Cultivation studies of edible ectomycorrhizal mushrooms: successful establishment of ectomycorrhizal associations in vitro and efficient production of fruiting bodies

Abstract

Most edible ectomycorrhizal mushrooms are harvested in forests or controlled tree plantations; examples include truffles, chanterelles, porcinis, saffron milk caps, and matsutake. This study explored recent advances in in vitro ectomycorrhizal cultivation of chanterelles and matsutakes for successful ectomycorrhizal seedling establishment and the subsequent manipulation of these seedlings for efficient fruiting body production. Chanterelle cultivation studies have been limited due to the difficulty of establishing pure cultures. However, once pure cultures were established in the Japanese yellow chanterelle (Cantharellus anzutake), its ectomycorrhizal manipulation produced fruiting bodies under controlled laboratory conditions. As C. anzutake strains have fruited repeatedly under ectomycorrhizal symbiosis with pine and oak seedlings, mating tests for the cross breeding are ongoing issues. As one of the established strains C-23 has full-genome sequence, its application for various type of ectomycorrhizal studies is also expected. By contrast, Tricholoma matsutake fruiting bodies have not yet been produced under controlled conditions, despite successful establishment of ectomycorrhizal seedlings. At present, the shiro structure of ≈1L in volume can be provided in two y incubation with pine hosts under controlled environmental conditions. Therefore, further studies that provides larger shiro on the host root system are desired for the outplantation trial and fruiting.

1. Introduction

Edible ectomycorrhizal mushrooms such as truffles (Tuber), chanterelles (Cantharellus cibarius Fr. and related species), porcinis (Boletus edulis Bull. and related species), saffron milk caps (Lactarius deliciosus [L.] Gray and related species), and matsutakes (Tricholoma matsutake [S. Ito & S. Imai] Singer and related species) have high economic value worldwide (Zambonelli & Bonito, 2012; Pérez-Moreno, Guerin-Laguette, Arzú, & Yu, 2020), with annual global sales estimated at several billion dollars. These mushrooms are harvested in forests or plantations for both domestic and export markets (Arora, 2008; Tsing, 2015, Pérez-Moreno et al., 2021). In Japan, imported matsutake mushrooms from countries such as China, USA, Canada, Morocco, Turkey, Mexico, and South Korea are valued at 80 to 100 million dollars annually (Aoki et al., 2022), which is equivalent to domestic T. matsutake production within Japan, mainly in Nagano and Iwate Prefectures (Yamanaka, Yamada, & Furukawa, 2020). During the past 20 y, truffle imports to Japan have increased such that their economic value in Japan now equivalent to around 10% that of the matsutake mushroom import value (MAFF; https://www.maff.go.jp/e/index.html).

To meet this increasing market demand, truffles have been harvested in tree plantations (Zambonelli, Iotti, & Murat, 2016). However, other edible mushrooms are largely harvested in forests because non-host and tree plantation cultivation techniques have not yet been established. Therefore, these mushrooms must be conserved in natural forests through controlled harvests or forest management to sustain moderate harvest levels for longer (decade-scale) periods (Guerin-Laguette, 2021; Hosford, Pilz, Molina, & Amaranthus, 1997; Pilz et al., 1999, Pilz & Molina, 2002; Furukawa, Masuno, & Takeuchi, 2016, Yamada, Furukawa, & Yamanaka, 2017, Yamanaka et al., 2020). In the Périgord truffle Tuber melanosporum, most harvests are derived from tree plantations. In their native European range, as well as in Australia, New Zealand, and North America, several truffle species such as Tuber aestivum (Wulfen) Spreng., Tuber borchii Vittad., Tuber brumale Vittad., and Tuber lyonii Butters are harvested in tree plantations established using mycorrhizal seedlings that were grown under greenhouse or laboratory conditions. These mycorrhizal seedlings are usually prepared through spore inoculation to non-mycorrhizal host seedlings. Several years after truffle-associated seedling outplantation, truffle mycelia that survive and adapt to host tree rhizospheres are able to produce fruiting bodies (Zambonelli et al., 2016; Guerin-Laguette, 2021). Other groups cultivated in tree plantations include porcinis, saffron milk caps, Suillus (e.g., Suillus bovinus [L.] Lam. and Suillus luteus [L.] Roussel), Tricholoma (e.g., Tricholoma portentosum [Fr.] Quél.), desert truffles (Terfezia), Lyophyllum shimeji Hongo, Rhizopogon roseolus (Corda) Th. Fr., and Astraeus (Wang & Chen, 2014; Yamada et al., 2017; Suwannasai, Dokmai, Yamada, Watling, & Phosri, 2020; Pérez-Moreno et al., 2020; Yamanaka et al., 2020; Guerin-Laguette, 2021). Most other edible mycorrhizal taxa including Caesar's mushrooms, chanterelles, and matsutakes are harvested in managed natural or semi-natural (plantation or secondary) forests because techniques for mycorrhizal seedling have yet to be established for these taxa (Pilz, Norvell, Danell, & Molina, 2003; Yamada, Ogura, & Ohmasa, 2001b, c; Endo, Gisusi, Fukuda, & Yamada, 2013; Endo et al., 2014; Yamanaka et al., 2020; Guerin-Laguette, 2021).

Yellow chanterelles have the highest economic value among edible mycorrhizal mushrooms globally, estimated at over 200,000 metric tons and 1.25 to 1.4 billion dollars wholesale annually (Pilz et al., 2003). Pure culture techniques for the representative European species, Cantharellus cibarius, were established by the early 1990s (Danell & Fries, 1990), leading to successful chanterelle fruiting from established mycorrhizal seedlings under greenhouse conditions (Danell & Camacho, 1997; Pilz et al., 2003). However, the outplantation of mycorrhizal seedlings for chanterelle production has been poorly studied. Pure culture of another European chanterelle, Cantharellus pallens Pilát, was reported in a single study (Danell & Fries, 1990). In the Japanese yellow chanterelle, Cantharellus anzutake W. Ogawa, N. Endo, M. Fukuda & A. Yamada, which is phylogenetically closely related to C. cibarius, successful pure culture establishment (Ogawa et al., 2019a), ectomycorrhization, and fruiting through mycorrhizal symbiosis in the laboratory (Ogawa et al., 2019b) have been reported. Other economically important chanterelle species in Europe (e.g., Cantharellus ferruginascens P. D. Orton and C. amethysteus [Quél.] Sacc.), North America (e.g., C. roseocanus [Redhead, Norvell & Danell] Redhead, Norvell & Moncalvo, C. formosus Corner, C. californicus D. Arora & Dunham, C. cascadensis Dunham, O'Dell & R. Molina, and C. enelensis Voitk, Thorn, Lebeuf & J.I. Kim), and China (e.g., C. yunnanensis W. F. Chiu) remain to be isolated, and very few cultivation trials have been reported. Many matsutake mushroom cultivation trials have been reported in Japan since the 1960s (Matsutake Research Association, 1964; Ogawa, 1978; Yamanaka et al., 2020); however, ectomycorrhizal synthesis of matsutake mushrooms was first validated only in the late 20th century (Yamada, Maeda, & Ohmasa, 1999b; Gill, Lapeyrie, Gomi, & Suzuki, 1999), followed by wide acceptance of the technique (Guerin-Laguette, Shindo, Matsushita, Suzuki, & Lapeyrie, 2004; Kobayashi, Watahiki, Kuramochi, Onose, & Yamada, 2007; Yamada et al. 2010; Vaario, Pennanen, Sarjala, Savonen, & Heinonsalo, 2010; Yamanaka et al., 2014, Jeon & Ka, 2016; Herrera, Wang, Zhang, & Yu, 2022) and matsutake fruiting under controlled environmental conditions with or without a host remains to be validated. Recently, fruiting of the matsutake species Tricholoma bakamatsutake Hongo on nutrient medium without a host plant was reported by a private company in Japan, which intended to cultivate T. bakamatsutake commercially. The cultivation of other economically important matsutake species such as Tricholoma murrillianum, Tricholoma magnivelare, Tricholoma mesoamericanum, Tricholoma anatolicum, and Tricholoma fulvocastaneum via mycorrhizal synthesis also remains poorly studied. Therefore, in this review, I examine recent advances in C. anzutake and T. matsutake cultivation, with a particular focus on host-plant associations in chanterelle and matsutake mushroom production, and discuss future directions in the cultivation of these economically important edible mushrooms and in related mycology fields.

2. Chanterelle cultivation

2.1. Foundations of chanterelle cultivation

Chanterelles consist of fungi in the family Cantharellaceae (Moncalvo et al., 2006; Hibbett et al., 2014), which is among the most ancestral clades of ectomycorrhizal basidiomycetous fungi (Miyauchi et al., 2020). Yellow chanterelles (Cantharellus) and other related groups such as Craterellus (e.g., Craterellus tubaeformis [Fr.] Quél. and Craterellus cornucopioides [L.] Pers.) have common fragrances that are similar to that of apricots, which makes them a delicacy among consumers. Their unique funnel shape and diverse colors (yellow, white, reddish, green, or black) also stimulate the senses.

The first report of pure chanterelle culture establishment described yellow colonies (representative strain 740b) obtained using a spore isolation technique; these colonies had a similar fragrance to the fruiting bodies (Fries, 1979). Strain 740b was identified as true C. cibarius using a DNA hybridization technique (Straatsma, Konings, & van Griensven, 1985). Danell and Fries (1990) reported 56 cultured C. cibarius and four C. pallens strains isolated from fruiting body tissues collected in coniferous forests. Based on established cultures, in vitro ectomycorrhization of C. cibarius was first achieved by Moore, Jansen, and van Griensven (1989). Ectomycorrhizal synthesis of this fungus with pine and spruce hosts was conducted to obtain mycorrhizal seedlings in a nursery study (Danell, 1994), producing several fruiting events under greenhouse conditions over a period of several months (Danell & Camacho, 1997; Pilz et al., 2003). However, later reports of pure culture, mycorrhization, and artificial fruiting of this fungal species are rare. The Japanese yellow chanterelle C. anzutake was previously identified as C. cibarius (Ogawa, Endo, Fukuda, & Yamada, 2018); pure cultures of this fungus were isolated from basidioma tissues and ectomycorrhizal root tips (Ogawa et al., 2019a). The established strains showed characteristics similar to those of C. cibarius, i.e., yellowish color and an apricot-like smell (Fig. 1A-D). Ectomycorrhizae of several C. anzutake strains were synthesized on a Pinus densiflora Siebold et Zucc. host in vitro, which were subsequently used as mother plants for the preparation of another ectomycorrhizal system on a Quercus serrata Murray host (Ogawa et al., 2019b). Under laboratory conditions, pine and oak seedlings grown in 4 L pots hosted 29 fruiting events within 2 y of observation (Fig. 1E-H). In North American yellow chanterelles, no cultivation based on pure culture techniques has been reported; most studies of these mushrooms have focused on their taxonomy and ecology, as well as forest management for their sustainable harvest and conservation (Pilz, Molina, & Mayo, 2006; Dunham, Kretzer, & Pfrender, 2003a; Dunham, O'Dell, & Molina, 2003b; Thorn, Kim, Lebeuf, & Voitk, 2017). The cultivation of other chanterelle taxa such as Craterellus has rarely been studied.

Fig. 1-Cantharellus anzutake. A: A colony of strain C-18 on MNC agar plate, which was isolated from a mycorrhizal root tip. B: Hyphae of strain C-18 showing clamp connection and intracellular oily droplet. C, D: Colonies of stains EN-60 and EN61 isolated from mycorrhizal root tips. E: Fruiting of strain EN-61 associated with Pinus densiflora host in a 250 mL pot, photographed Oct 2012. F: Fruiting of strain EN-51 associated with Quercus serrata host in a 4 L pot, photographed May 2015. Please note the deposited basidiospores on the soil surface, which were released form the basidioma. G, H: Fruiting of strain EN-61 associated with Pinus densiflora host in a 4 L pot, photographed May 2015. I, J: Fruiting of strain C-23 associated with P. densiflora host in a 4 L pot, photographed Aug 2021. In the details of experimental systems of A-D and E-H, please see Ogawa et al. (2019a) and Ogawa et al. (2019b), respectively. Bar: 1 cm.

2.2. Chanterelle cultivation bottleneck

The limited reports of chanterelle cultivation have highlighted the difficulty of obtaining pure cultures, compared to other edible ectomycorrhizal mushrooms such as those of the suilloid group, Tricholoma, Lactarius, and Boletus. Although C. cibarius mycelia grow moderately on nutrient media, success rates for pure culture establishment from basidiomata remain low (Straatsma et al., 1985; Danell & Fries, 1990) because basidioma tissue from C. cibarius harbors diverse bacteria and fungi (Danell, Alström, & Ternström, 1993; Danell, 1999; Rangel-Castro, Danell, & Pfeffer, 2002) that stunt its mycelial growth on nutrient agar medium. The same phenomenon has been observed in C. anzutake and probably occurs in other chanterelles and Craterellus species. Cantharellus anzutake rarely produces pure cultures from basidiomata (Ogawa et al., 2019a); even C. anzutake cultured from basidiomata under laboratory conditions (Ogawa et al., 2019b) exhibit bacterial and fungal contamination during tissue isolation procedures (unpublished data), as has been reported in C. cibarius (Pilz et al., 2003). However, multispore isolation techniques developed for C. cibarius by Fries (1979) have shown that young, fresh basidioma can produce pure cultures. In C. anzutake, pure cultures have been successfully obtained from field-sampled ectomycorrhizal root tips (Ogawa et al., 2019a). Cantharellus cibarius cultures are not presently available in any culture collections, although C. anzutake cultures are available in the National Institute of Technology and Evaluation Biological Resource Center (NBRC), Japan. The whole-genome sequence of C. anzutake strain C-23 (NBRC no. 113265) has been published on the Joint Genome Institute (JGI) Genome Portal (https://genome.jgi.doe.gov/portal/), and its genome structure has been analyzed as representative among Cantharellales species (Miyauchi et al., 2020). The taxonomic validity of cultured yellow chanterelles is easily verified by sequencing of DNA barcoding regions such as the internal transcribed spacer 2 (ITS2) region of the rDNA operon and tef-1 region through BLAST searching. Many contaminants of C. cibarius, C. anzutake, and likely other yellow chanterelles can be detected on nutrient agars. Some chanterelle cultures have developed fluffy aeration hyphae that fill the agar plate, suggesting contaminations. Yellow chanterelle mycelia typically exhibit clamp connections among dikaryotic hyphae and oily intrahyphal droplets with or without yellow pigments (Ogawa et al., 2019a; Fig. 1A-D).

2.3. Recent advances in chanterelle cultivation

Early C. anzutake chanterelle cultivation trials identified this fungus as C. cibarius Fr. (Ogawa et al., 2018). Pure cultures of European C. cibarius were established from basidioma tissue and basidiospores (Fries, 1979; Straatsma et al., 1985; Danell & Fries 1990). Therefore, Ogawa et al. (2019a) adopted a new approach to obtain pure cultures of Japanese “C. cibarius” (= C. anzutake) from ectomycorrhizal root tips. This technique has been successfully applied in diverse fungal taxa (Yamada & Katsuya, 1995; Yamada, Ogura Degawa, & Ohmasa, 2001a; Endo et al., 2013), and has produced larger quantities of pure C. anzutake culture than isolation from basidioma tissue, which resulted in the establishment of multiple cultures per year. The C. anzutake strain C-23 was even isolated from a Quercus crispula mycorrhizal root tip sampled from beneath epigeous basidiomata that were wounded by mycophagous insects.

Following the acquisition of many C. anzutake strains, Ogawa et al. (2019b) synthesized ectomycorrhizas of this fungus in vitro with pine hosts, and acclimated the established ectomycorrhizal seedlings to small (250 mL) pots; successful fruiting of the fungus led to the isolation of strains EN-51 (NBRC no. 113266) and EN-61 (no. 113270) (Fig. 1E). Fruiting events occurred within approximately 1 year following fungal inoculation to the small seedlings in vitro, as previously reported for Rhizopogon roseolus (Yamada et al., 2001b, c). Although potted ectomycorrhizal seedlings grow better in organic soil than in mineral soil, fruiting was observed only in pots containing mineral soil (Ogawa et al., 2019b). Next, mushroom production was upscaled by successively transplanting the seedlings into culture pots containing 1 and 4 L mineral soil. Within a 2-y incubation period, the four tested strains (EN-51, En-61, C-2, and EN-98) grown with pine and oak hosts in 4 L pots fruited a total of 29 times (Fig. 1F-H). Such repeated fruiting demonstrates the potential for successful C. anzutake cultivation using this new technique. Further experiments were conducted by adding strains and increasing pot replication using pine and oak hosts. By late 2021, these cultivation trials recorded more than 200 C. anzutake fruiting events (unpublished data), among which C-23 was the most fertile tested strain (Fig. 1I, J). Therefore, single-spore isolates derived from strain C-23 should be tested to determine their mating types and to select next-generation populations for crossbreeding in C. anzutake. Therefore, I recommend strain C-23 as a candidate model fungus for ectomycorrhizal symbiosis and edible mycorrhizal mushroom production.

Cantharellus anzutake is phylogenetically closely related to C. cibarius (Ogawa et al. 2018). In Asia another edible yellow chanterelle, Cantharellus yunnanensis W. F. Chiu, had been included in C. cibarius s.l. (Pilz et al., 2003). Therefore, I inferred that a Chinese yellow chanterelle sold under the name of C. yunnanensis is conspecific to C. anzutake, because both species were externally quire similar, and the several known ITS2 sequences under the name of C. yunnanensis on GenBank showed high homology to that of C. anzutake. In addition, the habitats of C. anzutake and C. yunnanensis shared the same biome, i.e., the Asian temperate broadleaf and mixed forests. However, a recent taxonomic study that designated the epitype of C. yunnanensis clearly distinguished these two Asian yellow chanterelles based on morphological and phylogenetic (tef-1α) analyses (Shao, Liu, Wei, & Herrera, 2021). The high degree of phylogenetic relatedness of C. anzutake and C. yunnanensis allows them to be misidentified as conspecific (Cao, Hu, Yu, Wei, & Yuan, 2021) even in phylogenetic analyses using the large subunit (LSU) and ITS2 rDNA regions and indicates recent speciation. Thus, accumulated knowledge about C. anzutake cultivation can be applied to C. yunnanensis.

2.4 Outdoor cultivation of yellow chanterelles

Once ectomycorrhizal seedlings have been established in the laboratory, they can be outplanted for outdoor mycorrhizal cultivation. Guerin-Laguette et al. (2014) successfully cultivated Lactarius deliciosus in New Zealand for several years following outplantation of mycorrhizal seedlings. Although similar practices have been reported for various fungal taxa, few have been commercially viable, with the notable exception of truffles. In the truffles, commercial harvests have been conducted in Europe, Australia, New Zealand, and North America, in plantations established by outplanting of mycorrhizal seedlings (e.g., Zambonelli et al., 2016). Although 600 Scots pine seedlings inoculated with C. cibarius were outplanted in 24 locations in southern Sweden, the long-term results have not been reported (Pilz et al., 2003).

At the KOA Corporation field near the Faculty of Agriculture of Shinshu University, C. anzutake cultivation trials have been conducted since 2014, comparing forest edge (N = 12) and inner forest plots (N = 12). In 2020, outplanted pine seedlings inoculated with several strains of C. anzutake (EN-51, EN-52 [NBRC no. 113267], EN-53 [no. 113268], EN-60 [no. 113269], EN-61, and C-2) grew to a height of 2-3 m at the forest edge; tree root systems showed increased C. anzutake biomass, as identified through microscopy and ITS sequencing (unpublished data). However, no fruiting was observed. Growing basidiomycetous ectomycorrhizal biomass is anticipated to fruit in response to induction signals, as shown in saffron milk cap and porcini mushrooms (De la Varga et al., 2013; Guerin-Laguette et al., 2014). If fruiting events are observed among the inoculated C. anzutake strains at the experimental site, large-scale plantations will likely be established, as occurred for C. cibarius in Sweden (Pilz et al., 2003).

3. Matsutake cultivation

3.1. In vitro ectomycorrhization of Tricholoma matsutake for mushroom cultivation

The first trial of ectomycorrhization of Tricholoma matsutake with a pine host was conducted by Masui (1927); however, that experiment was later deemed invalid because its microscopic observations of inoculated hyphal cells suggested Umbelopsis species (or “Mortierella”; Ogawa, 1978), which is a common contaminant during T. matsutake tissue isolation, rather than T. matsutake itself (Yamada et al., 2001a). Hamada (1974) recalled having first established T. matsutake cultures in 1940; these exhibited slow growth rates and white to pale cream colonies on nutrient agar medium. This description is adequate for the current identification of T. matsutake colonies through DNA barcoding. The taxonomic identity and ectomycorrhizal ability of strain NBRC no. 6933 (Institute for Fermentation, Osaka [IFO] no. 6933), which was originally established in 1952 (Ogawa & Hamada, 1975), was recently examined (Yamada, Kobayashi, & Murata, 2003). In the 1950s and 1960s, various approaches for increasing or recovering T. matsutake harvests in pine forests have been tested in Japan (Matsutake Research Association, 1964; Ogawa, 1978; Hosford et al., 1997), including in vitro ectomycorrhizal synthesis. However, very few data from these trials were reported. Wang, Hall, & Evans (1997) tested in vitro ectomycorrhization of T. matsutake with pine hosts and discussed their unique plant-fungus interactions, e.g., matsutake is involved in a symbiosis-saprobiosis-pathogenesis continuum. During this time, T. matsutake mycorrhizae were considered atypical ectomycorrhizae in situ, due to the absence of Hartig net development at the pine root cortex (Ogawa, 1978; Hosford et al., 1997), although this finding was technically incorrect according to the microscopy results. However, Yamada, Kanekawa, & Ohmasa (1999a) reported continuous Hartig net development, which was confirmed shortly thereafter by another research group (Gill et al., 1999; Gill, Guerin-Laguette, Lapeyrie, & Suzuki, 2000). Yamada et al. (1999b) conducted in vitro mycorrhizal synthesis using T. matsutake and Pinus densiflora and demonstrated ectomycorrhizal development, as was later confirmed (Gill et al., 2000; Guerin-Laguette et al., 2000; Vaario, Guerin-Laguette, Gill, Lapeyrie, & Suzuki, 2000). The understanding how we characterize T. matsutake mycorrhizae has been revolutionized within the past 30 y as described above, leading to the adoption of the concept of ectomycorrhizal symbiosis in mycorrhizal studies of this fungus. The establishment and outplantation of ectomycorrhizal seedlings has subsequently been conducted in matsutake cultivation research (Yamada, Maeda, Kobayashi, & Murata, 2006; Kobayashi et al., 2007; Yamanaka et al., 2014, 2020).

The basic biology of matsutake has also been clarified. T. matsutake is distributed in both Asia and Europe (Bergius & Danell, 2000; Matsushita et al., 2005), and extant T. matsutake associations with coniferous hosts have been found to have evolved from an ancestral fungus that may have been associated with oak hosts, based on retrotransposon DNA analyses of matsutake genomes (Murata et al., 2013b). In situ T. matsutake colonies can form fairy rings comprising basidiomata in a “shiro” structure including several genets (Murata, Ohta, Yamada, Narimatsu, & Futamura, 2005; Lian, Narimatsu, Nara, & Hogetsu, 2006). This is because basidiospores of T. matsutake contribute to subsequent generations within the present geographic range (Amend, Garbelotto, Fang, & Keeley, 2010). Tricholoma matsutake associates with oaks as well as pines at the foot of the Tibetan Plateau in China (Yamanaka, Aimi, Wan, Cao, & Chen, 2011), and its inoculation promotes the growth of pine seedlings in vitro (Guerin-Laguette et al., 2004). Cultured T. matsutake mycelia colonize arbuscular mycorrhizal plants, forming “root endophytes” in vitro (Murata et al., 2013a, 2014a, b). All of these findings have informed later cultivation studies.

3.2. Mycorrhizal seedlings as a Tricholoma matsutake cultivation bottleneck

In vitro ectomycorrhizal synthesis of T. matsutake with a pine host requires approximately 6 mo from inoculum preparation (cultured mycelium) and seed germination to data sampling of mycorrhizal properties, i.e, morphology and anatomy, and the quantitative data of fungal colonization and the host growth (Yamada et al., 1999b, 2006). Once the inoculated mycelium has attached to the lateral root surface of the host pine, Hartig net hyphae develop at the root cortex within only a few weeks (Vaario et al., 2000). However, using large culture vessels (e.g., 1L) for large mycorrhizal seedlings increases the total incubation period to 1 y or longer. As T. matsutake mycelia grow slowly (1-2 cm/mo), they require longer periods than chanterelles to expand through the soil and develop ectomycorrhizae. Throughout the incubation period, light, temperature, soil water content, and ambient CO₂ and O₂ conditions should be controlled to prevent microbial contamination, which can influence fungal and plant growth. Soil nutrient composition and physiochemistry (Saito et al., 2018), as well as genetic properties of the matsutake strains (Yamada et al., 2010) and host plants (Yamada, Endo, Murata, Ohta, & Fukuda, 2014) can affect experimental outcomes. No studies have yet clarified the dominant nitrogen form adsorbed by T. matsutake from soil in situ (Vaario, Sah, Norisada, Narimatsu, & Matsushita, 2019; Yamanaka et al., 2020), although adding dried yeast to soil has been highly effective as a nitrogen source for mycorrhizal synthesis in vitro (Yamada et al., 2006). To optimize these factors for further ectomycorrhizal synthesis experiments, relevant soil, fungal strain, and plant cell line (somatic plant strain) standards must be determined for the study objectives.

It remains difficult to acclimate T. matsutake mycorrhizae synthesized in vitro to open and non-axenic conditions. In many ectomycorrhizal fungal taxa such as Laccaria, Suillus, Rhizopogon, Hebeloma, Paxillus, Pisolithus, Lyophyllum, Amanita, Lactarius, Boletus, and Cantharellus, ectomycorrhizal seedlings synthesized in vitro are easily acclimated to non-axenic pot soil conditions (Cairney & Champers, 1999; Yamada et al., 2001b, c; Endo et al., 2013, 2014; Ogawa et al., 2019b; Guerin-Laguette, 2021). Even Tricholoma species such as T. portentosum, T. flavovirens, T. saponaceum, and T. terreum show similar properties (Yamada et al., 2001b, c; Yamada, Kobayashi, Ogura, & Fukada, 2007). However, T. matsutake ectomycorrhizal root tips disappear easily from the host root system when the shiro structure, i.e., mycelium-soil aggregates associated with plant host roots, are damaged upon seedling transplantation from an in vitro system to pots containing soil. The effects on T. matsutake macroscopic structure in mycorrhizal acclimation require careful manipulation during the establishment of large mycorrhizal seedlings for outplanting. The instability of T. matsutake on ectomycorrhizal root tips during seedling acclimation to open pot soil remains unexplained enough. Previous studies have reported that T. matsutake has significantly smaller shoot/root ratios in vitro than R. roseolus or C. anzutake (Yamada et al., 2010; Ogawa et al., 2019b), which suggests that the colonized fungus imposes a cost on host plant growth. It is generally thought that late-stage ectomycorrhizal fungi represent a cost to symbiont seedlings during forest succession (Deacon & Fleming, 1992; Smith & Read, 2008). Therefore, studies on ectomycorrhizal synthesis for the ecological group are limited, and mycorrhizal acclimations for those fungi are not always easy (e.g., Cairney & Chambers, 1999). In a well-known late-stage fungus, Boletus edulis, however, ectomycorrhizae synthesized in vitro can be acclimated in open pot soil even by washing the root system prior to transplantation (Endo et al., 2014). These experimental data highlight that T. matsutake has unique mechanisms associated with shiro structures that promote their survival and growth in situ.

Due to the difficulty of manipulating fungus-plant associations in T. matsutake, field trials have produced only limited or preliminary data. Outplanted pine seedlings synthesizing T. matsutake ectomycorrhizae in vitro in a soil volume of 1 L were found to sustain mycorrhizal status and shiro structures for 2 y (Kobayashi, Terasaki, & Yamada, 2015). However, the shiro did not expand in the area, and extended pine roots were colonized by native ectomycorrhizal fungi such as suilloids. Direct inoculation of cultured T. matsutake mycelia to non-ectomycorrhizal roots of adult pine trees in situ led to successful ectomycorrhization, but mycorrhizal status did not persist beyond 1 y (Guerin-Laguette, Matsushita, Lapeyrie, Shindo, & Suzuki, 2005).

3.3. Role of shiro structures in Tricholoma matsutake cultivation

The term shiro originated from the soil area (meaning probably in the color or the territory in Japanese) in pine forests where T. matsutake mycelia inhabit (occupy) and from which basidiomata of this fungus occur (Hamada, 1974; Ogawa, 1978). A small fairy ring of epigeous T. matsutake basidiomata ranging a few meters in diam is thought to be provided from the margin of a single shiro structure (single mycelial colony) (Hamada, 1953; Murata et al., 2005; Lian et al., 2006). Thus, shiro structures and fairy rings are often considered equivalent (e.g., Narimatsu et al., 2015); however, T. matsutake shiro structures consist of macroscopic (visible) mycelium-soil aggregates that develop from ectomycorrhizal root tips. Shiro mycelium is nutritionally supported by the host plant via connected ectomycorrhizal root tips; thus, even if a shiro structure does not produce a fruiting body, it can survive and grow in pine forest soil or in experimental pots for long periods, such that it is distinct in the meaning from a fairy ring.

Shiro research has taken a soil microbiological perspective because the dilution plating of shiro soil has led to the discovery of unique microbial arrays, with very low microbial detection, particularly beneath the basidioma stipe base, compared to soils sampled outside of the shiro (Ohara, 1966; Ohara & Hamada, 1967; Ogawa, 1978). Shiro antibiotic activity against soil microbes has been inferred from halo formations observed in soil bacteria from shiro soils spread on agar nutrient plates (Ohara, 1966). Although the nature of these antibiotic effects remain to be elucidated completely, Nishino et al. (2017) clarified that the main compound involved is aluminum oxalate, which was detected in T. matsutake shiro, and that this compound produced distinct halos in soil bacteria under natural concentrations. This insoluble aluminum compound has been suggested to be synthesized through a reaction of oxalate exuded from T. matsutake hyphae and soil aluminum phosphate, followed by absorption of the released phosphates by T. matsutake hyphae. This aluminum oxalate formation process provides yet another advantage in the nutrient-poor, acidic granite soils preferred by T. matsutake. Dissoluble aluminum ions, which are toxic to various organisms including plants, released from aluminum hydroxide under acid conditions are also converted into aluminum oxalate. Therefore, it is really the “killing three birds with one stone” effect (Hirai & Nishino, 2019). Due to this complicated chemical system, the shiro structure acts as the functional unit conferring ecological fitness to T. matsutake for in situ survival. These phenomena are also studied by Vaario et al. (2015) from the perspective of the effects of minerals produced by rock weathering on ectomycorrhizal fungi (Landeweert, Hoffland, Finlay, Kuyper, & van Breemen, 2001; Hoffland et al., 2004).

Single T. matsutake shiro structures can consist of several individuals (genets) of vegetative mycelia in situ (Murata et al., 2005; Lian et al., 2006). Since shiro formations are mainly studied in terms of epigeous basidiomata along with limited data available about shiro mycelia in the soil (Lian et al., 2006; Horimai et al., 2021), more comprehensive study is required for the structure of genetic heterogeneity and functional roles of sympatric genets to improve T. matsutake cultivation.

3.4. Shiro manipulation under controlled environmental conditions

Mycorrhizal synthesis experiments in vitro generally use small culture vessels containing 100-200 mL soil or other substrate with or without added nutrients (Yamada et al., 2006; Yamanaka, Maruyama, Yamada, Miyazaki, & Kikuchi, 2012; Murata et al., 2013a; Yamanaka et al., 2014; Endo et al., 2015; Saito et al., 2018; Horimai et al., 2020, 2021), which limit the volume of fully developed shiro structures in symbiosis with hosts to approximately 50 mL. Ectomycorrhizal systems can be established in even smaller culture vessels such as test tubes or Petri dishes to shorten the incubation period to a few mo; these are unsuitable for estimating shiro development despite the presence of measurable mycorrhizal root tips. Shiro structures can be increased in size by one order of magnitude using larger culture vessels (~500 mL) or bottles containing 4-5 L soil, which require 2-3 y of incubation (Guerin-Laguette, 2021; Horimai et al., 2021). Horimai et al. (2021) produced large shiro structures by transplanting an ectomycorrhizal seedling synthesized in vitro (~200 mL) into a larger culture vessel filled with 1 L soil under axenic conditions, and then into a culture bottle containing 4 L soil under non-axenic laboratory conditions. Kobayashi et al. (2007) synthesized ectomycorrhizae in vitro in culture vessels containing 1 L soil, with a ca. 1-y period from mycelium inoculation to mycorrhizal status estimation.

Mycorrhizal synthesis experiments typically use a single dikaryotic strain as inoculum for a single host seedling, such that the resulting shiro structure consists of a single fungal individual. It remains unknown whether such single-individual shiro development occurs naturally in situ. Matsutake mushrooms can be considered to have k- or C-selective survival strategies in situ (Yamada et al., 2014; Horimai et al., 2020), because they sustain a shiro structures on the meter-orders in area and are thought to live for decades (Hamada, 1953). Therefore, an established dikaryotic shiro mycelial system should encounter intraspecific spores on mycelial surfaces in situ, such that dikaryon-monokaryon mating will occur, producing fine-scale (millimeter- or centimeter-order) shiro mycelium heterogeneity. Such a process would explain the genetic mosaics observed among shiro structures in situ (Murata et al., 2005). Therefore, I hypothesize that mycelial genetic heterogeneity in shiro confers a functional advantage and can be produced experimentally through spore inoculation to an established shiro mycelium.

To test the this hypothesis (functional significance), I established a line of sibling strains from a single T. matsutake basidioma using the multispore isolation method, and then selected nine strains from among 100 tested strains based on their hyphal growth rates on modified Norkrans's C (MNC) agar medium, including rapid, moderate, and slow growth groups (N = 3 per group). The selected strains showed distinct physiological variation in response to changes in carbon and nitrogen levels in the MNC medium (Yamada et al., 2019). Therefore, I tested the mycelial growth patterns of the strains to evaluate their mycorrhizal symbiosis properties. Carbon/nitrogen balance significantly influences symbiosis efficiency in response to fungal nitrogen acquisition from soil, plant carbon fixation via photosynthesis, and interactive exchange at the interface of the Hartig net (Smith & Read, 2008). The nine sibling strains exhibited significant variation in ectomycorrhization in response to soil nitrogen levels (Horimai et al., 2020). The strain with the largest ectomycorrhizal biomass (#84) exhibited slow mycelial growth on MNC agar (Yamada et al., 2019), which suggests that preliminary ectomycorrhizal synthesis experiments are more important for identifying suitable fungal symbionts for cultivation than measuring mycelial growth rates on nutrient agar (Table 1). Next, I conducted a fungal combination experiment using three strains with different ectomycorrhization levels: #52 (moderate), #84 (high), and #99 (low) (Horimai et al., 2020). Paired inoculation of #52 and #99 led to higher ectomycorrhization levels than single-strain inoculation (Table 2). In a triple inoculation experiment with these strains in a single vessel, paired inocula (#52/#84, #52/#99, and #84/#99) showed higher ectomycorrhization levels than single inoculation with each strain (#52, #84, and #99) (Fig. 2). These results suggest the existence of a mechanism by which some combinations of sibling strains increase ectomycorrhizal biomass compared to a given single strain. Therefore, a new fungal strategy, competitive mycelium activation, was proposed such that the coexistence of genetically different intraspecific individuals (mycelia) increase their biomass relative to that of single individuals under ectomycorrhizal symbiosis (Horimai et al., 2020). When strains #52 and #99 were combined, strain #52 was activated by the presence of strain #99, as shown in polymerase chain reaction (PCR) detection results. Such a mechanism would have a significant impact on mycelium heterogeneity in T. matsutake shiro. By contrast, in shiro mycelium consisting of a single genet for many years, biomass could gradually decrease due to a lack of competitive activation.

Table 1. Comparison of three sibling strains of T. matsutake in their mycelial properties.

Mycelial properties	Relative comparison of three strains in their abilities
Growth rate on MNC agar	#52 > #99 > #84
Mycorrhization level	#84 > #52 > #99

This is a summarized result of Horimai et al. (2020)

Table 2. Comparison of ectomycorrhization level (number of ectomycorrhizal root tips) between paired strain inoculations and single strain inoculation to a single host.

Paired strain inoculations		Single strain inoculation
#52/#84	≈	Means of #52 and #84
#52/#99	>	Means of #52 and #99
#84/#99	≈	Means of #84 and #99

This is a summarized result of Horimai et al. (2020)

Fig. 2-Inoculation locations of Tricholoma matsutake mycelium in the triple inoculation experiment viewed from the top of the culture vessel (Horimai et al., 2020). Two configurations (arrangement-A, arrangement-B) were set up in this experiment, where three cultured strains were inoculated in a culture vessel as three pairs (arrangement-A) or singly (arrangement-B). The number in each circle of dashed line indicates the inoculated strain. Dashed straight lines show where the soil was separated, when the root system was measured, and the root tips were sampled for fungal DNA analysis (typing of fugal genet). As a result, the arrangement-A configuration showed significantly higher ectomycorrhizal biomass than that of the arrangement-B configuration. This figure is redrawn from Fig. 1 of Horimai et al. (2020).

To confirm another hypothesis (cause of the mycelial heterogeneity), I inoculated basidiospores onto established shiro structures following a previous study (Horimai et al., 2020). The spores colonized the shiro and developed new genets through ectomycorrhizal association with the host. Spore inoculation increased the amount of ectomycorrhizal biomass (Fig. 3A), supporting the competitive activation hypothesis and presenting a mechanism for the shiro genetic mosaic of shiro in situ. Thus, spore inoculation into ectomycorrhizal seedlings is a practical method for the effective production of large T. matsutake shiro structures on plant host root systems and to increase the genetic diversity of shiro mycelia (Horimai et al., 2021). This method apparently has a synergistic effect that stimulates competitive activation under environmental conditions.

Fig. 3-Tricholoma matsutake. A: A large shiro development by the spore inoculation to the previously established ectomycorrhizae from a dikaryotic mycelial strain #84 in the 4 L pot soil, photographed Dec 2015. Please see Horimai et al. (2021) in the details of experiment. B, C: Shiro development of strain #84 in the 4 L pot soil, photographed Nov 2017 (B) and May 2018 (C). Arrows with the same directions in B and C indicating the same point in the pot soil show the shiro development in the lower area of pot soil throughout the six months observation.

3.5. Key gaps in our understanding of T. matsutake biology

Dr. Minoru Hamada spent a long time observing and recording the phenology of autumn T. matsutake fruiting in the pine forests of Kyoto, Japan (Hamada, 1974). His ideas and knowledge in the biology of this fungus were studied deeply and widely by his disciples. Dr. Makoto Ogawa is one of them, who published a textbook “The biology of matsutake” (Ogawa, 1978). Subsequently, these ideas have been applied in studies conducted throughout Japan (Ito & Iwase, 1997; Saito, 2020) as well as in Oregon, USA (Hosford et al., 1997), New Zealand (Wang et al., 1997), and other regions worldwide. Despite huge knowledge of matsutake mushrooms obtained in the academic world, it may be a very small part of the real nature of matsutake (Satsuka, 2019). I pick up here only four aspects in the paucity of our scientific knowledges in the biology of matsutake described below.

Among the poorly understood aspects of T. matsutake biology, the reason for the broad variation in T. matsutake basidioma production among forest sites remains unknown, that is high or less productivity is found even within the same mountain range, where vegetation, topography, soil structure, and forest use history are very similar. In Japan, matsutake habitats generally occur on mountain slopes, often rocky or steep, with relatively dry, shallow, and poor organic soil layers where minerals are weathered (B layer) on parent rocks (C layer) under humid and high-precipitation climate conditions (Ogawa, 1978; Vaario, Yang, & Yamada, 2017). In addition, T. matsutake prefers habitats with acidic or neutral parent rock such as granite, sedimentary rocks, or volcanic ash and avoids regions with mafic, basic, and ultrabasic parent rocks. Multivariate statistical analyses of soil elements and other environmental factors have shown that T. matsutake exhibits specific environmental preferences; however, it remains difficult to clarify the effects of these soil elements on T. matsutake mycelial growth, as well as its colonization patterns in forests. In Japan, forest management for T. matsutake production typically begins with historical records of past T. matsutake harvests, which indicate the suitability of the topography and soil structure of the site for this purpose. Accordingly, geographic patterns suitable for production of T. matsutake and other matsutake mushrooms can have important impacts on local economies. Annual T. matsutake harvests fluctuate significantly with the climate (Furukawa et al., 2016; Vaario et al., 2017), such that ongoing global climate change is also an immediate concern.

In situ growth patterns of T. matsutake mycelia are yet to be documented by the quantitative approach. The real-time quantitative PCR (qPCR) is expected for the documentation because the quantity of shiro mycelium probably determine the productivity of fruiting bodies. To date, qPCR data are mostly limited to the in vitro experimental systems (e.g., Yamaguchi et al., 2016; Fig. 3B, C). The qPCR, however, may not precisely mirror the growth pattern for hyphal cell mass. Because each hyphal cell can increase the biomass without cell division prior to the start of a growth phase such as sexual reproduction (basidioma morphogenesis). The variation in qPCR conversion factors among growth conditions has also been described as a weakness of this method (Wallander et al., 2013). The start of active cell division is preceded by biosynthetic events including nutrient accumulation in the cell, i.e., biomass growth (Jomura et al., 2020). Therefore, there may be a time lag between sequential patterns in actual biomass accumulation and those estimated by qPCR. Due to almost no alternative techniques that specifically measure quantity of the shiro mycelium in situ, the qPCR approach is needed technical progress that allow reliable measurement for the shiro biomass under heterogeneous and changeable environmental conditions.

Metagenomic analyses of microbial communities in shiro soils have allowed us to visualize general patterns of species composition (Kataoka et al., 2012; Kim et al., 2013, 2014; Oh, Fong, Park, & Lim, 2016); however, because most of the key microbes detected in such studies appear not to be culturable through general microbial culture methods, we cannot easily apply these findings to T. matsutake cultivation, particularly in seedling experiments. Although there are technical limitations to our understanding of in situ mycelial growth patterns in T. matsutake, it is evident that its shiro biomass is mainly controlled by the photosynthetic activity of hosts and photosynthate allocation to the root system for ectomycorrhizal symbiosis. Other environmental parameters also affect shiro biomass growth patterns in T. matsutake. Therefore, it is important to determine which factors alter in situ mycelial growth patterns of T. matsutake to identify the mechanisms underlying shiro structures, as has been conducted for other edible ectomycorrhizal mushrooms (Iotti et al., 2014; Iotti, Leonardi, Vitali, & Zambonelli, 2018; De la Varga et al., 2013; Castaño et al., 2017; Parladé, Martínez-Peña, & Pera, 2017).

To date, data supporting the molecular mechanisms of matsutake-pine interactions are rare. Genome data for matsutake mushrooms and pine trees are available; however, it remains difficult to test specific genetic mechanisms using suitable mutants. Murata et al. (2019) used T. matsutake mutants to demonstrate a unique fungus-plant interaction described as a conversion from mutualism to parasitism. Although this concept is based only on data from in vitro testing, we may infer that the tight relationship of ectomycorrhizal symbiosis is anchored by an important genetic function. A recent study reported taxon-specific expression of small secreted proteins (SSPs) that likely function as effectors in fungus-plant interactions in Suillus, which may explain the association between the host of this fungal taxon and Pinaceae (Lofgren et al., 2021). Even in saffron milk caps, specific SSPs bearing LysM have been reported (Lebreton et al., 2022). The identification of specific SSPs that determine the conifer hosts of T. matsutake and control gene expressions to influence host colonization levels would dramatically facilitate T. matsutake cultivation. The unique mycorrhizal structure of T. matsutake on its pine host, i.e., slender ectomycorrhizal root tips with a thin fungal mantle, carbonized root cortex in older ectomycorrhizae, elaborate branching that forms mycorrhizal clusters known as witch's brooms, and the macroscopic shiro structure require elucidation at the molecular level (Yamada et al., 1999a, 2006; Horimai et al., 2021).

4. Conclusion

When I first discovered a large cluster of Cantharellus anzutake basidiomata as a child, I wondered why it grew in that specific pine forest. I next encountered a large cluster of this fungus in a mixed oak-pine forest with my students, whose work opened new avenues toward C. anzutake cultivation. During my study of T. matsutake ecology as a postdoctoral researcher, the elderly owner of a pine forest producing T. matsutake basidiomata (matsutake-yama) told me that although her father had clear-cut the trees several decades ago, T. matsutake returned to the site once the pine forest had recovered. This story further inspired me to investigate the nature of this fungus. Another elderly matstake-yama owner recently mentioned “breath of soil” referring to the flow of air from deep soil to the forest floor during the summer rainy season, which was a familiar notion to me, as I had experienced it in another matsutake habitat. Yet another elderly matstake-yama owner told me that he had lectured Dr. Makoto Ogawa in the 1970s in a pine-hemlock forest where T. matsutake mushrooms were harvested, a mere 50 m from his own house. These voices of matsutake-yama owners suggest that the current scientific understanding of T. matsutake represents only a fragment of the human knowledge about its cultivation. The key to successfully cultivating a variety of edible ectomycorrhizal mushrooms, including chanterelles and matsutakes, in association with their hosts will likely depend on how much research time can be allocated in the habitat of these fungi. The appropriate manipulation of different types of fungal cells, i.e., spores, monokarytotic and dikaryotic hyphae, symbiotic mycorrhizal hyphae, at specific experimental stages should lead to mushroom production within active mycorrhizal systems. Key points in the experimental cultivation of C. anzutake and T. matsutake mushrooms have been summarized in flowcharts (Figs. 4, 5), which will be useful for designing future experiments and cultivation systems.

Fig. 4-Flowchart of experimental approach for the cultivation of Cantharellus anzutake with host associations. The highlight of pale gray color with shadow indicates the process under mycorrhizal symbiotic state. As C. anzutake fruits under experimental conditions, crossbreeding experiment can be conducted as routine works.

Fig. 5-Flowchart of experimental approach for the cultivation of Tricholoma matsutake with host associations. The highlight of pale gray color with shadow indicates the process under mycorrhizal symbiotic state. Although T. matsutake cannot be applicable crossbreeding experiments due to not fruiting under experimental condition, the spore inoculation to the mycorrhizal system substitutes in part the crossbreeding experiments.