Biological material

Three strains of Lentinula edodes (L5, L9, L15) and one of Pleurotus ostreatus (PAsp14) were studied. All strains produced in this study were stored at Laboratorio de Cultivos Celulares of UPIBI-IPN, Mexico and cultured in malt extract agar (MEA) (35 g L^-1).

Genomic DNA Extraction

Mushroom strains were cultivated in liquid media (20 g L^-1 sucrose and 20 g L^-1 malt extract), and DNA was extracted by the gDNA Plant Kit (#CS 18000).

Molecular identification of parental strains

The ITS1-5.8r-ITS2 region was amplified with the ITS1 and ITS4 primers ^{(White et al., 1990)}. The PCR reactions consisted in 22 µL PCR SuperMix (Cat. No. 10572-014, Invitrogen), ITS1 and ITS4 primers (0.25 µM), DNA (3050 ng μL^-1) and distilled water. PCR steps were: initial denaturation, 5 min at 94 °C; 35 cycles (denaturation, 60 s 94 °C; annealing primers, 30 s 55 °C; extension, 60 s 72 °C) and final extension 5 min at 72 °C. PCR amplicons were verified in a 1.2 % agarose gel, TAE, and stained with GelRed™. Gels were run at 100 V for 1 h. The agarose gel was observed with a UV Transilluminator (ChemiDOC ™MP Imaging, Biorad^® System, Hercules, California, EUA). The PCR products were sequenced from both sides at IBT, UNAM, Mexico.

The ITS1-5.8s-ITS2 sequences obtained were assembled and edited with AliView software ^{(Larsson, 2014)}, deposited into the NCBI platform and compared with sequences of Lentinula and Pleurotus from the GenBank using the BLAST-n. Sequences of distinct species of the same genera were downloaded (Table 1). Eight species of Pleurotus and four species of Lentinula were used in this analysis. The sequences were aligned with MAFFT ^{(Katoh et al., 2019)} with minor manual adjustments in AliView ^{(Larsson, 2014)}. The Mesquite software ^{(Maddison and Maddison, 2019)} was used to generate compatible files with Bayesian analysis (MrBayes) in CIPRES Science Gateway V 3.3 ^{(Miller et al., 2010)}. The parameters used were GTR+invgamma, 1 × 10⁶ generations, sampling every 1000 generations with a burning fraction of 0.25. The phylogenetic tree generated in CIPRES was visualized and edited in ITOL ^{(Letunic and Bork, 2019)}.

Chemical dedikaryotization

Fresh cultures on MEA plates with mycelium just reaching the plate edge were blended (Waring Blender, Stamford, Connecticut, USA) 5-30 s for L. edodes or 60-120 s for P. ostreatus. The homogenized solution (50 μL) was added to flasks containing 50 mL of dedikaryotization solution (DS; 30 g L^-1 Peptone P, Oxoid ™, and 20 g L^-1 glucose) and stored at 24 °C for 16 d for L. edodes and 2 d for P. ostreatus, respectively.

In the case of P. ostreatus strains, when the mycelium was visible in the DS, liquid cultures were blended for a second time (1 min), and 25 μL were used to inoculate on MEA plates, which were incubated at 24 °C until mycelial colonies were observed. The same procedure was followed with L. edodes strains but skipping homogenization for a second time ^{(Valenzuela et al., 2017)}. The MEA plates were observed daily to identify monokaryotic colonies (lacking clamp connections).

Classification of neohaplonts and production of hybrids

To recognize the compatibility types of each parental strain, a single neohaplont was paired with all the neohaplonts of the same strain. MEA plates were then stored at 24 °C and observed under the microscope (10X ) to identify the presence or absence of clamp connections; then, the dikaryotic strains were transferred into other MEA petri dishes to confirm the presence of clamp connections

ISSR profile

Seven ISSR primers were used to obtain ISSR profiles of parental, reconstituted hybrid strains and neohaplonts (Table 2). The PCR reactions contained 22.5 μL PCR SuperMix (Cat. No. 10572-014, Invitrogen), DNA (30-50 ng µL^-1), ISSR primer (20 μM) and distilled water to obtain a final volume of 25 μL. PCR steps were: initial denaturation, 5 min at 94 °C; 35 cycles of denaturation, 60 s at 94 °C; anneal primers, 60 s at 55 to 64 °C; extension, 3 min at 72 °C, and final extension, 10 min at 72 °C. Amplicons were separated in 1.5 % agarose electrophoresis gels using TAE and stained with GelRed™. Gels were run at 100 V for 90 min and visualized with a UV Transilluminator (ChemiDOC™ MP Imaging de Biorad^® System, Hercules, California, USA) ^{(Sánchez et al., 2019)}.

Statistical analysis

An χ² test was applied to determine the symmetry in the recovery of neohaplonts of parental strain using the SPSS program ^{(Valenzuela et al., 2017)}. Data generated from seven ISSR primers for all strains were entered into a presence/absence matrix. Euclidean distance was calculated from the matrix, and the R function (fviz_dend) was used to generate a dendrogram in R. To evaluate genetic profiles, marker parameters were recovered from electrophoresis gels: total bands (TB), monomorphic bands (MB) and polymorphic bands (PB) were used to calculate the percentage of polymorphism (PP). To evaluate primer efficacy, other marker parameters were calculated: effective multiplex ratio (EMR), resolving power (RP), marker index (MI) and polymorphic information content (PIC) ^{(Roldán-Ruiz et al., 2000}; ^{Varshney et al., 2007)}.

Production of wheat grain

Wheat grains were hydrated for 24 h; then, 200 g of hydrated wheat grains were placed in a plastic bag and sterilized for 2 h at 121 ºC and 1.05 kg cm^-2. Once the bags were cooled , each bag was inoculated with a fresh culture on an MEA plate (mycelium just reaching the plate edge) of the parental and hybrid strains. Inoculated bags were incubated for two weeks at 25 ºC.

Production of substrates

Two types of substrates were prepared. Substrate M1 for the fruiting of Lentinula strains contained 40 % white oak sawdust, 40 % cotton husk, 16 % wheat bran, 3 % CaCO₃ and 1 % CaSO₄ with a final moisture of 64.5%. Substrate M2 for fruiting of Pleurotus strains contained 40 % pepper tree sawdust, 29 % cotton husk, 29 % wheat straw, 9 % millet, 3 % CaCO₃ and 1 % CaSO₄ with a final moisture of 72 %. Hybrid strains were cultivated on both substrates. Bags were filled with 0.5 kg of each substrate and sterilized for 2 h at 121 ºC and 1.05 kgf cm^-2. After that, bags were inoculated with wheat grain spawn fully colonized by the mycelium of the corresponding parental or hybrid strains and incubated for 2 weeks at 25 ºC.

Production of fruiting bodies

Bags with the fully colonized substrates by the mushroom mycelium were transferred into the fructification room at 18-20 ºC and a relative humidity of 80-95 %. Mushrooms were weighted and biological efficiency was calculated [(g of moist mushrooms/g of dry substrate) × 100].

Recovery of meiotic progeny and classification into compatibility types

The spore print of fruiting bodies of L. edodes × P. ostreatus hybrids was collected by placing the fruiting body on filter paper for 24 h at room temperature and later preserved at 2ºC. A segment of 1 cm² of filter paper with spore print was cut and vortexed with 1 mL of distilled water in a tube. Serial dilutions were done (10^-1 and 10^-5) and 50 µL of each dilution was inoculated on EMA Petri dishes; once spores germinated, mycelia were transferred to new AEM Petri dishes and verified for their monokaryotic status. The monosporic colonies were classified into four compatibility types by cross-mating all monokaryotic isolates.

Molecular identification of parental strains

The amplification by PCR of the ITS1-5.8r-ITS2 region produced a 756 bp amplicon for L5, L9 and L15 strains, and 676 bp for PAsp14. The sequences were uploaded to the NCBI platform and accession numbers were generated: PAsp14 (KX683410), L5 (KX683413), L9 (KX683412) and L15 (KX683411).

The phylogenetic tree (Figure 1) grouped the strains of the two genera (Pleurotus and Lentinula) in different clusters. The sporeless strain PAsp14 of Pleurotus clustered with sequences from species of P. ostreatus. The strains of Lentinula (L5, L9 and L15) clustered with sequences of L. edodes. Different phylogenies based on analysis of the ITS1-5.8r-ITS2 region allowed the identification of strains studied.

Figure 1 Phylogenetic tree based on the Bayesian inference analysis of the ITS1-5.8r-ITS2 region from strains studied compared to selected sequences from the GenBank. 

Dedikaryotization of parental strains and production of hybrid strains

The recovery of neohaplonts (nh) from three strains of Lentinula (L5, L9 and L15) and a Pleurotus strain (PAsp14) was successful. Neohaplonts were recovered for all strains: L5 (10 nh), L9 (9 nh), L15 (9 nh) and PAsp14 (16 nh). Matings among neohaplonts of the same strain allowed the identification of the compatibility types (TI and TII, respectively). The χ² test determined that recovery of the strains L9 (five neohaplonts of TI and four neohaplonts of TII) and L15 (seven neohaplonts of TI and two neohaplonts of TII) were symmetric, while for strains L5 (nine neohaplonts of TI and one neohaplont of TII) and PAsp14 (two neohaplonts of TI and 14 neohaplonts of TII) were asymmetric. Chemical dedikaryotization was feasible for L. edodes using short homogenization time (5 to 20 s) and glucose and peptone concentrations between 20 and 30 g L^-1. The effect of homogenization represented an essential factor for the recovery of the neohaplonts.

^{Fukumasa-Nakai et al. (1994)} reported recovery of the neohaplonts of L. edodes by production of protoplasts, while ^{Valenzuela et al. (2017)} reported recovery of neohaplonts of L. edodes using chemical dedikaryotization with short homogenization times (5 s), the shortest times being more effective. In this research, the shortest homogenization times allowed low damage to mycelia; consequently, the monokaryotic components were recovered. To recover the monokaryotic components of strain PAsp14, homogenization for 1 min was necessary, while ^{Valencia and Leal-Lara (1999)} reported homogenization times of 2.5 min to recover neohaplonts of Pleurotus spp. strains.

To produce intergeneric hybrids of L. edodes and P. ostreatus strains, one neohaplont was randomly chosen of each compatible type of each L. edodes strain (L5, L9 and L15) and mated with all the neohaplonts of PAsp14 in all possible combinations. Twelve intergeneric hybrids were obtained (Table 3).

The production of intergeneric hybrids has been previously reported. ^{Valenzuela et al. (2017)} produced intergeneric hybrids by mating neohaplonts of L. edodes × P. ostreatus. ^{Ramírez-Carrillo et al. (2007)} reported the production of interspecific (P. eryngii × P. ostreatus) and intergeneric (P. eryngii × L. edodes) hybrids. Hybrid strains between P. florida × Volvariella volvacea ^{(Chakraborty and Sikdar, 2008)} and P. florida × Calocybe indica ^{(Chakraborty and Sikdar, 2010)}. Pleurotus florida × L. edodes were also obtained by protoplast fusion ^{(Mallick and Sikdar, 2014)}. Chemical dedikaryotization and protoplast fusion allow the production of hybrids, either intra- or interspecies; however, hybrid production by chemical dedikaryotization offers the advantage of recovering the monokaryotic components that constitute the parental strain. However, producing hybrids by protoplast fusion does not involve the recovery of neohaplonts; thus, molecular characterization is restricted only to parental and hybrid strains ^{(Chakraborty and Sikdar, 2008}; ^{Mallik and Sikdar, 2014)}.

By the aforementioned techniques, interspecies and intergeneric strains can be produced, and thus, some incompatibility barriers are broken among different species. This process cannot be achieved by standard mating of monosporic isolates ^{(Worrall, 1997)}. Remarkably, ^{Valenzuela et al. (2017)} reported similar compatibility types among neohaplonts of L. edodes and P. ostreatus, which suggests a considerable similarity in the compatibility system of these two species.

Interspecific hybridization in fungi has been reported in nature as a mechanism for DNA exchange where two genomes from different species are merged into a single, often highly unstable, polyploid genome that evolves rapidly into stabler derivatives, the resulting interspecies hybrids can show novel combinations of genes and phenotypic variation that enhance fitness and allow colonization of new niches ^{(Steensels et al., 2021)}. In mushrooms, genetic improvement is widespread in producing interspecies hybrids to generate strains with high yields or different phenotypes.

Molecular characterization with ISSR markers

Seven ISSR markers were used to characterize 10 strains (Table 2). Figure 2 shows electrophoresis gel from amplified products using the ISR11 marker. The bands by ISSR markers had reproducibility in triplicated PCR essays.

Figure 2 Polymorphism profile of neohaplonts and dikaryotic strains using the ISR 11 marker. Strains: PAsp14 (P), PAsp14R (PR), PAsp14-nh1a (Pn1), PAsp14-nh12a (Pn12), L9 (L), L9R (LR), L9-nh5a (Ln5), L9nh8a (Ln8), L9-nh5a×PAsp14nh12 (H₁), L9-h8a×Pasp14nh12a (H₂), Control (C), 100 bp ladder (100pb), 1kb plus ladder (1kb). 

Similar results have been published by ^{Mallick and Sikdar (2014)} while ^{Ramírez-Carrillo et al. (2007)} reported that intergeneric hybrids of Pleurotus spp. × L. edodes presented Pleurotus phenotype. ^{Mallik and Sikdar (2014)} generated a band profile with six ISSR markers of L. edodes × P. florida hybrids and their parental strains. ^{Chakraborty and Sikdar (2008)} mentioned that recombinant bands proved that the hybrid lines were not heterokaryons of the two parental nuclei; instead, the dikaryons fused after protoplast fusion to form a synkaryon followed by recombination without meiosis and can be present in a possible elimination of different chromosomes.

Figure 3 shows the dendrogram constructed from Euclidean distance. Strains were separated into two groups. The yellow cluster grouped the Lentinula strains L9 (parental), L9R (reconstituted) and neohaplonts of L9 (L9nh5a and L9nh8a), while the blue cluster grouped the Pleurotus strains PAsp14 (parental), PAsp14R (reconstituted), neohaplonts (PAsp14nh1a and PAsp14nh12a) and intergeneric hybrids (L9-nh5a × PAsp14nh12a and L9-nh8a × Pasp14-nh12a), the whole cluster (dikaryotic strains) expressed P. ostreatus phenotype, showing a molecular similarity based on ISSR markers.

Figure 3 Dendrogram constructed from euclidean distance. Lentinula strains: L9, L9R, L9-nh5a, L9-nh8a and Pleurotus and hybrid strains: PAsp14, PAsp14R, PAsp14-nh1a, PAsp14-nh12a, L9-nh5a×PAsp14-nh12 (Hybrid 5 × 12) y L9nh8a×PAsp14-nh12a (Hybrid 8 × 12). 

^{Morales and Dujon (2012)} reported that the hybrids of natural hybridization of different yeast species show post-hybridization events for genome stabilization. These events included a chromosomal rearrange and modification of genetic contributions from parentals by aneuploidization and partial loss of chromosomes. The elimination of chromosomes in one species can stimulate the phenotypic expression of the other species that retained more chromosomes. For this reason, the dominance of P. ostreatus phenotype may be related to a loss of genetic material of L. edodes caused by a chromosomal rearrangement for genetic stabilization induced by intergeneric hybrid production.

The number of amplified products (TB) with distinct markers ranged from 12 to 21 different bands (Table 2). The UBC 807 was the primer that produced the largest number of bands (21), while the marker with fewer bands was UBC 811 (12). The seven markers studied generated 114 bands/ amplicons with a mean of 16.29 by marker; 104 bands (91.31 %) were polymorphic (with an average of 14.86 PB by marker) and 10 bands (8.69 %) were monomorphic. Polymorphism oscillated from 73.33 % for ISR markers up to 100 % polymorphism for markers UBC811, ISR 15 and ISR 02. The PPB is linked to genetic variability. ^{Yin et al. (2014)} reported an average polymorphism with ISSR markers of 74 % for 15 P. pulmonarius strains in China. A higher PPB value was observed in this study, i.e. 91.31 %, indicating that a high genetic diversity was generated when neohaplonts from different species were mated.

The PIC values indicate that both states (presence/ absence) are in different proportions. In this research, the PIC average was 0.37, and ranged from 0.30 (ISR12) to 0.41 (UBC 807, UBC 811 and ISR 02). ^{Mallick and Sikdar (2014)} reported an average PIC of 0.475; however, ^{Zhang et al. (2015)} reported PIC values with SCoT markers ranging from 0.20 to 0.50, while other researchers ^{(Gorji et al., 2011}; ^{Kumar et al., 2014)} reported PIC values with ISSR markers ranging from 0.15 to 0.49. The EMR values show the number of non-polymorphic loci analyzed by marker systems ^{(Varshney et al., 2007)}. In this research, the EMR average was 8.05 per primer, the major value (10.60) was observed with ISR 15, and the minimum (5.50) with ISR02. ^{Badfar-Chaleshtori et al. (2012)} mentioned that MI supplies an approximation of polymorphism generated per essay for a certain primer and they obtained values from 0.019 to 4.7. In this research, an MI average of 2.94 was obtained, the lowest one was 2.60 for ISR 12, while the highest was 4.10 for UBC 807. The EMR and MI parameters were used to calculate the resolving power (RP) for each marker, indicating the discriminatory potential of the primers chosen ^{(Kumar et al., 2014)}. The highest value of RP was obtained by marker UBC 807 (14) and the lowest one by marker ISSR 15 (7.8). ^{Badfar-Chaleshtori et al. (2012)} and ^{Kumar et al. (2014)} reported values of genetic variation in the population from one species ranging from 0.1 to 8.56.

Fructification of hybrid strains of L. edodes × P. ostreatus and classification of a meiotic progeny

The intergeneric hybrids of P. ostreatus × L. edodes were fructified on two different substrates: M1 (Pleurotus commercial substrate) and M2 (Lentinula commercial substrate) (Table 3). All hybrid strains produced fruit bodies with a P. ostreatus phenotype (Figure 4) regardless of the substrate type, this agrees with previous studies reporting that P. ostreatus phenotype was predominantly expressed by Lentinula × Pleurotus hybrids ^{(Mallick and Sikdar, 2014}; ^{Ramírez-Carrillo et al., 2007}; ^{Valenzuela et al., 2017)}.

Figure 4 Phenotipical diversity of parental (A-B) and hybrid strains (C-D). 

Noteworthy, although biological efficiencies produced by hybrid strains in this study varied widely depending on the type of substrate, similar biological efficiencies on both types of substrates were the pattern most frequently observed (Table 3). The biological efficiency ranged from 50 to 60 % on both substrates (M1 and M2) by hybrids L5-4 × P-1, L9-5 × P-1, L9-5 × P-12, L9-8 × P-1 and L15-48 × P-12. By contrast, the hybrid L5-4 × P-12 produced the highest yield on the M1 substrate (87.85 ± 5.70) among all hybrids, which was markedly higher than the biological efficiency produced on the M2 substrate by this strain (28.67 ± 7.3). On the other hand, the hybrid L15-8 × P-1 produced the highest yield on the M2 substrate (96.55 ± 10.40) among all hybrids, which was markedly higher than the biological efficiency produced on the M1 substrate by this strain (34.18 ± 3.91). This variation of biological efficiency depending on substrate composition may be related to the prevalence of certain genes due to genetic stabilization after the hybridization process, as ^{Morales and Dujon (2012)} observed with the natural hybridization of different yeast species.

Noteworthy, the analysis of the compatibility types in the progenies of interspecies hybrid strains of edible mushrooms has not been reported in previous studies. One hybrid strain (L9-5 × P-12) was selected randomly to classify its meiotic progeny. Monosporic strains were classified by mating tests in four compatibility types. Remarkably, this monokaryotic progeny derives from an intergeneric hybrid where both parental strains have a heterothallic tetrapolar compatibility system. The resulting meiotic progeny presented the same heterothallic tetrapolar compatibility system; 22 monosporic strains were recovered and classified (6 for T I, 3 for T II, 8 for T III, and 5 for T IV). Compatibility types I and II were identified as the parental types, corresponding to parental neohaplonts L9-5 and P-12, respectively, while III and IV were the recombinants. Noticeably, the mating capability in the resulting progeny was maintained as a heterothallic tetrapolar system, which is required for producing fruiting bodies.

This represents an important contribution to this type of studies since ^{Gallone et al. (2019)} and ^{Morales and Dujon (2012)} reported that interspecies hybridization with yeasts is a frequent event, but meiotic fertility is not so common; however, the interspecies hybridization in Basidiomycota has been reported in multiple articles. ^{Chen et al. (2016)} studied inter- and intra-specific diversity in Agaricus endoxanthus and closer species, which revealed a new taxon, A. punjabensis, since the ITS sequences from A. endoxanthus were highly variable but were separated from A. punjabensis. In addition, numerous heteromorphisms and allelic distribution at certain loci indicate that hybridization and recombination have likely occurred throughout the history of certain strains. Moreover, ^{Orihara et al. (2021)} reported a robust multilocus phylogenetic analysis of the genus Octaviania (Boletaceae) suggests that heterogenous sequences are probably the result of previous inter- and intra-specific hybridization and may have promoted the high genetic and species diversity found within the genus.