Fungal Diversity (2011) 51:279–296 DOI 10.1007/s13225-011-0136-7 Major clades in tropical Agaricus Ruilin Zhao & Samantha Karunarathna & Olivier Raspé & Luis A. Parra & Jacques Guinberteau & Magalie Moinard & André De Kesel & Gérard Barroso & Régis Courtecuisse & Kevin D. Hyde & Atsu K. Guelly & Dennis E. Desjardin & Philippe Callac Received: 17 February 2011 / Accepted: 3 September 2011 / Published online: 18 September 2011 # Kevin D. Hyde 2011 Abstract Agaricus (Basidiomycota) is a genus of saprobic fungi that includes edible cultivated species such as Agaricus bisporus, the button mushroom. There has been considerable ecological, nutritional and medicinal interest in the genus, yet the extent of its diversity remains poorly known, particularly in subtropical and tropical areas. Classification of tropical species has for a large part followed the classification of temperate species. The objective of our study was to examine to what extent this system of classification is appropriate for tropical Agaricus species. Species from temperate sections were therefore compared to the major clades of tropical species using a phylogenetic approach. ITS1+2 sequence data from 128 species were used in the phylogenetic analysis. Specimens included four species of genera closely related to Agaricus, 38 temperate species representing the eight classical sections of the genus, and 86 putative species of Agaricus from tropical areas of Africa, Asia and the Americas. Bayesian and maximum likelihood analyses produced relatively congruent trees and almost identical clades. Our data show that (i) only about one-third of tropical species belong to the classical sections based on temperate species; the systematics of the genus therefore needs to be expanded; (ii) among the remaining two-thirds of tropical species, those from the Americas and those from Africa and/or Asia group in distinct clades, suggesting that secondary diversification occurred in these two areas; (iii) in contrast, several clades of classical sections contain R. Zhao Key Laboratory of Forest Disaster Warning and Control in Yunnan Province, Faculty of Conservation Biology, Southwest Forestry University, Kunming 650224, China R. Courtecuisse Département de botanique, Faculté des sciences pharmaceutiques et biologiques, 59006 Lille, France S. Karunarathna : K. D. Hyde School of Science, Mae Fah Luang University, Chiang Rai 57100, Thailand K. D. Hyde Botany and Microbiology Department, College of Science, King Saud University, Riyadh 11442, Saudi Arabia O. Raspé : A. De Kesel National Botanic Garden of Belgium, Domein van Bouchout, 1860 Meise, Belgium L. A. Parra Avda. Padre Claret 7, 5º G, 09400 Aranda de Duero, Burgos, Spain J. Guinberteau : M. Moinard : G. Barroso : P. Callac (*) INRA, UR1264, Mycologie et Sécurité des Aliments, 33883 Villenave d’Ornon, France e-mail: callac@bordeaux.inra.fr A. K. Guelly Département de Botanique, Faculté des Sciences, Université de Lomé, B.P. 1515, Lomé, Togo D. E. Desjardin Department of Biology, San Francisco State University, San Francisco, CA 94132, USA 280 American and African+Asian species along with temperate species. In this study, we used approximately 50 distinct species from a small area of northern Thailand, most probably being novel species. This diversity indicates that Agaricus is a species-rich genus in the tropics as well as in temperate regions. The number of species and the hypothetical paleotropical origin of the genus are discussed. Keywords Agaricus . Basidiomycota . Tropical biodiversity . Biogeography . ITS . Phylogeny Introduction Agaricus L. is a genus of basidiomycetes that includes numerous species of which about 200–250 are presently known. Previous estimates of species are 300–400 worldwide, but the number is probably closer to 400 according to Bas (1991). Species live in various climates on all continents, with the exception of Antarctica. Agaricus species are saprobic and generally humicolous. Several species are collected (A. campestris L. : Fr.) or cultivated (A. bisporus (J.E. Lange) Imbach, the button mushroom) for consumption or for medicinal use (A. subrufescens Peck; Angeli et al. 2006; Bernarshaw et al. 2007). Despite their ecological and economical interest, the diversity of species remains poorly known, particularly in subtropical and tropical areas. Numerous species of Agaricus have however, been described from tropical areas over the last century (Baker and Dale 1951; Berkeley and Broome 1871; Heinemann 1956a, b, c, 1957, 1961, 1962a, 1962b, 1962c, 1971, 1978, 1980, 1982, 1990, 1993; Murrill 1918, 1942, 1945, 1946; Pegler 1966, 1968, 1969, 1977, 1983, 1986; Pegler and Rayner 1969; Peterson et al. 2000; and Rick 1906, 1919, 1920, 1930, 1939, 1961). Agaricus species have a limited number of characteristic phenotypic traits, and therefore identification of species can be challenging due to environmental effects and intraspecific variability; it is therefore often difficult to correctly identify specimens in the field. In temperate areas and particularly in Europe and North America, this situation is changing because of recent progress in the classification facilitated by molecular characterization and phylogeny. The genus Agaricus has been shown to be monophyletic (Vellinga et al. 2011). Among the eight sections recognized in the subgenus Agaricus (Parra 2008), Bivelares (Kauffman) L.A. Parra and Xanthodermatei Singer have been phylogenetically reconstructed (Challen et al. 2003; Kerrigan et al. 2006; Kerrigan et al. 2008) and others are under investigation. Circumscription of the species and sections has been improved, but phylogenetic reconstruction for groups of temperate species is far from complete. For example, Fungal Diversity (2011) 51:279–296 section Sanguinolenti Jul. Schäff. & Møller ex L.A. Parra seems polyphyletic and the circumscription of species as frequently encountered as A. campestris remains obscure. Species identification, as well as description of new species, should take characteristic traits into consideration, not only of the species but also of the phylogenetic group to which they belong. However, different traits can be selected to characterize the groups. For instance, the yellow vs. red discolouration of the flesh of the sporophore that was considered a major trait 50 years ago, although still important, should not be heavily weighted, since two non-related sections, Arvenses Konrad & Maubl. and Xanthodermatei, share this trait (Parra 2008). In contrast, certain odours appear to be synapomorphic and are crucial taxonomic traits, at least for temperate species (Parra 2008). Despite the number of species described, mainly between 1956–1993 by P. Heinemann and D.N. Pegler from tropical and subtropical areas, many species remain poorly known, and many remain to be described judging from the specimens we have collected and sequenced from these areas within our project of the European Distributed Institute of Taxonomy (EDIT). Similarly, tropical specimens are often difficult to identify or describe as novel species for two main reasons: (i) literature is scattered and type specimens are not easily available for comparison; and (ii) description and classification of tropical species, despite the introduction of the tropical subgenera Conioagaricus Heinem. and Lanagaricus Heinem. and the tropical section Brunneopicti Heinem & Gooss.-Font. (Heinemann 1956a), has been globally based on traditional systematics of temperate species, and therefore on the traits characterizing the temperate sections. Sequence data for less than ten identified tropical species are currently available in GenBank; therefore in this study we mostly use sequences from our collections in the analyses. The understanding of temperate sections has conceptually changed in recent years (Parra 2008), however the changes may neither be appropriate or sufficient to incorporate tropical diversity. The objective of the present study was to assess whether tropical Agaricus species are distributed in the same clades as the temperate species of the genus and to establish if there are any exclusively tropical clades. Our approach was to use a limited number of temperate species assigned to the different sections and a large sample of tropical species. We used tropical in the broad sense to include “tropical and subtropical” samples: most samples however were collected in tropical regions, but some were collected at high altitudes in Mexico, or outside the “tropics” on the western coast of the USA or in Brazil where climates can be considered as subtropical. We have chosen at this stage to obtain a better understanding of the phylogeny of the genus and its sections and have not Fungal Diversity (2011) 51:279–296 attempted to carry out monographic work. However, we have incorporated some named species as well as some type specimens of tropical taxa. The phylogeny is inferred via the analysis of the DNA sequences of ITS1 and ITS2 (internal transcribed spacers 1 and 2) separating the rRNA genes and that have already been largely used for the temperate species of this genus. This study will facilitate future taxonomic work on tropical Agaricus species and allow evolutionary considerations and biogeographical analyses. Materials and methods Fungal material Among 178 sequenced samples listed in Table 1, 156 have been collected and sequenced by the authors or are from their Herbaria. MATA816 provided by Gérardo Mata is an exception. The remaining sequences (22) were provided by Richard W. Kerrigan (SP307818, RWK2019, JH1) or were downloaded from GenBank. The 178 sequenced specimens included seven collections belonging to genera closely related to Agaricus (Heinemannomyces, Hymenagaricus, or unidentified), 38 temperate species of Agaricus, and 133 tropical collections of Agaricus from Africa, Asia, and the Americas. The seven collections of genera closely related to Agaricus have been used in preliminary analyses with species of other genera such as Hymenagaricus, Micropsalliota and Lepiota. Since they consistently formed a sister clade or one of the most closely related clades to the monophyletic genus Agaricus, they were selected as outgroups for the analyses. ZRL43, has a sequence similar to ecv3586 which Vellinga et al. (2011) identified as Heinemannomyces splendidissimus Watling and also found to be closely related to Agaricus. The 38 temperate species were chosen among about 100 sequenced species to represent all the major phylogenetic clades or taxonomic subgroups of the eight presently recognized sections of subgenus Agaricus (Challen et al. 2003; Kerrigan et al. 2006; Kerrigan et al. 2008; Parra 2008; and unpublished data). The type species, or in section Chitonioides Romagn. one of its most closely related species, was included for each section: A. bitorquis (Quèl.) Sacc. in section Bivelares, A. gennadii Chatin & Boud. in section Chitonioides, A. xanthodermus Genev. in section Xanthodermatei, A. sylvaticus Schaeff. in section Sanguinolenti, A. campestris in section Agaricus, A. litoralis (Wakef. & A. Pearson) Pilát in section Spissicaules (Heinem.) Kerrigan, A. arvensis Schaeff. in section Arvenses, and A. comtulus Fr. in section Minores Fr.. Four temperate samples (WC913, CA279, CA486 and CA684) were not formally identified and could represent new species, however, they were assigned to sections based on their morphology. The remaining 133 tropical collections of Agaricus were 281 deliberately not assigned to any section since the objective of the study was to clarify the classification of the tropical species and to establish to what extent the temperate Agaricus classification is useful for accommodating tropical species. DNA isolation Three DNA isolation methods were used depending on the laboratory. At the Institut National de la Recherche Agronomique (INRA), DNA was isolated following a CTAB protocol (Saghai-Maroof et al. 1984; Doyle and Doyle 1987) with ethanol precipitation and modified as follows: approximately 25 mg of dried mushroom were ground to a fine powder in liquid nitrogen. The samples were transferred into 2 mL reaction tubes. 700 μL of hot extraction buffer (CTAB 2% w/v; NaCl 1.4 M; Tris pH 8.0100 mM; EDTA 10 mM; ß-mercaptoethanol 2% v/v) were added. After 20 minutes of incubation at 56°C, cell debris, polysaccharides and proteins were separated from aqueous DNA portions through two purification steps with 700 μL chloroform:isoamylalcohol (24:1). DNA was washed with 700 μL precipitation buffer (CTAB 1% w/v; Tris pH 8.0 50 mM; EDTA 10 mM). The pellet was resuspended in 500 μL NaCl 1 M. DNA was precipitated with the addition of two volumes of absolute ethanol. The DNA pellet was washed 3 times in 1 mL 70% ethanol, air-dried and resuspended in 50 μL sterile H2O. At the National Botanic Garden of Belgium (BR), a similar protocol was used except that only ca. 10 mg of tissue was ground with a Retsch 300 mill; only 0.2% ßmercaptoethanol was added to the lysis buffer; samples were lysed for 1 hour at 60°C; proteins and polysaccharides were removed by two consecutive extractions with chloroform:isoamylalcohol (24:1), after which DNA was immediately precipitated by the addition of 0.8 volume isopropanol to the aqueous phase; the pellet was washed once in 600 μL 70% ethanol, air-dried, and resuspended in 100 μL TE pH 8.0; RNA was then digested with RNase A. DNA was isolated mostly from dry specimens, but in a few cases from CTAB-preserved tissue samples. At San Francisco State University (SFSU) and Hong Kong University (HKU), genomic DNA of the samples collected in 2004–2007 were isolated from dried fungal specimens using the E.Z.N.A. Forensic DNA Extraction Kit (Omega Bio-Tek, Norcross, GA, U.S.A.). PCR, primer design and sequencing PCR amplification was accomplished mainly with primers ITS5 and ITS4 (White et al. 1990). Because some specimens were old or because they exhibited length heteromorphisms, it was sometimes necessary to sequence ITS1 and ITS2 separately, with primers ITS5 and ITS2 (White et al. 1990) for the former, and ITS-PM (reverse complement of primer ITS1Rev323, Kerrigan 2007) and ITS4 for the latter. 282 Table 1 Collections of Agaricus and closely related genera N°a Sample Identificationb Country Collectorc Date AF432898 EU363033 JF797186 JF797187 AF432880 JF797188 DQ182532 AF432877 JF797181 Habitat Herbariume Lawn, Acacia LIP CGAB On roadside CGAB Under Cedrus Under Pinus Under Carpinus LAPAG LAPAG LAPAG Grass, oak LAPAG Under Robinia Forest edge In grassland CGAB CGAB CGAB CGAB In grassland CGAB Under Cedrus CGAB From GenBank From GenBank AM PW 11/1983 22/12/2008 ORo 2005 DQ185552 DQ182513 DQ185553 AY899273 AY899271 AY484684 JF797178 JF797179 JF797180 AY943974 DQ182531 Locationd Perpignan Canberra, Capital Territory From GenBank Seine-Maritime, Tancarville From From From From From LP LP OJ 06/11/2004 13/11/2003 15/09/2002 LP 27/10/1999 JF797189 JF797190 JF797191 JF727867 AY943973 JF797192 DQ185570 JG, AG anonymous TB JG, AG 23/09/2006 13/10/2004 27/08/2004 09/10/2010 JG 13/10/2007 JF797193 PC,JG 28/05/2008 GenBank GenBank GenBank GenBank GenBank From GenBank Madrid, Parque del Retiro Burgos, Barrio de Cortes Tremosnice near Caslav From GenBank From GenBank From GenBank From GenBank Gumiel de Mercado Gironde, Illats Paris, Salon MNHN Dordogne Brouchau Thenon Gironde, Préchac From GenBank Gironde, Léognan From GenBank Gironde, Labrède Fungal Diversity (2011) 51:279–296 COLLECTIONS FROM TEMPERATE AREAS (grouped by section) Section Bivelares 28 RWK1462 A. bitorquis USA 29 MATA681 A. tlaxcalensis T Mexico Section Chitonioides 30 PEAR83340 A. pearsonii France 31 CA684 A. sp. Australia 32 ARP173 A. bernardi USA 33 CA387 A. gennadii France Section Xanthodermatei 34 CA217 A. phaeolepidotus France 35 RWK1938 A. hondensis USA 36 CA186 A. freirei France 40 CA160 A. xanthodermulus T France 48 CA15 A. xanthodermus France Section Sanguinolenti 49 WC913 (A. fuscofibrillosus) USA 50 LAPAG341 A. sylvaticus Spain 51 LAPAG283 A. benesii Spain 52 LAPAG531 A. bohusii Czech Repub. 55 RWK1415 A. pattersonae USA 56 CA123 A. boisseletii France Section Agaricus 57 CA87 A. cupreobrunneus France 58 RWK1917 A. campestris USA 59 LAPAG141 A. langei Spain Section Spissicaules 72 CA486 A. sp. France 74 CA406 A. lanipes France 76 CA279 A. sp. France 78 CA829 A. litoralis f France 79 RWK1940 A. subrutilescens USA 80 CA583 A. aff.impudicus France 88 CA177 A. bresadolanus France Section Arvenses 94 CA590 A. augustus France Genbank N°a Sample Identificationb Country AY484683 AF482834 JF797194 JF797195 JF514525 JF727868 JF797203 JF715065 DQ232644 JF797182 species 1 to 4 JF691559 JF727840 JF727858 JF727859 JF514522 JF514543 JF727860 JF727863 JF727861 JF440300 JF514534 JF727862 JF514532 JF727844 JF514531 JF514518 JF691550 Collectorc Date Locationd Habitat Herbariume JG 16/10/2008 From GenBank From GenBank Gironde, Villenave d’Ornon Under Pinus CGAB JG EH LP JG LP-RK-PC 05/11/1997 12/02/1968 06/11/1997 30/09/2006 15/10/2003 In dune Nothofagus In meadow Glade of oak In lawn CGAB BR LAPAG CGAB LAPAG Under Pinus LAPAG In forest In forest Azadirachta Azadirachta In forest In grassland In forest Grazed bush In forest In forest In grassland In forest In grassland In urban park In forest In grassland In grassland In grassland In forest In grassland Old termite hill In forest In forest In forest BBH BBH LAPAF LAPAF Oléron Island Kowai, Monts Grey Honrubia de la Cuesta Gironde, Blanquefort Yvelines, Mantes la Jolie From GenBank LP 19/11/1999 Segovia, Pradales belong to genera closely related to Agaricus) ZR 10/06/2006 Chiang Mai, Mae Taeng DD 13/06/2006 Chiang Mai, Mae Taeng LP 21/05/2010 Lomé, University Campus LP 21/05/2010 Lomé, University Campus JG-GB 25/07/2010 Chiang Mai, University P JG-GB 25/07/2010 Chiang Mai, University P ZR 15/09/2006 Chiang Rai, Ob Luang P MG 11/1948 Panzi, Kivu NT 25/06/2010 Chiang Mai, Mae Taeng ZR 25/07/2010 Yunnan, Xishuangbana JG-GB-SK 27/07/2010 Chiang Mai, University P JG-GB 24/07/2010 Chiang Mai, University P JG-GB-SK 27/07/2010 Chiang Mai, University P EM 25/07/2004 Ouagadougou SK-KW 27/07/2010 Chiang Mai, Mae Taeng JG-GB 25/07/2010 Chiang Mai, University P SK-JG 27/07/2010 Chiang Mai, University P SK-JG 27/07/2010 Chiang Mai, University P SK-KW 22/0720/09 Chiang Rai, Khun Kone W SK-JG 27/07/2010 Chiang Mai, University P AK 08/06/1999 Niaouli, Plateau DD 07/06/2006 Chiang Mai, DSPNP TO 13/06/2006 Chiang Mai, DSPNP ZR 08/06/2006 Chiang Mai, Mae Taeng BBH BR BR BR BBH BBH BBH 283 WC777 A. fissuratus Denmark ECVel2339 A. inapertus USA CA640 A. arvensis France Section Minores 103 CA101 A. aridicola France 118 Horak68/79 A. viridopurpurascens T New Zealand 121 LAPAG77 A. pseudolutosus Spain 122 CA490 A. brunneolusg France 123 LAPAG339 A. comtulus France 125 GAL9420 A. campbellensis T New Zealand 124 LAPAG111 A. heinemannianus Spain Collections from tropical or subtropical areas (sorted by number of species; 1 ZRL3043 cf. He. splendidissimus Thailand 1 ZRL3062 cf. He. splendidissimus Thailand 2 LAPAF9 Hy. ardosiicolor Togo 2 LAPAF14 Hy. ardosiicolor Togo 3 CA833 Thailand 4 CA801 Thailand 4 ZRL3103 Thailand 5 Goossens5066 A. heterocystis T RDCongo 5 NTF9 Thailand 6 ZRL10.072 China 7 CA819 Thailand 8 CA799 Thailand 9 CA820 Thailand 10 ADK4732 A. subsaharianus T Burkina-Faso 11 NTT117 Thailand 12 CA800 Thailand 13 NTS116 Thailand 13 NTS115 Thailand 14 NT019 Thailand 15 NTS113 Thailand 16 ADK2564 (A. brunneopictus) Bénin 17 ZRL3031 Thailand 17 ZRL3064 Thailand 17 ZRL3041 Thailand 97 99 98 Genbank Fungal Diversity (2011) 51:279–296 Table 1 (continued) 284 Table 1 (continued) Sample 18 19 20 20 20 21 22 22 22 23 24 25 26 27 27 27 27 37 37 37 37 38 39 41 42 43 44 44 44 45 45 45 45 46 46 47 NTT34 LAPAF1 ZRL4017 NTT118 ZRL3005 CA856 ZRL2043 ZRL3086 ZRL2085 LAPAF2 ZRL3099 ZRL2132 LAPAF4 ZRL2123 NTS118 ZRL2128 ZRL3014 Grinling70109 NTT84 NTT38 NTS117 ZRL3044 NTF61 F2530 F2715 ZRL3095 Goossens5415 Hendrickx515 NTT50 NTS7 ZRL3094 NTT95 NT007 F2767 ADK2785 NTF58 Identificationb A. inoxydabilis A. campestroides A. A. A. A. A. A. trisulphuratus aff. trisulphuratus aff. trisulphuratus aff. trisulphuratus aff. trisulphuratus microvolvatulus T (A. caribaeus) A. aff. endoxanthus A. xanthosarcus T A. aff. volvatulus A. aff. volvatulus Country Genbank Collectorc Date Locationd Habitat Thailand Togo Thailand Thailand Thailand Thailand Thailand Thailand Thailand Togo Thailand Thailand Togo Thailand Thailand Thailand Thailand Congo Braz. Thailand Thailand Thailand Thailand Thailand Fr. Martinique Fr. Martinique Thailand RDCongo RDCongo Thailand Thailand Thailand Thailand Thailand Fr. Martinique Bénin Thailand JF514536 JF727841 JF691549 SK LP PS SK-KW-JG ZR JG JK ZR TB LP ZR ZR LP KH SK-JG EG ZR KG KW KW SK-JG TO NT JF JF ZR MG FH KW SK ZR SK SK JF AK NT 19/06/2010 12/05/2010 15/05/2007 27/07/2010 26/05/2006 25/07/2010 26/06/2005 22/07/2006 03/07/2005 12/05/2010 05/09/2006 21/08/2005 12/05/2010 11/08/2005 27/07/2010 18/08/2005 03/06/2006 15/01/1967 07/10/2010 20/06/2010 27/07/2010 10/06/2006 16/07/2010 15/04/2002 06/11/2002 13/08/2006 12/1954 15/04/1939 24/06/2010 11/05/2010 09/08/2006 17/07/2010 15/08/2009 19/06/2003 13/06/2000 10/07/2010 Chiang Mai, Mae Taeng Ola Chiang Mai, Mae Taeng Chiang Mai, University P Chiang Mai, Mae Taeng Chiang Mai, University P Chiang Mai, Mae Taeng Chiang Mai, Chiang Dao Chiang Mai, Mae Taeng Ola-Okpa-Fou Chiang Mai, Mae Taeng Chiang Mai, Mae Taeng Ola-Okpa-Fou Chiang Mai, Mae Taeng Chiang Mai, University P Chiang Mai, Mae Taeng Chiang Mai, Mae Taeng Brazzaville Chiang Mai, Doi Suthep Chiang Mai, Doi Suthep Chiang Mai, University P Chiang Mai, Mae Taeng Chiang Mai, Kiewtubyoung Tartane, pointe rouge Tartane, pointe rouge Chiang Mai, Mae Taeng Panzi, Kivu Mulungu Chiang Mai, New waterfall Chiang Rai, Khun Kone W Chiang Mai, Mae Taeng Chiang Rai, Doi Tung Chiang Rai, Khun Kone W Grand rivière Niaouli, Plateau Chiang Mai, DSPNP In forest In corn field In forest In grassland In forest In grassland In forest In forest In forest Path in forest In forest In forest Path in forest In forest In grassland In forest In forest In forest edge In forest In forest In grassland In forest In forest In forest In forest In forest Coffee plantation JF797202 JF691553 JF727842 JF691556 JF691558 JF727843 JF691557 JF514524 JF691555 JF514528 JF727856 JF727847 JF691554 JF514523 JF514533 JF727848 JF514527 In In In In In In In In forest forest forest forest forest forest forest forest Herbariume LAPAF BBH BBH BBH BBH BBH LAPAF BBH BBH LAPAF BBH BBH BBH BR BBH LIP LIP BBH BR BR BBH LIP BR Fungal Diversity (2011) 51:279–296 N°a Sample 53 54 54 60 61 61 61 62 63 64 65 66 67 68 69 70 71 73 75 75 77 81 82 83 83 84 85 86 86 86 87 89 90 91 92 92 ZRL3012 ZRL2136 ZRL2109 F2047 F2272 LD026 F3109 NTS05 F2389 F2187 NT020 MATA816 F1779 JH1 LAPAM1 CL/GUAD05.099 CJL090302-05 ADK2171 F2255 F2039 F2301 RC/GUY07.019 NYS122 NTT42 ZRL10.071 LAPAF3 ZRL3093 Rammeloo5756 Goossens5323 Goossens5406 F2467 LAPAM4 DeMeijer1904 RWK2019 F2285 RM05/156 Identificationb A. aff. argyropotamicus (A. aff. argyropotamicus) (A. magnivelaris) (A. johnstonii) A. cf. floridanus A. aff. rufoaurantiacus A. aff. rufoaurantiacus A. cf. goossensiae (A. parasilvaticus) (A. cf augustus) A. aff. impudicus A. kivuensis A. kivuensis T A. kivuensis T (A. porosporus) (A. argyropotamicus) A. deserticola A. fiardiih A. fiardiih Country Genbank Collectorc Date Locationd Habitat Herbariume Thailand Thailand Thailand Fr. Martinique Fr. Martinique Thailand Fr. Martinique Thailand Fr. Martinique Fr. Martinique Thailand Mexico Fr. Martinique USA Venezuela Fr. Guadeloupe Fr. Guiana Benin Fr. Martinique Fr. Martinique Fr. Martinique Fr. Guiana Benin Thailand China Togo Thailand Burundi RDCongo RDCongo Fr. Martinique Brazil Brazil USA Fr. Martinique Fr. Martinique JF691551 JF691552 ZR TBa ZR JF JF JC JF SK JF JF SK GM JF JH LP CLu JLC AK JF JF JF CL SY KW ZR LP OM JR MG MG JF JB AMe JS JF RC 03/06/2006 07/06/2005 02/08/2005 28/06/1999 12/07/2000 25/07/2010 Chiang Mai, Mae Taeng Chiang Mai, DSPNP Chiang Mai, Mae Taeng Fort-de-France Lamentin Chiang Rai, MFL University In In In In BBH BBH BBH LIP LIP 29/04/2010 02/06/2001 Chiang Rai, Muang Gros-Morne, Rivière Rouge In forest In forest 21/08/2009 07/2010 06/07/1998 07/2006 01/04/2006 29/11/2005 02/03/2009 20/06/1998 02/04/2000 27/06/1999 10/10/2000 27/02/2007 11/07/1999 20/06/2010 25/07/2010 12/05/2010 08/06/2006 17/11/1978 11/1953 12/1954 26/12/2001 17/04/2005 02/04/1991 07/2007 26/09/2000 27/08/2005 Chiang Mai, Mae Taeng Veracruz, Coatepec Prêcheur, Anse Couleuvre Suffolk Co., New York Choroni Sainte-Rose, Trace de Sofaïa Cayenne, La Mirande Wari Maro, Borgou Pointe banane Saint-Esprit, Bois la Charles Case-Pilote, Morne Rose Sinnamary, Parcelle Guyaflux Wari Maro, Borgou Chiang Mai, Doi Suthep Yunnan, Xishuangbannan Ola Chiang Mai, Mae Taeng Mugara Panzi, Kivu Panzi, Kivu Trinité, La Caravelle Rio de Janeiro Paraná, Paranagua Bernalillo Co., New Mexico Sainte-Anne Trinité, La Caravelle In forest JF727849 JF727850 JF514530 JF727851 JF727852 JF797197 JF727870 JF727853 JF896226 JF797183 JF727857 JF727869 JF514517 JF797198 JF727854 JF797199 JF514540 JF514538 JF797184 JF691548 JF514541 JF797200 JF797185 JF797196 JF896228 JF797201 forest forest forest lawn In grassland LIP In forest Mulch On soil In forest In forest Litter, humus In forest In forest In wood litter In forest Oil palm plantation Grevillea Coffee plantation In forest In forest edge In dune Sandy soil In forest In forest LIP LIP XAL LIP SFSU LAPAM LIP BR LIP LIP LIP LIP BR LAPAF BBH BR BR BR LIP LAPAM BR SFSU LIP LIP 285 N°a Fungal Diversity (2011) 51:279–296 Table 1 (continued) 286 Table 1 (continued) Sample Identificationb 92 93 93 93 93 95 96 96 96 96 96 100 100 100 101 102 102 102 104 105 105 106 106 107 108 109 109 110 111 111 112 113 114 115 116 116 F2286L NTF67 ZRL2036 NT001 ZRL2134 Thoen7297 ZRL2127 OR71 ZRL3028 CA798 ZRL2053 F2815 CL/MART03.055 F2343 SP307818 NTT37 ZRL2110 ZRL3039 CA848 ZRL3088 NTSCR1 NTT33 NTS73 NTF26 CA846 ZRL2044 ecv3614 NT055 ADK2905 CA847 ZRL3101 NT62 NTT72 CA843 ZRL3080 NTF063 A. A. A. A. A. A. A. A. A. fiardiih sp.h sp.h sp.h sp.h martinicensis martinicensis martinicensis martinezianus (A. cf bulbillosus) Country Fr. Martinique Thailand Thailand Thailand Thailand Senegal Thailand Thailand Thailand Thailand Thailand Fr. Martinique Fr. Martinique Fr. Martinique Brazil Thailand Thailand Thailand Thailand Thailand Thailand Thailand Thailand Thailand Thailand Thailand Thailand Thailand Benin Thailand Thailand Thailand Thailand Thailand Thailand Thailand Genbank JF514529 JF514542 JF691547 JF727855 JF896227 JF514537 JF727864 JF691543 JF514535 JF514526 JF727865 JF691540 HM488742 JF727846 JF514520 JF691544 JF727845 JF514539 JF727866 JF691542 Collectorc Date Locationd Habitat Herbariume JF SK RL SK KH DT KH OR ZR JG-SK DD JF-RC-CLu CL JF UP KW ZR TBa JG-GB ZR SK KW SK NT JG-GB ZR EV SK AK JG-GB ZR SK SK JG-GB ZR NT 29/09/2000 17/08/2010 25/06/2005 150/8/2009 28/08/2005 14/10/1984 16/08/2005 03/07/2010 05/06/2006 28/07/2010 27/06/2005 02/09/2003 02/09/2003 19/11/2000 14/12/2001 20/06/2010 03/08/2005 08/06/2006 25/07/2010 02/08/2006 08/03/2010 18/06/2010 30/06/2010 29/06/2010 25/07/2010 26/06/2005 Saint-Anne, Morne Manioc Chiang Rai, Muang Chiang Mai, Mok Fa W Chiang Rai, Khun Kone W Chiang Mai, Mae Taeng Dakar Chiang Mai, Mae Taeng Chiang Mai, Mae Taeng Chiang Mai, Doi Inthanon Chiang Mai, Nat.Park Chiang Mai, Doi Inthanon Prêcheur, Anse Couleuvre Prêcheur, Anse Couleuvre Prêcheur, Anse Lévrier Sao Paulo Chiang Mai, Mae Taeng Chiang Mai, Mae Taeng Chiang Mai, Mae Taeng Chiang Mai, University P Chiang Rai, Pamae Lao P Chiang Rai, Muang Chiang Mai, DSPNP Chiang Mai, Mae Taeng Chiang Mai, Doi Suthep Chiang Mai, University P Chiang Mai, Mae Taeng From GenBank Chiang Mai, DSPNP Wari Maro, Borgou Chiang Mai, University P Chiang Mai, Mae Taeng Chiang Mai, Mae Taeng Chiang Mai, Mae Taeng Hua-Hin, Golf Chiang Mai, Mae Taeng Chiang Mai, Mae Taeng In forest In forest In forest In forest In forest In market In forest In forest In forest In forest In forest In forest In forest In forest On soil In forest In forest In forest In forest In forest In forest In forest In forest In forest In forest In forest LIP 15/08/2009 19/09/2000 25/07/2010 13/09/2006 20/08/2009 07/03/2010 04/08/2010 10/07/2006 28/07/2010 In In In In In In In In In forest sandy soil forest forest forest forest grassland forest forest BBH BBH BR BBH BBH BBH LIP LIP LIP BBH BBH BBH BBH BR BBH BBH Fungal Diversity (2011) 51:279–296 N°a N°a Sample 117 119 120 126 127 128 128 128 ZRL3091 MATA774 ADK2751 ZRL3056 ZRL3102 LD030 ZRL2124 NTS106 Identificationb A. goossensiae Country Genbank Collectorc Date Locationd Habitat Herbariume Thailand Mexico Benin Thailand Thailand Thailand Thailand Thailand JF691546 JF727871 JF514519 JF691541 JF691545 JF514521 ZR GM-PC AK ZR ZR JC KH SK 02/08/2006 29/11/2007 05/06/2000 12/06/2006 15/09/2006 03/08/2010 12/08/2005 24/07/2010 Chiang Rai, Pamae Lao P Veracruz, San Andrés Tuxtla Calavi Campus, Atlantique Chiang Mai, Mae Taeng Chiang Mai, Ob Luang P Chiang Rai, MFL University Chiang Mai, Mae Taeng Chiang Mai, Mae Taeng In forest In dune In sandy soil In forest In forest On litter In forest In forest BBH XAL BR BBH BBH MFLU BBH Fungal Diversity (2011) 51:279–296 Table 1 (continued) a Number of species (as numbered in the phylogenetic tree); when there are redundant samples belonging the same presumed species, the first one appearing in the list was used in the phylogenetic analyses and the genbank accession number of its ITS1+2 sequence is given in the Table b T, Type (holotype, paratype or isotype); the species names in brackets require complementary studies of the collection to be confirmed or not. c AG, A. Guinberteau; AK, A. De Kesel; AM, A. Marchand; AMe, A. De Meijer; CL, C. Lechat; CLu, C. Lecuru; DD, D. Desjardin; DT, D. Thoen; EH, E. Horak; EG, E. Grand; EM, E. Maes; EV, E. C. Vellinga; FG, K. Grinling; FH, F. Hendrickx; GB, G. Barroso; GM, G. Mata; JB, J. Borovicka; JC, J. Chen; JF, J.P. Fiard; JG, J. Guinberteau; JH, J. Horman; JK, J. Kerekes; JLC, J.-L. Cheype; JR, J. Rammeloo; JS, J. W. Sparks II; KH, K. D. Hyde; KW, K. Wisitrassameewong; LP, L. Parra; MG, M. Gossens-Fontana; NT, N. Thongklong; OJ, O. Juhasz; OM, O. Myo Aung; OR, O. Raspé; ORo, O. Roblot; PC, P. Callac; PS, P. Sysouphanthong; PW, P. Wenzel; R. Courtecuisse; RK, R.W. Kerrigan; SK, S. Karunarathna; SY, S. Yorou Norou; TB, T. Boonpratuang; TBa, T. Baroni; TBo, T. Bouchara; UP, U.C. Peixoto; ZR, R.L. Zhao; d P, Park or National Park; W, Waterfall; MNHN, Museum National d’Histoire Naturelle; DSPNP, Doi Suthep Pui National Park. e SFSU: Harry D. Thiers Herbarium, Department of Biology, San Francisco State University. USA; BBH: Biotec Bandkok Herbarium, National Science and Technology Development Agency, Klong Luang, Pathumthani, Thailand; LIP: Herbier du Département de Botanique, Faculté des Sciences Pharmaceutiques et Biologiques, Université de Lille, Lille, France; BR : National Botanic Garden of Belgium Herbarium, Meise, Belgium; CGAB:Collection du germoplasme des agarics à Bordeaux, INRA, Bordeaux, France; XAL : Herbario del Instituto de Ecología, A.C., Xalapa, Veracruz, Mexico; LAPAM, LAPAF, LAPAG: Luis Alberto Parra private herbarium for America, Africa and Europe respectively, Aranda de Duero, Burgos, Spain. MFLU: Mae Fah Luang University, Chiang Rai Prov., Thailand. f syn. A. spissicaulis g syn. A. porphyrizon h A. subrufescens complex 287 288 At INRA, the reaction mix (final volume 25 μL) contained 5 μL PCR GO Taq buffer (5X, Promega), 2.5 μL dNTP mix (1.2 mM, Eurobio), 0.5 μL BSA (10 mg/mL, Promega); 1 μl of each primer (25 μM); 0.2 μL Taq polymerase (5 U/μL, Go Taq Promega); 1 μL DNA extract; ddH2O up to 25 μL. The PCR profile was 5 min at 95°C; 35 cycles (1 min at 94°C, 1.5 min at 55°C, 1.5 min at 72°C); 5 min at 72°C. At BR, amplifications were usually performed in 50 μL reactions using DreamTaq DNA polymerase (Fermentas), according to manufacturer’s instructions, except that reactions contained 0.2 mg/mL BSA (Fermentas) and 0.25 μM of each primer. Cycling conditions were as follows: 3 min at 95°C; 35 cycles of 1 min at 94°C, 1.5 min at 55°C, 1.5 min at 72°C; 5 min at 72°C. At SFSU and HKU, the thermal cycles consisted of 3 min at 94°C, 30–35 cycles of 1 min at 94°C, 50 sec at 52°C and 1 min at 72°C, with a final extension step of 72°C for 10 min. Sequencing was performed on ABI Prism Genetic analysers (Applied Biosystems) at the following institutions: Beckman Coulter Genomics, England; Macrogen, Republic of Korea; Genome Research Centre of the University of Hong Kong; Department of Biology, San Francisco State University. Phylogenetic analyses Based on our previous knowledge on the circumscription of species in Agaricus, sequences were considered as redundant in the three following cases: they were identical; they differed only by heteromorphisms and shared presumed alleles; they differed at a single polymorphic position and possibly by heteromorphisms at others. The entities defined above must be considered as presumed species because such a method might in some cases be too stringent for taxonomic purposes. After alignment using T-Coffee ver 8.99 (Notredame et al. 2000), corrections were made by hand, firstly because TCoffee did not interpret the heteromorphisms, secondly to suppress highly variable or ambiguous positions. The maximum likelihood (ML) analysis was performed on the ATGC bioinformatics platform. The phylogenetic tree was constructed using the ML method implemented in the PhyML ver 3.0 aLRT (Guindon and Gascuel 2003; Anisimova and Gascuel 2006). This method does not allow partitions. The GTR substitution model was selected with an estimated proportion of invariable sites of 0.438 and assuming 4 gamma-distributed rate categories to account for rate heterogeneity across sites. The gamma shape parameter was estimated directly from the data (gamma= 0.782), and the likelihood was increased by using the SPR tree improvement. Reliability of internal branches was assessed using the approximate Likelihood-Ratio test (aLRT) which assesses that the branch being studied provides a significant likelihood gain, in comparison with Fungal Diversity (2011) 51:279–296 the null hypothesis that involves collapsing that branch but leaving the rest of the tree topology identical (Anisimova and Gascuel 2006). The analysis included 100 bootstrap replicates. We also performed a Bayesian analysis with MrBayes 3.1 (Huelsenbeck and Ronquist 2001; Ronquist and Huelsenbeck 2003). We first estimated the best evolutionary model for ITS1 and ITS2 separately with jModeltest (Posada 2008), using the Bayesian information criterion, and restricting the explored evolutionary models to the ones implemented in MrBayes. The best models were GTR + Γ + I and HKY + Γ + I, respectively. We partitioned the data and set the evolutionary models accordingly in MrBayes. The model parameters were unlinked across partitions. We ran two parallel analyses, each with one cold and three heated chains, with tree sampling every 200th generation. Since the two runs converged slowly (the average deviation of split frequencies decreased slowly), we ran the analyses for 10×106 generations, and when summarizing the tree samples the first 12,000 trees were discarded. The burnin was determined by inspecting the log-likelihood by generation plot generated with Tracer ver 1.5 (Rambaut and Drummond 2007). Graphical representation and editing of the phylogenetic tree were performed with TreeDyn ver.198.3 (Chevenet et al. 2006). Results Sampling for analyses using ITS1+2 sequences We obtained complete sequence data from the 178 samples presented in Table 1. Ten of these were from types specimens, 51 tropical samples were collected in 2010, mainly in Thailand but also in China, Togo and Mexico; the oldest herbarium collection was from 1939. Among the 178 samples, 89 were from Asia, 37 from the Americas, 28 from Europe, 21 from Africa, and 3 from Oceania. Complete and non-redundant sequences for 128 samples, representing 124 presumed Agaricus species and four species of closely related genera, were used in the phylogenetic analyses; these were numbered from 1 through to 128 according to their phylogenetic placement. A total of 110 new sequences have been deposited in GenBank. The final 590-character alignment has been deposited in TreeBase (ID number 11249). Four nonAgaricus species represented the outgroups. Among the 124 Agaricus collections included in the study, 38 were from temperate areas and 86 from tropical or subtropical areas. Among the 86 tropical collections of Agaricus plus four collections of related genera, 25 have been identified to species (Table 1). Limited but important data concerning Fungal Diversity (2011) 51:279–296 289 redundancy can be summarized from this Table, as follows: first, pairwise sequence comparisons indicated that 60 putative species from the tropics were found once, 15 two times, 10 three times, and 5 four times or more. Second, for six presumed species of tropical Agaricus, redundant sequences from different continents were found: four in Africa and Asia (5: A. heterocystis Heinem. & Gooss.-Font.; 37: A. microvolvatulus Heinem.; 44: A. xanthosarcus Heinem. & Gooss.-Font.; and 111), one in Africa and America/Caribbean (46: A. aff. volvatulus Heinem. & Gooss.-Font.), and one in Asia and America/Caribbean (61). The 80 remaining species of Agaricus were found on a single continent: 47 in Asia, 22 in the Americas, and 11 in Africa; some of them have nearly but not entirely redundant sequences and belong to presumed closely related species, e. g. 26 and 27 (A. trisulphuratus Berk. group), or 92 and 93 (A. subrufescens complex); further studies are necessary to clarify their taxonomic status. Fig. 1 Most likely ML phylogram based on ITS1+2 sequences of 124 species of Agaricus. The SH-like branch support values above 50%, the bootstrap support values above 50% (after slash) and the posterior probabilities of 50% majority rule consensus Bayesian tree (in bold type) are shown. Red and blue branches refer to tropical and temperate species of Agaricus, respectively. Branch label indications are: the number of the species (outgroups from 1 to 4; Agaricus species from 5 to 128), the continent (Africa AF, the Americas AM, Asia AS, Europe EU, and Oceania OC), the section for the temperate species only (Bivelares BIV, Chitonioides CHI, Sanguinolenti SAN, Agaricus AGA, Spissicaules SPI, Arvenses ARV, Minores MIN), and the species name when known, T meaning type specimen. Sections are indicated as above and major tropical clades are numbered from TR I to TR VII for the well-supported ones and TR a to TR d for the others Phylogenetic results The most likely ML tree obtained by analysis is presented in Fig. 1. The branches having SH-like support values lower than 50% are collapsed. SH-like support values and Bayesian posterior probabilities of the 50% majority rule consensus tree are showed. Bootstrap support values were mostly lower than 50%; those above 50% are showed. Most of the major clades were similar and relatively well supported in the ML and Bayesian analyses with SH-like support values and posterior probabilities greater than 80%, while the bootstrap support values were sometimes much lower. Despite numerous polytomies and 290 Fungal Diversity (2011) 51:279–296 Fig. 1 (continued) topological differences, the ML and Bayesian trees remain congruent enough to have most of their major clades globally appearing in the same order. Because deep branches, except the branch bearing species 66 to 128, are very short, we preferred to emphasise the content of the clades (detailed in Table 2) rather than phylogenetic relationships between clades. One consequence of these short branches is the low bootstrap support values that we obtained, while aLRT SH-like values are less sensitive to the shortness of the branches. Phylogenetic distribution of the temperate species Temperate species, except A. aridicola Geml, Geiser & Royse (103), were distributed into eight accepted taxonomic sections of subgenus Agaricus. The sections Bivelares, Chitonioides, Xanthodermatei, Sanguinolenti and Agaricus are related. In both trees, they are monophyletic, except the section Sanguinolenti which is paraphyletic and constituted three clades. Among the seven clades of these five sections, four are entirely temperate and the three remaining ones have an early temperate branch. The section Spissicaules which is a less well-known section of the subgenus, is constituted by three clades and two branches that form a large polytomy with clade TR II and clade TR III in the Bayesian tree. In the ML tree, this section is paraphyletic with clade TR III. The possible polyphyly of section Spissicaules and the phylogenetic relationships with the closely related clades TR II and TR III remain unresolved. The sections Arvenses and Minores with the exception of the secotioid species A. aridicola are monophyletic in the Bayesian tree and in the ML tree. Twelve temperate species that have not been included in previous phylogenetic analyses were also taxonomically distributed in the eight sections. Temperate species, except A. aridicola and two species of the section Spissicaules which is not clearly resolved, are distributed in 12 clades (Table 2). All of these clades have SH-like support values greater than 80%, eight of them have Bayesian posterior Fungal Diversity (2011) 51:279–296 291 Table 2 Phylogenetic and geographic distribution of 90 tropical species into tropical clades and sections based on temperate species Sections, clades and branches ML/Bayesian branch supporta Species as numbered in the tree Species of other genera in a single clade Sections based on temperate species Bivelares Chitonioides Xanthodermatei Sanguinolenti I+II+III (3 clades) Agaricus Spissicaules (polytomies with 3 clades) Arvenses Number of tropical species Species found in: 2 continents ASb AFb AMb (see also Fig. 1 and Table 1) 100/100–100 1–4 4 0 3 1 0 98/98–100 81/55–98 85- 28–29 30–33 34–48 0 0 10 0 0 3 0 0 7 0 0 3 0 0 3 84–87, 83–56, 89/62–100 49–56 2 0 2 0 0 92–99 95/86–100, 88–70, 99/93–100 57–62 72–81 3 3 1 0 2 0 0 1 2 2 A. aff. argyropotamicus 99/87–100 92–99 4 0 2 1 1 A. fiardii 1 0 1 0 A. Goossensiae Minores 89–51 115–128 8 0 6 Isolated branch 103 0 0 0 (A. aridicola) Tropical branches and clades of presumed species found in Africa and/or in Asia (with the exception of Tropical clade TR I 91–82 11–20 10 0 8 Tropical clade TR III 100/100–100 83–88 6 0 2 Tropical clade TR V Tropical clade TR VI Tropical clade TR VII Tropical clade TR a (poorly supported) Tropical clade TR b (poorly supported) Isolated tropical branches Identified tropical species A. microvolvatulus, A. xanthosarcus A. aff. endoxanthus, A. aff. volvatulus species 88 in clade TR III) 2 0 A. inoxydabilis 3 1 A. kivuensis, A. aff. impudicus 0 0 1 0 0 0 0 0 93/89–100 93/55–92 99/91–100 59–60 104–107 108–111 112–113 8–9 4 4 2 2 0 1 0 0 4 4 2 2 82- 22–27 6 0 4 2 0 5, 6, 7, 10, 21 8 1 7 2 0 65, 102, 114 Tropical branches and clades of presumed species found in the Americas Tropical clade TR II 96–99 66–71 6 0 0 0 6 Tropical clade TR IV Tropical clade TR c (poorly supported) Tropical clade TR d (poorly supported) Isolated branch Total for tropical species of Agaricus In sections Out of sections cf. He. splendidissimus, Hy. ardosiicolor 97/96–100 85–61 89–91 63–64 3 2 0 0 0 0 0 0 3 2 67- 100–101 2 0 0 0 2 82 5–128 1 86 0 6 0 52 0 16 1 24 30 (35%) 56 (65%) 4 2 19 33 6 10 9 15 a A. campestroides, A. trisulphuratus A. heterocystis, A. subsaharianus A. aff. rufoaurantiacus, A. cf. floridanus A. deserticola A. martinicensis, A. martinezianus SH-like branch support values above 50%, bootstrap support values above 50% (after a slash), and posterior probabilities of 50% majority rule consensus Bayesian tree (after a dash) b AS=Asia, AF=Africa, AM=the Americas; the six species found in two continents are counted two times 292 probabilities greater than 80%, and four of the eight also have bootstrap support values greater than 80%. Phylogenetic and geographic distributions of the tropical species The four non-Agaricus species representing outgroups, formed a clade sister to Agaricus in both trees; although it was not necessary to impose them as outgroups in the ML analysis. This clade was supported at 100% by all the methods. These outgroups are from tropical Africa or Asia. Thirty five percent of the tropical species of Agaricus (30 species) were distributed amongst six of the eight accepted sections of the subgenus Agaricus (Table 2). In other terms, about one-third of tropical (including subtropical) Agaricus species nested in seven of the 12 clades based on temperate species. The 56 remaining tropical species were distributed in eleven exclusively tropical clades and in nine isolated branches detailed in Table 2. Among the eleven tropical clades detected in the analyses, seven that are well supported by both analyses are numbered as TR I to TR VII. Both SH-like branch support and Bayesian posterior probability were greater than 95% for four of these seven clades, greater than 90% for two, and greater than 80% for the remaining clade. Moreover, the four clades III, IV, V, and VII have high bootstrap support values of 100, 96, 89, and 91 respectively. Geographic information appears highly correlated with the phylogenetic data although African, Asian and American samples were not equally represented. Firstly, all basally joined species including those of the closely related genera, from 1 to 20, are tropical and were found in Africa and/or in Asia, but not in the Americas. Secondly, among the 11 tropical clades, three contained eight species collected in Asia, three contained 20 species found only in Africa and/or Asia, four contained 13 species found only in the Americas, and one contained six species from the three continents with a single species from the Americas (87 from Martinique). Samples from Africa were underrepresented, thus exclusively African clades possibly exist. Likewise, it is possible that the exclusively Asian clades would be in fact African and Asian. In Table 2, tropical clades and isolated branches of Africa and Asia are separated from those of the Americas. The separation between the American and the African+ Asian tropical species is much less pronounced in the sections based on temperate species than in the tropical clades although the tropical species are 50% less numerous in the former than in the latter. Five sections contain both American and African+Asian tropical species: the four monophyletic sections Xanthodermatei, Agaricus, Arvenses, Minores, and the possibly polyphyletic section Spissicaules. Section Sanguinolenti contains tropical species from Asia Fungal Diversity (2011) 51:279–296 only and the two remaining sections Bivelares and Chitonioides have no tropical species. Moreover, the only two species for which samples were found in the Americas and in Africa+Asia (46 and 61) belong to clades of the monophyletic sections Xanthodermatei and Agaricus. Discussion The number of recognized Agaricus species lies between 200 and 250 according to Bas (1991) and is estimated as ca 200 mostly temperate species in Kirk et al. (2008). These estimations are lower than our own calculation of 386 recognized species among which 203 are temperate and 183 are tropical. This estimation is updated with the most recently described species but remains approximate because some synonymies have not been detected between species described on different continents or climates. Thus, the numbers of tropical and temperate recognized species are quite similar. In our analyses, we used sequences from 86 tropical species and 38 temperate species carefully selected among about 100 species to represent their distribution in the eight presently recognized temperate sections of subgenus Agaricus (Challen et al. 2003; Kerrigan et al. 2006; Kerrigan et al. 2008; Parra 2008). Tropical and temperate samples are approximately equally represented. Concerning the number of species in existence, Bas (1991) recognized 200-250 described species in Agaricus, but estimated there were 300–400 species worldwide. We believe this is also an underestimate since it is similar to the number of presently recognized species and since probably most of the species collected in Thailand in the present study are new. The total number of species collected in only a small part of northern Thailand during the last five years, including the about 50 used in the present study, is likely to be more than the 70 species recognized in Europe over one century. From the present study, we have also established that a large part of the American samples are taxonomically different from those of Africa and Asia. The number of tropical species is certainly much greater than the number of temperate species and the total number of species in existence should be therefore much greater than 400. In the genus Agaricus ITS sequences were used for species characterization or for phylogenetic analyses of the sections of the genus; however, at the scale of the entire genus, it appears that most of the deep branches are short suggesting that major evolutionary radiations would have occurred in a relatively short time. In other respects, the low bootstrap support values of the deep branches are partly due to their shortness. The present analysis was not aimed at establishing a very strong phylogeny for the genus, but identifies the major clades and the distribution of the tropical species among these clades. Fungal Diversity (2011) 51:279–296 There is no major change in the classification of the temperate species into eight sections with the inclusion of tropical species plus an additional dozen temperate species never included in previous phylogenetic analyses (except for the secotioid species A. aridicola). The nLSU rDNA sequence of our collection CA101 of A. aridicola has previously been obtained by Moncalvo et al. (2002) and also used by Geml et al. (2004) and Capelari et al. (2006) in analyses. This species, exhibiting a positive Schaeffer reaction (orange, red or violaceous-purple discoloration when aniline and nitric acid are consecutively applied on the same spot of a sporophore) and an odor of almond, was primarily assigned to section Minores according to the data of Geml et al. (2004). However, tropical samples were mostly absent in this study and our present data better agree with the study of Capelari et al. (2006) who found that A. aridicola was more closely related to A. martinezianus than to sections Arvenses and Minores. With caution, we do not include A. aridicola in the clade TR V which would be much less supported with this species. In other respect, the ‘temperate’ status of this secotioid species is doubtful since it is rare in France but abundant in Israel according to Wasser (2002). To classify the species in climatic groups is not easy: the geographical range of some temperate species such as A. bisporus and A. bitorquis extends into tropical areas and, reciprocally, the tropical species A. subrufescens exists also in Europe (Kerrigan 2005). Moreover, some species such as the tropical species A. endoxanthus Berk. & Broome are sometimes found in greenhouses (Parra et al. 2002) in Europe and are suspected to have been introduced with plants. Our analyses are similar to those of Geml et al. (2004) and show that the three secotioid species, A. aridicola, A. deserticola G. Moreno, Esqueda & Lizárraga, and A. inapertus Vellinga, do not share a common ancestor. The secotioid form, considered as an adaptation to arid environment, has therefore evolved independently several times in a large clade that includes sections Arvenses and Minores, and five related tropical clades (samples from 89 to 128). Two-thirds (56/86) of the tropical or subtropical species did not cluster in the recognized sections based on temperate species. Nine species occurred on isolated branches, 12 form poorly supported clades, TR a to TR d, and 35 belong to seven well-supported clades, I to VII. We accept these seven tropical clades of which four are more strongly supported in the analyses (III, IV, V, and VII). Characterization of these tropical clades will be challenging. There are at least four examples where tropical species share some characteristics with the recognized temperate sections, although they do not belong to these sections. (i) Two samples both having affinity with A. impudicus (Rea) Pilát, a species of section Spissicaules, belong to two 293 different clades: the temperate species (81) to a clade of the section Spissicaules as expected, but the tropical species (85) to the most closely related tropical clade TR III. (ii) The sample of A. subsaharianus L.A. Parra, Hama & De Kesel (10) has traits that, until now characterized only sections Spissicaules, Minores and Arvenses (Hama et al. 2010), but A. subsaharianus is not related to these sections. (iii) A. heterocystis is not in section Arvenses as stated by Heinemann (1978) but on a non-related isolated branch. (iv) A. inoxydabilis Heinem. is not in section Sanguinolenti as stated by Heinemann (1980) but a member of clade TRI. Specimens belonging to the tropical clades can be used to identify these clades to the tropical section Brunneopicti or the subgenus Lanagaricus proposed by Heinemann (1956a). For the section Brunneopicti, the clade TR III is a plausible candidate since it contains A. kivuensis Heinem. & Gooss.-Font (87) that Heinemann included in this clade, but clade TR I could be the best candidate because it contains a specimen identified as A. brunneopictus Heinem. & Gooss.-Font (16), the type species of the section. Unfortunately we were unable to confirm this identification as the specimen we examined is immature. Further studies will be necessary to clarify what clade represents this section and how this section mainly based on the morphology of the cap can be better characterized. Heinemann acknowledged that this section was not well characterized (“Brunneopicti mais cette section est mal caracterisée”; Heinemann 1984). A sample of A. trisulphuratus (26), the type species of subgenus Lanagaricus was used in our analysis. It clustered in clade TR b which in our ML tree belongs to a subclade of subgenus Agaricus since it also includes sections Bivelares, Chitonioides and Xanthodermatei. In Vellinga et al. (2011) A. trisulphuratus also clusters in a clade of the subgenus Agaricus; unfortunately, in both cases the phylogenetic position of the clade is not well-supported. The confirmation of such a clade into subgenus Agaricus would imply the inclusion of all sections of subgenus Lanagaricus in subgenus Agaricus and disappearance of subgenus Lanagaricus. In our analysis, the clade TR b is poorly supported, but it contains two strongly supported African+Asian sister clades: one (25-26-27) with A. trisulphuratus and the other clade (2324) with A. campestroides Heinem.(23). The hypothesis that the whole clade TR b would represent the section Lanagaricus is reinforced by the uncertain classification of A. campestroides: this species was initially placed in section Agaricus (Heinemann 1956a) and later combined in genus Micropsalliota (Heinemann 1988). However, according to our microscopic studies it can be neither a member of section Agaricus nor Micropsalliota (data not shown). Although some of the tropical clades from the present study will probably use the section names or subgenera described by Heinemann, their characterization 294 and their circumscription need consideration. Only onethird of tropical Agaricus species are distributed in the recognized infrageneric organization of the genus based on temperate species. Although some species among the remaining two-thirds could be included in the tropical sections proposed by Heinemann, it appears necessary to expand the number of sections in the genus. The fact that numerous clades are exclusively tropical indicates that geography and climate have had a major impact on the evolution of the genus. Adaptation to temperate climates has never occurred in certain genera of Agaricaceae such as Micropsalliota (Zhao et al. 2010). A relative tolerance to cold is required in temperate climates and the ability to fruit at ambient temperatures that is required under warm climates is not advantageous in temperate regions because the summer is not the wettest season and consequently the most favorable season for fruiting. Largeteau et al. (2011) showed that the rate of wild isolates able to fruit at 25°C is 100% in A. bisporus var. burnettii Kerrigan & Callac, a variety that lives in hot climates, while it is only about 50% in the populations of A. bisporus var. bisporus living in temperate climates. Our data also indicate that clades of four of the eight recognized temperate sections include both American and African+Asian tropical species. To our knowledge, all temperate sections contain North American and Eurasian temperate species, and the distribution range of numerous temperate species also extends to both continents (Challen et al. 2003; Kerrigan et al. 2006; Kerrigan et al. 2008). These species probably migrated via land before the complete separation of these continents. Moreover, for some species, such as A. bisporus and A. bitorquis, subpopulations that adapted to (sub)tropical climates in Africa or in North America are known. For example, the geographic range of A. bisporus extends from the boreal region of Alaska (Geml et al. 2008) to the equatorial climate of Congo (Heinemann 1956a) in Africa and to the hot and dry climate of the Sonoran Desert of California in America (Callac et al. 1993); the ancestral condition of such species, however, remains unknown. Finally, a similar hypothesis for the amphiatlantic distribution of the temperate species and for the American and African+Asian distribution of the tropical species belonging to the same clades as the temperate species needs consideration. Forty-seven tropical species are distributed in 11 tropical clades. Seven of these clades contain species exclusively from Africa+Asia with the exception of one species in clade III, while the four remaining clades contain species exclusively from the Americas. The different phylogenetic distribution of the American species versus the African+Asian species appears to be a reliable pattern since four tropical clades contained only species found in America although the Asian samples were overrepresented compared to the American Fungal Diversity (2011) 51:279–296 samples. The distinct African+Asian and American tropical clades also suggest that species diversification probably occurred independently in Africa+Asia and in the Americas. Moreover, some clades belonging to the African+Asian group are not phylogenetically closely related, indicating that species diversification occurred several times on these continents. This is also possible in the Americas but less evident from the data. One of the four species used in the outgroup and that does not belong in Agaricus, was identified as Hymenagaricus ardosiicolor (Heinem.) Heinem. while another shares some characteristic traits with Heinemannomyces splendidissimus but others with Hymenagaricus and Xanthagaricus. However, when other GenBank sequences of Hymenagaricus species were included in the analyses (data not shown), they diverged more from the Agaricus clade than the four species that we used, with the exception of Hymenagaricus epipastus (Berk. & Broome) Heinem. & Little Flower recently reported from Thailand by Zhao et al. (2010). Further studies will be necessary to clarify the taxonomic rank of these taxa. In Vellinga et al. (2011) one of the two species Heinemannomyces splendisissimus or Clarkeinda trachodes (Berk.) Singer is sister to the monophyletic genus Agaricus depending on the analysis. Hymenagaricus epipastus, Hymenagaricus ardosiicolor, Heinemannomyces splendisissimus, Clarkeinda trachodes and two unidentified species in our outgroups are from tropical Africa or Asia. The geographical origin of Agaricus is unknown, however, the following evidences suggest a paleotropic origin. (i) Species of genera known to be the most closely related to Agaricus species are from tropical Africa or Asia. (ii) All the early branches and clades of our trees (from species 1 to 20) contain species exclusively from tropical Africa or Asia; and (iii) the first African+Asian clades appeared before the first American ones. This hypothesis remains to be confirmed because these basal branches and clades are not strongly supported in the analyses (Fig. 1). The analyses should also be extended to include species from tropical Australia and temperate areas of Asia and the South Hemisphere to better understand the biogeography of the genus and possibly correlate migrations with the known periods of connection between continents. Acknowledgments The authors are grateful to Jean-Pierre Fiard, Else Vellinga, Gerardo Mata, Marina Capelari, Marc-André Lachance, Rick Kerrigan, Peter Wenzel, Dario De Franceschi, Vincent Lefort, Stephane Welti and Saturnino (Nino) Santamaría. This work was supported by an Integration Research Grant from the European Distributed Institute of Taxonomy (EDIT). The National Science Foundation (USA) (PEET-grant DEB0118776 to Desjardin), the National Natural Science Foundation of China (Project ID: 31000013), and the project “value added products from Basidiomycetes: Putting Thailand’s biodiversity to use” (BRN049/2553) are thanked for providing partial support to this research. The Global Research Network for Fungal Biology and King Saud University are also thanked for support. Fungal Diversity (2011) 51:279–296 References Angeli JPF, Ribeiro R, Gonzaga MLC, Soares S de A, Ricardo MPSN, Tsuboy MS, Stidl R, Knasmueller S, Linhares RE, Mantovani MS (2006) Protective effects of β-glucan extracted from Agaricus brasiliensis against chemically induced DNA damage in human lymphocytes. Cell Biol Toxicol 22:285–291 Anisimova M, Gascuel O (2006) Approximate likelihood ratio test for branches: A fast, accurate and powerful alternative. Syst Biol 55:539–52 Bas C (1991) A short introduction to the ecology, taxonomy and nomenclature of the genus Agaricus. In: Van Grievsen LJLD (ed) Genetics and breeding of Agaricus. Proceedings of the First International Seminar on Mushroom Science. Pudoe Wageningen pp 21–24 Baker RED, Dale WT (1951) Fungi of Trinidad & and Tobago. Mycol Papers 33:1–123 Berkeley MJ, Broome CE (1871) The fingi of Ceylon (Hymenomycetes from Agaricus to Cantharellus. J Linn Soc 9:494–567 Bernarshaw S, Lyberg T, Hetland G, Johnson E (2007) Effect of an extract of the mushroom Agaricus blazei Murill on expression of adhesion molecular and production of reactive oxygen species in monocytes and granulocytes in human whole blood ec vivo Acta Pathologica. Microbiol Immunol 115(6):719–725 Callac P, Billette C, Imbernon M, Kerrigan RW (1993) Morphological, genetic, and interfertility analyses reveal a novel, tetrasporic variety of Agaricus bisporus from the Sonoran desert of California. Mycologia 85:835–851 Capelari M, Rosa LH, Lachance M-A (2006) Description and affinities of Agaricus martineziensis, a rare species. Fungal Diversity 21:11–18 Challen MP, Kerrigan RW, Callac P (2003) A phylogenetic reconstruction and emendation of Agaricus section Duploannulatae. Mycologia 95(1):61–73 Chevenet F, Brun C, Banuls AL, Jacq B, Christen R (2006) TreeDyn: towards dynamic graphics and annotations for analyses of trees BMC. Bioinformatics 7:439 Doyle JJ, Doyle JL (1987) A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochem Bull 19:11–15 Geml J, Geiser DM, Royse DJ (2004) Molecular evolution of Agaricus species based on IST and LSU rDNA sequences. Mycol Prog 3(2):157–176 Geml J, Laursen GA, Taylor DL (2008) Molecular diversity assessment of artic and boreal Agaricus taxa. Mycologia 100 (4):577–589 Guindon S, Gascuel O (2003) A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Syst Biol 52:696–704 Hama O, Maes E, Guissou M-L, Ibrahim DM, Barage M, Parra LA, Raspé O, De Kesel A (2010) Agaricus subsaharianus, une nouvelle espèce comestible et consommée au Niger, au Burkina Faso et en Tanzanie. Cryptogamie, Mycologie 31:221– 234 Heinemann P (1956a) Champignons récoltés au Congo Belge par Mme M Goossens-Fontana, II Agaricus Fr ss. Bull Jard Bot État 26:1–127 Heinemann P (1956b) Flore Iconographique des Champignons du Congo, 5º fascicule: Agaricus I. Ministère de l’Agriculture-Jardin Botanique de l’État Bruxelles Heinemann P (1956c) Champignons récoltés au Congo Belge par Mme M Goossens-Fontana, II Agaricus, Note complémentaire. Bull Jard Bot État 26:325–333 Heinemann P (1957) Flore Iconographique des Champignons du Congo, 6º fascicule: Agaricus II et Pilosace. Ministère de l’Agriculture-Jardin Botanique de l’État Bruxelles 295 Heinemann P (1961) Agarici Austro-Americani I Agaricus of Trinidad. Kew Bull 15:231–248 Heinemann P (1962a) Agarici Austro-Americani II Agaricus de Bolivie. Bull Jard Bot État 32:1–21 Heinemann P (1962b) Agarici Austro-Americani III Trois Agaricus de la Jamaïque. Bull Jard Bot État 32:23–28 Heinemann P (1962c) Agarici Austro-Americani IV Quatre Agaricus de Venezuela. Bull Jard Bot État 32:155–161 Heinemann P (1971) Quelques Psalliotes du Congo-Brazzaville. Cah Maboké 9:5–10 Heinemann P (1978) Essai d’une clé de determination des genres Agaricus et Micropsalliota. Sydowia 30:6–37 Heinemann P (1980) Les genres Agaricus et Micropsalliota en Malaisie et en Indonésie. Bull Jard Bot Belg 50:3–68 Heinemann P (1982) Quelques Psalliotes de Nouvelle Guinée (Papua New Guinea). Bull Jard Bot Belg 52:405–413 Heinemann P (1984) Agarici Austro-Americani VII. Agariceae des zones tempérées de l’Argentine et du Chili. Bull Jard Bot Belg 60:331–370 Heinemann P (1988) Novitates generis Micropsalliotae (Agaricaceae). Bull Jard Bot Belg 60:540–543 Heinemann P (1990) Agaricus singaporensis sp. nov. Bull Jard Bot Belg 60:417–419 Heinemann P (1993) Agarici Austro-Americani VIII Agaricaceae des régions intertropicales d’Amérique du Sud. Bull Jard Bot Belg 62:355–384 Huelsenbeck JP, Ronquist F (2001) MRBAYES: Bayesian inference of phylogeny. Bioinformatics 17:754–755 Kerrigan RW (2005) Agaricus subrufescens, a cultivated edible and medicinal mushroom, and its synonyms. Mycologia 97(1):12–24 Kerrigan RW (2007) Lectotypification of Agaricus brunnescens. Mycologia 99(6):906–915 Kerrigan RW, Callac P, Challen M, Guinberteau J, Parra LA (2006 “2005”) Agaricus section Xanthodermatei: a phylogenetic reconstruction with commentary on taxa. Mycologia 97(6):1292–1315 Kerrigan RW, Callac P, Parra LA (2008) New and rare taxa in Agaricus section Bivelares (Duploannulati). Mycologia 100 (6):876–892 Kirk PM, Cannon PF, Minter DW, Stalpers JA (2008) Dictionary of the fungi, Tenthth edn. CABI publishing, Wallingford Largeteau ML, Callac P, Navarro-Rodriguez A-M, Savoie J-M (2011) Diversity in the ability of Agaricus bisporus wild isolates to fruit at high temperature (25°C). Fungal Biology, doi:10.1016/j. funbio.2011.08.004 Moncalvo JM, Vilgalys R, Redhead SA, Johnson JE, James TY, Aime MC, Hofstetter V, Verduin SJW, Larsson E, Baroni TJ, Thorn RG, Jacobsson S, Clémencon H, Miller OK Jr (2002) One hundred seventeen clades of Euagarics. Mol Phylogen Evo l23:357–400 Murrill WA (1918) Agaricaceae of tropical North America VIII. Mycologia 10:62–85 Murrill WA (1942) New fungi from Florida. Lloydia 5(2):136–157 Murrill WA (1945) New Florida fungi. Quart J Florida Acad Sci 8 (2):175–198 Murrill WA (1946) New and interesting Florida fungi. Lloydia 9 (4):315–330 Notredame C, Higgins DG, Heringa J (2000) T-Coffee: a novel method for multiple sequence alignments. J Mol Biol 302:205–217 Parra LA (2008) Agaricus L Allopsalliota Nauta & Bas. Pars 1 Edizioni Candusso Alassio, Italy Parra LA, Villarreal M, Esteve-Raventos F (2002) Agaricus endoxanthus una specie tropicale trovata in Spagna. Rivista Micol 45 (3):225–233 Pegler DN (1966) Tropical African Agaricales. Persoonia 4(2):73–124 Pegler DN (1968) Studies on African Agaricales I. Kew Bull 21:499– 533 296 Pegler DN (1969) Studies on African Agaricales II. Kew Bull 23:219– 249 Pegler DN (1977) A preliminary Agaric flora of east Africa. Kew Bull Add Series 6:1–615 Pegler DN (1983) Agaric flora of the Lesser Antilles. Kew Bull Add Series 9:1–667 Pegler DN (1986) Agaric flora of Sri Lanka. Kew Bull Add Series 12:1–519 Pegler DN, Rayner RW (1969) A contribution to the Agaric flora of Kenya. Kew Bull 23:347–412 Peterson KR, Desjardin ED, Hemmes DE (2000) Agaricales of the Hawaiian Islands. 6. Agaricxaceae I. Agariceae: Agaricus and Melanophyllum. Sydowia 52(2):204–257 Posada D (2008) jModelTest: phylogenetic model averaging. Mol Biol Evol 25:1253–1256 Rambaut A, Drummond AJ (2007) Tracer v1.5, Available from http:// tree.bio.ed.ac.uk/software/tracer/ Rick J (1906) Pilze aus Rio Grande do Sul (Brazilien). Broteria 5:5–53 Rick J (1919) Contributio II ad monographiam Agaricinorum Brasiliensium. Broteriaser Bot 17:101–111 Rick J (1920) Contributio ad monographiam Agaricinorum Brasiliensium. Broteria ser Bot 18:48–63 Rick J (1930) Contributio IV ad monographiam Agaricinorum Brasiliensium. Broteria ser Bot 24:97–118 Fungal Diversity (2011) 51:279–296 Rick J (1939) Agarici Riograndenses IV. Lilloa 4(1):75–104 Rick J (1961) Basidiomycetes Eubasidii in Rio Grande do Sul— Brasilia 5 Agaricaceae. Iheringia ser Bot 8:296–450 Ronquist F, Huelsenbeck JP (2003) MRBAYES 3: Bayesian phylogenetic inference under mixed models. Bioinformatics 19:1572– 1574 Saghai-Maroof MA, Solima KM, Jorgenson RA, Allard RW (1984) Ribosomal DNA spacer-length polymorphisms in barley: Mendelian inheritance, chromosomal location, and population dynamics. Proc Natl Acad Sci USA 81:8014–8018 Vellinga EC, Sysouphanthong P, Hyde KD (2011) The family of Agaricaceae: phylogenies and two new white-spored genera. Mycologia 103(3):494–509 Wasser SP (2002) Family Agaricaceae (FR.) Cohn of Israel mycobiota I. Tribe Agariceae Pat., ed Nevo E & Volz PA, A.R.G. Gantner Verlag K.G., Ruggell, Liechtenstein White TJ, Bruns T, Lee S, Taylor JW (1990) Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ (eds) PCR protocols: a guide to methods and applications. Academic, New York, pp 315–322 Zhao R, Desjardin DE, Soytong K, Perry BA, Hyde KD (2010) A monograph of Micropsalliota in Northern Thailand based on morphological and molecular data. Fungal Divers 45:33–79