S'0-M•^

Genetics and breeding of Agaricus

Proceedings of the First International Seminar on Mushroom Science, Mushroom Experimental Station, Horst, the Netherlands, 14-17 May 1991

LJ.LD. van Griensven (Editor)

a Pudoc Wageningen 199 KVwÜNIVERSITKO

CIP-data Koninklijke Bibliotheek, Den Haag

Genetics

Genetics and breeding of Agaricus :proceeding s of the First International Seminar on Mushroom Science, Mushroom experimental Station, Horst, the Netherlands, 14-17 May 1991 / L.J.L.Dva n Griensven (ed.) - Wageningen : Pudoc. - III. ISBN 90-220-1045-7 bound NUGI 835 Subject heading: mushroom breeding.

ISBN 90-220-1045-7 NUGI 835

©Centr e for Agricultural Publishing and Documentation (Pudoc), Wageningen, Netherlands, 1991.

All rights reserved. Nothing from this publication may be reproduced, stored in acomputerize d system or published inan yfor m or inan ymanner , includingelectronic , mechanical,reprographi c or photographic, without prior written permission from the publisher, Pudoc, P.O. Box 4,670 0 AA Wageningen, Nether­ lands.

The individualcontribution s inthi spublicatio n andan yliabilitie sarisin gfro m them remainth e responsibility of the authors.

Insofar as photocopies from this publication are permitted by the Copyright Act 1912, Article 16Ban d Royal Netherlands Decree of 20 June 1974 (Staatsblad 351) as amended in Royal Netherlands Decree of 23 August 1985 (Staatsblad 471) and by Copyright Act 1912,Articl e 17,th e legally defined copyright feefo r anycopie sshoul db etransferre d to the Stichting Reprorecht (P.O.Bo x 882, I18 0A W Amstelveen, Netherlands). For reproduction of parts of this publication in compilations such asanthologie s or readers (Copyright Act 1912, Article 16), permission must be obtained from the publisher.

Printed in the Netherlands. CONTENTS

Preface vii

Keynote lecture A personal view on mushroom breeding from 1957 - 1991 Gerda Fritsche

Analysis of the genome of Agaricus bisporus A short introduction to the ecology, taxonomy and nomenclature of the genusAgaricus. C. Bas 21

Freeze-drying of fungal hyphae and stability of the product. CS. Tan, C.W. Van Ingen & J.A. Stalpers 25

A genetic linkage map for Agaricus bisporus. R.W. Kerrigan, P.A. Horgen & J.B. Anderson 31

Mitochondrial plasmids and homologous sequences inAgaricus. Mary M. Robinson & Paul A. Horgen 37

Mitochondrial genotypes and their inheritance in the cultivated mushroom Agaricus bisporus. Anton S.M. Sonnenberg, P.C.C. Van Loon & L.J.L.D. Van Griensven 42

Efficient protoplast formation and regeneration and electrophoretic karyotype analysis of Agaricusbisporus. J.C. Royer, W.E. Hintz & P.A. Horgen 52

Chromosome separation and assignment of DNA probes inAgaricus bisporus. Anton S.M. Sonnenberg, K. Den Hollander, A.P.J. Van De Munckhof & L.J.L.D. Van Griensven 57

The use of protoplast production, protoplast regeneration and restriction fragment length polymorphisms in developing a systematic and highly reproducible breeding strategy for Agaricus bisporus. P.A. Horgen, T. Jin & J.B. Anderson 62

Use of the polymerase chain reaction (PCR) inA. bisporusbreedin g programs. R.S. Khush, L. Morgan, E. Becker & M. Wach 73 Progress in the molecular analysis ofAgaricus enzymes. D.A. Wood & CF. Thurston 81

Genetics and Disease Control Fungi in the cultivation ofAgaricus bisporus-a n updated list of species. Albert Eicker & Martmari Van Greuning 89

Nature of disease resistance to compost-borne and airborne pathogens ofAgaricus bisporus. P.J . Wuest 97

Molecular dissection and control of virus disease inAgaricus bisporus. M.C. Harmsen & J.G.H. Wessels 103

Evidence for transmission of La France disease inAgaricus bisporus by ds RNA. Anton M. Sonnenberg & L.J.L.D. Van Griensven 109

Hydrophobin genes in mushroom development. J.G.H. Wessels 114

A DNA sequence inducing mushroom development in Schizophylum. Carlene A. Raper & J. Stephen Horten 120

Strategies for breeding and preparation of spawn Transformation strategies for Agaricus.M.P . Challen, B.G. Rao & T.J. Elliott 129

Towards a transformation system for Agaricus bisporus. John C. Royer & P.A. Horgen 135

The isolation of two tandemly linked glyceraldehyde-3-phosphate dehydrogenase genes from Agaricusbisporus. M.C. Harmsen, J. Scheer, T.A. Schuurs & J.G.H. Wessels 140

Maintenance, rejuvenation and improvement of Horst® Ul. Gerda Fritsche 145

The development of a set of characteristics for D.U.S. tests of cultivated mushroom varieties. A. Van Der Neut 153

List of authors 161 PREFACE

The First International Seminar on Mushroom Science was held in The Netherlands from May 14 to May 17, 1991 at the occasion of the retirement of Dr. Gerda Fritsche. A unique feature of the seminar was that it brought together all scientists working in the field of molecular genetics of Agaricus. The international gathering and free and mutual exchange of ideas has led to an intensified international cooperation. This will bring mushroom genetics where it belongs: at the forefront of science.

This book has become a comprehensive condensation of Agaricus genetics. It contains the scientific knowledge of the genome and has practical use in breeding and spawn preparation.

The invaluable help of Ms. Ineke Dohmen in the preparation of the manuscript is gratefully acknowledged.

L.J.L.D. Van Griensven

vu KEYNOTE LECTURE

A PERSONAL VIEW ON MUSHROOM BREEDING FROM 1957-1991

Gerda Fritsche

Professor van Griensven has invited me to tell you about my views on 34 years of personal involvement in the field of mushroom breeding. I was pleased to accept, since it has been a wonderful time during which many things have changed. Encouraged by Van Griensven, I should also like to tell you something about the period preceeding my mushroom career.

From 1929 -1955

I was born in 1929 in Merseburg, a medium-sized town in central Germany near Leipzig. When the Nazi's came to power, I was three years old. At ten years old, I saw the outbreak of the Second World War. After the War, my home town was allocated to the Russian zone of occupation, later East-Germany. In the meantime I was 16 years old and went to a highschool for girls. Relatives of my father, he himself originating from the countryside, found refuge in my parental home. They were evicted from hearth and home by the new authorities, because they owned more than 250 acres of land. I went to school until 1948. It was never a problem for me, what I was going to be one day. I wanted to be engaged in nature in the widest sense. My mother, watching me nursing our pot plants with great care, suggested botany as a field of study. A vocational bureau informed us, that I could become a teacher after studying in botany. Yet, I wanted to go into research. A study in horticulture was recommended. This study was preceeded by a horticultural training. I could practise this in a nursery, owned by relatives of a schoolfriend and charmingly situated in the beautiful south Harz countryside. After this training was finished, I applied in 1950 for a course of study in horticulture at the Humbold University in East-Berlin. I was refused without reason. Probably this was because my father was a physician and at that time children of university graduates were not allowed to study. Another reason was perhaps that I was not a member of the communist youth organization 'Free German Youth'. At the time, the refusal was not catastrophic for me, since the borders to the free West were still open. You only had to take the underground in East-Berlin and to travel to West-Berlin. But in West-Berlin, courses in horticulture started only in 1951. From 1950 till 1952, I was a horticultural assistant in a seed breeding nursery in the 'Flowertown' Erfurt. Plant breeding already fascinated me at that time. This period in practical plant breeding was very beneficial. In November 1952, I started my horticultural study at the Technical University in West-Berlin, together with some 20 fellow-students, most of them also originating from the East. We received a scholarship, one warm meal daily distributed by the Americans, and the opportunity to earn money from the 'TUSMA'. TUSMA stands in German for the sentence: 'Call And Students Do Everything'. Common problems united all the students together. Wonderful friendships were made. After a few years break, we have met every two years for a professional excursion and to talk about old memories. From 1956 - 1958

In 1956 I finished both my study in Horticulture at the Technical University of West-Berlin and training as a technical assistant at the Institute for Genetics and Breeding at the same University. The training was supposed to be a jumping-board to an appointment. My fellow-student Gertraud Lemke had the same training. As Professor Von Sengbusch had promised me that I could start in January 1957 to breed mushrooms, Professor Kappert, Director of the Berlin Institute, organized another short training course for me in a microbiological laboratory. Later, I have realized how important this additional training was and I am still grateful to Kappert for his support. In Berlin I was also helped by Professor Riethus, our teacher in vegetable growing. Guided by him, my fellow-student Hans-Joachim Tschierpe graduated on the role of C02 in mushroom growing (Tschierpe, 1959). Riethus organized a visit for me to a mushroom farm. Of course, the owner of the farm refused to show his spawn laboratory. This should remain a secret.

Professor Dr Reinhold Von Sengbush, director of the Max Planck Institute for Plant Breeding in Hamburg

When I started in January 1957 with my work, Von Sengbusch was already a famous plant breeder, from whom I could learn a lot. In 1928 he had developed sweet lupin (Von Sengbusch, 1931), which made it possible to use lupin as fodder. Besides, he also was the breeder of the famous strawberry Senga Sengana. Professor Von Sengbusch was director of the Max Planck Institute for Plant Breeding in Hamburg, where in 1957 hemp, spinach, tobacco and other plants were studied.

He wanted to start with mushrooms, because more than one generation could be produced each year. Von Sengbusch hoped that in the ten years before he had to retire he could be successful with mushrooms. Already in 1956 he had started cropping experiments in the cellar of a neighbouring farm in collaboration with Walter Huhnke. For mushroom breeding I needed an autoclave. There was already one there, because it was required for the hemp breeding programme. In the autoclave the fibres were opened up with sodium hydroxide. When I had to sterilize, I removed the sodium hydroxide from the autoclave and filled it with distilled water. I needed a clean room as well. I was given a room adjacent to a large room containing hemp stems. It could only be entered through a sluice, namely the lumber-room. The firm Hullen in Erlangen (in the south of Germany) supplied us with mushroom spawn in milk bottles. So we used milk bottles for the preparation of spawn as well. The first bottles were inoculated by Von Sengbusch and myself. They were all microbially infected, because we could not work germ-free. But soon I obtained an inoculation room constructed of metal walls and glass plates. To keep the room germ-free we used UV-lamps and Aerosept evaporizers. The latter evaporate a germicidal oil (Lemke, 1967). In the autumn of 1957 Huhnke and I visited a few English mushroom farms. The English mushroom growers gave us a hearty wellcome and showed us their farms. It was not possible, however, to visit a spawn laboratory. In 1958 an experimental farm with 6 growing rooms was built on the site of the Max Planck Institute together with a laboratory and offices. Von Sengbusch instructed me to collect a great number of monosporous cultures in order to find that excellent one that only seldom happens to appear. Fortunately, number 59 was already a very particular monospore culture. The primordia were deformed. The few fruit bodies which appeared, looked like puff-balls rather than cultivated mushrooms (type 59a). Later, its form slightly changed (type 59b). One day a clump-like fruit body of 350 grammes developed in a bed with our second type 59b. We multiplied it by tissue culture and called it type 59c (Fritsche & Von Sengbusch, 1963).

ill lll Ulf wÊHÊÈmm MpT Slfip *ips" *|jp||j||||i^r

II H • • »

Fruit bodies of type 59b (left), 59c (right) and intermediary types (middle). The team working on mushrooms grew rapidly. In December 1957 Gerlind Eger joined the group. Her task was to investigate the origin of fructification. Egers' work became well-known worldwide. I want to refer here to two of her numerous papers only, viz. one about the role of micro-organisms during pinhead formation (Eger, 1961) and one about the 'Halbschalentest' she developed (Eger, 1962). In 1958 Christine Breitenfeld joined the team of scientists. She had to develop methods to facilitate breeding. For instance, she looked after methods to increase the rate of spore germination. She left the Institute already in 1960, after she became married. From the publications available in 1957, the most comprehensive one was a thesis written by Kligman in the USA in 1943.Th e thesis was entitled: 'Some cultural and genetical problems in the cultivation of the mushroom, Agaricus campestris Fr.'. When I read the title, I thought that the mushroom involved was not the same one as we were interested in. A closer look revealed that this was not the case. The author proposed namely A. campestris bisporus as the name for the cultivated mushroom 'in order to recognize taxonomically its physiological and morphological distinctness from the wild four-spored A. campestris. The Latin name of the cultivated mushroom changed several times, even the generic name. So in 1957 Psalliota was used next to Agaricus. Later on, Agaricus bisporus became the most common name. But in the early eighties the name changed again. Especially in the USA the name Agaricus brunnescens Peck is used. However, in Europe the name Agaricus bisporusi s maintained, based on arguments put forward by Elliott in 1983. The comprehensive work of Kligman involved cytological studies, monosporous cultures, multisporous cultures, tissue cultures, mating sterile strains and mutations. Ferguson had discovered as early as 1902 that growing mycelium stimulates mushroom spore germination. In 1929 Lambert discovered monosporous cultures to be mostly fertile. Sinden developed grain spawn in 1932. When we started mushroom breeding in 1957, already three International Mushroom Congresses had taken place, viz. in England in 1950, in Belgium in 1953 and in France in 1956. The proceedings were published as Mushroom Science I, II and III. During the first Congress there was not a single lecture on genetics and breeding. Lambert (1950) only wrote some lines on this subject in his review about the developments in the preceding twenty years. The second Congress included two papers about breeding. Stoller and Stauffer (1953) classified their cultures obtained from spores according to morphological characteristics of the mycelium. Sigel and Sinden (1953) reported on experiences with a monosporous culture, often used on the farm where the authors were employed. Because the yield of that strain gradually declined, they multiplied it by multisporous cultures, which, to their surprise, were very variable. In Mushroom Science III Borzini and Scurti (1956) reported on their method to make monosporous isolates. Moreover, Heltay (1956) introduced the Hungarian research laboratory, founded in 1952. Until 1955 only compost spawn was used there. As starting material for breeding they especially used wild strains of A. bisporus.

From 1959 -1964

In 1959 I was allowed to attend the fourth International Mushroom Congress held in Danmark. For me, the most interesting of the seven papers on genetics and breeding was the one from Lambert (1959). The author reviewed the state of research in genetics, breeding and spawn production at that time. He emphasized the importance of analyzing the genetic basis of the cultivated mushroom. Uzonyi (1959) had obtained in Hungary good results with collected wild strains. She recommended to insert wild material in the breeding programmme, advice that is topical again right now. I would also like to mention the work of Riber Rasmussen, Amsen and Holmgaard (1959), who had studied the phenomenon 'open veil'. The authors could not find any relationship between 'open veil' and crop management or infections. In the progeny of normal fruitbodies from beds with open veil, however, mushrooms with open veil occurred again. An important cytological contribution in the late fifties was achieved by Evans (1959). He studied nuclear behaviour of the cultivated mushroom and published the results in Chromosoma. Evans assessed a haploid chromosome number of 12. Kligman (1943) had only found 9. In the Max Planck Institute Christine von Gayl, born Breitenfeld, was succeeded by Otto Till. Von Sengbusch instructed him to develop a substrate independent of horse manure composting. This was because horse manure compost had been found an unreliable substrate. The variation in the mushroom yields was considered unacceptable. Till (1961) mixed wheat straw with protein containing materials, such as cotton seed meal. This mixture was moistened and sterilized. Spawning and mycelial growth took place under sterile conditions. After the substrate was full- grown and cased, fruitbodies developed. Till had only a short time available for his work. He got cancer and died in 1962, 39 years old. The substrate was called after him Till-substrate'. In 1961 Gertraud Lemke joined the team in Hamburg. She was to concentrate on the problems of spawn production. For that purpose a new spawn laboratory was built in 1962. After Till's death, Huhnke and Lemke continued studies with the substrate he had developed. In 1961 the first article of Von Sengbusch and myself was published on the comparative performance of two strains (Fritsche & Von Sengbusch, 1961). It was followed by a publication on problems and preliminary results of our breeding work with the mushroom (Fritsche & Von Sengbusch, 1962). In the meantime we had gained experience with 4500 monosporous cultures. In 1962 I attended the fifth International Mushroom Congress in the USA. Here, Moessner (1962) presented a paper entitled: 'Preliminary studies of the possibility of obtaining improved cultures through mycelial fusion'. Moessner had placed the mycelium of two strains close to each other on the same agar plate, so that it could grow together. From the point of juncture he took mycelium and prepared spawn to test it in a growing room. Mycelium from each strain taken well back from the line of contact, was also tested in the growing room. This was done with eight different pairings. The fruitbodies of the strains differed in colour and shape. In a few cases Moessner got fruitbodies from mycelium that he took from the line of contact, with properties of both parental strains. He concluded that there may have been some genetical mixing through anastomoses in these cases. I was particularly interested in Moessners' work, because I too was engaged in questions concerning the possibility of crossings in A. bisporus.I had worked with a normal brown strain as well as with a puffball-shaped white one, that also produced deformed primordia. Spawn of both strains was mixed at spawning. Fruitbodies of both strains often appeared close together on a bed. In the offspring of brown fruitbodies the new combination normal white occurred. Also deformed primordia were found (Fritsche 1964, thesis). Moessners' as well as my own results demonstrated that new combinations of the cultivated mushroom are possible. Moessner and I had worked with heterokaryons. In order to investigate the genetical basis it was necessary to work with monosporous cultures, originating from the same basidium. During the American Congress, Lambert and Ayers (1962) reported on a technique that should facilitate this. It was based on the principle, that the spores of a basidium are released at the same time and remain together even during their fall. A paper, very important to mushroom breeders, was presented by Bels (1962). Bels proposed to establish an 'International Centre for the Investigation of Mushroom Strains', to bring an end to the uncertainty about mushroom strains. There was ample discussion, but up till now such a Centre has not been realised. Finally I would like to mention Stollers' paper (1962) concerning all aspects of spawn preparation. Stollers' advices became the basis for our own spawn production.

From 1965 - 1970

In 1965 Eger left our team and moved to Esser at the Ruhr University in Bochum. That same year the sixth International Mushroom Congress including a Symposium was held in The Netherlands. There, Pelham (1965) reported on his success to repeatedly isolate and germinate the spores from a single basidium. Pelham had developed an interesting method in which he used, among others, strips of cellophane to reduce the risk of contamination. I appreciated Pelhams' results particularly because the laminar flow system, that simplifies germ-free work, was not yet developed. Kneebone (1965) summarized the investigations of Penn State University, where an extensive collection of strains was maintained by periodic transfer to fresh media. In the Max Planck Institute we paid much attention to strain 59c that has already been mentioned. The fruitbodies could become economically valuable, because they had an excellent taste and they could be used sliced as vegetarian cutlets. In beds spawned with 59c, fruitbodies of type 59b appeared again with intermediary forms between 59b and 59c. We then multiplied these as tissue culture and obtained the various types back again. I explained this as a difference in ratio between various types of nuclei in the mycelium. During the sixth International Congress I reported on these studies. At that time we paid also much attention to maintenance of strains. The experiments involved three methods: 'Propagation by mycelium transfer, by tissue cultures and by multispore cultures'. The results were published in 1966 and 1967 (Fritsche, 1966a, 1966b, 1967). The tissue culture method seemed inferior to the other two, which can both be used, provided the material is regularly controlled. In 1968 the seventh International Congress was organized in our own city of Hamburg. Its organisation was mainly carried out by Von Sengbusch and Lemke. Our studies concerning the effect of media on mycelial morphology in the cultivated mushroom were ready to report (Fritsche, 1968). We had worked with strandy and fluffy mycelium. It was shown that compost-agar prevented degeneration of mycelium into the fluffy type better than did Biomalz-agar. Lemke (1968) reported on her observations on cold storage of wheat grain spawn. While both brown strains showed diminishing mycelial activity after 2 months of storage, this only happened after 4-6 months with the two white strains. Lemke supposed this difference to be due to different reactions against metabolic alcohol. An, in my opinion, very important article regarding the maintenance of strains was published by San Antonio and Hwang in 1970. The authors stored strains of A. bisporus as grain spawn up to one year under liquid nitrogen at -196°C. Several strains were checked after 2, 70 and 180 days of storage. The scientists found no significant difference in fruitbody yield and quality between spawn stored under liquid nitrogen and the control. Therefore, they concluded that this method might be very valuable for research and spawn production. In February 1968 Von Sengbusch reached the age of 70. What a performance to have organised a congress at that age! However, at this age he had to retire. The Max Planck Institute in Hamburg was founded for Von Sengbusch to enable him to practise science and breeding according to his own ideas. Obviously, the Institute had to close down when he retired. The Max Planck Society and the Government had already discussed years before, what should happen with the Institute. A solution was found by assigning the buildings to the State, which then established a Federal Research Station for Horticultural Plant Breeding. Also the fellow-workers could be employed there. True enough, but for financial reasons this did not hold for the mushroom department. The Max Planck Society continued to fund the breeding department from 1969 to 1971 expecting to find some other funds later. For the nearly 20 employees this was a time between hope and fear. Nevertheless, it was also a time during which Von Sengbusch turned out to be like a father, deeply concerned about his childrens' future. In 1970 a special issue of the Journal of the German Mushroom Growers was devoted to the future of the mushroom department. It was entitled: 'Mushroom research, yes or no?' German mushroom growers expressed their concern and Von Sengbusch (1970) reported on all work performed and the plans for the future. The introduction showed Von Sengbusch' foresight. He forecasted an avalanching increase in mushroom production and emphasized the necessity of strain protection. As I have already stated, Von Sengbusch paid a lot of attention to the creation of a reliable growing medium, which led to the development of the Till substrate. Later Huhnke slightly changed the technique by submitting the substrate to a fermentation process after sterilization, so that mycelial growth no longer had to take place under sterile conditions (Huhnke & Von Sengbusch, 1968). However, the new technique needed closer investigation before it might be introduced into practice. At the end of 1970, the decision was made to close the mushroom department, but nobody became unemployed. All personal found jobs at the Federal Research Station or somewhere else. Huhnke was the senior of the three scientists. The Max Planck Society continued to pay his salary until his retirement. Initially, he still worked with mushrooms. These activities faded away and were replaced by studies on meristem cultures. Lemke's spawn laboratory was equipped for this technique, which was new at that time. Lemke accepted an offer from Blanchaud in France, where she found new employment in the spawn and research section. Before she left, she wrote an article about practical experiences gathered with spawn production in Hamburg and dedicated this to Von Sengbusch (Lemke, 1972). I myself gladly accepted the opportunity to continue my original work in Horst. The Max Planck Society allowed me to take the most important breeding material with me.

From 1971 - 1980

I started my work at the Experimental Station in Horst on July 1, 1971, the first few months in the old buildings, but from October onwards in the new buildings in America, a village belonging to the municipality of Horst. Soon I noticed that the 'reliable substrate', for which Von Sengbusch exerted himself so intensively, existed here. The difference was that in Hamburg the compost was home-made in small heaps, whereas the Experimental Station in Horst purchased the substrate for the strain trials from the Cooperative in Ottersum. Here, several very large compost heaps are well mixed before delivery. Moreover, a large customer can make greater demands on horse manure suppliers than a small one. In my opinion, the Experimental Station in Horst offers very good conditions for successful breeding, because not only trays of 0.2 m2 growing surface are available, but also bed sections of 1.3 m2 and more. At that time ir. Van de Pol was Director of the Experimental Station in Horst. He was just engaged in setting up strain protection for mushrooms. Since the development of a new strain costs a lot of time and money, strain protection is an important prerequisite for breeding work. The money invested in breeding new strains ought to be earned back. Otherwise, soon, breeding work will not be done any more.

Ir. P.H. Van de Pol, director of the Mushroom Experimental Station from 1968 - 1981

In Horst, strain protection was first of all focused on the city mushroom (A. bitorquis). Studies of Dieleman-Van Zaayen (1972) had shown that this species is virus resistant. At that time Dutch mushroom growing was seriously affected by virus. Trials to cultivate A. bitorquis were performed at the same time and independent from each other by Hasselbach and Mutsers (1971) in Horst and Poppe (1972) in Belgium. In addition to its virus resistance, the insensitivity of the

10 fruitbodies to handling and their long shelf-life seemed to make cultivation of this species rewarding. The fact that A. bitorquisneed s a temperature approximately 5°C higher than A. bisporusi s only an advantage during warm summers and in southern countries. Disadvantages were its greater sensitivity for cultivation errors, deep primordia formation and a short stipe. The fruitbodies easily became dirty because the casing soil attached itself to the sticky skin of the cap. It was very exciting to do breeding work with A. bitorquis,a s it was possible to start with naturally occurring wild strains. Also, since A. bitorquisi s four-spored, it is easy to cross. Hasselbach and Mutsers already had developed an A. bitorquis strain, but this was not a product of cross-breeding. It received strain protection in 1972 under the name Horst® B 30 (Van de Pol, 1972). In 1974 a contract was concluded with Darlington (England). This company produced spawn of Horst® B 30 from then on, and sold it on a licence basis (Van de Pol, 1974).

Breeding work with A. bitorquis was fascinating. Homokaryons could be identified by slow-growing mycelium and young heterokaryons from successful crossings by fast-growing, mophologically changed mycelium (Fritsche, 1976). In 1975 Raper published results of her studies on the sexuality and life-cycle of A. bitorquis. She observed A. bitorquis to be heterothallic and bipolar, having a sexuality controlled by a single incompatibility locus with multiple alleles. Our work in Horst resulted in the development of the strains Horst® K 26 (Horbita) and K 32 (White Bitora), which came on the market in 1975 and were sold by Darlington (Van de Pol, 1975). Both strains are crossings between different wild origins and are distinguished from each other by mycelial growth, yield and properties of the fruitbodies. In the ensuing years, the production ofA. bitorquis (of the strains put on the market by various spawn companies) was mainly limited to virus-infected farms. These recovered after one cropping cycle. Commercially, however, A. bitorquiscoul d not compete withA. bisporus. From the A. bisporusstrain s originating from Hamburg a fast-pinning white one (Y 217) was suited for Dutch conditions. Spawn from Y 217 was put on the market in 1976 by Les Miz Holland B.V. (Van de Pol, 1976). Strain 59c, likewise originating from Hamburg, obtained strain protection under the name 'Aromata' (Van de Pol, 1972). However, it was not commercialized because our breeding programme was not successful in clearing up the disadvantages of this strain, such as regression to its original low-yielding form and the large amount of waste due to the formation of small or irregular fruitbodies. In 1971, a few months after my appointment in Horst the Eighth International Mushroom Congress was held in England. This Congress was a real breakthrough regarding A. bisporusgenetics . Three contributions dealt with this topic. The Rapers had worked with auxotrophic mutants (Raper & Raper, 1971). They confirmed a sexual process including recombination of inherited characters. Their conclusion was, that interstrain breeding is difficult but feasible. Elliott (1971) employed the technique of his predecessor Pelham (1965), already mentioned in this paper, to isolate and germinate the spores belonging to a single basidium. He succeeded with 'diads' as well as with 'triads' and in one case even with all four spores of a basidium. Elliott assessed A. bisporus to be secondarily homothallic. Miller and Kananen (1971) used monosporous cultures for their crossing experiments, which were isolated from four-spored basidia. To avoid pollution with

11 alien mycelium or spores, they carried out their fertility tests in glass jars covered with paper toweling, on a slightly changed Till substrate. They observed a perfect bipolar pattern of sexuality. During the Congress another nine papers concerning breeding or spawn were read, but in view of the time, I will not dwell on these. The ninth Congress took place in 1974 both in Japan and Taiwan. The alteration of 'International Congress on Mushroom Science' into 'International Scientific Congress on the Cultivation of Edible Fungi' was introduced deliberately, because in the host country Japan very little Agaricus bisporus is grown, while Lentinus edodes (Shiitake) was and will be the edible mushroom of the Japanese. Next to Shiitake, many papers dealt with Pleurotusspp . and Volvariella volvacea.O f eight lectures on breeding as well as on spore germination and spawn related to A. bisporus and A. bitorquis, I will only mention one. Miller, Robbins and Kananen (1974) investigated, how form and colour of the fruitbodies were inherited in the material of A. bisporus they had at their disposal. In the Fl they obtained fruitbodies uniform in shape and in the F2 a 3:1 segregation, so clear dominance. On the other hand, the colour showed a gradation of all colours, so that the authors presumed multiple gene action. In the mid-seventies the development of harvesting machines had made such progress, that many farms harvested mechanically. Only the so-called off-whites were appropriate, since only these strains produced sufficient high yields in the first two flushes. The canning industry complained about the inferior quality of the fruitbodies of these strains. The Experimental Station in Horst was asked to introduce the attractive properties of white strains, such as better colour of the flesh, slower discolouration of the lamellae and less shrinkage after canning, in the off-white strains. We decided to make crossings between existing commercial strains and to start selection procedures in the progeny. This was the first initiative to use cross­ breeding as a tool in the practical breeding of the cultivated mushroom. The methods we used in our work with A. bisporuswer e the same as we practised already in A. bitorquis. Also homokaryons of A. bisporus have a characteristically slow-growing mycelium, as had already been assessed by Kligman in 1943. It is true, that the heterokaryon of a successful pairing has no chacteristic mycelium, so that a fertility test in a growing room is necessary. Following the rule of Von Sengbusch, that only large numbers promise to be successful, we made many crossings by inoculating homokaryons of two varieties side by side on an agar plate (Fritsche, 1980). From each of 17 fertile pairings, about 40 monosporous cultures were isolated and tested. A description of our selection procedure has been published in the Mushroom Journal (Fritsche, 1983). Only a few strains passed all steps of selection successfully. Only six years after the beginning of our work the two strains Horst* Ul (Horronda) and Horst® U3 (Horwitu) were at the growers' disposal. In 1978 the tenth International Congress with Symposium on edible fungi was held in France. Of the 83 contributions at the symposium, 16 were concerned with genetics and breeding. Only nine dealt exclusively or partly with Agaricus. Two papers I would like to introduce shortly. Boissonet-Menes (1978) reported on protoplasts and the possibility to use them to overcome crossing-barriers. Elliott (1978) set out his breeding strategies in Agaricus bisporus. They consist of using homokaryons with different fungicide resistance instead of auxotrophic mutants for quick identification of heterokaryons in the laboratory.

12 Fruitbodies of Horst UI

From 1981 - 1991

In 1981 Van de Pol retired. His successor, Van Griensven started full of enthusiasm and new ideas. As far as breeding was concerned it was time to put spawn of Horst® Ul and U3 on the market, so Van Griensven negotiated with representatives of Darlington to conclude a contract. Darlington carefully introduced the two strains into commercial practice from November 1981 onwards, to obtain growing experiences. In the Experimental Station in Horst, Gerrits studied substrate requirements of both U-numbers, while Visscher performed casing soil trials. The four extension officers supported the farms growing Ul and U3. At first, U3 was grown more than Ul, but when some mushroom growers succeeded to grow the more difficult Ul with good results, many farms switched to the qualitatively better Ul. Also in 1981 the eleventh International Congress on Edible Fungi was held in Australia. I would like to mention the paper of May and Royse (1981), since new ways of genetic research were pointed out. The authors used electrophoretic methods for genetical studies in A. bispoms, A. campestrisan d Lentinus edodes. Taking into account the enormous development in the field of gene techniques, Van Griensven contacted Wessels from Groningen State University. A four year cooperation agreement between Groningen and Horst was set up, financed by third party funding, under the heading 'Application of molecular biological techniques to improve mushroom growing. Research in the field of breeding edible mushrooms'. In connection with this joint research, Sonnenberg came to Horst in September 1984. He graduated in Groningen, where he gained experience with protoplasting in Schizophyllum commune. It was a wise decission of Van Griensven to get to Horst a young scientist, familiar with modern molecular biological techniques. Sonnenberg

13 HorstU 3 tussen-1 ras

Fruitbodies of Horst U3 and its ancestors started his work enthusiastically and assiduously. As you know, he is to become my successor. The section 'strains and breeding' has grown steadily during recent years. While in 1983 only three people worked here, now it is about twice that number. In 1988 a modern laboratory was built, especially equipped for fundamental research. In my opinion, a new period in mushroom breeding has started. I watch all this with great interest. The new techniques open ways, to make a first selection already in the laboratory. Since such an early selection saves a breeder much time and material, we had already looked earlier for such possibilities in connection with conventional breeding, yet without success (Fritsche, 1970). The period between 1981 and 1991 also included some of my own work on the maintenance of strains. For this work I refer to my second paper during this Seminar. I also want to touch shortly upon the fact, that a small part of our time was dedicated to breeding work with mushrooms, other than A. bisporus and A. bitorquis. Work on A. arvensis was, after 13 years, finished by supplying Somycel with mycelium of strain Horst® R20 (Fritsche, 1989). Also breeding work on Lepista nuda and Coprinus comatus was recently concluded. Two strains are available of each species. During the twelfth International Congress on Edible Fungi, which took place in Germany in 1987, Van Loon and I reported on these studies (Van Loon & Fritsche, 1987; Fritsche & Van Loon, 1987). Comparison of the fourth International Congress on Mushrooms in 1959, which was the first I attended, with the twelfth Congress in 1987 reflects the enormous development in mushroom research. There were 66 papers in 1959, but 171 in 1987. Mushroom Science IV comprises 572 pages, Mushroom Science XII, on the contrary, 1786 pages, split up in two volumes. While in 1959 only 7 papers concerned the subject of breeding in the widest sense, there were already 22 lectures in 1987, including the key-note address during the opening of the congress.

14 Acknowledgements

At the end of my paper, I want to express my thanks. I'm pleased, to have participated in 34 years of booming mushroom growing and research. In my work I was not alone. I experienced the support of my teachers and superiors and I changed from a good team in Hamburg, to a similar team in Horst. For me, the nice relationships with people from commercial mushroom growing in The Netherlands and in Germany, as well as with employees from the Darlington and Somycel Spawn Companies, were very valuable. The small mushroom opened for me the door to the whole world. I experienced not only many nice and interesting things, but also became acquainted with many kind people, who enriched my life. I would like to mention four people by name. First of all Professor Dr L.J.L.D. Van Griensven, who stimulated my work enormously and to whom we owe this Seminar. It is an honour and a pleasure for me, that he has organized it on the occasion of my retirement. Mrs José Kuenen-Claes assisted me for almost 14 years. I know, I could always rely on her. With great care, she performed her task. She has also made valuable proposals for improvements. The same applies to Mr Peter Van Loon, who assisted me during eight years, till he got his own project in October 1990 (strain research). He is also a talented handy-man. My special thanks are also addressed to Drs J. Gerrits, not only because he translated this lecture from German into English. He was also the one to whom I went with questions and problems. He had always time for me. His advice was always a great help to me. I am also indepted to Mr A.R. Claxton (Darmycel) for making final corrections in the English text. Finally, I should like to wish everybody, who is engaged in research and the growing of edible mushrooms, a successful and happy future.

Summary

The author describes the developments, she witnessed during 34 years she was active in mushroom breeding. She started her career in January 1957 in the Max Planck Institute for Breeding of Cultivated Plants in Hamburg (Germany). The Institute was conducted by Professor Von Sengbusch, at the time already a famous plant breeder. The topic 'Cultivated Mushroom', on which hardly any breeding work was done so far, was jointly started. From small beginnings, a section was developed concerning breeding, and studies on a substrate (reliable for strain comparisons) and spawn. As the Institute closed down after Von Sengbusch retired, the author continued her work on mushroom breeding at the Mushroom Experimental Station in Horst (The Netherlands). First, strains of the virus-resistant species Agaricus bitorquis were developed. Later, work concentrated on A. bisporus.I n 1981, the two hybrid strains Horst® Ul and U3 were put on the market. These strains combined the good properties of white and off-white strains. On a small scale, attention was also paid toA. arvensis, Lepista nuda and Coprinus comatus. Based on numerous literature quotations, the author reviews the general development of mushroom breeding, of which the genetical basis was only elucidated in the early seventies. The striking developments in genetical research

15 have also penetrated mushroom work and open new prospects in mushroom breeding.

References

Bels, P.J., 1962. An international centre for the investigation of mushroom strains. Mushroom Science V, 231-240. Boissonett-Mens, Marcella, 1978. Les protoplastes de champignons. Mushroom Science X (part 1), 27-30. Borzini G. et J. Ceruti Scurti, 1956. Méthodes pour obtenir des cultures monospores de champignons de couche. Mushroom Science III, 138-146 Dieleman-Van Zaayen, Annemarie (1972): Spread, prevention and control of mushroom virus disease. Mushroom Science VIII, 131-154. Eger, Gerlind, 1961. Untersuchungen über die Funktion der Deckschicht bei der Fruchtkörperbildung des Kulturchampignons, Psalliota bispora Lge. Archiv für Mikrobiologie 39, 313-334. Eger, Gerlind, 1962. Der Halbschalentest in Wissenschaft und Praxis. Die deutsche Gartenbauwirtschaft 10, 15-17. Elliott, T.J., 1971.Se x and the single spore. Mushroom Science VIII, 11-18. Elliott, T.J., 1978. Breeding strategies in Agaricus bisporus. Mushroom Science X (part 1), 73-81. Elliott, T.J., 1983. The cultivated mushroom: is it Agaricus brunnescens? The Mushroom Journal 122, 69. Evans, H.J., 1959. Nuclear behaviour in the cultivated mushroom. Chromosoma (Berl.) 10, 115-135. Ferguson, Margaret C, 1902. A preliminary study of the germination of the spores of Agaricus campestrisan d other basidiomycetous fungi. Washington, Bul.U.S. Bur. Plant Ind. No.1 6 Fritsche, Gerda & R. Von Sengbusch, 1961. Leistungsvergleich zweier Champignonsorten. Der Züchter 31,233-238 . Fritsche, Gerda & R. Von Sengbusch, 1962. Die züchterische Bearbeitung des Kulturchampignons {Psalliota bispora Lge.). Probleme und erste eigene Ergebnisse. Der Züchter 32, 189-199. Fritsche, Gerda & R. Von Sengbusch, 1963. Beispiel der spontanen Entwicklung neuer Fruchtkörperformen beim Kulturchampignon. Der Züchter 33, 270- 274. Fritsche, Gerda, 1964. Versuche zur Frage der Merkmalsübertragung beim Kulturchampignon Agaricus (Psalliota) bisporus (Lge.) Sing. Der Züchter 34, 76-93. Fritsche, Gerda, 1965. Beitrag zur Mutationsforschung. Mushroom Science VI, 27- 47. Fritsche, Gerda, 1966a. Versuche zur Frage der Erhaltungszüchtung beim Kulturchampignon. I. Vermehrung durch Teilung des Mycels. Der Züchter 36, 66-79. Fritsche, Gerda, 1966b. Versuche zur Frage der Erhaltungszüchtung beim Kulturchampignon. II. Vermehrung durch Gewebekulturen. Der Züchter 36, 224-233. Fritsche, Gerda, 1967. Versuche zur Frage der Erhaltungszüchtung beim Kulturchampignon. III. Vermehrung durch Vielsporaussaat. Der Züchter 37,

16 109-119. Fritsche, Gerda, 1968. Untersuchungen des Nährbodeneinflusses auf verschiedene Myzelformen des Kulturchampignons. Mushr.Science VII, 515-529. Fritsche, Gerda, 1970. Prüfung eines physiologischen Unterscheidungsmerkmales zweier Champignonstämme hinsichtlich seiner Eignung zur Frühselektion bei Kreuzungen. Theoret. and Applied Genetics 40, 169-172. Fritsche, Gerda, 1976. Welche Möglichkeiten eröffnet der viersporige Champignon "Agaricus bitorquis (Quel.) Sacc." dem Züchter ? Theoret. and Applied Genetics 47, 125-131. Fritsche, G., 1980. Stand van de kruisingswerkzaamheden wit ras x tussenras respectievelijk bruin ras. De Champignoncultuur 24/1,11-15 . Fritsche, Gerda, 1983. Breeding Agaricus bisporus at the Mushroom Experimental Station, Horst. The Mushroom Journal 122, 49-53. Fritsche, Gerda, & Peter Van Loon, 1987. Breeding experiments with the wood blewit (Lepista nuda). Mushroom Science XII (part I), 227-235. Fritsche, Gerda, 1989. Ontwikkelingswerk met de akkerchampignon (Agaricus arvensis, Schaeffer ex.secr.) De Champignoncultuur 33/1,7-13 . Hasselbach, O.E., & P. Mutsers, 1971. Agaricus bitorquis (Quel.) Sacc., een warmteminnend familielid van de champignon. De Champignoncultuur 6, 211-219. Heltay, J., 1956. Report of the situation and problems of the hungarian mushroom research- and experimental work. Mushroom Science III, 199-217. Huhnke, W., & R. Von Sengbusch, 1968. Champignonanbau auf nicht kompostiertem Nährsubstrat. Mushroom Science VII, 405-419. Kligman, A.M., 1943. Some cultural and genetic problems in the cultivation of the mushroom Agaricuscampestris FR. Americ. Journal of Botany 30, 745-763. Kneebone, L.R., 1965. Spawn research at the Pennsylvania State University. Mushroom Science VI, 265-281. Lambert, E.B., 1929. The production of normal sporophores in monosporous cultures ofAgaricus campestris. Mycologia XXI/6, 333-335. Lambert, E.B., 1950. Comment of twenty years of research in mushroom culture. Mushroom Science I, 7-8 Lambert, E.B., 1959. Improving spawn cultures of cultivated mushrooms. Mushroom Science IV, 33-51. Lambert, E.B., & T.T. Ayers, 1962.Technique s for isolation of paired spores from a single basidium of the cultivated mushroom. Mushroom Science V, 185-187. Lemke, Gertraud, 1967. Über die Wirksamkeit von Aerosept-Verdampfern zur Raumluftdesinfektion bei der Champignon-Brutherstellung. Der Champignon 75, 24-28. Lemke, Gertraud, 1968. Beobachtungen bei der Kühllagerung von Körnerbrut. Mushroom Science VII, 543-552. Lemke, Gertraud, 1972. Praktische Erfahrungen bei der Champignonbrutherstellung. Der Champignon 126,5-23 . Loon, Van P.C.C. & G. Fritsche, 1987. Breeding experiments with Coprinus comatus. Mushroom Science XII (part I), 209-215. May, B. & D.J. Royse, 1981.Applicatio n of the electrophoretic methodology to the elucidation of genetic life histories of edible mushrooms. Mushroom Science XI (part 2), 799-817. Miller, R.E., & D.L. Kananen, 1971.Bipola r sexuality in the mushroom. Mushroom

17 Science VIII, 713-718. Miller, R.E. & W.A. Robbins & D.L. Kananen, 1974. Inheritance of sporophore color and "wild" morphology in Agaricus bisporus. Mushroom Science IX (part I),39-45 . Moessner, E.J., 1962. Preliminary studies of the possibility of obtaining improved cultures through mycelial fusion (anastomoses). Mushroom Science V, 197- 203. Pelham, J., 1965.Technique s for mushroom genetics. Mushroom Science VI, 49-64. Pol, Van De P.H., 1972. Nieuws van het Proefstation. De Champignoncultuur 16/9, 393-395. Pol, Van De P.H., 1974. Horst B30 komt in de handel. De Champignoncultuur 18/8, 255. Pol, Van De P.H., 1975. Bitorquis-rassen K26 en K32 in het verkeer. De Champignoncultuur 19/9,241-243 . Pol, Van De P.H., 1976. Ras Y217 binnenkort in de handel. De Champignoncultuur 20/3, 69. Poppe, J.A., 1972. Un exellent Agaricus tetrasporique cultivable commercialiment avec success. Mushroom Science VIII, 517-525. Raper, John R. & Carlene A. Raper, 1971. Life cycle and prospects for interstrain breeding inAgaricus bisporus. Mushroom Science VIII, 1-9. Raper, Carlene A., 1975. Sexuality and life-cycle of the edible, wild Agaricus bitorquis.Journ . of General Microbiology 95,54-66 . Riber Rasmussen, C, M.G. Amsen & Grethe Holmgaard, 1959. "Open Veiled" or "Hard Gilled" mushrooms. Mushroom Science IV, 416-429. San Antonio, J.P. & S.W. Hwang, 1970. Liquid nitrogen preservation of spawn stocks of the cultivated mushroom, Agaricus bisporus (Lange) Sing. Amer.Soc.Hort.Sci. 95 (5), 565-569. Sengbusch, Von R., 1931.Bitterstoffarm e Lupinen IL Der Züchter 3/4, 93-109. Sengbusch, Von R., 1970. Champignonforschung ja oder nein ? Der Champignon 111, 4-37. Sigel, Edith M. & J.W. Sinden, 1953. Variations in cultures made from the strain of mushrooms used at the Butler County Mushroom Farm INC. Mushroom Science II, 65-68. Sinden, J.W., 1932. Mushroom spawn and method of making same. United States Patent 1, 869,517. Stoller, B.B. & J.F. Stauffer, 1953. Studies on naturally occuring and ultraviolet radiation induced strains of the cultivated mushroom Agaricus campestris L. Mushroom Science II, 51-65. Stoller, B.B., 1962. Some practical aspects of making mushroom spawn. Mushroom Science V, 170-184. Till, O., 1961. Champignonkultur auf sterilisiertem Nährsubstrat. Die deutsche Gartenbauwirtschaft 9/10, 215-216. Tschierpe, HJ., 1959. Die Bedeutung des Kohlendioxyds für den Kulturchampignon. Die Gartenbauwissenschaft 24(6), 18-75. Uzonyi, Adele Latkoczky, 1959. Data on the determination of the cropping value of wild mushroom strains taken into culture. Mushroom Science IV, 379-392.

18 ANALYSIS OF THE GENOME OF AGARICUS BISPORUS A SHORT INTRODUCTION TO THE ECOLOGY, TAXONOMY AND NOMENCLATURE OF THE GENUS AGARICUS

C. Bas

Rijksherbarium, Leiden, The Netherlands

Summary

A short introduction is given to the distribution, ecology, taxonomy and nomenclature of the genusAgaricus. Th e estimated number of species ofAgaricus in existence is discussed. Continuation of the use of the name Agaricus bisporus is advised. Attention is drawn to a few, excellently edible, temperate species that seem to be suitable for cultivation but have not or hardly been tried. Keywords:Agaricus, distribution, ecology, species number, taxonomy, nomenclature, cultivation.

Distribution and ecology ofAgaricus

The genus Agaricus has a world-wide distribution. It occurs on the arctic tundra as well as in tropical rain forests. Although all (or nearly all) of its species are saprophytic the genus shows a wide ecological amplitude. Its representatives are found on the turf of alpine meadows as well as on grassy dunes and saltish seaside grass-lands, on humus and forest litter of coniferous as well as deciduous woods, on all kinds of accumulated vegetable matter, and on nearly all types of soil. It seems, however, to avoid, very acid and wet soils. In view of the manner in which A. bisporusan d A. bitorquis are being cultivated it is remarkable that, as far as I know, there are no truly fimicolous species in Agaricus. In other words, in nature there are no species of Agaricus growing directly on dung. Quite a few, however, prefer habitats enriched by dung or urine.

The number aïAgaricus specie s in existence

The number of species of Agaricus in existence is difficult to estimate because from a mycological point of view large areas of the world are underexplored. In the Netherlands Agaricus is represented by 45 to 50 species, in Europe by 70 to 90 species, whereas at the moment the total number of species in the world recognized by specialists lies between 200 and 250. It is not unreasonable to assume that the true total number of species of Agaricus will be somewhere between 300 and 400 and probably closer to the latter than to the former number.

Genus Micropsalliota Höhn

In the very closely, in my eyes too closely, related genus Micropsalliotaanothe r 50 species are known, all from tropical or subtropical regions. This genus may have to be fused with Agaricus because its supposed generic characters occur scattered in

21 Agaricus as well. Very little is known about the edibility of Micropsalliota species, but we may assume that many are edible. For the mushroom growers, however, they will be of little importance because of their tiny, sometimes almost membranaceous fruit-bodies.

History of the nameAgaricus

Nomenclaturally speaking, the genus Agaricus had a somewhat problematic childhood. From Linnaeus onwards to about the middle of the 19th century most fungi with fruit-bodies with gills were placed in the genus Agaricus. Gradually, however, well-recognizable groups of species were taken out of it and placed in new genera, to begin with genera like Russula, Amanita, Coprinus, etc. During this process also the group of species already recognized by Fries (1821: 280) as Agaricus tribus Psalliota was raised to generic rank under the name Psalliota (Kummer, 1871:23 ,72) . Finally, the entire genus Agaricus had been subdivided into new genera and the nameAgaricus had disappeared. P. Karsten (1879) was the first to become conscious of the fact that Psalliota represented the very core of the old genusAgaricus and restored the nameAgaricus for the genus Psalliota. Fortunately most authors agreed that Agaricus campestris had to be the lectotype of both Agaricus and Psalliota. The use of the name Psalliota persisted, however, till around 1950 (Moeller, 1950/51). Finally sophisticated nomenclatural reasons made it necessary to conserve the name Agaricus L.: Fr. with A. campestris as type species, so that we may assume now that the generic name Agaricus is fixed for ever.

Nomenclature of A. bisporus

Also the name of the most important species of the genus, viz.Agaricus bisporus, has caused some headaches. The early mushroom growers called theAgaricus under cultivation Agaricus or Psalliotacampestris. However, the trueA campestrisL. : Fr. is a rather widespread species from grasslands that is easily distinguished from A. bisporus. It has very bright flesh-pink gills when young, a different type of ring on the stem, 4-spored basidia and no sterile cells (cheilocystidia) along the gill edge. Jacob Lange (1926: 8) was the first author who clearly defined the cultivated, 2- spored Agaricus. He named it Psalliota hortensis var. bispora. Twenty years later Imbach (1946: 15) raised this variety to specific rank, which made Agaricusbisporus (J. Lange) Imbach the correct name. There is, however, a threatening cloud on the horizon. In 1900(: 16) Peck described from North America a brown, 2-spored Agaricus under the name A. brunnescens. According to Malloch (1976: 912) this species is identical withA. bisporusan d its name the oldest one for the cultivated species. But according to Singer (in Singer & Harris, 1987: 40) A. brunnescens and A. bisporus are different species. A third opinion is needed.

It is advisable that all workers onAgaricus continue to use the name A. bisporus. If Singer is right there is no reason to switch over toA. brunnescens,i f Malloch is right the name A. bisporusshoul d be conserved against A. brunnescens. The International Code of Botanical Nomenclature has explicitly opened the possibility of

22 conservation of names of economically important species. A proposal for the conservation of the name Agaricus bisporus should be written, however, by an experienced nomenclaturist because pitfalls are plentiful.

Taxonomy and subdivision ofAparicus

The taxonomie position of Agaricus is without great problems. Within the order Agaricales it belongs to the family Agaricaceae. In that family it is placed together with Micropsalliotaan d some small genera in tribus Agariceae which is distinguished from the other tribus, containing the Lepiota-like fungi, by the dark brown colour of the spores. The infrageneric classification of Agaricus is fairly simple also. In most systems published the grouping of the species is more or less the same, but differences of opinion exist about the taxonomie levels at which these groups should be recognized. The classification reproduced here is that of Singer (1986: 486). It does not deviate seriously from that published by Heinemann (1979: 8), the world specialist onAgaricus.

GenusAgaricus L.:Fr .

Subgenus Agaricus Section Agaricus cosmopolitan " Sanguinolenti " " Arvenses " Xanthodermi " Brunneopicti (sub)tropical

Subgenus Lanagaricus "

" Conioagaricus "

Whereas subgenus Agaricus contains the more typicalAgaricus species as we know them, subgenus Lanagaricus covers the (sub)tropical species with a rather loose, woolly outer layer on cap and lower part of stem and subgenus Conioagaricus accommodates the (sub)tropical species with very short to round cells on the cap.

More species to be tried for cultivation

In this concise paper, I have tried to emphasize the great number of species and the wide range of the ecological spectrum of the genus Agaricus. Probably it contains many good to excellent, edible species that can be successfully cultivated, but its great potentials have hardly been touched. One needs no more than the fingers of one or two hands to count the species that have been seriously tried. Not knowing much of the edibility of the tropical representatives, I restrict myself to comments on a few species I know well, that in my eyes should not be neglected. The most widely cultivated A. bisporusi n section Agaricus has the disadvantage of being 2-spored and not suitable for cross-breeding. Besides the already successful A. bitorquis,ther e are a few other 4-spored species close or very close to A. bisporus, viz. A. subfloccosus (rare), A. devoniensis (common on coastal dunes), and A.

23 subperonatus (perhaps just a variant of the rather common, nitrophilous A. vaporarius). The widely eaten A. campestrisseem s to be difficult to bring in culture, but it is a species with many variants and several relatives that may be easier to grow. Agaricus arvensisha s the disadvantage of a strong yellow discoloration of the fruit- body when bruised, but A. nivescens and A. macrosporus are magnificently white, hardly discolouring relatives, whereas the delicious A. augustus has large fruitbodies with a beautiful brown cap. Many species in section Arvenses have a very pleasant smell. More and more sophisticated research is done onA. bisporus,bu t let us hope that the unexploited gold still growing in field and forest will not be forgotten.

References

Fries, E.M., 1821.System a mycologicum 1. Lundae. Heinemann, P., 1979. Essai d'une clé de détermination des genres Agaricus et Micropsalliota. Sydowia 30: 6-37 ('1977'). Imbach, E.J., 1946. Pilzflora des Kantons Luzern und der angrenzenden Innerschweiz. Mitt, naturf. Ges. Luzern 15:5-85 . Karsten, P.A., 1879. Ryslands, Finlands och den Skandinaviska halföns Hattsvampar I. Bidr. Kann. Fini. Nat. Folk 32: I-XXVIII, 1-571. Kummer, P., 1871.De r Führer in die Pilzkunde. Zerbst. Lange, J.E. 1926. Studies in the agarics of Denmark VI. Dansk bot. Ark. 4 (12): 1- 52. Malloch, D., 1976. Agaricus brunnescens: the cultivated mushroom. Mycologia 68: 910-919. Moeller, F.H., 1950/51. Danish Psalliotaspecie s I and II. Friesia 4: 1-60, 135-220. Peck, C.H., 1900. New species of fungi. Bull. Torrey bot. Club 27: 14-21. Singer, R., 1986. Agaricales in modern taxonomy, Ed. 4., Koeltz, Koenigstein. Singer, R. & B. Harris, 1987. Mushrooms and truffles, Ed. 2, Koeltz. Koenigstein.

24 FREEZE-DRYING OF FUNGAL HYPHAE AND STABILITY OF THE PRODUCT

CS. Tan1, C.W. van Ingen2 & J.A. Stalpers1 ^entraalbureau voor Schimmelcultures, P.O. Box 273, 3740 AG Baarn, The Netherlands foundation for the Advancement of Public Health and Environmental Protection, P.O. Box 457, 3720 AL Bilthoven, The Netherlands

Summary

Hyphae of the Basidiomycetes Coprinus psychromorbidus, Lepista nuda, Perenniporiasubacida, Schizophyllum commune and of the Ascomycetes Altemaria dianthicola, Cercospora rautensis, Chaetomium globosum, Curvularia afßnis, Microsporum audouinii and Phoma terrestris were freeze-dried and viability was recorded. Optimal results were obtained when young colonies, cooled at a rate of - l°C/min, were freeze-dried entirely. Hyphae of Ascomycetes as well as Basidiomycetes did survive freeze-drying although generally better results were obtained with the Ascomycetes studied. The freeze-dried product was stable when stored for a period of 2 months at 30°C. Influence of the addition of trehalose varied with the organism studied. Keywords: freeze-drying, hyphae, trehalose.

Introduction

Methods of long-term preservation in which metabolism is inhibited during storage are preferable to maintenance on agar slants. The last mentioned method is laborious because regular transfers onto fresh media are required. In addition, repeatedly subcultured strains are subjected to mutations. Current methods for long-term preservation are freeze-drying and preservation at temperatures below -130°C. Cryopreservation methods give high survival rates and are universally applicable. However, some disadvantages are encountered. Methods using nitrogen are rather expensive and a regular supply of liquid nitrogen is not always guaranteed. The dependence of ultra-deep freezers on electricity may be a problem in developing countries. Organisms must be dispatched under frozen conditions or activated before transport. Freeze-drying is a good alternative. Ampoules can be stored easily in dense packing without any special requirements. Cultures need not to be revived on agar slants prior to dispatch. The product is light, inactive and dry, enabling easy distribution by mail. However, so far only conidia or spores can be freeze-dried successfully. Most attempts to revitalize dehydrated hyphae have failed, except for some successes reported for vesicular arbuscular mycorrhizal fungi (Tommerup, 1988), Claviceps (Pertot et al., 1977) and some Basidiomycetes (Bazzigher, 1962).

25 Previous studies have shown, that the survival rate of frozen fungal cells increases considerably when cells are cooled at a rate of -l°C/min (Mazur et al., 1972; Hwang et al., 1976; Grout and Morris, 1987). Several authors have stressed the beneficial effect of trehalose (Crowe et al, 1984, 1990; van Laere, 1990; Wiemken, 1990). Addition of peptides to the resuscitation medium is aimed to improve recovery (MacLeod and Calcott, 1976). After optimization of these parameters, hyphae of a test series of Ascomycetes and Basidiomycetes were freeze-dried. Trehalose was added to the growth medium as well as to the protectant to allow absorption by the cell; this would result in protection of cellular as well as organelle membranes.

Materials and methods

Cultures: Altemaria dianthicola Neergaard, CBS 112.38; Cercospora rautensis Massai., CBS 555.71; Chaetomium globosum Kunze : Fr., CBS 143.38; Coprinus psychromorbidus Redhead & Traquair, CBS 865.85; Curvulariaaffinis Boedijn, CBS 154.34; Lepista nuda (Bull.:Fr.)Cooke, CBS 589.76; Microsporum audouinii Gruby, CBS 317.51; Perenniporia subacida (Peck) Donk, CBS 463.50; Phoma terrestris H.N. Hansen, CBS 335.29; Schizophyllum commune Fr. : Fr., CBS 103.20.

Media: MEM 4%: 400 ml malt extract (TNO, Zeist, The Netherlands: 10% sugar), 600 ml water; MEA 4%: MEM 4% and 15 g agar; MEA 2%: 200 ml malt extract, 800 ml water, 15 g agar; MEMtreh: MEM 4% and 2% trehalose; OA: 30 g oat flakes (Quaker Oats B.V., Dordrecht, The Netherlands) in 11 water, 15 g agar; MP: 20 g powdered malt extract (Difco), 1 g peptone (Difco), 20 g glucose, 11 water.

Small colonies (0.5 to 1.5 mm in diam.) were obtained according to the following procedure. Organisms were grown in Petri dishes on an uncoated PT cellophane membrane on 15 ml MEA 4%. All strains were incubated at 24CC except Co. psychromorbidus (17°C). After one week for S. commune and Ch. globosum up to three weeks for the other organisms, the cultures were lifted from the cellophane and suspended in 10 ml MEM 4% in a Waring blender. The suspensions were mixed at low speed for 2 x 1 sec 5 ml suspension was added to 75 ml of either MEM 4% or MEMtreh in 300 ml Erlenmeyer flasks. The suspensions were incubated on a rotary shaker (150 rpm) at 24°C, except Co. psychromorbidus which was incubated 7 d at 24°C and subsequently at 17°C. Colonies of the ascomycetous strains were harvested 3 d later and colonies of the Basidiomycetes 10 d later. Three h prior to harvesting, 5% trehalose was added to a culture incubated in MEM 4%. After harvesting, individual colonies were transferred to 300 ßl 12% skimmed milk (Elk, DMV Campina B.V., Eindhoven, The Netherlands) or to 300 ßl 5% dextran/7% trehalose in a 1.5 ml vial (Müller & Müller, Deutschland) closed with a rubber lyophilization stopper diam. 12.7 mm (Heivoet, Aiken, Belgium).

26 Colonies were cooled at a rate of -l°C/min to a temperature of -45°C and subsequently to -75°C in 30 sec using a Sylab Icecube 1610. Freeze-drying was performed in a Leybold Heraeus GT 4 (condensor temperature -55°C; vacuum 2x10s milliter) for 2 h at -40°C followed by 20 h at -20°C and 8 h at 20°C resulting in a residual moisture content (RMC) of 2%. The ampoules were closed under vacuum.

Revival was recorded 16 h after freeze-drying as well as after a storage period of 2 months at 30°C. For revival, pellets were suspended in 10 ml MP and incubated for 16 h at 24°C on a rotary shaker (150 rpm). Afterwards, the mycelia of A. dianthicola, Ce. rautensis,Ch. globosum, Cu. affinis and Ph. terrestris were transferred to 15 ml OA; M. audouinii was transferred to 15 ml MEA 4%; the other organisms were transferred to MEA 2%.A. dianthicola, Ce. rautensis,Cu. affinis, M. audouinii and S. commune were incubated at 24°C, Ch. globosum at 28°C; Co. psychromorbidus at 17°Can d the others at 22°C.

RMC was estimated by the titrimetric procedure of Karl-Fischer (Anonymous, 1990) with a Mitsubishi CA05 Moisture Analyzer coulometric titrator with microprocessor control (Mitsubishi Chemical Industries Ltd, Tokyo, Japan). The freeze-dried material was dissolved in Coulomat A (Riedel de Haën A.G., Seelze, Germany). The generator solution cell was filled with 100 ml Coulomat C (Riedel de Haën A.G., Seelze, Germany); the cathode solution cell was filled with 5 ml Coulomat A.

Results and discussion

Hyphae were cooled at a rate of -l°C/min prior to drying which is supposed to enhance survival (Mazur et al., 1972; Grout and Morris, 1987). With faster cooling rates intracellular ice crystals are produced, which cause irreversible damage to the cell. With hyphae, an additional problem is encountered. With the preparation of suspensions of hyphal cultures , hyphae need to be broken in small fragments. Such fragments are open ended. During cooling, crystals, produced in the medium, nucleate the hyphae via these open ends. The ice crystals enter the adjacent cell through the septal pore and subsequently expand continuously throughout the whole hyphal system. As a consequence, hyphal fragments are principally unsuitable to lyophilize. For that reason, young colonies with undamaged cells containing living cytoplasma, were freeze-dried entirely. Most strains used in this study had completely lost their ability to sporulate. Chaetomium globosum and Schizophyllum commune did produce asco- and basidiospores, respectively. In the experiments however, young colonies consisting of sterile hyphae only were freeze-dried. The sporulating strains were included to verify whether the ability to sporulate was retained after freeze-drying.

27 Table 1. Survival of colonies 16 hours after freeze-drying and after a storage period of 2 months at 30°C. Colonies are incubated in malt-medium (M) or malt-medium supplemented with 2% trehalose (M2%treh). In addition, 5% trehalose is added 3 hours prior to freeze-drying to a culture incubated in malt-medium (M(3h5%treh)).

M M2%treh M(3h5%treh) sm td sm td sm td + + + - + + + Alternariadianthicola 16 h 0 16 0 15 13 9 5 8 16 3 9 6 2m30°C2 9 4 5 13 0 0 5 12 0 7 5

Curvularia affinis 16 h 6 5 0 10 1 9 1 10 1 10 3 1 2m30°C 17 0 2 0 6 2 3 4 10 6

Phoma terrestris 16 h 6 6 3 5 7 3 0 14 0 11 0 10 2m30°C0 8 0 10 0 7 0 9 0 7 8 0

Cercospora rautensis 16 h 20 0 15 0 9 0 15 5 15 5 15 5 2m30°C 13 0 7 2 10 0 16 1 13 0 13 0

Chaetomium globosum 2m30°C 9 0 2 0

Coprinuspsychromorbidus 16 h 1 16 7 3 2 7 2 14 7 12 6 8 2m30°C4 12 19 7 0 17 0 21 11 4 0 19

Schizophyllum commune 16 h 15 0 12 0 27 0 24 0 5 0 42 0 2m30°C 28 0 27 0 24 4 26 0 22 0 23 0 sm = 12% skimmed-milk; td = 5% dextran/7% trehalose; - = number of negative colonies; + = number of positiv colonies

Most strains could be revived (Table 1). The Ascomycetes as well as the Basidiomycetes showed the same characteristics as before freeze-drying, including the production of ascomata by Chaetomium and the basidia by Schizophyllum. The freeze-dried materials were stable when stored at 30CC for a period of 2 months, except Ph. terrestriswher e only colonies preincubated for 3 h in 5% trehalose and protected by 7% trehalose/5% dextran could be revived.

28 The survival rates of the Basidiomycetes studied were found to be lower than those of the Ascomycetes. One of the 6 Ascomycetes studied could not be revived while 2 of the 4 Basidiomycetes had lost viability. Results were negative for the Ascomycete M. audouinii and for the Basidiomycetes L. nuda and Pe. subacida (results not shown). Difficulties in obtaining young colonies of good quality for the Basidiomycetes might have been the cause. Influence of the addition of trehalose varied with the organism studied. A. dianthicola showed a higher survival rate when preincubated in media supplemented with trehalose and when protected by 12% skimmed-milk. As was mentioned before, Ph. terrestris survived a storage period for 2 months at 30°C when preincubated in 5% trehalose for 3 h and when protected by 5% dextran/7% trehalose. Co. psychromorbidus showed the highest survival rate either when preincubated in MEM 4% and protected by a lyoprotectant containing trehalose or when preincubated for 3 h in 5% trehalose and protected by 12% skimmed milk. Survival for the other organisms was independent on the composition of both the preincubation medium and the lyoprotectant.

Acknowledgement

The authors are indebted to B. Stevens for his technical assistance.

References

Anonymous, 1990. Water determination. Pp. 1619-1621. In USP XXII. The United States Pharmacopeia. Bazzigher, G., 1962. Ein vereinfachtes Gefriertrocknungsverfarhen zur Konservierung von Pilzkulturen. Phytopathol. Z. 45:53-56. Crowe, J.H., L.M. Crowe & D. Chapman, 1984. Preservation of membranes in anhydrobiotic organisms: the role of trehalose. Science 223:701-703 . Crowe, J.H., J.F. Carpenter, L.M. Crowe & T.J. Anchordoguy, 1990. Are freezing and dehydration similar stress vectors? A comparison of modes of interaction of stabilizing solutes with biomolecules. Cryobiology 27: 219-231. Grout, B.W.W. & G.J. Morris, 1987. Freezing and cellular organization. In: B.W.W.Grout & G.J. Morris (Ed.): The effects of low temperatures on biological systems. Edward Arnold, Cambridge, p. 147-173. Hwang, S.W., W.F. Kwolek & W.C. Haynes, 1976. Investigation of ultra low temperature for fungal cultures III. Viability and growth rate of mycelial cultures followingcryogenic storage. Mycologia 68: 377-387. Laere, A.J. van, 1990. Metabolism and function of trehalose. In: A. Reisinger, and A. Bresinsky (Ed.): Fourth International Mycological Congress. Regensburg, p. 198. MacLeod, R.A. & P.H. Calcott, 1976. Cold shock and freezing damage to microbes. In: T.R.G. Gray & J.R. Postgate (Ed.): The survival of vegetative microbes. Twenty sixth symposium of the Society for General Microbiology. Cambridge University Press, Cambridge, p. 81-109. Mazur, P. S.P. Leibo & E.H.Y. Chu, 1972. A two-factor hypothesis of freezing injury. Evidence from Chinese hamster tissue culture cells. Exp. Cell Res. 71: 345-355.

29 Pertot, E., A. Pue & M. Kremser, 1977. Lyophilization of nonsporulating strains of the fungus Claviceps. Eur. J. Appl. Microbiol. 4: 289-294. Tommerup, I.C., 1988. Long-term preservation by L-drying and storage of vesicular arbuscular mycorrhizal fungi. Trans. Brit. Mycol. Soc. 90: 585-591. Wiemken, A., 1990. Trehalose in yeast, stress protectant rather than reserve carbohydrate. Antonie van Leeuwenhoek Ned. Tijdschr. Hyg. 58: 209-217.

30 A GENETIC LINKAGE MAP FOR AGARICUS BISPORUS

R. W. Kerrigan, P. A. Horgen, and J. B. Anderson

Centre for Plant Biotechnology, and Dept. of Botany, Erindale College, University of Toronto, Mississauga, Ontario, Canada

Introduction

A limited number of genetic markers, including allozymes, auxotrophic lesions,an d morphological aberrations, have been reported in the cultivated mushroom Aaaricus bisporus since 1972. These markers enabled for the first time some fundamental, albeit rather rudimentary genetic characterizations of the biological material and its behavior. Auxotrophy was usedt o demonstrate recombination of the genetic material during the presumably meiotic process of basidiosporogenesis (Raper gtaL, 1972). With the advent of allozyme markers it became possible to identify homokaryons and successful heterokaryon-crosses with high confidence, and to initiate genotypic characterization of strains (Royse & May, 1982a, 1982b). Allozyme analysis also provided limited evidence for independent assortment and genetic linkage during the meiotic process (Spear et al.. 1983). These were important milestones in the field of mushroom strain development and protection.

DNA markersfo r A. bisporus were first described by Castlee la_ L (1987). Only with thedevelopmen t ofthes emarkers ,includin grestrictio nfragmen t length polymorphisms (RFLPs) and random amplified polymorphic DNAs (RAPDs), has it become possible to address genetical issuesi nA . bisporus ina comprehensive , if not exhaustive, manner. With a large (and potentially limitless) set of markers, more complex issues can be addressed: What is the overall structure of the nuclear genome? How many chromosomes and linkage groups are present? Isthei r composition homogeneous or not? How does the genome behave during meiosis? Dochromosome s assort with full independence? What isth e frequency of crossovers? And,mos t importantly: Canth e agronomic phenotypes of offspring be associated with the inheritance of particular markers and chromosomal segments?

Conventionally these questions areaddresse d bydevelopin g data onth e inheritance of genetic markers and organizing these data into a map of genetic linkage relationships. These relationships are statistical, but normally reflect the analogous physical linkage relationships of markers distributed on several linear DNA molecules (i.e. chromosomes). We describe below the approach we used to produce the first comprehensive map of genetic linkages in A. bisporus. The map, some unusual results, and their applications and implications will then be presented.

Methods and Results

Two things must bedevelope d in aconventiona l linkage mapping study: a progeny set and a marker set. We have described elsewhere the details of producing our set of homokaryons (Kerrigan §1 aL, submitted) and will simply outline the process here. In fungi, the free-living homokaryon stage permits the analysis of haploid (1N) postmeiotic nuclei,rathe r thanth e morecumbersom e analysis of diploid (2N) genomes common to most plants and animals. These homokaryons are more comparable to eukaryotic gametes than to 2N progeny, and we suggest they instead be called hemi-progeny, to avoid confusion (Kerrigan & Ross, 1987).

31 The secondarily homothallic lifestyle of A. bisporus unfortunately complicates the ordinarily simpler task of fungal genome analysis: homokaryons areuncommo n among the spore-offspring of this species,an dther e is nomorphologica l criterion (sucha sth e clamp connection of the dikaryotic basidiomycetes) by which they can beidentifie d in culture. For this reason we constucted a cross (AG 93b) between an auxotrophic mutant (H1-1, = ATCC 24662: Raper e£ aL 1972) and a prototrophic protoplast regenerate (H89-65) of a field-collected heterokaryon (AG 89: see Malloch et sL 1987), reasoning that negative selection on minimal medium would identify among hemi-progeny the expected half inheriting the auxotrophic marker, plus a few homozygous heterokaryons, which are rare (Royse & May, 1982b; Summerbell et al.. 1989). If this strategy had worked, it would have permitted undistorted mapping of the whole genome with the exception of segments linked to the auxotrophic lesion. However, it did not work: the trait was not penetrant within the recombined hybrid genotypes produced meiotically by AG 93b.

Consequently we used the following approach. Basidiospores from 93b were germinated on PDA containing antibiotics, with a single marked, unrelated heterokaryon introduced at one edge of each plate as a stimulus. The frequency of successful germination after 21 dwa s quite low,c_a0.0077 . 639 single-spore isolates (SSIs) were transfered to CYM and 576 remained sufficiently vigorous to permit further analysis. The weekly increment of radial growth on CYM and MinM was recorded for each SSI shortly after isolation (stability of these traits is reviewed elsewhere).

Beginning with the most slowly growing SSIs, allozyme phenotypes at two polymorphic loci were determined for 317 SSIs. A novel, nearly null allele restricted analysis of the PEP1 locus, but homo- or hemi-allelic isolates at PEP2 could be recognized unambiguously among the more common heterozygous progeny. 52 of these putative homokaryons (PHs) were found, primarily among the slowest growth-rate classes of SSIs. During mapping it became clear that virtually all of these were true homokaryons, providing further evidence that PEP2 is a locus that rarely if ever becomes homozygous due to recombination with its centromere in heterokaryon progeny of A. bisporus. We estimate that more than 90% of all homokaryons in the sample were recovered. The impracticality of exhaustively recovering the few faster-growing homokaryons introduces a potential genetic bias into our sample set, which, however is probably moretha n offset by numerous genotypes that permit little or no growth of the SSI. Skewed segregation ratios of certain markers would not be surprising under these constraints.

In addition to 50 PH SSIs identified in time for further study, 3 PH protoplast regenerates were identified among 275 produced from 14 93b progeny. It is unfortunate that the efficiency of homokaryon production via protoplast production of 93b progeny is so low, since it would be very interesting to compare meiotic events preceeding production of 1N vs^ 2N basidiospores. Ultimately 53 post-meiotic homokaryons, the parental 93b heterokaryon, and the two pre-parental homokaryons were subjected to DNA marker analysis.

The production of Aaaricus DNA restriction fragments, their incorporation into pUC1 8t o produce recombinant plasmids,thei r incorporation into clonal lines of E^coli . and their use as anonymous probes in Southern hybridizations against blotted, restriction endonuclease digested genomic DNAs was first described by Castle el aL (1987). We evaluated all of the plasmids previously found to detect RFLPs in A. bisoorus. aswel l asalmos t 100 others containing inserts of A. bisporus (AG 33 or AG H1-1) or A. bitorauis (AG 4) DNA. We found 35 interprétable RFLPs segregating among 93b hemi-progeny. Many probes gave multiple signals, which allowed very efficient linkagemappin g butgreatl ycomplicate d ourcomplementar y physical mapping

32 of chromosomes. In one extreme case, hybridization of plasmid p33n25 at reduced stringency to the genomic DNAs produced more than 25 distinct bands, among them 11 scorable markers on several linkage groups. Multiple copies of DNA segments appear to be unusually diverse and abundant in the genome of A. bisporus. An explanation for this will make more sense following other of our results.

A new source of DNA markers, called RAPDs, has been described quite recently (Williams elaL, 1990). These rely onth e polymerase chain reaction (PCR) andsingle , short primers of arbitrary sequence to amplify specific, anonymous portions of the genome. We evaluated 28 primers; 12 gave products, while five produced multiple stable polymorphisms. Almost all RAPD polymorphisms involved the presence or absence of a PCR product of one size, rather than variation in length of an amplified product. Most RAPDs are therefore dominant markers, whereas most allozymes and RFLPs are codominantly expressed. Intotal , 22 RAPD markers were scored. Finally, someindividua l RAPDproduct s wereisolated ,reamplified ,ge lpurified ,an dlabelle dan d used as probes in southern hybridizations with genomic DNAs. Insimpl e cases these results duplicated the corresponding RAPD marker data, but in others, families of dispersed, repeated polymorphic elements were detected. As of this writing, 66 allozyme, RFLP, RAPD, and RAPD-RFLP markers have been scored over the 93b hemi-progeny set, and statistically analyzed for linkage.

An initial observation was that the segregation ratios of several markers (and groups) deviated significantly from classical Mendelianexpectations . As noted above, this could be expected when not all genotypes are equally likely to be sampled, and might also be expected when spore germination occurs at such a low rate. This segregational skew will seriously compromise statistical tests of linkage that rely upon Mendelian expectations (SLL. Fincham e_î aL, p. 94).

Two approaches to linkage analysis were followed. A direct statistical test of joint segregation was made for all pairs of markers, using the Chi-square test. The observed segregational skew necessitatedcomputatio n ofth eexpecte dsize so fal lfou r genetic classes of hemi-progeny for each pair of informative (heteroallelic parental) markers. This expectation was based on the frequencies of each allele in the subset of hemi-progeny for which neither marker score was missing (missing data amounted to £âi 10% of the total). The analysis was performed by software written in the dBASE III language by RWK.

Significance levels of p.< 0.05 and p.< 0.01 were used as thresholds to identify probable andhighl y probable linkage relationships, respectively. At levels ofp . < 0.01, indicating outcomes with lesstha n onechanc e in onehundre d of occurring bychance , 11 linkage groups were detected. 56 of 66 markers exhibited linkage at this stringency. Relaxing the stringency of the linkage support criterion to ß < 0.05 caused all but two markers to appear linked. However, another unusual phenomenon became apparent when these weaker 'linkages' were considered: the assumption of linear correspondence between linkage group and physical chromosome was often violated. At this level of resolution, the map of linkage relationships begins to resemble a single branching network rather than a set of discrete linear gene orders.

A second approach to the analysis of linkage used an indirect statistical test: the comparison of (vanishingly small) likelihoods of exactly reproducing the actual data set given alternative hypothetical linkage relationships. We used Mapmaker Macintosh, a generous gift of Scott Tingey of Du Pont Company, to compute and compare the logs of these likelihood ratios (= LOD scores). Our early experience with the Mapmaker algorithms isthat , inth e simultaneous analysis of multipoint data they are quite useful in determining most probable gene orders, however in the two-point analysis of potential linkage relationships the default threshold of anLO Dscor e of 3.0

33 is far less stringent than a Chi-square based significance threshold of ja < 0.01. At an LOD threshold of 3.0 Mapmaker recognized only four linkage groups (and two unlinked markers) inth e 93b data set. At an LOD threshold of 2.0, corresponding to odds of 100 to 1 or better in support of linkage, the entire genome of A. bisporus cross 93b collapsed into a single effective linkage group.

Discussion

The observed level of segregational interdependence over virtually the whole genome is most unusual. To illustrate this point, the more typical genetic behavior of the oomycete fungus Bremia lactucae may be compared. When the same LOD threshold of 2.0 was used to evaluate two-point data from B. lactucae. 13 linkage groups were found,whil e moretha n 40% of all markers remained unlinked (Hulbert §1 âL, 1988). The sizeo f the B.lactuca e haploid genome is only about 46% greater than that of A. bisporus. hardly sufficient to account for the very different levels of joint segregation observed in the two organisms. The small number of unlinked markers suggest a low frequency of crossovers ( = small map length) in A. bisporus. while the reduced number of linkage groups detected at conventional LOD thresholds corresponds to the 'linkage networks' seen in the Chi-square analysis at 0.01 < fi < 0.05.

Our preferred explanatory hypothesis for this very unusual genetic behavior proposes that many chromosomal translocations have occurred over evolutionary time in populations of A. bisporus. Multiple translocation heterozygosities could occur in certainheterokaryons , including AG 93b. That segregation in such genomes leads to lethal deletions insom e cases, andt o non-linear linkage relationships in others, is well known (Strickberger, 1976). This would account for at least part of the inviability observed in 93b basidiospores. DNA sequence duplications will arise in this way if they are tolerated by the organism, and would lead to an elevated number of low-repeat DNA markers, as we have found in this species.

Afurther , potentially relatedclas so f phenomena, called 'pseudolinkages,' mayals o be operating in this cross or species. We may postulate, given gene duplications and reduced selection against deleterious mutations under secondary homothallism (Kerrigan, 1989), that a certain level of intragenomic complementation occurs. If complemented lesions, occurring independently in, for example, unlinked translocated-duplicated segments of the genome, are reciprocally arranged, and translocation heterozygosity exists, then hemi-progeny may beles s likely to be viable unless both chromosomes are inherited jointly from a single parent. We think that translocation and fitness interaction effects are most likely responsible for the extension of statistical 'linkages' beyond the physical limits of chromosomes.

Physical mapping of markers to electrophoretically separated chromosomes cang o far to resolve such uncertainties. Ina paralle l study led by John Royer of our research group, at least 10 and probably 13 chromosomes have been separated by Contour-clamped Homogeneous Electrical Field electrophoresis, and then transferred to hybridization membranes. All of our single-copy RFLP markers plus several others have been hybridized to these blots. The correspondence of individual 'core' linkage groups to specific chromosomes is now fairly well established. We are presently concentrating on physically mapping through one or more suspected translocations, to test the hypothesis presented above. At this writing we have only one distinctly anomalous result to report: markers p4n6 and p1n147 both mapped to asingl e point on linkage group G, but these probes hybridized to two physically separate chromosomal bands! Assuming that this result is reproducible, which is now being determined, an absolute 'psei olinkage' between these chromosomes must exist.

34 'Pseudolinkage' is far more than a nuisance. If widespread in A. bisoorus. it will demand an additional level of comprehension and competence in engineering successful gene combinations for improved cultivar strains. Knowledge of both physical and fitness relationships between loci will be required for breeding success rates comparable to those of other vegetable crops. We have observed correlations betweenth e inheritance of certainmarker s from 93b andth e vigor of homokaryons on either CYM or MinM, and have also located a primary determinant of colony color in homokaryons grown on PDA. We have no doubt that the next major period of A. bisporus strain development will be marked by the successful association of neutral genetic markers with linked determinants of numerous economically important traits. Wewoul dsuggest , however, that, onceagain ,th e unusual life-cycle of thespecie sha s been found to give rise to quirks that will make selective breeding of A. bisoorus an unusually challenging enterprise.

References

Castle, A.J., P. A. Horgen, & J. B. Anderson, 1987. Restriction fragment length polymorphisms in the mushrooms Aqaricus brunnescens and Aaaricus bitorauis. Applied and Environmental Microbiology 53:816-822. Hulbert, S. H., T. W. Ilott, E.J . Legg, S. E. Lincoln, E. S. Lander, & R.W . Michelmore, 1988. Genetic analysis of the fungus, Bremia lactucae. using restriction fragment length polymorphisms. Genetics 120:947-958. Kerrigan, R. W., 1989. Evolution and Aaaricus bisporus. Ph.D. dissertation, University of California, Santa Barbara. Kerrigan, R.W. , L. M. Bailer, P.A . Horgen, & J. B. Anderson, (submitted). Strategies for the efficient recovery of Aaaricus bisporus homokaryons. Kerrigan, R. W., & I. K. Ross, 1987. Dynamic aspects of basidiospore number in Aaaricus. Mycologia 79:204-215. Malloch, D., A. Castle, &W . Hintz, 1987. Further evidence for Aaaricus brunnescens Peck as the preferred name for the cultivated Aaaricus. Mycologia 79:839-846. Raper, C. A., J. R. Raper, & R. E. Miller, 1972. Genetic analysis of the lifecycle of Aaaricus bisoorus. Mycologia 64:1088-1117. Royse,D . J.,& B. May, 1982a. Use of isozyme variation to identify genotypic classes of Aaaricus brunnescens. Mycologia 74:93-102. Royse, D. J., & B. May, 1982b. Genetic relatedness and its application in selective breeding of Aaaricus brunnescens. Mycologia 74:569-575. Strickberger, M. W., 1976. Genetics. Second Edition. Macmillan Publishing Company, Inc., New York. pp. 495-529. Spear, M. C, D. J. Royse, & B. May, 1983. Atypical meiosis andjoin t segregation of biochemical loci in Aaaricus brunnescens. Journal of Heredity 74:417-420. Summerbell, R. C, A. J. Castle, P. A. Horgen, & J. B. Anderson, 1989. Inheritance of restriction fragment length polymorphisms in Aaaricus brunnescens. Genetics 123:293-300. Williams, J. G. K., A. R. Kubelik, K. J. Livak, J. A. Rafalski, & S. V. Tingy, 1990. DNA polymorphisms amplified by arbitrary primers are useful as genetic markers. Nucleic Acids Research 18:6531-6535.

FIGURE 1 (next page)

Genetic linkage mapo f A. bisporus 93b. Mapmaker Macintosh Multipoint Analysis was used to produce a strictly linear maximum likelihood representation of statistical correlations in the marker segregation data. Linkage groups C-D, F-H, and E-l are known to span more than one chromosome. Raising the LODthreshol d breaks these compound groups into "core" groups corresponding in most cases to single chromosomes. Non-linear linkages and pseudolinkages are not represented by this method.

35 (3) PEP1 (4.9%) 5.1 (2) R4_3 (4.9%) 5.1 (8) PEP2 ( 0.0%) 0.0 P1N31 • (35) P33N7 (0.0%) 0.0 W (4,1 %) 4.2 • P1N150 - (36) R21_5S1 { 0.0 %) 0.0 (5) P1N17 ( 0.0 %) 0.0 (6) (10.5%) 11.8 • P1N148 (7.7%) 8.4 J/- (7) J (10) P33N25J1 (15.8%) 19.0 ï\ (9) P33N25_4

(16.9%) 20.6 •

(1.9%) 2.0 • • (24) R18_4 (3.8%) 4.0 • • (25) P33N16 • (26) P33N5 - (53) R22_2 ( 4.5 %) 4.7 - (54) R9_1 (0.0%) 0.0 -^- - (55) P1N55 (4.1 %) 4.3 -y^ - (56) P33N25J (2.1%) 2.1 *•$ • (57) R21_2S2

Jh

- (28) P33N25_2

(8.0%) 8.7 - (0.0%) 0.0 • - (30) P33N25_3 (2.3%) 2.3 - - (31) P33N10 •- (32) P1N38

D --

(0.0%) 0.0 -y Ç^ (17) P33N25 (0.0%) 0.0 S/L . 4 n -/f

F-fl_• (40) P33N25_7 (5.2%) 5.5 • B - (41) P1N36 (5.0%) 5.3 • • (42) P33N25_6

(43) P1N147 ( 0.0 %) 0.0 -^ (44) P1N121 \ (45) P4N6

(47) R1B_3

(31.2%) 49.0 (48) P4N14 (8.7%) 9.6 • (2.5%) 2.6 • (49) P1NY H (50) P1N1 G

36 MITOCHONDRIALPLASMID S ANDHOMOLOGOU S SEQUENCESI N AGARICUS

Mary M.Robiso n &Pau l A.Horge n Mushroom Research Group,Centr efo r Plant Biotechnology University of Toronto,Erindal e Campus,Mississauga ,Ontario , Canada

Summary Manyisolate so fAgaricu sbitorqui shav ebee nfoun d tocontai n autonomously- replicatingplasmid so fdifferen t sizes. Thetw oplasmid s found inon eparticula r isolate havebee nexamine d indetai lan dhav ebee nfoun d tob elinea rmolecule s residing within themitochondrion . Theseplasmid s sharestructura l andgeneti c characteristics withlinea rmitochondria l plasmids found inothe rfung i andplants , including thepossessio n of twoope nreadin gframe s (ORFs) thatcoul dencod e virus-like DNA andRN Apolymerases . Although plasmids have notbee n found in several species of Agaricus.includin gA.bisporus .mitochondria l DNA (mtDNA) sequencesexis ti n these speciestha tar ehomologou st oth eOR Fo npE M corresponding to anRN Apolymerase . Sequence analysis of apEM-homologou s mtDNA sequencefro m A.bisporus suggests thatth emitochondria lsequenc ei s relatedbu tno tderive dfro m thepE Msequence . Keywords:Agaricus .plasmid ,mitochondrion , RNA polymerase Introduction Extrachromosomal elements,o rplasmids ,hav ebee nreporte d in awid evariet y ofbot hplant s andfungi . Manyo f theseplasmid shav ebee ndetermine d tob elinea r molecules of DNA,an do f these,mos t appear toresid ewithi n the mitochondrion, apparently in association withth emtDN A (Meinhardt etal . 1990). Examination ofDNA sfro m isolateso f thefiel d mushroom, Agaricus bitorquis. byagaros e gelelectrophoresis ,ha sreveale dth eexistenc e ofplasmid s within several of theseisolates . Plasmidso f asimila rsiz ema yexis twithi n several different isolates;alternatively , moretha non e sizeo fplasmi dma yexis t within asingl e isolate(Meye re tal . 1988). Oneisolate ,Ag4 ,contain stw oplasmid si nrelativel y high copy number andwa schose n for further investigation. Thelarge rplasmid , pEM,i sabou t 7.4kbi n length andha s been the subject of most subsequent work. The smallerplasmid ,pMPJ ,i sabou t 3.9kbi nlength . Both plasmids havebee n demonstrated byrestrictio n digestst ob elinea rmolecules ,an dt oresid e exclusively within themitochondrio n (Fig.1,Moha ne t al. 1984). Both molecules alsoposses s inverted repeated sequences atthei rtermin i(Fig.1) ,a characteristi c sharedwit h otherlinea rmitochondria l plasmids (Meinhardte t al. 1990).

Examination ofDNA sfro m bothcommercially-grow n and wild-collected isolates of Agaricus bisporus has not,t odate ,reveale d theexistenc e of any plasmids within that species. However, iti spossibl etha tth ecop y numbero fa plasmid may bes olo wa st opreven tdetectio n byethidium-bromid estaine dgels . Probing Southern blotso f A.bisporus DNAs withpE Mfragment s hasno treveale d any lowcop y numberpEM-homologou splasmids .

37 Plasmids in other specieso f Agaricus Other specieso f Agaricus.fro m alleigh tSection s of thegenus ,hav ebee n checked for thepresenc eo fplasmids .S ofar , plasmidsappea ri n atleas t four species,othe rtha n A.bitorquis.includin g Agaricus bernardii.Agaricu spattersonae . Agaricus fuscofibrillosus andAgaricu s subperonatus.representin g threeSections . Plasmids arequit eprevalen ti nisolate so f A.bitorquis ,eve n thosefro m diverse geographic origins.Interestingly , manyo fth eplasmid si n A.bitorquis have similar restriction digestpattern s andhomolog y topEM ,a sreveale db y Southern blotting. Sofar , all isolates ofA.bitorqui s obtainedi n California, aswel l asa fe w Ontario isolates,contai n plasmids that arehomologou st opE Man dhav eth e samerestrictio n digestpattern ,wherea spE Mwa sobtaine dfro m aisolat eoriginatin g inIllinoi s (ATCC 24666).Plasmid s of other species, such asthos ei n A.pattersonae. A.bernardiian dA.subperonatu sals ohav ehomolog y topE Mbu t lackth e similarity inrestrictio n digestpatterns . Again,th enumbe ro fplasmid s found canb e considered only asa lowe restimat eo f thosetha texist . Iti sunknow n whether allo f theseplasmids ,eve nthos eexhibitin ghomolog y topEM ,ar elinea rmolecule so r associated withth emitochondrion . Generic andstructura lorganizatio n ofpE M BothpE M andpMP Jhav epropertie s apparently shared by all well-characterized linear mitochondrial plasmids (Meinhardt et al. 1990). Thetermin i aremad eu p of invertedrepeate d sequences thatar esevera lhundre d basepair s long,a smentione d earlier (Mohan et al. 1984). Thetermina l 5'nucleotide s appeart ob ecovalentl y linkedt oproteins ,wherea s the3 'nucleotide s areavailabl e for exonucleolytic digestion (Robison, M.Sc. 1988). Sequencingo fthre einterna l contiguousEcoR I fragments of pEM,representin g 80%o f themolecule ,reveale d two longope n reading frames (ORFs)presen to nopposit e strands of themolecul e(Fig.2) . Transcription wouldpotentiall y proceed from thetermina l regions towardsth e middle,reachin g translational stopcodon s 105bpapar t (Robisone t al. 1991). Aminoaci d sequencesderive dfro m ORF1an dORF 2o fpE Mwer ecompare d to thosefro m sequencescatalogue d inGenBank . Therear e strong similarities between pEMORFl anda particula r typeo fDN Apolymeras eencode d bycertai n viruses, such assom e bacteriophages andadenovirus2 . Likewise,pEMORF 2i s very similar toa nunusua ltyp eo f RNApolymeras e encoded by bacteriophagesT3 , T7 andSP6 ,a swel l asth enuclear-encode d mitochondrial RNApolymeras eo f yeast (Saccharomycescerevisiae ) (Robison et al. 1991). Not surprisingly, the translated sequenceso fORF 1an dORF 2ar eeve n more similart otranslate d ORFs ofothe r linearmitochondria l plasmids thathav ebee nshow nt oencod eth esam e typeso fDN Aan dRN Apolymerases . However,n o similarity exists atth e nucleotideleve lbetwee npE Man dan yothe rsequence d linearmitochondria l plasmid (Robison et al. 1991). Theprevalenc e of thesetype so f plasmids in widely- divergent specieso fplant san dfungi , andth eabilit y of theseplasmid st oencod e highly homologousproduct s but without similarity atth e nucleotide level, suggests thatlinea rmitochondria l plasmidsdelineat ea larg euniqu eclas so fmolecule stha t havebee n associated withthei rhos t mitochondria for aver ylon gtime ,perhap s sinceth eappearanc eo f theeukaryoti ccell .

Homologyo fpE Mt omitochondria l DNAs Therelationshi po fpE Mt oit shos t mitochondrial DNA hasbee nconsidere db y determining ifan yhomolog y exists between pEMan dth emtDNA so f Ag4o rothe r isolateso f A.bitorquis. The twointerna lEcoR Ifragment s ofpE Mtha tcorrelat e approximately withth eRN Apolymerase ,hybridiz et omtDN Afragment s from all isolateso f A.bitorquis that havebee nexamine d sofa r (Meyere t al. 1988, unpub. data) Over6 5isolate so f A.bisporusals ocontai n apEM-homologou smtDN A

38 fragment (unpub.data) . However, the smallerinterna lfragmen t ofpE Mtha t correlates with theDN Apolymeras edoe sno thybridiz e tomtDNA sfro m these species,exceptin g DNA from aparticula rmitochondria l typeo fA.bitorqui s found in someOntari oisolates . ThemtDNA so fA.bisporu sappea r tob erelativel y homogeneous in termso f sizean drestrictio n fragment patterns,wherea s themtDNA s of A.bitorquis vary widely (Hintze t al. 1985). Interestingly, themtDN Afragmen t generated byEcoR I digestion that hybridizes topEM' s RNApolymerase ,i sth e samesiz efo r allisolate s within each species (A.bisporus andA.bitorquis ^bu tdifferen t between thetw o species (unpub.data) . Aswell ,mtDN A sequences immediately flanking thepEM - homologous sequencei non eisolat eo fA.bisporus .ar efoun d tob elinke d toth e pEM-homologous sequences in allothe risolate so f A.bisporus (unpub.data) , suggesting thatmitochondria l genomicorganizatio n isver yconserve d atthi slocu s in this species. Surprisingly, oneo f theflankin g sequences from the A.bisporus isolatei s alsolinke dt oth epEM-homologou s sequencei n allisolate so f A.bitorquis (unpub.data) ,suggestin g notonl yconservatio n of genomicorganizatio n for this locus within aspecies ,bu t between species aswell . Theflankin g sequence thati s linked toth epEM-homologou s sequencei n both speciesi spotentiall y functional because itcontain s thegen efo r ATPase subunit 8(Hint ze tal . 1988). Thus,i ti s possible that thepEM-homologou s sequencei sals ofunctiona l becauseo fth ehig h degreeo f conservation ofbot h its sequence andposition .

Asurve y of mtDNAsfro m speciesrepresentin g all Sections within thegenu s Agaricus ispresentl y underway todetermin eth eubiquit y of asequenc e sufficiently similar topE M tob edetectabl e bySouther n hybridization. Sofar , hybridization to pEM hasbee ndetecte d in atleas ton especie si neac h Section. Theintensit yo f hybridization, and thusprobabl y thedegre eo f homology, varies between Sections. Minimal hybridization isdetectabl e tomtDNA sfro m members ofXanthodermati , whereas mtDNAs from members ofHortenses ,Arvenses ,Agaricu s and Spissicaules givequit e strongsignal s (unpub.data) . Atthi spoint ,i t appears thata mtDNAsequenc e homologous toth eRN Apolymeras e ofpE Mexist si n mitochondria from allSection so fAgaricus .suggestin g eithertha t 1) pEM-like Plasmidshav ebee n residents of allAgaricu smitochondri a atsom epoint ,i f plasmids areth e source of thesequence ,o rtha t 2)th esequenc eha s been conserved during theevolutio n of Agaricus mtDNA becausei ti sa nimportan t locuswithi n those mitochondrial genomes. Sequence analysis of A.bisporus mitochondrial fragment TheDN A sequenceo fa pEM-homologou s mtDNA fragment from acommercia l isolateo f A.bisporus thatdoe s notcontai n anyplasmid s wasdetermine d and compared toth ecorrespondin gpE M sequence. Thetw osequence s areove r80 % similar at thenucleotid e level. TheA.bisporu s sequencecorrespond s toth eC - terminal portion of pEM's RNA polymerase,representin g about 60%o f the expected length of thatgene ,an dwoul dcontai n allo fth econserve dregion s thought tob eessentia l totha t typeo f gene. However, several nucleotide changes have introduced frame shifts and stopcodon s sotha tth eA.bisporu s gene apparently couldno tencod ea complet ereadin gframe . ThemtDN A sequences immediately flanking itar eno trelate dt oan ypE M sequencean dd ono tcontai n anyunusua l sequencepatterns ,suc ha sdirec trepeats ,tha t arecharacteristi c of insertion or transposition events (Robison et al. 1991). Thehig hdegre eo f similarity atth e nucleotide level between pEM andth eA.bisporu s sequence suggests aclos e relationship between the twosequences ,bu tth elac ko f plasmids in A.bisporus.a s well asth elimite dexten t of thehomology , suggests thatth esimilarit y mayb emor e aresul to fconvergen tevolutio n ofth esam etyp eo fgenes ,rathe rtha n shared ancestry.

39 A B

21.7—

:Mit

pEM pMPJ

Fig. 1.A .Undigeste d totalDNA sfro m Agaricus bitorquisisolat e Ag4(lan e 2)an d anisolat e of Agaricus bitorquis thatdoe sno tcontai n plasmids (lane 3).Arrow s indicateplasmi d molecules.B .DN Aisolate dfro m intact mitochondria showing association of plasmid molecules with mtDNA.C .Electro n micrograph of native pEM.D.&E .Electro n micrographs ofpE Mi nracquet-for m after reannealingo f previously-denatured termini.Lengt h bari sO.lum .

40 Plasmid pEM DNApolymeras e 3'

3" -< 5' ' » ' RNApolymeras e ' » ' TIR TIR

Fig.2 .Schemati crepresentatio n ofplasmidpE Mshowin gtermina linverte d repeated sequences (TIR),covalently-attache d proteins andth edirectio no f transcription ofreadin g frames. References Hintz,W. , M.Mohan ,J . Anderson &P .Horgen , 1985.Th e mitochondrial DNAs of Agaricus:heterogeneit y inA.bitorqui san dhomogeneit y inA.brunnescens . Curr. Genet. 9:127-132. Hintz, W., J. Anderson &P . Horgen, 1988.Physica l mapping of the mitochondrial genomeo fth ecultivate dmushroo mAgaricu s brunnescens (=A.bjsporus). Curr. Genet. 14:43-49. Meinhardt, F.,F . Kempken, J. Kamper &K . Esser, 1990.Linea r plasmids among eukaryotes: fundamentals andapplication . Curr. Genet. 17:89-95. Meyer, R., W. Hintz, M. Mohan, M. Robison, J. Anderson &P .Horgen , 1988. Homology ofAgaricu smitochondria l plasmidswit hmitochondria l DNA.Genom e 30:710-716. Mohan, M., R. Meyer,J . Anderson &P .Horgen , 1984.Plasmid-lik eDNA s in the commercially important mushroom genus Agaricus. Curr.Genet . 8:615-619. Robison, M., 1988.Mitochondria l plasmid of Agaricus bitorquis. M.Sc. thesis. University ofToronto ,Toronto ,Canada . Robison, M., J. Royer &P .Horgen , 1991. Homology between mitochondrial DNA of Agaricus bisporusan da ninterna l portion of alinea r mitochondrial plasmid of Agaricus bitorquis. Curr.Genet , (inpress) .

41 MITOCHONDRIAL GENOTYPES AND THEIR INHERITANCE IN THE CULTIVATED MUSHROOM AGARJCUS BISPORUS

Anton S.M. Sonnenberg, P.C.C. Van Loon & L.J.L.D. Van Griensven

Mushroom Experimental Station, 5960 AA Horst, The Netherlands

Summary

Twenty nine commercial strains of Agaricus bisporus were examined for their mt genotype. Only three genotypes (I, II and III) were found and the majority of the strains shared one genotype. The smooth-white strains contain mt genotype I whereas the off-white strains contain mt genotype II. The hybrid strains, which are probably all crosses between off-white and white strains, contain either mt genotype I or II. Two older strains, that are not cultivated in Europe anymore, contain mt genotype III. Of all three mt genotypes restriction site maps were constructed. Mt genotype II was almost identical to the map published by Hintz et al (1988). Comparison of the three maps showed variable and conserved regions. In all genotypes inverted repeats were found and in mt genotype II both orientations between the repeats were present. To study mitochondrial inheritance in A. bisporus, mt genotypes were examined in hybrids obtained by mating or by electrofusion of protoplasts. In mating experiments, pairs of homokaryons were chosen in such a way that every mt genotype combination was represented. In repeats of the matings, every repeat resulted in a heterokaryon with the same mt genotype, indicating that inA. bisporus mitochondria are inherited uniparentally. From six heterokaryons obtained after protoplast fusion, only two had the same mt genotype as the heterokaryons obtained after mating. Two other heterokaryons contained mt genotypes not transmitted during mating. The two remaining heterokaryons contained recombinant mt genotypes. Possible mechanisms of mitochondrial inheritance inA. bisporusar e discussed.

Introduction

Comparison of restriction fragment length sizes (RFLPs) of mitochondrial DNA's is increasingly used as an additional tool in systematic and evolutionary biology of fungi (Hoeben & Clark-Walker, 1986; Weber et al, 1986; Smith & Anderson, 1989; Bruns & Palmer, 1989; Carter et al, 1990). The differences in mitochondrial genomes seem generally due to variation in the size of intergeneric regions and the number and sizes of introns that genes carry, whereas their coding capacity is rather constant (Grossmann & Hudspeth, 1985). The range of fungal mitochondrial (mt) genome sizes is large, varying from 18 to 174 kb. The mt DNA of the white button mushroom, A. bisporus,i s one of the largest found in fungi, i.e. 136 kb (Hintz et al, 1988).

42 In breeding ofA. bisporus,littl e or no attention has been paid to the mt genome, so far. As part of our breeding program, we were interrested in the effect of mt genotypes on the vegetative and generative fase of A. bisporus strains. In order to study this effect, one has to know what mt genotypes are present in commercial lines and how mt DNA's are inherited. For this, 29 commercial strains were examined for their mt genotype by comparing restriction fragment length sizes of mt DNA on agarose gels. In addition, mt genotype inheritance was studied in heterokaryons obtained after mating and after electrofusion of homokaryotic protoplasts.

Results and discussion

Characterization of mitochondrial genotypes

The analysis of mt genotypes can be done by isolating mt DNA on CsCl/bisbenzimide gradients, digestion of the mt DNA with endonucleases and comparison of restriction fragment sizes. Alternatively, differences in mt genotypes can be detected in Southern blots of total DNA, digested with endonucleases and hybridized with mt DNA probes. Because both methods are rather laborious, a different strategy was used to screen cultivars for the occurrence of different mt genotypes. Restriction enzymes that recognize 4 base-pairs, cut nuclear DNA more

Fig. 1. Identification of mt genotypes. Total DNA of Horst Ul and U3 was digested with 3 different endonucleases and separated on a 1% agarose gel. Lane 1 and 2; 3 and 4; 5 and 6: total DNA digested with endonucleases Haelll, TaqI and Cfol respectively. Lane 1, 3, 5: Ul; lane 2, 4 and 6: U3; lane 7: molecular weigth markers: Xx EcoRI x Hindlll.

43 frequently than mt DNA. This results in restriction fragments that are on average smaller for nuclear DNA than for mt DNA. When total DNA is digested with these restriction enzymes and separated on a 1% agarose gel, the nuclear DNA fragments are sufficiently separated from the larger mt DNA fragments. In this way, differences in mt genotypes can be detected without isolation of mt DNA or the need for preparation of Southern blots. Figure 1 shows the mt genotypes (mt DNA I and mt DNA II) of the two most common hybrids, Ul and U3. The patterns were obtained by digesting total DNA with the endonuclease Haelll, TaqI or Cfol and separation of the fragments on a 1% agarose gel. The validity of the method was tested by comigration of restriction fragments obtained by digesting CsCl/bisbenzimide gradient purified mt DNA I and mt DNA II.

Table 1. Mitochondrial genotypes in commercial strains ofA. bisporus.

Strain strain-type mt genotype

Horronda (Ul) Hybrid i

Horwitu (U3) ȕ ii

Claron AX30 Jï i

Ciaron AX31 ïî i

Claron AX60 >î ii

Claron A5.1 î> i

Claron A5.3 ïï ii Le Lion XI )» i

Le Lion X20 ïï i

Le Lion X13 îï ii

Somycel 205 îî i

Somycel 208 î) i

Euro-Semy 170 ïl ii

Euro-Semy 280 îî i

Euro-Semy 285 Jï i

Le Champignon 102A >ï i Le Champignon 200 )) i Le Champignon 222 » i

Royal 26A »ï i Le Lion Lj white i Le Lion Bw » i Somycel 53 » i Le Lion B^ off-white ii

Somycel 76 »T ii

Claron A3.6 »t ii Le Lion Q, brown ii

Royal 30A Î5 ii Mori Ha-Q cream m Le Lion Hi »? m

44 In this way, 29 strains were examined for their mt genotype (table 1). Three different mt genotypes were detected. Only two different mt genotypes were found in 19 hybrids tested with a majority of the strains having mt genotype I. The smooth-white and the off-white strains have different mt genotypes indicating that they belong to different groups. This is supported by the fact that they have different karyotypes (Sonnenberg et al, these proceedings). Two traditional strains, that are not cultivated in Europe anymore (bottom of table 1), showed a restriction fragment pattern that was different from mt genotype I and II and will be referred to as mt genotype III. In 1981, the first hybrids were developed by Fritsche (1983). The two hybrids were constructed by mating infertile monospore cultures of smooth-white and off-white strains. In both hybrids a different mt genotype was inherited, i.e. Horst Ul mt genotype I and in Horst U3 mt genotype II. Both hybrids differ in cultivation characteristics and fruiting body morphology. Probably all commercially available hybrids are derived from matings between smooth-white and off-white strains. When the hybrids, listed in table 1, were examined for their cultivation characteristics and fruiting body morphology, a striking correlation was observed between the characteristics of a strain and its mt genotype. This suggests that many hybrids, if not identical to Horst Ul and Horst U3, may be derived from these strains. After all, if all hybrids had been obtained by independent matings between smooth-white and off-white strains, such a correlation would not be observed. The low genetic variability in mt genotypes points to the importance of wild isolates, that contain more genetic variability, for the improvement of commercial strains in future breeding programs.

Fig. 2. Restriction fragments of mt DNA of 3 mt genotypes found in commercial lines of A. bisporus. Lane 1-4: BamHI digest, lane 6-9: EcoRI digest, lane 11-14: PvuII digest. Lane 1, 6 and 11: mt genotype I; lane 2, 7 and 12: mt genotype II; lane 4, 9 and 14: mt genotype III; lane 3, 8 and 13: recombinant mt genotype. Lane 5: X x Hindlll, lane 10: k x EcoRI x Hindlll, lane 15: High molecular weigth markers.

45 To characterize the mt genotypes more precisely, mt DNA of all three genotypes were isolated from CsCl/bisbenzimide gradients. After digestion of the mt DNA's with endonuclease BamHI, EcoRI and PvuII, fragment sizes were compared on agarose gel (fig 2). Although many restriction sites seem to be similar, clearly restriction fragment length polymorphisms could be seen between the three mt genotypes. In order to construct restriction site maps, Southern blots were prepared from gels containing mt DNA fragments obtained after single or double digestion with endonucleases mentioned before. The blots were hybridized with biotinylated BamHI fragments of mt genotype II. These fragments were isolated from gel. Four BamHI probes were a gift of Dr Anderson of the University of Toronto. The restriction site maps produced in this way are not prefect due to the difficulty in isolating all BamHI fragments. However, the maps are appropiate to compare the three mt genotypes. The mt genotype II was almost identical to the physical map of the mt genome constructed by Hintz et al. (1988) whereas the mt genotype I is most likely the same mt genotype that has been found in strain Ag2 by Malloch et al. (1987). The physical maps revealed that all mt DNA's were circular and that the differences between the genotypes are found on the same regions of the map. The latter may indicate that there are variable and conserved regions in mt DNA's of A. bisporus. As has been found in the mt DNA mapped by Hintz et al. (1988), inverted repeats were present in mt genotype II. We have indications that these repeats are also present in the two other mt DNA's. Inverted repeats can be sites of intramolecular recombination. A recombination leads to an inversion of the DNA between the repeats which leads to two populations of mt DNA's (Hudspeth et al, 1983). Although Hintz et al. (1988) published that only one orientation was present in the mt DNA examined, we found indication that in at least mt DNA II both orientations were present. It has been reported that restriction fragments that contain an inverted repeat are sensitive to orientation (Hudspeth et al, 1983). If recombination has occured, a restriction fragment containing the inverted repeat and derived from a mt DNA with one orientation may differ in length from the same restriction fragment derived from the mt DNA with the other orientation. The two largest BamHI fragments both contain one inverted repeat (Hintz et al, 1988). We therefore re-examined the large BamHI fragments on agarose gel under condition that were optimal for separation of large DNA fragments. The analysis revealed that indeed both large BamHI fragments were present as double bands. Each BamHI fragment, containing an inverted repeat, hybridized to all four bands, indicating that both orientation were present. This means that intramolecular recombination is frequent or that there is no selection for either orientation.

Mitochondrial inheritance

To study the inheritance of mt DNA's in A. bisporus, mating experiments were done with homokaryons differing in mating-type and mt genotype (table 2). All crosses showed a fast growing segment arising from the junction zone between the paired homokaryons. The time required for the formation of vigorously growing mycelium varied considerably between the different crosses. The fast growing sectors were subcultured twice and the succes of a mating was confirmed by its RFLP pattern on a Southern blot of total DNA digested with EcoRI. For each

46 cross, at least two nuclear probes were used that show unique RFLP patterns for each homokaryon used. In all crosses only one mt genotype was recovered, indicating that in A. bisporus mitochondria are inherited uniparentally. In crosses that were repeated, each time the same mt DNA was inherited suggesting a mechanism that favours the transmission of one mt genotype.

Table 2. Inheritance of mt genotypes after mating.

mt genotype in parents heterokaryon

2066,,,x 39, I 2004a x 97„ III 2004,,,x PF52/108,, III 39,x 337,, (3) II 39,x 97„ (6) I subscribt indicates mt genotype of parent; numbers between parentheses represent number of replications

In Basidiomycetes studied most closely, the inheritance of mitochondria is determined by nuclear migration. After a compatible mating, nuclear migration occurs through the residient cells of each recipient monokaryon (Baptista-Ferreira et al, 1983). Because mitochondria do not migrate, the mating results in the formation of two heterokaryons differing in mt genotype. In A. bisporus nuclear migration does not occur (Raper & Raper, 1972). In the matings listed in table 2, we have also no indication for nuclear migration. A possible mechanism for the transmission of one mt genotype to the heterokaryon is the unequal distribution of nuclei in cells involved in anastomosis. The mycelium derived from such heterokaryotized cells will always inherit the cytoplasm of the recipient homokaryon thus explaining the inheritance of the same mt genotype in repeated matings. To investigate the possibility of an unequal migration of nuclei as a mechanism for the inheritance of mitochondria, we examined the mt genotypes in heterokaryons obtained after electrofusion of protoplasts. If protoplasts are fused, cytoplasms of two different homokaryons are mixed, thus overruling the mechanism of nuclear migration. Two typical fusion experiments were performed. Protoplasts of homokaryons 39 and 97, and of 39 and 337 were electrofused, as described before (Sonnenberg & Wessels, 1987). In each fusion experiment, three heterokaryons were isolated. The heterokaryotic nature of the fusion products were tested on the basis of RFLP patterns as mentioned before. In addition, all 6 fusion products produced fruiting bodies on compost indicating that these were true heterokaryons. Two of the six heterokaryons obtained after fusion had the same mt genotype as the heterokaryon obtained after mating (table 3). Two other heterokaryotic fusion products contain a mt genotype that was not inherited after mating. The two remaining fusion products contained mt DNA's that were different from both mt genotype I and mt genotype II. The results of the fusion experiments clearly differ from those obtained in mating experiments. Similar observations were

47 made in mating and fusion of Chlamydomonas cells (Matagne, 1981). In this organism, the transmission of chloroplasts is determined by the nuclear genotype. In fusion experiments, however, about one third of the fusion products transmitted chloroplasts of both parents. The rest of the population was equally distributed between fusion products transmitting the chloroplast of one parent or the other exclusively.

Table 3. Mitochondrial genotypes of heterokaryons obtained after fusion of protoplasts.

Parents used in fusion mt genotype of fusion experiments product fusion product

39, x97 „ BIO-A I/IP BIO-B II BIO-D I/IP

39, x 337,, B19-D II B19-E II B19-G I subscribt indicates mt genotype of homokaryotic pro­ toplast; a: recombinant mt genotype

Further investigation showed that in these cases the transmission of chloroplasts was dependent of the input frequency of chloroplasts in the fusion product. The influence of the input frequency on the inheritance of mitochondria has also been observed in yeast (Birky et al, 1978). The inheritance of either one or the other mt genotype after fusion of A. bisporus protoplasts could also be explained by differences in input frequency of a certain mt genotype. The protoplasts of A. bisporus differ considerably in size. This means that after fusion of protoplasts of different sizes, the contribution of each cytoplasm, and therefore the input frequency of of each mt genotype, will differ considerably in each heterologous fusion product. This does, however, not explain the new mt genotype found in two fusion experiments. Examination of restriction fragments showed that the new mt DNA contained fragments that are unique for both mt DNA I and II (fig 2). This may indicate that a recombination has occured between mt DNA I and II. A recombination between different mt DNA's has not been observed before in heterokaryons obtained after mating. In matings there will be little opportunity for recombination of mitochondria because their is no indication for mitochondrial migration. The regeneration of protoplasts of A. bisporus starts after 2 to 4 days (Sonnenberg et al, 1988). During this period there would be sufficient time available for mt DNA's to recombine. Another possibility is that the electric pulses not only induce the fusion of protoplasts but also of mitochondria, thus facilitating the recombination of mt DNA's. The results of the fusion experiment may indicate that an unequal distribution of

48 nuclei plays a role in the inheritance of mitochondria. These experiments are, however, far from conclusive. To further test the hypothesis that an unequal nuclear migration is involved in the inheritance of mitochondria, we constructed two homokaryons that have a new nuclear/mitochondrial combination. From two heterokaryons produced after mating (39 x 97 and 39 x 337; table 2), protoplasts were prepared and after regeneration, colonies were isolated, derived from single regenerated protoplasts (protoclones). About 10% of the régénérants are homokaryotic (Sonnenberg et al, 1988). Because the heterokaryon from which the protoplasts are derived contains only one mt genotype, both homokaryons will have the same mt genotype. In this way, a homokaryon 39 was constructed with mt genotype II and a homokaryon 97 with mt genotype I. With these new homokaryons and the original homokaryons, two mating experiments were done, reciprocal for the mt genotype. If an unequal nuclear distribution plays a role, i.e. homokaryon 97 always donates nuclei to homokaryon 39, one would expect that in the first mating mt genotype I would be inherited and in the second mating mt genotype II. However, in both matings, mt genotype I was inherited in heterokaryons (table 4).

Table 4. Inheritance of mt genotypes in two mating experiments, reciprocal for the mt genotype.

mt genotype in parents heterokaryon

39,x 97„ (6) I

39„x 97, (3) I subscripts indicate mt genotype of parents; numbers between parentheses are number of replicates

This indicates that it is not likely that an unequal migration of nuclei in matings does play a role in the inheritance of mitochondria. The two mating experiments, reciprocal for mt genotype, suggest that somehow mt genotype I is dominant over mt genotype II. This, however, is inconsistent with the mt genotype found in matings between homokaryons 39 and 337 (table 2). In three independent matings between these homokaryons, always mt genotype II was inherited. This leaves the possibility that one nuclear/mt genotype is dominant over an other nuclear/mt genotype. To understand the mechanism by which mitochondria are inherited in matings between homokaryons of A. bisporus, more nuclear/mt combinations have to be tested and, in addition, matings in which other mt genotypes are involved. Without knowing the mechanism by which mitochondria are inherited inA. bisporus, however, it is possible to make nuclear/mitochondrial combinations that are not, or difficult to obtain by matings. For this, the isolation of homkaryotic protoplasts of heterokaryotic mycelium and the fusion of protoplasts can be used. In this way, the effect of mt genotypes on the

49 vegetative and generative fase of A. bisporus can be studied and in particular in those strains commercially grown.

Acknowledgement

We like to thank Dr J.B. Anderson for providing nuclear and mt probes. We are also grateful to mrs Jose in 't Zandt and mrs Jose Kuenen for their technical assistance.

References

Baptista-Ferreira, J.L.C., S. Economou & L.A. Casselton, 1983. Mitochondrial genetics of Coprinus: Recombination of mitochondrial genomes. Curr. Genet. 7:405-407. Birky, C.W., C.A. Demko, P.S. Perlman & R. Strausberg, 1978. Uniparental inheritance of mitochondrial genes in yeasts: dependence on input bias of mitochondrial DNA and preliminary inverstigations of the mechanism. Genetics 89:615-651. Bruns, T.D. & J.D. Palmer, 1989. Evolution of mushroom mitochondrial DNA: Suillius and related genera. J. Molec. Evol. 28:349-362. Carter, D.A., S.A Archer, K.W. Buck, D.S Shaw & R.C. Shattock, 1990. Restriction fragment length polymorphisms of mitochondrial DNA of Phytophthora infestans. Mycol. Res. 94:1123-1128. Fritsche, G., 1983. Breeding Agaricus bisporus at the Mushroom Experimental Station. Mushroom Journal 122:49-53. Grossman, L.T. & M.S. Hudspeth, 1985. Fungal mitochondrial genomes. In: J.W. Bennet & L.L. Lasure (Ed.): Gene manipulations in fungi. Academic Press, London, p.66-103 . Hintz, W.E.A., J.B. Anderson & P.A. Horgen, 1988. Physical mapping of the mitochondrial genome of the cultivated mushroom Agaricus brunnescens (=A. bisporus). Curr. Genet. 14:43-49. Hoeben, P. & G.D. Clark-Walker, 1986. An approach to yeast classification by mapping mitochondrial DNA from Dekkera/Brettanomyces and Eeniella genera. Curr. Gnet. 10:371-379. Hudspeth, M.E.S., D.S. Shumard, C.J.R. Bradford & L.I. Grossman, 1983. Organization olAchfya mtDNA: A population with two orientations and a large inverted repeat containing the rRNA genes. Proc. Natl. Acad. Sei. USA 80:142- 146. Malloch, D., A. Castle & W. Hintz, 1987. Further evidence for Agaricusbrunnescens Peck as the preferred name for the cultivated Agaricus. Mycologia 79:839-846. Matagne, R.F., 1981. Transmission of chloroplast alleles in somatic fusion products obtained from vegetative cells and/or 'Gametes' of Chlamydomonas reinhardi. Curr. Genet. 3:31-36. Raper, C.A. & J.R. Raper, 1972. Genetic analysis of the life cycle of Agaricus bisporus. Mycologia 64:1088-1117. Smith, M.L. & J.B. Anderson, 1989. Restriction fragment length polymorphysms in mitochondrial DNAs of Armillaria: identification of North American biological species. Mycol. Res. 93:247-256. Sonnenberg, A.S.M. & J.G.H. Wessels, 1987. Heterokaryon formation in the

50 basidiomycete Schizophyllum commune by electrofusion of protoplasts. Theor. Appl. Genet. 74:654-658. Sonnenberg, A.S.M., J.G.H. Wessels & L.J.L.D. Van Griensven, 1988. An efficient protoplasting/regeneration system for Agaricus bisporus and Agaricus bitorquis. Curr. Microbiol. 17:285-291. Weber, CA., M.E.S Hudspeth, G.P. Moore & L.I. Grossman, 1986. Analysis of the mitochondrial and nuclear genomes of two Basidiomycetes. Coprinuscinereus an d Coprinus stercorarius. Curr. Genet. 10:515-525 .

51 EFFICIENTPROTOPLAS TFORMATIO NAN DREGENERATIO N AND ELECTROPHORETICKARYOTYP EANALYSI SO FAGARICU S BISPORUS

J. C. Royer, W. E.Hint z and P. A.Horge n Mushroom Research Group,Centr e for Plant Biotechnology University of Toronto,Erindal e Campus,Mississauga ,Ontario , Canada

Summary Anefficien t procedurefo r generation andregeneratio n ofprotoplast s from shake flask cultureso f A.bisporu s wasdeveloped . OrthogonalFiel d Electrophoresis (OFAGE)o f highmolecula rweigh tDN Ai nlyse dprotoplast sresolve dth e genomeo f A.bisporu s intobetwee n 10an d 13 chromosomes,rangin g in sizefro m approximately 1M bt o5 Mb . Theelectrophoreti c karyotypeso ftw ohomokaryon s werehighl y polymorphic. Ahybri do f thetw ohomokaryon scontaine d acombinatio n of thetw o electrophoretic patterns,thoug h therati oo f nucleartype swa sno tbalanced . Anumbe r of geneticallymappe d RFLPan dRAP Dmarkers ,an dth erDN Arepea thav ebee n localized tospecifi c chromosomes. Keywords:Agaricus .protoplast , electrophoretic karyotype,OFAGE ,CHE F Introduction Orthogonalfiel d electrophoresis (OFAGE)ha sbee nutilize dfo r theestimatio no f chromosome numbero f severalfilamentou s fungi andyeast s (Orbach et al., 1988). Electrophoretic karyotype analysis hasfoun d utilityi ngeneti cmappin g studies, analysiso fdisperse drepeated element s (Hamere tal. , 1989)an didentificatio n of minichromosomes (Masele tal. , 1990). Though ametho dfo r determining electrophoretic karyotypes bychemica l treatmento fhypha ewa srecently publishe d (McCluskey et al., 1990), highmolecula r weightDN Afo r chromosome separation gels hastypicall y beenobtaine d from lysedprotoplasts . Previous procedures for protoplast formation andregeneratio n from Agaricus bisporuspublishe d byou r laboratory haveno tbee nefficiën t enough for transformation experiments (Anderson et al., 1984;Castl ee tal. , 1987). Themetho d of Sonnenberg etal. , (1988)resulte di n veryefficien t protoplastregeneration, bu tlo wnumber so fprotoplast s for karyotype analysis. Inthi spaper , wedescrib e aprocedur efo r thegeneratio n of largenumber so f protoplasts from Abisporus . Theelectrophoreti c karyotypes of 2homokaryon s anda hybrido f thetw ohomokaryon s arepresented . In addition, wepresen t Southern hybridization datawhic hlocaliz eRFL Pan dRAP Dmarker sfro m distinct genetic linkage groups (seeKerriga n etal. , thisvolume ) tospecifi c chromosomes. Experimental Organisman dcultur econditions . Strain Ag50i sa derivativ eo fth eU- 3hybri dgenerate d byG .Fritsch e atth eDutc h MushroomExperimen t Station.Strai n Agl-1 (ATCC24662 )i s anauxotrophi c homokaryon generated byRâpe re t al.,(1972 )usin gX-ra y mutagenesis. StrainAg - 89-65i sa homokaryo n derivedfro m awil d isolate byprotoplas t formation and regeneration (Castlee t al., 1988). Strain Ag93bi sa hybri d of Agl-1 and Ag89-65 constructed by Castlee t al.(1988) . Cultures weremaintaine d oncomplet e yeast medium (CYM,Râpe re t al., 1972)aga rplates .

52 Alternatingfield electrophoresi s Protoplast plugpreparatio n andlysi swer e asdescribe d byOrbac h et al. (1988), excepttha tEDT Awa somitte dfrom th eprotoplas t buffer andth elo wmeltin g temperature agarosepreparation . Mycelialplug swer eprepare d according to McCluskey etal .(1990) . ChromosomalD NA swer e separated on aCHE F apparatus (BioRad). Separations wereperforme d at 14°C ,wit h0. 5 xTB E (Maniatis et al.,1982) asth erunnin g buffer. Gels wereelectrophorese d at5 0volt swit h aswitc h rampo f 5- 3 0min .fo r 10days . Buffer waschange d at4 da y intervals. Gels were stainedwit hethidiu m bromidean dvisualize do n aU V transilluminator. DNAtransfe r and hybridization. Gelswer eirradiate do n aU V transilluminator for 5minute s and treatedwit h0.1 2N HClfo r 30minute sprio r toDN Atransfe r toGeneScree n Plus according toth e manufacturer's instructions. Hybridizations wereperforme d at65 °C a sdescribe d in Maniatis et al. (1982). RFLP loci wereisolate d byCastl e et al. (1987), whileRAP D products (Williamse t al., 1991)wer e generated according toKerriga n etal .(1991) . Ribosomal DNA was localized usinga prob efro m Hintze tal . (1989),whic h contains theribosoma l repeato f A.bitorquis . Results and discussion Protoplast formation andregeneratio n Themos tcritica lparamete r for thereleas eo flarg eamount so fprotoplast s isth e generation of large amountso f actively growingmycelium . Wehav ebee nabl et o achieve this through aninoculu m build-upprocedure . Aprimar y cultureo f the fungus isgenerate d byth einoculatio n of 100m lo f CYMwit ha homogenize d agarplu g (ca.2 cm)fro m aCY M agarplate . After 2-4week so f shaking (100RPM ) at25 ° C,th e primary culture ishomogenize d (2, 10 second,hig h speed bursts ona blende r witha 10secon dinterval ) and2 0m laliquot s ofth ehomogenize dcultur ear euse d toinoculat e a second seto f 100m lCY Mcultures . After 2-4weeks ,a secondar ycultur ei s homogenized, and2 0m laliquot s areuse dt oinoculat e tertiarycultures . Weroutinel y use2- 4da yol dtertiar yo rquaternar y shakeflas k cultures for optimal protoplast formation.

A second criticalparamete r inprotoplas tformatio n isth eprotoplas t buffer. Weteste d anumbe ro f buffers atth e sameosmoti c strength,an dfoun d thathighes t levelso f protoplasts were achieved usingMgS04 . Theseresult scontraste d with thoseo f Sonnenburg et al.(1988 ) whoobtaine d higherlevel s with sucrose. Inou rhands ,th e highest levelso fprotoplast s were generated with0. 8 MMgS0 4i n thepresenc e of2 0 mMsodiu mcitrate . (A0. 5 Mstoc k of sodiumcitrat ewit hth ep Hadjuste d to5. 5i s utilized. Thep Ho f theMgS04-sodiu m citrate solution isapproximatel y4.0 . Increasingth ep Ho f thesolutio n above4. 0ha da negativ eeffec t on protoplastyield) . Wehav eutilize dNovoZy m 234a t 10mg/m la sth edigestio n buffer inal lo fou r experiments. Ammendmanto f NovoZym23 4wit h purified chitinase hadn oeffec t on protoplast yield. In addition,pretreatmen t of myceliumwit h2-mercaptoethanol , which enhancesprotoplas t yield from anumbe ro ffung i (Wange t al., 1988;Roye re t al., 1991) hadn oeffec t on protoplast yield. Ouroptimize d procedurei sa sfollows : Mycelium from a2- 3da y tertiary shake flask culture isdispense d into2,5 0 mlpolypropylen e centrifuge tubes,an dcentrifuge d at high speed(settin g7 )fo r 10 mini n aclinica l centrifuge (IEC). Mycelium iswashe di n

53 protoplast buffer (0.8 MMgS04 , 0.02 M sodiumcitrate ,final p H4.0 ) and recentrifuged. Twenty mlo f NovZym(1 0mg/m li nprotoplas t buffer) is centrifuged (setting7 , 10min) ,filtere d through a0.2 2\i filte r andadde dt oeac htube . The mixturei sincubate d withgentl e shakingfo r 40mi na troo mtemperature . The resultingprotoplas t solution ispasse dthroug h several layers of cheesecloth, and finally a 10| ifilter an dcentrifuge d (setting4,1 0min. ) Eachpelle ti swashe dwit h3 0 mlo f STC(1. 0M sorbitol,0.0 2M Tris ,p H7.5,0.0 5 MCaCl2 ) and recentrifuged (setting 3, 10min. ) before counting on ahemocytometer . Using thisprocedure ,w e canroutinel y generatebetwee n 5x lu** and2 x 10^protoplast s from a 100m ltertiar y culture (0.3g dr y weight)o fth efas t growinghomokaryo n Agl-1. Approximately 10** protoplasts aretypicall y achievedfro m theothe rstrain sw ehav e examined. Protoplastregeneratio n Weteste da numbe ro fdifferen t osmoticstabilizer sfo r theireffec t onprotoplas t regeneration. Thehighes tlevel so fregeneratio n (between2 and5 %) wer eachieve d onsucros e(0. 6M) . Wehav eobserve dtha tprotoplas tregeneratio n isinhibite db y highprotoplas tconcentrations . Toovercom etins, w eroutinel ydilut eprotoplast st oa concentration of 10*>pe rm li nliqui dCY Mplu s0. 6M sucros e(CYMS) . The regeneratedprotoplast sar ethe nconcentrate dan dapplie dt oselectiv emédiu ma s describedbelow .W ehav eobtaine dsomewha thighe rregeneratio n frequencies when compostextrac t (Sonnenberge tal. , 1988)i sincorporate di n theregeneratio nmedium . However, thecompos t extractappear s tointerac twit hman yo f theantibiotic s usedi n transformation experiments. Thisresult si nth erequiremen tfo rmuc hhighe rantibioti c concentrations toinhibi t growth. Ourregeneratio n frequencies arelowe rtha nthos e obtained for A.bisporu s by Sonnenberge ta L(1988 ) andconsiderabl y lowertha n thosei n ascomycete systems,bu t shouldb eadequat e fortransformatio n (Royere taL , thisvolume) .

Alternatingfield electrophoresis . Themetho do f (McCluskey et al., 1990)di dno treleas esufficien t quantitieso fintac t A.bisporu s chromosomalDN A toallo wit sapplicatio n inCHE Fanalysi s (datano t shown). Therefore, lysedprotoplast s serveda sth e sourceo f highmolecula r weight DNAi n allchromosom e separation gels. Ininitia lexperiments ,protoplast s were washed in sorbitol (1.0M) ,Tri s (20mM ,p H7.5 ) andEDT A (50mM ) (Orbach et al., 1988). Weobserve d thatprotoplast s lysedi nth epresenc eo f 50m MEDTA . Asa result,EDT A wasomitte dfro m theprotoplas tbuffer s andagaros epreparations . Theelectrophoreti ckaryotype s ofth e2 homokaryon san dth ehybri d are showni n Fig 1. Tenband s within therang eo f 1 to5 M bwer eobserve d onethidiu m bromide stainedgels . Ahighe rleve lo fintensit y atthre epostion si n Ag 1-1 wasconsistentl y observed suggesting thepresenc eo fmultipl eband s atthes elocations . Anumbe ro f sizepolymorphism s existed between thekaryotype s of the2 homokaryon s (Fig. 1). Asexpected ,th ehybri dphenotyp e appearedt ob ea composit eo fth ekaryotypi c patterns of thetw ohomokaryons . However,ethidiu m bromidestainin gintensit y suggested that theA g 1-1 pattern predominated in thehybrid . (Fig.1) . SegregatingRFL Pan dRAP D (Williamse tal. , 1991)sequence swhic h hadbee n assigned to 1 of 9linkag e groups byKerriga n et al., (1991),a swel l asa plasmi d containing therDN Arepea twer euse da shybidizatio n probest ocorrelat e the established linkagema pwit hth ephysica lmap . Mosto fth eRAP Dproduct s hybridized tomultipl echromosoma l bands. Atleas ton eo fth eavailabl eprobe s hybridized toal lo f theresolvabl echromosome s exceptchromosom eXi n of Ag 1-1. Inthre ecases ,tw oprobe s which appearedt ohybridiz et oth e sameban di n theA g 1-1 separation hybridizedt o2 distinc tband si nA g89-65 . Thehybridizatio n signalsi nth e

54 Ag 1-1 Separation corresponded toth eregions o fhig hintensit y in theethidiu m bromide stainedA g 1-1 separation. Weinterpre t theseresults a sindicatin gdoublets , andhav e assigned twochromosome s toeac ho f thesepostion s (Fig.1 ) Wehav eresolve d the genomeo f A.bisporu s intoa tleas t 10chromosomes .Base d onth eintensit y ofethidiu m bromidestaine d gels,an dth ehybridizatio n dataw e estimate that A.bisporu s contains 13 chromosomes.Thi s numberi shighe rtha na previouschromosom e estimateo f9 determine d bycytologica l staining (Saksenae tal. , 1976),an di sals ogreate r than thenumbe ro f linkage groups determined thus far through analysis ofmeioti c segregation in theA g93 bcros s (Kerrigan et al., 1991). Thehaploi dgenom e size,a sestimate db y summingth emolecula rweight s ofth e individualchromosome s is 34Mb .Thi si si nver yclos eagreemen t with anestimat e determinedpreviousl y byreassociatio n kinetics (Arthure t al., 1982).

1 2 3 4

5.7M b 4.6M b 3.5 Mb

vm,ix 2.2M b

1.6 Mb

1.1Mb

Fig. 1.Electrophoreti c karyotype of A.bisporus . Lane 1,Agl-1 ; lane2 , Ag89-65; lane 3,Ag93b ;lan e4 ,Saccharomvce scerevisea e andSchizosaccharomvce s pomhe.

55 References: Anderson, J. B.,D . M. Petsche, F. B.Herr , &P . A.Horgen , 1984. Breeding relationships among several specieso f Agaricus. Can.J .Bot .62:1884-1889 . Arthur, R., F. Herr, N. Strauss, J. B. Anderson &P . A. Horgen, 1982. Characterization ofth egenom eo fth ecultivate d mushroom,Agaricu s brunnescens. Exper. Mycol 7:127-132. Castle,A .J. , P. A.Horge n &J . B.Anderson , 1987. Restriction fragment length polymorphisms in themushroom s Agaricusbrunnescen s andAgaricu sbitorquis . Appl.Environ .Microbiol .53:816-822 . Castle,A .J. , P. A.Horge n &J . B.Anderson , 1988. Crosses among homokaryons from commercial andwild-collecte d strainso fth emushroo m Agaricus brunnescens (=A . bisporus'). Appl.Environ . Microbiol. 54:1643-1648. Hamer, J. E., L. Farrall, M. J. Orbach, B. Valent, &F . G. Chumley, 1989. Host species-specific conservation of afamil y ofrepeate dDN A sequencesi nth e genomeo f afunga l plantpathogen . Proc.Natl .Acad . Sei. USA 86:9981-9985. Hintz, W. E., J. B.Anderso n &P . A.Horgen , 1989. Relatedness of three species of Agaricus inferred from restriction fragment lengthpolymorphis m analysiso f the ribosomal DNArepea t andmitochondria l DNA. Genome32:173-178 . Maniatis,T. ,E.F .Fritsc h andJ . Sambrook, 1982. Molecular cloning: a laboratory manual. Cold SpringHarbo rLaboratory , ColdSprin g Harbo, New York. Masel, A., K. Braithwaite,J . Irwin, andJ . Manners, 1990. Highly variable molecularkaryotype s inth eplan tpathoge nColletotrichu m gloeosporioides. Curr Genet. 18:81-86. McCluskey, K., B.W . Russell andD. Mills, 1990. Electrophoretic karyotyping without thenee dfo r generatingprotoplasts . Curr.Genet . 18:385-386. Orbach, M.J. , D.Vollrath , R. W.Davi s &C .Yanofsky , 1988. An electrophoretic karyotypeo f Neurospora crassa. Mol.Cel lBiol . 8:1469-1473. Raper, C.A. ,J . R.Rape r &R .E .Miller , 1972. Genetic analysis of the life cycleo f Agaricus bisporus. Mycologia 64:1088-1117. Royer, J. C, &P . A.Horgen , 1991. Towards a transformation system for Agaricus bisporus (Thisvolume) . Royer, J. C, K. Dewar, M. A.Hubbe s &P . A.Horgen , 1991. Analysis of a high frequency transformation systemfo r Ophiostomaulmi .th ecausa l agento fDutc h elm disease. Mol.Gen .Genet . 225:168-176. Saksena, K. N., R. Marino, M. N.Halle r &P . A.Lemke , 1976. Study on development of Agaricus bisporus byfluorescen t microscopy and scanning electron microscopy. J. Bacteriol. 126:417-428. Sonnenberg, A. S., J. G. Wessels &L . J. van Griensven, 1988. An efficient protoplasting/regeneration system for Agaricusbisporu s andAgaricu s bitorquis. Curr. Microbiol. 17:285-291. Wang,J. , D.H .Holde n &S . A.Leong , 1988. Gene transfer system for the phytopathogenic fungus Ustilago maydis. Proc.Natl .Acad . Sei. USA 85:865-869. Williams, J. G. K., A. R. Kubelik, K. J. Livak, J. A. Rafalski, & S. V. Tingy, 1990. DNApolymorphism s amplified by arbitrary primers areusefu l asgeneti c markers. Nucleic Acids Res.18:6531-6535 .

56 CHROMOSOME SEPARATION AND ASSIGNMENT OF DNA PROBES IN AGARICUS BISPORUS

Anton S.M. Sonnenberg, K. den Hollander, A.P.J, van de Munckhof & L.J.L.D. Van Griensven

Mushroom Experimental Station, 5960 AA, Horst, The Netherlands

Summary

An efficient protoplasting system was used for karyotyping strains of A. bisporus. Using two-step cultures, routinely 2 x 10s to 1 x 10' protoplasts were isolated from homokaryotic and heterokaryotic strains after a short period of in vitro cultivation. Pulsed field electrophoresis revealed the presence of at least 11 chromosomes in one of the haploid genomes examined. Four different karyotypes were detected: 3 homokaryons most common in hybrids (Horst ®U1 and Horst ®U3) and a homokaryon isolated from a strain collected in nature. Several RFLP probes were localized on different chromosomes. In addition, RAPD markers, excised from gel, purified and reamplified, were used as probes. One of these markers hybridized with a chromosome of one strain only and not to chromosomes of other strains. Unique probes like these, can be used to identify chromosomes in crosses and their offspring. A GPD gen, isolated from strain Horst ®U3 (Harmsen, these proceedings), was localized on the smallest chromosome.

Introduction

The estimation of chromosome numbers in Agaricus bisporus was done cytologically by Jiri (1967) and Evans (1959). They reported chromosome numbers of 8 and 12 respectively. Because chromosomes of fungi are relatively small, it is difficult to visualize all chromosomes present in fungal cells. With the development of pulse-field electrophoresis, it is now possible to separate DNA molecules of up to 10 megabase pairs (Mbp). Especially the 'clamped homogeneous electric field' (CHEF) electrophoresis technique has been shown to be very useful for the separation of large chromosomes in agarose gels (Chu et al, 1986, Orbach et al, 1988, Brody & Carbon, 1989). With the Separation of intact chromosomes, it is now possible to karyotype strains by comparing chromosome numbers and their electrophoretic mobility. In addition, Southern blots of chromosomes separated on agarose gels are very usefull to correlate linkage groups with the physical structure of the fungal genome. By hybridization with different DNA probes, sets of genetic markers can be assigned to each chromosome which facilitate the construction of a genetic map. These maps can be used to find linkages between genetic markers and phenotypic traits and can be used to improve breeding programs.

57 In this paper we report preliminary results of the separation of chromosomes of homo- and heterokaryotic strains and the assignment of DNA probes inA. bisporus.

Results and discussion

A prerequisite for the separation of chromosomes is the isolation of intact DNA. This can only be done by purifying chromosomes from protoplasts. We will therefore briefly outline the isolation of protoplasts, modified after a protoplasting sytem published before (Sonnenberg et al, 1988).

Cultivation of strains and isolation of protoplasts

Stockcultures, maintained at 4 °C, are used to inoculate 10 to 14 Petri dishes, containing MMP agar medium (1% malt extract, 0.5% mycological pepton, 1% agar, pH 7.0). The agar is covered with a cellophane sheet. After 10 to 14 days incubation at 24 °C, the mycelium is removed and fragmented in a Waring blendor in DT80 medium (Sonnenberg et al, 1988). The suspension is used to inoculate 6 Fernbach flasks, each containing 150 - 200 ml DT80 medium. The flasks are incubated stationary, 4 to 5 days at 24 °C. The mycelium is collected by filtration over a nylon cloth (300 /im mesh) and rinsed with sterile destilled water. The mycelium is then resuspended in 0.6 M sucrose, 10 mM KMOPS (3[N- morpholino]propanesulfonic acid), pH 6.0 containing 1 to 2 mg/ml laboratory prepared cell wall lytic enzymes (TE). After 3 to 4 h incubation at 24 °C, the suspension is filtered over 500 and 150 urn mesh nylon cloth and glaswool succesively. The protoplasts are then pelleted by centrifugation and washed twice in 0.6 M sucrose. The yield is 2 x 108t o 1x 10'protoplasts . The protoplasts are diluted with melted Insert agarose (FMC) to a final concentration of 4 x 10s protoplasts/ml.

The cell wall lytic enzymes are isolated from the culture filtrate of Trichoderma harzianum, grown on a medium containing cell walls isolated from A. bisporus fruiting bodies (Sonnenberg et al, 1988). Although the use of a laboratory prepared enzym mixture is rather laborious, we do not use commercial enzymes such as Novozyme 234. In our hands, TE is more effective in the production of protoplasts and can be used in a much lower concentration than the commercially available enzymes. Moreover, the activity of each batch of TE is very constant, contrary to different batches of commercial enzymes.

Pulsed field electrophoresis

The embedded protoplasts were incubated overnight in 1% proteinase K, at 50°C , washed twice in 50 mM EDTA and equilibrated with the electrophoresis buffer (0.5 x TBE). Samples were loaded on 1% agarose gels and chromosomes were separated by clamped homogeneous electric field electrophoresis (CHEF II, BioRad). The electric parameters were 50 V, 10 to 30 min pulses during 9 to 11 days and 30 to 50 min pulses during 2 to 4 days succesively. Figure 1 shows the banding pattern of a heterokaryon and the constituent homokaryons. In this gel and a previous gel 4 homokaryons were examined. All 4 showed a different karyotype. Two homokaryons were derived from a off-white strain (97 and 337) and one from a white strain (39).

58 The fourth homokaryon was obtained from a wild isolate collected in the Netherlands (PF100).

Fig. 1. Chromosome separation ofA. bisporusb y CHEF. Lane 1: S. cerevisiae;lan e 2 and 3: A. bisporus homokaryons 97 and PF100; lane 4: heterokaryon (97 x PF100); lane 5: S. pombe.

The chromosome banding pattern of homokaryon 97 in fig 1 showed at least 11 different bands. Two bands had a higher intensity after ethidium bromide staining than the other bands, indicating the presence of different chromosomes with similar length. This means that in this homokaryon a minimum of 11 and a maximum of 13 chromosomes are present. Using chromosomes of Saccharomycescerevisiae and Schizosaccharomycespombe as size markers, chromosome sizes were estimated of the homokaryons shown in fig 1. The estimated total genome size based on CHEF analysis is 31 Mb (table 1), which agrees with the 34 Mb estimated by Arthur et al. (1982), obtained using DNA:DNA reassociation analysis. In order to assign DNA probes to chromosomes, Southern blots were hybridized with cloned EcoRI fragments, used in RFLP analysis. Four probes were assigned to 2 chromosomes so far. We were also interested in the use of random amplified DNA segments (RAPD's) as markers for the construction of genetic maps. These markers are useful if they represent one locus on the fungal genome. To test this, single bands of amplified DNA were excised from gel, purified and reamplified. The DNA's were biotinylated and hybridized to Southern blots. These blots contained chromosomes of three homokaryons with different karyotypes (39, 97 and PF100) and two heterokaryons derived from matings (39 x 97 and 97 x PF100). Different hybridization patterns were observed. Some probes hybridized to all chromosomes, whereas other probes hybridized to 4 or 5 chromosomes. These probes may represent DNA segments that are present as repetitive sequences throughout the entire genome. An other probe hybridized to only one chromosome of all karyotypes. These probes may be used for the construction of genetic maps. One probe, excised from gel as a band with a unique length for homokaryon 97, hybridized to one chromosome of homokaryon 97 only and not to chromosomes of other homokaryons. Unique probes like these may be used to identify chromosomes in putative fusion products, in matings and their off-spring.

59 Table 1. Estimation of chromosome sizes of 2 homokaryons of A bisporus.

97a PF100"

3.4" 3.4 3.2 3.1 3.1 (2) 3.1 (2?) 2.7 2.7 2.4 (2) 2.5 2.3 2.3 (2) 2.0 1.9 1.9 1.8 1.7 1.7 1.5 1.6 1.4 1.4

Tot. 31 Mb 28 (31?) Mb a: strain designation; b: length of chromosomes in Mega-basepairs (Mb); numbers between parentheses indicate chromosomes with similar length

Finally, a glyceraldehyde 3-phosphate dehydrogenase (GPD) gen, isolated from strain Horst U3 by M.C. Harmsen (University of Groningen), was localized on the smallest chromosome. Further attempts will be made to resolve all chromosomes present in homokaryons ofA. bisporusint o single bands on CHEF gels. Southern blots of these gels can than be used to assign sets of DNA markers to each chromosome. These markers will be used to construct a genetic map. For this, we have already isolated a set of 100 homokaryotic monosporecultures of a genetically defined heterokaryon.

Acknowledgement

We like to thank Dr J.B. Anderson for providing nuclear probes and Dr H.C. Harmsen for providing the GPD gen. We are also grateful to Mrs José in 't Zandt and Mrs José Kuenen for their excellent technical assistance.

References

Arthur, R., F. Herr, N. Straus, J. Anderson & P. Horgan, 1982. Characterization of the genome of the cultivated mushroom, Agaricus brunnescens. Exp. Mycol. 7:127-132. Brody, H & J. Carbon, 1989. Electrophoretic karyotype ofAspergillus nidulans. Proc. Natl. Acad. Sei. USA 86:6260-6263.

60 Chu, G, D. Vollrath & R.W. Davis, 1986. Separation of large DNA molecules by contour-clamped homogeneous electric fields. Science 234:1582-1585. Evans, H.J., 1959. Nuclear behaviour in the cultivated mushroom. Chromosoma 10:115-135. Jiri, H., 1967. Zytologische Studien über die Gattung Agaricus. Mushroom Science 6:77-81. Orbach, M.J., D. Vollrath, R.W. Davis & C. Yanofsky, 1988. An electrophoretic karyotype of Neurospora crassa.Mol . Cell. Biol. 8:1469-1473. Sonnenberg, A.S.M., J.G.H. Wessels & L.J.L.D. Van Griensven, 1988. An efficient protoplasting/regeneration system for Agaricus bisporus and Agaricus bitorquis. Curr. Microbiol. 17:285-291.

61 THEUS EO FPROTOPLAS TPRODUCTION ,PROTOPLAS T REGENERATION ANDRESTRICTIO N FRAGMENTLENGT HPOLYMORPHISM S INDEVELOPIN G ASYSTEMATI CAN DHIGHL Y REPRODUCIBLE BREEDING STRATEGYFO R AGARICUS BISPORUS

P. A.Horgen , T. Jin and J. B. Anderson Mushroom Research Group,Centr e forPlan t Biotechnology University ofToronto , Erindale Campus,Mississauga , Ontario Canada L5L 1C6

Summary Thebispori clif e cycleo f Agaricus bisporuspresent s seriousnatura l barriers toth e development of asimple ,systemati cbreedin g strategy. Isolation ofparenta l homokaryons from afertil e heterokaryon hasbee nproblemati c andth everificatio n of ahomokaryoti c cultureb ycolon y morphology canb ehighl y subjective. Wehav eadapte d simple, unambiguous approaches for theisolatio n of parentalhomokaryon s from afertil e heterokaryotic culture. Restriction fragment lengthpolymorphism s (RFLPs) havebee n successful indistinguishin g thetw ouniqu enuclea r typeso f afertil e heterokaryon. Many genetic locidetecte d byth eRFL Pprobe s areheteroalleli c inan ygive n heterokaryon. Whenmyceliu mfro m afertil e heterokaryon isdigeste dwit h acell-wal l hydrolyzing enzyme,protoplast s arereleased . If thoseprotoplast s areallowe d toregenerate (fo r most strainsexamined) ,approximatel y 10%o f theregenerated culture s arehomokaryotic . This hasbee n verified byth eus eo f specific RFLPprobes . With thismethodology , parental homokaryons can beisolate dfro m selectedcommercia l strainsan dfro m selectedwild - collected isolateso f A.bisporus . Compatible homokaryons can bemate d andth ene w intraspecies hybridsca nb everifie d again byusin gselecte d heteroallelic RFLPprobes . These new strains thenca nb efruite d on asmal l scalei npolystyren ecups . Wehav e utilizedthi s selectivebreedin g strategy tofollo w both theinheritanc eo f nuclearRLF P geneticmarker s andmor erecently ,mitochondria l genetic markers. Homokaryons carrying different mitochondrial genotypes wereuse di nmatin gexperiment s togai na n understanding ofcytoplasmi c inheritancei nth ecommercia l mushroom. Abrie fovervie w of ourmitochondria l inheritance study ispresented . Keywords: homokaryon,hybrids , RFLPrecombinant, DNA ,mitochondria linheritance . Introduction Despiteth ecommercia l valueo f Agaricus bisporus £=A.brunnescens V incompariso n to othervegetabl ecrop san dt oothe rcommerciall y importantfungi , mushroom scientists haveno tbee nabl et osystematicall y manipulate thebutto n mushroom ina classi c "crops" breedingprogram . Themajo r successesi nmushroo m breedinghav e beenfe w andthe y haverequired a nenormou s amounto feffort . Oneo fth emajo rreasons fo r these difficulties inestablishin g areproducibl e andsystemati c breedingprogra m for mushrooms isth esecondaril y homothallic natureo f theA ^bisporu s life cycle. Figure1 showsa simplistic diagramo fth elif e historyo f A.bisporu s anda comparativ e life cyclediagra m for othermushroom-producin g basidiomycetes. Themos tdramati cdifferenc e between A» bisporus andothe ragaric si stha tther ei sa n absenceo f uninucleatepropagule s in the life cycleo f thebutto n mushroom. Themajorit y of A,, bisporus basidiospores contain two haploidnucle iwherea sth e basidiospores ofothe rmushroom s areinitiall y uninucleate (Figure 1). Furthermore,i n A.bisporus .ther e appears tob ea mechanis m which ensures thatth ebinucleat e basidiosporereceive s ahaploi dnucleu sfo reac ho fth eparenta l homokaryons (Râpere t al., 1972;Eliot ,1985) . Thismean stha tth evas tmajorit y of these basidiospores aresel f fertile andwil lgerminat e tofor m aheterokaryo n whichwil l

62 binucleate basidiospores Button Mushroom

fruiting spore germination body

fertile mycelium (heterokaryon)

B

uninucleate basidiospores «d£l*P^ otner Mushrooms TT fruiting bod spore germination U y monokaryon fusion

C» c5T4-o • l • o l>To fertile mycelium (dikaryon) V

Figure 1 -Diagramati crepresentation o f mushroom life cycle.(A).Th euppe rdiagra m illustrates thesecondaril y homothallic life historyo fth ebutto n mushroom,Agaricu s bisporus. (B).Thelowe rdiagra millustrate s atypica l4 sporedheterothalli c life history which isrepresentativ e of A.bitorqui s andman y (butno tall )othe rgille d basidiomycetes.

63 completeth elif e cycle. Whilether ei sa smal lproportio n of thebasidiospore s which are homokaryotic,th eisolatio no f homokaryotic basidiospores is anextremel y laborious task. Thetraditiona l methodo f homokaryon verification, theinabilit yt ofruit , wasextremel y timeconsuming . Othermethod s thoughtt overif y thehomokaryoti c statear eno t reliable.

A 12345678 9 10 1112 1314 15161718192 0212 2 232425

m-m -«räfi*•<-..• - •'**• -»

B 1 2 3 4 5 6 7 8 9 101112131 4 15 1617 181 9202122232 4 25

à».*l£*.4^K£i*' i»

4**» »!

Figure 2- A nillustratio n of restriction fragment length polymorphisms (RLFPs)i n Agaricus bisporus. (A).Commercia l isolates lanes 2- 25 . (B).Wild-collecte d isolates lanes 2- 25 . DNAs wereisolated , digested with EcoRI,separate d on agarose gels southern blotted andhybridize d toth erecombinan t plasmidpAg3 3N1 0 (Castlee tat. , 1987).

Experimental Thepotentia l sources of A.bisporu s breeding stock: Most mushroom scientists believe thatth emoder n cultivation of A.bisporu s hadit s origins inFranc e about 300year sag o(Rinker , 1989), Thecommercia l strainsi n cultivation today aremos tlikel yderivativ eo fth einitia lisolates . Mosto f thedat a published todat esuggest s thatther ei slittl egeneti cdiversit yi ncommercia l isolates ofth e button mushroom (Royce andMay , 1982;Castl ee t al., 1987;Hint ze t al., 1985;Kerriga n and Ross, 1989;Kerriga n et al.,1991) . Our group has beeninstrumenta l tobegi n using restriction fragment length polymorphisms (RFLPS)o fmushroo m DNA as useful indicators of geneticdiversit y (orsimilarity) . Thismethodolog y whichcombine sth eus e ofrecombinan tDN Atechnique s withDNA/DN Ahybridizatio n hasbee nestablishe d asa 64 powerful new tool in modern genetics (Patterson et.al. , 1989). Wear e nowcapabl eo f generating "DNA fingerprints" of individual mushroom strains (homokaryons aswel la s heterokaryons) andthes e toolsoffe r tremendous technical advancement tomushroo m science. Selectedrecombinan t clonesca n beuse da sprobe s tocompar e anumbe ro f mushroom isolates. Figure 2illustrate s howRFLP s can beuse d to differentiate individuals and showstha t whilether e isa lac ko f geneticdiversit y incommercia l isolates (Figure 2A), there ismuc h morediversit y in wild-collected populations of A^ bisporus. Results from ourgrou pindicate s thatthes eobservatio n hold true for bothnuclea rDN A (Castle et al.1987 ; 1988)a swel l asmitochondria l DNA:(Figur e 6: Hintze l al, 1985; Hintz et. al., 1989).I f mushroom scientistsrestric t tobreedin g only with strains that arei n cultivation, they areworkin gwit h anextremel y narrow geneticpool . An important question that might beaddresse d iswhethe r iti spossibl et ofin d logical and systematic approachest obree ddiversit y inwild-collecte d strainswit hcommerciall y selected advantageous traits. Can abreedin g strategy bedevelope d for thebutto n mushroomwhic h utilizes toolso f themoder n biologist tobrea kdow n existing natural barriersi n mushrooms?

Resultsan dDiscussio n Abreedin g Approach for the1990s : In anexaminatio n of thedat a from ourRFL Pstudie s onfertil e heterokaryons,i tbecam e apparent thatfo r manyo fou r specific clonedDN Aprobes ,w eofte n observed multiple hybridization signals which appeared asmor etha non e bandi nou rDNA/DN A hybridizations with nuclear DNA (Figure 3,Castle ,e l al.,1987 ;Castl e et al,1988 ; Summerbell et al.,1989) . Further analysesindicate d thatthes emultipl e hybridization signalsrepresente d theheterozygou s condition ofth eheterokaryoti c (binucleate) mycelium thateventuall y produced thefruitin g bodies (Castlee t al.,1988 ;Summerbel l et al.,1989) . This phenomenon can beillustrate d in Figure 2 andi nFigur e 3B, Lane 1 (Figure3B ) represents theheteroalleli c condition of thefertil e heterokaryon for this specific recombinant DNA probe,wherea s laneF (Figur e 3B) showsonl y twoband s aswel la s lane E(Figure3 B )show sonl y twobands . These doubletsrepresen t thehomoalleli c condition for this specific RFLPgeneti c marker. These kindso f observations werever y common for anumbe ro f ourclone d RFLPprobe s (Castle et al., 1987:Castl ee t al., 1988; Summerbell et al., 1989). An important development from theseobservatio n wastha tw e can nowroutinel y use this methodology toverif y thehomokaryoti c condition of aculture .

The somatic generation of homokaryons Cell walldegradin g enzymes whichremov e thecel l wallfro m thehypha eo f fungi result in theproductio n of wall-less cellscalle dprotoplasts . Protoplast production andprotoplas t regeneration has been aver y useful tooli nth ebiotechnologica l manipulations of higher plants andfung i (Peberdy, 1989). Protoplast production andregeneratio n hasbee n developed andoptimumize d for the button mushroom (Anderson et al.,1984 ;Sonneber g et al., 1988;Roye r andHorgen , thisvolume) . Inou rlaboratorie s weroutinel y can generate 109protoplast s from 0.3 gdr yweigh to f hyphae andhav e observed regeneration frequencies from 2-10% (Royer andHorgen , this volume). Threepossibl e typeso f protoplasts can begenerate dfro m abinucleate ,heterokaryoti c mycelum (Figure 3A): (1) The mostfrequen t typeo fregenerat e willhav ebot hnuclea rtype san dwil l produce mycelium exactly thesam ea sth estartin gmaterial , (2) someo f theprotoplast s produced will lack nuclei orothe rke ycellula r organelles,thes e protoplasts will notregenerate , (3)a small proportion of theprotoplast s willhav eonl y 1 nucleartyp e andwil lb ehomokaryotic . In amushroo m breedingprogram , thetyp e (3)protoplas t would be themos t important type toverif y andrecover .

65 A ^ C o • oo • o • r^V (.°)(1) w®o<2 ) ® ®o(3) ( ®

Figure 3- Protoplas tproductio n andregeneration. (A) .Diagramati can ddeterminatio no f thehomokaryoti c staterepresentatio n ofprotoplas tproduction : (1)represent s aprotoplas t that isheterokaryoti c containingbot hnuclea r types. (2)represent s aprotoplas t lackingan y nuclei (willno tregenerate) . (3)represent s aprotoplas tcontainin g onlyon eo fth eparenta l nucleartypes . (Willregenerat e tofor m oneo f thetw ohomokaryons) . The technique illustrated inFigur e 4show s how weverif y nuclear status of protoplast regenerates. Approximately 10% ofregenerated protoplast s arehomokaryotic . (B)Th eus eo f RFLPs toillustrat e thehomokaryoti c andheterokaryoti c condition inA .bisporus . Lane 1 (on far left) represents atypica lhybridizatio n profile withon eo fou rnuclea rprobes , (showing multiplehybridizatio n signals "bands"). LanesB - H represent profile s from protoplast regenerates. TotalDNA swer eextracte d from allcultures ,digeste d withEcoR I electrophoresised intoagarose ,souther n blottedt oa reusabl enylo n filter andhybridize d to plasmidpAg33N1 0 which waslabelle d invitr ob ynic ktranslatio n (Castlee t al., 1987). Inthi sautoradiogra mth eprofile s inlane sE an dF represen t thehomokaryoti c condition anddemonstrate s thatthi sRFL Pprob ei sheterozygou s for thetw oparenta l nucleartypes .

Utilizing theabundanc eo fRFLP so f thetyp edescrib eabov e allowed ust oeasil y identify thehomokaryoti cprotoplas tregenerate s (Figure 3B) b yDNA/DN A hybridizations. Depending uponth e strain used,u pt o 10% of theprotoplast s regenerated produced homokaryons. Homokaryons produced andverifye d in thismanne rca n then serve as stockfo r acontrolle d breeding programi nA, ,bisporus . (Castleet . al., 1988). In thepast ,homokaryon s havebee nrecovere d from germinating sporeso rfro m excising hyphal tips andidentifie d by slowerradia lgrowt ho f themycelium ,o rb yth einabilit y to fruit. Growth studies inou rlaboratorie s have showntha tman yhomokaryons , especially ifrecovere dfro m wild-collected strains,d ono talway sexhibi tthi sslo w growth (Figure 4A. We suggest thatne wparenta l homokaryotic stocksca n systematically be generated from commercialan dwild-collecte d heterokaryons usingth emethodolog y described. Parental homokaryotic stocksca nb e storedi nliqui dnitroge n andgermplas mbank sca nb e established from breedingpurposes .

66 Temperature (°C)

B ______

( | r-^ZiZjT— 1AO/OS 1 A~~ nm """"»»w (mm) P 45.00 1 ™,0"myb.„,

- H . r—il c^r 36oo

/ 'SA W li BTTMMII- Jfcfl UP 1 MKJUT 37'°° i »x| AIJ mU imilÊË'] l-JI 1 Bmll }H"}

i"*-ifrï^™WEg!5S____iy_^__|___H |9wf] I ^E^K^F~ 9'°° "»-ID-». ^'«aal^BÄjJMÄa •»-20-^. rUBköE«*Hj "»-22~>. "'"Waj -«-24-^ ^____yH__j_Kj ~- — "*-26-k ~*&Wte£££ i «•s***^- ~ - ^Jtf

Figure4 . Radial growth studies of selected homokaryons andintrospecie s hybridso f Agaricusbisporus . (A)Compariso n ofradia l growth after 28days . Note that some homokaryons growmor e slowly thanother s which have similarradia l growth characteristics toth eheterokaryo n Ag95 (thickestline) . (B)Thre edimensiona l analysiso f 4 hybrids generated by Castlee t al. (1988),compare d toAg2 7 (U3hybrid) . Radial growth (xaxis )i splotte d againsttemperatur e (yaxis ) which isplotte d in thethir d dimension (zaxis )tim ei ndays . Thedat a wasplotte d usinga Wing Zprogra m (Informix, International Inc.)o n anAppl e Macintosh computer system (Stockton etal. ,i n preparation). Thehitogra mo n thelef t ofeac hgroupin grepresent sradia l growth measured for the Ag37 strain (U3hybrid) .

67 Producing andverifyin g ane wintraspecie s hybrid In a systematic breeding program,initia lmating sca n bemad eo npetr iplate s asshow ni n Figure 5. Twoplug so f homokaryoticmyceliu mca nb einoculate dont oa petr i platean d theresultan t colonies can beallowe d togro w together. Plugs can betake n from thezon e of hyphal anastomosis and transfered toa secon dpetr iplat e and again allowed togrow . Oneca n often distinguish thene wintraspecie s heterokaryon by atyp eo f "hybrid vigor"a s shown inFigur e 5a (arrow). Amor edefinitiv e methodfo r verifying thene w heterokaryotic condition is tous eth eRFL Panalysi sdescribe d above and showni n Figure 5b toconfir m thene w heteroallelic condition. Finally,on emus td ofruitin g trialso na small scale toestablis h thefertil e natureo f thene wintraspecie s hybrids (Figure 5c). The methods for this approach havebee ninitiall y described in Castlee t al. (1988). In developing these methodologies for breeding mushrooms, wegenerate d anumbe ro f intraspecies hybrids bymatin g parentalhomokaryon s isolated from commercial strainsan d wild-collected strains. Sixteen genetically stablehybrid s wereisolate d from mating6 homokaryons in 65pairwis ecombination s (Castleet.al. , 1988). All 16o f theinitia l hybrids aswel l asth eparenta l heterokaryons weresubjecte d topreliminar y production tests. Thesetrial s weredon e in 2'X 2 'tray scontainin g astandar dcompos t substrate and

Table 1. Some selectedDat afro m Production Trials ComparingErindal eHybrid swit h Selected Commercial Strainsan dWild-Collecte d Parents

Yield %afte r Strain 4 weeks Color

M2 104.74 13.55 M8 84.76 10.21 Ag2 98.86 10.61 Ag89 111.27 17.22 Ag95 88.22 14.00 Ag95-31 101.21 12.69 Ag95- 3 2 72.01 12.24 Ag95- 3 3 96.74 10.88 PS-2 75.80 39.84

This isa summar y of selected datafro m aproductio n trialcarrie d out byDr .Kur t Dahlberg,Campbel l SoupResearc h Institute,Napoleon , Ohio. M2i s aCampbel l strain which consistantly gives growers ahig h yield. TheM 8 strain is also aCampbel l Soup strain;thei r whitest strain. Ag2i sth ecommercia lparen tan dAg8 9i sth e wild-collected parento f ourAg9 5hybrid . Ag95i sth eorigina lintraspecie s hybrid and Ag95-x represents single spore isolates from theorigina l Ag95. PS-2i s abrow n strain. The color scalei s based on analysis done by Campbell Soup. Thelowe rth enumber , themor ewhit eth e mushroom. Theyiel dvalu e isbase do nth efres h weighto fth emushroom s divided byth e dry weighto f the compost. 68 cased withpeat/lime .Al lbu ton eo fth ehybrid s fruited givingfurthe r confirmation thatth e original crosses were successful. Weobtaine d data onyield ,colo ran dquality . Ason e mighthav epredicte dfro m theseinitia lexperiment swit hgeneticall y diversestartin g material,th eyiel dan dqualit y rangedfro m equivalent toth ebes tcommercia l strainsi nus e today (underth eenvironmen t of thisparticula rexperiment ) tover ypoo r (1 mushroom/2 ' X 2'tray) . Ourmos tpromisin g hybrid, Ag95,wa snearl y aswhit e in capcolo r and was asproductiv e asth ebes t commercial strainexamine d (Table 1 ). Thishybri dwa s acros s between aparenta l homokaryonisolate dfro m acommercia l strain anda homokaryo n isolated from awild-collecte d strain (Castlee t al.,1988) . All 16o f thehybrid s generated in ourinitia l studies areth epropert y of CanadianMushroo mGrower s Association. Permission toreleas ethes e strainsmus tb eobtaine d bywritin gth eExecutiv e Vice President of theCanadia n MushroomGrower s Association, 310- 1101Princ e ofWale s Drive,Ottawa ,Ontari oK2 C3W 7CANAD A In addition tofruitin g trials,w eals oexamine d thevegetativ e growth rate andhav e identified thisa sa trai ttha tca n bemanipulate d genetically. Ourresult s have shown that someo f the 16hybrid s generated (including Ag95an dit sprogen y havea broade rgrowt h temperature optima andgenerall y showfaste r lineargrowt h atal ltemperature s thaneithe r of theorigina l parents (Figure 4b) . Themethodologie s described anddevelope d inou r laboratories areno w beingadapte d bya numbe ro f thelarges t Spawn Companies inNort h America. Mitochondrial inheritance andbreedin g inA; ,bisporu s Themitochondrio n in eukaryotic organismsrange s in sizefro m 16 kilobas epair s (kbp) in animals tonearl y 2500kb pi n higherplants . In fungi, therang ei s from approximately 19kb p tonearl y 200kpb ,wit h thelarges t beingreporte d for Agaricus species (Hintz et al., 1985). Despite anearl y 10fol d difference in the sizeo f fungal mtgenomes ,essentiall y similar setso f genes areencode d on fungal mtDNA s (Dawson et al., 1986). Extremely littleinformatio n existso nmitochondria l inheritance andmitochondria l transmissioni n basidiomycetes Somedat aexist s for Coprinus. (Ma y & Taylor, 1988),Armillari a (Smith et al.,1990).an d wehav egenerate d somedat afo r thewil d Agaricus bitorquis. (Hintz et al., 1988b} Furthermore,w ehav ephysicall y mapped themitochondria l genome of the commercial strain,A g 50 (aU 3 hybrid). (Hintze t al., 1988a; Robison M.an dP . Horgen, thisvolume) . Ourprimar y studieso nmushroo mm tDNA ssuggeste d thatonl y asingl e mitochondrial typeexiste d incommercia l isolates ofA .bisporus , whilea numbe ro f polymorphic types were found in A.bitorqui s (Hintz et al,.1985 ;Hint z et al., 1989). During our mapping study of A.bisporu s .w ediscovere d adeletio n variento f theAg5 0type ,on e that wecal l Ag2 (Hintz et al.,1988a) . Morerecently , wehav e found anumbe r ofpolymorphi c types in wild-collected strains of A.bisporu s (~8) . Studies havebee ninitiate d onth em tinheritanc e ofA .bisporus . Iti sentirel y possible thatm tDN A typeca n affect strainperformanc e of A.bisporu s whengrow n under commercial conditions. Ourpreliminar y data suggesttha tm tgenotyp eca naffec t the phenotypiccharacteristic s sucha scolon y growhrate so npetr iplates . Wehav earbitraril y selected threedifferen t mt genotypes,Ag89 ,a nAg5 0typ e (Ag 85) (Hintze l al.,1988a) , theAg 2type ,whic hrepresent s anapproximatel y 20kb pdeletio n of thestandar dAg5 0 type (Hintze t al.,1988a )an da wild-collecte d mtgenotype ,Ag89 . Figure 6show sa comparison of theDN Apolymorphism s ofthes ethre e mtgenotypes . Homokaryons were isolated from protoplasts from thethre efertil e heterokaryons carrying thethre edifferen t mt genotypes byth emethod s described earlieri nthi schapter . Nuclear RFLP differences were used toverif y homokaryon isolation (Figure2 ,3) . We were successful in isolating5 of the6 possibl e homokaryotic typesfro m thethre efertil e heterokaryons used.

69 RFLPS Fruiting

Figure5 - Th eproductio n ofa nintraspecie s heterokaryon (hybrid), (a). Twocompatibl e homokaryons re allowed togro wtogethe ro na petr iplat e sotha tkaryogen y occurs. Two mmplug s aretake nfro m therigh tan dlef t margins of themarke dcultur e andtransferre d to a secondplat e(o nright) . In addition 2mmplug s aretake n from the zoneo f anastomosis andplace di n themiddl eo f thepetr i plate(to p tobotto mo n right). Thecolonie s onth e rightan do n thelef t ofth epetr iplat eo nth erigh trepresen t theorigina l homokaryons whereasth ecolonie si nth emiddl etop/botto mrepresen tputativ ehybrids . Noticeth e fast growing ("hybridvigour" ) of thelowe rcolon y (seearrow) , (b) Verifying thehybri d statuso f RLFPanalysi s homokaryon xi s mated with homokaryon yt ofor m new heterokaryon xy. Similarly, homokaryon yi smate dwit h homokaryon zt ofor m new heterokaryon yz. Thelas t stepi sestablishin g that ane whybri d hasforme d ist ofrui t the putative heterokaryon. Thisi sa photograp h ofErindal ehybri dAg95 . Theresult s thatw ehav eobtaine d todat e suggestth e following: (A) Whenever anAg8 5o r anAg8 9homokaryo n iscrosse d with anAg 2type ,th ene w hybrid strains alwayscarrie d theAg 2mitochondria l typebase do nRFL Panalysi s (Figure6 )an dactua lfruitin g trialst overif y theheterokaryoti c statefollowe d byRFL P analysis. (B) IfAg8 5homokaryon s aremate dwit hAg89 ,th eresultan t hybrid alwayssho wth e Ag85 mtgenotype . Thismitochondria linheritanc e studywhic hi si nit sfina l stagesillustrate stha tth ebreedin g approach describedi nthi spape ri sapplicabl e tobot happlie dan dbasi cmushroo m scientific needs.

70 10 O) CM 00 00 O) O) O) X < < <

Figure6 - Diagrami crepresentation o fth eAgaricu smitochondria lplasmi dpE M(Moha ne t al., 1984). The spherical structure of the5 1end so f this7. 5k blinea rplasmi d (p) represent protein. Sequenceanalysi sindicate s atleas ttw oope nreading frame s on different strands of themolecule ,on eencodin g for aputativ eDN Apolymeras e (Robison et al.,manuscrip t submitted).

References:

Anderson, J.B., Petsche, D.M., Herr, F.B., &Horgen , P.A. (1984). Breeding relationships among several species ofAgaricus. . CSJLLËOL 62,1884-1889. Castle, A.J., Horgen, P.A., &Anderson , J.B. (1987). Restriction fragment length polymorphisms inth emushroom s Agaricusbrunnescen s andAgaricu s bitorquis. Applied Eviron. Microbiol. 53,816-822. Castle, A.J., Horgen, P.A., & Anderson, J.B. (1988). Crosses among homokaryons from commercial andwild-collecte d strainso fth emushroo m Agaricus brunnescens (=A bisporus). Applied Eviron. Microbiol. 54,1643-1648. Elliott,T.J . (1985b). The general biologyo f mushrooms inTh eBiolog y and Technology fifths Cultivated Mushroom (Flegg,P. , Spencer, D.,Wood , D.,eds.) , John Wiley and Sons, New York pp. 9-22. Hintz, W.E.A., Mohan, M., Anderson, J.B. &Horgen , P.A. (1985). The mitochondrial DNAso fAgaricus : heterogenecity in A.bitorqui s andhomogenecit yi n A. brunnescens. Current Genetics 9.127:132. Hintz,W.E.A. , Anderson, J.B. &Horgen , P.A. (1988a). Physical mapping of the mitochondrial genomeo fth ecultivate dmushroo m Agaricusbrunnescen s£=A » bisporusl. Current Genetics 14,43-49. Hintz, W.E.A., Anderson, J.B. &Horgen , P.A. (1988b). Nuclear migration and mitochondrial inheritance in themushroo m Agaricus bitorquis. Genetics 119,35-41 . 71 Hintz, W.E.A., Anderson, J.B. &Horgen , P.A. (1989). Relatedness of three species of Agaricus inferred from restriction fragment lengthpolymorphis m analysiso f the ribosomalDN Arepea t& mitochondria l DNA. Genome 32 ,173-178. Kerrigan, R.W. &Ross ,I.K . (1989). Allozymeso f awil d Agaricus bisporus population: newalleles ,ne wgenotypes . Mycologia 81,433-443 . Kerrigan, R.W., Royer, J., Bailer, L., Kohli, Y., Horgen, P.A. and J. B. Anderson. 1991. AGeneti c mapo f thecultivate dmushroom .Agaricu s bisporus. Submittedt o Genetics. May,G . &J .Taylor , 1988. Patterns of mating andmitochondria l DNAinheritanc e inth e agaric basidiomycete.Coprinu s cinereus. Genetics 118,213-220. Patterson, A., Lander, E., Hewett, J., Patterson, S., Lincoln, S. &Tanksley , S. (1989). Resolution ofqualitativ e traitsint oMedelia nfactor s byusin ga complet elinkag ema po f restriction fragment length polymorphisms. Nature 335,721-730. Peberdy, John,F. ;(1989) . Presidential address: Fungiwithou t coats-protoplasts astool s for mycologicalresearch . Mycological Research 93,1-2 Raper, CA., Miller, R.E. &Râper ,J.R . (1972). Genetic analysis of the life cycleo f Agaricus bisporus. Mycologia 64.1088-1117. Rinker.D .J . (1989). Fungi potpourri. Highlights. Vol 12no .1 March1989 . Royce,D.J .& May , B. (1982). Useo f isozyme variations toidentif y genotype classesi n Agaricus brunnescens. Mycolgia 74.93-102 . Smith, M.L., Duchesne, L.C., Bruhn, J.N. and J. B.Anderson . 1990. Mitochondrial Genetics inNatura l Population of thePlan tPathoge n Armillaria. Genetics 126, 575- 592. Sonnenberg, A.S.,Wessels , J.G. &va n Grrensven, L.J. (1988). An efficient protoplasting/regenerationsyste mfo r Agaricus bisporus and Agaricus bitorquis. Current Microbiology 17,285-291 . Summerbell, R.C., Castle, A.J., Horgen, P.A. & Anderson, J.B. (1989). Inheritance of restriction fragment length polymorphisms inAgaricu s brunnescens . Genetics 123, 293-300.

72 USE OF THE POLYMERASE CHAIN REACTION (PCR) IN AGARICUS BISPORUS BREEDING PROGRAMS

R.S. Khush, L. Morgan, E. Becker & M. Wach

Monterey Laboratories, P.O. Box 189, Watsonville, CA 95076, USA

Summary

Although Agaricus bisporusi s the most extensively cultivated mushroom worldwide, modern approaches to breeding this economically important fungus have been largely ignored. In part, this is due to the organism itself which has a recalcitrant reproductive cycle and few genetic markers. Recently Restriction Fragment Length Polymorphisms (RFLP'S) have been utilized effectively to differentiate between isolates and to verify crosses between homokaryotic cultures. We are utilizing Polymerase Chain Reaction technology (PRC) for use in developing a set of random markers known as RAPD's (Random Amplified Polymorphic DNA). Random DNA segments amplified using single primers appear to be an effective and highly sensitive way of identifying and monitoring crosses between homokaryotic isolates and for "fingerprinting" strains. In addition, we are using PCR to amplify segments of DNA which contain genes of interest. By gathering consensus data from comparable sequences found in other organisms, primers prepared to conserved regions may be used to amplify the intervening DNA. Preliminary data suggests that our laboratory has successfully carried out this protocol for the Agaricus tyrosinase gene.

Introduction

At the American Society of Human Genetics Conference in October of 1985, the scientific community was introduced to the technology we now refer to as the Polymerase Chain Reaction (PCR). Our Laboratory has been developing PCR technology for use in an Agaricus breeding program over the last 2 years. The technology itself is simple. The target DNA is thermally denatured and a short oligonucleotide primer is allowed to anneal to the denatured target DNA. Via a thermostable DNA polymerase, the primer is extended, and the cycle is allowed to repeat. Amplification of the target sequence is exponential so that after 35-45 cycles there is sufficient DNA to observe directly by gel electrophoresis. Depending on the sequence of the primer, the amplified target DNA may be cut with a restriction enzyme to generate "sticky ends" or may be blunt end cloned into the vector of choice.

The use of cloned fragments of chromosomal DNA as genetic markers has evolved into a branch of genetics known as RFLP (Restriction Fragment Length Polymorphism) mapping. This technique relies on the inherent genetic variation found throughout eukaryotes. Inversions, translocations, deletions or transpositions

73 as well as single base pair changes accumulate during the course of selection and evolution. Digestion of higher eukaryotic DNA by a single restriction enzyme typically gives rise to millions of discrete fragments in a continuous range of sizes.

Chromosomal clones may be then be used as probes to elucidate the mutations which have evolved. PCR can be used in an analogous manner. Random oligonucleotide primers, selected singly or in pairs, can be used to amplify small fragments of DNA. Any polymorphism within the simplified sequence will alter the pattern of target DNA as observed by gel electrophoresis. These primers may be mapped in a manner identical to RFLP's. This technique of DNA fingerprinting has been dubbed RAPD for Random Amplified Polymorphic DNA.

We have used both RFLP and RAPD markers to analyze selected commercial and wild strains ofAgaricus bisporus,i n an effort to streamline our breeding efforts. We hope that this work will lead to the eventual linkage of markers to phenotypic traits.

Results and Discussion

Genomic DNA from three commercial and five wild strains of Agaricusbisporus were digested with EcoRI and subjected to Southern hybridization using a single copy genomic DNA probe, pAG33nlO (Summerbell et al, 1989) kindly provided by Dr. J. Anderson. The commercial lines included two near isogeneic off white hybrids, (Amycel Ul and 208) and a brown strain (Amycel 456). Wild isolates were collected from California and Israel as part of the Agaricus Recovery Program administered by Dr. R. Kerrigan. No differences in RFLP patterns were observed between commercial hybrids, however, the probe did reveal a polymorphism between the hybrid strains and the commercial brown mushroom (Figure 1A). Considerable variation was observed among the wild strains and none displayed an RFLP pattern consistent with any of the cultivated strains.

The same group of DNA's were subjected to RAPD analysis. Six oligonucleotide lOmers, synthesized to contain 60% GC, were used as single primers to generate RAPD fingerprints (Table 1). Analysis of the amplified sequences, suggests that as with RFLP probes, certain primer sequences are better suited to fingerprint analysis than others. RAPD 1 is able to differentiate between the hybrid strains (Figure IB), while RAPD 5 is unable to differentiate between the commercial brown and all but one of the wild strains (Figure 1C).

Once primers have been selected which differentiate strains of A. bisporus, RAPD fingerprinting can be used to differentiate between homokaryotic and heterokaryotic cultures. Regenerated protoplasts are identified as homokaryotic by either RAPD or RFLP analysis and are subjected to paired crosses. Mycelium is isolated from matings which exhibit enhanced growth at the interface of the two homokaryotic cultures. These crosses are verified by DNA analysis prior to further experimentation. By using this approach, germplasm from both wild and commercial cultures can be subjected to rapid screening, with only the desired crosses being carried further into the breeding program.

74 12 345678

Figure IA. Southern blot of genomic DNA isolated from eight A. bisporus strains, digested with EcoRl and hybridized with a single copy probe, shows RFLPs. Strains 1 and 2 are commercial white isolates, strain 3 is a commercial brown isolate, and strain 4-8 are wild isolates.

As genetic maps, such as that being prepared by Dr. R. Kerrigan and co-workers, become more sophisticated, we hope to use PCR technology to analyze the segregation of phenotypic traits as well; particularly in F2 populations.

In addition to our work with RAPD markers, we have utilized PCR technology to tentatively identify a previously uncloned Agaricusgene .

Tyrosinase, is the gene responsible for enzymatic browning in the mushroom. It is a multisubunit copper containing monoxygenase primarily involved in the biosynthesis of melanins. The tyrosinase gene has been isolated, cloned and sequenced from several organisms including Streptomyces spp., Neurospora crassa, mouse and humans.

75 Figure IB. RAPD fingerprints of the eight strains shown in figure A obtained with primers 1, 2 and 3. Lane 2-9: primer 1, lane 10-17: primer 2 and lane 18-25: primer 3.

Figure 1C. RAPD fingerprints of the eight strains shown in figure A obtained with primers 4, 5 and 6. Lane 2-9: primer 1, lane 10-17: primer 2 and lane 18-24: primer 3.

By aligning the amino acid sequences from these organisms and identifying overlapping regions of homology or highly conserved sequences, we were able to design oligonucleotide primers for use in the Polymerase Chain Reaction. The left hand primer was a 18 base pair sequence containing 4 third codon degenerations and coding for a sequence just upstream from the enzyme's active site. The right hand primer, located approximately 350-900 bp downstream, was a 17 bp conserved sequence with 3 degeneracies (Table 2).

The primers were utilized in pairs to amplify Agaricus bisporusstrai n 456 genomic DNA. PCR revealed 2 bands (450 and 850 bp respectively) which, based on predicted size were selected for sequence analysis (Table 3). Both amplified fragments contain each of the two primers at the fragment ends and show

76 characteristics attributed to tyrosinase. However, due to the unconserved nature of the tyrosinase genes (Table 4), our analysis must be considered as preliminary. Confirmation awaits amino acid sequence analysis currently underway.

Polymerase chain reaction technology allows the breeder unequalled power in carrying out complex recombinational analysis. While it will be some time before RAPD primers are linked to genetic maps and subsequently to phenotypic traits, the ability to monitor crosses, particularly among inbreed and near isogenic lines, will increase the efficiency of manyAgaricus breeding programs.

References

Summerbell, R.C., A.J. Castle, P.A. Horgen & J.B. Anderson, 1989. Inheritance of restriction fragment length polymorphisms in Agaricus brunnescens. Genetics 123: 293-300.

77 Table 1. Nucleotide sequence of primers used to generate "RAPD" fingerprints via the Polymerase Chain Reaction.

RAPD 1 GGCAGGTAAG RAPD 2 CCGTGACTCA RAPD 3 GCCGCCACCA RAPD 4 CCAGGTGAGT RAPD S AGCCAAGCGG RAPD 6 ACGTA.GCGTC

Table 2. Oligonucleotide primers generated from conserved Tyrosinase sequences.

LEFTPRIME R

nc tcctcttcatcacctggcacaggccctac sg cctcgttcctgccctggcaccgcagatac mm cagggtttctgccttggcacagacttttc hs cagcttttctgccttggcatagactcttc primer gaattcctgccctggcac deg. a cat

RIGHTPRIME R

nc tgttttggttgcaccatgttaacgtcga sg tgttctggctgcaccacgcctacgtcga mm tttttcttcttcaccatgcttttgtgga concensus tgca ca cgctaag ct t deg. tt c primer 5'aagcttàgcgtggtgca3 ' deg. gaa

78 Table 3. Homologous regions between amplified fragments and Tyrosinase genes.

450b.p .frag . FLTWVQ —IGL HLEG WVHHAK 850b.p .frag . FLTWHQ --LHL WLHHAK N. crassa FITWHR —LAL SLED FWLHHVNVD- M.musculu s FLPWHR —LLL TLEG FLLHHAFVD- H. sapiens FLPWHR—LL R S. glaucesens FLPWHR —LLE HLEG FWLHHAYVD- S. antibioticus FLPWHR —LLE HLEG FWLHHAYID-

79 Table 4. Alignment of known tyrosinase sequenses.

nc stdikfaitgvpttpssngavplrrelrdlqqn 13 mm gvddreswpsvfynrtcqcsgnfmgfncgnckf gfggpnctekr vlirrnifdlsvs 57 hs ddreswpsvfynrtcqcsgnfmgfncgnckfgfwgpncter r llvrrnifdlsap 55 sg sa nc ypeqfnlyllglrdfggldeakldsyy qva gihgmp fkpwag vpsdtd 28 mm eknkffsyltlakhtissvyviptgtygqmnngstpmfndiniydlfvwmhyy vsrdtl 116 hs ekdkffayltlakhtissdyvipigtygqnùcngstpmfndiniydlfvwmhyy vsmdal 114 sg tvrknqatltadekrrfvaavlelkrsgrydefvtthnafiigdtd 21 sa tvrknqasltaeekrrfvaallelkrtgrydafvtthnafilgdtd 22 nc wsqpgssgfggycthssil FITWHR pylalyeqaly asvqavaqkfpvegglrakyv 50 mni lgg seiwrdidfaheapg FLPWHR lflllweqeir el 153 hs lgg seiwrdidfaheapa FLPWHE lfllrweqeiq kl 151 sg agertghrsps FLPWHR rylleferala sv 32 sa ngertghrsps FL?WHR rflleferalq sv 32

nc aaakdfrapyfdwasqppkgtlafpeslssrtigyvdvdgktksinnplhrftfhpvnps 69 mm tgdenftvpywdwrdaencdicwdey lggrhpenpnllspasffsswqiicsrsesyn 211 hs tgdenîtipywdwrdaekcdicidey mggçhptnpnllspasffsswq 199 sg dasvalpywdwsadrtarasIwapdflggtgrsldgrvmdgpfaasagnwpinvrvd g 58 sa dasvalpywdwsadrstrsslwapdflggr.grsrdggvmdgpfaasagnwpinvrvdg 56

nc pgnfsaawsrypstvrypnrltgasrderiapiladslasLrnnvsllllsykdfdafsy 94 mm shgvlcd gtpegpllrnpgnhdkaktprlpssadvefclsltqyesgsmdrtanfsfrn 270 hs sg raylrrslgtavrelptraevesvlgmatydtapwnsasdg frn 81 Sä rrflrralgagvselptraevdsvlamatydmapwnsgsdg frn 77

nc nrwdpntnpgdfgsledv hneihdrtggnghmsslevsafdpl FWLHHVNVD rlw 111 mm tlegfasplt giadpsqssmhnalhifm ngtmsqvqgsandpi PT.T.-HKaPVT) sif hs sg hlegwrgvnl hnrvhvwvg ggma tgmspndpv FWLHHAYVD klw 95 sa hlegwrgvnl hnrvhvwvg ggma tgvspndpv FWLHHAYID klw 91

nc siwqdlnpnsfmtprpapystfvaqegesqskstplepfwdksaanfwtseqvkdsitfg 127 mm egwlrrhrpllevypeanapiginrdsymvpfiplyrngdffitskdlgydysylgesdp 385 hs sg aewqrrhpgsgylpaagtpdvvdlndrmkpwndtspadlldhtahytfdtdl 115 sa aewgrrhpsspylpgggtpnwdlnetmkpwndttpaalldhtrhyt.fdvl 111

nc yaypetqkwkyssvkeyqaairksvtalygsnvfl 141 mm gfyrnyiepyleqasriwpwllgaalvgaviaaalsglssrlclqkkkkkkqpqeerqpl 445 hs sg sa

nc mm lmdkddyhsllyqshl* hs sg sa

N.C. Neurospora crassa S.G. Streptomvces alaucesens M.H. Mus muscuius (mouse) H.S. Homo sapiens

80 PROGRESS IN THE MOLECULAR ANALYSIS OF AGARICUS ENZYMES

D.A. Wood1 & CF. Thurston2

1 Microbiology & Crop Protection Department, Horticulture Research International, Littlehampton, West Sussex, BN176LP ,U K

2 Division of Biosphere Sciences, King's College, Kensington, London, W87AH , UK

Summary Agaricus bisporus is cultivated on composted wheat straw. The fungus has been shown to efficiently degrade both lignocellulosic and microbial biopolymers in this substrate. Many of the extracellular enzymes responsible for substrate bioconversion have been identified. Two,laccas e and endocellulase, have been studied in detail since they are also developmentally regulated during fruit body production. Laccase is the most abundant extracellular protein produced byth e mycelium. The enzyme is specifically inactivated during fruiting. Studies have been made of the regulation of activity and production in liquid cultures and solid substrate fruiting cultures. Various forms of the enzyme protein have been identified, purified and antibodies raised against them. The antibodies have been used to find clones coding for laccase in cDNA expression libraries. Verification of the clones as the laccase gene has been achieved aswel l as cDNA and genomic DNA sequence analysis to study gene structure and regulation. Endocellulase has been shown to be cyclically regulated in phase with the cycles of fruit body biomass. Its role in the life cycle has been deduced and correlated with intracellular fruit body cyclically regulated carbon metabolising enzymes. Regulation of cellulase production has been studied in liquid culture and in solid substrate fruiting cultures. Enzyme activity has been purified and shown to comprise 5isoforms . Enzymatic and structural properties of the isoforms have been analysed. Antibody to the 'mixed' isoform bands has been used to measure regulation of enzyme protein levels in fruiting cultures. The antibody has been used to find various putative cellulase clones in a cDNA expression library. One clone (cell) has been verified as a cellulase gene. Genomic clones have been identified and sequenced. The application of this work to strain manipulation of the fungus will be discussed. Keywords: Agaricus. enzymes, molecular analysis, laccase, endocellulase

Introduction Knowledge of the physiology and biochemistry of the various stages of the life cycle of Agaricus bisporus is essential to be able to define genetic targets for the scientific improvement of the agronomic performance of the organism. Two

81 extracellular enzymes of A.bisporus f laccase and endocellulase, are of interest since they are both developmental^ regulated during fruit body production and associated with the nutritional physiology of the fungus. Studies on the properties and regulation of these enzymes have been carried out to determine their physiological roles,protei n structure and genetic structure and control. Analysiso f their genetic structure should provide useful information towards the development and enhancement of a DNA transformation system and to providing candidate genes for strain manipulation. Progress and future work for the analysis of the regulation and molecular properties of these enzymes and their genes is reported aswor k from ourjoin t laboratories.

Progress in molecular analysis

Laccase Developmental regulation and physiological role. Regulation of laccase activity during fruiting of A.bisporu s was observed byTurne r (1974). Wood& Goodenough (1977) confirmed that laccase activity increased during mycelial colonisation of compost and then declined rapidly over 10-20fol d during fruiting. Axenic cultures at various stages showed correlation of laccase regulation with fruit body enlargement. Assays of the quantity of enzyme protein showed that enzyme activity losswa s greater than enzyme protein loss indicating evidence for specific inactivation (Wood, 1980a). The electrophoretic properties of native and denatured enzyme samples before and after fruiting (high activity, HA, and low activity LA), and amino acid analysis, kinetic properties and spectral properties showed large molecular changes occur (Wood, 1980b;Woo d£ lâ L 1990). Antibodies showed that high and low forms exhibit antigenic cross-reactivity but not identity (Wood.s iâ L 1990). The physiological role of laccase in the life cycle of A.bisporu s remains unknown. Its abundance, over 2% of cell protein, and pronounced regulation at fruiting, indicate that its activity should be of some importance for the fungus. It is of interest to note that the extracellular laccase of Schizophyllum commune is also under regulation (activity loss) during fruiting (Leonard &Philips , 1973) Production, purification and properties. A. bisporus laccase has been purified from both liquid and solid substrate cultures (Wood, 1980b;Woo d£ lâ L 1990). Thefirst purifie d preparation studied wasfro m malt extract (ME) grown cultures. The enzyme was characterised as a laccase on the basis of substrate specificity and inhibitor profile. It was characterised as a glycoprotein of about 100kD a molecular weight,containin g 15%carbohydrate . Crude or purified enzyme resolved into several active isoforms on native PAGE gels. Denatured purified enzyme also resolved into several polypeptides. It was concluded that enzyme production was constitutive. The purified malt enzymewa sunlik e other known fungal laccases in being yellow and containing only2 copper atoms per mole enzyme protein as opposed to the classical fungal laccases containing 4 copper atoms (Reinhammar, 1984). Agaricus laccase has alsobee n purified inbul k from fruiting cultures (HA and LAforms) . The low activity form exhibits some properties similar to the high

82 activity form such as molecular weight, pH optimum and substrate specificity. The high activity form is blue, and gives a blue copper signal in e.p.r. analysis. The low activity form is yellow, and gives no e.p.r. signal. Amino acid analysis profiles of malt enzyme and high activity enzyme are similar. The comparison of the properties of two forms, HA and LA, substantiate the suggestion that the LA form may arise by inactivation followed by proteolytic cleavage (Wood, 1980a). The denatured malt extract and HA compost forms of the enzyme differ considerably on SDS-PAGE analysis. The HA form exhibits a predominant poly­ peptide of 65 kDa molecular weight. This polypeptide has been electrophoretic- ally purified and antibodies to it, and to a synthetic peptide of 15 residues of its N- terminal end have been prepared. There are differences in the antigenic behaviour of compost laccase and malt extract laccase, and between their crude and purified forms. Patterns shown by immuno blotting analysis may be due to the generation of partially cleaved enzyme molecules combined with varying levels of glycosylation. Treatment of HA main polypeptide with N-glycanase produced a loss of molecular weight of about 5 kDa.

Biosynthesis. XS methionine labelling has been used to analyse biosynthesis of laccase in malt extract cultures. Following short term labelling (up to 24 hr), immunoprecipitation with antilaccase antibody (ME) a single labelled polypeptide was precipitated from the medium. The mycelial extracts gave a doublet species. Both the medium and mycelial polypeptides were about 68 kDa molecular weight. Poly(A)-containing RNA from mRNA was translated jn vitro and a single polypep­ tide 57 kDa was precipitated consistent with a carbohydrate content of 15% for the exported form of the enzyme.

Isolation and analysis of laccase recombinant DNA. Isolation of cDNA for laccase mRNA was achieved by screening an expression library with affinity purified anti-laccase antibody. Identification of laccase cDNA was confirmed initially by demonstration of two characteristic copper-binding motifs in the deduced amino acid sequence (Wood si aL, 1990). Our analysis of laccase gene sequence is not yet complete. Briefly, further confirmation of cDNA clones has been obtained by alignment of deduced amino acid sequence with sequence of part of a cyanogen bromide fragment of pure laccase protein. The overall laccase sequence shows extensive lack of homology with other fungal laccases that have been sequenced (e.g. Germann si üL 1988). Sequencing of genomic clones has revealed the presence of numerous small introns in the laccase gene sequence and indicated the presence of more than one allele of the laccase gene in the strain studied.

Cellulase

Developmental regulation and physiological role. TurnersiaL (1975) showed that an endocellulase activity increased several fold during fruiting. Wood & Goodenough (1977) substantiated the observation of enzyme activity increase and showed by sampling of axenic cultures that regulation of endocellulase activity was also associated with fruit body enlargement. Mathematical modelling of fruit body production by Chanter (1979) showed that

83 the growth medium was apparently depleted at a uniform rate. The model led to a prediction that 'flushes' were dependent on periodic accumulation of a substrate whichwa s required to reach a certain concentration threshold to allow fruit body biomass construction. Each subsequent 'flush' required reaccumulation of this substrate. Biochemical studies showed that the fluctuation of the carbon compounds trehalose and glycogen, closely followed this model (Hammond, 1985). Various intracellular fruit body carbon metabolising enzymes showed cyclical regulation with the 'flushing' cycles. These enzymes included glucose 6-phosphate isomerase, mannitol dehydrogenase, trehalase, glucose 6-phosphate dehydrogenase and glycogenPhosphorylas e (Hammond, 1985;Well s.slâL , 1987). Claydon.eJ.aL (1988) analysed in more detail the pattern of cellulase production in 'flushing' cycles. Endocellulase activity was present in direct proportion to harvested fruit bodybiomass . Bymanipulatin g fruit body biomass levels itwa s demonstrated that fruit body biomasswa spositivel y regulating the substrate located extracellular endocellulase activity. This regulation pattern could be coupled to that of the intracellular fruit body enzymes to account for the levelso f trehalose and glycogen and the depletion rate of cellulose (Wood.e lÜL , 1988). Fruit bodies may act as a sink, driving mechanisms for increasing carbon supply. The physiological role of cellulase is to replenish the mycelial levels of carbon compounds as these are translocated from the mycelium during fruiting. Removal of the sink may cause carbon catabolite repression to control (shut-down) further cellulase increase. Immunological assays have been used to show that cellulase activity isdirectl y proportional to cellulase protein levels. This suggests that biosynthesis of cellulase protein, and possibly inactivation/proteolysis occurs, during the flushing cycles.

Production, purification and properties. Endocellulase production has been studied using liquid grown cultures (Manning &Wood , 1983). Enzyme production was inducible only by growth on various forms of cellulose or cellobiose. Enzyme yieldswer e highest from growth on insoluble celluloses. Culture filtrates contain other cellulolytic activities but cellobiohydrolase activity has not yet been detected. Maximal production of endocellulase was on low (0.1-0.3%) concentration of celluloses. This observation wasuse d to maximise production yield. The activity comprised both adsorbable and non-adsorbable forms, the ratio of whichwa s dependent on growth conditions. The physiological role and molecular properties of the two forms have not been further studied. Evidence was obtained for carbon catabolite repression of endocellulase production and activity losswa s also observed in such cultures. Endocellulase has been purified from bulk cultures grown on insoluble cellulose. It has been difficult to obtain satisfactory fractionation. Recently purification has been achieved with DEAE-Sepharose and preparative native PAGE gels. Enzyme yield is about 0.2 mg/litre culture supernatant. Unpurified and purified liquid culture enzyme activity and compost extracts all show multiple activity bands, stained with Congo Red on native PAGE gels. 5 bands of increasing mobility (1-5) have been detected. This 'mixed-band' preparation has been used to raise a polyclonal anti-cellulase antibody which reacts with each isoform. Each isoform resolves into further multiple bands on isoelectric focusing, indicating considerable charge heterogeneity. The isoforms exhibit differences in catalytic and structural properties. Band 4i s the most active

84 against carboxymethylcellulose, crystalline cellulose and hydroxyethylcellulose, and also has the highest Km. Analysis of denatured protein by SDS-PAGE gels shows that each band differs, quantitatively in polypeptide composition but share common bands. Isoforms 1,2 contain predominantly polypeptides of 54 and 40 kDa molecular weight, isoforms 4,5 of 22 kDa. Sequence analysis shows that the 54 and 40 kDa bands share a common N-terminal sequence of about 15 amino acid residues. Isoform 3 contains no single predominant band and is the most thermostable activity.

Isolation and analysis of cellulase recombinants. These studies have generally followed the pattern described above for laccase and similarly are as yet incomplete. cDNA clones expressing polypeptide detected by the "mixed-band" anti-cellulase antibody have identified two genes (cell and cel2) that apparently encode two polypeptides of the endocellulase complex. Both of these genes encode mRNA that is expressed differentially in cellulose grown cultures, both encode polypeptides that are identical in size to polypeptides obtain by antibody precipitation of products from jn vitro translation of mRNA (again from cellulose- grown mycelium only), but at present, the relationship between primary translation products and the polypeptides found in the (extracellular) endocellulase complex is obscure. Lastly, for cell the deduced amino acid sequence has revealed that this polypeptide has a cellulose-binding-domain joined to the rest of the sequence by a proline, serine/threonine-rich region homologous to that found in the cellulases of Trichoderma and Phanerochaete (Béguin, 1990).

Conclusions

Further work is aimed at studying biosynthesis and gene expression of these enzyme systems during the life cycle to reveal the level(s) of regulation. DNA sequencing of both cDNA and genomic DNA clones is aimed at establishing the full genetic structure for these enzyme genes. For cellulase it will be of interest to establish if the cyclically regulated extracellular endocellulase genes are similarly genetically regulated, via common upstream sequences, to the intracellular fruit body enzyme genes. Possession of suitable 'upstream' regions of genomic DNA should be of value in development of a DNA transformation system. It may then be useful to alter the regulation and production of laccase and cellulase to see if beneficial changes in the agronomic properties of Agaricus strains can be achieved.

Acknowledgements

The authors wish to thank their colleagues at HRI and King's including Steve Matcham, Norman Claydon, Monica Allan, Edisher Kvesitadze, Caroline Perry,. Selina Raguz and Ernesto Yague, for their contributions to this work.

85 References

Béguin, P., 1990. Molecular biology of cellulose degradation. Ann. Rev. Microbiol. 44:219-248. Chanter, D.O., 1979. Harvesting the mushroom crop: a mathematical model. J. Gen. Micro. 115:79-87. Claydon, N., M. Allan & D.A. Wood, 1988. Fruit body biomass regulated production of extracellular endocellulase during periodic fruiting by Agaricus bisporus. Trans. Br. Mycol. Soc. 90:85-90. Germann, U.A., G. Muller, P.E. Hunhiger & K. Lerch, 1986. Characterisation of two allelic forms of Neurospora crassa laccase. J. Biol. Chem. 263:885-896. Hammond, J.B.W., 1985. The biochemistry of Agaricus fructification. In: D. Moore, L.A. Casselton, D.A. Wood & J.C. Frankland (Ed.): Developmental biology of higher fungi. Cambridge University Press, Cambridge, p. 389-401. Leonard, TJ. & TJ. Philips, 1973. Study of phenoloxidase activity during the reproductive cycle in Schizophyllum commune. J. Bact. 114:7-10. Manning, K. & D.A. Wood, 1983. Production and regulation of extracellular endocellulase by Agaricus bisporus. J. Gen. Microbiol. 129:1839-1847. Reinhammar, B., 1984. Laccase. In: R.Lontie (Ed.): Copper proteins and copper enzymes. Vol. 111. CRC Press, Boca Raton, Florida, p. 1-35. Turner, E.M., 1974. Phenoloxidase activity in relation to substrate and development stage in the mushroom, Agaricus bisporus. Trans. Br. Mycol. Soc. 63:541-547. Turner, E.M., M. Wright, M. Ward, DJ. Osborne & R. Self, 1975. Production of ethylene and other volatiles and changes in cellulase and laccase during the life cycle of the cultivated mushroom Agaricus bisporus. J. Gen. Micro. 91:167-176. Wells, T.K., J.B.W. Hammond & A.G. Dickerson, 1987. Variation in activities of glycogen Phosphorylase and trehalase during periodic fruiting of the edible mushroom Agaricus bisporus (Lange) Imbach. New Phytol. 105:273-280. Wood, D.A. & P.W. Goodenough, 1977. Fruiting of Agaricus bisporus. Changes in extracellular enzymes during growth and fruiting. Archiv, for Microbiol. 114:161-165. Wood, D.A., 1980b. Production, purification and properties of extracellular laccase of Agaricus bisporus. J. Gen. Microbiol. 117:327-338. Wood, D.A., 1980a. Inactivation of extracellular laccase during fruiting of Agaricus bisporus. J. Gen. Microbiol. 117:339-345. Wood, D.A., N. Claydon, K. Dudley, S. Stephens & M. Allan, 1988. Cellulase production in the life cycle of the cultivated mushroom Agaricus bisporus. In: J.P. Aubert, P. Béguin & J. Millet (Ed.): Biochemistry and genetics of cellulose degradation. Academic, London, p. 53-70. Wood, D.A., C. Perry, CE. Thurston, S.E. Matcham, K. Dudley, N. Claydon & M. Allan, 1990. Molecular analysis of lignocellulolytic enzymes of the edible mushroom Agaricus bisporus. In: T. Kent Kirk & H.M. Chang (Ed.): Biotech­ nology in pulp and paper manufacture. Butterworth-Heinemann, Boston, p. 659-666.

86 GENETICS AND DISEASE CONTROL FUNGI IN THE CULTIVATION OF AGARICUS BISPORUS -AN UPDATED LIST OF SPECIES

Albert Eicker & Martmari van Greuning

Department of Botany, University of Pretoria, Pretoria 0001, Republic of South Africa

Summary

Of the numerous fungal species associated with composting and the cultivation of Agaricus bisporus, a few are pathogens causing serious disease while many others are weed moulds influencing the mushroom negatively. For both the mushroom scientist and the mushroom grower it is of vital importance to be able to correctly identify the most common pathogens and weed moulds with special emphasis to their correct nomenclature.

Introduction

Many fungi are associated with the cultivation of Agaricus bisporus (Lange) Sing. Many of these fungi are harmless, posing no threat to the mushroom and have no influence on its yield. Many are also beneficial, playing a vital role in the composting process (Eicker, 1977; Eicker, 1980; Fergus, 1978). However, some fungi are undesirable since some can attack the mushroom directly while others can compete with the mushroom for nutrients or be antagonistic to it. The first group comprises the pathogens which cause serious mushroom diseases. These fungi parasitize the mushroom mycelium in the beds, or they attack the basidiocarps causing them to die, deteriorate or become completely unfit for •huma n consumption. The second group includes the saprophytic fungi, also known as weed moulds. They occur in the compost or casing medium where they compete with the mushroom mycelium for available nutrients. The presence of certain weed moulds may be associated with certain nutritional, chemical or physical conditions of the compost or casing medium, indicating inefficiency of composting or poor growing techniques. The presence of a specific weed mould actually serves as an indication that certain modifications are necessary in composting, hygiene or cultural practices. These moulds assume dominance over the mushroom since they flourish only when the compost is not properly prepared. A weed mould may thus be regarded as an indicator mould, as a competitor mould, or both.

The chief objective of disease control should be prevention. Both the source of the disease and its spread should be eliminated as far as possible by the preparation of good quality compost. However, since pasteurization and not sterilization is part of the composting process and most troublesome fungi occur naturally on the materials from which the compost is prepared, propagules such as conidia, chlamydospores, papulaspores or oospores of unwanted fungi may survive.

89 Thus, although weed moulds and pathogens occur mainly in poorly prepared compost, they may also be present in good quality compost (Fletcher, 1987; Gandy, 1974).

The way to control invasion by these fungi is to regulate the environmental conditions in such a way that the habitat is unsuitable for the germination of their conidia and for their growth. Each fungus has more or less distinctive preferences with regard to the temperature, humidity, pH and nutrition under which it thrives. Information on these aspects would enable the grower to adapt external conditions and to prepare a substrate of such characteristics that growth of the mushroom mycelium is promoted with the exclusion of other fungi.

To successfully control pathogens, weed moulds and antagonistic fungi it is important to know their precise identity. In 1983 Betterly compiled a list of the most common weed moulds. Since then various weed moulds and other antagonistic species were recorded (Betterly & Brown, 1988; Botha & Eicker, 1987; Eicker et al, 1989; Fletcher, 1987; Fletcher, 1989; Geels et al, 1988). This paper is an attempt to list all those fungal species which, in our view, are authentic pathogens and weed moulds of Agaricusbisporus an d at the same time update their nomenclatural status.

Results and discussion

Table 1 is an updated list of troublesome fungi in the cultivation of Agaricus bisporus. Unless otherwise indicated, we have studied all these fungi and sub­ cultures of them are kept in the culture collection of the Department of Botany, University of Pretoria. We have recently also studied compost samples received from England from which we isolated Pénicillium fellutanum Biorge. This weed mould, which we call the compost pénicillium, is a very aggressive antagonist.

The scientific name which we believe to be the correct one is suggested in Table 1. The synonyms and/or incorrect names previously applied to the fungus in "mushroom" literature are listed. It is likely that this list is not complete and we would appreciate comments from fellow scientists.

Table 1. Pathogens and weed moulds in the cultivation ofAgaricus bisporus.

FUNGUS COMMON NAME

Acremonium lamellaeocola (F.E.W.Smith) W.Gams Flock Gill mould Cephalosporiumlamellaecola F.E.W.Smit h "

Aphanocladium album (Preuss) W.Gams '5 Cap spotting Aphanocladium disease "

Arthrobotrys oligospora Fresen. Brown mould

90 Beauveriabrongniartii (Sacc.) Petch Flour mould Botrytisbrongniartii Sacc. Meria sp. Sporotrichum epigaeum Brunaud

Botryotrichumpiluliferum Sacc. & March Pearly white plaster mould Teleomorph: Chaetomium piluliferum J.Daniels

Botrytiscinerea Pers . ex Nocca & Balb. " Brown mould Teleomorph: Various spp. of Sclerotinia and Botryotinia

Cephalotrichummicrosporus (Sacc.) Morton Black whisker & G.Smith mould Doratomyces microsporus (Sacc.) Morton & G.SM Stysanus microsporus Sacc. Stysanusstemonitis (Pers. ex Steud.) Corda Trichurus spiralisHasselbr .

Chaetomium olivaceum Cooke & Ellis Olive green mould Gill mould

Chromelosporiumfulvum (Link ex Fr.) McGinty Cinnamon Hennebert & Korf brown mould Botrytisfulva link Botrytisluteo-brunnea Krzem . & Bad. Byssusfulva Micheli Dematium ollarePers . Chromelosporiumollare (Pers.) Hennebert Mycotypha dichotoma Wolf Sporotrichumfulvum (Link.) Fr. Teleomorph: Peziza ostracodermaKor f

Crysonilia crassa(Shea r & B.O. Dodge) v. Arx Red bread mould Monilia sitophila (Mont.) v. Arx Monilia crassaShea r & B.O. Dodge Teleomorph:Neurospora crassa Shear & B.O. Dodge

Chrysosporium luteum (Cost.) Carmichael Mat disease Myceliophtoralutea Cost. (Vert-de-gris) Scopulariopsis lutea (Cost.) Tubaki Sporotrichum carthusio-viride Rai & Mukerji Teleopmorph: Unknown, probably in the Gymnoascaceae

91 Chiysosoporium merdariium (Link ex Grev.) Confetti Carrn.'! Sporotrichummerdarium Link ex Grev Teleomorph: Gymnoascus uncinatus Eidam

*Cladobotrywn mycophilum (Oudem.) W. Gams & Cobweb Hoozemans Dactylium mycophilum Oudem. Diplocladium elegansBai n & Sart. Teleomorph:Hypomyces odoratusArnol d

Cladobotryum varium Nees !5 Cobweb Botrytisvariospermum Lin k Cladobotryum variospermum(Link ) Hughes Teleomorph: Hypomyces aurantius (Pers. ex S.F.Gray) Tul.

Cladobotryumdendroides 1&i Cobweb (Buil. ex Mérat) Gams & Hoozem. Dactylium dendroides (Buil. ex Mérat) Fr. Teleomorph:Hypomyces rosellus (Alb. & Schw. ex Fr.) Tul.

Coprinus spp.including Ink caps C. radiatus Bolton ex Fr. 7 C. fimetarius (L.) Fr245

Dichobotrysabundans Hennebert 10 none

Diehliomyces microsporus (Diehl & Lamb.) Gil.2 False truffle Pseudobalsamia microspora Diehl & Lamb.

Hormiactis alba Preuss !8 Cap spotting Ramularia sp.

*Mortierella bainieri Cost.4 Shaggy stipe

Mortierella reticulata van Tieghem & Le Monnier !3 none

*Mycogonepemiciosal Magn. Wet bubble Teleomorph:Hypomyces sp.

92 Oedocephalum glomerulosum (Bull, ex Chev.) Brown mould Sacc.1 3 Oedocephalum fimetarium Sacc.7 Teleomorph: Iodophanustestaceus (Mougeot ex Fr.) Korf

Oidiodendronsp 7 ? Vern Astley

Papulasporabyssina Hotson Brown plaster mould Myriococcumpraecox Fr. Teleomorph: Unknown, probably basidiomycetous

Pénicillium spp 7 Green moulds

P. chermesinum Biourge !5 Smoky mould

P.fellutanum Biourge ** Compost pénicillium

Pythium oligandrum Dreschler Black compost disease Pythium artotrogus (Mont.) De Bary Pythium hydnosporum (Mont.) Schröter

Schizophyllum ?commune Fr. None

Scopulariopsis fimicola (Cost. & Matr.) Vuill. White plaster mould Oospora fimicola (Cost. & Matr.) Cub. & Megl.

! Scytalidium hyalinum Campbell & Mulder ' " none

Sepedonium niveum Massée & Salmon u Yellow mould Sepedonium chrysospermum(Bull , ex Fr.) Link7 Teleomorph: Hypomyces sp.

Sporendonema purpurascens (Bon.) Mason & Hughes Lipstick mould Geotrichum candidum Link ex Pers. Geotrichumpurpurascens (Bon ) Sacc. Oosporalacti s (Fres.) Sacc.

Sporotrichumroseum Link " Flour mould Teleopmorph: Corticium sp.

93 Trichotheciumroseum (Pers.) Link ex Gray " Pink flour mould Teleomorph: Hypomyces trichothecioides Tubaki

Trichoderma aureovirideRifai l5 Green mould Teleomorph: Hypocreaaureovirides Plowr. & Cooke

Trichoderma hamatum (Bonord.) Bain Green mould Teleomorph: ? Hyporeasemiorbis (Berk. ) Berk.

T. harzianum Rifai Green mould Teleomoph:Hypocrea vinosa Cook and other Hypocreaspp .

T. pseudokoningü Rifai Green mould Teleomorph: Hypocrea sp.

T. viride Pers. ex S.F. Gray Trichoderma blotch Trichoderma spot Trichodermalignorum Tode Teleomorph:Hopocrea rufa (Pers. ex Fr.) Fr. and other Hypocreaspp .

*Trichoderma koningii Oudem. Green mildew Teleomorph:Hypocrea ceramica Ellis & Everh. Hypocreabrunneo-lutea Do i

*Verticilliumfungicola (Preuss) Hassebrauk Dry bubble varfungicola W. Gams & Van Zaayen Acrostalagmusfungicola Preuss Cephalosporiumconstantinii Smit h Verticillium malthousei Ware Verticillium psalliotae Treschow

Teleomorph: ? Nectria sp.

"Verticilliumfungicola Cap spotting var aleophilum W. Gams & Van Zaayen s* '

"Verticillium psalliotae Treschow Cap spotting

94 * Pathogens ** Eicker isolated Pénicilliumfellutanum Biourge from heavily infected mushroom compost received from Middlebrook Mushrooms, England. *** Dr James Sinden drew our attention to this fungus, which is intimately associated with Agaricus bisporus rhizomorphs in compost, inhibiting mushroom yield severely.

! Cultures not studied by authors.

1 Betterley, 1983; 2 Geels et al., 1988; 3 Betterley & Brown, 1989; 4 Fletcher, 1973;5 Fletcher et al, 1989; ' Seaby, 1987; 7 Harvey et al, 1982; 8 Fermor, 1979; 9 Gams & Van Zaayen, 1982; I0 Coetzee, 1987; " Singer & Harris, 1987; 12Nair , 1984; 13 Botha & Eicker, 1986; " Eicker, Smit & Wuest, 1989.

References

Betterley, D.A., 1983. Indicator and weed molds associated with Agaricus brunnescens.Mushroo m News 31(3): 9-12. Betterley, D.A. & Brown, M.F., 1989. A Mortierella species (Zygomycetes) associated with cultivation of Agaricus brunnescens. Mushroom Science 12(2): 771-777. Botha, W.J. & Eicker, A., 1987. Notes on .the physiology and morphology of Sepedonium niveum, a newly recorded competitor mould in mushroom compost. Proceedings, International Syposium Scientific and Technical Aspects of Cultivating Edible Fungi. Elsevier Science Publishers B.V., Amsterdam: Developments in Crop Science 10:331-339 . Coetzee, J.C., 1987. Die biologie van enkele fungussoorte wat by die kommersiële verbouing van Agaricus bisporus (Lange) Imbach in Suid-Afrika problème skep. (The biology of some fungus species causing trouble in the commercial cultivation of Agaricus bisporus (Lange) Imbach)) MSc Thesis, University of Pretoria. Eicker, A., 1977. Thermophilic fungi associated with the cultivation of Agaricus bisporus(Lange ) Singer. Journal of South African Botany 43: 193-207. Eicker, A., 1980. Mesophilic fungi associated with the cultivation of Agaricus brunnesscens. Transactions of the British Mycological Society 74: 465-470. Eicker, A. Smit, Martmari & Wuest, P.J., 1989. A scanning electron microscope study of troublesome fungi in the Agaricus bisporus mushroom industry. Mushroom Science 12 (2): 789-801. Fermor, T.R., 1970. Hormiactis alba, an uncommon fungal pathogen of the cultivated mushroom, Agaricus bisporus. Annual Report Glasshouse Crops Research Institute, 191-193. Fletcher, J.T., 1973. Shaggy stipe, a new disease of cultivated mushroom caused by Mortierella bainieri.Plan t Pathology 22:25-27 . Fletcher, J.T., 1987. Weed moulds. Mushroom Journal 174: 198-200. Fletcher, J.T., P.F. White & R.H. Gaze. 1989. Mushrooms - pests and disease control. 2nd edition. Intercept, Andover. Gams, W. & A. Van Zaayen., 1982. Contribution to the taxonomy and pathogencity of fungicolous Verticillium species. I. Taxonomy. Neth. J. Plant Pathol. 88: 184- 187.

95 Gandy, D.G., 1974. Weed moulds. Mushroom Journal 23:428-429 . Geels, F.P., J. van Geijn & A.J. Rutjens., 1988. Pests and diseases. In L.J.L.D. Van Griensven (ed.), The cultivation of mushrooms, pp. 361-422. Darlington Mushroom Laboratories Ltd., Rustington, UK. Harvey, C.L., P.J. Wuest & L.C. Schisler., 1982. In P.J. Wuest & G.D. Bengtson (ed.), Penn State Handbook for Commercial Mushroom Growers, pp. 19-33. University Park: The Pennsylvania State University Press. Nair, T., 1984. Diseases of cultivated mushrooms. Agfacts H8.AB.30, first ed. Department of Agriculture, New South Wales. Seaby, D.A., 1987. Infection of mushroom compost by Trichoderma species. Mushroom Journal 179: 355-361. Koeltz Scientific Books, Germany. Singer, R & B. Harris., 1987. Mushrooms and Truffles. Koelz Scientific Books, Germany.

96 NATURE OF DISEASE RESISTANCE TO COMPOST-BORNE AND AIRBORNE PATHOGENS OFAGARICUS BISPORUS

P.J. Wuest

Department of Plant Pathology, Penn State University, 211 Buckhout Laboratory, University Park, P.A. 16802, USA

Summary

Disease resistance is a legitimate tool for disease management in A. bispoms. Resistance to bacterial blotch, Verticillium,ma t (confetti), truffle, Trichoderma and nematodes has been reported for existing spawn strains, but disease resistance has not been a primary consideration of those developing or buying spawn strains. A proposal for international collaboration is put forth as a way to catalogue disease resistance. Keywords: mushrooms, pest management, Pseudomonas tolaasii, Verticillium fungicola, Trichodermaviride, Chrysosporium luteum.

Introduction

Mushroom diseases have been recognized as a threat to mushroom crops for more than a century yet there are no spawn strains of A. bisporusadvertise d and sold on the basis of disease resistance in 1991.

Farmers informally score the susceptibility of commercial spawns to a variety of diseases. Diseases of particular concern include: bacterial blotch, Verticillium, mummy and virus. Certain strains are perceived as being more or less susceptible to one or more diseases, diseases which periodically infest their crops. Farmer evaluations of susceptibility are based on the amount of the crop infected and showing symtoms, the yield and quality of the harvest, and the opinions of neighboring farmers. Some commercial spawns have lost their portion of the market based on such evaluations, and farmer perceptions have evolved into the reality of choosing a specific spawn to plant.

Disease resistance in mushroom strains is important to farmers, spawn makers and the public since genetically mediated disease resistance is a low impact, pesticide free, environmentally benign method of disease management, and these attributes reflect the environmental consciousness of many people. Rotem and Palti (1980) considered disease management via cultural modifications and codified attributes which relate to this tactic. Such management of disease is an alternative to the intradiction or prevention of disease with agrichemicals, as are the successful and widely used ecological controls described by Sinden (1971). Nonchemical approaches to disease management include the use of spawn strains resistant to disease. Low impact, nonchemical and sustainable are key concepts in pest management, and the attributes delineated by Rotem & Palti for cultural

97 modifications in disease management have application to gene-based disease resistance in mushrooms.

The prospects for cultural control are improved when: 1) Crop value and level of potential crop loss is low; 2) Cost of chemical control, relative to overall growing expenses, is high; 3) Changes... of field conditions are many; 4) Educational level of the farmer is high; and others. Mushroom farming satisfies 3 and 4, but not 1 and 2. Mushroom farmers prefer redundancy in disease management systems because crop loss potentials are high. Use of disease resistant germplasm is a preferred approach in disease management, yet disease resistance is an ancillary attribute in spawn strains since yield, quality, size and production traits are given precedence in selecting new germplasm.

One purpose of this paper is to identfy spawn cultivars reported to be disease resistant. How disease resistance can be seen, quantified and used is a second goal of this report. Comprehensive reviews of disease symptoms, signs, etiology and epidemiology are available, are readers are referred to them for these aspects of mushroom diseases (Gandy, 1985, Fletcher et al., 1989 and Harvey et al., 1982) and Carlile (1988) provides a generalized view of disease resistance in green plants.

Diseases of the Sporocarp

Bacterial Blotch

Olivier (1984) reported on disease resistance, and selected data from that paper are presented in Table 1. These data suggest that white strains have greater general resistance (nonspecific resistance) to bacterial blotch than cream or brown strains. It is also apparent from the differences in the percentages or caps with symptoms that disease resistance differs in each two-strain comparison.

Table 1. Response ofA. bisporus strains to bacterial blotch.

Strain % Diseased caps

White S 56 40 LLC45 5

CreamLL C75 0 LLB3 100

Brown LLB86 60 S 665 100

S= Somycel; LL= LeLion; Inoculum- 10'ml- Ref: Olivier, 1984.

98 Lesion size differences are not reported, and such comparisons would provide additional insight about the ability of each strain to limit infection (specific resistance). Cole & Skellerup (1986) suggested that they observed specific resistance at the cellular level, but allow that additional studies are needed to validate this supposition. Lesion size may be considered a superfluous parameter considering that one lesion can downgrade a mushroom from fresh market quality (unblemished) to a processing grade, or less. Choosing disease symptoms of significance exclusively from a marketing perspective is too constrained an approach in assessing disease resistance. Symptom expression reflects the activities of one or more genes, and mushroom breeders may be able to use this parameter in developing breeding strategies.

Verticillium Disease

In recent experiments by the author, Table 2, disease resistance is quantified. Unlike bacterial blotch where crop loss due to a lowering of the harvest weight is not a consideration, Verticilliumcause s crop losses in many ways as enumerated by Gandy (1971). Data in Table 2 reflect the percentage of mushrooms affected by Verticillium, and all mushrooms with spotted caps or split stipes were harvested separately and weighed, as were unblemished mushrooms. It appears the off white hybrid (OWH, Ul) is more susceptible to Verticilliumtha n the white hybrid (WH, S381). A significant increase in the number of dry bubbles per unit area with the OWH affected the number of spotted mushrooms and those with stipe infections, especially on mushrooms formed after 1st break. The density of dry bubbles per unit area impacts the percentage of mushrooms with other Verticillium symptoms.

Table 2. Response of two hybrid strains of A. bisporus to Verticillium fiingicola.

Strain* Healthy" Diseased" %Spot' Bubbles m2 yield yield

WHS381 D 2.11 4.65 32.4 3.4 H 8.96 0.10 9.1 0.8

OWHU-1 D 1.86 4.46 43.7 16.8 H 9.21 1.47 6.7 3.2

': S=Sylvan Spawn Company, Kittaning, PA, USA; U-l via Amycel, Forrestville, PA, USA. ": Yield expressed Kg m2; c: %spot = %spot/(Diseased + Healthy) x 100

Additional insight on disease resistance, disease increase and crop loss is illustrated in Figure 1. This experiment included two white strains, 191 and 615, and two brown strains, 816 and 901 (North, 1987). Cultures 615, 816 and 901 originated with Mushroom Growers Cooperative Assoc, Kennett Square, PA and 191 was from Amycel, Forrestville, PA, USA. Yield loss (g) of statistical significance occurred in only one white strain, although yield of unblemished mushrooms was lower in all

99 treatments infested with V. fungicola.

Viewing Fig. lb allows one to conclude that the brown strains, 816 and 901, experienced lower rates of disease increase during the 4 - break harvest than the white strains, 191 and 615. These data indicate brown strains possess more general resistance to Verticillium disease than the white strains. Identifying the specific phenotypic expression of this resistance is not possible from the data presented. Olivier (1986) reported that brown strains were resistant to the dry bubble syndrome of Verticillium disease, and this could help explain the lower rate of disease increase in the brown strains used by North.

Whether a breeding program is well served by only weighing the symptomatic mushrooms remains an open question, but further study may clarify this approach to quantifying this disease. Considering that symptoms of Verticillium disease can be categorized into symptoms of the sporophore and symptoms of the tissue that may develop into sporophores (dry bubbles), one may ponder the relationship between these two symptoms and their contributions to crop loss. There appears to be a relationship between dry bubble occurrence and the rate of disease increase. Dry bubbles serve as inoculum sources. Propagules of the pathogen splash onto the pilei when irrigation water is applied to the crop. General resistance slows disease development and it will be of greater significance as the harvest period of a crop is extended in time. Specific resistance affects lesion appearance and development, and it will be of greater significance where cropping economics dictate brief harvest periods of 2 or 3 breaks.

Diseases of the Vegetative Stage

Pathogens in this category usually infest compost either when the compost is being nurtured or processed, or at spawning. An earlier report (Wuest & Harvey, 1978) indicated that resistance to T. viride, C. luteum, and D. microsporia existed. Pathogenesis of compost-borne fungal pathogens is based on the production of metabolites by each pathogen (Harvey, 1978). These metabolites interfere or redirect metabolic activities of the suscept. Observations of pathogenesis revealed the lack of rhizomorph formation, death of mycelium and rhizomorphs, lack of primordia formation, reduced yields, and others. Our level of understanding suggests resistance is general and nonspecific, but additional information may provide another perspective into the nature of disease resistance.

Disease Resistance in Disease Management Programs

Resistance to major and minor diseases exists, but rarely is it put forward as a legitimate method of disease management. Information on disease resistance is lacking and no one has developed a comprehensive enumeration of disease resistance to all diseases for any strain.

Global collaboration may afford an opportunity to develop more information on disease resistance. Disease management by use of resistance is within our grasp, and it is a significant goal. Disease resistance will take its place with crop production criteria in developing new strains of A. bisporus when this trait is prioritized and

100 viewed as a legitimate and useful tool for disease management in the mushroom crop.

Acknowledgements

The author express his thanks to Ms. Harvey and Mr. North for the fine research both of them conducted during their graduate studies at Penn State University. It is their research that provided much of the information in this manuscript. Mr. Harry Muthersbaugh, Supervisor, Mushroom Research Center, Penn State University provided a great deal of assistance and moral support in the course of this research, and he still is not a fan of disease research next to physiological or genetic research.

References

Carlile, W.R., 1988. Production and use of crop varieties (cultivars) resistant to disease. In: Control of crop disease. p45-56. E. Arnold, London. Cole, A.L.J. & M.V. Skellerup., 1986. Ultrastructure of the interaction ofAgaricus bisporusan d Pseudomonas tolaasii.Trans . Brit, mycol. Soc. 87(2): 314-316. Fletcher, J.T., P.F. White & R.H. Gaze, 1989. Mushrooms-Pest and Disease Control, 2nd ed. Intercept Ltd., Newcastle upon Tyne, England. 174pp. Gandy, D.G., 1971. Observations on the development of Verticillium malthousei in mushroom crops and the role of cultural practices in its control. Mushroom Sei. 8: 171-181. Gandy, D.G., 1985. Bacterial and fungal diseases. In: The Biology and Technology of the Cultivated Mushroom. Ed. P.B. Flegg, D.M. Spencer, and D.A.Wood. J. Wiley & Son, Inc. London. p261-277. Harvey, C.L., 1987. Redaction of three strains of the cultivated mushroom, A. brunnescens Peck, to four compost-invading fungi. M.S. Thesis. Department of Plant Pathology, Penn. State University. 56pp. Harvey, C.L., PJ. Wuest & L.C. Schisler, 1982. Diseases, weed molds, indicator molds and abnormalities of the commercial mushroom. In: Penn State Handbook for Commercial Mushroom Growers, Ed. P.J. Wuest and G.D. Bengston. Penn. State University. North, L.H., 1987. Studies on Verticillium disease of Agaricus bisporus (Lange) Imbach. M.S. Thesis. Department of Plant Pathology, Penn State University. 40pp. Olivier., J.M., 1984. Bacterial blotch in French caves. In: Symposium Proc, Bacterial Blotch, Ed. R.L. Edwards. ISMS, c/o HRI, Worthington, England. p31-45. Olivier, J.M., 1986. Evolution of the phytopathological situation in French caves. Dev. Crop Sei. 10: 351-360. Rotem, J. & J. Palti, 1980. Epidemiological factors as related to plant disease control by cultural practices. In: Comparative Epidemiology, a tool for better disease management. Ed. H. Palti and J. Kranz. Pudoc, Cntr. Agr. Publ. & Doc., Wageningen, Netherlands. pl04-116. Sinden, J.W., 1971. Ecological control of pathogens and weedmolds in mushroom culture. Ann. Rev. Phytopathology 9: 411-532. Wuest, J.W. & C.L. Harvey, 1987 The nature of disease resistance in strains of the cultivated mushroom,Agaricus brunnescensPeck . Mushroom Sei. 10(1): 741-746.

101 5000

4000-

o 3000- •J

z < 2000

1000-

615 816 MUSHROOM STRAIN Figure la. Influence of Verticillium fungicola on the yield (g) of four Agaricus bisporus cultivars following inoculation with 5.4 x 106/m2 phialospores when mycelium reached the surface of the casing; Experiment 2. Significant difference, p = 0.05, indicated by *.

uQ <

BREAK Figure lb. Incidence of Verticillium disease on Agaricus bisporus assessed on the peak picking day of each break. Inoculum was applied when mycelium reached the surface of the casing; Experiment 2.

102 Molecular Dissection and Control of Virus Disease in Agaricus bisporus

M.C. Harmsen and J.G.H. Wessels

Department of Plant Biology, University of Groningen, Kerklaan 30, 9751 NN Haren, The Netherlands

SUMMARY

La France disease of the cultivated mushroom Agaricus bisporus is associated with the presence of ten differently sized dsRNAs in diseased fruit bodies. One of these dsRNAs is also present in healthy mushrooms. The dsRNAs appear to be encapsidated by virus particles of 25 and 34 nm. The ten dsRNAs do not share large sequence homologies, and at least five of them encode proteins as was shown by in vitro translation analyses of denatured dsRNA. A small minor dsRNA (S3, 0.4 kb) that sometimes is maintained in high concentrations in diseased mushrooms was found to be an internally deleted form of M2 dsRNA (1.3 kb). We have constructed cDNA clones for all dsRNAs. Clones with full-length cDNA inserts from L3, M1 and M2 dsRNA were isolated and sequenced. Computer analyses suggested single large open reading frames (ORFs) for all three of these dsRNAs. No homologies with sequences in databanks could be detected. Labelled cDNA clones of L3, M1 and M2 dsRNA were used as hybridization probes for the detection of La France disease in fruit bodies, mycelium, spawn and compost. It was shown that dsRNAs connected to La France disease were detectable in compost at early stages of the cultivation cycle. In the future the regular use of this diagnostic test may prevent spreading of La France disease and thus lead to elimination of the disease.

INTRODUCTION

La France disease of the cultivated mushroom Agaricus bisporus was first observed by Sinden and Häuser (1950). The main symptom is loss of crop which may be accompanied by malformed fruit bodies. It appeared that diseased fruit bodies contain virus particles: these were in fact the first fungal viruses, or mycoviruses, to be described (Hollings, 1962). In almost 30 years that have since elapsed particles of different sizes and shapes have been reported to be present in diseased and healthy mushrooms as well as in mycelium (reviewed by van Zaayen, 1979). It still remains to be properly shown that one or some of these viruses are responsible for La France disease. Most studies only concern ultrastructural analyses of occurring viruses. One of the particles is a bacilliform virus (19x50 nm) that contains a single-stranded RNA genome (Tavantzis et al., 1980). It is not always present in diseased mushrooms and therefore this virus is unlikely to cause La France disease. Two spherical particles that seem to be related to with the disease measure 25 and 34 nm and contain a double-stranded RNA genome (Barton & Hollings, 1979) which is commonly found for mycoviruses (Buck, 1986).

103 RESULTS AND DISCUSSION

Molecular analysis of A. bisporus dsRNA.

Our work was primarily concerned with molecular analyses of dsRNA associated with A. bisporus and the subsequent development of a diagnostic test for the presence of dsRNAs causing La France disease. The test should preferably be applicable to compost at early stages of the cultivation cycle. Double-stranded RNA can be quickly and efficiently extracted from phenol- extracted fungal material by using chromatography on cellulose in the presence of 15% ethanol. After elution, the dsRNA can be analyzed by conventional agarose gel electrophoresis. In the U.S.A. six or more dsRNAs were found associated with symptomatic mushrooms (Marino et al., 1976, Wach et al., 1987). In the Netherlands diseased mushrooms appeared to contain ten major and several minor dsRNAs (Table 1, Harmsen et al., 1989). One of the major dsRNAs (L6) was also routinely detected in healthy sporophores from high- yielding crops. Thus, L6 dsRNA is either not involved in La France disease or it supplements other dsRNAs in causing the disease. The dsRNA in sporophores and mycelium were found to be similar.

Table 1. Lengths, transcripts and coding capacity of A. bisporus dsRNA.

transcript in vitro translation dsRNA size (kb)2 length (kb)3 product (kDa)4

XLd-4)1 15 ND ND LO1 6.5 ND ND L1 3.6 ND 130 L2 3.0 ND 84? L3 2.8 2.8 88 L4 2.7 ND 84? L5 2.5 ND - L6 2.35 2.35 and 1 - L7' 2.0 ND ND L81 1.9 ND ND M1 1.6 1.6 - M2 1.35 1.35 45 S1 0.86 ND 28 S2 0.78 0.78 28 S31 0.40 0.40 -

1 These dsRNAs were only observed in minor amounts, except for some samples in which large quantities of S3 dsRNA were found. 2 Comparison was with A DNA cut with fcoRI and Hinó\\\. 3 Transcripts were detected in total single-stranded RNA from diseased mushrooms by hybridization with cDNA clones. ND: not determined. 4 A "-" sign is used for those dsRNAs for which no in vitro translation product has been identified yet.

Cross-hybridizations between the major dsRNAs revealed no significant sequence homologies, suggesting that they were all unique in sequence, except for S3 dsRNA that strongly hybridized to M2 dsRNA. The coding capacity of the dsRNAs was determined by in vitro translation of denatured total and individually purified dsRNA segments. At least five dsRNAs were found to code

104 for proteins in vitro, whereas a major product of 59 kDa after translation of denatured total dsRNA could not be assigned as yet. Furthermore, two closely migrating products of 84 and 88 kDa were found after translation of denatured total dsRNA, one of which (88 kDa) could be assigned to L3 dsRNA, whereas the other might be coded for by L2 or L4 dsRNA. We anticipate that future experiments will show that all ten major dsRNAs code for proteins. We have obtained evidence that L6 dsRNA is encapsidated by 25 nm particles (Harmsen, 1990). Recent experiments provide strong evidence that nine major dsRNAs, but not L6 dsRNA, are encapsidated by 34 nm particles (T.R. van der Lende et al., Abstr. First Int. Sem. Mush. Sei., 1991). This strongly suggests a viral basis for all ten major dsRNAs and that at least the 34 nm particles are involved in La France disease.

Sequence analysis of A. bisporus dsRNA.

Total dsRNA was reverse-transcribed and the cDNA was cloned in E. coli. Clones for all major dsRNAs were isolated. However, only for three dsRNAs (i.e. L3, M1 and M2) full-length cDNA inserts were obtained. The other dsRNAs probably contain regions with strong secondary structures or with a base- composition that is unfavourable for the AMV reverse transcriptase to read through. The cDNA clones of L3 (pAV260353), M1 (pAV250226) and M2 (pAV322) were completely sequenced and putative ORFs were detected (Harmsen et al., 1991). Each of these dsRNAs contained only one large ORF (Fig. 1 ), coding for proteins of 87 kDa (pL3), 40 kDa (pM1) and 38 kDa (pM2), respectively. As mentioned before, in vitro translation products have been detected for L3 and M2 dsRNA, thus the predicted proteins may actually be formed. However, their function remains obscure as they did not resemble any known protein present in databanks. Neither did the nucleotide sequence of the dsRNAs reveal homologies with other nucleic acids.

pL3 Wen 87 kDa

pHI 359aa 40 kDa pM2 340aa 38kD a

_^^^—_K07nt 1307n t

Fig. 1. Comparison of the dsRNAs L3, M1 and M2. Each dsRNA is represented by a line (drawn to scale) in which the thickened part corresponds to the predicted ORF (pL3, pM1 and pM2). The total number of amino acids (aa) and deduced molecular weights are given for the putative proteins.

Full-size transcripts from dsRNAs S3, S2, M2, M1, L6 and L3 were detected in total single-stranded RNA preparations from diseased mushrooms by hybridization with labelled cDNA clones (Table 1). These Northern analyses show that at least six of the dsRNAs are actively transcribed in diseased fruit bodies. Whether these transcripts all serve as messengers for proteins or act as templates for the synthesis of dsRNA is unknown. In contrast to the full- sized transcripts for most of the dsRNAs a truncated transcript of

105 approximately 1 kb was found for the L6 dsRNA. Because this cannot be an intermediate in replication it must be messenger RNA. By using the derived restriction map of the M2 cDNA clone and Southern analysis, the position of S3 dsRNA was mapped (Harmsen et al., 1991). It was found to be an internally deleted variant of M2 dsRNA that comprised approximately 0.15 kb from one and 0.25 kb from the other end of M2 dsRNA. Close to these "breakpoints" strong secondary structures were predicted that may be involved in the formation of S3 dsRNA (Harmsen et al., 1991).

Development of a diagnostic test

Hybridization of gel-separated nucleic acids from diseased and healthy fruit bodies with 32P-labelled cDNA clones from dsRNAs L3, M1, M2 and L6 showed that only the L6 probe hybridized to extracts from healthy A. bisporus. Thus, clones from L3, M1 and M2 dsRNA were suitable for detection of the virus associated with the disease. Similar results were found with mycelial extracts. Using a P-labelled probe a minimum of 1 pg total dsRNA could be detected, which was approximately 104 times less than could be detected by agarose gel electrophoresis and staining with ethidium bromide (>10 ng). Initially, compost containing mycelium was analyzed similarly. However, this only gave good results with gel-separated samples and 32P-labelled probes; spotted samples gave unacceptable high background signals. An additional purification step using a small ion-exchange column (Chelex-100) greatly reduced the background and thus made spot assays of compost samples possible.

• • • • • • f* • • • ® . • • # ''-• © • • o •i

Fig. 2. Spot hybridization for the detection of dsRNAs in mycelium containing compost. Extracted compost samples were spotted on a nylon membrane and hybridized to a probe comprising the cDNA inserts of the dsRNAs L3, Ml and M2 and labelled with DIG-dUTP. Hybrids were traced with a conjugate of anti- DIG-antibodies and alkaline phosphatase using a chemiluminescent substrate (AMPPD) and autoradiography to detect the signal.

A comparison was made between the electron microscopical analysis of 36 samples of suspected mushrooms and hybridization analysis of the compost substrate. In the majority of cases the presence of particles coincided with the detection of dsRNAs in the compost. In two cases only dsRNAs were detected but no particles were found, suggesting that at some stages the virus titre may be too low to enable detection at the EM level. In an attempt to determine the earliest moment of detectable dsRNAs, compost was inoculated with healthy

106 spawn mixed with a small amount of diseased spawn. At weekly intervals compost samples were analyzed and already one week after spawning dsRNAs indicative of La France disease were detectable. Thus, at early stages the potential disease could be detected. Strong hybridization signals were found after two weeks of growth, which would make the hybridization test suitable for the screening of full-grown compost which is used by an increasing number of growers in the Netherlands. As the use of 32P-labelled probes needs specialized laboratory facilities, we also tested the effectiveness of probes labelled with biotin-dUTP and digoxigenin-dUTP (DIG-dUTP). Hybridization was done similarly as with 32P probes and detection was done with conjugates of alkaline phosphatase (AP) and (strept)avidin for biotinylated probes and anti- DIG-antibodies for DIG-probes. Initially, a combination of NBT and BCIP was used as a substrate for the AP to develop a purple colour at places were hybrids had formed. Recently, a chemiluminescent substrate (AMPPD) for AP became available. The best results were obtained with biotin- or DIG-probes using the chemiluminescent substrate (Fig. 2). This test is now routinely used at the Mushroom Experimental Station in the Netherlands for the screening of suspected compost samples. Some attempts were made to use the polymerase chain reaction (PCR) for diagnostic purposes. Two PCR amplification primers were derived from the sequence of L3 dsRNA and used for the detection of this dsRNA in suspected compost samples (Harmsen, 1990). Although very sensitive with purified dsRNA (detection of a few hundred dsRNA molecules) the detection in compost was less sensitive than the hybridization method, probably due to unknown inhibiting compounds. Provided a better purification is found PCR may be developed as an alternative or additional testing method to screen samples for dsRNAs associated with La France disease.

ACKNOWLEDGEMENTS

We thank Drs. A.S.M. Sonnenberg and L.J.L.D. van Griensven of the Mushroom Experimental Station for the supply of compost and fruit body samples and the assistance in some of the experiments. This research was supported by the Netherlands Foundation for Technical Research (STW) and coordinated by the Foundation for Biological Research (BION).

REFERENCES

Barton, R.J. & M. Hollings, 1979. Purification and some properties of two viruses infecting the cultivated mushroom Agaricus bisporus. J. Gen. Virol. 42: 231-240. Buck, K.W., 1986. Fungal virology - An overview. In: K.W. Buck (Ed.) Fungal Virology. CRC Press, Boca Raton, p. 1-84. Harmsen, M.C., 1990. Ph.D. Thesis, University of Groningen. Harmsen, M.C., L.J.L.D. van Griensven & J.G.H. Wessels, 1989. Molecular analysis of Agaricus bisporus double-stranded RNA. J. Gen. Virol. 70: 1613-1616. Harmsen, M.C., B. Tolner, A. Kram, S.J. Go, A. de Haan & J.G.H. Wessels, 1991. Sequences of three dsRNAs associated with La France disease of the cultivated mushroom {Agaricus bisporus). Curr. Genet, in press. Hollings, M., 1962. Viruses associated with a die-back disease of cultivated mushroom. Nature 196: 962-965. Marino, R., K.N. Saksena, M. Schuier, J.E. Mayfield & P.A. Lemke, 1976. Double-stranded ribonucleic acid in Agaricus bisporus. Appl. Environ. Microbiol. 31: 433-438.

107 Sinden, J.W. & E. Hauser, 1950. Report on two new mushroom diseases. Mushroom Sei. 1: 96-100. Tavantzis, S.M., C.P. Romaine & S.H. Smith, 1980. Purification and partial characterization of a bacilliform virus from Agaricus bisporus: A single- stranded RNA mycovirus. Virology 105: 94-102. Wach, M.P., A. Sriskantha & C.P. Romaine, 1987. Double-stranded RNAs associated with La France disease of the commercial mushroom. Phytopath. 77: 1321-1325. van Zaayen, A., 1979. Mushroom viruses. In: P.A. Lemke (Ed.): Viruses and Plasmids in Fungi. Marcel Dekker Inc., New York. p. 239-324.

108 EVIDENCE FOR TRANSMISSION OF LA FRANCE DISEASE INAGAJUCUS BISPORUS BY DS RNA

Anton S.M. Sonnenberg & L.J.L.D. Van Griensven

Mushroom Experimental Station, 5960 AA Horst, The Netherlands

Summary

Experimental evidence is presented that La France disease or Die Back can be transmitted by hyphal anastomosis between diseased and healthy strains. By using isoenzymes and RAPD's as nuclear markers and RFLP's as mitochondrial markers it was shown that the disease is transmitted by virus particles and ds-RNA only (there is no evidence of the presence of plasmids in A. bisporus). Because the viral complex could be reisolated from the acceptor strain which showed the disease symptoms, all the postulates of Koch were met as closely as possible. If no unknown cytoplasmic factors are involved, these experiments prove that La France disease is caused by, and not only associated with, the viral complex as described above.

Introduction

The main symptoms of La France disease or Die-Back of the white button mushroom Agaricus bisporusar e a dramatic reduction in yield and the formation of abnormal fruiting bodies. The disease is always accompanied by the presence of isometric virus particles (Van Zaayen, 1979) and 10 ds RNAs (Harmsen et al, 1989). Therefore it is generally assumed that the disease is of viral nature. However, direct evidence for viruses being the causative agent is still lacking. This is mainly due to difficulties in isolating and purifying intact viral particles and their subsequent use for infection of protoplasts. The disease is usually spread by infected spores. It is assumed that a healthy crop is infected by hyphal anastomosis between germinated infected spores with uninfected mycelium. In an attempt to prove that the viral complex, i.e. ds RNAs and viral particles, indeed is the causative agent for the disease symptoms, we used this natural way of transmission of the disease. By using nuclear and mitochondrial markers, we showed that the disease is transferred by the transmission of virus particles and ds RNA's only.

Results and discussion

For the anastomosis experiment between healthy and diseased strains, Horst® U3 was used as a donor strain and Horst® Ul as acceptor strain. U3 was derived from a tissue culture of a fruiting body removed from a crop that showed all the symptoms of La France disease.

109 The tissue culture contained 10 distinct ds RNAs and 2 types of isometric particles (25 and 34 nm in diameter). The acceptor strain Ul contained only one ds RNA (L6), normally found in healthy strains in The Netherlands (Harmsen et al., 1989). Both strains were inoculated on Petri dishes filled with compost. Before use, the compost was sterilized and inoculated with the thermophilic fungus Scytalidium thermophilum. After 3 days of incubation at 45 °C, the fungus was inactivated by shortly heating the compost at 100 °C. This pretreatment stimulates the colonization of the compost byA. bisporus (Straatsma et al., 1989). After the healthy and diseased strains had made contact, small pieces of colonized compost on both sides of the contact zone were removed at different times and different distances of the contact zone. The samples were transferred to 1% Malt extract, 0.5% mycological Pepton plates covered with cellophane. After 14 days of incubation at 24 °C, the mycelia were removed and freeze-dried. From all samples ds RNAs were isolated by batch absorption to cellulose (Harmsen et al., 1989). All samples removed from the U3 side of the contact zone contained all ds RNAs normally associated with the disease (Fig. 1).

Fig. 1. Ds RNA analysis of the diseased donor strain U3 (lane 10 - 18) and the healthy acceptor strain Ul (lane 1 - 9). Samples were removed after 2 days (lane 1- 3; lane 10 - 12), 3 days (lane 4 - 6; lane 13 - 15) or 5 days (lane 7 - 9; lane 16 - 18) after the colonies had made contact. The samples were removed at different distances from the contact zone : lane 1, 4, 7, 10, 13 and 16 at 0.5 cm; lane 2, 5, 8, 11, 14 and 17 at 1 cm; lane 3, 6, 9, 12, 15 and 18 at 2 cm.

The analysis of the samples removed from the Ul side of the contact zone showed that all samples removed after 5 days of contact (0.5, 1 and 2 cm from the contact zone) contained a high concentration of all 10 ds RNAs. Of the samples obtained after 2 and 3 days, only those that were removed 0.5 cm from the contact zone contained a high concentration of all ds RNAs. In the samples removed 1 and 2 cm from the contact zone only a low or hardly detetectable concentration of the disease specific ds RNAs was found.

110 The analysis shows that the concentration of the ds RNAs in the Ul samples depends on the time of contact with the diseased strain U3 and the distance of the sample point to the contact zone. This means that the ds RNAs are migrating slowly through the residient hyphal cells. In an attempt to find nuclear markers for the donor and acceptor strain, several RFLP markers were tested. None of these markers could discriminate between the Ul and U3, indicating that these strains are genetically highly related. Only one isoenzyme, i.e. a-esterase, was found that exhibited a different banding pattern for Ul and U3. All samples removed from the Ul side of the contact zone showed an a-esterase banding pattern that was identical to the control Ul, indicating that these samples indeed represent the acceptor strain Ul. However, the isoenzyme marker does not permit the discrimination between all four nuclei involved. In order to find nuclear markers for all nuclei involved, four homokaryons, representing the two hybrids, have to be available. The parental homokaryons of Ul (Fl) are still maintained in our collection. The U3, however, is derived from a fertile monosporeculture of a cross between a white and an off-white strain (F2). The parental homokaryons of U3 were therefore obtained by protoplasting U3 mycelium and the isolation of homokaryotic protoclones. Several primers (10-mers) were used to produce random amplified DNA segments (RAPD's) with PCR techniques. With one primer, RAPDs were produced that differ for all four nuclei involved (Fig 2). With these nuclear markers it was shown that the samples removed from the Ul site of the contact zone did not contain nuclei of the donor strain U3.

Fig. 2. Analysis of nuclear genotypes with random amplified DNA's (RAPD). Lane 1 and 9: 123 bp DNA ladder; lane 2 and 3: nuclear types of Ul; lane 7 and 8: nuclear types of U3; lane 4: control Ul; Lane 5: control U3; lane 6: Ul, sampled after 5 days of contact, 0.5 cm from the contact zone.

Ill Fig 3. Analysis of mt genotype. A Southern blot of total DNA, digested with Haelll, was hybridized with a biotinylated mt DNA probe. Lane 1: A.x EcoRI x Hindlll; lane 2: controle Ul, lane 3: Ul, sampled after 2 days of contact at 0.5 cm from the contact zone; lane 4: Ul, sampled after 5 days of contact at 0.5 cm from the contact zone; lane 5: control U3.

Ul and U3 have different mt genotypes (Sonnenberg, this proceedings). By hybridizing a biotinylated mt DNA probe with a Southern blot of total DNA digested with Haelll it was shown that no mitochondria were transmitted between the donor and acceptor strain (Fig. 3). For fructification trials, all samples were subcultured and, before preparation of spawn, re-examined for the presence of ds RNAs. In all samples removed after 5 days of contact between the acceptor and donor strain still high concentrations of all ds RNAs were present. In samples, removed after 2 and 3 days of contact, however, no disease specific RNAs could be detected on agarose gel. This means that the concentration of the ds RNAs is either below the detection level or they are completely absent. The loss of ds RNAs after subculturing has been observed before and may indicate that the expansion of hyphae is faster than the synthesis of new ds RNAs. This is supported by the observation that many protoclones isolated from heterokaryons, containing ds RNAs, do not have detectable levels of ds RNAs. Microscopic observations have shown that most protoplasts are derived from apical and adjacent cells. Three samples, removed after 2, 3 and 5 days from the Ul site of the contact zone were used to prepare spawn and to raise fruiting bodies. The crops were examined for the presence of disease symptoms and fruiting bodies analysed for ds RNAs and virus particles. The fruiting bodies obtained from samples removed after 2 and 3 days of contact contain only one ds RNA (L6) and a low concentration of 25 nm particles. No disease symptoms were observed (Table 1). In fruiting bodies obtained after 5 days of contact, all disease specific ds RNAs and both 25 and 34 nm particles were found. In addition, symptoms were seen typical for La France disease, i.e. fruiting bodies with long bend stipes that easily tumble when thouched by hand.

112 Table 1.Analysi s of fruiting bodies derived from cultures infected via anastomosis.

sample dsRNAs virus particles symptoms

Ul (2d, 1cm)* L6 25 nm no Ul (3d, 1cm) L6 25 nm no Ul (5d, 0.5cm) 1-10 25 + 34 nm yes

a: symbols between parentheses indicate time (days) of sampling after the colonies had made contact and distance from the contact zone

Conclusions

By the use of nuclear and mitochondrial markers, we showed that La France disease can be transferred by the transmission of a viral complex only. The involvement of plasmids is unlikely because until now, no plasmids have been found in A. bisporus(Meye r et al. 1988). Because the causative agents could be reisolated from the acceptor strain which showed the disease symptoms, all postulates of Koch were met as closely as possible. If no unknown cytoplasmic factor is involved, these experiments prove that La France disease is.cause d by, and not only associated with, the viral complex.

Acknowledgement

We like to thank Dr J.B. Anderson, University of Toronto, for providing mt DNA probes. We are also grateful to Mrs José in 't Zandt and Mrs Karen den Hollander for their technical assistance.

References

Harmsen, M.C., L.J.L.D. Van Griensven & J.G.H. Wessels, 1989. Molecular analysis ofAgaricus bisporusdouble-strande d RNA. J. Gen. Virol. 70:1613-1616. Meyer, R.J., W.A. Hintz, M. Mohan, M. Robinson, J.B. Anderson & P.A. Horgen, 1988. Homology of Agaricus mitochondrial plasmids with mitochondrial DNA. Genome 30:710-716. Straatsma, G, J.P.G. Gerrits, M.P.A.M. Augustijn, H.J.M, op den Camp, G.D. Vogels & L.J.L.D. Van Griensven, 1989. Population dynamics of Scytalidium thermophilum in mushroom compost and stimulatory effect on growth rate and yield ofAgaricus bisporus. J. gen. Microbiol. 135:751-759. Van Zaayen, A., 1979. Mushroom viruses. In: P.A. Lemke (Ed.): Viruses and plasmids in fungi. Marcel Dekker, New York. p. 239-324.

113 Hydrophobin Genes in Mushroom Development

J.G.H. Wessels

Department of Plant Biology, University of Groningen, Kerk/aan 30, 9751 NN Haren, The Netherlands

SUMMARY

Schizophyllum commune contains a gene family, [Sc1,3,4], the members of which code for small hydrophobic cysteine-rich proteins (nydrophobins). These genes are silent in young cultures but become transcriptionally active at the time of formation of aerial hyphae and fruit-bodies. In the monokaryon only Sc3 is activated, in the dikaryon Sei and Sc4, which are controlled by the mating-type genes, are also activated, together with a number of other fruiting-specific genes. The hydrophobins are excreted into the medium but in hyphae growing at the surface they are insolubilized within the wall domain. Hyphae only containing pSc3 in their walls form aerial mycelium. Hyphae containing pSd and pSc4 in their walls form fruit-bodies. Apart from the mating-type genes, also the FBF and THN genes control the activities of the hydrophobin genes. Frequently occurring mutations in these controlling genes lead to inability to produce fruit-bodies and/or formation of aerial hyphae.

INTRODUCTION

In contrast to Agaricus bisporus and other edible mushroom species, Schizophyllum commune is genetically well known (for review see Raper, 1988) and produces fruit-bodies within a few days on simple synthetic media. The basic molecular biology and genetic regulation of growth and fruiting in this species has been studied for some time (for reviews see Wessels, 1987, 1991). A transformation system has become available (Munoz-Rivas et al., 1986; Mooibroek et al., 1990) and the cloning of several genes has been accomplished (Dons et al., 1984a; Mulder & Wessels, 1986; Froeliger et al., 1987; Giasson et al., 1989). Our work regarding fruiting genes of S. commune has now progressed to the point that functions can be assigned to some of these genes. The results may be of significance in understanding the biology of mushroom development in general (Wessels 1990).

RESULTS AND DISCUSSION

Differential gene regulation and morphogenesis

Genes differentially regulated during morphogenesis in S. commune were originally isolated by screening a cDNA library made on poly(A)+RNA from a fruiting dikaryon for the presence of RNA sequences which are expressed in this dikaryon but not in a co-isogenic monokaryon which, at the same stage of development, forms aerial mycelium only (Dons et al., 1984a; Mulder & Wessels, 1986). A cDNA library was used because previous experiments had shown that about 5 % of the mRNA mass in such a fruiting dikaryon was unique and consisted of a small number of very abundant mRNAs (Hoge et al., 1982).

114 Table 1. Abundance of hydrophobin mRNAs in various strains of S. commune, correlated with the occurrence of fruit-bodies and aerial hyphae.

Genotype* Fruit- Aerial mRNA abundance* bodies mycelium Sei Sc4 Sc3

A41 B41 _ + ND* 3.6 49.5 A43 B43 - + ND 5.1 63.5 A41 B41IA43 B43 + +/ - 30.6 98.4 41.4 Aeon Boon + +/ - 34.2 96.2 36.6 Aeon Boon fbf - + ND 2.7 84.0 A41 B41 thn - - ND ND 2.8 A43 B43 thn - - ND ND ND A41 B41 thnlA43 B43 thn - - ND ND ND

All strains were co-isogenic except for the genes indicated and grown under the same conditions for 4 days in surface culture. + mRNA abundance is given as percent of total RNA (x 1000). Maximum variation in mRNA values for replicate cultures was 5%. * Not detected, i.e. an abundance of less than 0.5 x 103 %.

Nine cDNA clones were isolated. These clones were used to probe the abundance of the mRNAs at various stages of development. Most abundant at the time of fruiting were the transcripts of the Sei and Sc4 genes which peaked at 0.5% and 3.5% of the total mRNA mass on the fourth day of culture when fruit-bodies were developing. In addition, a transcript from the Sc3 gene was cloned because this transcript, also present in the dikaryon, reached a very high level in the monokaryon (about 1% of the mRNA mass). The emphasis in this review is on these three genes because they are members of a gene family which code for hydrophobic wall proteins to which functions can be assigned (see below). Table 1 summarizes the mRNA levels in co-isogenic strains variously expressing the Sei, Sc4 and Sc3 genes, together with the abilities of these strains to form fruit-bodies and aerial mycelium. The fruiting heterokaryotic dikaryon A41 B41/A43 B43 was synthesized by mating monokaryons with different mating-type genes (A41 B41 and A43 B43). The homokaryotic mimic of the dikaryon {Aeon Bcon) carries constitutive mutations in both mating-type genes (Raper, 1988; Ruiters et al., 1988). Both these dikaryons make few aerial hyphae when fruiting profusely. The recessive mutation thn suppresses formation of aerial hyphae in monokaryons and suppresses formation of both aerial hyphae and fruit-bodies in dikaryons (Schwalb and Miles, 1967; Wessels era/., to be published). The recessive mutation fbf has no phenotype in monokaryons but suppresses fruiting in dikaryons while allowing for abundant formation of aerial hyphae (Springer and Wessels, 1989). The expression of the Sc1 and Sc4 genes is evidently controlled by the mating-type genes and by the THN and FBF genes, and the correlation with morphogenesis suggests that their mRNAs are involved in fruiting. Accumulation of mRNA from the Sc3 gene appears to be controlled by the THN gene only and its expression coincides with formation of aerial hyphae. It should be noted that after 2 days of growth all cultures looked much the same with no aerial hyphae nor fruit-bodies while the mRNAs for all three hydrophobin genes were at a very low level. Nuclear run-on experiments have shown that transcriptional activation is, at least partly, responsible for the

115 accumulation of the mRNAs between days 2 and 4.

Structure of the [Sc1,3,4] gene family

Sequencing of cDNA and genomic clones revealed a remarkable homology between the general structures and the coding sequences of the Sc7, Sc4, and Sc3 genes (Schuren & Wessels, 1990). As shown in Fig. 1, these three genes clearly belong to a family of evolutionary-related genes coding for small homologous proteins each with 8 cysteines conserved with respect to their positions. We call these proteins hydrophobins because they are quite hydrophobic with average hydrophobicity indices of +0.54, +0.59, and +0.90 for pSc1, pSc4, and pSc3, respectively (+0.28, +0.35, and + 0.68 if the putative signal peptides are omitted).

4 \ 49 Sc1 109 aa W \ '/ H2N| 'j COOH 1 II 1 '"1 1 II 1 Sc 4 111 aa w \8/ , H2N i mm. i yyj \ -i COOH i ii i i H i \ 48 / So 3 125 aa ^y \y \y H2N| \ IIMH g| |j H III II H H II i II i 1 II 1

Fig.1 . Comparison of the coding sequences of the S. commune hydrophobin genes. The shaded blocks indicate stretches with identical amino acids or conservative substitutions (small interruptions in the sequences are introduced to obtain optimal alignment). Vertical bars under the blocks indicate the positions of the cysteines. Stars indicate positions of possible /V-glycosylation sites. The positions of introns in the genes is shown by triangles within which the lengths of the introns are given in base pairs. The dashed lines in the amino-terminal parts of the sequences indicate the approximate cleavage sites of the leader peptides. (Data from Dons et al., 1984b and Schuren & Wessels, 1990.)

The putative signal peptides suggest that the hydrophobins are secreted. Indeed, antibodies raised against synthetic peptides, representing the non­ homologous amino acid sequences around the first intron in the Sc7 and Sc4 genes (Fig. 1), do react with proteins in the medium and in the cell wall of cultures that express these genes. Furthermore, it was earlier found (de Vries & Wessels, 1984) that at the time of fruiting the dikaryon excretes a number of unique small proteins in great abundance.

Location and function of the hydrophobins

Proteins corresponding to the cloned Sc3 and Sc4 sequences have now been identified as major components of the medium and the cell walls (to be published). From the radioactivity incorporated from [35S]sulfate into the hydrophobins at the fourth day of growth we calculate that at that time the

116 monokaryon and the dikaryon were directing 5.6% and 8.1%, respectively, of their protein synthesizing activities towards synthesis of these hydrophobins. In the medium the hydrophobins are present as monomers and small aggregates but in the cell walls they are present in a form that resists extraction with hot sodium dodecylsulfate. They can be extracted by cold formic acid but then are still of high-molecular weight. After oxidation of cysteine and cystine to cysteic acid with performic acid, dissociation occurs into proteins with apparent molecular weights of 19,000 (pSc4) and 28,000 (pSc3). No other major proteins were found in these SDS-extracted walls. If pSd was present, its concentration was too low to be detected. Only emerged hyphae contained hydrophobins in their walls; submerged hyphae excreted all hydrophobins into the medium. Apparently, the hydrophobins, which are very small proteins, can be excreted through the wall into the medium. At the substrate/air interface, however, they are deposited in the walls of the hyphae as insoluble complexes. This may provide the wall with a hydrophobic surface and confer to the hyphae the ability to grow into the air.

Regulation and role of hydrophobins in development

The most water-repellent hydrophobin, which is encoded by the Sc3 gene, could be necessary for formation of the individually growing aerial hyphae that make up the woolly surface mat. The two other hydrophobins, encoded by the Sei and Sc4 genes, which are under the control of the mating-type genes, may have the additional role of causing hyphae to adhere to each other during fruit-body formation, e.g. by hydrophobic interactions or formation of interhyphal disulfide bridges. For their growth these emerging structures are completely dependent on translocation of materials from the submerged feeding mycelium. The inactivity of the hydrophobin genes during the early growth phase of the mycelium may thus ensure that at least a minimum amount of assimilating mycelium can be formed before the emergence of aerial structures which draw upon this supportive mycelium. The first developmental switch in the developing mycelium would be the activation of the hydrophobin genes. In dikaryotic colonies which are brought to light the hydrophobin genes Sei and Sc4 (and other fruiting-specific genes) are only activated at some distance from the advancing colony front where fruit-body initials will arise (Ruiters & Wessels, 1989a). The hydrophobins are now excreted into the medium, potentiating hyphae to grow into the air. This developmental activation of the hydrophobin genes also occurs in shaken cultures where formation of aerial structures is prevented (Wessels et al., 1987). The second developmental switch would only occur at the substrate/air interface with the anchorage of the hydrophobins in the wall. Of course, once aerial hyphae have emerged the wall would be the only domain where the hydrophobins could accumulate. The hyphae that make up the fruit-bodies continue to express the Sc7 and Sc4 genes, but less so the Sc3 gene (Ruiters & Wessels, 1989b). In these fruit-bodies the hydrophobins are excreted into the hyphal walls but they could also be present in the spaces between the constituent hyphae.

General significance of hydrophobins

We have preliminary evidence that hydrophobins similar to those characterized in S. commune are of general occurrence in fungi. Recently, a gene responsible for the hydrophobic rodlet layer of Aspergillus nidulans conidiospores was cloned and shown to be homologous to the S. commune hydrophobin genes (W.E. Timberlake, personal communication). Because these

117 genes are apparently conserved between an ascomycete and a basidiomycete, it is probable that they will also be found to be conserved within the basidiomycetes, including cultivated species. Their proper formation would then be crucial to the formation of mushrooms in general. Possibly some of the degeneration phenomena noted in these cultivated species have a genetic basis similar to the frequently occurring mutations in the THN and FBF genes of S. commune which lead to inability of all or some of the hydrophobin genes to become expressed, resulting in a loss of fruiting capacity.

ACKNOWLEDGEMENTS

I thank S.A. Äsgeirsdottir, F.H.J. Schuren, and Dr. O.M.H, de Vries for critically reading the manuscript and for allowing me to use their unpublished results.

REFERENCES

Dons, J.J.M., J. Springer, S.C. de Vries & J.G.H. Wessels, 1984a. Molecular cloning of a gene abundantly expressed during fruiting body initiation in Schizophyllum commune. J. Bacteriol. 157: 802-808. Dons, J.J.M., G.H. Mulder, G.J.A. Rouwendal, J. Springer, W. Bremer & J.G.H. Wessels, 1984b. Sequence analysis of a split gene involved in fruiting from the fungus Schizophyllum commune. EMBO J. 3: 2101- 2106. Froeliger, E.H., A.M. Munoz-Rivas, CA. Specht, R.C. Ullrich & C.P. Novotny, 1987. The isolation of specific genes from the basidiomycete Schizophyllum commune. Curr. Genet. 12: 547-554. Giasson, L., CA. Specht, C. Milgrim, C.P. Novotny & R.C Ullrich, 1989. Cloning and comparison of Aa mating type alleles of the Basidiomycete Schizophyllum commune. Mol. Gen. Genet. 218: 12-11. Hoge, J.H.C., J. Springer, & J.G.H. Wessels, 1982. Changes in complex RNA during fruit-body initiation in the fungus Schizophyllum commune. Exp. Mycol. 6: 233-243. Mooibroek, H., A.G.J. Kuipers, J.H. Sietsma, P.J. Punt & J.G.H. Wessels, 1990. Introduction of hygromycin B resistance into Schizophyllum commune: preferential mythylation of donor DNA. Mol. Gen. Genet. 205: 103-106. Mulder, G.H. & J.G.H. Wessels, 1986. Molecular cloning of RNAs differentially expressed in monokaryons and dikaryons of Schizophyllum commune in relation to fruiting. Exp. Mycol. 10: 214-227. Munoz-Rivas, A., CA. Specht, B.J. Drummond, E. Froelinger, C.P. Novotny & R.C. Ullrich, 1986. Transformation of the basidiomycete Schizophyllum commune. Mol. Gen. Genet. 222: 41-48. Raper, CA., 1988. Schizophyllum commune, a model for genetic studies of the Basidiomycetes. In G.S. Sidhu (Ed.): Genetics of Plant Pathogenic Fungi. Academic Press, London, p 511-522. Ruiters, M.H.J. & J.G.H. Wessels, 1989a. In situ localization of specific RNAs in whole fruiting colonies of Schizophyllum commune. J. Gen. Microbiol. 135: 1747-1754. Ruiters, M.H.J. & J.G.H. Wessels, 1989b. In situ localization of specific mRNAs in developing fruit bodies of the basidiomycete Schizophyllum commune. Exp. Mycol. 13: 212-222. Ruiters, M.H.J., J.H. Sietsma & J.G.H. Wessels, 1988. Expression of dikaryon-specific mRNAs of Schizophyllum commune in relation to incompatibility genes, light, and fruiting. Exp. Mycol. 12: 60-69.

118 Schuren, F.H.J. & J.G.H. Wessels, 1990. Two genes specifically expressed in fruiting dikaryons of Schizophyllum commune: homologies with a gene not regulated by mating type genes. Gene 90: 199-205. Schwalb, M.N. & P.G. Miles, 1967. Morphogenesis of Schizophyllum commune I. Morphological variation and mating behavior of the thin mutation. Am. J. Bot. 54: 440-446. Springer, J. & J.G.H. Wessels, 1989. A frequently occurring mutation that blocks the expression of fruiting genes in Schizophyllum commune. Mol. Gen. Genet. 219: 486-488. Vries, O.M.H, de & J.G.H. Wessels, 1984. Patterns of polypeptide synthesis in non-fruiting monokaryons and a fruiting dikaryon of Schizophyllum commune. J. Gen. Microbiol. 130: 145-154. Wessels, J.G.H., 1987. Mating-type genes and the control of expression of fruiting genes in basidiomycetes. Antonie van Leeuwenhoek J. Microbiol. 53: 307-317. Wessels, J.G.H., 1990. Towards molecular genetics in the basidiomycetes. In: H. Heslot, J. Davies, L. Bobichon, G. Durand & L. Penasse Eds): Genetics of Industrial Microorganisms 90, Proceedings Vol. 2. Société Française de Microbiologie, Strasbourg, p 577-583. Wessels, J.G.H., 1991. Fungal growth and development: a molecular perspective. In D.L. Hawksworth (Ed.): Frontiers in Mycology. C.A.B. International, Kew., p 27-48. Wessels, J.G.H., G.H. Mulder & J. Springer, 1987. Expression of dikaryon- specific and non-specific mRNAs of Schizophyllum commune in relation to environmental conditions and fruiting. J. Gen. Microbiol. 133: 2557- 2561.

119 A DNA SEQUENCE INDUCING MUSHROOM DEVELOPMENT IN SCHIZOPHYLUM

Carlene A. Raper & J. Stephen Horten

Department of Microbiology and Molecular Genetics University of Vermont, Burlington VT 05405, USA

Summary

Mushroom production (fruiting) normally occurs in mated isolates in the basidiomycete Schizophyllum commune. We have isolated a cloned sequence, called FRT1, wich upon transformation into the genome of unmated, homokaryotic isolates induces the de novo development of mushrooms. Thus, FRT1 overides the normal requirement for fruiting of a mating interaction that activates the mating-type genes (MAT-on). It also enhances regular fruiting in MAT-on dikaryons after mating in this fungus. FRT1 was isolated from a homokaryon with a MAT-on phenotype due to constitutively functioning mutations in the mating-type genes. At least one, possibly three, other DNA sequences with similarity to FRT1 have been identified within the genome of origin by DNA-DNA hybridization analyses. These sequences are linked and can be spontaneously and simultaneously excised from the genome with a correlated switch from a fruiting to a nonfruiting phenotype. A sequence adjacent to FRT1, and similar to a previously isolated gene determined to be specifically transcribed during mushroom differentiation, has also been identified. DNA hybridization experiments and DNA-mediated transformation analyses have implicated sequence divergence of FRT1 between different strains of S. commune. A working hypothesis that accomodates the current information has been proposed to explain the role of FRT1 as an important regulating element in the pathway for fruiting in this organism. The possibility that comparable genes may exist in related Basidiomycetes such as Coprinuscinereus an d edible mushroom species is also being explored. Keywords: mushrooms, fruiting, cloning, development, Schizophyllum.

Introduction

Schizophyllum commune is a wood rotting Basidiomycete related to various species of edible mushrooms including Agaricus bisporus, Agaricus bitorquis, Pleurotus ostreatus, Flammulina velutipes and Lentinus edodes. Because of its tractability for laboratory studies, S. commune has been used as a model system for elucidating the genetics, physiology and biochemistry af mating interactions leading to development of the sexual reproductive structures known as mushrooms or fruiting bodies (see Raper, 1978 & 1988; Wessels, 1987; Stankis, Specht & Giasson, 1990, for reviews). Studies in this highly accessible fungus have provided valuable guidelines for comparable studies in commercially important fungi.

120 As in most species of edible mushrooms, fruiting in S. commune is normally under the control of the multiallelic mating-type genes (Raper, 1966). It is known to involve a number of other genes as well as environmental factors (Raper & Krongelb, 1958). With the advent of the techniques of molecular genetics, it has become possible to isolate some of these genes involved in fruiting and examine their activities at the molecular level. A number of genes that are specifically transcribed in the differentiated tissue of fruiting bodies in 5. commune have been cloned and are being characterized (Mulder & Wessels, 1986; Wessels, this issue). These genes, called Sc genes, are thought to encode protein products essential to the differentiation of fruiting tissue. Recently, we have cloned a DNA sequence, called FRT1, that regulates the induction of mushroom development. A preliminary characterization of this sequence and speculations as to how it may interact with other genes for fruiting is presented here.

Results and Discussion

Isolation of the mushroom-inducing DNA sequence FRT1

The clone containing FRT1 was selected from a bank of clones containing random fragments from the haploid genome of a fruiting homokaryon (H9-1) of S. commune. This particular strain contains mutations in the A alpha and B beta mating-type genes (MAT) which result in constitutive functioning for these genes, i.e. this homokaryotic strain mimics the dikaryon in that the developmental pathway of fruiting is "turned on" as a result of the activation of the mating-type genes (MAT-on). The cosmid vector used for preparation of this genomic clone bank contains the S. commune TRP1 gene (for tryptophan synthesis) as a selective marker in DNA-mediated transformation experiments. Cells used as recipients in these experiments were homokaryotic with wild-type alleles for mating (MAT-off) and carried a mutation in the TRP1 gene that resulted in a Trp- phenotype which could be complemented by the selective marker of the cosmid vector. Trp* transformants, selected for growth on medium without tryptophan, were screened for change in developmental phenotype. The clone carrying the FRT1 sequence was recognized by its ability to induce fruiting in a large majority of Trp+ transformants, subcultures of which are shown in Figure 1. The induced phenotype is stable through meiosis, and the FRT1 sequence appears to be active when integrated in various genomic locations. Through a series of mating experiments, integration of cloned FRT1 was shown to enhance the fruiting that normally occurs in the MAT- on dikaryon, suggesting that it plays a key role in the main pathway for fruiting.

FRT1i s one of a family of similar DNA sequences

This isolated sequence has been mapped to a 1.4 kb fragment of DNA within the cosmid clone. When used as a probe against blots of genomic DNAs isolated from the strain of origin and other strains, the hybridization signals observed indicated at least one, possibly up to three additional genomic sequences with strong similarity to FRT1. Hybridization of the FRT1 probe was significantly stronger to DNA from the strain of origin as compared to the genomic DNAs of four other strains tested, including those strains used as recipients in the transformation experiments described above (e.g. strain 72-4, see below).

121 Restriction fragment length polymorphisms were also seen. These observations suggest some sequence divergence, possibly heteroallelism, for the FRT1 loci.

Figure 1. Subcultures of homokaryotic transformants of strain 72-4. The colonies at left were transformed with TRP1 DNA only; the colonies at right were transformed with TRP1 and FRT1 DNA.

FRT1an d its similar sequences are linked within the strain of origin and there is evidence for heteroallelism between strains

FRT1 and its similar sequences within the genome of the strain from which it was isolated were shown to be inherited together. In a sample of 24 progeny (Trp-, MAT-off) from an outcross of H9-1 (Trp+, MAT-on) with strain 72-4 (Trp-, MAT- off) 18 had FRT1 sequences of the H9-1 type and 6 had the more faintly hybridizing FRT1 sequences of the 72-4 type. Two FRT1 sequences, the one we have isolated and studied and another one with strong similarity, are so closely linked that they reside no more than 20 kb apart within the same cosmid clone that was originally isolated - this other sequence has not yet been tested for its ability induce fruiting.

None of the MAT-off progeny from this cross, including those with sequences of the H9-1 FRT1 type, fruited on their own. How can this be reconciled with the fact that cloned FRT1 (derived from H9-1) did induce fruiting when inserted bij transformation into the MAT-off 72-4 genome? It occurred to us that perhaps some degree of heteroallelism at these FRT1 loci might be required for fruiting. A consequence of mating between compatible homokaryons, heteroallelic for the matingtype (MAT) genes, may not only be the turning on of the MAT pathway to establish and maintain a dikaryon, but may also be the pairing of alternate versions of alleles for fruiting. The artificial insertion of cloned FRT1 into a genome containing an alternate version of this sequence may have accomplished what the complementation of two different genomes normally does in dikaryotic association. In order to test this idea, we compared the effects of inserting cloned FRT1 into progeny containing the H9-1 FRT1 type (i.e. into self FRT1) with the effects of inserting the same sequence into progeny containing the 72-4 FRT1 type. Upon transformation with this sequence, all five recipients of the self type tested failed to develop fruiting bodies while all four of the nonself type tested were induced to fruit. This result, together with the observed differential hybridization of the FRT1 probe to DNA of the H9-1 versus 72-4 genomes, supports the concept of a

122 requirement for heteroallelism at this locus to activate fruiting in S. commune.

A spontaneous switch in phenotype from fruiting to nonfruiting in a MAT-on homokaryotic strain is correlated with an apparent deletion of FRT1 and its linked similar sequences from the genome

As mentioned above, the H9-1 strain from which FRT1 was isolated carries mutations in the mating-type genes that render it MAT-on. Although haploid and homokaryotic, it looks like a dikaryon (with clamp connections and two nuclei per cell). It also forms fruiting bodies all by itself without the necessity of mating. Occasionally, isolates of this strain produce sectors that are no longer capable of self fruiting (Figure 2A). Four subcultures of nonfruiting mycelia, each independently derived from the H9-1 strain, were compared with five nonsectored fruiting mycelia of H9-1 for the presence of FRT1, using Southern analyses. Fragments of the digested genomic DNA of all five of the fruiting subcultures hybridized to the FRT1 probe as expected, whereas no hybridization was detected within the genomes of the four nonfruiting derivatives. This indicated the absence of FRT1 and its related sequences in these genomes (Figure 2B). Tests to date indicate that none of the nonfruiting derivatives revert to the fruiting phenotype. These results suggest that FRT1 and its similar sequences were spontaneously excised from the H9-1 genome to render it incapable of eliciting fruiting and that at least one of these sequences, if not all, is/are essential for the development of fruiting bodies. Although essential, FRT1 is not sufficient: activation of the mating- type genes is also required. These two gene systems must act together to initiate fruiting under normal circumstances.

tT.%%m

Figure 2. A, at left, nonfruiting sector (arrow) from MAT-on fruiting homokaryon. B, at right, blot of genomic DNAs of subcultures of MAT-on homokaryon's digested with Hind III and Eco Rl and probed with radiolabeed FRT1 DNA. The first three lanes contain DNA isolated from three independently derived nonfruiting sectors; the remaining five lanes contain DNAs from subcultures of fruiting mycelia.

123 A sequence similar to another isolated gene that is specifically transcribed during fruiting is closely linked to FRT1

A DNA sequence similar to, but not the same as, 5c7 (Mulder and Wessels, 1986) has been shown to reside within 5 kb of FRT1 on our isolated clone. Scl is one of a group of genes that has been shown by these investigators to produce significantly elevated transcript levels in fruiting tissue. In other words, this group of genes appears to be developmentally regulated. It is conceivable that the sequence related to Sc7 may also be regulated during fruiting and that FRT1 plays some role in this. This possibility will be investigated.

Much of the information discussed in this presentation is documented in Horton & Raper, 1991.

A working model to explain FRT1 activity

We propose a testable hypothesis that we feel best accommodates all of the information to date. In this hypothesis, FRT1 is implicated as a key link in the pathway to development of fruiting bodies. It is cast as a gene that is both regulated and regulates. We suggest that FRT1-typ e sequences exist as an allelic series and that each allele is regulated by a linked allele-specific sequence that is, itself, subject to regulation, directly or indirecly, by the activity of the mating-type genes. We further propose that FRT1 regulates the production of fruiting bodies; it may do so by activating, directly or indirectly, other genes for fruiting, such as members of the Sc gene family that are specific to fruiting. In order to reconcile the differential activity of endogenous FRT1 versus our cloned FRT1, we suggest that cloned FRT1 was separated from its allele-specific regulator (a probable repressor element) in the act of cloning. This is conceivable in view of the analysis of our FRT1-containin g clone which located the biologically active sequence within 0.7 kb of one end of the cosmid vector. According to our model, endogenous FRT1 is normally repressed in the absence of MAT activity; MAT-on effects derepression of FRT1 thus permitting expression of this sequence. Cloned FRT1 is expressed in the absence of MAT activity when paired with an alternate allele in a tranformed genome because it lacks its specific linked repressor; it is not expressed when paired with the endogenous version of self in transformants because it is repressed by its allele-specific repressor present in that genome (see below).

MKT-off // allele specific EKEL-off -// ro fruiting ran repressor-on MKT-an > allele specific -// ran-cn —> fruiting EKT1 repressor-off MKP-off—(allele specific repressor — cloned IRTl-an > fruiting of cloned FRFL not present)

124 This model will be tested by isolating and characterizing the putative FRT1 repressor, examining the nature of the sequence related to Sc7, and characterizing the other sequences with similarity to FRT1, not only from the strain of origin, but from other strains (e.g. 72-4) as well. Once the relationships between these genetic elements for fruiting in S. commune are better understood, we will investigate their possible conservation in related edible species.

References

Horton, J.S. & CA. Raper, 1991. A mushroom-inducing DNA sequence isolated from the Basidiomycete Schizophyllumcommune. Genetics, submitted. Mulder, G.H. & J.G.H. Wessels, 1986. Molecular cloning of RNAs differentially expressed in monokaryons and dikaryons of Schizophyllum commune in relation to fruiting. Experimental Mycology 10:214-227 . Raper, CA., 1978. Sexuality and breeding. In: S.T. Chang & W.A. Hayes (Ed.): The biology and cultivation of edible mushrooms. Academic Press, NY. p. 83-117. Raper, CA., 1988. Schizophyllum commune, a model for genetics studies of the Basidiomycotina. In: G.S. Sidhu (Ed.): Genetics of plant pathogenic fungi, Advances in Plant Pathology, Vol.6, Academic Press, San Diego, p. 511-522. Raper, J.R., 1966. Genetics of Sexuality in Higher Fungi. Ronald Press. NY, 283 pp. Raper, J.R. & G.S. Krongelb, 1958. Genetic and environmental aspects of fruiting in Schizophyllum commune. Mycologia 59: 707-740. Stankis, M.M., CA. Specht, & L. Giasson, 1990. Sexual incompatibility in Schizophyllum commune: from classical genetics to a molecular view. In: CA. Raper & D.I. Johnson (Ed.): Developmental systems in fungi, Seminars in Developmental Biology, Vol.1 (3), Saunders Scientific Publications, Philadelphia, p. 195-206. Wessels, J.G.H., 1987. Mating-type genes and the control of expression of fruiting genes in Basidiomycetes. Antonie van Leeuwenhoek, 53:307-316 .

125

STRATEGIES FOR BREEDING AND PREPARATION OF SPAWN TRANSFORMATION STRATEGIES FOR AGARICUS

M.P. Challen, B.G. Rao & T.J. Elliott

Microbiology & Crop Protection Department, Horticulture Research International, Littlehampton, West Sussex, BN1 7 6LP

Summary

Strain improvement inAgaricu sbisporu s isconstraine d byit ssecondaril y homothallic breeding system. Transformation can help circumvent this difficulty by allowing the direct introduction of genes of agronomic value into the mushroom genome. Unfortunately, in this, as in many other areas of its biology, the mushroom has proved intractable. We are attempting to develop a transformation system for Agaricus bisporus and other edible fungi, based on a positively selectable marker. Our approach is currently two-pronged and aims to exploit resistance to two antimetabolites, carboxin and 5-fluoroindole, using the ink-cap Coprinus bilanatus as a model. This speciesi s2-spore d and secondarily homothalliclik eAgaricu s bisporus. Carboxin is an oxathiin fungicide developed for the control of rust diseases but is also particularly toxic to the homobasidiomycetes. A cloned gene for carboxin resistance would seem an ideal selectable marker. The antimetabolite 5-fluoroindole blocks tryptophan synthesis by feedback inhibition and can be cloned directly by exploiting trp. auxotrophy. One from a number of carboxin resistant mutants has been selected on the basis of its expression in the dikaryotic phase and the resistance shown to segregate as a single gene dominant. A genomic cosmid library from this mutant hasbee n screened for the presence of this gene by sib-selection and transformation. A heterologous probe for a carboxin resistant gene from Ustilago maydis has also been used to screen the library. A 5-fluoroindole resistance gene has been cloned from Coprinus cinereus and used to transform C. bilanatus. It is hoped that these studies will lead to the successful transformation of A. bisporus. Keywords: Agaricus. mushroom, transformation, carboxin, 5-fluoroindole.

Introduction

Agaricusbisporus .th e cultivated mushroom,ha sa secondaril yhomothalli c life-cycle with a unifactorial mating-type system (Elliott, 1972;Râpe r et al. 1972). In such life- cycles outcrossing is limited and breeding advance therefore constrained. The development of a DNA mediated transformation system for A.bisporu s may provide the opportunity of circumventing the complexities of the mushroom's breeding system by allowing the direct introduction of novel genetic information into the genome. The development of such a system is dependent on the availability of a selectable marker or markers which can be used to distinguish the transformed from the non- transformed in a large population of cells. Two type of markers are generally used: auxotrophy or resistance. Examples of the use of auxotrophy are tryptophan in Schizophyllum commune

129 (Munoz-Rivas et al. 1986),an d Coprinus cinereus (Binninger et al. 1987)an d adenine inPhanerochaet e chrysosporium (Alice t al. 1989). The difficulty with transformation based on selection for complementing auxotrophy is that it depends on the pré­ existence of mutant strains. In A.bisporu s there are very few auxotrophies identified and in general these strains grow poorly. Resistance markers allowtransformatio n ofwild-type s and are the favoured option for the development of a novel system. Benomyl resistance (Henson^Lâi, 1988) and hygromycin resistance (Punt et al. 1987) have been and are being widely used in a range of fungi. The use of benomyl is inappropriate for A. bisporus which is naturally resistant to high concentrations. A. bisporus is sensitive to hygromycin but vectors based on it haveye t to be made to work in the mushroom. This may be due to the fact that the common hygromycin-based vectors e.g. pAN7-l use promoters derived from Aspergillus (Punt et al. 1987). We have therefore tried to develop transformation in A. bisporus using two novel markers of basidiomycete origin (i) carboxin resistance and (ii) 5-fluoroindole resistance. These studies have been carried out using the two-spored secondarily homothallic ink-cap fungus, Coprinus bilanatus which serves as a genetic model for A. bisporus (Challen, 1988) and is also much more amenable to laboratory manipulation.

Use of carboxin resistance Carboxin is an oxathiin fungicide developed for the control of rusts and smuts. It is also active against homobasidiomycetes. Its mode of action is to interfere with respiration by blocking succinic dehydrogenation. Resistance has been shown to result from the modification of enzymes in the succinic dehydrogenase complex (White, 1971). Studies at HRI(L) have shown that carboxin resistant mutants could be generated in A. bisporus (Challen & Elliott, 1987). In one of these mutants in which the segregation of resistance was tested, both sensitive and resistant progeny were recovered. The predominance of resistant progeny suggested that resistance was determined by a single nuclear gene. The establishment of carboxin resistance in A. bisporus and its known efficacy against basidiomycetes prompted the thought of attempting to clone this resistance gene for use in transformation. Four mutants for carboxin resistance were produced in C.bilanatus . Monokaryons carrying resistance were mated to wild-type to determine dominance. All four mutants formed resistant dikaryons which grew at levels of carboxin three times greater than wild-type. However, when these dikaryonswer e cultured in the absence of carboxin,thre e proved unstable and reverted to monokaryotic growth. The fourth was stable remaining both dikaryotic and resistant. Segregation of this gene, carbr. was studied and itwa sshow n tobehav e as a single gene dominant (Challen & Elliott, 1989). A genomic library was constructed using the DNA from a monokaryotic strain carrying carboxin resistance. Lorist cosmids developed by Gibson et al (1987) have been used to construct genomic libraries in C. cinereus (Mutasa et al. 1990) and were used here for C. bilanatus. These Lorist cosmids use the phage lamda origin of replication and give more uniform copy numbers when compared with Col El based cosmids. They carry kanamycin resistance and the SP6 and T7 phage promoters adjacent to cloning sites which facilitates chromosome walking experiments. A Lorist library which is fully representative of the C. bilanatus genome has been

130 screened for expression ofcarboxi n resistanceb ysib-selectio n and transformation into a carboxin sensitive trp-2" strain of C. cinereus. Putative carboxin resistant transformants have been recovered but none have been shown to contain transforming DNA. Homologous transformation of carboxin sensitive C. bilanatus has also been attempted. No transformants have been identified. Transformants prototrophic for tryptophan auxotrophy have been recovered and confirmed by Southern analyses. The cloning of the trp-2 gene from C. bilanatus using transformation in C. cinereus confirms the validity of this approach and the representative nature of the library. In a further effort to identify the carbr gene in C. bilanatus. a carbr gene isolated from Ustilago maydis has been used to probe the library. We have made the assumption, based on what is known of the genetics of carboxin resistance in Aspergillus (Tuyl, 1977), that our C. bilanatus mutant gene might be functionally- equivalent to the U. maydis gene. Four hybridising clones have been identified but no transformants have so far been obtained using one of these clones. This U. maydis gene gives good transformation frequencies in U. maydis (Hargreaves.e l ai, pers. comm.). We have also attempted to use the U. maydis gene directly to transform C. cinereus again to no avail; a not unexpected result. The reasons for our lack of success in cloning the carboxin gene are not clear and may only be established when the molecular biology of this species has been developed by other means. A similar approach to clone a carbr mutant gene of C. congregatus (Loftus & Ross, pers. comm.) has also been unsuccessful.

Use of 5-fluoroindole (5-FI) resistance

The genetics of Coprinus cinereus is extensively developed and its molecular biology becoming increasingly so (Pukkila & Casselton, in press). The tryptophan biosynthetic pathway iswel lunderstoo d (Fig 1.) and four structural genes,trp-1 .trp-2 . trp-3 and trp-4 have been characterised (Tilby, 1976).

I I I Feedback inhibition | I 5FI -• 5FT

12 3 4 5 Chorismate -» AA -» PRA -• CDRP -• NGP -• TRP \ P / trp-3 trp-4 trp-2 trp-2 \ trp-1 / / * indole7

Fig.1. Tryptophan biosynthesis in Coprinus cinereus. AA anthranilic acid; PRA N-(5'-phosphoribosyl) anthranilic acid; CDRP l-(0-carboxyphenylamino)-l- deoxyribulose-5-phospate; INGPindoleglycero l phosphate: 5FI5-fluoroindole ; 5FT 5-fluorotryptophan. Enzyme steps. 1 anthranilate synthase; 2 N-(5'-phosphoribosyl) transferase; 3N- (5'-phophoribosyl) anthranilate isomerase; 4 indoleglycerol phosphate synthase; 5 tryptophan synthase (Veal & Casselton, 1985).

131 Three of these genes, trp-1. trp-2 and trp-3 have been cloned in C. cinereus (Pukkila & Casselton, in press). There is a mutant of this fungus designated iar 5 which confers resistance to the metabolic inhibitor 5-FI and has been mapped by classical genetics to theHp ^ locus (Veal & Casselton, 1985). Tryptophan synthetase (trp-1) converts 5-FI to 5-fluorotryptophan which acts as a feedback inhibitor of anthranilate synthetase (trp-3) and switches off tryptophan biosynthesis. The iar 5 mutant allele of the trp-3 gene is a dominant mutation giving rise to a feedback resistant enzyme. Transformation based on the use of the cloned iar 5 gene should therefore be possible. To evaluate the possibility of exploiting the iar 5 gene in transformation we have used the following approach. Tryptophan mutants were not available in Coprinus bilanatus so trp-2" and trp-3" mutants were generated by UV-mutagenesis. Trp mutants can be characterised on the basis of known information about the trp biosynthetic pathway, by a combination of auxanography and detection of anthranilic acid excretion (Tilby, 1978). Only trp-3" mutants will be able to grow on anthranilic acid and indole. Trp-2" mutants will not grow on anthranilic acid but will grow on indole and anthranilic acid is not accumulated. Anthranilic acid accumulation is identified by fluorescence under UV. The establishment of trp-2" and trp-3" mutants in C. bilanatus makes possible heterologous transformation between C. cinereus and C. bilanatus. The C. cinereus trp-2 gene has been used to transform the trp-2"mutan t of C. bilanatus (Burrows £l ai, 1990 and Table 1). Transformation frequencies were usually low due to the difficulty of obtaining high numbers of protoplasts. When sufficient protoplasts were available the transformation frequency was 90//igDNA comparable to the levels seen in early transformation experiments with C. cinereus (Binninger et al. 1987). The reciprocal transformation of C. cinereus trp-2"usin g the C. bilanatus trp-2 gene has also been done (Challen and Elliott, unpublished). These heterologous transformations and earlier studies (Casselton & Fuente Herce, 1989) show that the genes of homobasidiomycetes are transferable between species and that 5-FI resistance should be similarly transferable.

Table 1. Transformation of a trp-2" mutant of Coprinus bilanatus with the heterologous trp-2 gene of Coprinus cinereus fBurrow s et al, 1990).

Experiment Number of ^gDNA TRP+ transformants viable protoplasts total /fig DNA 1 7.3 x 104 1.0 7 7.0 2.5 14 5.6 2 3.9 x 104 1.0 11 11.0 2.5 25 10.0 3 21.0 x 104 5.0 27 5.4 4 150.0 x 104 5.0 452 90.4

132 To cloneth eia rS gene,a cosmi dgenomi clibrar y of awild-typ eC .cinereu s strain wasprepare d and the trp 3wild-typ e allele identified bytransformatio n into atrp-3 " auxotrophicstrain . This cloneo f trp-3wa suse d asa prob e toidentif y putative iar 5 fragments from genomic digests of a 5-FI resistant mutant. Hybridising fragments wererescue d from agarose gels and used to transform the trp-3"auxotrophi c strains toprototrophy . Prototrophiccolonie swer ethe nteste d for 5-FIresistance . Theia r 5 gene has now been used to transform C. bilanatus both in our laboratory and by Burrows et al (unpublished). We are currently trying to extend this work to transform A. bisporus which is particularly sensitive to 5-FI.

Conclusions

Transformation of the fungi has a relatively short history and hasye t tob ewidel y applied to the basidiomycetes. There is a pressing need for a transforming vector which is easy to use, easy to select and has general applicability to the basidiomycetes. The work described here is directed towards that end. Introducing the DNA is of itself only part of the story. The integration of the hygromycin gene into the genome of S. commune has been demonstrated but the resistance is not expressed (Mooibroek et al. 1990). Good expression may depend on the use of basidiomycete promoters. Constructs using the hygromycin gene coupledwit h suchpromoter sprovid ea nalternativ e approach toth edevelopmen to f transformation whichi si nadditio nt oth e directclonin go fhomobasidiomycet egene s described here. Afurthe r problemtha tma ylimi ttransformatio n inA .bisporu si sth e difficulty inproducin gsufficien t numberso fviabl eprotoplasts . Protoplastproductio n andregeneratio n isstil lproblemati ci nthi sspecie salthoug h progressha sbee n made inthi sarea . Itma ywel lb enecessar yt odevelo pDN Adeliver ysystem swhic har eno t protoplast dependant. It is only a matter of time, however, before transformation is available in the armoury of the 'mushroomologist'; for the breeder, for use in directed strain improvement and for mushroom scientists, in general, for use in the continuing dissection of the biology of this fascinating species.

Acknowledgements

We wish to thank Prof. Lorna Casselton of Queen Mary & Westfield College, Londonwhos e idea itwa st o developth e useo f 5-fluoroindole resistance. The iar 5 gene was cloned by D.M. Burrows, a doctoral student working under the joint supervision of L.A. Casselton and T.J. Elliott.

References

Alic, M., J.R. Kornegay, D. Pribnow & M.H. Gold, 1989. Transformation by complementation of an adenine auxotroph of the lignin-degrading basidiomycete Phanerochaete chrysosporium.Applie d and Environmental Microbiology 55,406 - 411. Binninger, D.M., C. Skrzynia, PJ. Pukkila & L.A. Casselton, 1987.DNA-mediate d transformation of the basidiomycete Coprinus cinereus. The EMBO Journal 6, 835-840.

133 Burrows, D.M., T.J. Elliott & L.A. Casselton, 1990. DNA-mediated transformation ofth e secondarilyhomothalli cbasidiomycet eCoprinu sbilanatus .Curren t Genetics, 17, 175-177. Casselton, L.A. & A. de la Fuente Herce, 1989.Heterologou s gene expression in the basidiomycete Coprinus cinereus. Current Genetics 16,35-40 . Challen, M.P. & T.J. Elliott, 1987. The production and evaluation of fungicide resistant mutants inth e cultivated mushroom Agaricusbisporus .Transaction s of the British Mycological Society 88,433-439 . Challen, M.P., 1988. Biology and genetics of the secondarily homothallic basidiomycete Coprinus bilanatus nom. prov. MIBiol Thesis, Institute of Biology, London, 170pp. Challen, M.P. & TJ . Elliott, 1989.Segregatio n of genetic markers in the two-spored secondarilyhomothalli cbasidiomycet e Coprinusbilanatus .Theoretica l and Applied Genetics 78, 601-607. Elliott, T.J., 1972. Sex and the single spore. Mushroom Science 8, 11-18. Gibson, T.J., A.R. Coulson, J.E. Sulston & P.F.R. Little, 1987.Lorist2 , a cosmid with transcriptional terminators insulatingvecto r genes from interference by promoters within the insert: effect on DNA yield and cloned insert frequency. Gene 53,275 - 281. Henson, J.M., N.K. Blake & A.L. Pilgeram, 1988. Transformation of Gaeumannomyces graminis to benomyl resistance. Current Genetics 14, 113-117. Mooibroek, H., A.G.J. Kuipers, J.H. Sietsma, P.J. Punt & J.G.H. Wessels, 1990. Introduction ofhygromyci n Bresistanc eint oSchizophyllu m commune: Preferential methylation of donor DNA. Molecular and General Genetics 222, 41-48. Munoz-Rivas, A., CA. Specht, B.J. Drummond, E. Froeliger, C.P. Novotny & R.C. Ullrich, 1986. Transformation of the basidiomycete, Schizophyllum commune. Molecular and General Genetics 205, 103-106. Mutasa, E.S.,A.M .Tymon ,B . Göttgens, F.M. Mellon, P.F.R. Little &L.A . Casselton, 1990. Molecular organisation of an A mating type factor of the basidiomycete fungus Coprinus cinereus. Current Genetics 18,223-229 . Pukkila, P.J. & L.A. Casselton, in press. Molecular genetics of the Agaric, Coprinus cinereus. In:J.W . Benett &L.L . Lasure (Ed): More Gene Manipulations in Fungi. Punt, P.J., R.P. Oliver, M.A. Dingemanse, P.H. Pouwels & C.A.M.J.J. van den Hondel, 1987.Transformatio n ofAspergillu s based on the hygromycin B resistance marker from Escherichia coli. Gene 56, 117-124. Raper, CA., J.R. Raper & R.E. Miller, 1972. Genetic analysis of the life-cycle of Agaricus bisporus. Mycologia 64, 1088-1117. Tilby, M.J., 1976.Tryptopha n biosynthesis in Coprinus lagopus: A genetic analysis of mutants. Journal of General Microbiology, 93, 126-132. Tilby, M.J., 1978. Inhibition of Coprinus cinereus by 5-fluoroindole. Archives of Microbiology 118,301-303 . Tuyl, J.M. van, 1977. Genetics of fungal resistance to systemic fungicides. Mededelingen Landbouwhogeschool, Wageningen, Nederland, 77-2, 1-136. Veal, D. & L.A. Casselton, 1985.Regulatio n of tryptophan metabolism in Coprinus cinereus: Isolation and characterisation of mutants resistant to 5-fluoroindole. Archives of Microbiology 142, 157-163. White, G.A., 1971.A potent effect of 1,4 oxathiin systemic fungicides on succinate oxidation by a particulate preparation from Ustilago maydis. Biochemical Biophysical Research Communications 44, 1210-1219.

134 TOWARDS ATRANSFORMATIO N SYSTEMFO RAGARICU S BISPORUS

John C. Royer and P. A. Horgen Mushroom Research Group,Centr e forPlan t Biotechnology University of Toronto,Erindal e Campus,Mississauga , Ontario, Canada

Summary Experiments aimeda ttransformin g A.bisporu s usinga numbe r ofdifferen t vectors encodingfo r complementation of auxotrophy andantibioti cresistanc e aredescribed . A numbero fpotentia lproblems ,an dfutur e research strategiesar eals opresented . Keywords:transformation , gene-transfer, Agaricus.protoplast . Introduction Selection approaches for genetransfe r systemi nfilamentou s fungi haveemphasize d 3 basic strategies (for review, seeFincham 1989). One Transformation of auxotrophs to prototrophy byth eintroductio n of acomplementar y gene,Tw o Theintroductio n ofa genefo r antibiotico rinhibito rresistanc efro m aprokaryoti co reukaryoti c source. Three Theadditio no f ane w genetha talter sth enutritiona lcapabilitie s ofth efungu s (Fincham, 1989). Anumbe ro fdifferen t techniques havebee nutilize d todelive r the transforming DNAint oth efunga l cell for eventual integration intoth ehost' schromosomes : (a) The mostcommonl y usedmethodolog y isth eincubatio n of thetransformin g DNAwit h protoplasts (cellswhic har etreate dwit henzyme st oremov e theircel lwalls )i nth e presenceo fpolyethylen e glycol (PEG)an dcalciu mchloride . Transformation frequencies ranging from lesstha n 1 transformant per\ig (Turgeo n et al., 1987)t ogreate r than 10^ transformants per (i.gtransformin g DNA (Churchille tal. , 1990)hav ebee nachieved ,(b ) DNA hasbee n successfully introducedint ointac tcell so f afe w speciesi n thepresenc eo f lithium salts (Itoe t al., 1983). Thisprocedur e isrelativel y rapid, howeverth e transformation frequencies areconsiderabl y lowertha nthos eachieve d byth eprotoplas t method, (c) Anotherapproac h thatha sbee n successful bothwit hprotoplast s andwit h walledcell si sth eus eo felectro-injectio n orelectroporatio n (Ward,e t al., 1989). (d) Themos trecen t transformation methodology isth eballisti c gun,whic h shootslate x beads coated with DNAint ocell s (Howe 1988).

Anumbe ro f Ascomycetes havebee ntransforme d withrelativel y littledifficult y using themethodologie s described above. Within thehomobasidiomycetes , transformation of auxotrophs toprototroph y hasbee n achievedi nSchizophyllu m commune (Munoz-Rivas, et al.,1987) , Coprinuscinereu s (Binninger,e t al., 1988)an dPhanerochaet e chrysosporium (Alice t al.,1989,1990) . Procedures utilizingdominan t selectable markers inbasidiomycete s haveemerge donl yrecently . Wange t al. (1988) transformed theheterobasidiomycet e Ustilagomaydi swit h avecto rcontainin gU .maydi sregulator y sequences fused toth ehygromyci nresistance gene . Theectomycorrhiza l fungus Laccaria laccatawa stransforme d tohygromyci nresistanc e (Barrete tal. , 1990)usin gpAN7 , which contains Aspergillus nidulans regulatory sequencesPunt ,e t al. 1987). Mooibroek et al. (1990) showed thatpAN 7coul d becotransforme d intoS .commune . Expressiono f hygromycin resistance was severely depresseddu e tomethylatio n of theforeig n DNA. Thismethylatio n couldb ereduce db yincorporatin g asequenc eo f S.commun eDN Aint o the transforming vector. Finally,Casselto n (pers.comm. ) has achieved successful transformation of C.cinereu s toresistanc et o5-fluoroindol e using amutan t anthranilate synthase gene.

135 Thedevelopmen to f atransformatio n systemwoul d greatly expandth epotentia l for genetic improvemento f A.bisporus . In addition,i twoul d beinvaluabl efo r studieso f thebasi c biology ofth eorganism , andcoul db eusefu l for theproprietar y markingo f mushroom strains. Wehav epursue d apositiv e selection scheme becausemutation s have beenextremel y difficult togenerat ei nA .bisporu s (Râpere t al., 1972),an d because transformation wouldb emos tusefu l ifi tcoul db eapplie dt oprototrophi ccommercia l strains. Wehav e stressedth eprotoco l involving PEG/CaC12treatmen to f protoplasts sincethi si sth emos tcommo n and successful methodfo r other fungi. Wehav eattempte d totransfor m protoplasts of A.bisporu swit h anumbe r of different vectors. Todate ,w ehav e notbee n ablet oconfir m successful transformation. In this paperw edescrib eth etransformatio n procedures andth evector s thatw ehav eutilize das , wella spotentia lproblem si ntransformation , andou rcurren t andfutur e research strategies. Results andDiscussio n Transformation protocol Thetransformatio n procedure thatw ehav eattempte di sbase do ntha tuse d for Chryphonectria parasitica (Churchille tal. , 1990). Thisprocedur e isquit e standard,an d resultsi nprotoplast s thatexhibi t someclumping ,bu tappea rt ob ehealth y when viewed underth emicroscope . Thetransformatio n reactions areperforme d in5 0m l polypropylene tubes,an dcentrifugation s aredon eo na clinica lcentrifug e (IEC). Approximately 10°protoplast s (seeRoye re tal. ,thi svolume ) areresuspende d in50 0 (J.1 of STC [Sorbitol (1.0M) ,Tri s (0.02M ,p H7.5) , CaCl2 (0.05 M)]. DNA (10-100 ng) isadde dan dth eprotoplast s areincubate do nic efo r 20min . Anequa lvolum eo f polyethylene glycol4,00 0(BDH ,2 0% )i nT C (.02 MTris ,p H7.5 , .05 MCaCl2 )i s mixedin ,an dth eprotoplast s areincubate dfo ra nadditiona l 20mi no nice . The protoplast-PEG suspension is sequentially diluted andmixe dwit h 1,5an d 30m lo f 0.6M sucrose andcentrifuge d for 30mi na tsettin g2 o n aclinica lcentrifuge . The protoplasts areresuspende di n 1 mlo f CYMS[Complet eYeas tMediu m(Râpe re tal. , 1972),containin g0.0 6M sucrose]an dadde dt oa 50 0m lflas k containing 50m lo f CYMS. After 2days ,th eregenerate dprotoplast s areconcentrate d bygravit y filtration on 1.2 \ifilters . Theregenerate dprotoplast s arerelease dfro m thefilte r by swirling iti n3 m l ofCYM ,an dth eresultin g suspension isplate do nCY Mplu s selectionusin g a5 m l overlay of CYMcontainin g 1%lo wmeltin gtemperatur eagarose .

Antibioticrequirement s Wehav eutilize d hygromycin B (Sigmachemica l Co. St.Loui sMO ,USA) ,G41 8 (Geneticin ,Gibco ,Gran d Island, N.Y. ,USA )an dPhleomyci n (Cayla,Cedex , France) and5-fluoroindol e (Sigma)i nou rselections . Experiments haverevealed 2 importan t observationsregarding th e antibioticsuse di nthi s research. First considerable variability inresistanc et oth eantibiotic sexist s amongprotoplasts ,eve n thosegenerate d from a homokaryon. Second,th econcentratio n of antibioticrequire dt oinhibi tgrowt hi s dependent upon theconcentratio n ofprotoplast s or(protoplas tregenerates ) appliedt oth e solidmedium . Aconcentratio n of 30H-g/rn lo feithe rG41 8o rhygromyci n willinhibi t growtho f mostprotoplas tregenerates . However, wefin d considerable background growth atconcentration s ashig h as 150Jig/m lwhe nth eregenerates fro m asingl e transformation experiment (between 1 and5 x l(foregenerates ) areapplie dt oa singl e plate. Weroutinel yus e20 0ng/m lo fG41 8o rhygromycin . Even atthi sconcentration , it isessentia lt oinclud ea contro lo f untransformed protoplasts toavoi dpickin g up false positivetransformants . Aconcentratio n of 10|0.g/m lphleomyci n appearst ob e sufficient toinhibi t growth of A.bisporus .

136 Vectors Table 1 listsal lo fth evector stha tw ehav eutilize dt odat ei nou r transformation attempts with A.bisporus . Wehav eattempte d tocomplemen t theauxotrophi c mutation inA g 1-1 with2 j genesfro m theadenin ebiosyntheti c pathway, and 1 genefro m theuraci l Table 1. Vectorsteste dfo r transformation.

Vector Description Reference pIT221 hph/1yeas tpromote r Queenere t al.,198 5 pPS21 hph/Cephalosporiu m promoter Skatrud et al., 1985 pPS57 ht)h/Pénicilliu mpromote r Skatrud, unpublished pCM54 hph/Ustilag opromote r Tsukuda et al., 1988 p3Sr2 A.nidulan s acetamidase gene Kelly &Hynes , 1985 pADE2 S.commun e genes Froeliger et al., 1987 pADE5 pURAl pAN7 hph -A.nidulan spromoter / Punt et al., 1987 terminator pAN8 phleomycin resistancegene - Punt, pers. comm. A.nidulan s promoter/terminater pura4-D18/HP kan^.CaM vpromoter / Gmunder &Kohli , terminater 1989 pura4-D18/ABD kan.CaM v promoter/ Gmunder &Kohli , terminater 1989 pDBl,2 5-fluoroindoleresistance , Casselton, unpublished C.cinereu s gene pGAGl.2 pura4-D18/HP+ Agaricu s Royer, unpublished repetitiveDN A sequences Promoter hph &ka n/Agaricu s fragments Royer &Horgen , library unpublished pEM/kan pEMsequences/ka n Horgen, unpub. lhph =hygromyci n resistancegen e ^kan= kanamycin/G41 8resistanc egen e biosynthetic pathwayo fS .commune . AUo f theothe rvector steste dencod e forpositiv e selection. Twoo f thevector s areexclusivel y offunga l origin. These arep3SR2 ,whic h isa gen e from A. nidulans encoding growth onacetamid e (Kelly andHynes , 1985),an d pDBl which isderive d from C.cinereu s andallow s growth on5-fluoroindole . Wehav e obtainedan dteste da numbe ro fconstruction scontainin ga bacteria l genefo r hygromycin resistancefuse d withregulator y sequences from variousAscomycetes ,an d 2vector s containingcauliflowe r mosaic viruspromoter s fused toa bacteria l geneencodin g resistance toG418 . Wehav e successfully transformed Ophiostomaulm iwit h severalo f these vectors (pPS57,pHlS , pPS57,pIT221 , pAN8,an d pura4-D18/HP), confirming that thegene s arefunctiona l (Royere t al., 1991,unpublishe dresults) . pCM54contain s a Ustilagomaydi spromote r fused toth ehygromyci n resistance genea swel l asa sequence which confers autonomous replication inU .maydi s (Tsukuda et al.,1988) . Wehav econstructe d severalvector scontainin gbacteria l antibioticresistanc egene s andvariou s sequencesfro m Agaricus. Apromote r libraryo f A.bisporu s sequenceswa s constructed indi epromoterles s vectorpIT123 . Inaddition ,w ehav eclone d2 different

137 repetitive DNA sequences from A.bispom san dA .bitorqui sint opura4-D18/H P (pGAGland2). Wehav e attempted transformations usingpDB2 ,pD18/HP ,an dpGAG l by electroporationwit h aBioRa dGen ePuiser . Protoplasts (10^)an dDN A (25(lg )wer e resuspended in sorbitol (1.0M) ,Tri s (0.05M ,p H7.5) , and CaCl2 (0.005M )an d pulsed at2 5\iFd ata rang eo f0 to 1200Volts . Protoplasts lostviabilit y at approximately 1000Volts ,dependin g on theconcentration .

Potentialbiologica lblock st otransformatio n andfutur e strategies Theinabilit y totransfor m Agaricusi slikel ydu et oon eo rmor eo f the following problems:(1) ,th einabilit yo f thetransformin g DNAt oente rth ecell ,(2 )th einabilit yo f thetransformin g DNAt ointegrat eint oth echromosome ,o rt oreplicate ,o r(3 )th e inabilityo f thetransformin g DNAt ob eexpressed ;eithe rdu et olac ko frecognitio n ofth e promoter sequenceso rmodificatio n ofth eforeig n DNA. Theobserve d lowrat eo f recombination in A.bispom sfrui t bodies (Summerbelle t al., 1989)an dth e methylation problems with S.commun e (Mooibroek et al.,1990 ) suggesttha tbot h (2)an d (3)coul d bepotentia lproble mareas . Asystemati cexaminatio n of thepotentia lproblem si nA .bisporu s transformation shouldb eundertaken . Weca ntes tfo r transient expression ofreporte r genes such asB - glucuronidase orGU S (Jefferson et al., 1986).I f weca n achieveexpressio n ofGU S constructsi nA .bisporus .w eca nconfir m thatth eDN Aca nente rth ecel lan dtha tth e regulatory signalsca nb erecognized . Ifw ecanno tsho wtransien texpressio n of theDN A using standardtransformatio n procedures,i tma yb eworthwhil e testing amor enove l method, such asth e "genegun "t ointroduc e theDNA . Thougha dominan t selection schemewoul db emos tusefu l forAgaricus .i tma yb e worthwhilet odevelo p asyste m involvingcomplementatio n of anauxotrophi c mutant. Anauxotrop h couldb egenerate d anda complementin g genefro m A.bisporu scoul db e isolated bytransformatio n with acosmi dlibrar y of thefungus . The methodologies developed could bedirectl y appliedt oothe rselectio n schemes usingdifferen t vectors. In addition, anauxotrop h complementation procedure wouldallo wcotransformatio n of drug resistance markers,an da nunambiguou s assessment of theirpotentia l for expressioni n Agaricus. If weca nintroduc e avecto rtha t willtransientl y express inAgaricus .w ewil lnee dt o concentrateo nmethod st oallo wit sintegratio n intoth efunga l chromosome. Wehav e recently devised anove lapproac h which wecal l "HelperTransformation" . If weassum e thatrecombinationa lenzyme s aredeficien t in A.bisporus .an dtha trecombinationa l enzymes willfunctio n across species,w ema yb eabl et otransfor m Agaricususin gth e recombinational systemo fanothe rfungus . Ametho do f selectingagains tth esecon d fungus, such asa differenc e inantibioti c sensitivityo rnutritiona lrequirement is a prerequisitefo r thistechnique . Theascomycet eO .ulm itransform s extremely efficiently in ourhands . Thisfungu s isextremel y sensitive tobenomyl ,a compoun d which hasn o effect on Agaricus athig hconcentrations . Wear ei nth eproces so f performing experiments in which protoplasts from A.bisporu s andO .ulm i aremixed , transformed with ahygromyci n resistance gene,an dplate do nmediu mcontainin g both hygromycin andbenomyl . Results of such experiments are forthcoming. References Alic, M.,E . K. Clark, J. R. Kornegay &M .H . Gold, 1990. Transformation of Phanerochaete chrysosporium andNeurospor a crassa withadenin ebiosyntheti c genes fromSchizophvllu m commune. Curr.Genet .17:395-311 . Alic, M, J. R. Kornegay, D.Pribno w &M .H . Gold, 1989. Transformation by complementation of anadenin eauxotrop h ofth elignin-degradin g basidiomycete Phanerochaete chrysosporium. Appl.Environ . Microbiol.55:406-411 .

138 Barrett, V.,R . K.Dixo n &P . A.Lemke , 1990. Genetic transformation of amycorrhiza l fungus. Appl.Microbiol .Biotechnol . 33:313-316. Binninger, D. M., C. Skrzynia, P.J . Pukilla, L. A. Casselton, 1986. DNA-mediated transformation ofth ebasidiomycet e Coprinuscinereus . EMBOJ . 6:835-840. Churchill, A. C.L. ,L . M. Ciufetti, D.R .Hansen , H.D . Van Etten &N . K. Van Alfen, 1990. Transformation of thefunga l pathogen Cryphonectriaparasitic awit h avariet yo f heterologous plasmids. Curr. Genet. 17:25-31. Fincham, J. R. S., 1989. Transformation in Fungi. Microbiol. Rev. 53:148-170. Froeliger, E., A. M. Munoz-Rivas, C. A. Specht, R. C. Ullrich & C.P . Novotny, 1987. Theisolatio n of specific genesfro m thebasidiomycet e Schizophvllum commune. Gurr. Genet. 12:547-554. Gmunder,H . &J . Kohli, 1989. Cauliflower mosaicviru spromoter s direct efficient expression of abacteria l G418resistanc e genei n Schizosaccharomvces pombe. Mol. Gen. Genet.220:95-101 . Howe, C.J. , 1988. Organelle transformation. T. I.G . 4:150-152. Ito,H. , Y.Fukuda ,K . Murata, and A.Kimura , 1983. Transformation of intact yeast cells treatedwit h alkalications . J.Bacterid . 153:163-168. Jefferson, R. A., S. M.Burges s &D .Hirsh , 1986. B-glucuronidasefro m Escherichia colia sa gene-fusio n marker. Proc.Natl .Acad . Sei.US A 83:8447-8451. Kelly,J . M.& M .J . Hynes, 1985. Transformation of Aspergillus niger by theamd s geneo f Aspergillus nidulans. EMBOJ .4:475-479 . Mooibroek, H., A. G. J. Kuipers, J. H. Sietsma, P. J. Punt &J . G. H. Wessels, 1990. Introduction of hygromycin Bresistanc eint oSchizophvllu m commune:Preferentia l methylation of donorDNA . Mol.Gen .Genet .222:41-48 . Munos-Rivas, A., C. A. Specht, B.J . Drummond, E. Froeliger, C. Novotny &R . C. Ullrich, 1986. Transformation of thebasidiomycet e Schizophvllum commune. Mol. Gen. Genet. 205:103-106. Punt, P. J., R. P. Oliver, M. A. Dingemanse, P. H. Powels & C. A. M. J. J. van den Hondel, 1987. Transformation of Aspergillus basedo nth ehygromyci n Bresistanc e markerfro m Escherichiacoli . Gene56:117-124 . Queener, S. W., T. D. Ingolia, P. L. Skatrud, J. L. Chapman & K. R. Raster, 1985. A systemfo r genetictransformatio n ofCephalosporiu m acremonium. Microbiology 1985. American Society for Microbiology, pp.486-472. Raper, C.A. ,J . R.Rape r& R .E .Miller , 1972. Genetic analysiso f thelif e cycleo f Agaricus bisporus. Mycologia 64:1088-1117. Royer, J. C, K. Dewar, M. A.Hubbe s &P . A.Horgen , 1991. Analysis of a high frequency transformation systemfo r Ophiostomaulmi .th ecausa l agento fDutc hel m disease. Mol.Gen .Genet . 225:168-176. Royer, J. C, W.E .Hint z &P . A.Horgen . Efficient protoplast formation and regeneration andelectrophoreti c karyotypeanalysi so fAgaricu sbisporus .(Thi svolume ) Skatrud, P.L. , S.W . Queener, L. G. Garr, &D .L . Fisher, 1987. Efficient integrative transformation of Cephalosporium acremonium. Curr.Genet . 12:337-348. Summerbell, R. C, A. J. Castle,P . A.Horge n &J . B.Anderson , 1989. Inheritance of restriction fragment lengthpolymorphism s inAgaricu s brunnescens. Genetics 123:293- 300. Tsukuda, Y., S. Carleton, S.Fotheringha m &W .K .Holloman . 1988. Isolation and characterization of anautonomousl y replicating sequencefro m Usûlagomaydis . Mol. Cell Biol. 8:3703-3709. Turgeon, B.G. , R. C.Barber , O.C . Yoder, 1985. Development of a fungal transformation systembase do n selection of sequences withpromote ractivity . Mol.Cel l Biol. 7:3297-3305. Wang, J., D.H . Holden & S.A .Leong , 1988. Gene transfer system for the phytopathogenic fungus Ustilago mavdis. Proc.Natl .Acad . Sei.US A 85:865-869. Ward, M., K.H . Kodama &Lor i J. Wilson, 1989. Transformation of Aspergillus awamorian dA .nige rb yelectroporation . Exper. Mycol. 13:289-293.

139 The Isolation of Two Tandemly Linked Glyceraldehyde-3- phosphate Dehydrogenase Genes from Agaricus bisporus

M.C. Harmsen, J. Scheer, T.A. Schuurs and J.G.H. Wessels.

Department of Plant Biology, University of Groningen, Kerk/aan 30, 9751 NN Haren, The Netherlands

SUMMARY A genomic library of nuclear DNA from Agaricus bisporus was constructed in ^FIX II. By screening under low stringency conditions with the GPD gene from Aspergillus nidulans a strongly hybridizing clone MABU3- 412) was isolation from the library. This clone contained the reading frames of two tandemly linked GPD genes (GPDI and GPD/I) that were separated by less than 0.3 kb. At high stringency conditions these genes showed no cross-hybridization. Only GPD/I was actively transcribed in mycelium and fruit bodies.

INTRODUCTION

Glyceraldehyde-3-phosphate dehydrogenase (GPD or GAPDH, E.C. 1.2.1.12) plays a key role in the second part of glycolysis. The active enzyme is a tetramer composed of identical subunits. In yeast, GPD may comprise up to 5% of the cellular dry weight (Krebs, 1953), whereas 2-5% of the poly(A)+RNA can be GPD mRNA (Holland and Holland, 1978). Thus, the GPD gene represents a constitutively and highly expressed gene. We were interested in the isolation of a "strong" promoter of an endogenous gene from A. bisporus for the development of an efficient transformation system based on expression of a bacterial antibiotic resistance gene. Efficient vectors employing GPD promoter sequences from endogenous genes have been successfully used for transformation of Saccharomyces cerivisiae (Bitter & Egan, 1984) and Aspergillus nidulans (Punt et al., 1987). The amino acid sequences of the eukaryotic GPD genes that have been determined to date all show a high degree of homology as do their nucleotide sequences. Therefore, we anticipated that the GPD gene of A. bisporus could be isolated from a genomic library by hybridization with another fungal GPD gene under conditions of low stringency.

METHODS AND RESULTS

Total DNA was extracted from fruit bodies (of Agaricus bisporus, strain Horst® U3) by conventional methods. Nuclear and mitochondrial DNA were separated by CsCI equilibrium density centrifugation in the presence of bis- benzimide. In a 0.3% agarose gel the total DNA had a length that exceeded 100 kb, which would make it suitable for the construction of a genomic library. For this purpose 20 pg of nuclear DNA was partially digested with Sau3A to yield fragments with sizes between 10 and 25 kb. The first two nucleotides of the Sau3A sites of the genomic fragments were filled-in using Klenow polymerase, dCTP and dTTP, which prevents self-ligation and makes size-fractionation unnecessary. Approximately 0.5 pg of filled-in DNA was ligated into 1 pg of Xho\ restricted ylFIX II DNA (Stratagene) of which the first two bases had also been filled-in with dCTP and dTTP by the manufacturer to prevent self-ligation. The ligated DNA was packaged into

140 phage heads (GigaPack Goldll system from Stratagene) and used to infect E. coli P2392 and LE392. The primary library that was obtained contained at least 100,000 independent recombinants. Assuming an average insert size of 15 kb and a genome size of 3 * 107 bp (Horgen et al., 1984) this would mean that the library represents approximately 100 times the genome, which is sufficient for the isolation of single copy genes with 99.9% probability. Prior to screening the library, Southern blots of digested nuclear DNA from A. bisporus were hybridized to suitable restriction fragments corresponding to the coding regions of the GPD genes from the ascomycete Aspergillus nidulans (pAN5-22, Punt et al., 1988) and the basidiomycete Schizophyllum commune (p121-9, unpublished). Hybridizations were done at 54°C using the method of Church and Gilbert (1984), except that post- hybridization washings were done with 0.2% SDS in 0.1 M sodium phosphate buffer pH7.2 at the same temperature. When using the 1.2 kb Sca\-Sac\ fragment of pAN5-22 only strongly hybridizing bands were observed (Fig. 1b), whereas with the 1.4 kb Pvu\\-Hinó\\\ fragment of p121- 9 additional fainter bands were observed. In both cases, hybrids could still be detected after increasing the washing temperature to 58°C.

A (B ïîiiiiiiiitiii iiiiiiiiiisiïi

kb — •23 23 — .."•' -ÊÊk •• ÉÊk •• . _ 9.4 9.4 ~ — 6.6 6,6 — ] x M; — 4.4 4.4 — M: ••-••. ".;";'* m m 2.3 2.3 v ; : • = 2.0 2.0 - "f "' 'è "..

Fig. 1. Southern analysis of digested nuclear DNA from Agaricus bisporus by hybridisation at 54°C with restriction fragments corresponding to the coding region of the GPD genes of A) Schizophyllum commune and B) Aspergillus nidulans.

Therefore, plaque lifts of the constructed genomic library were hybridized at the same stringency (54°C) as described above with the Aspergillus GPD probe. As a result /1ABU3-412 was isolated (Fig. 2) which has a 15 kb genomic insert of which a 5 kb BamH\ fragment (subcloned in pGEM11Zf(-)

141 as pGBB5) hybridized strongly to the GPD genes from both A. nidulans and S. commune (not shown). AABU3-412

1 kb l_l T EBEE E B E B T I I I I I I J L4 I

B / G SXN S S E \ B pGBB5 L is. TE r GPD\ GPDU

Fig. 2. Partial restriction map of MBU3-412 and of its derived subclone pGBB5 (the 5 kb BamH\ fragment). Enzymes abbreviations: B, BamH\; E, fcoRI; N, Nco\; S, Sa/l; T, Not\ and X, Xho\. Open and closed triangles indicate the start codons and stop codons, respectively, of two putative GPD genes (GPD/ and GPDII).

The complete sequence of the pGBB5 genomic insert was determined. Deduced reading frames predicted the presence of two proteins (GPDI and GPDII), which were not identical, but each containing up to 70% sequence identity in amino acids when compared to known GPD protein sequences. The stop codon of GPDI and the start codon of GPDII were separated by no more than 264 bp. The general structures of GPDI and GPDII were very similar in that both genes had 8 introns located at the same position, while the GPDII gene had one additional intron adjacent to the position of the second intron of the GPDI gene. The introns had sizes of 48 to 65 bp and most of them had the CTNA(T/C)N8.,0AG consensus sequence for lariat formation at the 3' splice site also found in the basidiomycete Schizophyllum commune (Dons et al., 1984) and other filamentous fungi (Ballance, 1986). A CAAT and TATA-like element were found upstream of GPDII in the intergenic region, whereas these seemed to be absent upstream of the GPDI gene. Repeated AC dinucleotide sequences were found in both genes in a region that was located between 30 and 50 bp upstream of the ATG start codon. In general, transcription starts on an AC pair which suggests that, in principle, both genes could be transcribed. Nevertheless, to our surprise we detected a transcript of the GPDII gene only. In mycelium grown on agar plate and in fruit bodies the GPDI gene is apparently not transcribed. It is, however, imaginable that the GPDI gene is expressed during another stage of the life cycle or under different environmental conditions.

142 Experiments are being done to determine the promoter functioning in expressing the GPDII gene and to use this promoter for heterologous gene expression in A. bisporus.

PROBE GPD\ Gmw fMFFMMFF

26S rRN A 18SrRNA

Fig. 3. Transcription analysis of the GPD genes from A. bisporus. Total ssRNA of mycelium (M) or fruit bodies (F) was separated in a formaldehyde- agarose gel (1%) and transferred to a Hybond N+ hybridization membrane. Hybridization at 65°C was with the 2.5 kb BamH\-*Xho\ (GPD\) and the 2.5 kb Xho\-+BamH\ (GPDII) fragments of pGBB5 (cf. Fig. 2).

ACKNOWLEDGEMENTS

We thank Mr. Schuren for providing the GPD gene from Schizophyllum commune (p121-9) and sharing his results before publishing and Mr. Punt from TNO for providing pAN5-22. This research was supported by the Netherlands Foundation for Technical Research (STW) and coordinated by the Foundation for Biological Research (BION).

REFERENCES

Bailance, D.J., 1986. Sequences important for gene expression in filamentous fungi. Yeast 2:229-236. Bitter, G.A. & Egen, K.M., 1984. Expression of heterologous genes in Saccharomyces cerevisiae from vectors utilizing the glyceraldehyde-3- phosphate dehydrogenase gene promoter. Gene 32: 263-374. Church, G.M. & Gilbert, W., 1984. Genomic sequencing. Proc. Natl. Acad. Sei. USA. 81: 1991-1995. Dons, J.J.M., Mulder, G.H., Rouwendal, G.J.A., Springer, J., Bremer, W. & Wessels, J.G.H., 1984. Sequence analysis of a split gene involved in

143 fruiting from the fungus Schizophyllum commune.EMBO J. 3: 2101-2106. Holland, M.J. & Holland, J.P., 1978. Isolation and identification of yeast messenger ribonucleic acids coding for enolase, glyceraldehyde-3- phosphate dehydrogenase and phosphoglycerate kinase. Biochem. 17: 4900-4907. Horgen, P.A., Arthur, R., Davy, 0., Mourn, A., Herr, F., Strauss, N. & Anderson, J., 1984. The nucleotide sequence homologies of unique DNAs of some cultivated and wild mushrooms. Can. J. Microbiol. 30: 587-593. Krebs, E.G., 1953. Yeast glyceraldehyde-3-phosphate dehydrogenase I. Electrophoresis of fractions precipitated by nucleic acid. J. Biol. Chem. 200: 471-478. Punt, P.J., Oliver, R.P., Dingemanse, M.A., Pouwels, P.H. & van den Hondel, C.A.M.J.J., 1987. Transformation of Aspergillus based on the hygromycin B resistance marker from Escherichia coli. Gene 56: 117-124. Punt, P.J., Dingemanse, M.A., Jacobs-Meijsing, B.J.M., Pouwels, P.H. & van den Hondel, C.A.M.J.J., 1988. Isolation and characterization of the glyceraldehyde-3-phosphate dehydrogenase gene of Aspergillus nidulans. Gene 69: 49-57.

144 MAINTENANCE, REJUVENATION AND IMPROVEMENT OF HORST® UI

Gerda Fritsche

Mushroom Experimental Station, Horst, The Netherlands

Summary

The methods for maintenance of strains, used in Horst, are described. To study the phenomenon 'degeneration', abnormal fruitbodies were multiplied by tissue culture. In some cases these cultures produced normal fruitbodies, in other cases the abnormalities were reproduced. Possible causes of this different behaviour are put forward. Horst® Ul can be rejuvenated by renewal of heterokaryons as well as by multispore or monospore cultures. In the first case the two homokaryons forming Ul were inoculated side by side. It was shown that the newly formed heterokaryons had a different performance. Spawn from a well-performing young heterokaryon is already on the market for some time now, under the old name Horst® Ul. Multispore cultures of Ul behaved differently. In several cases, strong yield depressions were observed as well as abnormal fruitbodies. Some monospore cultures have already successfully completed five selection steps. It was tried to combine specific good properties of fast pinning white strains with the good fruitbody quality of Ul by crossing Ul with some of these strains. In the Fl the desired phenotype was not found. Spores were collected from 17 of the hybrids. Only one of the samples showed encouraging results in the progeny (F2).

Introduction

Horst® Ul was the first commercialized product of crossing a 'white' and an 'off- white' strain. It came on the market in 1981, together with Horst® U3. In commercial practice both strains were called 'hybrids' to distinguish them from white and off-white strains. Our methods of breeding Ul and U3 were published in 1981 (Fritsche, 1981).

Maintenance of Ul

Methods used in Horst

Darmycel, the licensee for strains developed in Horst is given new mother cultures every 6 months. The mycelium originates from compost cultures (Fritsche, 1981). It is subcultured with 1.5 to 2 years intervals. Every culture receives a letter code. After the mycelium of a given culture has been checked for good mushroom production and quality, it can be used for inoculating the mother cultures for Darmycel. For safety reasons Ul mycelium is also kept on wheat agar (Fritsche, 1981) and under liquid nitrogen.

145 Experiences with abnormal cultures

As all living matter, mushroom strains are subject to instability. Variations in type of mycelium, yield and shape of the fruitbodies were found in spawn of a great variety of strains and from several suppliers (Van Griensven 1988). When in 1987 abnormalities in strain Ul occurred, tissue cultures were made from the abnormal fruitbodies (plectenchym inoculated). Table 1 shows the results of yield trials with such tissue cultures. G888 is a tissue culture from a fruitbody with a thick stipe and a flat cap, that easily breaks from the stipe. Tissue culture G889 was isolated from normal fruitbodies from the same crop, but from another bed with no visible abnormalities. The yield was normal, while the beds with abnormal fruitbodies produced far less than normal. G898 and G896 also originate from an abnormal and from a normal fruitbody respectively, but both were obtained from another commercial mushroom farm.

i ». » *••*•*• • * v ... • t i J * > **ß' W 1 f ^ - "9 *' w * <;S8H 11 G8V8 Proel 14391

Figure 1. Fruitbodies of tissue cultures from aberrant fruitbodies of Ul (G. Numbers) and commercial spawn of Ul.

As table 1 and fig. 1 show, G888 and G898 produced abnormal fruitbodies. They were similar to those from which they were isolated. The cap breaks off easily, likely due to a cavity where the stipe is connected to the cap. The yield of G888 and G898 was only about half of that of the commercial Ul. G889 and G896 originating from normal fruitbodies produced normal fruitbodies and a yield equal to the control Ul.

146 Table 1. Fruiting trials of tissue cultures (G-numbers) of Ul derived from aberrant and normal fruit bodies collected at two farms (F and H) and Ul stored for 30 2 months under liquid N2 (Y). Each experiment was done in four replicates of 0.2 m

kg/nri fruit body strain origin exp.1 exp.2 morphology

Ul commercial spawn 22.2 21.3 normal G888 F (aberrant) 11.5 11.6 aberrant G889 F (normal) 22.5 19.1 normal G898 H (aberrant) 12.8 11.6 aberrant G896 H (normal) 23.1 21.4 normal Y liquid N2 - 13.6 open veil G913 Y (open veil) - 21.8 normal

Table 1 also shows culture Y originating from Ul, which has a low yield and fruitbodies with 'open veil'. Culture Y originated from compost tube Y that was inoculated with mycelium which was kept under liquid nitrogen from 1984 until 1987. From the four compost tubes inoculated with this mycelium, two produced normally in a yield trial (20.5 and 20.3 kg/m2), whereas the other two produced significantly less (15.3 and 14.3 kg/m2). 'Open veil' only showed up in the latter two cultures. Also in exp.2 the yield of Y was only 13.6 kg/m2, whereas G913, originating from a Y culture with 'open veil', produced only normal fruitbodies and a normal yield, being 21.8 kg/m2 (table 1). The experience, that in some cases obvious defects of a mycelium disappear after subculturing via plectenchym, was also evident with protoclones. Sonnenberg (1985) had isolated numerous protoclones of Ul, all of them with more or less severe 'open veil' and a low yield. It is well known from higher plants that properties of protoclones can differ from the original strain. After vegetative multiplication of the potato strain 'Bintje', the parental type turned up again sometimes (Sree Ramulu et al, 1984). With protoclone P12 it was investigated whether this applies to the cultivated mushroom as well. Multiplications by mycelial transfer and tissue culture were compared. Table 2 shows that multiplication by subculturing mycelium did not lead to improvement. Even after 10 subcultures the yield was still low and 'open veil' still occurred. Tissue cultures behaved differently. These cultures showed a normal yield and only normal fruitbodies.

How can it be explained that defects are sometimes remedied by tissue subculturing and sometimes not? When the abnormal situation is not restored, I think that the cause is degeneration i.e. a stable genetic change for the worst. When it, however, goes back to normal, the cause was weakening of the mycelium as a result of stress. Here, culture Y (table 1) was exposed to strong temperature changes by temporary storage at -196°C. The mycelium of protoclone P12 (table 2) originated from a protoplast and was thus temporarily deprived of its cell wall.

147 Table 2. Fruiting trials of P12 (protoclone: culture derived from a single regenerated protoplast of Ul) and tissue cultures of P12 (G-numbers). a: number between parenthesis indicates times of subculturing of P12. Each experiment was done in four replicates of 1.3 m2.

kg/iTl 2 fruit body Strain origin exp.1 exp.2 morphology

Ul commercial spawn 20.4 19.6 normal P12 protoclone (l)a 12.8 15.8 open veil P12 il (5) 13.6 - open veil P12 11 (10) - 16.8 open veil G864 tissue cult. P12 18.7 20.3 normal G865 n " G864 - 20.2 normal G866 M Il ii - 20.8 normal

Rejuvenation of m

It is generally assumed that aging mushroom strains gradually loose their productivity and become more and more sensitive to degeneration. In this case, younger, equally productive material, should be made available. Therefore, the Experimental Station in Horst has paid much attention to the rejuvenation of Ul.

Renewal of heterokaryons

Ul originates from the pairing of two homokaryons, one descending from a white strain and the other one from an off-white strain. Both homokaryons are in the Experimental Stations' collection. In 1987 they were inoculated again side by side in six culture tubes on wheat agar (Fritsche, 1981). Each tube received a specific letter code. Table 3 shows further performances of the six young heterokaryons (code E- K) compared to the much older heterokaryons from 1978. The growth tests were carried out on compost-agar (Fritsche, 1981). It can be seen that only the growth rate from code H differed significantly from the other codes by growing more slowly. This heterokaryon had also a very low yield. Both properties indicate that no stable heterokaryon was formed. However, this could not be the case with code J, because its growth rate was normal. Yet, its yield, being only 11.3 kg/m2, was very much lower than that of the other heterokaryons.

When the shape (convexity) of the cap was measured (convexity = cap height : cap width x 100), no significant differences were found. Spawn from a young heterokaryon of Ul is on the market since spring 1990.

148 Table 3. Mycelial growth rate, fruit body yield and convexity of the cap of "old" (B to D) and "young" (E to K) heterokaryons Ul derived from individual matings of homokaryons (39 x 97). Ul derived from commercial spawn. Growth rate was expressed as colony diameter (mm) on compost-agar after 14 days growth at 24°C (average of 3 x 10 Petri dishes). Fruit body yield of each heterokaryon was tested in four replicates of 1.3 m2. Of each replicate three flushes (96 fruit bodies each) were used to measure the convexity of the cap (ratio cap heigh/cap width x 100). hetero­ year of colony dia­ yield karyon isolation meter (mm) (kg/m2) convexity

Ul 1978 - 19.6 54 B tl 81 19.5 55 C tt 80 20.0 57 D It 70 19.4 55 E 1987 81 19.9 55 F H 75 20.3 54 G n 80 19.7 56 H H 58 1.1 - J H 81 11.3 56 K H 81 20.5 55

Multispore cultures

Rejuvenation of strains is often done by subculturing multispore isolates. Instead of making a young heterokaryon from old homokaryons, this method has the advantage of generative renewal. As the nuclei have gone through meiosis, new combinations of properties might be expected. This is of more concerns in monospore cultures than in multispore cultures where competition with mycelium from other spores exists and an average phenotypic pattern will result. However, several research workers have assessed that some multispore cultures can strongly differ from parental strains (Sigel & Sinden, 1955; Lambert, 1959; Bretzloff, Robbins & Curme, 1962). We had a similar experience in our own trials with Ul. From each of three spore samples, two multispore cultures were made. Both multispore cultures from one of the three samples yielded significantly less (7.5 and 0.1 kg/m2) than the multispore cultures from the other samples (14.7 till 15.8 kg/m2), which did not differ from commercial Ul spawn. The two low-yielding multispore cultures also showed a slow mycelial growth rate and abnormal fruitbodies. These were very big and firm, discoloured, very scaly and had a cap hardly wider than the thick long stipe. In another series of trials, both multispore cultures from Ul yielded normally, but sometimes yield depression and abnormalities in colour and shape of the fruitbodies still occurred after subculturing the mycelium (Fritsche, 1990).

Monospore cultures

Since 1977 a total of nearly 300 monospore cultures originating from Ul, were isolated and tested. Our original purpose of selecting a monospore culture superior

149 to Ul, was later reversed by finding a new strain equal to Ul. The advantage should primarily be the younger age. As expected, great variability between monospore cultures existed. Many infertile strains were stored for future crossings. Our selection procedures were published in 1981 (Fritsche, 1981). In one year, only two selection steps can be completed. Three monospore cultures have already successfully passed all five selection steps. They are now one by one compared with Ul in a single growing room, each occupying half the room. Another seven monospore cultures are in the fifth selection step (5x6 m2), while an equal number is completing the fourth step (8 x 1.3 m2). Obviously, we succeeded to have young monospore cultures, phenotypical identical to Ul, in stock. At need, these can be put on the market instead of Ul.

Improvement of Ul

Table 4. Mycelial growth rate in casing soil and fruit body characteristics of fertile monospore cultures isolated from two spore samples (QK and QL). The spores were obtained from two fruit bodies derived from two independent matings between different pairs of homokaryons of Horst* Ul and Le Lion B92. a: numbers indicate the degree of colonization of the casing soil ranging from 0 (not colonized) to 5 (strongly colonized), b: numbers indicate the quality of the fruit bodies (size, firmness, smoothness); the numbers range from 1 (small, soft, scaly) to 10 (large, firm smooth). The fruiting trials were done in two replicates of 1.3 m2.

n mono- mycelial spore spore growth in fruit body fruit body fruit body sample cultures casing soil size firmness smoothness

QK 17 range 0.8 - 4.5 4.8 - 7.5 5.7 - 8.0 4.2 - 7.8 means (s.e.) 3.2 (1.1) 6.5 (0.6) 6.8 (0.6) 5.6 (1.0) QL 9 range 1.5 - 4.3 3.8 - 7.0 4.0 - 6.0 6.7 - 8.5 means (s.e.) 3.0 (0.9) 4.9 (0.9) 5.1 (0.6) 7.4 (0.6) controls Horst® Ul 3.3 + 3.8 8.0 + 8.2 8.0 + 8.4 4.0 + 4.4 Le lion B92 4.5 + 4.8 5.0 + 5.7 4.1 + 4.6 8.0 + 8.8

We try to combine the good quality of the fruitbodies of Ul with good properties of white strains, such as easier cropping , faster mycelial growth, early pinhead formation, smooth cap and a long stipe. This is done by crossing Ul with fast- pinning white strains. Among 161 crossings we found no single one meeting our requirements. From 17 cross-products spores were collected and from each sample about 40 monospore cultures were isolated (F2). Of 633 monospore cultures in the first selection step, only 185 were selected for the second step. It was shown that the selections often grew faster in the casing soil than Ul and had a smoother cap. However, they usually did not match the good quality of Ul. Table 4 depicts results of one of the trials (second step of selection, test in 2 x 1.3 m2) and shows the mycelial growth in casing soil. Furthermore the quality of the

150 fruitbodies is given. The monospore cultures from both spore samples varied considerably in all properties concerned. The high value for B92 mycelial growth was only matched by one new monospore culture, whereas the majority grew slower into the casing soil than Ul. No single new monospore culture matched the fruitbody size of Ul and only one had the same firmness as Ul. Most monospore cultures had better smoothness of the cap than Ul, but only three were as good as B92. It was striking, that for some properties all monospore cultures originating from one spore sample were the same. In one case this concerned stroma formation (fig.2) and in another case 'open veil'. Anderson (1989) pointed out already the problems of the low combination rate inA.bisporus and its tendency to harbor some genetic defects. The monospore cultures originating from another spore sample had the same fruitbody quality as Ul, with the advantage of a smooth cap. Nevertheless, their yield was lower. Additional monospore cultures originating from this spore sample were isolated later to obtain high-yielding ones. The results of the first trials are encouraging. m*

Figure 2. A monospore culture with abundant "stroma" formation alongside a normal monospore culture.

Acknowledgements

Discussions on the behaviour of protoclones of higher plants with ir. A. Houwing are greatly acknowledged. I thank also drs. J.P.G. Gerrits for translating the manuscript into English and dr. A. Sonnenberg for advice. The careful assistance of mrs. J. Kuenen-Claes and mr. P. van Loon is greatly acknowledged.

References

Anderson, J.B., 1989. What is new in the genetic engineering world. Proceedings of the 8th North American Mushr. Conference, 85-92.

151 Bretzloff, C.W., W.A., Robbins & J.H., Curme, 1955. Observations of multisporous isolates from the cultivated mushroom Agaricus bisporus (Lange) Sing. Mushroom Science V, 188-196. Fritsche, G.F.G., 1981. Some remarks on the breeding, maintenance of strains and spawn of Agaricus bisporus and A.bitorquis. Mushroom Science XI, part I, 367- 385. Fritsche, G.F.G., 1990. Veredelingswerk met de U-rassen. De Champignoncultuur 34 (3), 115-121. Lambert, E.B., 1959. Improving spawn cultures of the cultivated mushroom. Mushroom Science IV, 33-51. Sigel, E.M., & J.W. Sinden, 1955. Variations in cultures made from the strain of mushrooms used at the Butler County Mushroom Farm Inc. Mushroom Science II, 65-68. Sonnenberg, A.S.M., 1985.Persona l communication. Sree Ramulu, K.P. Dijkhuis & S. Roest, 1984. Genetic instability in protoclones of potato {Solanum tuberosum L.cv.'Bintje'): new types of variation after vegetative propagation. Theor.Appl.Genet.68, 515-519. Van Griensven, L.J.L.D., 1988. Stabiliteit van champignonrassen. De Champignoncultuur 32 (1), 15-19.

152 THE DEVELOPMENT OF A SET OF CHARACTERISTICS FOR D.U.S. TESTS OF CULTIVATED MUSHROOM VARIETIES

A. van der Neut

Centre for Variety Research and Seed Technology (CRZ), Wageningen, The Netherlands

Summary

A number of characteristics were examined for the discrimination between cultivated mushroom varieties of Agaricus bisporus. Only a few were found to meet the requirements for D.U.S. tests: cap border, cap thickness and flushing pattern. The investigated varieties could be arranged morphologically and by esterase isozym analysis in four groups: white varieties, off-white varieties, hybrid varieties of the Horronda type and hybrid varieties of the Horwitu type. Keywords: characteristic, cultivated mushroom, discrimination, identification, plant breeders' rights, variety.

Introduction

The Dutch plant breeders' rights (PBR) are based on the International convention for the protection of new varieties of plants of December 2, 1961 (UPOV = Union internationale pour la protection des obtentions végétales) (UPOV, 1985; Fikkert, 1985). Since then, three times revised, lately on May 19, 1991 (UPOV, 1991). The protection provided for in the UPOV convention is specifically tailored to the needs of agricultural and horticultural industries, and to what were considered to be the needs of the community and its food supply. The scope of protection is limited to the variety (product), the specific assemblage of plant material selected by the breeder which represents the variety. Under the patent system the available scope of protection is limited only by the valid claims of the granted patent (Greengrass, 1990). Patents are granted for the protection of inventions (processes and/or products). Such a invention has to be industrially applicable. novel: the description of the invention shall not be disclosed to the public. subject of a enabling disclosure: it must be so described that a person skilled in the art to which the patent application relates, can reproduce or repeat the invention. inventive stap: it should not constitute a obvious step forward from the existing known technology to a person skilled in the art described in the patent application. In many countries plant varieties are considered not to fulfil one or other of these criteria and thus to be ineligible for patent protection. The UPOV convention provides that protection shall only be granted after examination of the variety. The criteria for granting PBR include novelty: the variety shall not be marketed before the date of application. distinctness (D): the variety has to be distinct from any other variety whose

153 existence is a matter of common knowledge at the date of application. The distinctness is established by examining a set of characteristics. uniformity (U): the variety has to be sufficiently uniform in the relevant characteristics. The uniformity is subject to the variation expected from particular features of propagation. stability (S): the variety is stable, if the relevant characteristics remain unchanged after repeated propagation of the variety. As distinct from patent legislation, the UPOV convention permits the farmer to propagate plant material of a protected variety for his own use. Also a breeder can use freely a protected variety for breeding purposes. Characteristics for D.U.S. testing must allow precise recognition and description (UPOV, 1979). They does not necessarily refer to the commercial value of the variety. For many crops there are UPOV guidelines with tables of characteristics and their testing criteria, but for cultivated mushrooms such guidelines are lacking. Till now, only a few UPOV member states (Hungary, Japan (A. bisporus), The Netherlands and United States of America) provide protection of cultivated varieties of the genus Agaricus (UPOV, 1986). Since Dutch PBR were open to cultivated mushrooms {Agaricus),eleve n applications have been submitted: four for A. bisporusvarieties , six for A. bitorquisvarietie s and one for an A. arvensis variety. Thus far, there are PBR for three A. bitorquisan d twoA. bisporusvarieties . The cultivated mushroom turned out to be a complicated crop in variety research. In the Netherlands given variety descriptions have led to legal procedures (Anon., 1984). It is therefore necessary to develop a set of unequivocal diagnostic characteristics.

Materials and methods

The experiments have been carried out in 1987, 1988 and 1990. The mushrooms were grown in fresh compost of the Netherlands Cooperative Mushroom Growers' Association (CNC) in units of 1,3 m2 (130 kg; 1987) and 0,2 m2 (22 kg; 1988, 1990), CNC casing soil of 4 cm (1987), respectively 5 cm (1988, 1990) thickness. The units were inoculated with 650 ml (1987) and 110 ml (1988, 1990) commercial spawn. After pinning the average temperature of air and compost was kept at 19 °C. Each variety was grown in four to six replicates. The flushing patterns were determined by picking flats every day, when they appear in that physiological stage. The morphological characteristics were observed using samples of 40 mushrooms of each replicate. The obtained data were analyzed using statistical programs of Genstat and/or SPSS-X. A significance level of 1% was used. Agaricus bisporus was used in three experiments. In the first experiment: the white varieties Le Lion B92, Somycel 53 and Somycel 91, the off-white varieties Le Lion B86 and Somycel 76, the hybrids Horronda and Horwitu. In the second experiment: the white varieties Somycel 53, Somycel 91, Royal 70 and Royal 75, the off-white varieties Le Lion B86, Le Lion B62, Somycel 76, Claron A3.01, Royal 20A and Royal 29A, the hybrids Horronda, Horwitu, Le Lion XI, Le Lion X13, Claron AX30, Claron AX60, Royal 21A, Royal 23A and Royal 24A. In the third experiment: the hybrids from white and off-white varieties Horronda, Horwitu, Le Lion XI, Le Lion X13, Le Lion X20, Somycel 112, Somycel 205, Somycel 208, Claron AX30, Claron AX31, Claron A5.1,Claro n A5.3, Royal 26A, Le Champignon 102A, Le Champignon 200, Le Champignon 222, Euro-Semy 170 and

154 Euro-Semy 285. A survey of investigated characteristics is given in table 1.

Table 1. Characteristics investigated in the experiments 1, 2 and 3 (+investigated; -=not investigated).

Characteristics 1 2 3 Cap colour + - - Scale colour + - - Stipe colour + - - Gill colour + - - Fading of cutting surface + - - of longitudinal section Attachment of gills near the stipe + - - Cap shape (longitudinal section) + - + Stipe shape (longitudinal section) + - + Stipe base + - - Scaliness + + - Stipe structure (longitudinal section) + + - Weight of fruiting body + + - Central hole of the cap + + - Cap diameter + + - Cap thickness + + + Stipe length + + - Stipe diameter + + - Firmness of button cap + - - Firmness of button stipe + - - Position of veil on the stipe (button) + - - Position of veil remnant on the stipe (flat) + - - Nature of veil remnant on the stipe (flat) + - - Cap border (flat) + + + Cap diameter/cap thickness ratio + + - Cap diameter/stipe length ratio + + - Cap diameter/stipe diameter ratio + + - Cap thickness/stipe length ratio + + - Cap thickness/stipe diameter ratio + + - Stipe length/stipe diameter ratio + + - Stipe length/length stipe base to veil + - - remnant ratio Flushing pattern + + +

155 Results and discussion

Experiment 1

The statistical analyses indicated that there were effects for flush, variety and flush*variety, but not for replicates. Therefore the results were studied for both flush and variety separately. In the first experiment both buttons and flats were studied. In the button stage the fruiting body rapidly grows to the next stages by stipe elongation and cap expansion (Bonner et al, 1956). Characteristics observed in this stage were influenced by samples with fruiting bodies which did not fulfil the correct physiological requirements. Often the stipe was already elongating or the veil was breaking. For that reason, in the other two experiments only flats were observed (table 1). The RHS Colour Chart {RHS, without year) was used to determine colour characteristics. The colour chart contains as little as five whitish hues and few brownish hues, which were insufficient to distinguish whitish mushrooms. The colours of cap, scales, stipe and gills seemed to be dependent of the developmental stage of the fruiting body. The use of a colorimeter which quantifies colours offers good opportunities for colour detection, but needs further research. Fading along longitudinal section of a fruiting body, attachment of gills near the stipe, nature and position of veil and veil remnant on the stipe (annulus) did not discriminate between white, off-white varieties and their hybrids. The firmness of button cap and stipe were not useful as characteristic, because the used method of pressing the button between fingertips was unreliable. It was not possible to discriminate the varieties by shape of cap, stipe and stipe base in longitudinal section of the fruiting body. Only a very slight variation was visually observed both within and between varieties. In experiment 3 these characteristics have been again analysed by image analysis. Except stipe length/length stipe base to veil ratio, the remaining characteristics seemed distinctive, but a second examination was necessary to establish absence of experiment*variety interaction.

Experiment 2

The characteristics which seemed to be distinctive in experiment 1, were investigated (table 1). By means of the varieties which were examined in both experiments, the experiment*variety effect of each characteristic was calculated. This experiment revealed that the weight of the individual fruiting body was not a helpful distinctive characteristic. The variance of most of the varieties was much higher than the overall variance and the coefficient of variation was very high (34% to 93%). Scaliness of the cap was qualitatively measured within the categories 'smooth to very few scales', 'medium' and 'scaly'. The characteristic is useful to distinguish the white varieties (smooth to very few scales) from the off-white varieties (scaly), if grown under the same conditions, but the hybrid varieties showed much variation between flushes and both experiments. Environmentally induced variation, especially in case of hybrid varieties, makes the characteristic unsuitable for PBR testing.

156 Table 2. Outline of white, off-white varieties and their hybrids, based on the three characteristics cap border, cap thickness and flushing pattern (experiments 1 and 2).

Cap Cap Flushing Flush 1 Flush 2 Flush 3 Border Thickness Pattern

Smooth Thick Late Somycel 53

Medium Early Le Lion B92 Le Lion B92 Royal 70 Royal 70 Royal 75 Royal 75 Somycel 91 Somycel 53 Somycel 91

Medium Le Lion B92 Royal 70 Royal 75 Somycel 91

Late Somycel 53

Partly Medium Medium Claron AX60 Frayed Horwitu Le Lion X13 Royal 23A Royal 24A

Frayed Thick Very Claron AX30 Claron AX30 Claron AX30 Late Horronda Horronda Horronda Le Lion XI Le Lion XI Le Lion XI Royal 21A Royal 21A Royal 21A

Medium Early Claron A3.01 Claron A3.01 Claron AX60 Le Lion B62 Le Lion B62 Horwitu Le Lion B86 Le Lion B86 Le Lion X13 Royal 20A Royal 20A Royal 23A Royal 29A Royal 29A Royal 24A Somycel 76 Somycel 76

Medium Claron AX60 Claron A3.01 Horwitu Le Lion B62 Le Lion X13 Le Lion B86 Royal 23A Royal 20A Royal 24A Royal 29A Somycel 76

157 Although the hollowness of the stipe in longitudinal section (stipe structure) for the off-white and hybrid varieties tended to have more hollow stipes in the second and third flush, this characteristic showed an experimental effect. Therefore, it was rejected for PBR testing. The cap border of flats was noticed for little remnants of the veil in the categories 'smooth', 'partly frayed' and 'frayed'. Consistent results in both experiments proved this characteristic to be useful for PBR testing. The white varieties (smooth) were distinct from the off-white varieties (frayed). The hybrid varieties had a frayed border as the off-white varieties, except Horwitu, Royal 23A, Royal 24A, Le Lion X13 and Claron AX60 in the first flush where they had a partly frayed border. Of the quantitative characteristics, only the cap thickness had survived as reproducable and reliable for PBR tests, by lack of a significant experiment*variety interaction. Both experiments make clear that varieties have their own flushing behaviour. Each flush gave different results between varieties. The investigations thus far showed that variety groups could be distinguished: white varieties (Le Lion B92, Royal 70, Royal 75, Somycel 53 and Somycel 91), off-white varieties (Claron A3.01, Le Lion B62, Le Lion B86, Royal 20A, Royal 29A and Somycel 76), hybrid varieties of the Horronda type ( Claron AX30, Horronda, Le Lion XI and Royal 21A) and hybrid varieties of the Horwitu type (Claron AX60, Horwitu, Le Lion X13, Royal 23A and Royal 24A). Within groups differences between varieties were either absent or insufficient. Moreover, esterase isozyme analysis by PAG electroforesis confirmed this grouping.

Table 3. Outline of hybrid varieties from white and off-white varieties, based on the three characteristics cap border, cap thickness and flushing pattern (experiment 3).

Cap Cap Flushing Flush 1 Flush 2 Flush 3 Border Thickness Pattern

Smooth Thick Medium Claron AX30 Claron A5.1 Claron A5.3 Le Champ.200 Le Champ.222 Le Lion X20 Royal 26A Somycel 112

Late Le Champl02A Le Lion XI Somycel 205

Medium Early EuroSemyl70 Horwitu Le Lion X13

158 Partly Thick Late ClaronAX30 Royal 26A Frayed Ciaron A5.1 Ciaron A5.3 Le Champ.200 Le Champ.222 Le Lion X20 Royal 26A Somycel 112

Very Claron AX31 Claron AX31 Claron AX30 Late EuroSemy285 EuroSemy285 Claron AX31 Horronda Horronda Claron A5.1 Somycel 208 Le Champl02A Claron A5.3 Le Lion XI EuroSemy285 Somycel 205 Horronda Somycel 208 LeChampl02A Le Champ.200 Le Champ.222 Le Lion XI Le Lion X20 Somycel 112 Somycel 205 Somycel 208

Medium Early EuroSemyl70 EuroSemyl70 Horwitu Horwitu Le Lion X13 Le Lion X13

As far as the same varieties were used, esterase isozyme analysis showed largely a similar grouping. These results are supported by Royse and May (1982), who concluded that the cultivated mushroom is a near monoculture. Table 2 outlines the examined varieties based on the three characteristics cap border, cap thickness and flushing pattern.

Experiment 3

This experiment is part of a joint research project between CRZ and the Mushroom Experimental Station Horst in order to reveal a possible PBR test by combination of molecular genetics and morphology. Besides this, the project focuses attention on wether DNA-analysis for the identification of mushroom varieties with phenotypic different expression under the same breeding conditions is reliable. In this experiment 18 hybrid varieties from white and off-white varieties were grown. These varieties were examined for the three characteristics: cap border, cap thickness and flushing pattern. The results are presented in table 3. Results of the DNA-analysis of these 18 hybrids will be reviewed elsewhere. Fruiting bodies from the third flush were also tested for shape differences by image analysis. The resulting analysis will be published separately.

159 Cap border tended now in the direction of smooth, comparing with the earlier experiments. The hybrid varieties, Euro-Semy 170, Le Lion X13 and Horwitu were strikingly similar, and at the same time very different from the other varieties, which were only gradually distinct from each other. They resembled mostly the Horronda type.

Conclusion

The white, off-white varieties and their hybrids are very poor in morphological and physiological characteristics for PBR tests. Thus far the only reliable and reproducable characteristics are cap border, cap thickness and flushing pattern. The near monoculture of the cultivated mushroom which brings on a very limited gene pool, contributes to the lack of distinctive characteristics. Progress in finding characteristics to establish genetic relationships may be expected from quantification of characteristics (image analysis) and from DNA analysis to establish genetic relationships. Such investigations are in progress.

References

Anonymous, 1984. De hoger beroepsprocedure. De Champignoncultuur 28:463-465 . Bonner, J.T., K.K. Kane & R.H.Levey, 1956. Studies on the mechanics of growth in the common mushroom, Agaricus campestris. Mycologia 48:13-19. Fikkert, K.A., 1985. Zaaizaad en Plantgoedwet. In: Schuurman & Jordens (Ed.): Nederlandse Staatswetten 163-11. W.E.J. Tjeenk Willink, Zwolle. Greengrass, B., 1990. The interface between plant breeders' rights and other forms of intellectual property protection and the future. In: UPOV Seminar on the nature of and the rationale for the protection of plant varieties under the UPOV convention, September 19 to 21, 1990, Budapest. Geneva. Royal Horticultural Society in association with the Flower Council of Holland, without year. RHS Colour Chart. London, Leiden. Royse, D.J. & B. May, 1982. Use of isozyme variation to identify genotypic classes ofAgaricus brunnescens.Mycologi a 74:93-102. UPOV, 1979. Revised general introduction to the guidelines for the conduct of tests for distinctness, homogeneity and stability of new varieties of plants. TG/1/2. Geneva. UPOV, 1985. International union for the protection of new varieties of plants. 293(E). Geneva. UPOV, 1986. List of the taxa protected in the member states of UPOV and in the signatory states of the 1978 act of the convention. C/XX/6. Geneva. UPOV, 1991. Diplomatic conference for the revision of the international union for the protection of new varieties of plants. DC/91/138. Geneva.

160 LIST OF AUTHORS (in alphabetic order)

J.B. Anderson University of Toronto, Canada C. Bas Rijksherbarium, The Netherlands E. Becker Monterey Laboratories, USA M.P. Challen Horticulture Research International, UK K. Den Hollander Mushroom Experimental Station, The Netherlands A. Eicker University of Pretoria, Republic of South Africa T.J. Elliot Horticulture Research International, UK G.F.G. Fritsche Mushroom Experimental Station, The Netherlands M.C. Harmsen University of Groningen, The Netherlands W.E. Hintz University of Toronto, Canada P.A. Horgen University of Toronto, Canada J.S. Horten University of Vermont, USA T.Jin University of Toronto, Canada R.W. Kerrigan University of Toronto, Canada R.S. Khush Monterey Laboratories, USA L. Morgan Monterey Laboratories, USA CA. Raper University of Vermont, USA M.M. Robinson University of Toronto, Canada J.C. Royer University of Toronto, Canada J. Scheer University of Groningen, The Netherlands T.A. Schuurs University of Groningen, The Netherlands A.S.M. Sonnenberg Mushroom Experimental Station, The Netherlands J.A. Stalpers Centraalbureau voor Schimmelcultures, The Netherlands S.L. Tan Centraalbureau voor Schimmelcultures, The Netherlands C.F. Thurston King's College London, UK A.P.J. Van De Munckhof Mushroom Experimental Station, The Netherlands A. Van Der Neut Centre for Variety Research and Technology, The Netherlands M. Van Greuning University of Pretoria, Republic of South Africa L.J.L.D. Van Griensven Mushroom Experimental Station, The Netherlands C.W. Van Ingen Foundation for the Advancement of Public Health and Environmental Protection, The Netherlands P.C.C. Van Loon Mushroom Experimental Station, The Netherlands M. Wach Monterey Laboratories, USA J.G.H. Wessels University of Groningen, The Netherlands D.A. Wood Horticulture Research International, UK P.J. Wuest Pennsylvania State University, USA

161