Mycol. Res. 100 (9). 1082-1086 (1996) Printed in Great Britain 1082 Intracellular infection of the cultivated mushroom Agaricus bisporus by the mycoparasite Verficillium fungicola var. fungicola J. W. DRAGT', F. P. GEELS', W. C. DE BRUIJNZ AND L. J. L. D. VAN GRIENSVEN' 'Mushroom Experimenfal Sfafion, P.O. Box 6042, 5960AA Horsf, The Netherlands AEM-unit, Laboratory for Clinical Pafhology, Erasmus University, Rotterdam, The Netherlands Infection of the cultivated mushroom Agaricus bis~orus by the mycoparasite Verticillium fungicola var. fungicola was studied using light microscopy and TEM. Light microscopy indicated that V. fungicob hyphae grew on and in the hyphae of A . bisporus. Appressoria of Verticillium were also seen. TEM revealed the presence of V. fungicola hyphae inside the hyphae of A . bisporus. Dry bubble disease of the cultivated mushroom, Agaricus bisporus (J. E. Lange) Imbach, is caused by Verticillium fungicola (Preuss) Hassebr. var. fungicola (Van Zaayen & Gams, 1982). It is a very damaging disease and has been a problem for the mushroom industry since it was first reported in 1892 (Costantin & Dufour, 1892). In The Netherlands, the disease has been commonly prevented by strict hygiene, disinfection of the casing with formaldehyde and, when necessary, by the application of the chlorinated aromatic compound prochloraz (Sporgon) to the casing. However, the use of prochloraz has gradually discontinued due to the elevated residues of the fungicide in waste materials from mushroom farms. This has created a need to breed new mushroom strains resistant to V. fungicola. The wild type A. bisporus ARP collection (Kemgan, 1995), which contains approximately 200 different strains from all over the world, has been screened for the presence of natural resistance t o V . fungicola. Amongst many susceptible strains, a limited number were found to exhibit a partial resistance to the infecting mycoparasite. Since it is intended to incorporate several resistance factors of A. bisporus into one single strain, it appears imperative to gain an in depth knowledge of the pathology of V. fungicola. The course of events during pathogenesis has already been studied and the different symptoms have been described, mainly macroscopically (Ware, 1933; Fekete, 1967). Some microscopical studies have also been published, including those using scanning electron microscopy (Gandy, 1985; North & Wuest, 1993). No evidence of specialized penetration structures or of direct penetration has been found (North & Wuest, 1993). The pathological effects have been described as mainly due to the production of extracellular enzymes by V. fungicola, leading to necrosis and browning of the mushroom (Trigiano & Fergus, 1979; Gray & Morgan-Jones, 1981; Kalberer, 1984; Thapa & Jandaik, 1989). This communication describes the infection of A. bisporus by V . fungicola both by light microscopy and TEM. Basidiomes of A. bisporus (strain Horstm UI, Somycel SA, Langeais, France) were grown in growing rooms at the Mushroom Experimental Station on compost obtained from the Dutch Mushroom Growers Cooperative (CNC), Milsbeek as described before (Geels, Hesen & Van Griensven, 1994). Mycelial cultures of V. fungicola were grown on malt extract agar (MEA), containing 10 g 1-I malt extract (Oxoid) and 12 g I-' purified agar, autoclaved at 120 'C for 20 min. V. fungicola (isolate CBS 648.80) (Van Zaayen & Gams, 1982) was grown on MEA at 20'. Conidial suspensions were prepared by gently scraping the surface of 3-wk-old colonies in tap water. The density of the conidial suspension was determined with a haemocytometer and adjusted to 5 x 105 conidia ml-' tapwater. Growing basidiomes of A. bisporus (1 cm diam.) were sprayed with the conidial suspension (200 ml m-') and were harvested at 3 and 7 d after inoculation. Tissue for histological examination was taken from infected areas on the surface of the pileus. Samples were prepared for light microscopy using glycol methacrylate embedding techniques. Pieces of tissue (2 x 5 x 10 mm) were fixed in 4% (w/v) formaldehyde and 1 % (w/v) glutaraldehyde in 0.1 M Millonig buffer (pH 7.2) for 16 h at 20' (Millonig, 1961). Fixed tissue was dehydrated in an ethanol series (50, 70, 80, 90, 96, 100% (v/v)), and then immersed in 100% (v/v) ethanol ovemight at 20'. This was followed by embedding in Technovit 7100 (Kulzer, Friedrichsdorf, Germany). Semi-thin sections (I pm) were cut with a glass knife on a rotary microtome (Microm, HM 360), and stained for light microscopy using the Periodic Acid Schiff (PAS) method (Lillie & Fulmer, 1976) and subsequently with haematoxylin (Shandon, Pittsburgh, U.S.A.). Infected tissue was then examined using a Zeiss Axioskop light microscope at x 1000 magnification. For TEM, pieces of tissue (2 x 2 x 5 mm) from the outer part of the pileus were fixed as described previously, and subsequently postfixed in 1% (w/v) OsO, and 0.05 M K,Fe(CN), .3H,O in 0.1 M Millonig buffer, ovemight at 20' (De Bruijn, 1973). This was followed by dehydration in an