Introduction

Matsutake [Tricholoma matsutake (S. Ito et S. Imai) Singer] is a traditional special delicacy in Japan. Since the mid-1970s Japan has annually imported more than 3,000 tons of matsutake from China and other countries (^{Wang, 1995}). Matsutake has become one of the most widely known edible mycorrhizal mushrooms in the world (^{Ogawa, 1978}; ^{Wang, 1995}; ^{Hosford et al., 1997}; ^{Danell, 2002}). This is due to its unique flavor, high commercial value, and wide natural distribution. Exploration continues for new matsutake-producing areas, and Mexico, Bhutan, Sweden, and Guatemala are all matsutake exporting countries (^{Danell, 2002}; ^{Hosford et al., 1997}).

Matsutake (matsu = pine, take = mushroom) is a Japanese common name for T. matsutake growing exclusively in pine woods in Japan. In China, matsutake has a variety of names including “song-koumo” (pine-Tricholoma), “song-rong” (pine-mushroom), “song-jun” (pine-fungus), and “qing-gang-jun” (oak-mushroom) (^{Wang, 1995}). Harvesting and trading of matsutake are important economic activities locally in China. This country started exporting T. matsutake to Japan from the north-east in the late 1970s (about 10-30 tonnes annually). Since the 1990s another large T. matsutake-producing area has been south-western China (Yunnan, Sichuan, and Tibet). China has become the world’s largest matsutake exporting country, with more than 1,000 tonnes of matsutake exported to Japan annually (^{Wang, 1995}; ^{Yang et al., 2007}). Harvesting, processing, and trading of matsutake has become an important industry, and generated multi-million foreign dollars for the Chinese economy. The matsutake-producing areas in China are all remote mountainous regions where agriculture is poor, and hence they provide important income for local communities. In Shangri-La County, Yunnan, 80% of farmers are involved with commercial harvest, processing, and trading of matsutake, generating up to RMB $50,000-60,000 annually per family; 5 to 20 times than a normal farming family’s annual income (Figures 1 and 2).

Figure 1 Local Tibetan famers selling Tricholoma matsutake to a middle-man at a wild mushroom market, Zhongdian, Yunnan. 

Figure 2 Tricholoma matsutake at a wild mushroom market, Zhongdian, Yunnan. 

Species complex

T. matsutake is considered to be the true (“hong” in Japanese) matsutake in Japan. It grows in many parts of Asia (Japan, China, North Korea, South Korea, Russia, and Bhutan), Europe (Turkey, Sweden, Finland, Norway, Germany, Czechoslovakia, Austria, Switzerland, and Italy), and North Africa (Algeria and Morocco). There are closely related species in China. Tricholoma bakamatsutake Hongo (false matsutake) is mainly found in oak forests and has a stronger matsutake-like aroma but less commercial value. Tricholoma fulvocastaneum Hongo is another related species with lower commercial value. Three new species were recently discovered in China; T. lavendulophyllum F. Q. Yu, T. longicystidium F. Q. Yu, and T. pseudomatsutake F. Q. Yu. These belong genetically and morphologically to the matsutake group (Yu, 2007). T. matsutake (S. Ito et S. Imai) Singer var. formosana Sawada was described by ^{Sawada (1931)}, based on collections of matsutake from Taiwan, and is now considered synonymous with T. matsutake.

Tricholoma magnivelare (Peck) Redhead (pine mushroom or white matsutake) is a closely related species with good commercial value, found in Canada, the USA, and Mexico (^{Hosford et al., 1997}). Tricholoma caligatum is another closely related species, mainly found in southern Europe (^{Kytövuori, 1989}), with no commercial value due to its bitter taste. These two species are not found in China.

Matsutake and related species are found mainly in north-eastern and south-western China. In the north-east, matsutake is found in pine forests along the eastern mountain ranges; the Xiaoxingan-ling and Changbai Mountains, between 200 m and 1,500 m a.s.l. In south-western China matsutake is found under pine and oak forests at high elevations between (1,200-) 2,000 and 3,500 (-4,200) m a.s.l. in Yunnan, Sichuan, and Tibet. Matsutake is also found on Mt. Yushan, in Taiwan.

Based on molecular analysis (^{Murata et al., 2008}), T. matsutake can be divided into two genotypes; genotype A refers to the mushrooms in Japan, the Korean Peninsula, and north-eastern China, and genotype B refers to those in south-western China. Genotype A may be further divided into three sub-genotypes: sub-genotypes 1 and 2 refer to the mushrooms in Japan, sub-genotype 3 to those on the Korean Peninsula, and sub-genotype 4 to those in south-western China. We believe that the biogeography of matsutake genotypes can be satisfactorily explained by the vicariance model. It can also be related to their different host associations. The matsutake mushrooms of genotype A are mainly associated with Pinus densiflora, while those of genotype B are associated with a diversity of species in Pinus, Quercus, Lithocarpus, and Castanopsis (^{Wang, 1995}; Yu, 2007).

Shiro: cradle of matsutake

Tricholoma matsutake and related species form distinctive fungal colonies in the soil called shiro (= white, castle or place) in Japanese. These colonies are white in pine forests, to pale grey in oak forests, and consist of a compact mass of mycelia which colonizes everything in the shiro soil including plant roots (fine rootlets and thick mother roots, soil granules and rocks, and gaps between soil granules (Figure 3). The shiro can be divided into a few zones, including a zone of maximum mycelial growth and root infection where fruit bodies are produced. After fructification, roots collapse and die, and the soil becomes extremely hydrophilic. Fruit bodies and primordia are only found in shiros. The shiro is the matsutake mushroom birth-place.

Figure 3 Tricholoma matsutake shiro in a Pinus yunnanensis forest, Huize, Yunnan. 

Shiro development is closely correlated with the age of the host plants and usually begins when Pinus densiflora trees are 10 to 20 years old. The fruit bodies begin to be productive three to four years after the shiro is formed. The shiro advances between 100 and 200 mm per year. Matsutake production reaches its peak when shiros become circles, arcs, or other irregular shapes. Finally the shiro collapses and disappears, and production of fruit bodies declines to nothing. Shiros can survive and still produce fruit bodies for several years even if host trees are clear-felled (^{Ogawa, 1978}; ^{Wang, 1995}).

Shiro occurrence is related to environmental conditions, including topography, vegetation, and soils. Shiros usually occur on the middle to upper parts of hills, on sunny and semi-sunny slopes (^{Wang, 1995}; ^{Gong et al., 1999}), where plant canopy provides only 50 to 80% cover and the shrub layer is relatively sparse. Suitable soils are well-drained, poor, rocky-sandy or sandy loam, dry-moist and not wet, with pH 4.5 to 5.5, and a thick litter layer. In general, shiros occur in very harsh environments, probably with reduced competition from other fungi, microorganisms, insects and other natural enemies.

In north-eastern China, matsutake always occurs in pure stands of Pinus densiflora or where these trees are mixed with a few individuals of Quercus mongolica, and with bush and herb layers containing fewer than 20 species, including Rhododendron species, Lespedeza bicolor, Atractilis japonica, Atractylodes japonica, Artemisia species, Imperata cylindrica, Peucedanum terebinthaceum, Pedicularis sp., Platycodon grandiflorus, Polygonatum odoratum, and Carex callitrichos. These are considered indicator plants for matsutake-producing areas (^{Wang and Xie, 1982}; ^{Sun et al., 1989}; ^{Fu et al., 1996}). In eastern China (Anhui Province) and Taiwan, matsutake is found under Pinus taiwanensis (^{Chen, 1983}). In south-western China, matsutake forests can be divided into two kinds; pure pine stands or pine mixed with oak and other broadleaf trees, and pure oak forests or oak mixed with pines. The pine species in this area are P. densata, P. yunnanensis, and P. armandii. The oak species include Q. semicarpifolia (Figure 4), Q. aquifoliodes, Q. senescens, Q. panosa, Q. gillilana, and Q. spinnosa, and other common trees within these forests are Castanopsis delavayi, Cyclobalanopsis delavayi, Lithocarpus sphaerocapus, and L. dealbatus. The bush and herb layers are less developed and include Rhododendron species, grasses such as Carex and Festuca, and fern species. Canopy density ranges from 0.6-0.7. Rhododendron species are always found in matsutake-producing areas in China, indicating that soils are acidic and poor.

Figure 4 Tricholoma matsutake growing under Quercus semicarpifolia trees (3000 m a.s.l), Zhongdian, Yunnan. 

Tricholomauber matsutake is also rarely found associated with other tree species. It is sometimes found under Pinus pumila and P. koraiensis, for example, in north-eastern China (Figure 5), under Picea meyeri in Shanxi Province, China (^{Liu, 1991}), under Pinus yunnanensis var. pygmaea in south-western China, and under Tsuga chinensis in Taiwan (^{Hu, 1994}).

Figure 5 Tricholoma matsutake growing under Pinus koraensis trees, Yichun, Heilongjiang. 

Fruiting

Matsutake primordia develop just below the surface layer of the shiro, and take 10 to 20 days to open into mushrooms depending on soil temperature, moisture, and other factors. Matsutake fruits between July and early September in north-eastern China and July to October in south-western China, and yields are closely tied to weather patterns, especially rainfall and temperature. Generally, T. matsutake yields are highest when there is plentiful rain in spring, a relatively hot and dry summer, and a moist cool autumn. According to Sun (personal communication) in Heilongjiang Province, north-eastern China, good harvests can be expected if the 100 mm soil temperature is between 16 and 19 ^o C, and 70 to 100 mm of rain falls in around 10 showers between the last 10 days of July and the first 10 days of August. Alternatively if the soil temperature is between 19 and 22 ^o C, good production will follow if there is a rainfall of no more than 100 mm, spread over eight showers. However, if the soil temperature has a conspicuous increase (over 19 ^o C) or drops too low (below 15 ^o C) for several days after the primordia have been formed, the primordia will abort and there will be no fruiting that year.

In Japan and Korea, matsutake production starts when P. densiflora trees are around 20 years old and reaches a maximum in 40 to 50 year old forests, gradually declining over the following 30 to 40 years as the litter layer builds up and the pines grow old; finally being replaced by other plant species such as deciduous oaks. In China, matsutake production starts when pine trees are 45 to 45 years old and ends when they are around 120 years old (^{Sun et al., 1989}; ^{Wang, 1995}).

Matsutake: not a normal ectomycorrhizal fungus

Following over 100 years of research history of matsutake in Japan, the symbiotic association of T. matsutake and P. densiflora has been recognized as a parasitic ectomycorrhiza or an ectomycorrhiza with a parasitic nature (^{Masui, 1927}; ^{Ogawa, 1982}; ^{Tominaga and Komeyama, 1987}). Mycologists working recently in Japan (^{Gill et al., 2000}; ^{Guerin-Laguette et al., 2002}; ^{Yamada et al., 1999a}; ^{Yamada et al., 1999b}; ^{Vaario et al., 2002}) have reexamined mycorrhizal roots formed by this association in situ and in vitro, and concluded that they are typical ectomycorrhizae, although atypical behaviors do exist (^{Gill et al., 2000}; ^{Hur et al., 2001}; ^{Vaario et al., 2002}; ^{Yamada et al., 1999a}). The saprophytic nature of T. matsutake was discovered and studied by ^{Hiromoto (1963 a, b)} and others (^{Terashita and Kono, 1987}; ^{Terashita et al., 1995}; ^{Wang, 1995}; ^{Hur et al., 2001}; ^{Vaario et al., 2002}, ²⁰¹¹; ^{Wang and Hall, 2006}; ^{Yamada et al., 2006}, ²⁰¹⁰). However, all attempts to grow T. matsutake using typical methods for cultivating saprotrophic mushrooms have failed (^{Ogawa, 1978}; ^{Wang, 1995}).

Our research suggests that matsutake is not a typical ectomycorrhizal fungus and her biotrophy involves mutualism, parasitism, and saprophytism. We have previously discussed this in detail (^{Wang, 1995}, ^{Wang et al., 1997}; ^{Wang and Hall, 2006}).

Conservation and cultivation

Increasing commercial harvest of matsutake from natural forests has damaged matsutake environments and populations in China to the extent that harvests have dramatically decreased over the past 20 years. The Shangri-La region is the most important matsutake producing region in China. Production in this area dropped from over 700 tonnes in 1999 to around 500 tonnes after 2010 (Yang et al., 2009). The issue of the conservation and cultivation of matsutake has been raised previously. Attempts at production of matsutake-infected seedlings for the establishment of plantations have had little success in or outside China (^{Ogawa, 1978}; ^{Wang, 1995}; ^{Hu, 1994}; ^{Gong et al., 1999}; ^{Guerin-Laguette et al., 2000}; ^{Yamada et al., 2002}), even though the technology for the infection of matsutake seedlings has made good progress in Japan and other countries (^{Yamada et al., 2006}, ²⁰¹⁰; ^{Park et al., 2007}). Research of T. matsutake in China has concentrated on its conservation and ways of stimulating fruiting in forests (^{Wang and Xie, 1982}; ^{Wei et al., 1985}; ^{Fu et al., 1996}; ^{Gong et al., 1999}), and similarly in Japan and Korea (^{Ogawa, 1978}; ^{Park et al., 1997}; ^{Koo and Milek, 1998}) it has been reported that the number of shiros increased by 9.8%, and fruit bodies increased by 44.7%, in experimental plots in Yunnan following three years of management of matsutake forests. The contracting of matsutake forests to individuals is an effective way of managing matsutake resources in Yunnan (^{Su, 2002}). Recently, local regulations have been issued and a few matsutake reserves have been established in Jilin, Helongjiang, and Yunnan Provinces (^{Zhou, 2002}). However, the protection and management of matsutake forests are still big challenges in China, due to a variety of social, cultural, and institutional factors (Yang et al., 2009).

The recent application of molecular technology has considerably progressed our understanding of the biology of matsutake (^{Lian et al., 2003}; Yu, 2007; ^{Xu et al., 2008}, 2010; ^{Wang et al., 2013}; ^{Wang and Chen, 2014}). Hopefully, advances in molecular technology will help to provide better ways of protecting and cultivating the matsutake mushroom.