with biological properties, above all in terms of antioxidant and antimicrobial properties. Future studies are needed to further improve our knowledge of the metabolic pathways and the complexity of the compounds measured. This will also allow in understanding the mechanisms of action at the basis of the observed effects.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.