Mushroom Biology and Mushroom Products. Sánchez et al. (eds). 2002

UAEM. ISBN968-878-105-3

Molecular phylogeny and cultivation

of Agaricus SPECIES

J. Geml 1,2 and D. J. Royse 1

1Department of Plant Pathology, 211 Buckhout Lab

The Pennsylvania State University, University Park, 16802 PA, U.S.A.

2Department of Botany, Szent István University,

Ménesi u. 44., Budapest, 1118, Hungary

>,

ABSTRACT

A phylogenetic analysis was performed on 47 isolates of 34 Agaricus species using sequences from the internal transcribed spacer-2 and the partial large subunit of ribosomal DNA. Our data confirm the monophyly of the genus Agaricus, which is composed of three subgenera: Agaricus, Lanagaricus and Coniogaricus. Within the subgenus Agaricus the following clades were found: 1) Arvenses (A. abruptibulbus, A. albolutescens, A. arvensis, A. augustus, A. blazei, A. diminitivus, A. excellens, A. fissuratus, A. macrocarpus, A. macrosporus, A. nivescens, A. osecanus, A. semotus, A. sylvicola), 2) Campestres (A. californicus, A. campestris, A. cupreo-brunneus), 3) Fuscovelati (A. fuscovelatus, A. lilaceps), 4) Hortenses (A. bisporus, A. bitorquis, A. devoniensis, A. impudicus, A. spissicaulis, A. subfloccosus, A. subperonatus), 5) Subrutilescentes (A. lanipes, A. maskae, A. subrutilescens). All four previously classified sections of the subgenus Agaricus (Agaricus, Arvenses, Sanguinolenti and Xanthodermatei, sensu Heinemann 1978) represent non-monophyletic groups with shared morphological and chemical features that may have evolved independently. These species also were evaluated for cultural characteristics in small-scale pilot trials. While isolates of A. bisporus, A. blazei, A. subfloccosus,and A. fissuratus showed the fastest growth rates on agar media and in compost prepared for A. bisporus, significant intraspecies variation was found in these values. Isolates of A. bisporus, A. bitorquis, and A. arvensis produced the highest yields, while A. sylvicola, A. arvensis,and A. bitorquis showed the largest mean weight of basidiomes.

INTRODUCTION

Agaricus bisporus (Lange) Imbach is the most widely cultivated edible mushroom, accounting for 32% of the more than 6 million metric tons of mushrooms produced worldwide in 1997 (Chang 1999). While A. bisporus has maintained its leading role in the past decade, the mushroom industry is undergoing considerable changes due to the rapid expansion of specialty mushroom (Lentinula, Pleurotus, Flammulina etc.) production. In order to capitalize on the increasing demand for specialty mushroom products worldwide, Agaricus spp. growers are seeking to diversify their product line by offering a greater variety of species and products to consumers.

The number of commercial strains of A. bisporus available to meet specific demands for fresh and processed products is limited, and these cultivars posses narrow genetic diversity (Royse and May 1982, Loftus et al. 1988, Sonnenberg 2000). Researchers (Kerrigan and Ross 1989, Kerrigan 1990, Kerrigan et al. 1993, Rimóczi 1994ab, Rimóczi 2000, Xu et al. 1997, Callac et al. 2000) have sought for unique germplasm of A. bisporus and have described geographical distribution, ecology, lifecycles, genetic structure, and phylogeny of wild populations of this species. Some other species of Agaricus, namely A. arvensis, A. bitorquis, A. macrosporus, A. subfloccosus and A. subrufescens were investigated for potential commercial production and for use as breeding stocks (Elliott 1978, Fritsche 1978, Fermor 1982, Kerrigan 1983, Martinez-Carrera et al. 1995, Noble et al. 1995, Geml

1

Mushroom Biology and Mushroom Products. Sánchez et al. (eds). 2002

UAEM. ISBN968-878-105-3

and Rimóczi 1999, Kerrigan et al. 1999, Calvo-Bado et al. 2000). Another species of Agaricus, A. blazei, has gained popularity among consumers because of its purported medicinal value (Mizuno 1995, Iwade and Mizuno 1997). Other species of Agaricus, with unique culinary or medicinal value, may soon be domesticatable as increasing knowledge of parameters necessary for their fructification becomes available (Chang and Hayes 1978, Chang and Miles 1989, Miles and Chang 1997). Some of these species may be exploitable in breeding programs of A. bisporus, especially considering the recent success in Agrobacterium-mediated transformation of this species (Challen et al. 2000, Chen et al. 2000).

Use of ribosomal DNA (rDNA) sequences to infer phylogenetic relationships among Agaric fungi now is widely exploited (for example, see Bruns et al. 1991, Hillis and Dixon 1991, Hibbett et al. 1995, Lutzoni and Vilgalys 1995, Moncalvo et al. 1995, Nicholson 1995, Bunyard et al. 1996, Hibbett et al. 1997, Hopple and Vilgalys 1999, Pine et al. 1999, Thon and Royse 1999, Moncalvo et al. 2000ab). For phylogenetic analyses of Agaricus species, researchers have used sequence variations found in rDNA (Mitchell and Bresinsky 1999, Calvo-Bado et al. 2000), mitochondrial plasmid pEM (Robison and Horgen 1999) and the mitochondrial atp6 gene (Robison et al. 2001). However, these investigations have used a limited number of species and the phylogenetic relationships of many available Agaricus spp. remain unclear. The work reported herein was designed to gain a better understanding of the evolutionary relationships among available species of the genus Agaricus, and to observe mycelial growth and mushroom development of some species on compost prepared for A. bisporus.

MATERIALS AND METHODS

Sequencing and phylogenetics

Forty-seven isolates of 34 Agaricus species were investigated including 32 isolates of 25 species obtained from culture collections and research groups from France, Hungary and the United States. Sequence data of an additional 15 Agaricus isolates were obtained from Genbank. Isolates were grown in 50 ml potato dextrose yeast extract broth for 3 to 6 weeks depending on the growth rate of the mycelium. The mycelium was filtered from the broth and DNA was extracted using the PUREGENE DNA Isolation Kit (Gentra Systems, Minneapolis, MN). Internal transcribed spacer-2 (ITS-2) and partial large subunit (LSU) regions were PCR amplified using four primers. Amplification products were purified directly from reactions using the Wizard® PCR Prep system (Promega, Madison, WI). Purified amplification products were sequenced using the Applied Biosystems (ABI) BigDye terminator kit and an ABI 377 automated DNA sequencer (Perkin-Elmer, Foster City, CA). Each sample was sequenced in both directions with the same primers that were used for PCR. Sequence ends were trimmed, manually edited and assembled into contigs using the SeqMan II module in the Lasergene package (DNAStar Inc., Madison, WI). Sequences were then aligned using the Clustal W algorithm (Higgins et al. 1991) of MegaAlign 4.03 (DNAStar Inc., Madison, WI) and manual editing. Sequence data of eight non-Agaricus species were downloaded from Genbank and were included in the analysis. These species of reportedly closely related genera (Moncalvo et al. 2000b) were chosen to investigate the phylogenetic origin of the genus Agaricus. As a more distantly related species, Stropharia coronilla was chosen as outgroup. Phylogenetic analysis was performed using PAUP version 4.0b4a (Swofford 2000). A neighbor-joining (NJ) tree was constructed using the Kimura 2-parameter model. The stability of clades was evaluated by bootstrap analysis with 1000 replications (Felsenstein 1985, Hills and Bull 1993).

1

Mushroom Biology and Mushroom Products. Sánchez et al. (eds). 2002

UAEM. ISBN968-878-105-3

Cultivation experiments

The maximum radial growth rates of isolates on agar were determined following the inoculation of 2% malt extract agar, supplemented with 1‰ yeast extract (MYA), with a single 2 mm agar plug taken from the leading edge of a colony. Four replicate plates were prepared per culture and incubated at 25 °C for 3 to 10 weeks. Mycelial growth (maximum radius) was measured at 3 to 10 days intervals (depending on the growth rate). Isolates were examined for growth rate on conventional phase II wheat straw-based compost prepared at the Mushroom Test Demonstration Facility of the Pennsylvania State University according to the short method of composting by Sinden and Hauser (1953). One spawn grain was placed in the center of a sterile 50-ml Petri dish containing 20 g of phase II compost. Four replicate plates were prepared per culture and the inoculated substrate was incubated at 25 °C (2 wk). Mycelial growth (radius; maximum length) was measured at 3 to 7 day intervals (depending on growth rate).Mushrooms were grown in small-scale trials conducted in controlled environmental chambers at the Mushroom Research Center of The Pennsylvania State University. The substrate (2.5 kg) was filled into plastic containers (three replicates per strain), and grain spawn was mixed into the substrate at 1.5% (v/w). Standard cultivation methods for A. bisporus were used for all strains (Wuest and Bengston 1982). Fruit bodies were harvested at a slightly open, “portobello” stage. The lower part of the stipe was trimmed removing the adherent casing material. The number and weight of basidiomes harvested from each container were recorded and average yield per 100 kg wet substrate, average basidiome weight, and average number of days to fruiting were calculated. In all production experiments a commercial hybrid strain of A. bisporus (Korona 2) served as a control.

RESULTS

Phylogeny

Amplification of the ITS-2 and partial LSU yielded fragments of approximately 500 and 550 bp, respectively. The assembled sequences ranged in size from 910 to 1020 bp. An alignment of 928 base pairs, including gaps, was generated for phylogenetic analysis. The neighbor-joining analysis resulted in a single tree (Figure 1). Several clades were found as follows: 1) Arvenses group - bootstrap value of 56%, 2) Campestres group - bootstrap value of 52%, 3) Fuscovelati group – bootstrap value of 69% 4) Hortenses group - bootstrap value of 83%, and 5) Subrutilescentes group - bootstrap value of 88%.

1

Mushroom Biology and Mushroom Products. Sánchez et al. (eds). 2002

UAEM. ISBN968-878-105-3

Figure 1. Phylogram generated by neighbor-joining analysis of sequences from ITS-2 and partial LSU rDNA showing recognized clades of Agaricus spp.

Bootstrap values are based on 1000 replications (only values of main groups are shown).

1

Mushroom Biology and Mushroom Products. Sánchez et al. (eds). 2002

UAEM. ISBN968-878-105-3

Mycelial growth and mushroom production

Isolates of A. blazei, A. bisporus, and A. fissuratus showed the fastest growth on MYA, with average values of 1.29, 1.18, and 0.61 mm/day, respectively (Figure 3). However, the highest growth rate means on compost were of A. bisporus, A. blazei, and A. subfloccosus, 5.09, 4.29, and 3.16 mm/day, respectively (Figure 4). The highest yields were obtained from A. bisporus, A. bitorquis, and A. arvensis; 36, 35, and 23 kg mushroom per 100 kg wet compost, respectively. Isolates of A. sylvicola, A. arvensis, and A. bitorquis produced mushrooms with the greatest average basidiome weight, 98.4, 73.2, and 47.2 g; while the longest time periods before fruiting were recorded for A. sylvicola, A. bitorquis, and A. blazei, i.e. 55, 53, and 48 days, respectively (Figure 2). Although a small number primordia of A. fissuratus were observed on compost, they did not develop into mature mushrooms.


Figure 2. Mean yield, basidiome weight and number of days to fruiting (rounded to nearest whole number).

ARV= A. arvensis, BIS= A. bisporus, BIS-C= A. bisporus Control, BIT= A. bitorquis, BLA= A. blazei, SUF= A. subfloccosus, SYL= A. sylvicola.

1

Mushroom Biology and Mushroom Products. Sánchez et al. (eds). 2002

UAEM. ISBN968-878-105-3

Figures 3-4. Maximum radial growth rate of Agaricus species on malt-yeast extract agar (MYA) and mushroom compost.

ALB= A.albolutescens, ARV= A. arvensis, AUG= A. augustus, BER= A. bernardii, BIS= A. bisporus, BIS-C= A. bisporus Control, BIT= A. bitorquis, BLA= A. blazei, DIM= A. diminitivus, EXC= A. excellens, FIS= A. fissuratus, FUF= A.fusco-fibrillosus, FUV= A. fuscovelatus, LIL= A. lilaceps, MAC= A. macrocarpus, MAS= A. macrosporus, NIV= A. nivescens, SUF= A. subfloccosus, SUR= A. subrutilescens, SYL= A. sylvicola. Meansand standard errors are shown as lines and bars (within diamonds). Vertical end points of diamonds form the 95% confidence interval for the mean. Dotted line shows overall mean for each experiment.

1

Mushroom Biology and Mushroom Products. Sánchez et al. (eds). 2002

UAEM. ISBN968-878-105-3

DISCUSSION

Our data support the theory of monophyletic evolution of the genus Agaricus - bootstrap value of 97% - as previously proposed by others (Heinemann 1978, Cappelli 1984, Kerrigan 1986, Singer 1986, Bohus 1995, Mitchell and Bresinsky 1999). However, four previously classified sections of the subgenus Agaricus (Agaricus, Arvenses, Sanguinolenti and Xanthodermatei, sensu Heinemann 1978) likely represent paraphyletic groups with shared morphological and chemical features that may have evolved independently. Furthermore, our results suggest possible evolutionary groups within the genus.

The largest of these groups is the Arvenses clade, that contained 14 species (A. abruptibulbus, A. albolutescens, A. arvensis, A. augustus, A. blazei, A. diminitivus, A. excellens, A. fissuratus, A. macrocarpus, A. macrosporus, A. nivescens, A. osecanus, A. semotus, A. sylvicola). These species were previously recognized as members of the section Arvenses (Heinemann 1978), sections Arvenses and Minores (Cappelli 1984) and the group Arvenses (Kerrigan 1986). Despite the observed morphological and genetic differences, many recognized taxa of the section Arvenses are able to interbreed (Calvo-Bado et al. 2000). Therefore, the appropriate taxonomic level of this section is yet to be clarified based on the concordance of morphological, biological and phylogenetic species concepts as proposed by (Taylor et al. 2000).

The Campestres group (A. californicus, A. campestris, A. cupreo-brunneus) included morphologically similar species. Although A. californicus often mistaken in the wild for A. campestris, it was placed earlier in the section Xanthodermatei, based on its inedibility, phenolic odor and some other features (Kerrigan 1986).

Another well-supported group found in our work is Hortenses (A. bisporus, A. bitorquis, A. devoniensisA. impudicus, A. spissicaulis, A.subfloccosus and A. subperonatus). Our results, in agreement with Heinemann (1978), Kerrigan (1986) and Robison et al. (2001), but in disagreement with Cappelli (1984), confirms the close relatedness of A. bisporus, A. subfloccosus and A. subperonatus. The placement of A. impudicus could not be made with confidence until additional isolates of this species are investigated.

Although species of the Subrutilescentes clade (A. lanipes, A. maskae and A. subrutilescens) were placed in different sections/groups previously by Heinemann (1978) and Kerrigan (1986), the grouping of these species is well-supported. In addition, several similar morphological and ecological characteristics have been reported for A. lanipes and A. subrutilescens (Cappelli 1984, Kerrigan 1986).

Classification of A. bernardii, A. fusco-fibrillosus, A. hondensis, A. placomyces, A. silvaticus and A. xanthoderma still is not clear. Sequence data from additional Agaricus species of the section Sanguinolenti (A. annae, A. benesii, A. bohusii, A. depauperatus, A. haemorrhoidarius, A. mediofuscus, A. squamulifer etc.) are needed to further elucidate the evolutionary relationships of this diverse group. Also, future studies should include species of other subgenera (Conioagaricus Hein. and Lanagaricus Hein.) in order to further elucidate the evolution of the genus Agaricus.

Knowledge currently is very limited about the growth and production requirements of all but a few species of Agaricus. Our research clearly indicates, there are several species in the genus Agaricus, that can be cultivated with the commercial method developed primarily for A. bisporus. Some isolates of these species showed values of the same level or even superior to A. bisporus in some aspects, i.e., growth on agar media, basidiome weight etc.; and the full potential of these species is

1

Mushroom Biology and Mushroom Products. Sánchez et al. (eds). 2002

UAEM. ISBN968-878-105-3

yet to be determined by adapting production methods to their ecological requirements. Others, showing slow growth or producing no basidiomes in these experiments (A. albolutescens, A. augustus, A. bernardii, A. diminitivus, A. excellens, A. fusco-fibrillosus, A. fuscovelatus, A. lilaceps, A. macrocarpus, A. macrosporus, A. nivescens, and A. subrutilescens), could still possess yet-to-be-discovered valuable features that might be exploited in mushroom breeding.

Because the species of Agaricus we examined in this study are distributed predominantly in the temperate regions, future phylogenetic studies should include tropical species from the subgenus Agaricus, as well as from the subgenera Conioagaricus and Lanagaricus. A better understanding of the phylogenetic relationships and cultivation requirements of species of Agaricus may help the selection and breeding of commercial lines and help to improve commercial production of these mushrooms.

ACKNOWLEDGEMENTS

This work is part of a Ph.D. program of J. Geml, lead by Dr. Imre Rimóczi at the Department of Botany of the Szent István University (Hungary), and was supported by a joint scholarship of the Hungarian Fulbright Commission and the Soros Foundation. Special thanks go to Dr. Philippe Callac (INRA, France), Dr. Richard W. Kerrigan (Sylvan Inc., USA), Dr. Mark G. Loftus (Amycel–Spawn Mate, USA), Vija Wilkinson (PSU Mushroom Culture Collection, USA), the Korona Spawn Plant (Hungary), and the Hungarian Museum of Natural History for providing cultures. Thanks go to Dr. David M. Geiser for assistance with the phylogenetic analysis, and to Patrick Collopy and Dr. Qing Shen for technical assistance and advice.

REFERENCES

Bohus, G. 1995. Agaricus Studies XIII. Hungarian Mycological Society 34:5-37.

Bruns, TD, T.J. White and J.W. Taylor. 1991. Fungal molecular systematics. Ann. Rev. Ecol. Syst. 22:525-564.

Bunyard, BA, M.S. Nicholson and D.J. Royse. 1996. Phylogeny of the genus Agaricus inferred from restriction analysis of enzymatically amplified ribosomal DNA. Fung. Gen. Biol. 20:243-253.

Callac, P, M. Imbernon, J. Guinberteau, L. Pirobe, S. Granit, J.M. Olivier and I. Theochari. 2000. Discovery of a wild Mediterranean population of Agaricus bisporus, and its usefullness for breeding work. Mush. Sci. 15:245-252.

Calvo-Bado, L, R. Noble, M.P. Challen, A. Dobrovin-Pennington and T.J. Elliott TJ. 2000. Sexuality and genetic identity in the Agaricus Section Arvenses. Appl. Environ. Microbiol 66:728-734.

Cappelli, A. 1984. Agaricus L.: Fr., Fungi Europaei, Libreria editrice Biella Giovanna, Saronno, Italia. 560.

Challen, MP, K.E. Gregory, S. Sreenivasaprasad, C.C. Rogers, S.B. Cutler, D.C. Diaper and T.J. Elliott. 2000. Transformation technologies for mushrooms. Mush. Sci. 15:165-172.

Chang, ST. 1999. World production of cultivated edible and medicinal mushrooms in 1997 with emphasis on Lentinus edodes (Berk.) Sing. in China. Int. J. Medicinal Mushrooms 1:291-300.

Chang, S.T. and W.A. Hayes (eds). 1978. The Biology and Cultivation of Edible Mushrooms. Academic Press. 819.

Chang, S.T. and P.G. Miles. 1989. Edible Mushrooms and Their Cultivation. CRC Press Inc. 345 p.

Chen, X, M. Stone, C. Schlagnhaufer and P. Romaine. 2000. A fruiting body tissue method for efficient Agrobacterium-mediated transformation of Agaricus bisporus. Appl. Environ. Microbiol. 66:4510-4513.

Elliott, TJ. 1978. Sexuality in the genus Agaricus. Mush. Sci. 10:41-50.

Felsenstein, J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39:783-791.

Fermor, TR. 1982. Agaricus macrosporus: An edible fungus with commercial potential. Scient. Hortic. 16:273-282.

Fritsche, G. 1978. Tests on breeding with Agaricus arvensis. Mush. Sci. 10:91-102.