Genetic and Chemical Diversity of Edible Mushroom Pleurotus Species

The genus Pleurotus is one of the most widely cultivated and edible mushrooms with various cultivators. Three molecular characteristics were used to evaluate the genetic diversity of 132 tested samples. Phylogenetic analysis showed five clades for tested samples of the genus Pleurotus by the combined ITS and LSU sequences with strong bootstraps and Bayesian posterior probability supports. A total of 94 polymorphic fragments ranging from 10 to 100 bp were observed by using an intersimple sequence repeat (ISSR) marker. The DNA fragment pattern showed that P. ostreatus cultivator (strain P9) was clearly distinguished from wild strain based on their clear banding profiles produced. DNA GC content of the genus Pleurotus varied from 55.6 mol% to 43.3 mol%. Their chemical composition was also determined, including sugar, amino acid, polar lipid, mycolic acid, quinone, and fatty acid, which presented some high homogeneity. Most of the tested samples contained mycolic acid; glucose and arabinose as the main sugars; aspartic acid, arginine, lysine, tyrosine, and alanine as the main amino acids; and C16:0, C18:0, C18:2cis-9,12, anteiso-C14:0, and summed feature 8 as the main fatty acids. In addition, their polar lipid profiles were investigated for the first time, which significantly varied among Pleurotus species. The genus Pleurotus contained menaquinone-6 as the sole respiratory quinone, which showed a significant difference with that of its closely related genera. These results of this study demonstrated that the combined method above could efficiently differentiate each Pleurotus species and thus be considered an efficient tool for surveying the genetic diversity of the genus Pleurotus.


Introduction
The species of the genus Pleurotus are among the most cultivated and consumed edible mushrooms in the world [1]. Currently, several Pleurotus species can be grown commercially to high yields, such as Pleurotus ostreatus and Pleurotus eryngii. Their production has exceeded 15 million tons each year, which was ranked second following Lentinula edodes [2]. In addition to their nutritional value, the genus Pleurotus is a natural source of prebiotics and antioxidants [3] and is thus of great interest to the food industry. Also, the genus Pleurotus showed a broad spectrum of biological activities [4,5], including antitumor, antidiabetic, and antibacterial activities.
Currently, the morphological characteristics have laid down the foundation for Pleurotus identification. Nonetheless, the characteristics of Pleurotus cultivars are unstable and dependent on the environmental conditions, which are too limited in terms of accurate identification [6]. With the development of molecular technology, highly conserved regions in ribosomal DNA are served as the reference points to investigate genetic and evolutionary relationships within species for providing unequivocal species delimitation rule [7,8], such as 18S, internal transcribed spacers (ITS), and large subunit rDNA (LSU) ( Figure S1).
Besides, molecular fingerprinting is also an efficient tool in investigating the genetic diversities for the purpose of breeding programs because of its properties such as independence of environmental parameters and the high levels of detectable polymorphism [9], including simple sequence repeat (SSR), random amplified polymorphic DNA (RAPD), amplified fragment length polymorphism (AFLP), sequencerelated amplified polymorphism (SRAP), and intersimple sequence repeat (ISSR) markers. Among them, an obvious advantage of the ISSR marker is that it has no sequence data for primer construction and is randomly distributed throughout the genome. ISSR primers (AG, GA, and (GATA)n repeats) that are anchored to genomic DNA making targeted simple sequence repeats could generate a wide array of amplification products, which provide sufficient information in determining genetic relationships [10]. DNA GC content as an important molecular characteristic has been widely used in taxonomic descriptions of species and genera [11]. However, only a few species of Pleurotus have been investigated to analyze their GC content [12,13].
Chemical characteristics were provided as supplementary criteria to more accurately describe species, which are being given more and more attention [14]. These approaches were used way before DNA-based methods and are mainly regarded as obsolete or less informative nowadays, yet not necessarily useless. Sugar and amino acid profiles as a complementary tool have been applied to identify varieties of Gibberella fujikuroi [15]. Studies on chemical characteristics of the genus Pleurotus have made some progress [16,17], but the study is not wide enough.
In the present study, a total of 132 tested samples were collected from Korean Agricultural Culture Collection (KACC, Korea), National Agrobiodiversity Center (NAAS, Korea), Korean Culture Center of Microorganisms (KCCM, Korea), Culture Center of Microorganisms Jilin Agricultural University (CCMJ, China), and market (Homeplus in Yongton-si of Korea) (Table S1). Subsequently, all tested samples were investigated to analyze their genetic and chemical diversity and to compare the differences between them. This information will promote the efficient identification of the genus Pleurotus.

Materials and Methods
2.1. Strains and Cultivation. All strains were cultivated in potato dextrose broth (PDB, BD products, Franklin Lakes, NJ, USA) at 32°C for 2 days as seed culture, and then, the seed culture (3% v/v) was transferred into 50 mL PDB at 32°C for 10 days. Mycelium was obtained by centrifugation at 10,000 g for 10 min and freeze-dried for further investigation as tested samples.

Phylogenetic
Analysis. The genomic DNA from an overnight culture was obtained by using the Qiagen Genomic Tip 500G Kit, following the manufacturer's instructions, except that the lysozyme was replaced with lysostaphin to a final concentration of 200 μg/mL and then was stored at -20°C. ITS and LSU regions were performed by using universal primers (primers ITS4 and ITS1 for ITS, primers LR0R and LR3 for LSU). Then, amplification products were sequenced by Biofact Co. Ltd, (Seoul, Korea). All generated sequences were submitted to GenBank and are listed in Table S1.
The sequences of tested samples together with reference sequences from GenBank (Table S2) were aligned by using BioEdit and ClustalX. Alignment was manually adjusted to allow maximum alignment and to minimize gaps. Maximum parsimony analysis was applied to the combined ITS and LSU sequences. The sequence of Agaricus bisporus obtained from GenBank was used as an outgroup. PAUP version was carried out as a phylogenetic tree construction procedure. All characters were equally weighted with gaps as missing data. The phylogenetic tree was estimated by using TBR branch swapping and 1,000 random sequence additions. Max-trees were set to 5,000, branches of zero length were collapsed, and all parsimonious trees were saved. The robustness of clades was tested by using a bootstrap (BT) analysis with 1,000 replicates. Then, several factors of Maximum Parsimonious Tree (MPT) were calculated, such as homoplasy index (HI), descriptive tree statistics tree length (TL), rescaled consistency index (RC), retention index (RI), and consistency index (CI). The best evolution for each data set was determined for Bayesian inference (BY) by using MrModeltest 2.3. Bayesian inference was performed by using MrBayes 3.2.3 with a general time reversible (GTR) model of DNA substitution and a gamma distribution rate variation across sites. Four Markov chains were run for 2 runs from random starting trees for 2 million generations, and trees were sampled every 100 generations by discarding the first one-fourth generations as burn-in [18]. A majority rule consensus tree of all remaining trees was calculated. Branches were considered significantly supported with 80% of maximum parsimony (MP) and 0.95 of Bayesian posterior probabilities (BPP).

Genetic Diversity Analysis.
A total of 24 primers were collected and applied in this study (Table 1). Then, primer, annealing temperature (47°C, 50°C, 53°C, 60°C, 65°C, and 70°C), and the number of cycles (10, 15, 20, 25, 30, and 35) were studied. The ISSR amplifications were carried out in a 20 μL reaction volume containing 50 ng of the template DNA, 0.75 μM of the primers, 2.5 mM of MgCl 2 , 0.2 mM of dNTPs (TaKaRa, Japan), and 0.5 U of Taq DNA polymerase. The amplification conditions were as follows: an initial denaturation at 94°C for 5 min followed by the corresponding cycles each at 94°C for 30 s, 45 s at the annealing temperature, and 72°C for 90 s, followed by the final extension for 7 min at 72°C. The ISSR amplification products were detected on 1% agarose electrophoresis gels, and the images were captured using the ChemiDoc XR (Bio-Rad, USA). Unambiguous and reproducible bands in successive amplifications were selected for scoring. Each fragment was scored as "1" and "0" for the presence/absence, and a genetic distance matrix from raw data was constructed by using the PHYLIP 1.0 package [19]. Cluster analysis was performed BioMed Research International by the unweighted pair group method with arithmetic averaging (UPGMA) [20].
2.4. DNA GC Content Analysis. The genomic DNA was obtained by using the Qiagen Genomic Tip 500G Kit and purified by using genomic DNA purification kits (Intron Biotechnology Inc., Seongnam, South Korea). Briefly, the purified DNA was decomposed into nucleotides, and the DNA GC content was determined by the method of Tamaoka and Komagata [21] using reversed-phase highperformance liquid chromatography (RP-HPLC).
2.6. Amino Acid Analysis. The amino acid profiles of each tested sample were determined by using HPLC analysis. Chromatography conditions were in accordance with the Agilent method [23]. Briefly, an amount equivalent to 2.5 μL of each sample was injected on a Zorbax Eclipse-AAA column (5 μm, 150 × 4.6 mm) (Agilent), at 40°C, with detection at 338 nm. Mobile phase A was 40 mM NaH 2 PO 4 , adjusted to pH 7.8 with NaOH, while mobile phase B was acetonitrile/methanol/water (45/45/10 v/v/v). The separation was obtained at a flow rate of 2 mL/min with a gradient program that allowed for 1.9 min at 0% B followed by a 16.3 min step that raised eluent B to 53%. Then, washing at 100% B and equilibration at 0% B were performed in a total analysis time of 26 min. The amino acid was identified by comparing calibration chromatogram established by 10 known amino acids, such as arginine (Arg), alanine (Ala), aspartic acid (Asp), valine (Val), cysteine (Cys), glutamic acid (Glc), glycine (Gly), lysine (Lys), threonine (Thr), and tyrosine (Tyr).

Mycolic Acid Analysis.
To investigate the distribution of mycolic acid in the genus Pleurotus, mycolic acid was extracted, purified, and analyzed by using TLC with petroleum ether/acetone (95/5 v/v) as the developing solvents. After air drying, dots were visualized by iodine fumigation [26].

Fatty Acid Analysis.
Fatty acid profiles were determined by gas-liquid chromatography and identified by using the Sherlock Microbial Identification System (MIDI) [27]. The relative percentage of each fatty acid was calculated by internal normalization of the chromatographic peak area.

Results and Discussion
3.1. Phylogenetic Analysis. In this study, about 100 equally parsimonious trees were produced by maximum parsimony analysis with RI = 0:652, TL = 1132, RC = 0:302, CI = 0:367, and HI = 0:632. The same topology was also obtained by Bayesian analysis with an average standard deviation of split frequencies (0.009). As shown in Figure 1, a well-resolved phylogenetic tree was constructed by using the combined sequences of ITS and LSU.  3 BioMed Research International All tested strains were split into five clades, and most of these clades were recovered by the combined ITS and LSU sequences with strong bootstraps and Bayesian posterior probability supports. Clade I was formed and comprised four subclades, and subclade A was composed of Pleurotus smithii and Pleurotus australis, indicating that they were closer in a relationship. Subclade B included Pleurotus abalonus and Pleurotus cystidiosus. Previously reported P. abalonus was also considered to be a subspecies of P. cystidiosus (P. cystidiosus subsp. abalonus) [28], which strongly supports results of our study. However, there were significant differences in morphology between them. P. cystidiosus exhibited a specific anamorphic stage of the genus Pleurotus, which is manifested by the presence of arthroconidia on conidiophores assembled in the coremium macrostructures, suggesting the ability for asexual reproduction independently, as a possible mechanism for  Figure 1: Phylogenetic analysis of species in the genus Pleurotus. Strict consensus tree was generated by maximum parsimony based on combined ITS+LSU sequences. Parsimony bootstrap proportions (before the/) higher than 80% and Bayesian posterior probabilities (after the/) more than 0.95 were indicated along branches. The reference sequences of Pleurotus (bold) were downloaded from GenBank (Table S2). 4 BioMed Research International effective dispersal of these species in the wild [29]. P. tuberregium was separated as an individual subclade (C) in clade I. Pleurotus rattenburyi was identified as novel species by Redhead and Norvell, which was closely related to Pleurotus purpureo-olivaceus [30], and they made up subclade D in this study. Clade II was composed of Pleurotus dryinus, Pleurotus citrinopileatus, and Pleurotus cornucopiae. Among them, P. dryinus was formed as an individual subclade (E). Both P. citrinopileatus and P. cornucopiae made up subclade F, suggesting a closed relationship between them at the generic level. P. citrinopileatus and P. cornucopiae were classified as belonging to the same intersterility group, despite the confirmed differences in their ITS sequences [31]. Then, Petersen and Krisai-Greilhuber reconsidered P. citrinopileatus species status and define it as P. cornucopiae var. citrinopileatus [32]. However, since differences between both species are detected not only at the molecular level but also in their morphology, further examinations and additional crosses between representatives of both species are needed to definitively confirm or deny the existence of interspecific reproductive barriers among them. Clade III was composed of Pleurotus fossulatus, Pleurotus djamor, Pleurotus elongatipes, Pleurotus salmoneostramineus, Pleurotus calyptratus, Pleurotus flabellatus, Pleurotus ostreatoroseus, Pleurotus incarnatus, and Pleurotus nebrodensis, which comprised three subclades (G, H, and I). P. salmoneostramineus was revised as P. djamor, but they are not in the same subclade. Similar results also prevailed in P. fossulatus and P. nebrodensis. The origin and taxonomic status of P. flabellatus were difficult to determine by previous data, which was associated with P. cornucopiae [33], or considered an intermediate between the P. ostreatus and the P. eryngii clades [34]. Previous studies suggested that P. calyptratus is a variation within the P. djamor [35]. However, both species were divided into two subclusters on the phylogram in this study. Meanwhile, they are also differentiated by their habitat; for example, P. calyptratus is distributed in the temperate climate zone, while P. djamor is common in a warm tropical climate zone. Their edibility is also different; for example, P. djamor has been cultivated commercially and edible. In turn, P. calyptratus are hard and are not edible. It was confirmed that P. calyptratus and P. djamor represent a single species. Most interestingly, the taxonomic position of P. elongatipes remained uncertain, which was proposed as Hypsizygus elongatipes in Index Fungorum and P. elongatipes in MycoBank. In our study, two samples of P. elongatipes were gathered together with a 100% bootstrap value and 1.00 Bayesian posterior probability, and it strongly grouped with P. salmoneostramineus (91% MP, 0.94 BPP), which has given qualified support to views of MycoBank. P. nebrodensis also formed a monophyletic clade by a weak support (less than 80% MP), suggesting that it is slightly distant from other species in clade III. Clade IV included Pleurotus opuntiae, P. sapidus, Pleurotus pulmonarius, and Pleurotus eous. Our analysis inferred that P. opuntiae, P. sapidus, P. pulmonarius, and P. eous formed a well resolved monophyletic clade with strong support (100% MP, 1.00 BPP). Our study indicated that P. sapidus was distinct from P. cornucopiae at the genetic level, which differed from the view of Albert et al. [35], suggesting that both represent a single species. Clade V as the largest clade was composed of  [36] or separated [34]. In our analysis, P. cornucopiae on clade III and P. columbinus and P. ostreatus on the clade V were included in two distinct clades. It was to be noted that the close relationship between P. columbinus and P. ostreatus has been previously reported [37] and that they have recently been described as sexually compatible species [38]. Sequences of Pleurotus florida and Pleurotus floridanus obtained from GenBank were gathered together with P. ostreatus by a weak support (less than 80% MP). Previous data has been described that P. floridanus as the invalid name was revised as P. ostreatus [30] and P. florida was geographical isolates from the P. ostreatus complex, also called P. ostreatus [39]. It was suggested that they might have a close genetic relationship, which strongly supported our results.

Genetic Diversity Analysis.
A total of 24 primers were applied to investigate genetic diversities of the tested samples by using the ISSR marker ( Figure S2). Among them, primers R1 and R3 failed to produce any PCR products and were discarded from further analysis. From the total number of bands obtained and the percentages of polymorphisms of each tested primer, 52 out of 76 bands were considered polymorphic, generating a polymorphism information content (PIC) overall average of 68.4%. The maximum percentage of polymorphism was observed using primer R17 (90.0%), while the lowest percentage was observed using primers R10 and R24 (50.0%). To increase the richness and legibility of fragments, optimal annealing temperature cycle number was detected at 53°C and 25, respectively ( Figure S3). The annealing temperature was often higher than Tm, which has more polymorphic fragments than that of other annealing temperatures [10]. In this study, a total of 94 polymorphic fragments ranging from 10 to 100 bp were observed, which have a higher polymorphism level than those in a previous study [40]. As shown in Figure 2, the average coefficient is 0.67, ranging from 0.37 to 0.97. Based on the UPGMA dendrogram, all tested samples were clustered into several clusters corresponding to their respective taxa; polymorphic fragments were almost consistent with the coefficient of 0.65 at the intraspecies level. A special polymorphic fragment was detected in the P. ostreatus cultivator (strain P9), which has significant differences in wild strain. The breeding stage might have a significant impact on the genetic diversity of P. ostreatus, which was consistent with a previous report [41]. The ISSR marker was also considered to be a more useful method in distinguishing P. eryngii isolates and varieties [42,43]. The genetic similarity among P. eryngii, P. fuscus,
3.3. DNA GC Content Analysis. DNA GC content of the genus Pleurotus varied from 43.3 mol% in P. opuntiae to 55.6 mol% in P. fuscus var. ferulae. Meier-Kolthoff et al. confirmed that the threshold value of GC content was 3-5% within species [11]. Exceptions did exist, especially with fungi, which sometimes altered inconsistently with the above-said patterns. A threshold value of GC content was 8% in yeast [46] and 1% in the genus Trichosporon [47]. Only a few species of the genus Pleurotus have been investigated ( Table 2). In addition, GC content of other species of the genus Pleurotus, such as P. nebrodensis, has not yet been determined despite their importance for different industries. In this study, the range of GC content was slightly different from the values reported previously. This might be caused by incubation conditions of the mycelium, which was supported by Cui et al. [48].

Sugar Analysis
. Sugar profiles of all tested samples showed some high homogeneity ( Table 3). All of them contained Glu, Ara, and Xyl. Part sugars slightly varied among species, such as Gal, Rib, and Rha. This phenomenon also appeared in that of yeast [49]; for example, Xyl and Ara existed in different yeast species. Our results showed that the types of sugar in tested samples were in agreement with previous results [50]. Currently, sugar might play an important role in food and provide the majority of energy for human being. Meanwhile, Jacob et al. confirmed that Pleurotus was low in sugar and became a popular low-calorie food [51].
3.5. Amino Acid Analysis. A total of 10 types of amino acid were detected in the tested samples (Table 3), such as Asp, Arg, Lys, Tyr, Ala, Cys, Gly, Thr, Glc, and Val. All of them contained Asp, Tyr, and Ala. Then, other types of amino acid varied with the different species of the genus Pleurotus. For example, P. eryngii was distinguished from P. eous by the presence of Thr. Patil et al. found similar amino acid profiles in P. ostreatus, which strongly supports our results [52]. Previous reports also confirmed that cultivation materials of the genus Pleurotus have a significant effect on its content of amino acids; for example, P. citrinopileatus cultivated on paddy straw and other agrowaste combination has a higher content of amino acids than that on paddy straw alone [53].
3.6. Polar Lipid Analysis. Currently, only a few articles described the polar lipid of the genus Pleurotus, but most focused on its content and activity [54]. Polar lipid profiles of the genus Pleurotus were investigated for the first time in this study (Table 3). Both phosphatidylcholine (PC) and phosphatidyl-N-methylethanolamine (PME) were detected in all tested samples. And other types of polar lipid varied with the different species of the genus Pleurotus, such as diphosphatidylglycerol (DPG), phosphatidylglycerol (PG), phosphatidylserine (PS), phosphatidylethanolamine (PE), unidentified phospholipid (PL), unidentified aminolipid (AL), unidentified glycolipid (GL), and unidentified lipid (L). P. tuber-regium could be distinguished from others by the absence of DPG. Similar results also appeared in P. elongatipes, which was the lack of PE.
3.7. Quinone Analysis. So far, there were no correlated publications reporting quinone profiles of the genus Pleurotus. This study first investigated quinone profiles of the genus Pleurotus (Table 4). Our results showed that all tested samples contain MK-6 as the sole respiratory quinone, which was consistent with previous studies that menaquinone often exists in Gram-positive species with a high GC content [55]. As far as the genus Pleurotus is concerned, our results showed that there is no remarkable correlation between quinone types and Pleurotus species. Subsequently, closely related genera were also investigated to determine quinone types, such as Lentinellus, Hohenbuehelia, Resupinatus, and Phyllotopsis. Results revealed that quinone type is different from that of the genus Pleurotus (Table 4), which was used as chemical characteristic in distinguishing the genus Pleurotus from its closely related genus. Environmental factors such as nutritional components, oxygen, and temperature are known to affect lipid content and composition in living organisms, including fungi [49].
3.8. Mycolic Acid Analysis. Mycolic acid has been characterized in Mycolata taxon [56]. However, there are no previous reports that record mycolic acid distribution in the genus Pleurotus. In our studies, the distribution of mycolic acid in the genus Pleurotus was detected for the first time. Results showed that mycolic acid was detected in all tested samples, except in P. tuber-regium. It was suggested that P. tuberregium could be distinguished from others by using mycolic acid as a key indicator.

10
BioMed Research International proportion of a specific fatty acid when compared with others. The application of fatty acid composition data has now extended to studies of physiology, chemotaxonomy, and intrageneric differentiation, as well as human nutrition. But its content in mushrooms is extremely flexible and always influenced by environmental conditions such as media composition, pH, temperature, and growth stage [57].

Conclusion
Referring to these findings from this study, surveying the genetic variation through phylogenetic analysis, ISSR marker, and GC content could be useful in efficiently differentiating each Pleurotus species. The subsequent analysis of their chemical characteristics was consistent with the above results. The combined molecular and chemical analysis could provide a solid and reliable tool for Pleurotus classification and taxa delimitation.