LIBRARY Michigan State University PLACE IN RETURN BOX to roman this checkout m you: record. TO AVOID FINES Mum on or before date duo. DATE DUE DATE DUE DATE DUE MSU I. An Nflmaflvc WM Oppommuy Institution GENETIC ANALYSIS OF VEGETATIVE INCOMPATIBILITY POLYMORPHISMS AND HORIZONTAL TRANSMISSION IN THE CHESTNUT BLIGHT FUNGUS, CR YPHONECTRIA PARASITICA By David Henry Huber A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Botany and Plant Pathology 1996 ABSTRACT GENETIC ANALYSIS OF VEGETATIVE INCOMPATIBILITY POLYMORPHISMS AND HORIZONTAL TRANSMISSION IN THE CHESTNUT BLIGHT FUNGUS, CR YPHONECIRIA PARAsmCA By David Henry Huber Vegetative incompatibility polymorphisms in Cryphonecm'a parasitica are known to have quite variable effects upon the horizontal (cytoplasmic) transmission of hypoviruses. This study examined the effects of individual vegetative incompatibility (Vic) genes upon the horizontal transmission of nuclei, dsRNA hypoviruses, and a senescence-inducing agent in C. parasitica. A genetic analysis identified three new Vic loci named vic3, vial and vic5, each with two alleles, that produce incompatibility through allelic interactions. The effects of these three vic loci as well as vicI and vic2 upon heterokaryon formation were tested using complementary color mutations (cre and br) under nonselective growth conditions. Heterokaryons were found to form between strains homoallelic at all We loci, when grown on potato dextrose agar and chestnut tissue. Heteroallelism at any of the five vic loci produced barrages and prevented heterokaryon formation on chestnut tissue. Hyphal tips from heterokaryons contained both nuclear types at variable ratios. The effects of all five vic loci upon the horizontal transmission of hypoviruses were examined. Heteroallelism at only Vic] produced a nonreciprocal (unilateral) transmission where the vial-2 recipient does not become infected with virus, but the vicI -1 recipient always becomes infected. Heteroallelism at vic2 prevented viral transmission but epistasis from other vic genes could decrease this barrier. Evidence indicates that allele WC] -1 is epistatic over allele vic2-1 when both occur in a recipient thereby reducing the transmission barrier caused by vic2. Alleles vic4-2 and vic5-2 may also cause a unilateral reduction in the vic2 transmission barrier. Heteroallelism at vic3 also causes unilateral transmission. Heteroallelism at vicI was found to be epistatic over heteroallelism at vic3. Heteroallelism at Wet! and vic5, individually and together, did not hinder hypovirus transmission. An unknown hypovirulence—induCing agent was found to be phenotypically Similar to the suppressive senescence syndrome found in other fungi. Transmission of this agent occurred between compatible strains, was uninhibited by heteroallelism at vic4, but was prevented by vic2. The senescence phenotype was found to be characterized by elevated levels of respiration through the alternative oxidase, and to produce conidia with variable degrees of senescence, implicating mitochondrial dysfunction in the senescence syndrome. Copyright by David Henry Huber 1996 Dedicated with love to Mom and Dad for their love and for their encouragement through the years of my fascination with the wonders of life “Be exalted, O God, above the heavens; let your glory be over all the earth." (Psalm 57:5) ACKNOWLEDGEMENTS I would like to express my heart felt thanks to several people who have had an impact on me and my research. First, many thanks to my advisor Dr. Dennis Fulbright for his guidance and insight throughout this work, unflagging enthusiasm, generous patience and material support. Dr. Helmut Bertrand served on my committee and provided valuable scientific counsel. Dr. David Jacobson served on my committee, participated in many helpful discussions, and helped to substantially improve this dissertation. Thanks also to Dr. C. A. Reddy for serving on my committee. My primary financial support was provided by the National Science Foundation Center for Microbial Ecology for which I am grateful. The training environment provided by the CME has changed my outlook on the biologieal sciences and altered my course-a sincere compliment. The Department of Botany and Plant Pathology also contributed financial support through several assistantships. Finally, thanks to the members and associates of the Fulbright lab for their friendships and help through the years. Dr. Frank Louws, in particular, was a bright light. vi TABLE OF CONTENTS LIST OF TABLES ......................................... x LIST OF FIGURES ....................................... xiii CHAPTER 1 * INTRODUCTION ......................................... l Vegetative incompatibility and heterokaryon formation ................ 5 The genetic basis of vegetative incompatibility ..................... 8 Podospora anserr'na .................................... 9 Neurospora crassa .................................... l4 Cryphonectria parasitica ................................ 16 The occurrence and horizontal transmission of viruses and genetic elements in filamentous fungi ................................... 18 Horizontal transmission of viruses and genetic elements in C. parasitica . . . . 25 The pathogen: Cryphonecm’a parasitica ........................ 28 The hyperparasites ...................................... 32 Vegetative incompatibility in fungi: function and consequences .......... 36 Dissertation content ..................................... 40 Literature Cited ........................................ 41 CHAPTER 2 DETECTION OF VEGETATIVE INCOMPATIBILITY IN CRYPHONEC’IRIA PARASITIC‘A ............................................ 55 Abstract ............................................ 55 Materials and Methods ................................... 57- Strains ........................................... 57 Sexual crosses ....................................... 58 Barrage tests ........................................ 61 Results ............................................. 63 Discussion ........................................... 73 Literature Cited ........................................ 78 CHAPTER 3 VEGETATIVE IN COMPATIBILITY POLYMORPHISMS IN CRYPHONEGRIA PARASIHCA: NEW VIC LOCI AND HETEROKARYON FORMATION ..... 80 Abstract ............................................ 8O vii Materials and methods ................................... 84 C. parasitica strains and culture conditions .................... 84 Sexual crosses ....................................... 84 Mutagenesis ........................................ 86 Heterokaryon determination .............................. 86 Vegetative incompatibility determination ...................... 87 Results ............................................. 90 Heterokaryon formation ................................ 90 Restriction of heterokaryon formation by vegetative incompatibility ..... 93 Genetic analysis confirming the identity of vicI and vic2 ............ 94 Genetic analysis identifying two new vic loci, vial and vic5 .......... 95 Unusual mycelial interactions associated with heteroallelism at vic4 and vic5 ..................................... 101 Identification of the remaining nonparental vc genotypes from cross F3.16 br nic X Al.l3 cre met .......................... 102 Genetic analysis identifying a new locus vic3 .................. 104 Linkage analysis .................................... 110 Discussion .......................................... 1 13 Identification of three new vic loci ......................... 113 Heterokaryon formation under nonselective conditions ............ 116 The effects of vic genes upon heterokaryon formation ............. 117 The function of vegetative incompatibility genes ................ 119 Literature Cited ....................................... 121 CHAPTER 4 THE EFFECTS OF VEGETATIVE INCOMPATIBILITY GENES UPON THE HORIZONTAL TRANSMISSION OF VIRUSES IN THE CHESTNUT BLIGHT FUNGUS, CRYPHONECIRIA PARASITICA, ARE LOCUS SPECIFIC AND MODIFIED BY EPISTASIS ................................. 125 Abstract ........................................... 125 Materials and Methods .................................. 129 C. parasitica strains and culture conditions ................... 129 Transmission tests ................................... 130 Results ............................................ 135 Hypovirus effects upon fungal morphology, and the detection of transmission ..................................... 135 The effect of genetic background upon transmission when all Vic loci are homoallelic ............................ 140 The effects of via! and vic5 upon transmission ................. 142 The effect of Vic] upon transmission ........................ 142 The effect of vic2 upon transmission ........................ 146 Other genetic background effects upon the transmission barrier caused by vic2 .................................... 152 The effect of vic3 upon transmission, and modifications due viii to epistasis. ..................................... 153 Prevention of transmission due to undescribed incompatibility genes . . . 157 Horizontal transmission efficiencies of different dSRNA genomes ..... 157 Horizontal transmission on live chestnut tissue ................. 160 Discussion .......................................... 161 Individual vic genes have different effects upon horizontal transmission . 161 Nonreciprocal transmission caused by vic loci .................. 165 Evidence that epistasis between vic genes modifies cytoplasmic transmission ..................................... 166 The implications of cytoplasmic transmission variability for vr'c gene action ................................... 169 Transmission efficiencies of different viruses .................. 171 Primary function of vegetative incompatibility genes ............. 171 Literature Cited ....................................... 174 CHAPTER 5 THE TRANSMISSION AND PHENOTYPIC EFFECTS OF A SENESCENCE SYNDROME IN CRYPHONECTRIA PARASIHCA .................. 180 Abstract ........................................... 180 Materials and Methods .................................. 182 C. parasitica strains and culture conditions ................... 182 Transmission assay ................................... 184 Virulence assays .................................... 184 Alternative oxidase respiration assay ........................ 185 Results ............................................ 186 The senescence phenotype .............................. 186 Serial transmission of the senescence phenotype ................ 189 The effects of vegetative incompatibility loci upon transmission of senescence .................................... 189 Respiration, senescence, and reduced virulence ................. 191 Discussion .......................................... 194 Literature Cited ....................................... 201 Table 1.1 2.1 2.2 2.3 2.4 3.1 3.2 3.3 3.4 3.5 LIST OF TABLES Examples of horizontal (cytoplasmic) transmission of parasitic genetic elements in filamentous fungi. All transmissions were observed between individuals of a species unless otherwise indicated ............... Strains of Cryphonectria parasitica used in this study ............. Vegetative compatibility types of the progeny from cross 389.7 X EP243 based upon the wood/agar vc assay. .............. Barrage tests of strains representing vegetative compatibility types Q, theta and the vc types with which they had been reported to be compatible (i.e. "multiple-merge" response) by Kuhlman et al. (1984). Barrage tests were based upon the PDA vc assay ................ Barrage tests of strains representing vegetative compatibility types Q, theta, and the vc types with which they had been reported to be compatible (i.e. multiple merge response) by Kuhlman et al. (1984). Barrage tests were based upon the wood/agar vc assay. ............ Strains used in this study ............................... Ratios of cre (cream) to br (brown) nuclear types from heterokaryons composed of vegetatively compatible strains and one putative heterokaryon from strains heteroallelic at vic5 .................. Number of progeny in vegetative compatibility groups from three sexual crosses ...................................... Number of progeny in eight different vegetative compatibility genotypes that segregated from cross 389.7 cre nic X 80-2c br. ....... Number of progeny in eight different vegetative compatibility genotypes that segregated from cross A1.13 cre met X F3.16 br nic. . . . 59 68 7O 71 88 92 97 100 3.6 3.7 3.8 3.9 4.1 4.2. 4.3 4.4 4.5 4.6 4.7 4.8 4.9 Number of progeny in each vegetative compatibility group that segregated from cross Ep29 X Al. 10 cm nic. ........................ 106 Number of progeny in each vegetative compatibility genotype that segregated from crosses K2.30 cm X 12.69 br nic and K2.30 cre X 12.23 br nic .............................. 107 Number of progeny in four different vegetative compatibility genotypes that segregated from cross F4.13 br X Kl.43 cre ........ 109 Linkage analysis of the loci used in this study ................. 111 Strains used in this study .............................. 132 Transmission of dsRNAs where donor and recipient strains are homoallelic at all vic loci (=vegetatively compatible). ............ 141 Transmission of dsRNAs where the donor and recipient strains are heteroallelic at vegetative incompatibility loci vic4 and vic5. ........ 143 Transmission of dsRNAs where donor and recipient strains are heteroallelic at vegetative incompatibility locus vicI . The presence of additive and epistatic effects between vic genes was tested by varying the alleles present at vic4 and vic5. ................... 144 Transmission of dsRNAs where donor and recipient strains are heteroallelic at vegetative incompatibility locus vic2. The presence of additive and epistatic effects between We genes was tested by varying the alleles at Vic], vial and vicS ..................... 148 Transmission of dsRNAs where donor and recipient strains are heteroallelic at both vegetative incompatibility loci vr'cI and vic2. The presence of additive or epistatic effects between Vic genes was also tested by incorporating heteroallelism at vic4 and vic5. ........ 150 Transmission of dsRNAs where donor and recipient strains are heteroallelic at vic3. The presence of additive and epistatic effects between Vic genes was tested by varying the alleles at vial and vic2. . . . 154 Transmission of dsRNAs where the vic gene differences between vegetatively incompatible donor and recipient strains are not known. . . . 158 Transmission of hypovirus CHVl-7l3 between donor and recipient strains conducted on live chestnut tissue. ......................... 159 xi 4.10 Summary of the effects of each vic locus upon the horizontal (cytoplasmic) transmission of hypoviruses in Cryphonecm'a parasitica . . 162 5.1 Strains used in this study .............................. 183 5.2 Transmission of the senescence inducing agent between donor and recipient strains that differ at several vegetative incompatibility (Vic) loci. ........................................ 190 5.3 Alternative oxidase as percent of total respiration, virulence, and senescence phenotypes of strains that have been cytoplasmically infected with the senescence inducing agent from strain KFC9 ....... 193 xii 2.1 2.2 3.1 3.2 4.1 LIST OF FIGURES Barrage tests using the wood/agar vc assay between parental strains 389.7 and EP243 and two progeny from the cross of these strains. The order of the strains in the plate is presented below the photo. Weak and prominent incompatibility reactions are present. The weak incompatibility reactions occur between strain pairs 389.7/D3.3l and EP243/D129. The other incompatibility reactions are of the prominent type. .............................. Barrage tests of weak and strong incompatibility reactions using the wood/agar vc assay demonstrating the efficacy of the assay with virus-infected strains. The order of the strains in each plate is presented to the right of the photo. A. Barrage tests using strain 389.7(713). B. Barrage tests using strain 389.7(GH2). .................... Heterokaryon formation between strains 389.7 ore and F2.36 br homoallelic at all Vic loci (A), and the prevention of heterokaryon formation between389.7 cre and F3.4 br heteroallelic at only vicI (B). The orange sector produced between the strains in (A) is heterokaryotic and exhibits wild type color due to complementation between the ore and br mutations ......................... An orange sector of abnormal mycelium has formed between strains 389.7 ore and F3.10 br heteroallelic at only vic5. The orange sector has thinner mycelium and less conidia than the normal mycelium of the cre and br strains. The plate has back lighting .............. Nonreciprocal transmission of 80-2 dsRNA between donor and recipient strains caused by heteroallelism at vicI. The strain on the left of each plate is the donor (infected with dsRNA); the strain on the right is the recipient (dsRNA-free). The recipient has become infected in the pairing in the right plate but not in the left plate. The left plate contains strains F2.36(80—2) and EP388. The right plate contains Strains EP388(80-2) and F2.36. ................. xiii .65 .72 91 103 136 4.2 4.3 5.1 Phenotypic changes in Strain F2.36 caused by hypovirus infection. A. uninfected strain F2.36; B. strain F2.36(713) infected with CHV1-713; C. Strain F2.36(GH2) infected with CHV3-GH2; D. Strain F2.36(80-2) infected with dsRNA 80—2. ............... 137 The banding patterns of dsRNA isolated from the infected strains of C. parasitica shown in Figure 2 electrophoresed in a 5% polyacrylamide gel and stained with ethidium bromide. Lanes: 1, F2.36; 2, F2.36(713); 3, F2.36(GH2); lane 4, F2.36(80-2). Sizes of dsRNA molecules present in strain F2.36(GH2) are indicated in kilobase pairs (kb). .......................... 138 Growth phenotypes of (A) Strain EP289 and (B) strain EP289(KFC9). Strain EP289(KFC9) was infected with the senescence agent from strain KFC9 and exhibits the senescence growth form ...................................... 187 xiv Chapter 1 Introduction The origin and consequences of genetic diversity represent two central issues in biology. How does genetic variability arise? How is the genesis of novel genetic information controlled? The primary source of genetic variation in eukaryotes lies with the complementary processes of mutation and mixis. In many filamentous fungi genetic variation can also occur in somatic cells through the natural formation of heterokaryons and heteroplasmons. These novel genetic states are quite unusual when considered in the context of most other eukaryotes (Andrews, 1991). The transmission of genetic variation in most eukaryotes occurs primarily in the vertieal dimension through sexual or clonal reproduction, that is, from parent to offspring; some evidence also suggests that horizontal gene transmission‘, that is, transmission from individual to individual, has also occurred although rarely (Kidwell, 1993). The situation with many filamentous fungi is quite different. In filamentous fungi the opportunity for horizontal transmission is associated with the body plan of the organism. Mycelia are frequently coenocytic in nature and form interconnected networks through the ramification and anastomosing of hyphae. The same processes governing vegetative cell fusions within ' The term horizontal transmission has been used in differed, though related, senses by various authors. Kidwell (1993) defines it as 'nonsexml transfer of genetic information between genomes"; Smith et al (1992) used the term to refer to gene transfer between qecies. The term has also been used for the transmission of plasmids via conjugation between bacteria (Simonsen 1991), and for the hat-mission of chromosomal genes between bacteria of the same species (Guttrnan and Dykhuizen 1995). Griffiths (1995) uses the urn in reference to the transmission of plasmids between fungal individuals within a species and between species. In all cases, genetic elemnts are transferred between individuals rather than from parent to offspring. It is in this sense that the transmission of viruses and melei bemeen fungal individuals is referred to here as horizontal. 2 a Single individual (genet) may also occur between genetically different individuals when they make contact. Horizontal gene transmission is therefore made possible through the direct cellular fusions of one mycelium with another, and the subsequent mixing of their protoplasms. This provides the opportunity for a remarkable cohabitation of nuclear and mitochondrial genomes within the shared cytoplasm of the somatic cells, and for coenocytic fungi, stable heterokaryons and heteroplasmons can sometimes form. The traditional concept of organismal individuality would therefore seem to be quite compromised in fungi. For many years, following the classic study by Buller (1931), the concept of the ”compound mycelium” held that this creation of composite mycelia occurred frequently and improved the fitness of the organism. More recently, the compound mycelium concept has been challenged by the recognition that many filamentous fungi possess genetically regulated mechanisms that Significantly limit the occurrence of intraspecific somatic cell fusions (Todd and Rayner, 1980). Somatic or vegetative cell fusion and the genetic mechanisms that regulate the viability of the fusions have been found in the Ascomycetes (Glass and Kuldau, 1992) and Basidiomycetes (Casselton and Economou 1985) although the function of these systems is not the same for each of these groups. Interestingly, other organisms whose lifestyle and body plan provide unique opportunities for somatic cell interactions also exhibit similar systems: the plasmodial slimemolds (Myxomycetes) possess an elaborate somatic cell fusion system (Lane 1987), and evidence also indicates that the Zygomycetes (Griffin and Perrin 1960), and the filamentous bacteria, Actinomycetes, are capable of undergoing cell fusions (Waksman 1967). Bus (1987) points out that even colonial sessile marine animals 3 such as the Ascidians and Bryozoans possess somatic cell compatibility systems. In the Ascomycetes, vegetative cell incompatibility systems have been found to limit the viability of intermycelial vegetative cell fusions to some, but not all, close relatives, and to mediate the degree of cytoplasmic continuity between fused hyphae (Glass and Kuldau, 1992). Although the vegetative cell incompatibility systems in the Ascomycetes have the effect of maintaining the genetic (nuclear and mitochondrial) integrity of the individual (genet), it is not clear from present knowledge whether this constitutes their raison d 'étre. Several hypotheses have been proposed to explain the primary function of the genes controlling vegetative incompatibility, including such disparate ideas as reproductive organ development and ecological competition. Another hypothesis for the primary function of these genes is that they have evolved to limit the infectious transmission of cytoplasmic parasites (Caten 1972; Nauta and Hoekstra 1994). Limitations upon cytoplasmic continuity between mycelia not only constrain the movement of nuclei and mitochondrial genes, but also limit the horizontal transmission of intracellular parasitic genetic elements. Viruses and plasmids are common in fungi, and several transposons are also known (Kistler and Miao, 1992; Nuss and Koltin, 1990; Griffiths, 1995). These mobile genetic elements are capable of being transmitted between individual mycelia through vegetative cell fusions. In fact, horizontal movement without direct cytoplasmic contact has not been detected. Although horizontal transmission, particularly of viruses, has been documented many times in fungi, the effects of the vegetative cell fusion systems upon the transmission process are still not well understood. The horizontal transmission of double-stranded (ds) RNA viruses in the chestnut blight pathogen, Ctyphonecm'a parasitica, has been studied 4 using many fungal strains from nature. These Studies have Shown that the effects of different vegetative incompatibility genotypes (polymorphisms) upon viral infection are complex and quite variable, suggesting that the polygenic nature of the polymorphisms is responsible. The elucidation of the influence of vegetative cell incompatibility systems in the Ascomycetes upon horizontal transmission must be set in the context of the genetic basis of the incompatibility. One of the unique features of Ascomycete vegetative incompatibility is that it is regulated by several (many) genetic loci, each of which is capable of independently eliciting an incompatibility reaction. Preliminary studies in several Ascomycetes further indicate that each genetic locus may have a unique effect upon the fused ee11s, such as differences in the rapidity or extent of cell death (Garnjobst and Wilson 195 6). Therefore the effects of vegetative cell incompatibility upon the cytoplasmic transmission of intracellular genetic elements must be considered on a gene by gene basis. The purpose of this research was to investigate the effects of the vegetative incompatibility system of the Ascomycete C. parasitica upon horizontal (cytoplasmic) transmission. C. parasitica has been chosen for this study because its virulence can be reduced by cytoplasmic infection with viruses and transmissible cytoplasmic respiratory- deficiency mutations. Both of these debilitating genetic elements are strictly intracellular and their ' horizontal transmission has been reported to be limited by vegetative incompatibility (Anagnostakis and Day 1979; MacDonald and Fulbright, 1991). This work consists of two objectives. The first concerns the elucidation of the genetic basis Of 5 vegetative incompatibility by identifying new vegetative incompatibility (Vic) genes. The second objective concerns the effects that individual vic genes have upon horizontal transmission of different genetic elements, including nuclei, dsRNA viruses, and a cytoplasmic senescence-inducing agent. Vegetative incompatibility and heterokaryon formation Gene transfer in filamentous fungi commonly occurs through nuclear transfer. The state where two or more different nuclei cohabit a common cytoplasm is termed a heterokaryon. The nature of the heterokaryotic condition in fungi is quite variable. For example, in Basidiomycetes the mating process requires the formation of a specialized heterokaryotic condition known as a dikaryon where strict developmental control, operating via clamp connections, maintains one of each nuclear type in each hyphal cell. Dikaryons are formed between strains of complementary mating types, and are the necessary prelude to sexual recombination. Ascomycetes do not require the formation of somatic dikaryons for sexual reproduction although ascogenous hyphae are dikaryotic, and many species can form different types of heterokaryons during somatic growth. The formation of heterokaryons in Ascomycetes results from hyphal fusions between conspecifics or even between different species. Vegetative incompatibility results when hyphal fusions create heterokaryons of incompatible nuclei but the heterokaryotic state is limited by the death of the fused cells. Recent reviews of vegetative incompatibility are Esser and Blaich (1994), Leslie (1993), and Glass and Kuldau (1992). Two types of heterokaryons are known to form in Ascomycetes. In one type of 6 heterokaryon nuclei from each individual migrate into the hyphal tips. Proliferation of the heterokaryon is therefore possible via hyphal tip extension and branching. Fungi in which this type of heterokaryon has been found are Neurospora crassa (Beadle and Coonradt, 1944), Aspergillus nidulans (Pontecorvo, 1953), Penicillium cyclopium (Rees and Jinks, 1952), and Sclerorinia sclerotiorum (Ford et al., 1995). In the second type of heterokaryon the different nuclear types are restricted to the fused hyphal cells and do not migrate throughout the mycelium or into the hyphal tips. Although these heterokaryotic cells do not grow or divide, they are capable of nutritionally sustaining mycelial growth well beyond the fusion cells (Adams et al., 1987). This second type of heterokaryon has been described in Venicillium dahliae (Puhalla and Mayfield, 1974), Gibberella firjikuroi (Puhalla and Spieth, 1985), Gibberella zeae (Adams et al., 1987), and Magnaporthe grisea (Crawford et al., 1986). Heterokaryons have also been formed through protoplast fusion. Adams et al. (1987) found that heterokaryons formed by protoplast fusion in G. zeae differed from heterokaryons produced by hyphal fusions. Heterokaryosis following hyphal fusions of vegetatively compatible strains in G. zeae is confined to the fused cells. When heterokaryons were formed through protoplast fusion between vegetatively compatible complementing auxotrophic strains, the resulting colony initially produced prototrophic hyphal tips, but subsequently segregated into auxotrophic hyphal tips, suggesting nuclear segregation. In contrast, although heterokaryosis would not occur between vegetatively incompatible strains following mycelial contact, stable self-perpetuating heterokaryons of incompatible Strains would form after protoplast fusion. The resulting colonies had an 7 unusual, inhibited growth morphology, and produced prototrophic conidia. Adams et al. (1987) concluded that heteroploids had formed rather than heterokaryons between the vegetatively incompatible protoplasts. Tolrnsoff ( 1983) has suggested that aneuploidy and heteroploidy may be natural mechanisms of variation in fungi. The physiological basis of the cell death caused by vegetative incompatibility has not been examined in detail. Gamjobst and Wilson (1956) observed that the protoplasm of the fused cells of incompatible strains of N. crassa became granular and vacuolated. This degeneration involved one or a few cells in each hypha, and was sharply demarcated by septa with plugged pores. When cell degeneration occurred more slowly, two or three consecutive septae would become plugged, along with less degeneration in the more distal cells. Eventually the vacuoles disappeared and the protoplasm shrank leaving the old cell walls in place. The viable cells behind the plugged septa would then regrow, often within the old cell walls. Hyphal regrowth within old cell walls has also been observed by Jacobson (1993). Microinjection of protoplasm from one hypha into an incompatible hypha Showed that the protoplasm was sufficient to induce cell death in the recipient cell (Garnjobst and Wilson, 1956). Transmission electron micrographs of compatible and incompatible cell fusions in C. parasitica showed that the incompatible fusions resulted in a granular appearance of the protoplasm, vacuolation, and subsequent collapse of the cell walls (Newhouse and MacDonald 1991). The association of vegetative cell death with organ development in P. anserina, and the complexity of the genetic regulation of cell death in this Species, suggests that it may be appropriate to consider the cell death response of vegetative incompatibility as analagous to apoptosis (programmed cell death) which has 8 been found to be important in the development of higher eukaryotes (Ellis et al. , 1991). Heterokaryon formation, whether restricted to fused hyphae or maintained in hyphal tips, is controlled by vegetative incompatibility genes. To understand the consequences of vegetative cell fusions in fungi, we must understand the gene interactions that limit the viability of the fusions. The Genetic Basis of Vegetative Incompatibility Genetic analyses of vegetative incompatibility in Ascomycetes have been conducted on only a few species. The best understood vegetative compatibility (vc) systems are those of Podospora ansen'na and Neurospora crassa. These two species along with C. parasitica will be reviewed here. In addition, the Basidiomycetes have a fascinating compatibility system controlled by mating type genes that regulates fusions between vegetative hyphae, but toward quite different developmental ends (Casselton and Economou, 1985). The plasmodial protists (Myxomycetes) present another interesting system where the viability of somatic cell fusions are controlled by multiple loci (Lane, 1987). Basidiomycete mating type compatibility and Myxomycete somatic cell compatibility will not be reviewed here. Vegetative incompatibility polymorphisms in Ascomycetes are produced by polygenic systems. In the following reviews the term allelic vegetative incompatibility refers to an incompatibility reaction between strains caused by heteroallelism at any one or more vegetative incompatibility (Vic or her) loci. Strains homoallelic at each vic locus are vegetatively compatible and do not exhibit the incompatibility reaction. The term 9 nonallelic vegetative incompatibility refers to an incompatibility response produced by the presence of certain alleles at different loci in a common cytoplasm, resulting from either hyphal fusions or recombination. Podospora anserina The best understood vegetative (protoplasmic) incompatibility system is that of P. anserina. The genetic picture of vegetative incompatibility that has emerged for this Species includes allelic gene interactions, nonallelic gene interactions and suppressor genes. The suppressor genes, in particular, have provided intriguing clues as to the biological function of vegetative incompatibility genes. The detection of vegetative incompatibility in P. anserina has been based upon the development of reaction zones called barrages between contacting mycelia. Rizet (1952) was the first to subject the barrage phenomenon in P. marina to genetic analysis. Thirteen loci have been discovered so far which interact in an allelic or nonallelic fashion to cause vegetative incompatibility reactions. Nine loci are known to be involved in five nonallelic gene interactions. Five of the nine loci (c,d,e, r, v) were found in wild-type strains, and four of the loci (fig,k,l) were produced by mutagenesis. Allelic incompatibility is controlled by five loci (b,q,s,v,z), one of which, v, also functions in the nonallelic system. Nonallelic vegetative incompatibility involves lethal cellular reactions resulting from the interactions between the gene combinations, C/D, C/E, G/F, UK, or V/R, where each capital letter represents a particular allele at a locus designated by that letter. These lethal cellular reactions will occur when hyphae from two strains fuse, each of which carries one 10 of the alleles of these pairs, or when recombination reassorts these alleles into a single individual. Several lethal interactions are produced from c/d and c/e interactions because these three loci are multiallelic. Cell death resulting from any of the nonallelic gene interactions is suppressed by the cooccurrence of two mutations, a recessive mutation in the modA gene, and a dominant mutation in the modB gene (Boucherie and Bernet, 1974; 1980). Bernet (1992a) has shown that the nonallelic vegetative incompatibility genes c,d,e,r and v (hereafter, c-v) have pleiotropic effects. Gene interactions between the pairs of loci c/d, c/e, and r/v were found to be necessary for the development of aerial organs, including protoperithecia, perithecia, and aerial hyphae. Strains with the double mutation modA modB also demonstrated a connection between nonallelic incompatibility and aerial organ development because of their pleiotropic effects: modA modB abolished nonallelic incompatibility as well as the development of protoperithecia and aerial hyphae (Boucherie and Bernet, 1974; 1980). Bernet (1992a) has shown that each of the gene interactions c/d, c/e, and r/v are involved in cell death. Evidence suggests that for each of these three gene interactions one of the gene products (d, e, r) is synthesized in the vegetative cells while the other gene product (c, v) is synthesized in the perithecia. Bernet (1992a) concludes that the development of reproductive organs are aided by the degradative functions of the mod and c-v germ which transform fertile stationary-phase cells into a source of nutrients for the developing organs. The suppression of both nonallelic vegetative incompatibility and developmentally associated cell death by modA modB mutations suggests that these two processes are related. 11 Other mutations have been found in P. anserina which involve the regulation of cell death. Mutations resulting in large perithecia (lpr) have been described which also have pleiotropic effects (Bernet, 1991). The lpr mutations (lprA, B, C, D, E, F) lead to the increased development of protoperithecia and aerial hyphae and also to the early death Of stationary-phase cells. The mutations modA modB, which suppressed nonallelic vegetative incompatibility, also suppressed the Ipr mutations. In addition, IprB maps at the d incompatibility locus. Bernet (1992a) has proposed an interesting regulatory system for the development of perithecia based upon the phenotypic evidence from the nonallelic incompatibility genes and the suppressor genes. He notes that the products of the nonallelic incompatibility genes c and v are diffusible through hyphae whereas the gene products of d, e, r, modA, and modB are not. Furthermore, Asselineau et al. (1981) Showed that the gene products of d, e, r, and modB may be associated with the plasma membrane. Bernet (1992a) therefore proposes that the products of genes c and v may be analagous to peptide hormones and the products of d, e, and r to their receptors. The polypeptide products of genes c and v may then diffuse away from perithecia through the surrounding hyphae, bind to their receptors, and signal the inception of a cell death process which will provide the developing perithecia with nutrients. This suggested mechanism would also account for the asymmetry in the cell death response observed between strains incompatible at these loci (Labarere et a1. , 1974). What then is the relationship between nonallelic vegetative incompatibility which results from anastomoses between different strains and the cell death associated with aerial organ formation? Bernet ( 1992a) suggests that nonallelic vegetative 12 incompatibility results from mutations in genes whose normal function lies in the application of cell death to development. Recent work is beginning to elucidate the biochemical basis of nonallelic vegetative incompatibility (Begueret et al., 1994; Saupe et al., 1994; Saupe et al., 1995). Several alleles at loci c (het-c) and e (her-e) have been cloned. The e locus polypeptide has two structural features homologous to known proteins. The carboxy terminal region contains a repeat of 42 amino acids that shows similarity to B subunits of trimeric G proteins. The number of repeats varies among e alleles from 3 to 12. The amino terminal region has sequence Similarity to GTP—binding consensus sequences of GTPases. Single amino acid differences in the GTP—binding domain prevents the incompatibility reaction. The Structure of these proteins suggests that they are involved in signal transduction. Four wildtype c alleles have been cloned that have distinct specificities for e alleles (Saupe et al., 1995). The polypeptides differ from each other by l to 15 amino acids, and Show similarity to a protein from pig brain that catalyzes the exchange of glycolipids between cellular membranes (Saupe et al. , 1994). Gene disruptions of c caused defective ascospore production, indicating that c has an essential role in development. Allelic vegetative incompatibility in P. anserina is controlled by five known loci, b, q, s, v, and z. The most detailed molecular understanding of allelic vegetative incompatibility comes from the cloning of the three 3 locus alleles: s, S and the neutral allele A" (Turcq et al., 1990; Deleu et al., 1993). Alleles s and S encode 30 kDa polypeptides which differ at 14 amino acids. The function of the polypeptides could not be inferred from present databases. Sequencing of s and S alleles from four different 13 strains each showed that the amino acid sequences of the two alleles is completely conserved. The neutral allele, .3", differed from s by containing a direct duplication of 46 bp, and by failing to produce any detectable protein. Remarkably, Deleu et al. (1993) found that a Single amino acid difference between the s and S alleles is sufficient to induce vegetative incompatibility. Four sequences with 3 Specificity were also cloned from the species Podospora comata. In contrast to P. anserina, these four protein sequences differed by up to 12 amino acids even though they conferred the same incompatibility specificity. Turcq et al. (1991) inactivated the s and S alleles by gene disruption, and were thereby able to inhibit incompatibility. They also claimed that the 3 genes were not essential for cell viability. Unfortunately, no data (and no methods) were incorporated into their paper to support this claim. Allelic incompatibility is suppressed in P. anserina by mutations in the gene modD while nonallelic incompatibility is not so suppressed. The initial work on the modD mutations indicated that they affected the transition from the stationary phase into a developmental phase where aerial and reproductive organs are formed (Labarere and Bernet, 1979, Durrens et al., 1979, Durrens and Bernet, 1982). Specifically, Labarere and Bernet (1979) showed that exit from stationary phase depends upon proteolytic activity which is dependent upon the modD gene. This proteolytic activity has also been found to be suppressed by the modB gene, which as mentioned above, also suppresses the proteolytic activities caused by the nonallelic incompatibility genes. Because of this confluence in the effects of the mod genes, Bernet (1992b) has suggested that the only difference. between allelic and nonallelic vegetative incompatibility is the trigger l4 mechanism. Bernet (1992b) further speculates that the effects of the various allelic incompatibility genes upon the modD gene might represent redundancy in the developmental control system. Neurospora crassa Neurospora crassa was one of the first Ascomycetes where heterokaryosis was documented (Beadle and Coonradt, 1944), and subsequently became the first fungus where heterokaryon incompatibility was genetically characterized. Ten vegetative (heterokaryon) incompatibility (he!) loci have been identified in N. crassa (Perkins, 1988); two new her loci have recently been reported (Min et al. 1994; Ohmberger et al. 1994). These loci produce incompatibility through allelic gene interactions although there is now evidence that nonallelic incompatibility is also present in this species (Jacobson, 1993). Gamjobst (1953, 1955) identified and named the vegetative incompatibility genes her-c and her-d, the first such genes to be identified for any fungus. The identification of her-i by Pittenger (1964) and her-e by Wilson and Gamjobst (1966) followed. Holloway also (1955) studied the actions of four loci that either prevented the formation of heterokaryons or inhibited the growth of heterokaryotic mycelium. These genes were referred to as incompatibility genes, but they have not been incorporated into other analyses of vegetative incompatibility so their relationship to identified he: genes is unknown. The mating type locus of N. crassa also has a vegetative incompatibility function (Perkins 1988). The number of her genes was extended to 10 (hetS-hetIO) by Mylyk (1975) using duplication-producing chromosomal rearrangements which allowed the detection of segmental aneuploids 15 heterozygous at different incompatibility loci. The nature of the incompatibility reaction in N. crassa has been addressed by several studies. Gamjobst and Wilson (1956) showed that the incompatibility response followed hyphal fusion between strains differing at her-c and her-d. The physiological basis of incompatibility was addressed by Wilson et a1. (1961) who did reciprocal injection Studies using the cytoplasm from strains which represented all combinations of the alleles of her-c and her-d. Lethal cytoplasmic reactions occurred when the cytoplasm from a hypha of one strain was injected into incompatible recipient strains but no lethal reactions appeared when the cytoplasms from compatible strains were mixed. Several types of degradative treatments were also done to the extracted cytoplasms to identify the nature of the active compounds. Proteases were found to eliminate the lethal effects which indicated that the unknown factor(S) was probably a protein. Suppressor mutations of vegetative incompatibility have recently been reported in N. crassa (Arganoza et a1. , 1994b). The specificity of these mutations was found to vary greatly: some suppressor mutants suppressed only certain alleles of a her locus, others suppressed both alleles at the locus, and others suppressed the incompatibility due to heteroallelism at three her loci. Remarkably, spontaneous suppressor mutations were found to arise fairly frequently. Heterokaryon incompatibility due to the mating type genes A and a (recently termed idiomorphs) has been found to be suppressed by the to! gene (Newmeyer, 1970). Jacobson (1991) has Shown that other Neurospora species have a mutant rol gene. Introgression of the rolT gene from the pseudohomothallic Species N. tetrasperma into N. crassa resulted in suppression of mating type vegetative l6 incompatibility while introgression of the N. crassa r01c gene into N. tetrasperma allowed mating type vegetative incompatibility to appear, and thereby eliminated pseudohomothallism. Cryphonectria parasitica The genetic basis of the vegetative incompatibility system of C. parasitica has been studied by Anagnostakis (1977, 1980, 1982, 1988). The genes directly responsible for initiating the vegetative incompatibility reactions have been termed vie genes in C. parasitica rather than her genes as in Neurospora although the two designations may refer to homologous genes. Prior to the present Study two vic loci had been identified using a classical genetic approach. These two loci are unlinked and have been designated vicI and vic2 (earlier classification labelled them B and C, respectively). Each locus has two known alleles that produce an incompatible reaction through allelic interactions. Nonallelic vegetative incompatibility reactions have not been found in C. parasitica. Reference to a third incompatibility gene, vic3, has recently appeared in publication (Rizwana and Powell 1992). However, the use of the name in this paper is incorrect as demonstrated by the genetic analyses of Anagnostakis (1982) and this dissertation (chapter 3). The mating-type locus in C. parasitica does not have a vegetative incompatibility function. Individual vegetative incompatibility genotypes (polymorphisms) have been referred to as vegetative compatibility (vc) types by most authors and designated by a number or a letter. The number of vegetative incompatibility loci in C. parasirica has been estimated 17 to be between five and seven. This estimate is based upon the recovery Of 106 vc types from more than 1200 progeny from sexual crosses between vc type 5 and vc type 10 Strains (Anagnostakis 1982). If incompatibility were exclusively produced by allelic interactions, seven different vic loci would be needed to account for this many vc types (27= 128 total vc types from seven loci). Anagnostakis also estimated the total number of We genes segregating in this cross by the proportion of progeny compatible with the male parent. In this estimate only five vic loci are required to account for these ratios. Explanations offered to account for these discordant estimates were the presence of nonallelic or epistatic interactions, or different levels of vegetative incompatibility comparable to that seen in Myxomycetes. It should be noted that none of these proposed explanations was directly supported by these data. Two simpler explanations are possible. First, the proportion of progeny compatible with the male parental vc type may underrepresent the total number of vic genes segregating in the cross if some of the genes are closely linked. Secondly, the methods used to detect vegetative incompatibility are not always sufficiently sensitive (see Chapter 2). The vegetative compatibility genotypes of vc types 5 and 10 are of interest because they represent strains, relative to each other, with the largest number of heteroallelic vic loci yet known, and both of these vc types have been collected several times from natural cankers in Connecticut (Anagnostakis, 1982). Heterokaryon formation was found to occur in C. parasitica using four different auxotrophic mutants derived from a single strain and placed under selective conditions (Puhalla and Anagnostakis 1971). However, attempts at forming heterokaryons using morphological and auxotrophic mutants under nonselective conditions were not very 18 successful, and led the authors to conclude that heterokaryosis is not common in this species in nature. Anagnostakis (1981) later found a strain that was a Stable heterokaryon with regard to mating type. Parasexual recombination has recently been reported in this species after protoplast fusion (Rizwana and Powell 1995). The specificity of vegetative incompatibility in C. parasitica has been reported to be unstable after UV mutagenesis (Rizwana and Powell, 1992). Mutagenesis of We] and vic2 genes was inferred from the gain and loss of incompatiblity between strains which differed at these loci. In both cases the change was transient because the incompatibility of the mutant would revert back to its original vc type upon subculturing. Changes in vc type were also reported in this study from sectors of the strains. Two points should be kept in mind concerning this study. First, the cell death phenomenon which constitutes the vegetative incompatibility reaction is undoubtedly a complex physiological process that includes many genes. Random mutagenesis could have affected other genes involved in the regulation or expression of the incompatibility reaction. The suppressor mutations which affect vegetative incompatibility in P. anserina and N. crassa demonstrate this complexity. Secondly, genetic support for the inference that the mutagenesis specifically affected vicI and vic2 is lacking. The occurrence and horizontal transmission of viruses and genetic elements in filamentous fungi Vegetative incompatibility genes not only limit heterokaryon formation but also limit the intermycelial transmission of mobile cytoplasmic genetic elements. Ascomycetes and Basidiomycetes are parasitized by viruses and other types of genetic elements that reside 19 in the cytoplasm. Reviews of these parasitic elements and their effects upon their hosts can be found in Ghabrial (1980), Buck (1986), Nuss and Koltin (1990), Kistler and Miao (1992) and Griffiths (1995). Fungal viruses may or may not be encapsidated, and most are composed of dsRNA (Buck, 1986). Several intracellular locations have been found for fungal viruses including mitochondria (Polashock and Hillman, 1994), free in the cytoplasm (Yamashita et al. 1973), and in vesicles of host origin (Hansen et al 1985). The distribution of viruses within hyphae is known to be variable, and the titer of a virus can vary at different locations in the mycelium (Ghabrial, 1980). Intracellular distributions of viruses are also subject to the developmental state of the mycelium. Ultrastructural studies found viruses to be free in the cytoplasm in younger hyphae, but also present within membrane-bound vesicles and vacuoles in older hyphae (see references in Buck, 1986). What is the general occurrence of viruses and plasmids of fungi in nature? Survey work to meaningfully address this question is now becoming available. The most information about the diversity and abundance of parasitic genetic elements in nature has been accumulated on dsRNA viruses. Buck (1986) states that more than 100 different fungi are known to harbor viruses. The within-species diversity of viruses and dsRNA elements is best known for C. parasitica where many different dsRNAs have been found (Nuss, 1992; Paul and Fulbright, 1988; Fulbright, 1990; Enebak et al., 1994a; Enebak et al.1994b). The geographic distribution of the dsRNAs in populations of C. parasitica appears to include much of the range of the fungus in eastern North America, and C. parasitica populations in Europe, China, and Japan (Fulbright et al. , 1983; Anagnostakis, 20 1987; Heiniger and Rigling, 1994). An interesting recent Study of sequence divergence among the dsRNAs from C. parasitica isolates collected in a limited geographic area Showed genetic drift although most of the nucleotide changes did not alter the deduced translation products (Chung et al. , 1994). A survey of wildtype isolates of Neurospora showed that seven of 36 carried dsRNA (Myers et al. 1988). All of the dsRNA-containing strains were from geographically different regions, and three of the seven dsRNAs Showed cross homology. A survey of Usrilago maydis isolates from nature Showed that all of them carried dsRNA (Seroussi et al., 1989). DsRNAS have been found in all of the major taxonomic groups of the rusts and in most of the species surveyed (Zhang et al. , 1994). Plasmids also appear to be quite common in filamentous fungi (Griffiths, 1995). Fungal plasmids occur as either circular or linear types, and are almost exclusively located in the mitochondria (Griffiths, 1995). Several plant pathogenic fungi have been found to harbor plasmids, including Gaeumannomyces graminis, Fusarium oxysporum, Necrria haemarococca, and Rhizocronia solani (reviewed in Samac and Leong, 1989). A recent study of 61 field isolates of Fusariwn oxysporum from Japan found that each of six formae speciales carried a different linear plasmid (I-Iirota et al., 1992). A survey of 114 field isolates of Rhizocronia solani Showed that nearly half harbored plasmids and that homologous plasmids were frequently found in the same vegetative compatibility group (Miyasaki et al., 1990). The most extensive search for plasmids in fungi has been carried out in Neurospora species (reviewed in Griffiths, 1995). In Neurospora, seven different homology groups have been found that represent circular plasmids, and four homology groups have been 21 found among the linear plasmids (Griffiths, 1995).~ Yang and Griffiths (1993) conducted a worldwide survey with 171 strains of N. inrermedia and N. crassa and found that most of them carried either linear or circular plasmids, and some Strains carried both. Using Southern hybridizations they found that some plasmids are globally widespread. Another extensive survey of plasmids in natural populations of Neurospora species showed that the distribution of seven different homology groups is world wide and includes several species (Arganoza et al., 1994). Distinct frequency distributions have been found for two plasmids with different phenotypic effects upon a single host species. In a Hawaiian population of N. inrermedia, Debets et al. (1995) found that a cryptic circular plasmid was present in 74% of the isolates while a senescence-inducing linear plasmid was present in only 38% of the isolates, and both frequencies were maintained over time. The infectious transmission of fungal viruses and plasmids is only known to occur through hyphal fusions which permit cytoplasmic contact (plasmogarny) between individual mycelia and has been documented in a number of fungi (Table 1). Extracellular routes for infection are not known even though a few examples of lytic fungal viruses are known (Ghabrial, 1980). The cytoplasmic spread of a virus through a Single mycelium has been observed in C. parasitica to occur more rapidly than hyphal tip growth (Martin and Van Alfen, 1991; personal observations). Viral transmission can also occur vertically (intergenerationally) via asexual and sexual spores although sexual Spores are virus-free in some species (Buck 1986). What limits the horizontal transmission of viruses and genetic elements in fungi? Because infection requires intracellular passage any processes which prevent cytoplasmic 22 Table 1. Examples of horizontal (cytoplasmic) transmission of parasitic genetic elements in filamentous fungi. All transmissions were observed between individuals of a species unless otherwise indicated. fungal species genetic element‘ reference Schizophyllrmr commune VLPS and plaques Koltin et al. (1973) Rhizocronr'a solam' degenerative disease Castanho and Butler (1978) Usrilago maydis VLPs Wood and Bozarth ( 1973) Melampsom linr' dsRNA Lawrence et al. (1988) Agaricus bisporus virus Gandy (1960) Collerom'chum h'ndenmrhiamm virus Delhotal et al. (1976) Hebninrharporium vicron'ae disease Lindberg (1959) Gaeumarmomyces gramr'nis VLPs Rawlinson et al. (1973) Ophiosroma ulmi d-factors, unencapsidated Rogers et al. (1986) dsRNA Ophr'osroma nova-ubni degenerative disease Charter et al. (1993) Cryphonecrria parasitica dsRN A hypoviruses; Anagnostakis and Day (1979); respiratory defects; Mahanti et al (1993); Huber et al. senescence agent (1994); Monteiro-Vitorello et al. (1995) Aspergillus amrrelodami vegetative death (vgd) Caten (1972) . cytoplasmic mutation Aspcrgr'llus niger virus Lhoas (1970) Penicillium srolonr'ferum dsRNA virus Lhoas (1971) Penicillium chrysogenum lytic plaques Lemke et al. (1973) Padospora anserina senescence Marcou (1961) Neurospora cmssa kalr'lo DNA plasmid, Debets et al. (1994); Kinsey (1990); Neurospora inrermedia N. inter-media ~ N. cmssa senescence; Tad transposon kalr'lo DNA plasmid; mt chromosome, mt plasmids kalilo DNA plasmid Griffiths et a1. (1990) Griffiths et al. (1990); Debets et al. (1994); Collins and Saville (1990) Griffiths et al. (1990) ' Designations for the transmissible genetic elements or agents are those used by the authors. 23 contact between mycelia would effectively prevent infection. Therefore, vegetative incompatibility is generally regarded as a primary limitation. However, the effects of vegetative incompatibility upon horizontal transmission of viruses and genetic elements have not been thoroughly Studied in any fungal species. Certainly one of the hindrances has been the polygenic nature of the vegetative incompatibility systems which produce a high number of incompatibility phenotypes. The effects of vegetative incompatibility upon horizontal transmission have been considered primarily in Aspergillus amsrelodami, Ophr'osroma ulmi, N. crassa, and C. parasitica. The cytoplasmic transmission of the spontaneous mutation known as vegetative death (vgd) in A. amsrelodami was examined in relationship to different vegetative incompatibility genotypes (Caten, 1972). Vegetatively compatible Strains freely permitted transmission of the vgd mutation while differences at two or more vegetative incompatibility loci prevented transmission. Infectivity was also found to be differentially limited by specific her genes. The herB locus prevented cytoplasmic transmission, while herA reduced transmission by 80% (Caten, 1972; Handley and Caten, 1973). Caten (1972) suggested an additive effect upon the inhibition of transmission due to increasing numbers of heteroallelic her loci. Cytoplasmic transmission of the d-factor in 0. ulmi has also been found to be limited by vegetative incompatibility. Brasier (1984) has identified four different classes of vegetative incompatibility reactions in 0. ulmi that can be distinguished phenotypically: wide (w) incompatibility reaction zones, a narrow (n) reaction zones, and two types referred to as line and line-gap (l/lg) reaction zones. The w reactions were quite 24 restrictive of d-factor transmission while the n reactions allowed transmission about half of the time. Some of the weakest incompatibility reactions (l/lg) permitted d-factor transmission in 100% of the transmission tests. Genetic analysis indicated that the w reactions are caused by a single w locus, the n reactions may be caused by more than one other locus, and the lg reactions are caused by a Single weak locus. The horizontal transmission of plasmids and the effects of vegetative incompatibility upon plasmid transmission in fungi have been documented in several studies. Recently, Debets et al. (1994) tested the transmission barrier to plasmids imposed by her-c, her-d, her-e, and the mating type locus in N. crassa. The barrier created by her-c reduced transmission more than the other three loci. Mitochondrial plasmids in Neurospora have also been _ shown to transfer from one mitochondrial genotype to another at a high frequency during vegetatively incompatible mycelial interactions (Collins and Saville, 1990). This study also found a nonparental combination of nuclei and mitochondria in one conidium following an incompatible mycelial contact. Collins and Saville (1990) conclude that vegetative fusions may be an important source of mitochondrial genetic variation in natural populations. Circumstantial evidence for plasmid transmission in Neurospora in nature is also indicated by the association of plasmids with different mitochondrial DNA types, and by the occurrence of homologous plasmids in different species (Arganoza et a1. , 1994a; Yang and Griffiths, 1993). The horizontal transmission of the senescence plasmid kalr'lo has been observed in the laboratory between the species N. inremredia and N. crassa (Griffiths et al., 1990). 25 Horizontal transmission of viruses and genetic elements in C. parasitica The horizontal (cytoplasmic) transmission of viruses and genetic elements has received more attention in C. parasitica than in any other fungus due to the biological control known as transmissible hypovirulence. French workers first discovered that the hypovirulence phenotype could be transmitted to virulent strains through hyphal fusions in the laboratory, and that vinrlent strains in planra could be converted to the hypovirulent phenotype upon contact with hypovirulent strains (Grent, 1965; Grent and Berthelay- Sauret, 1978). The Spread of hypovirulence through the C. parasirica population in Italy is thought to have dramatically reduced the severity of the disease (see references in Heiniger and Rigling, 1994). Interest in the infectious spread of hypovirulence has not only been motivated by the widespread appearance of hypovirulence in Italy, but also by the localized presence of hypovirulence in North America (Fulbright et al. 1983; Griffin 1986; MacDonald and Fulbright 1991). This has provided the incentive for studies of the effects of vegetative incompatibility upon the horizontal transmission of dsRNA. How important is vegetative incompatibility in restricting the horizontal transmission of viruses in C. parasitica? Two general observations can be made. First, dsRNA viruses are able to freely pass between vegetatively compatible strains by way of hyphal fusions. Secondly, some vegetative incompatibility reactions prevent horizontal transmission, but others do not. Results obtained from dsRN A virus transmission studies have commonly relied on pairing several strains of different compatibility types harboring dsRNA with a number of strains representing many compatibility types which do not harbor dsRNA. The results repeatedly have demonstrated that horizontal transmission is variable given a 26 particular dsRNA-containing donor and various potential recipients of different vc types: some strains (vc types) will always become infected with dsRNA, other vc types will become infected at a reduced frequency, and still others will never become infected. This result has been observed many times (e. g. Anagnostakis and Day 1979; Anagnostakis 1983, 1984a; Kuhlman and Bhattacharyya 1984; Kuhlman et al. 1984). Differences in infectivity were related by Anagnostakis (1983) to "strong” versus "weak” barrage reactions which were distinguished by whether or not pycnidia formed along the border of the reaction zone. Most weak barrage reactions permitted rapid transmissiOn while strong barrages were associated with little or no infection. Recently, Liu and Milgroom (1996) have presented interesting evidence that the successive addition of heteroallelic vic loci presents an increasingly effective barrier to transmission. The weakness of this study is that the vc genotypes of the Strains are not known so that differences in transmission cannot be definitively attributed to the cumulative effects of vic loci rather than the particular effects of individual vic loci. Horizontal transmission has also been found to occur in nontransitive steps (Anagnostakis, 1983; Kuhlman et a1. 1984; Fulbright et al.,1988). Nontransitive transmission ocurs when strain A can infect strain B, and strain B can infect strain C, but strain A cannot (or infrequently) infect strain C. Anagnostakis (1983) and Fulbright et al. (1988) referred to this phenomenon as a transmission network and suggested that such networks could facilitate the movement of viruses through populations of incompatible strains. Two other studies further characterized the relationship between vegetative incompatibility polymorphisms and susceptibility to dsRNA infection by advancing two 27 concepts concerning horizontal transmission. First, Kuhlman and Bhattacharyya (1984) and Kuhlman et al. (1984) presented the concept of "broad conversion capacity" which refers to the ability of a particular dsRNA-containing strain to infect other Strains in several vegetative compatibility groups. Secondly, these studies described “conversion groups” which highlighted the potential complex nontransitive nature of transmission in populations with several incompatibility genotypes. Conversion groups are composed of more than one vegetative compatibility genotype where dsRNA transmission readily occurs among the vc types of the conversion group but generally not (or infrequently) between vc types in different conversion groups. Nine different conversion groups were described using cluster analysis based upon frequency and rate of conversion (Kuhlman et a1. , 1984). Each of the nine conversion groups overlapped with at least one other conversion group; that is, catain strains were present in more than one group, thereby linking all nine groups together in a network (Kuhlman et al., 1984). Presumably, a dsRNA virus could enter any strain in this network of conversion groups and eventually become transmitted to all of the vegetative compatibility groups within the nine conversion groups. Whether this can occur in nature has not been tested. The characterization of conversion groups also raised important conceptual and methodological issues. Kuhlman et a1. (1984) suggested that vegetative compatibility groups are not discrete units but rather form a continuum. Apparently, this concept arises from the described conversion groups and from ”multiple-merge” strains that they have identified. The multiple-merge strains were described as appearing to be vegetatively compatible ( =merging) with strains from different compatibility groups. However, they 28 describe their compatibility assays as being unreliable and inconsistent. In chapter 2, I have examined strains from some of their multiple-merge vc types and provide a different interpretation. Although the concept of conversion groups is sound (and provided with a genetic basis by this dissertation, Chapter 4), the Kuhlman et al. ( 1984) concept of a vegetative compatibility continuum based upon multiple-merge strains still lacks support. The pathogen: Cryphonecrrr’a parasitica Cophonecrria parasitica (Murr.) Barr is a filamentous ascomycete in the Diaporthales. The pathogen was originally described by Murrill (1906) and named Diaporrha parasitica Murrill. Anderson and Anderson (1912) transferred the species to the genus Endorhia. Barr (1978) produced a monograph on the Diaporthales of North America and transferred Endorhr'a parasitica and four other species of Endorhia to the genus Cryphonecrria. Subsequent studies by Micales and Stipes (1987) concur with Barr, but Griffin et al. (1986) prefer the genus Erldorhia. In Barr's (1978) classification the genus Endorhr'a is in the family Gnomoniaceae (subfamily Mamianioideae, tribe Endothieae) and Cryphonecrria is placed in a separate family, Valsaceae (subfamily Valsoideae, tribe Diaportheae). C. parasitica was introduced into the United States near the turn of the century, probably from China or Japan on imported chestnut trees (Anagnostakis, 1987). The first published report of the disease and the original description of the fungus as a new species were based on diseased American chestnut trees (Castanea denrara) within the New York Zoological Gardens (Merkel, 1906). The disease spread rapidly throughout the natural range of the American chestnut in eastern North America. Both natural agents and human 29 activities are attributed to be responsible for the rapid dissemination which by 1945 included the entire natural range of the American chestnut (Griffin, 1986). This pathogen was also introduced into Europe and first noticed on European chestnut trees (Castanea sariva) in Italy in 1938 (Griffin, 1986). The extent and severity of the disease in C. denrara and C. saliva populations rank chestnut blight as one of the worst plant disease pandemics witnessed and caused by humans. C. parasitica reproduces asexually through conidia (pycnidiospores) produced in pycnidia which develop in stroma. Sexual reproduction occurs through ascospores that are produced within perithecia embedded in the stroma. Dissemination of the conidia, which are extruded as Sticky masses on tendrils, has been documented on insects, birds and mammals (Anagnostakis, 1987). Wind dispersal of the conidia is not thought to be significant. Ascospores are ejected from perithecia and may be wind dispersed. Infection of the tree frequently occurs at the base of branches or at wound sites. The cankers that develop from virulent infections appear as depressed regions of bark where necrosis of the underlying tissues has resulted from mycelial ingress and expansion. The vascular carnbium is killed by virulent strains of the fungus so that several cankers on a large tree or single cankers on small trees will girdle and kill the distal portions of the tree. C. denrara remains as a frequent component of the eastern deciduous forest in North America because epicormic sprouts develop when the above-ground portion of the tree dies. The sprouts are, in turn, infected by the fungus after several years, then die, and are replaced by new epicormic sprouts, repeating the cycle. The mating system of C. parasitica includes both outcrossing and self-fertilization, and 30 the unusual biological phenomenon of multiple paternity has also been demonstrated in the laboratory (Anagnostakis, 1982b). A Study of the outcrossing rate by Milgroom et al. (1993) found that natural populations did exhibit a mixed mating system (both outcrossing and self-fertilization). A comparison of the genetic diversity of C. parasitica in China with the North American population using restriction fragment length polymorphisms Showed greater genetic diversity in China as would be expected due to founder effects in the North American population (Milgroom et al. , 1992). The genetic diversity of populations has also been examined using vegetative incompatibility groups. Vegetative compatibility diversity has been found to be high in some North American populations (MacDonald and Double, 1978; Anagnostakis and Kranz, 1987; Milgroom et al., 1991). However, Milgroom et al. (1993) found that the outcrossing rate suggested by molecular markers was much higher than that indicated by vegetative compatibility diversity. Other work has also shown that the vegetative compatibility diversity of a population may significantly under represent the resident genetic diversity (Liu and Milgroom, 1992). The frequency of vegetative compatibility groups has also been found to be skewed in two population surveys. Anagnostakis and Kranz (1987) and MacDonald and Double (1978) report observations of populations over several years where a single vc type was much more common than others. The diversity of vc groups has been found to be higher in Connecticut than in Europe (France, Corsica, and Italy) (Anagnostakis et al., 1986). The chestnut blight disease is of particular interest to the study of host/parasite relationships because of the occurrence of horizontally transmissible hyperparasites that reduce fungal virulence. C. parasitica was first officially recorded in Europe (Italy) on 31 Castanea sariva by Biraghi in 1938 (Heiniger and Rigling, 1994). In 1951, Biraghi found superficial, healing cankers on C. sariva. Grente (1965) later isolated unusual strains of C. parasitica from healing cankers that were reduced in virulence, and when coinoculated with normal strains in trees, resulted in the development of nonlethal cankers. The term hypovirulence was coined by Grente (1965) to describe these strains. Day et a1. (1977) found that the transmissible hypovirulence phenotype was associated with dsRNA. Proof that dsRNA caused hypovirulence was recently provided by Choi and Nuss (1992) who transformed C. parasitica with a full-length cDNA of hypovirus CHVl-713 which produced the associated hypovirulence traits and caused the reappearance of cytoplasmic dsRNA. However, not all cases of hypovirulence can be attributed to dsRNA. Transmissible hypovirulence has also been found to be caused by unknown cytoplasmic factors (Fulbright, 1985; Mahanti et al. 1993; Huber etal., 1994). In the work by Fulbright and colleagues, hypovirulent strains collected in the field were able to transmit a hypovirulence phenotype to virulent Strains but no dsRNA was detected. Some of these strains from nature have a senescence phenotype reminiscent of Neumspora senescence that is probably caused by the hypovirulence agent (Chapter 5). To test the possibility that the senescence agent causes mitochondrial dysfunction, Monteiro-Vitorello et al. (1995) induced mitochondrial mutations that debilitated the fungus, causing abnormal respiration and reducing virulence, and were horizontally transmissible. Curiously, strains infected with certain dsRNAs produce conidia bearing a nuclear mutation called flar that reduces vinrlence and segregates as a single locus (Anagnostakis, 1984b; personal observations). 32 Nuclei with the flat mutation are horizontally transmissible under laboratory conditions (personal observations). The hyperparasites . All of the viruses and viruslike genetic elements described from C. parasirica thus far are composed of dsRNA. An excellent review of these dsRNA viruses was published by N uss (1992). The recent characterization of three dsRNAs from C. parasitica that are associated with, or known to cause, hypovirulence has lead to the erection of a new virus family, the Hypoviridae (Hillman etal., 1995). This new family of viruses is interesting in several respects. The dsRNAs are unencapsidated but associated with pleomorphic membranous vesicles of host origin. The lack of a capsid precludes the dsRNAs from being infectious through extracellular routes as is typical for other viruses. Therefore hypoviruses must infect new hosts through an intracellular route, that is, through cytoplasmic contact between hosts. Fusions between hyphae of different fungal individuals provide the means for horizontal (infectious) transmission. The dsRNA genome of hypoviruses is not segmented but internal deletions will produce defective replicating molecules of dsRNA that vary in size, number, and concentration (Shapira et al. , 1991). The Structural organization of hypoviruses includes a 3' poly(A) tract base paired to a 5 ' poly(U) tract at one end, and a consensus 28 nucleotide 3' terminal sequence (Nuss, 1992). The most thoroughly characterized of the hypoviruses is CHVl-713 which is of European origin. The L dsRNA of CHV1-7l3 is 12712 base pairs long, excluding the polyA-polyU domain (Choi and Nuss, 1992). The 33 L dsRNA contains two open reading frames, each of which encodes a polyprotein. Both polyproteins undergo proteolytic (autocatalytic) processing to produce the active protein species. Sequence Similarities between several coding domains within the L—dSRNA and conserved motifs within potyvirus proteins include a putative RNA helicase domain and RNA-dependent RNA polymerase domain (Nuss, 1992). Hypovirus CHV3-GH2 from Michigan has been partially sequenced and found to have a large open reading frame with Significant Similarities to CHV1-713 (Durbahn, 1992; Smart and Fulbright, 1996). Although the hypoviruses are unlike conventional viruses in terms of lacking both a capsid and an extracellular infection capability, justification for considering them to be viruses is based upon their similarity to viral genomes in genetic organization and mode of expression (Nuss, 1992). Viruses and dsRNA elements infecting C. parasitica are widespread in the native and naturalized range of the fungus although their frequency in fungal populations can vary widely. For example, Double et al. (1985) surveyed over 1000 isolates of C. parasitica from West Virginia and only found nine that contained detectable levels of dsRNA. But two other surveys which totaled 360 isolates from virulent cankers in West Virginia, Virginia, and Maryland found that 25% of the isolates contained the SR-2 dsRNA species (Likins, 1990; Sillick and MacDonald, 1988). Hypoviruses constitute the best characterized and probably most frequent group of viruses that infect C. parasirl'ca. Studies of the relationships among dsRNA viruses and elements in several regions of North America have revealed considerable diversity. A survey of dsRNAs from six localities in Michigan showed that hypovirus CHV3-GH2 34 hybridized to dsRNAs from five of the localities (Paul and Fulbright, 1988). However, dsRNA from Michigan isolate RC1 did not hybridize to other Michigan dsRNAs or to dsRNAs from West Virginia and Tennessee (Paul and Fulbright, 1988). Further North American geographical diversity is demonstrated by the lack of cross-hybridization between dsRNAs from fungal isolates from Maryland (SR-2), Pennsylvania (D2), and West Virginia (C-18) (Enebak et al., 1994b). European and North American dsRNAs were shown to be nonhomologous in hybridization studies by L’Hostis et al. (1985), although hypovirus CHV2-NB58 from New Jersey has been found to hybridize with two dsRNAs from Europe (Hillman et al., 1992). CHV2-NB58 may have originated from earlier releases of European hypovirulent strains in the same region (Hillman et al. 1992). Recently new types of dsRN A viruses and elements have been found in C. parasitica. A hypovirulent West Virginia isolate (018) was found which contained a dsRNA similar to reoviruses (Enebak et al. 1994a). This dsRNA has 11 segments, at least seven of them are genetically unique, and it is associated with icosahedral particles in the mycelium. Another dsRNA element has been found in the mitochondria of a hypovirulent isolate from New Jersey (Polashock and Hillman, 1994). This dsRNA is small (2728-bp), lacks polyadenylated termini, is apparently unencapsidated, and is ancestrally related to the yeast cytoplasmic T and W dsRNAs rather than the Potyviridae. What effects do dsRN A viruses and elements have upon their fungal host? Are the effects the result of a general debilitation, or are specific cellular functions afflicted? Although the causal connection between dsRNA and hypovirulence in C. parasitica has been established (Choi and Nuss, 1992), the mechanism by which virulence is reduced is 35 not yet known. Infection with certain dsRNAs has been found to reduce fungal growth rate on synthetic media and live chestnut tissue, and to alter gross morphology on synthetic medium (Anagnostakis and Waggoner, 1981; Elliston, 1985; Hillman et al., 1990; Nuss, 1992). Each of these abnormalities may be related to reduced virulence. Other phenotypic and molecular alterations due to dsRNA infection that may be _ ecologically important have also been found. In particular, the production of both conidia and ascospores can be reduced by some species of hypovirus (Elliston, 1985; Nuss, 1992). The first evidence for changes of gene expression in the fungal host was provided by Powell and Van Alfen (1987a, 1987b). Exciting recent work by Choi et al. ( 1995) has Shown that the accumulation of a GTP-binding protein (G protein) is reduced in hypovirus- infected strains. This is significant because G proteins are important components of signal transduction pathways in eukaryotes, and their suppression could potentially broadly affect gene expression, hence pathogenesis, in the fungus. The reduction of several types of host molecules has also been demonstrated in hypovirus-infected, hypovirulent strains including pigment (Hillman et al., 1990), oxalate (Havir and Anagnostakis 1983; Vannini et al., 1993), extra and intracellular laccase (Choi et al. 1992; Rigling and Van Alfen 1993), a cell surface protein (Carpenter et al. 1992), cutinase (Varley et al. 1992), and a putative mating type pheromone (Zhang et al. 1993). The Specific reduction of laccase is of particular interest because it has been implicated in lignin degradation and pathogenesis in other fungi (Nuss, 1992). Choi et al. (1992) have demonstrated the repression of laccase mRNA levels by hypovirus CHV1-713. Larson et al. (1992) have Shown that dsRNA interferes with the cellular Signalling processes that are necessary for laccase induction. 36 The effects of hypoviruses and dsRNA elements upon C. parasitica are not uniform however. Some viruses such as CHV3-GH2 reduce virulence but do not reduce sporulation, pigmentation, or laccase production (Durbahn 1992). Elliston (1978, 1985) has demonstrated that dsRNAs can also have quite varied effects upon fungal virulence, ranging from severely debilitating to benign. A recently described dsRNA element from the central Appalachians had no apparent effect upon its fungal host (Enebak et al, 1994b). Vegetative incompatibility in fungi: function and consequences The biological purpose and consequences of vegetative incompatibility systems in filamentous fungi have received attention in only a few species although vc groups have been frequently used as phenotypic markers in population studies. Several hypotheses have been proposed to explain the function and evolution of vegetative incompatibility in fungi. These will be briefly discussed. I will refer to the first hypothesis as the parasite defense hypothesis. This idea was proposed by Caten (1972) who suggested that vegetative incompatibility responses evolved as a cellular defense reaction against cytoplasmic infection. Since fungal viruses and parasitic genetic elements require cytoplasmic contact between mycelia for transmission, any process that prevents or limits such contact will limit infection. Numerous studies with C. parasirica (reviewed above) and some work with 0. MM have Shown that vegetative incompatibility can limit the horizontal transmission of dsRNAs and the d— factor, respectively. Recent work with Neurospora has shown that vegetative incompatibility can limit the transmission of plasmids as well (Debets et al., 1994). 37 The second hypothesis is related to the first but sufficiently distinct to require separate treatment. I will call this hypothesis the somatic cell parasite hypothesis after Buss (1987). Cellular defense mechanisms might also have evolved to protect the genetic integrity of individuals against less fit, and thereby parasitic, somatic cells (or nuclei) rather than viruses or other genetic elements that cause reduced fitness. Buss ( 1987) has developed the idea that the evolution of cellular differentiation in multicellular organisms required a means for protecting the individual from harmful somatic cell variants that might arise. The coenOcytic nature of many fungi further provides the opportunity for nuclei and mitochondria of less fit individuals to gain reproductive access. The Situation with fungi is, by analogy at least, Similar to that of the sedentary, colonial, marine metazoa. Buss (1987) considers the restriction of somatic cell parasitism by compatibility systems in organisms capable of fusion to be a control upon the units of selection. He writes: The coupling of historecognition with intraspecific competition strongly implies that the fusion/rejection loci of clonal invertebrates are genes which act to control the units of selection. Fusion results in a competition between cell lineages, and rejection results in competition between individuals. The decision to fuse or to reject is a decision to compete at the level of the cell or at the level of the individual. (p. 150) Two studies have attempted to model the effects of vegetative incompatibility upon limiting parasitic nuclear invasion. Hartl et al. (1975) modeled the situation where a parasitic nuclear gene is competitively superior in a heterokaryon, but less fit as a homokaryon. They showed that a parasitic nuclear gene could theoretically explain the evolution of two vc groups. A recent Study by Nauta and Hoekstra (1994) considered whether protection from a parasitic nuclear gene could account for the evolution of numerous vc groups, as are found in nature. Nauta and Hoekstra found that the conditions 38 necessary for the evolution of numerous vc groups are quite restrictive, suggesting that protection from somatic cell parasites may not provide a sufficient selective pressure. Nauta and Hoekstra (1994) also modeled the effects of parasitic cytoplasmic elements upon the evolution of vc group diversity and found that selection caused by cytoplasmic parasites may be more likely to favor the evolution of high numbers of vc groups although this model was still unsatisfactory. Both of these models do not consider important aspects of the biology of vegetative incompatibility. In considering the selective pressure caused by less fit nuclear genes (nuclei), neither model addressed the situation where heterokaryons in some Ascomycetes such as Gibberellafigiilarroi are restricted to the fused cells and do not proliferate, thereby eliminating or significantly limiting the possibility of somatic cell parasitism (Puhalla and Spieth, 1985). Concerning the model based upon cytoplasmic parasites, Nauta and Hoekstra note that they did not consider that some incompatibility barriers do not prevent viral infection, as is well documented in C. parasitica. The third hypothesis suggests that intraspecific competition provides the selection pressure for the evolution of vc polymorphisms (Rayner 1991; Rayner et al. , 1984). Individuals of sessile, colonial organisms with limited substrate, such as fungi (and also plants and some marine metazoans), are incapable of relocation to new substrate in the same manner as mobile organisms. In this case, vegetative incompatibility in fungi delimits the boundaries of the foraging space for a particular individual. Pleiotropy could also explain the existence of vegetative incompatibility where the incompatibility genes have other primary cellular functions. For example, the numerous 39 loci capable of initiating vegetative incompatibility reactions could represent mutations in genes which normally function as homo- or heteromultimeric proteins with other purposes (Deleu et al., 1993). The vegetative incompatibility function of the mating type genes in Neurospora is a good example of pleiotropy (Perkins, 1988). The interesting work by Bernet and colleagues (see references in Bernet 1992a, 1992b) has also provided evidence that some vegetative incompatibility genes have cellular functions other than intraspecific fusion and rejection responses, particularly in regard to the development of reproductive organs (reviewed above). As already mentioned, the work by Turq et al. (1991) addresses the question of pleiotropic effects (i.e. cell viability, reproductive organ development), but it cannot be evaluated because of the absence of data and methods. Surprisingly, none of the works with which I am familiar have considered the possibility that the genes involved in allelic and nonallelic vegetative incompatibility may have different primary functions (delimitation of conspecifics versus development, respectively) that converge upon a common programmed cell death response. Esser (1971) and Esser and Blaich (1973, 1994) postulated that nonallelic vegetative incompatibility in P. anserina may exist to limit outcrossing by producing sterility barriers between Strains. This has been criticized because P. anserina is pseudohomothallic (Bernet, 1992a). However, Jacobson (1995) has recently demonstrated that vegetative incompatibility specifically limited outcrossing in the pseudohomothallic species N. rerrasperma. Vegetative incompatibility has not been found to limit fertilization in ascomycetes such as N. crassa and C. parasitica. Lastly, another possible function for vegetative incompatibility may be the maintenance 40 of physiological or developmental independence (Rayner, 1991; Rayner et al., 1984). Two different mycelia growing on a common substrate could experience different microenvironmental conditions that require different responses. Developmental or physiological Signals translocatcd through one mycelium in response to one environment may be inappropriate for the neighboring mycelium. Dissertation Content The questions addressed in this dissertation consider the genetic regulation of vegetative cell fusions and the concomitant effects of vegetative cell incompatibility upon the horizontal transmission of cytoplasmic genetic elements. Chapter 2 presents a new method for detecting vegetative incompatibility between strains of C. parasitica and applies this method to the evaluation of Strains purportedly exhibiting compatibility with more than one vc genotype. Chapter 3 consists of a genetic analysis which identified three new vic loci: vic3, vic4, and vic5. This Study also examined the relationship between vegetative incompatibility and heterokaryon formation using complementary pigmentation biosynthesis mutations. Chapter 4 examines the effects of individual vic genes upon the horizontal (cytoplasmic) transmission of two different dsRNA hypoviruses and an uncharacterized dsRN A element. Chapter ‘5 examines the effects of several vic genes upon the horizontal transmission of an unidentified senescence-inducing agent, and provides a preliminary physiological characterization of the senescence syndrome. 41 LITERATURE CITED Adams, 0., N. Johnson, J .F. Leslie, and LP. Hart. 1987. Heterokaryons of Gibberella zeae formed following hyphal anastomosis or protoplast fusion. Exp. Mycol. 11: 339-353. Anagnostakis, S.L. 1977. Vegetative incompatibility in Endorhia parasitica. Exp. Mycol. 1: 306-316. Anagnostakis, S.L. 1980. Notes on the genetics of Endorhr’a parasitica. Neurospora Newsletter 27: 36. Anagnostakis, S.L. 1981. A stable heterokaryon of Endorhia parasirica. Mycologia 73: 570-576. Anagnostakis, S.L. 1982a. Genetic analyses of Endorhia parasitica: linkage data for four single genes and three vegetative compatibility types. Genetics 102: 25-28. Anagnostakis, S.L. 1982b. The origin of ascogenous nuclei in Endorhia parasitica. Genetics 100: 413-416. Anagnostakis, S.L. 1983. Conversion to curative morphology in Endorhia parasitica and restriction by vegetative incompatibility. Mycologia 75: 777-780. Anagnostakis, S.L. 1984a. The mycelial biology of Endorhia parasiticaJI. vegetative incompatibility, pp. 499-507 in The Ecology and Physiology of the Fungal Mycelium, edited by DH. Jennings and A.D.M. Rayner. Cambridge University Press, Cambridge. Anagnostakis, S.L. 1984b. Nuclear gene mutations in Endorhia (Cryphonecrria) parasitica that affect morphology and virulence. Phytopathology 74: 561-565. Anagnostakis, S.L. 1987. Chestnut blight: the classical problem of an introduced pathogen. Mycologia 79: 23—37. Anagnostakis, S.L. 1988. Cryphonecrrr'a parasitica, cause of chestnut blight, pp. 123-136 in Advances in plant pathology, edited by 6.8. Sidhu. Academic Press, Ltd., London. Anagnostakis, S.L. and P.R. Day. 1979. Hypovirulence conversion in Endorhia parasirica. Phytopathology 69: 1226-1229. Anagnostakis, S.L., B. Hau, and J. Kranz. 1986. Diversity of vegetative compatibility groups of Cryphonecrrr’a parasitica in Connecticut and Europe. Plant Disease 70: 536-538. Anagnostakis, S.L. and J Kranz. 1987. Population dynamics of Cryphonecrria parasitica in a mixed-hardwood forest in Connecticut. Phytopathology 77: 751-754. 42 Anagnostakis, S.L. and P.E. Waggoner. 1981. Hypovinrlence, vegetative incompatibility, and the growth of cankers of chestnut blight. Phytopathology 71: 1198-1202. Anderson, P.J. and H.W. Anderson. 1912. Endothia virginiana. Phytopathology 2: 261- 262. Andrews, J .H. 1991. Comparative Ecology of Microorganisms and Macroorganisms. Springer-Verlag, New York. Arganoza, M.T., J. Min, Z. Hu, and RA. Aldns. 1994a. Distribution of seven homology groups of mitochondrial plasmids in Neurospora: evidence for widespread mobility between species in nature. Curr. Genet. 26: 62-73. Arganoza, M.T., J. Ohmberger, J. Min, R.A. Akins. 1994b. Suppressor mutants of Neurospora crassa that tolerate allelic differences at Single or at multiple heterokaryon incompatibility loci. Genetics 137: 731-742. Asselineau, D., J. Bernet, and J. Labarere. 1981. Protoplasmic incompatibility in Podospora anserina: possible involvement of the plasma membrane in the trigger mechanism. J. Gen. Microbiol. 125: 139-146. Barr, ME. 1978. The Diaporthales in North America. Mycologia memoir 7. J. Cramer Publish, Lehre, Germany. Beadle, G.W. and V.L. Coonradt. 1944. Heterocaryosis in Neurospora crassa. Genetics 29: 291-308. Begueret, J., B. Turcq, and C. Clave. 1994. Vegetative incompatibility in filamentous fungi: her genes begin to talk. Trends Genet. 10: 441-446. Bernet, J. 1991. Aerial organs and cell death in Podospora anserina mutants: relationship with protoplasmic incompatibility. Exp. Mycol. 15: 215-222. Bernet, J. 1992a. In Podospom amen'na, protoplasmic incompatibility genes are involved in cell death control via multiple gene interactions. Heredity 68: 79-87. Bernet, J. 1992b. A gene suppressing the allelic incompatibility specified by genes at five different loci in Podospora anserina. J. Gen. Microbiol. 138: 2567-2574. Boucherie, H. and J. Bernet. 1974. Protoplasmic incompatibility and female organ formation in Podospora anserina: properties of mutations abolishing both processes. Mol. Gen. Genet. 135: 163-174. Boucherie, H. and J. Bernet. 1980. Protoplasmic incompatibility in Podospora anserina: 43 a possible function for incompatibility genes. Genetics 96: 399-411. Brasier, C.M. 1984. Inter-mycelial recognition systems in Cerarocystis ulmi: their physiological properties and ecological importance, pp. 451-498 in The Ecology and Physiology of the Fungal Mycelium, edited by DH. Jennings and A.D.M. Rayner. Cambridge University Press, Cambridge. Buck, K.W. 1986. Fungal virology--an overview, pp.2-84 in Fungal Virology, edited by K.W. Buck. CRC Press, Boca Raton, Fla. Buller, A.H.R. 1931. Researches on fungi. Vol. 4. Longmans Green, London. Bus, L.W. 1987. The Evolution of Individuality. Princeton University Press, Princeton, New Jersey. pp. 201. Carpenter, C.E., R.J. Mueller, P. Kazmierczak, L. Zhang, D.K. Villalon, and N .K. Van Alfen. 1992. Effect of a virus on accumulation of a tissue-Specific cell-surface protein of the fungus Cryphonectria (Endorhia) parasitica. Mol. Plant-Microbe Interact. 4: 55-61. Casselton, L.A. and A. Economou. 1985. Dikaryon formation, pp. 213-229 in Developmental biology of the higher firngi, edited by D. Moore, L.A. Casselton, D.A. Wood, and J .C. Frankland. Cambridge University Press, Cambridge. Castanho, B. and EB. Butler. 1978. Rhizoctonia decline: a degenerative disease of Rhizoconia solani. Phytopathology 68: 1505-1510. Caten, C.E. 1972. Vegetative incompatibility and cytoplasmic infection in fungi. J. Gen. Microbiol. 72: 221-229. Charter, N.W., K.W. Buck, and C.M. Brasier. 1993. De-novo generation of mitochondrial DNA plasmids following cytoplasmic transmission of a degenerative disease in Ophiosroma nova-ulmi. Curr. Genet. 24: 505-514. Choi, G.H., B. Chen, and D.L. Nuss. 1995. Virus-mediated or transgenic suppression of a G-protein a subunit and attenuation of fungal virulence. Proc. Natl. Acad. Sci. USA 92: 305-309. Choi, G.H., T.G. Iarson, D.L. Nuss. 1992. Molecular analysis of the laccase gene from the chestnut blight fungus and selective suppression of its expression in an isogenic hypovirulent strain. Mol. Plant-Microbe Interact. 5: 119-128. Choi, G.H., and D.L. Nuss. 1992. Hypovirulence of chestnut blight fungus conferred by an infectious viral cDNA. Science 257: 800-803. 44 Chung, P—H., P.J. Bedker and 8.1. Hillman. 1994. Diversity of Cryphonectria parasitica hypovirulence-associated double-stranded RNAs within a chestnut population in New Jersey. Phytopathology 84: 984-990. Collins, R.A. and 8.]. Saville. 1990. Independent transfer of mitochondrial chromosomes and plasmids during unstable vegetative fusion in Neurospora. Nature 345: 177-179. Crawford, M.S., F.G. Chumley, C.G. Weaver, and B. Valent. 1986. Characterization of the heterokaryotic and vegetative diploid phases of Magnaporthe grisea. Genetics 114: 1111-1129. Day, P.R., J.A. Dodds, J.E. Elliston, R.A. Jaynes, and S.L. Anagnostakis. 1977. Double-stranded RNA in Endothia parasitica. Phytopathology 67: 1393-1396. Debets, P., X. Yang, and A.J.F. Griffiths. 1995. The dynamics of mitochondrial plasmids in a Hawaiian population of Neurospora intermedia. Curr. Genet. 29: 44-49. Debets, P., X. Yang, and A.J.F. Griffiths. 1994. Vegetative incompatibility in Neurospora: its effect on horizontal transfer of mitochondrial plasmids and senescence in natural populations. Curr. Genet. 26:113-119. Deleu, C., C. Clave, J. Begueret. 1993. A single amino acid difference is sufficient to elicit vegetative incompatibility in the fungus Podospora anserina. Genetics 135: 45-52. Delhotal, P., F. Legrand-Pernot, H. Lecoq. 1976. Etude des virus de Colletotrichum lindemuthianum. 11. Transmission des particules virales. Ann. Phytopathol. 8: 437. Double, M.L., W.L. MacDonald, and R.L. Willey. 1985. Double-stranded RNA associated with the natural population of Endothia parasitica in West Virginia. Phytopathology 75:624 (Abst.). Durbahn, C.M. 1992. Molecular characterization of dsRNA-associated hypovirulence in Michigan isolates of Cryphonectria parasitica. PhD dissertation, Michigan State University, East Lansing. Durrens, P. and J. Bernet. 1982. Podospora anserina mutations inhibiting several developmental alternatives and growth renewal. Curr. Genet. 5: 181-185. Durrens, P., F. Laigret, J. Labarere, and J. Bernet. 1979. A Podospora mutant defective in protoperithecium formation, ascospore germination, and cell regeneration. J. Bacteriol. 140: 835-842. Ellis, R.E., J. Yuan, H.R. Howitz. 1991. Mechanisms and functions of cell death. Annu. Rev. Cell Biol. 7: 663-698. 45 Elliston, LE. 1978. Pathogenicity and sporulation of normal and diseased strains of Endothia parasitica in American chestnut, pp. 95-100 in Proceedings of the American Chestnut Symposium, edited by W.L. MacDonald, F.C. Cech, J. Luchok, and H.C. Smith. West Virginia University Press, Morgantown, WV. Elliston, LE. 1985. Characteristics of dsRNA-free and dsRNA-containing strains of Endothia parasitica in relation to hypovirulence. Phytopathology 75:170-173. Enebak, S.A., B.I. Hillman, W.L. MacDonald. 1994a. A hypovirulent isolate of Cn'phonectria parasitica with multiple, genetically unique dsRNA segments. Mol. Plant Microbe Interact. 7: 590-595. Enebak, S.A., W.L. MacDonald, and BI. Hillman. 1994b. Effect of dsRNA associated with isolates of Cryphonectria parasitica from the central Appalachians and their relatedness to other dsRNAs from North America and Europe. Phytopathology 84: 528- 534. Esser, K. 1971. Breeding systems in fungi and their significance for genetic recombination. Mol. Gen. Genet. 110: 86-100. Esser, K. and R. Blaich. 1994. Heterogenic incompatibility in fungi, pp. 211-232 in fire Mycota I: Gromh, Drfi‘erentiation, and Sexuality, edited by Wessels and F. Meinhardt. Springer-Verlag, Berlin. Esser, K. and R. Blaich. 1973. Heterogenic incompatibility in plants and animals. Advances in Genetics 17:105-152. Ford, E.J., R.V. Miller, H. Gray and J .E. Sherwood. 1995. Heterokaryon formation and vegetative compatibility in Sclerotinia sclerotiorum. Mycol. Res. 99: 241-247 . Fulbright, D.W. 1985. A cytoplasmic hypovirulent strain of Endothia parasitica without double-stranded RNA (dsRNA). Phytopathology 75: 1328. (Abst.) Fulbright, D.W. 1990. Molecular basis for hypovirulence and its ecological relationships, in New Directions in Biological Control edited by R. Baker and P. Dunn. Alan R. Liss, NY. Fulbright, D.W., C.P. Paul, and S.W. Garrod. 1988. Hypovirulence: a natural control of chestnut blight, pp. 121-139 in Biocontrol of Plant Diseases Vol. 11, edited by K.G. Mukerji and K.L. Garg. CRC Press, Inc., Boca Raton, Florida. Fulbright, D.W., H.W. Weidlich, K.Z. Haufler, C.S. Thomas, and C.P. Paul. 1983. Chestnut blight and recovering American chestnut trees in Michigan. Can. J. Bot. 61:- 3164-3171. 46 Gandy, D.G. 1960. A transmissible disease of cultivated mushrooms (“watery stipe”). Ann. Appl. Biol. 48: 427. Gamjobst, .L. 1953. Genetic control of heterokaryosis in Neurospora crassa. Am. J. Bot. 40: 607-614. Gamjobst, L. 1955. Further analysis of genetic control of heterocaryosis in Neurospora crassa. Am. J. Bot. 42: 444-448. Gamjobst, L. and J .F. Wilson. 1956. Heterokaryosis and protoplasmic incompatibility in Neurospora crassa. Proc. Natl. Acad. Sci. USA 42: 613-618. Ghabrial, SA. 1980. Effects of fungal viruses on their hosts. Ann. Rev. Phytopathol. 18: 441-461. Glass, N.L. and A.G. Kuldau. 1992. Mating-type and vegetative incompatibility in filamentous ascomycetes. Ann. Rev. Phytopathol. 30: 201-224. Grent, J. 1965. Les formes Hypovirulentes d’Endothia parasitica et les espoirs dc lutte contre le chancre du chataignier. CR. Hebd. Seances Acad. Agr. France 51: 1033-1037. Grent, J. and S. Berthelay-Sauret. 1978. Biological control of chestnut blight in France, pp. 30-34 in Procwdings of the American Chestnut Symposium, edited by W.L. MacDonald, F.C. Cech, J. Luchok, and C. Smith. West Virginia University, Morgantown, WV. Griffin, D.M. and H.N. Perrin. 1960. Anastomoses in the Phycomycetes. Nature 187: 1039-1040. Griffin, G.J. 1986. Chestnut blight and its control. Hort. Rev. 8: 291-335. Griffin, G.J., J.R. Elkins, and M.K. Roane. 1986. Chestnut blight, other Endothia diseases and the genus Endothia. APS monographs. APS Press, St. Paul, Minn. Griffiths, A.J.F. 1995. Natural plasmids of filamentous fungi. Microbiol. Rev. 59: 673- 685. Griffiths, A.J.F., S.R. Kraus, R. Barton, D.A. Court, and H. Bertrand. 1990. Heterokaryotic transmission of senescence plasmid DNA in Neurospora. Curr. Genet. 17:139-145. Guttman, D.S. and DE. Dykhuizen. 1995 . Clonal divergence in Escherichia coli as a result of recombination, not mutation. Science 266: 1380-1383. 47 Handley, L. and C.E. Caten. 1973. Vegetative death: a mitochondrial mutation in Aspergillus amstelodami. Heredity 31: 136 (Abst.). Hansen, D.R., N.K. Van Alfen, K. Gillies and N.A. Powell. 1985. Naked dsRNA associated with hypovirulence of EndOthia parasitica is packaged in fungal vesicles. J. Gen. Virol. 66: 2605-2614. Hartl, D.R., E.R. Dempster, and S.W. Brown. 1975. Adaptive significance of vegetative incompatibility in Neurospora crassa. Genetics 81: 553-559. Havir, EA. and S.L. Anagnostakis. 1983. Oxalate production by virulent but not by hypovirulent strains of Endothia parasitica. Physiol. Plant Pathol. 23: 369-376. Heiniger, U. and D. Rigling. 1994. Biological control of chestnut blight in Europe. Annu. Rev. Phytopathol. 32: 581-599. Hillman, B.I., D.W. Fulbright, D.L. Nuss, and N.K. Van Alfen. 1995. Hypoviridae, pp. 261-264 in Virus Taxonomy, edited by F.A. Murphy, C.M. Fauquet, D.H.L. Bishop, S.A. Ghabrial, A.W. Jarvis, G.P. Martelli, M.A. Mayo, and MD. Summers. Springer- Verlag, New York. Hillman, B.I., R. Shapira, and D.L. Nuss. 1990. Hypovirulence-associated suppression of host functions in Ctyphonectria parasitica can be partially relieved by high light intensity. Phytopathology 80: 950-956. Hillman, B.I., Y. Tian, P.J. Bedker, and M.P. Brown. 1992. A North American hypovirulent isolate of the chestnut blight fungus with European isolate-related dsRNA. J. Gen. Virol. 73:681-686. Hirota, N.~, T. Hashiba, H. Yoshida, T. Kikumoto, and Y. Ehara. 1992. Detection and properties of plasmid-like DNA in isolates from twenty three formae speciales of Fusarium oxyspomm. Ann. Phytopathol. Soc. Jpn. 58: 386-392. Holloway, B.W. 1955. Genetic control of heterocaryosis in Neurospora crassa. Genetics 40: 117-129. Huber, D.H., D.W. Fulbright, H. Bertrand, C.M. Vitorello, J. Bell and B. Shaw. 1994. Seneseence-like phenotypes in dsRNA-free hypovirulent strains of Cryphonectria parasitica. Phytopathology 84: 1063 (Abst.). Jacobson, D.J . 1995. Sexual dysfunction associated with outcrossing in Neurospora tetrasperma, a pseudohomothallic ascomycete. Mycologia 87: 604-617. Jacobson, DJ. 1993. Evidence of nonallelic heterokaryon (vegetative) incompatibility in 48 Neurospora. Seventeenth Fungal Genetics Conference, Pacific Grove, CA. (Abst). Jacobson, DJ. 1991. Control of mating-type incompatibility by the to! gene in Neurospora crassa and N. tetrasperma. Genome 35: 347-353. Jacobson, DJ. and K.L. Klomparens. 1993. Microscopic and ultrastructural examination of heterokaryon incompatibility in partial diploids heterozygous at het loci in Neurospora crassa. Seventeenth Fungal Genetics Conference, Pacific Grove, CA. (Abst). Kidwell, MG. 1993. Lateral transfer in natural populations of eukaryotes. Annu. Rev. Genet. 27: 235-256. Kinsey, J .A. 1990. Tad, a LINE-like transposable element of Neurospora, can transpose between nuclei in heterokaryons. Genetics 126: 317-323. Kistler, H.C. and V.P.W. Miao. 1992. New modes of genetic change in filamentous fungi. Annu. Rev. Phytopathol. 30: 131-152. Koltin Y., R. Ben'ck, J. Stamberg, and Y. Ben Shaul. 1973. Virus-like particles and cytoplasmic inheritance of plaques in a higher fungus. Nature (London) New Biology 241: 108-109. Kuhlman, E.G. and H. Bhattacharyya. 1984. Vegetative compatibility and hypovirulence conversion among naturally occurring isolates of Cryphonectria parasitica. Phytopathology 74: 659-664. Kuhlman, E.G., H. Bhattacharyya, B.L. Nash, M.L. Double, W.L. MacDonald. 1984. Identifying hypovirulent isolates of Cryphonectria parasitica with broad conversion capacity. Phytopathology 74: 676-682. L'Hostis, B., S.T. Hiremath, R.E. Rhoads, and S.A. Ghabrial. 1985. Lack of sequence homology between dsRNA from European and American hypovirulent strains of Endothia parasitica. J. Gen. Virol. 66:351-355. Labarere, J. and J. Bernet. 1979. Protoplasmic incompatibility in Podospora anserina: a possible role for its associated proteolytic activity. Genetics 93: 525-537. Labarere, 1., J. Begneret, and J. Bernet. 1974. Incompatibility in Podospora anserina: comparative properties of the antagonistic cytoplasmic factors of a nonallelic system. J. Bacteriol. 120: 854-860. Lane, E.B. 1987. Somatic incompatibility in fungi and myxomycetes, pp. 239-258 in Evolutionary biology of the fitngi edited by A.D.M. Rayner, C.M. Brasier, and D. Moore. Cambridge University Press, Cambridge. 49 Larson, T.G., G.H. Choi, and D.L. Nuss. 1992. Regulatory pathways governing modulation of fungal gene expression by a virulence-attenuating mycovirus. EMBO J. 11: 4539-4548. Lawrence, G.J., M.G. Boelen, and A. Pryor. 1988. Transmission of double-stranded RNAs in flax rust, Melampsora lini. Can. J. Bot. 66:61-66. Lemke, P.A., C.H. Nash, and S.W. Pieper. 1973. Lytic plaque formation and variation in virus titre among strains of Penicillium chrysogenum. J. Gen. Microbiol. 76: 265-275. Leslie, J.F. 1993. Fungal vegetative compatibility. Annu. Rev. Phytopathol. 31: 127-150. Lhoas, P. 1971. Transmission of double stranded RNA viruses to a strain of Penicillium stolontferum through heterokaryosis. Nature 230: 248-249. Lhoas, P. 1970. Use of heterokaryosis to infect virus-free strains of Aspergillus niger. Aspergillus Newsletter 11:8-9. Likins, T.M. 1990. Occurrence of dsRNA in isolates of Endothia parasitica from sites in Michigan and West Virginia. M.S. thesis. West Virginia University, Morgantown. Lindberg, GD. 1959. A transmissible disease of Helminthosporium victoriae. Phytopathol. 49: 29. Liu, Y.C. and M.G. Milgroom. 1996. Correlation between hypovirus transmission and the number of vegetative incompatibility (vic) genes different among isolates from a natural population of Cryphonectria parasitica. Phytopathology 86: 79-86. Liu, Y.C. and M.G. Milgroom. 1992. Analysis of genetic diversity within vegetative compatibility groups of Cryphonectria parasitica using DNA fingerprinting. Phytopathology 82: 245. (Abst.) MacDonald, W.L. and M.L. Double. 1978. Frequency of vegetative compatibility types of Endothia parasitica in two areas of West Virginia, pp. 103-105 in Proceedings of the Ameriean Chestnut Symposium edited by W.L. MacDonald, F.C. Cech, J. Luchok, and H.C. Smith), West Virginia University Books, Morgantown, WV. MacDonald, W.L. and D.W. Fulbright. 1991. Biological control of chestnut blight: use and limitation of transmissible hypovirulence. Plant Disease 75: 656-661. Mahanti, N., H. Bertrand, C.B. Monteiro-Vitorello, and D.W. Fulbright. 1993. Elevated mitochondrial alternative oxidase activity in dsRNA-free, hypovirulent isolates of Cryphonectria parasitica. Physio]. Mol. Plant Pathol. 42: 455-463. 50 Maroon, D. 1961. Notion de longevite et nature cytoplasmique du determinant de senescence chez quelques champignons. Ann. Sci. Nat. Bot. 12:653-764. Martin, RM. and N.K. Van Alfen. 1991. The movement of viral-like RNA between colonies of Cryphonectria parasitica. Mol. Plant-Microbe Interact. 4: 507-511. Merkel, H.W. 1906. A deadly fungus on the American chestnut. N. Y. Zoo. Soc. Annu. Rep. 10: 97-103. Mieales, J.A. and R.J. Stipes. 1987. A reexamination of the fungal genera Cryphonectria and Endothia. Phytopathology 77: 650-654. Milgroom, M.G., W.L. MacDonald, and M.L. Double. 1991. Spatial pattern analysis of vegetative'compatibility groups in the chestnut blight fungus, Cryphonectria parasitica. Can. J. Bot. 69: 1407-1413. Milgroom, M.G., S.E. Lipari, and K. Wang. 1992. Comparison of genetic diversity in the chestnut blight fungus, Cryphonectria (Endothia) parasitica, from China and the US. Mycol. Res. 96: 1114-1120. Milgroom, M.G., S.E. Lipari, R.A. Ennos, and Y-C. Liu. 1993. Estimation of the outcrossing rate in the chestnut blight fungus, Cryphonectria parasitica. Heredity 70: 385- 392. Min, 1., C. Xu, M.P. Lamberti, M.T. Arganoza, and R.A. Akins. 1994. Cloning and mapping of heterokaryon incompatibility genes in Neurospora crassa by using ectopically integrating plasmids. Fifth International Mycological Congress, Vancouver, British Columbia, Canada, (Abst.). Miyasaki, A., C.L. Chen, and T. Hashiba. 1990. Detection and properties of plasnrid-like DNAs in isolates from nine anastomosis and intraspecific groups of Rhizoctonia solani. J. Gen. Microbiol. 136: 1791-1798. Monteiro-Vitorello, C.B.M., J.A. Bell, D.W. Fulbright, and H. Bertrand. 1995. A cyt0plasmieally transmissible hypovirulence phenotype associated with mitochondrial DNA mutations in the chestnut blight fungus Cryphonectria parasitica. Proc. Natl. Acad. Sci. USA. 92: 5935-5939. Murrill, W.A. 1906. A new chestnut disease. Torreya 6: 186-189. Myers, C.J., A.J.F. Griffiths, S.R. Kraus, and R.R. Martin. 1988. Double stranded RNA in natural isolates of Neurospora. Curr. Genet. 13: 495-501. Mylyk, C.M. 1975. Heterokaryon incompatibility genes in Neurospora crassa detected 51 using duplication producing chromosome rearrangements. Genetics 80: 107-124. Nauta, M.J. and R.F. Hoekstra. 1994. Evolution of vegetative incompatibility in filamentous ascomycetes. I. deterministic models. Evolution 48(4): 979-995. Newhouse, J .R. and W.L. MacDonald. 1991. The ultrastructure of hyphal anastomoses between vegetatively compatible and incompatible virulent and hypovirulent strains of Cryphonectria parasitica. Can J. Bot. 69: 602-614. Newmeyer, D. 1970. A suppressor of the heterokaryon incompatibility associated with mating-type in Neurospora crassa. Can. J. Genet. Cytol. 12: 914-926. Nuss, D.L. 1992. Biological control of chestnut blight: an example of virus-mediated attenuation of fungal pathogenesis. Microbiol. Rev. 56: 561-576. Nuss, D.L. and Y. Koltin. 1990. Significance of dsRNA genetic elements in plant pathogenic fungi. Annu. Rev. Phytopathol. 28: 37-58. Ohmberger, J ., M.T. Arganoza, and B. Akins. 1994. Cloning of a new het gene, het-12, in Neurospora crassa. Fifth International Mycological Congress, Vancouver, British Columbia, Canada. (Abst.). Paul, C.P. and D.W. Fulbright. 1988. Double-stranded RNA molecules from Michigan hypovirulent isolates of Endothia parasitica vary in size and sequence homology. Phytopathology 78:751-755. Perkins, DD. 1988. Main features of vegetative incompatibility in Neurospora. Fungal Genetics Newsletter 35: 44-46. Pittenger, T.H. 1964. Spontaneous alterations of heterokaryon compatibility factors in Neurospora. Genetics 50: 471-484. Polashock, 1.]. and BI. Hillman. 1994. A small mitochondrial double-stranded (ds) RNA element associated with a hypovirulent strain of the chestnut blight fungus and ancestrally related to yeast cytoplasmic T and W dsRNAs. Proc. Natl. Acad. Sci. USA 91:8680-8684. Pontecorvo, G. 1953. The genetics of Aspergillus nidulans. Adv. Genet. 5 : 141-238. Powell, W.A. and N.K. Van Alfen. 1987a. Differential accumulation of poly(A)+ RNA between vinrlent and double-stranded RNA-induced hypovirulent strains of Cryphonectria (Endothia) parasitica. Mol. Cell. Biol. 7: 3688-3693. Powell, W.A. and N .K. Van Alfen. 1987b. Two nonhomologous viruses of Cryphonectria (Endothia) parasitica reduce accumulation of specific virulence-associated polypeptides. 52 J. Bacteriol. 169: 5324-5326. Puhalla, J .E. and S.L. Anagnostakis. 1971. Genetics and nutritional requirements of Endothia parasitica. Phytopathology 61: 169-173. Puhalla, J .E. and J .E. Mayfield. 1974. The mechanism of heterokaryotic growth in Verticillium dahliae. Genetics 76: 411-422. Puhalla, LE. and P.T. Spieth. 1985. A comparison of heterokaryosis and vegetative compatibility among varieties of Gibberella fujikuroi (Fusarium moniliforme). Exp. Mycol. 9: 39-47. Rawlinson, C.J., D. Hornby, V. Pearson, and J .M. Carpenter. 1973. Virus-like particles in the Take-all fungus Gaeumannomyces graminis. Ann. Appl. Biol. 74:209-243. Rayner, A.D.M. 1991. The phytopathological significance of mycelial individualism. Annu. Rev. Phytopathol. 29: 305-323. Rayner, A.D.M., D. Coates, A.M. Ainsworth, T.J.H. Adams, E.N.D. Williams, and N.K. Todd. 1984. The biological consequences of the individualistic mycelium, pp. 509- 540 in The Ecology and Physiology of the Fungal Mycelium, edited by D.H. Jennings and A.D.M. Rayner. Cambridge University Press, Cambridge. Rees, H. and LL. links. 1952. The mechanism of variation in Penicillium heterokaryons. Proc. R. Soc. London Ser. B 140: 100-106 Rigling, D. and N .K. Van Alfen. 1991. Regulation of laccase biosynthesis in the plant- pathogenic fungus Cryphonectria parasitica by double-stranded RNA. J. Bacteriol. 173: 8000-8003. Rizet, G. 1952. Les phenomenes de barrage che Podospora anserina. I. Analyse genetique des barrages entre souches S et s. Rev. Cytol. Biol. Veget. 13: 51-92. Rizwana, R. and W.A. Powell. 1992. Ultraviolet light induced changes in vegetative compatibility groups in Cryphonectria parasitica. Phytopathology 82: 1206-1211. Rizwana, R. and W.A. Powell. 1995. Ultraviolet light-induced heterokaryon formation and parasexuality in Cryphonectria parasitica. Exp. Mycol. 19: 48-60. Rogers, H.J., K.W. Buck, and C.M. Brasier. 1986. The molecular nature of the d-factor in Ceratocystis ulmi, pp. 209-220 in Fungal Virology, edited by K.W. Buck. CRC Press, Boca Raton, Florida. Samac, D.A. and S.A. Leong. 1989. Mitochondrial plasmids of filamentous fungi: 53 characteristics and use in transformation vectors. Mol. Plant-Microbe Interact. 2: 155-159. Saupe, S., B. Turcq, and J. Begueret. 1995. Sequence diversity and unusual variability at the het-c locus involved in vegetative incompatibility in the fungus Podospora anserina. Curr. Genet. 27: 466-471. Saupe, S., C. Descamps, B. Turcq, and J. Begueret. 1994. Inactivation of the Podospora anserina incompatibility locus het-c, whose product resembles a glycolipid transfer protein, drastically impairs ascospore production. Proc. Natl. Acad. Sci. USA 91: 5927- 5931. Seroussi, E., T. Peery, I. Ginzberg, and Y. Koltin. 1989. Detection of killer-independent dsRNA plasmids in Ustilago maydis by a simple and rapid method of extraction of dsRNA. Plasmid 21: 216-225. Shapira, R., G.H. Choi, B.I. Hillman, and D.L. Nuss. 1991. The contribution of defective RNAs to the complexity of viral-encoded double-stranded RNA populations present in hypovirulent strains of the chestnut blight fungus, Cryphonectria parasitica. EMBO J .10: 741-746. Sillick, J.M. and W.L. MacDonald. 1988. The occurrence of dsRNA-containing strains of Endothia parasitica in different age chestnut stands. Phytopathology 78:863 (Abst.). Simonsen, L. 1991. The existence conditions for bacterial plasmids: theory and reality. Microbial Ecology 22: 187-205. Smart, CD. and D.W. Fulbright. 1996. Molecular biology of fungal diseases, pp. 57-70 in Molecular Biology of the Biological Control of Pests and Diseases of Plants, edited by M. Gunasekaran and DJ. Weber. CRC Press, Inc., Boca Raton, Florida. Smith, M.W., D. Feng, and R.F. Doolittle. 1992. Evolution by acquisition: the case for horizontal gene transfers. Trends Biochem. Sci. 17: 489-493. Todd, N.K. and A.D.M. Rayner. 1980. Fungal individualism. Sci. Prog. Oxford 66:331- 354. . Tolmsoff, W.J. 1983. Heteroploidy as a mechanism of variability among fungi. Annu. Rev. Phytopathol. 21: 317-340. Turcq, B., C. Deleu, M. Denayrolles, and J. Begueret. 1991. Two allelic genes responsible for vegetative incompatibility in the fungus Podospora anserina are not essential for cell viability. Mol. Gen. Genet. 228: 265-269. Turcq, B., M. Denayrolles, J. Begueret. 1990. Isolation of the two allelic incompatibility 54 genes s and S of the fungus Podosopora anserina. Curr. Genet. 17: 297-303. Vannini, A., CD. Smart, and D.W. Fulbright. 1993. The comparison of oxalic acid production in vivo and in vitro by virulent and hypovirulent Cryphonectria (Endothia) parasitica. Physio]. Mol. Plant Pathol. 43: 443-451. Varley, D.A., G.K. Podila, and S.T. Hiremath. 1992. Cutinase in Cryphonectria parasitica, the chestnut blight fungus: suppression of cutinase gene expression in isogenic hypovirulent strains containing double-stranded RNAs. Mol. Cell. Biol. 12: 4539-4544. Waksman, S.A. 1967. The Actinomycetes: A summary of current knowledge. The Ronald Press Co., New York. Wilson, J.F., L. Gamjobst, and EL. Tatum. 1961. Heterokaryon incompatibility in Neurospora crassa: micro-injection studies. Am. J. Bot. 48: 299-305. Wilson, J .F. and L. Gamjobst. 1966. A new incompatibility locus in Neurospora crassa. Genetics 53: 621-631. Wood, HA. and R.F. Bozarth. 1973. Heterokaryon transfer of virus-like particles associated with a cytoplasmically inherited determinant in Ustilago maydis. Phytopathology 63: 1019. Yamashita, S., Y. Doi, K. Yora. 1973. Intracellular appearance of Penicillium chrysogenum virus. Virology 55: 445-452. Yang, X., and A.J.F. Griffiths. 1993. Plasmid diversity in senescent and nonsenescent strains of Neurospora. Mol. Gen. Genet. 237:177-186. Zrang, L., D-H Kim, A. Churchill, Y.B. Sun, P. Razmierczak, R. Baasiri, and N.K. Van Alfen. 1993. Expression of putative mating type a pheromone genes of Cryphonectria parasitica are repressed by a fungal virus. Fungal Genetics Conference Asilomar, 17th, p. 67 (Abst.), Pacific Grove, CA. Zhang, R., M.J. Dickinson, and A. Pryor. 1994. Double-stranded RNAs in the rust fungi. Annu. Rev. Phytopathol. 32: 115-133. Chapter 2 Detection of Vegetative Incompatibility in Cryphonectn'a parasitica Abstract Vegetative incompatibility in Cryphonectria parasitica is usually identified by barrage development, that is, a zone of visibly inhibited mycelial growth where conspecific mycelia make direct contact. The development of observable barrages between certain incompatible genotypes can be difficult to identify on the traditional assay medium, potato dextrose agar. This study has found that barrage testing conducted on autoclaved chestnut cortex/ phloem tissue is more sensitive and reproducible than tests performed on potato dextrose agar. The new wood/agar vegetative compatibility (vc) assay permitted the identification of previously unrecognized nonparental vc types among the progeny of a sexual cross. Representative strains from two vc types reported to be compatible with several different vc types were found, instead, to be incompatible with these other vc types with the new assay. In addition, barrage development was detectable in strains infected with debilitating hypoviruses using the wood/agar vc assay. Vegetative (heterokaryon) incompatibility in Ascomycetes is the condition whereby stable vegetative cell fusions cannot form between conspecifics due to the presence of particular alleles at any of several incompatibility loci (Glass and Kuldau, 1992; Leslie, 1993). The incompatibility reaction which limits heterokaryon formation occurs after hyphal fusion and generally results in either cell death or heterokaryons with abnormal growth. In allelic incompatibility systems, such as in Neurospora crassa and Cryphonectria parasitica, compatible vegetative cell fusions require both interacting individuals to be homoallelic at all vegetative incompatibility loci; heteroallelism at any of the incompatibility loci results in a rejection response. Nonallelic vegetative incompatibility systems, such as in Podospora anserina, render vegetative cell fusions 55 56 incompatible when particular alleles are present at different loci. The detection of vegetative incompatibility has generally been accomplished with heterokaryon tests or barrage tests. Heterokaryon tests rely upon the complementation of mutations in compatible strains that results in a positive selection phenotype, and the absence of selected growth when complementation is prevented by incompatible cell fusions. In contrast, the barrage test detects incompatible vegetative cell fusions by the development of a zone of inhibited mycelial growth and dead cells (called a barrage) between two anastomosing mycelia, and can be done with wild type strains. The use of barrage tests to detect vegetative incompatibility depends upon the visible development of the incompatibility reaction. In Cryphonectria parasitica, the chestnut blight pathogen, barrage tests have been relied upon to test for vegetative incompatibility and to identify vegetative compatibility (vc) groups. The customary method for barrage testing with C. parasitica utilizes potato dextrose agar (PDA) as the growth substrate and subjects paired cultures to four days of growth in. the dark, followed by a 16 hour daily photoperiod for several more days (Anagnostakis 1977, 1984). Under these conditions, vegetatively incompatible strains develop a visible barrage zone that may be bordered by pycnidia where they make contact while compatible strains produce a confluent mycelium. While many workers have used this method (e.g. Anagnostakis,1977; Kuhlman et al., 1984), my experience has been that mycelial interactions are often ambiguous under these conditions, making interpretation of the interactions difficult. Atypical incompatibility reactions have also been noted in other studies. For example, compatibility groups have been described where certain 57 member strains sometimes produce barrages among themselves (Kuhlman and Bhattacharyya, 1984). In another study, the term ”multiple-merge" vc group was used for C. parasitica strains which were apparently vegetatively compatible with more than one vc group although the authors indicated that the incompatibility reactions were difficult to evaluate (Kuhlman et al., 1984). Barrage reactions have also been classified as "strong" versus ”weak” to discriminate between obvious visible differences in the development of the reaction zone (Anagnostakis, 1983). I have found that the identification of the so- called weak incompatibility reactions can be especially problematic. This work was undertaken in conjunction with a genetic analysis of vegetative incompatibility polymorphisms in C. parasitica in order to find a method for producing more distinctive and reproducible barrage reactions to facilitate identification of vegetative incompatibility genes. The enhancement of the detection of weak incompatibility reactions was of particular concern. The new technique was then applied to several instances of anomolous vegetative incompatibility reactions reported in the literature to see if these strains had unusual incompatibility phenotypes or if the anomolous incompatibility reactions had a methodological basis. MATERIALS AND METHODS Strains. Cryphonectria parasitica strains used in this study are presented in Table 1. Cryphonectria hypovirus CHV3-GH2 was originally obtained from a nonlethal canker on a chestnut tree in Grand Haven, Michigan (Fulbright et a], 1983). Cryphonectria hypovirus CHV1-713 was obtained from strain Ep713. Hypovirus CHV1-713 was 58 transferred to strain 389.7 through three steps of transmission which were accomplished by pairing a donor (hypovirus infected) strain with a virus-free recipient on potato dextrose agar (PDA). First, Ep713 was used to infect strain Ep78; the resulting strain, Ep78(713) was then used to infect P1.9. Strain P1.9(713) was used to infect 389.7. Strain P1.9 is an ascosporic progeny fi'om a cross between strains Ep155 and 80-2c. Hypovirus CHV3- GH2 was transmitted into strain 389.7 by first infecting Ep289 with CHV3-GH2 by pairing strain GH2 with Ep289. The resulting strain, Ep289(GH2), was then used to infect 389.7. Transfer of the viruses was confirmed by dsRNA extraction of the infected strains following the procedures of Morris and Dodds (1979) as modified by Fulbright et al. (1983). Sexual crosses. The protoperithecia] strain was grown on autoclaved chestnut stems about 6 cm long which were placed in disposable 15 mm x 100 mm Petri dishes with about 20 m] of molten 1.5% water agar (Sigma) to hold the stems in place. Prior to autoclaving, strips of bark were cut off the stems which exposed strips of cortex and phloem tissues alternating between strips of intact bark. Crosses were performed by spreading conidia as sperrnatia on the protoperithecia] strain that had been growing on the stems for two to three weeks. The inoculated plates were kept in an incubator with an eight hour photoperiod at 22° C. Ascospores were collected by carefully extracting intact perithecia which were embedded in mycelium beneath the bark using dissecting needles. Perithecia crushed or found to be oozing ascospores during removal were not used. The intact perithecia were then placed into a drop of sterile water on a microscope slide to wash off conidia which 59 Table 1. Strains of Cryphonectria parasitica used in this study. Strains Source‘I Vegetative Hypovirus compatibility present types " 389.7 I vc 5 - Ep243 11 VC 56 - Ep289 II vc 71 - Ep78 ATCC 38752 vc 17 - Ep155 ATCC 38755 vc 40 - Ep713 ATCC 52571 vc 40 CHV1-7l3 80-2c III unnamed P1.9 Ep155 X 80-2c recombinant - D1.29 389.7 X Ep243 recombinant - D3.31 389.7 X Ep243 recombinant - GH2 IV unnamed CHV3-GI-I2 Ep78(713) this study vc 17 CHV1-713 P1.9(713) this study recombinant CHV1-713 Ep289(GH2) this study vc 71 CHV3-GH2 389.7(GH2) this study vc 5 CHV3-GH2 389.7(713) this study vc 5 CHV1-713 16-7-3 V vc theta - 9-11-2 V vc Q - 114 II vc 7 - 157 II vc 42 - 158 II vc 43 - 159 II vc 44 - 60 Strains Source'I Vegetative Hypovirus compatibility present types " 37 II vc 19 - 241 II vc 54 - ‘ Strains are from the following sources: 1, UV mutagenized conidium from Ep389 described in chapter 3; 11, Sandra Anagnostakis, Connecticut Agricultural Experiment Station; 111, single conidium from strain 80-2 described in chapter 3; IV, Dennis Fulbright, Michigan State University; V, William MacDonald, West Virginia University. " All of the vc types designated by a number represent the Connecticut system of vegetative compatibility grouping. Vc types theta and Q represent the West Virginia system of vegetative compatibility grouping. The recombinant (nonparental) vc types are progeny from either cross 389.7 X Ep243, or progeny from cross Ep155 X 80-2c. Strain 80-2c has not been given a vc type designation although the vc genotype has been described in chapter 3. 61 may have been sticking to the surface of the perithecia. This procedure was done under a dissecting microscope (40X). If many conidia were found to wash off of a perithecium, the perithecium was transferred to one or more additional drops of sterile water to wash off the mq'ority of loose conidia. The washed perithecia were then transferred to a clean drop of water on a microscope slide and crushed with sterile dissecting needles under a dissecting scope (40X). The ascospores and asci that spilled from the ruptured perithecium were removed with a pipetman and spread with water on Petri plates of 1.5 % water agar (Sigma) supplemented with nicotinamide at 0.002 mg/ ml. Single ascospores were collected from the water agar soon after germination (24-36 hours after plating) using a dissecting scope (40X) and a finely beveled cutting tool. Ascospores germinate more quickly than conidia on water agar thereby insuring that ascospores could be distinguished from conidia] contaminants. Germinated ascospores were subsequently transferred to PDA, generally eight per plate. Barrage tests. Barrage tests were conducted on American chestnut stems harvested throughout the year. Cut stems were placed at 4° C with their bottom ends in water if they were not immediately prepared for barrage testing. The stems were cut transversely into sections about 1.5-2 cm long. The diameter of the stems varied from about 10-15 mm; the primary size limitation for the stems concerned the depth of the Petri dish. When stems were too thick to place in standard dispomble 15 mm x 100 mm Petri dishes without coming into contact with the lid, they were bissected longitudinally. To prepare the stems for barrage testing, the outer layer of periderm and bark was carefully cut off with a sharp knife, leaving an inner layer of cortex and phloem still attached to the stem. The cuts 62 were imprecise as to which tissue layers remained, although both cortex and phloem tissues remained attached to the stem to varying degrees. C. parasitica was found to grow well on the cortex and phloem tissues that occur between the secondary xylem and the outer bark, but grew poorly on the secondary xylem and the intact outer bark themselves. These small pieces of stem were autoclaved for twenty minutes at about 120° C. Eight pieces of autoclaved stem were then arranged in 1.5% molten water agar (Sigma) in disposable 15 mm x 100 mm Petri dishes with the exposed cortex/phloem surface sticking above the agar (Figure 1). The development of barrages occurred on the exposed cortex/phloem surface of the stems. Pairs of fungal strains were placed on the agar at either end of the stem pieces so that the mycelia would grow and contact each other on the stem surface. It was useful to dip the transversely cut ends of the stems in the molten agar when embedding them beeause the fungus grows across the agar more quickly than across the secondary xylem surface. With this method, eight unknown strains could be assayed against a single tester strain in one petri dish. In order to insure that the opposing strains in a barrage test made contact on the surface of the wood, it was preferable to use strains that were at approximately the same age or stage of growth (i.e. active growth or stationary phase) so that neither strain overgrew the wood faster than the other. Every barrage test was performed at least twice. Barrage tests were conducted in plates subjected to three different light regimens. Plates were either placed under a daily photoperiod of 12 hours light, or four to six days of darkness followed by a 12 hour photoperiod, or total darkness. Cultures were kept at room temperature (about 25° C). Cultures subjected to light were 63 placed under cool white flourescent lights (34 watts) at a distance of 25-30 cm. This method for detecting vegetative incompatibility will be called the wood/agar vc assay. Barrage tests were also conducted on potato dextrose agar (PDA; Difco) as described by Anagnostakis (1977, 1984). Small cubes of PDA (2 mm’) with mycelium were placed about 5-7 mm apart on PDA for the barrage tests. Inoculated plates were exposed to the same light regimens as described above and were kept at room temperature (about 25 ° C). A variation of this test was also used where red dye was added to the PDA (Rizwana and Powell, 1992; Kohn et al. , 1990). Red dye (FD&C #40) and red food color (McCormick) were both used in separate experiments (35 drops per liter PDA) to test their enhancement of the visibility of barrage reactions. RESULTS Visible vegetative incompatibility reactions were found to be more discernible when strains made vegetative contact in the wood/agar vc assay rather than in the traditional PDA vc assay. The appearance of the barrages produced by the progeny from cross 389.7 X Ep243 were of two general types. A prominent type of barrage appeared in the wood/agar vc assay after five to seven days that consisted of a conspicuous gap between the mycelia at their zone of contact (Figure 1). When strains producing this prominent barrage were subjected to the PDA vc assay, a distinct region of thin mycelium without obvious pycnidia developed as a reaction zone where the mycelia made contact after about seven days. Following 1-2 weeks exposure to light, the barrage on PDA was often bordered by rows of pycnidia. 64 A second type of barrage was also present between some progeny when they made contact in the wood/agar vc assay. This barrage was morphologically like the first, but always appeared as a narrower gap (about 0.5 mm wide) between the confronting mycelia (Figure 1). In the PDA vc assay, this weak incompatibility reaction sometimes appeared as a subtle narrow gap in the development of pycnidia across the expanse of both mycelia, but did not appear as a break in the confluence of the two mycelia. Repeated pairings of weakly incompatible strains showed that these strains would sometimes appear to be compatible, that is, lack a barrage, when tested with the PDA vc assay. Following prolonged exposure to light (two months or more), pycnidia did not develop as a border along the weak type of incompatibility reaction on PDA as they did with the strong type of incompatibility reaction. The visibility of the weak incompatibility reactions was not enhanced by the addition of either of the red dyes to PDA. The development of barrages in both the wood/agar vc assay and PDA vc assay was subject to modification by environmental conditions. Mycelia] growth in the wood/agar vc assay showed some variability that may be attributable to differences in the chestnut tissue, but this is uncertain. In some eases, mycelial growth was thin and not vigorous. When this occurred, barrages were difficult to discern and the barrage test was rejected and repeated. Barrages developed in the wood/agar vc assay under all light conditions tested. However, the weakest barrages, particularly between strain 9-11-2 (vc type Q) and strain 241 (vc type 54) were most clearly produced in complete darkness. Barrages produced in the wood/agar vc assay were most discrete when the mycelium grew robustly, and were not enhanced by the development of rows of bordering pycnidia as occurs on 65 389.7 389.7 EP243 03.31 389.7 D1.29 389.7 EP243 EP243 D3.31 EP243 D1.29 1:igure 1. Barrage tests using the wood/agar vc assay between parental strains 389.7 and Ep243 and two progeny from the cross of these strains. The order of the strains in the plate is presented below the photo. Weak and prominent incompatibility reactions are Present. The weak incompatibility reactions occur between strain pairs 389.7/D331 and lEp243/D1.29. The other incompatibility reactions are of the prominent type. 66 PDA. The barrage tests using the PDA vc assay showed variable results. Cultures kept continuously in the dark were unusable for discerning incompatibility as previously reported (Anagnostakis, 1977). Cultures subjected to either a 12 hour photoperiod, or four to six days of darkness followed by several days of a 12 hour photoperiod, were relatively comparable to each other in the development and variability of the strong incompatibility reactions. The strong type of barrage became more evident on PDA when the mycelia produced relatively more pycnidia and less aerial mycelium. Barrages on PDA were also easier to observe with increasing time: one week or more of exposure to light produced the clearest barrage development. As stated above, the weak incompatibility reactions were sometimes (but not always) visible on PDA as a subtle decrease in pycnidia at the border of the confronting mycelia. Detection of this gap in the distribution of pycnidia along the contact zone was dependent upon the presence of environmental conditions that favored pycnidia] development. Therefore, when environmental conditions were such that dense aerial mycelium developed rather than pycnidia, this weak barrage was not visible in the PDA vc assay. The wood/agar vc assay demonstrated the presence of two additional vegetative compatibility types among the progeny of a cross between strains 389.7 (vc 5) and Ep243 (vc 56) that were previously missed in sexual crosses between these two compatibility types (Anagnostakis, 1988; Huber and Fulbright, 1994). The progeny from this cross consisted of the two parental vc types which formed conspicuous barrages with the Opposite parental type, and two nonparental vc types, each of which was strongly 67 incompatible with one parental type and weakly incompatible with the other (Table 2, Figure 1). Several replicates of barrage tests of the progeny using the PDA vc assay showed that the weak incompatibility reactions were visible in some tests but not in others. The previously published reports of crosses between these two vc types indicated that they differed at a single vegetative incompatibility (vic) locus. However, this new assay indicated that strains 389.7 and Ep243 were heteroallelic at two vic loci, one which produces a relatively prominent barrage and one which produces a very weak barrage. Both the PDA vc assay and the new wood/agar vc assay were also applied to strains that represented two ”multiple-merge” vc types (Q and theta) and the vc types with which they were reported to be compatible (Kuhlman et al., 1984). Because both vc types Q and theta were reported to be mutually compatible with vc type 54, all of the vc types apparently compatible with either Q or theta were tested in all combinations to look for other unusual compatibility reactions. The wood/agar vc assay was found to resolve ten different pairings of strains from the two multiple-merge groups that the PDA vc assay typed as either compatible or equivocal (Tables 3 and 4). Vc type theta was previously reported by Kuhlman et al. (1984) as being compatible with vc types 7 and 54. I found that the PDA vc assay did show incompatibility between vc types theta and 54, but did not produce a clear barrage between theta and 7 (Table 3). However, the wood/agar vc assay produced clear barrages between these strains (Table 4). The barrage produced between vc theta and vc 7 was very narrow and similar in appearance to the weak barrages observed among some of the progeny from cross 389.7 x Ep243. Vc type Q was reported by Kuhlman et al. (1984) as being 68 Table 2. Vegetative compatibility types of the progeny from cross 389.7 X Ep243 based upon the wood/agar vc assay. parentalxumes 389 E9243 compatible progeny l4 1 1 weakly incompatible progeny' 15 16 ' A weak vegetative incompatibility reaction is characterized by a narrower barrage in the wood/agar vc assay. Each progeny that produced a weak incompatibility reaction with one parental vc type always produced a strong incompatibility reaction (broader barrage) with the other parental vc type. The progeny in each of these two nonparental vc types are compatible among themselves and strongly incompatible with the other nonparental vc type. 69 compatible with vc types 19, 42, 43, 44, and 54. I found that the PDA vc assay showed two of these combinations to be compatible and two to be equivocal (Table 3). In contrast, barrage tests using the wood/agar vc assay showed that vc type Q was incompatible with vc types 19, 42, 44, and 54 (Table 4). Vc type 54 was found to be very weakly incompatible with vc type Q such that visible barrage formation was dependent on environmental conditions: particularly weak (narrow) barrages were evident between these vc types when they were kept in total darkness, but barrages were not always visible when assays were conducted in light. Vc type Q and vc type 43 were found to be vegetatively compatible indicating that they represent a single vc type (Table 4). Barrage tests were also performed on strains which contained hypoviruses CHVl-713 and CHV3-GH2 (Figure 2). Both of these viruses are known to cause altered growth of the fungal host. Incompatibility tests conducted with the wood/agar vc assay showed that barrages were still produced by the hypovirus-infected strains. The barrages produced by the virus-infected strains were sometimes broader in appearance and less discrete than those produced by virus-free strains (Figure 2), but still discernible. Occasionally particular subcultures would grow too thinly in the wood/agar vc assay for clear barrage development, but replicate tests were always found with more vigorous growth. 70 Table 3. Barrage tests of strains representing vegetative compatibility types Q, theta, and the vc types with which they had been reported to be compatible (i.e. "multiple-merge" response) by Kuhlman et al. (1984). Barrage tests were based upon the PDA vc assay. 241 159 158 157 114 37 16-7-3 9-11-2 9—11-2 + :t + i - i - + 16-7-3 - - - - j; - + 37 - + - i — + 114 - - - - + 157 - + - + 158 i - + 159 :t + 241 + Note: + = compatible; - = incompatible; j; = results were equivocal because some replicate tests appeared compatible and others weakly incompatible. 71 Table 4. Barrage tests of strains representing vegetative compatibility types Q, theta, and the vc types with which they had been reported to be compatible (i.e. ”multiple-merge” response) by Kuhlman et a]. (1984). Barrage tests were based upon the wood/agar vc assay. 24] 159 158 157 114 37 16-7-3 9-11-2 - + + 9-11-2 16-7-3 37 - - - - - + 114 - - - - + 157 -- - - + 158 - - + 159 - + 241 + + Note: + = compatible; - = incompatible 72 EP243 389.7(713) 389.7 D3.31 389.7(713) D1.29 389.7 389.7(GH2) EP243 D3.31 389.7(GH2) D1 .29 Figure 2. Barrage tests of weak and strong incompatibility reactions using the wood/agar vc assay demonstrating the efficacy of the assay with virus-infected strains. The order of the strains in each plate'is presented to the right of the photo. A. Barrage tests using strain 389.7(713). B. Barrage tests using strain 389.7(GH2). 73 DISCUSSION This study has found that vegetative incompatibility polymorphisms in C. parasitica are more easily and reliably detected using the wood/agar vc assay rather than the traditional PDA vc assay. The new assay even detected a new vegetative incompatibility (vic) locus that was undetected in two previous studies. In addition, the new assay showed that strains representing vc types previously considered to be compatible with multiple vegetative compatibility types were found to be incompatible with these other types. A genetic analysis of vegetative incompatibility found errors in two previous studies which analyzed the compatibility types of progeny from crosses between vc types 5 and 56 (Anagnostakis, 1988; Huber and Fulbright, 1994). These studies, which relied upon PDA as the growth substrate for the barrage tests, erroneously classified the progeny as exclusively parental vc types. When the progeny of this cross were subjected to the wood/agar vc assay, two nonparental vc types appeared. Each of these new recombinant vc types produced a weak type of incompatibility reaction with one of the parental types, rendering it difficult to detect and giving the appearance of only two classes of progeny using the PDA vc assay. Weak vegetative incompatibility reactions in C. parasitica have been previously idmtified by the absence of pycnidia bordering the barrage (Anagnostakis, 1983, 1988). The class of weak incompatibility reactions identified in this study using the wood/agar vc assay not only lacks pycnidia production, but also typically does not produce a distinct zone of inhibited mycelial growth on PDA as is characteristic of barrages. Differences in the visible manifestations of vegetative incompatibility reactions have also been observed in other Ascomycete species. For example, four morphological types 74 of barrage-like reactions have been described in Ophiostoma ulmi that differ in the extent to which the mycelium is affected (Brasier, 1984). These morphological differences in the incompatibility reactions in 0. ulmi have been attributed to the effects of different vegetative incompatibility genes (Brasier, 1984). Strains of C. parasitica that were apparently vegetatively compatible with several different vc types have been reported by Kuhlman et al. (1984). These strains were referred to as having a ”multiple-merge" capability which refers to their failure to develop barrages with mutually incompatible strains. A reevaluation of the incompatibility relationships among strains representing these vc types using the PDA vc assay demonstrated several instances of either compatible or equivocal interactions. However, repeating these tests with the new wood/agar vc assay demonstrated that the equivocal incompatibility reactions, and three cases of compatible reactions, were actually weakly incompatible. The weakest incompatibility reaction (i.e. least prominent barrage) occurred between strains in vc types 54 and Q. The barrage produced between these strains was consistently observed when they were grown on chestnut tissue without light; but when grown with light, the visible barrage was sometimes not present. These analyses show that the weakest incompatibility reactions are the most easily obscured by environmental conditions. The observations by Kuhlman and Bhattacharyya (1984) that replications of incompatibility tests between certain strains did not consistently show incompatibility are similar to my observations of weak interactions. Proffer and Hart (1988) have also observed the mutiple-merge phenomenon among strains of Leucocytospora kunzei , but they did not report attempts to vary the conditions under which incompatibility was observed. 75 Are the vegetative incompatibility groups in C. parasitica discrete or do nontransitive effects mitigate incompatibility? Kuhlman eta]. ( 1984) concluded that the existence of multiple-merge vc types demonstrates that the vegetative incompatibility system in C. parasitica does not produce discrete incompatibility groups, but instead produces a continuum of overlapping vc groups. My evaluation of strains from the putative multiple- merge vc types Q and theta does not support this concept. However, the suppression of specific vegetative incompatibility genes is conceivable as a consequence of suppressor mutations such as those recently found in Neurospora crassa that prevent vegetative incompatibility between certain incompatibility genotypes (Arganoza et a]. , 1994). The presence of similar mutations in natural populations of C. parasitica is unknown but a formal possibility. Some of the strains used in the studies by Kuhlman and Bhattacharrya ( 1984) and Kuhlman et al. (1984) could earry such mutations. It should also be noted that the strains used in this study representing vc types Q and theta are not the same strains used by Kuhlman et a]. (1984). Therefore, it is possible that differences exist between individual strains within vc types Q and theta. In conclusion, the published reports of the multiple-merge phenomenon and the failure to consistently detect barrages between strains may be the result of two simultaneously confounding factors: differences between strains at weak incompatibility loci, and environmentally-induced inconsistencies in growth. A reevaluation of the other strains which fit the multiple-merge phenomenon is needed before the concept of nondiscrete vegetative compatibility types is accepted. The possibility of suppressor mutations should also be considered. The infection of fungal strains with dsRNA viruses can also contribute to the apparent 76 nondiscrete quality of some vegetative compatibility types in two ways. First, viral infection can significantly affect the growth and morphology of the mycelium, thereby obscuring barrage development on PDA (Anagnostakis, 1977). However, this problem was found to be diminished with the new wood/agar vc assay. Secondly, the horizontal transmission of viruses between different vegetative compatibility types may approximate a continuum due to nontransitive transmission effects. Here, the horizontal transmission of viruses needs to be kept conceptually distinct from vegetative incompatibility. While virus transmission is inhibited by some vic loci and modifiable by epistatic interactions between vic genes, it has not been found to be inhibited between all vegetative incompatibility polymorphisms (Anagnostakis, 1983; Fulbright et al., 1988; this dissertation). The development of visible vegetative incompatibility reactions as a function of the growth environment is not unique to Cryphonectria parasitica. Vegetative incompatibility tests for Ophiostoma ulmi are conducted on elm sapwood agar medium because some types of incompatibility reactions cannot be observed on other media, such as potato dextrose agar, earrot agar, and a malt extract medium (Brasier, 1984). Croft and Dales ( 1984) have reported that heterokaryon incompatibility sometimes occurs in Aspergillus without any apparent mycelial reaction. Brasier (1984) has suggested that this situation with Aspergillus may also be due to the type of medium used. Visible vegetative incompatibility reactions are also not present between incompatible strains of Cochliobolus heterostrophus when grown on an agar medium (leach and Yoder, 1983). Previous work on C. parasitica has demonstrated that the light regimen can influence barrage 77 development (Anagnostakis, 1977). The influence of the environment upon the development of visible barrages does not imply that the vegetative incompatibility reactions themselves are different at the cellular level. Barrages are a visible phenotype, produced in response to the cell death which occurs between fused incompatible cells, that would depend upon the degree of hyphal contact between confronting mycelia. Less contact between mycelia would provide less opportunity for hyphal fusions and, presumably, correspondingly fewer cells involved in the incompatibility response. The frequency of hyphal fusions has been shown to be affected by growth substrate in the Basidiomycete Corticium vellereum (Bourchier, 1957). It has not been tested whether the frequency of hyphal fusions is affected by growth substrate in C. parasitica. Barrage development on the chestnut cortex/phloem tissue may be more pronounced because of denser mycelial growth, alterations in the frequency of hyphal fusions, or unknown factors. An environmental influence upon the physiology of the incompatibility reaction itself has not been investigated to my knowledge. Since the detection of weak vegetative incompatibility reactions in C. parasitica using the PDA vc assay has been found to be unreliable, the use of the wood/agar vc assay is recommended for vegetative incompatibility testing. 78 LITERATURE CITED Anagnostakis, S.L. 1977. Vegetative incompatibility in Endothia parasitica. Exp. Mycol. 1: 306-316. Anagnostakis, S.L. 1983. Conversion to curative morphology in Endothia parasitica and its restriction by vegetative compatibility. Mycologia 75: 777-780. Anagnostakis, S.L. 1984. The mycelial biology of Endothia parasitica. II. Vegetative incompatibility, pp. 499-507 in The Ecology and Physiology of the Fungal Mycelium, edited by D.H. Jennings and A.D.M. Rayner. Cambridge University Press, Cambridge. Anagnostakis, S.L. 1988. Cryphonectria parasitica, cause of chestnut blight, pp. 123—136 in Advances in Plant Pathology, vol. 6, Genetics of Plant Pathogenic Fungi, edited by G.S. Sidhu. Academic Press. Arganoza, M.T., J. Ohmberger, J. Min, R.A. Akins. 1994. Suppressor mutants of Neurospora crassa that tolerate allelic differences at single or at multiple heterokaryon incompatibility loci. Genetics 137 : 731-742. Bourchier, R.J. 1957. Variation in cultural conditions and its effect on hypha] fusion in Corticium vellereum. Mycologia 49:20-28. Brasier, C.M. 1984. Inter-mycelial recognition systems in Ceratocystis ulmi: their physiological properties and ecological importance, pp. 451-497 in The Ecology and Physiology of the Fungal Mycelium edited by D.H. Jennings and A.D.M. Rayner. Cambridge University Press, Cambridge. Croft and Dales. 1984. Mycelia] interactions and nritochondrial inheritance in Aspergillus, pp. 433-450 in The Ecology and Physiology of the Fungal Mycelium edited by D.H. Jennings and A.D.M. Rayner. Cambridge University Press, Cambridge. Fulbright, D.W., C.P. Paul, and S.W. Garrod. 1988. Hypovirulence: a natural control of chestnut blight, pp. 121-139 in Biocontrol of Plant Diseases, Vol. 11 edited by K.G. Mukerji and K.L. Garg. CRC Press, Inc., Boca Raton, Florida. Fulbright, D.W., Weidlich, W.H., Haufler, K.Z., Thomas, CS, and Paul, C.P. 1983. Chestnut blight and recovering American chestnut trees in Michigan. Can. J. Bot. 61: 3164-3171. Glass, N .L. and Kuldau, G.A. 1992. Mating type and vegetative incompatibility in filamentous ascomycetes. Annu. Rev. Phytopathol. 30: 201-24. Huber, D.H. and D.W. Fulbright. 1994. Preliminary investigations on the effect of 79 individual vic genes upon transmission of dsRNA in Cryphonectria parasitica, pp. 15-19 in Proceedings of the American Chestnut Symposium edited by M. L. Double and W.L. MacDonald. West Virginia University Books, Morgantown, WV. Kohn, L.M., Carbone, 1., Anderson, J .B. 1990. Mycelia] interactions in Sclerotinia sclerotiorum. Exp. Mycol. 14: 255-267. Kuhlman, E.G., and Bhattacharyya, H. 1984. Vegetative compatibility and hypovirulence conversion among naturally occurring isolates of Cryphonectria parasitica. Phytopathology 74: 659-664. Kuhlman, E.G., Bhattacharyya, H., Nash, B.L., Double, M.L., and MacDonald, W.L. 1984. Identifying hypovirulent isolates of Cryphonectria parasitica with broad conversion capacity. Phytopathology 74: 676-682. Leach, J. and O.C. Yoder. 1983. Heterokaryon incompatibility in the plant-pathogenic fungus Cochliobolus heterostrophus. J. Heredity 74: 149-152. Leslie, J.F. 1993. Fungal vegetative compatibility. Annu. Rev. Phytopathol. 31: 127-50. Morris, TI. and J .A. Dodds. 1979. Isolation and analysis of double-stranded RNA from virus-infected plant and fungal tissue. Phytopathology 69: 854-858. Proffer, TL and J .H. Hart. 1988. Vegetative compatibility groups in Leucocytospora kunzei. Phytopathology 78: 256-260. Rizwana, R., and Powell, W.A. 1992. Ultraviolet light-induced instability of vegetative compatibility groups of Cryphonectria parasitica. Phytopathology 82: 1206-121]. Chapter 3 Vegetative Incompatibility Polymorphisms in Cryphonectria parasitica: new vic loci and heterokaryon formation. Abstract The genetic basis of vegetative incompatibility polymorphisms and their effects upon heterokaryon formation were examined in the chestnut blight pathogen, Cryphonectria parasitica. Three new vegetative incompatibility (vic) loci, designated vic3, vic4, and vic5, were identified; two alleles were found at each of the three new vic loci. These three loci, like vicI and vic2, function in an allelic manner: compatibility requires that each vic locus be homoallelic, whereas heteroallelism at any vic locus produces an incompatible reaction. Vegetatively compatible strains were found to form heterokaryons when grown under nonselective conditions on PDA and in the wood/agar vc assay. Two pigmentation mutations, .cre (cream) and br (brown), were found to be tightly linked and to complement and form the wild type orange color when they occurred as heterokaryons or as meiotic recombinants. Hyphal tips subcultured from heterokaryons carried both nuclear types, and conidia consisted of either single nuclear type. Nuclear ratios in the hyphal tips from the heterokaryons varied, and corresponded to variations in mycelial color ranging from orange to orange/brown to brown. Heterokaryons were prevented from forming between strains heteroallelic at vicI, vic2, vic3, or vic4. Heteroallelism at vic4 is occasionally associated with abnormal, flat sectors between anastomosing mycelia. Heteroallelism at vic5 prevented heterokaryon formation in the wood/agar vc assay, but morphologically abnormal orange mycelium frequently formed between strains grown on PDA. The effects of 1ch , vic2, and via? upon heterokaryon formation were epistatic over the less restrictive effects of vic4 and vic5. The effect of heteroallelism at both vic4 and vic5 was considered additive since no abnormal sectors appeared between contacting mycelia. Filamentous Ascomycetes exhibit the ability to discriminate between conspecifics when they differ from each other at specific genetic loci (Glass and Kuldau, 1992; Leslie, 1993). Vegetative cell fusions between genetically compatible conspecifics may permit the formation of heterokaryons and heteroplasmons whereas genetic incompatibility will limit the formation of these chimeric genetic states. The vegetative or heterokaryon 80 81 incompatibility systems in Ascomycetes which mediate these cellular acceptance/ rejection responses have been found to be controlled by numerous genes. Both allelic and nonallelic gene interactions have been found that initiate incompatibility reactions. Allelic recognition systems, such as in Neurospora crassa and Cryphonectria parasitica, require homoallelism at each of the compatibility loci to produce compatible vegetative cell fusions; incompatible cellular fusions follow from heteroallelism at any of these loci. Nonallelic vegetative incompatibility, such as in Podospora anserina, occurs when particular alleles at different loci interact. Genetic analyses of vegetative incompatibility in Ascomycetes have shown that numerous genetic loci typically control the elicitation of the incompatibility reaction. In N. crassa ten het (heterokaryon incompatibility) loci have been identified (Perkins, 1988); in Aspergillus nidulans eight het loci have been found (Anwar et al., 1993). In natural populations of Podospora anserina, five incompatibility loci that cause allelic incompatibility, and five that cause nonallelic incompatibility, have been identified; one of these loci is involved in both incompatibility systems (Bernet, 1992). Previous studies of C. parasitica have only identified and named two vic loci, although estimates that at least seven vic loci exist have been made based upon the number of nonparental vc types that have segregated from sexual crosses (Anagnostakis, 1982). What is the relationship between vegetative cell incompatibility and the maintenance of individual genetic integrity in filamentous Ascomycetes? Fusions between vegetative cells allow the opportunity for protoplasmic contents to move from one individual (genotype) to another. Organelles and genetic elements that have been observed to move 82 through vegetative hyphal fusions include nuclei (Beadle and Coonradt, 1944), mitochondria] chromosomes (Collins and Saville, 1990; Mahanti and Fulbright, 1995), plasmids (Griffiths et al., 1990), and viruses (Day et a1. 1977; Buck, 1986; Nuss and Koltin, 1990). The effects of vegetative incompatibility upon the horizontal transmission of nuclei and cytoplasmic genetic elements are generally restrictive although many interesting exceptions have been found that indicate that incompatibility has quite variable effects upon transmission. For example, under laboratory conditions evidence indicates that individual incompatibility genes can have different effects upon the horizontal transmission of nuclei (Pittenger and Brawner, 1961), plasmids (Debets et al., 1994), viruses (Anagnostakis, 1983; Brasier, 1984) and other genetic elements (Handley and Caten, 1973). The effects of vegetative incompatibility upon cytoplasmic transmission are of particular interest in Cryphonectria parasitica (Murr.) Barr. , the chestnut blight pathogen, where several different types of transmissible genetic elements capable of reducing fungal virulence have been found (Nuss, 1992; Monteiro-Vitorello et a], 1995; this study). Cryphonectria parasitica (Ascomycota, Diaporthales) was introduced into North Ameriea near the turn of the century and subsequently infectiously spread throughout the Castanea dentata (American chestnut) population in about four decades (Anagnostakis, 1987; Griffrn, 1986). The blight has also widely infected Castanea sativa in EurOpe (Heiniger and Rigling, 1994). Subsequently, a natural form of biological control which reduces fungal virulence has appeared on both continents. The primary agents known to be responsible for reducing fungal aggressiveness are cytoplasmically transmissible dsRNA 83 viruses (hypoviruses). These viruses have spread through much of the C. parasitica population in Europe but are only locally present in the North American population (MacDonald and Fulbright, 1991; Heiniger and Rigling, 1994). The lack of infectious spread of hypoviruses in the North American population of C. parasitica has been attributed to the high number of resident vegetative incompatibility polymorphisms which have been thought to significantly restrict cytoplasmic transmission (Anagnostakis et al., 1986). However, confounding the attribution of reduced virus infectivity to vegetative incompatibility is the variability in transmission efficiency found among the many vegetative incompatibility polymorphisms in the North American population. This suggests that an understanding of the effects of vegetative incompatibility upon horizontal transmission depends upon understanding the effects of individual incompatibility genes. Before the effects of vegetative incompatibility upon genotype integrity could be fully examined, it was imperative to identify all heteroallelic vic loci in each strain. Therefore, I sought to identify additional vic loci in C. parasitica for the purpose of studying their individual contributions to the maintenance of genotype integrity. In pursuit of this objective, I have been able to identify and name three new vic loci, and to examine the relationship between vegetative incompatibility and heterokaryon formation. 34 MATERIALS AND METHODS C. pamsitica strains and culture conditions. The strains of C. parasitica used in this study, including their genetic markers and sources are listed in Table 1. Strain 80-2c is a single conidial isolate derived from strain 80-2. Strain 80-2 was collected from the field in West Virginia by W. L. MacDonald (West Virginia University) and has an irregular phenotype that varies from a flat morphology to abundant aerial mycelium. This strain has a variable color ranging from brown to white, often showing a blotchy appearance. It also harbors an uncharacterized dsRNA that probably originated from a dsRNA of Italian origin (Euro 7) used in field studies in West Virginia (W. L. MacDonald, personal communication). Strain 80-20 lacks dsRNA, has a normal growth phenotype, and has a dark brown pigmentation mutation (br). All C. parasitica strains were grown and stored on potato dextrose agar (PDA; Difco, Detroit, M1) at 4° C. Actively growing cultures were kept under cool white flourescent lights (34 watt) at a distance of 25-30 cm at room temperature (25° C). Long term storage was accomplished using the silica gel method used for Neurospora (Perkins, 1977). Sexual crosses. The protoperithecial strain was grown on autoclaved chestnut stems about 6 cm long which were placed in disposable 15 mm x 100 mm Petri dishes with about 20 ml of molten 1.5% water agar (Sigma) to hold the stems in place. Prior to autoclaving, strips of bark were cut off the stems to expose strips of cortex and phloem tissues alternating between strips of intact bark. Crosses were performed by spreading conidia as spermatia on the protoperithecial strain that had been growing for two to three weeks. The inoculated plates were kept in an incubator with an eight hour photoperiod at 22° C. 85 Perithecia developed most abundantly along the margin of intact bark and exposed cortex/phloem tissues. The time between application of the conidia (spermatia) and the appearance of perithecia was quite variable and ranged from three weeks to more than eight weeks. Ascospores were collected by carefully extracting intact perithecia which were embedded in mycelium beneath the bark using dissecting needles. Perithecia crushed or found to be oozing ascospores during removal were not used. The intact perithecia were then placed into a drop of sterile water on a microscope slide to wash off conidia which may have been sticking to the surface of the perithecia. This procedure was done under a dissecting microscope (40X). If many conidia were found to wash off of a perithecium, the perithecium was transferred to one or more additional drops of sterile water to wash off the majority of loose conidia. The washed perithecia were then transferred to a clean drop of water on a microscope slide and crushed with sterile dissecting needles under a dissecting scope (40X). The ascospores and asci that spilled from the ruptured perithecium were removed with a pipetman and spread with water on Petri plates of 1.5 % water agar (Sigma) or 1.5% noble agar (Difco; Detroit, MI). The water agar or noble agar were appropriately supplemented with growth nutrients when the progeny carried any of the auxotrophic mutations. Methionine was added to the noble agar at 0.1 mg/ml, and nicotinarnide was added at 0.002 mg/ ml. Single ascospores were collected from the agar soon after germination (24-36 hours after plating) using a dissecting scope (40X) and a finely beveled cutting tool. Ascospores germinate more quickly than conidia on water agar thereby insuring that ascospores could be distinguished from conidial contaminants. 86 Germinated ascospores were subsequently transferred to PDA, generally eight per plate. Mutagenes's. Conidia from strain EP389 were subjected to UV mutagenesis to obtain an auxotrophic mutation for use as an additional genetic marker. Conidia were collected from EP389 and 2x108 conidia were resuspended in 20 ml of sterile, distilled water in a 15 X 100 mm plastic disposable Petri dish. The conidia were exposed to 254 nm UV-light which was held 12 cm above the petri dish. During mutagenesis, conidia were kept suspended in the water with a stir bar. A red light was used for viewing the procedures in the dark. Conidia were exposed to UV light for 160 seconds which produced a 90% kill. The mutagenized conidia were then diluted and spread on plates of PDA (about 100 per plate) amended with sodium desoxycholate (50 mg/ml) to reduce colony size. Individual colonies which appeared after one week were transferred to a separate Petri plate containing water agar. Colonies which failed to grow on water agar were subsequently tested for specific nutritional requirements on water agar supplemented with various amino acids and vitamins following the method of Holliday (1956). A new strain designated 389.7 was thereby created that carried a nicotinamide requirement (nic) that segregated as a single locus (Tables 1 and 9). Heterokaryon determination. When vegetatively compatible brown colored (hr) and cream colored (cm) progeny derived from cross 389.7 cre X 80-2c br grew in contact with each other, orange color deve10ped at the confluence of the two mycelia. To determine if this orange mycelium was a heterokaryon composed of the two different mutant nuclei, hyphal tips were collected from subcultures of the orange mycelium that had been placed upon water agar supplemented with methionine (0.1 mg/ml) or nicotinamide (0.002 87 mg/ml) as needed and allowed to grow for one to two weeks. Hyphal tips about 1 mm long were cut off along with small blocks of agar and placed on PDA. The cultures were kept under lights as described above which enhanced conidiation. Conidia from the hyphal tip cultures were collected with a sterile loop, resuspended in sterile water, and spread on PDA. Individual conidia were collected soon after germination (24-30 hours after being spread on PDA) and subcultured onto PDA (eight per plate). The conidial cultures were kept under cool white fluorescent lights at room temperature (25° C) until their color developed. Vegetative incompatibility determination. Barrage tests were performed using the wood/agar vc assay described in chapter 2 rather than the traditional PDA vc assay described by Anagnostakis (1977, 1988). Heterokaryon determination was conducted under nonselective conditions on PDA which relied upon the complementation of ore and br color mutations. Vegetatively compatible ore and br strains developed orange colored mycelium at the confluence of the mycelia. Subcultures of cream and brown colored strains were placed about 5mm apart near one edge of a PDA plate and placed under cool white flourescent lights as described above. These strains grew across the PDA in continuous contact with each other. After 7 to 10 days, the color of the strains had developed sufficiently to clearly detect the orange color if present. Vegetatively incompatible cre and br strains did not develop orange mycelium when they made contact. When scoring the vic genotype of ascosporic progeny, up to eight pairings of unknown strains with tester strains were performed in one Petri plate. This was accomplished by pairing a subculture of an unknown strain about 5 mm from a subculture of a tester strain. Table 1. Strains used in this study. =—-=-= __Strain W‘ Snares" EP389 vicI-I, 2-1, 3-1, 4—1, 5-1, cre, MATT-l ATCC 38980 389.7 vicI-I, 2-1, 3-1, 4-1, 5-1, cre, nic, MATT-I UV mutagenesis of EP389 EPBS8 vicI-Z, 2-1, 3-1, 4-1, 5-1, met, MATI-Z ATCC 38979 EP289 vicI-I. 2-2, 3-1, 4-1, 5-1, met, MATI-Z Conn. Agri. Exp. Sta.c 22508 vicI-Z, 2-2, 3-1, 4-1, 5-1, met, MATI-Z ATCC 22508 EP243 vicI-I, 2-1, 3-1, 4-2, 5-2, MATI-Z Conn. Agri. Exp. Sta.c EPZ9 Viol-2, 2-2, 3-2, 4-1, 5-2, MATI-I, MATT-2° ATCC 38754 EPlSS MATI-Z ATCC 38755 80-2c vicI-Z, 2-1, 3-1, 4-2, 5-2, br, MATI-Z conidium of 80-2c Al.8 vicI-Z, 2-2, 3-1, 4-1, 5-1, cre, met 389.7 X 22508 Al.lO vicI-2, 2-1, 3-1, 4-1, 5-1, cre, nic 389.7 X 22508 Al.l3 vicI-I, 2-2, 3-1, 4-1, 5-1, cre, met 389.7 X 22508 F2.4 vicI-I, 2-1, 3-1, 4—1, 5-2, are, nic 389.7 X 80-2c F2.13 Viol-2, 2-1, 3-1, 4-1, 5-2, cre, nic 389.7 X 80-2c F2.l4 _ vicI-Z. 2-1. 3-1, 4-2, 5-2, cre, nic 389.7 X 80-2c F2.36 vicI-I, 2-1, 3-1, 4-1, 5-1, hr 389.7 X 80-2c F3.2 vicI-I, 2-1, 3-1, 4-2, 5-1, cre 389.7 X 80-2c F3.4 vial-2, 2-1, 3-1, 4-1, 5-1, br, nic 389.7 X 80-2c F3.10 vicI-I, 2-1, 3-1, 4-1, 5-2, br, nic 389.7 X 80-2c F3.]2 vicI-I, 2-1, 3-1, 4-2, 5-1, crc, nic 389.7 X 80-2c F3. 13 vicI-Z, 2-1, 3-1, 4-2, 5-1, cre 389.7 X 80-2c F3. 15 vicI-Z, 2-1, 3-1, 4-1, 5-2, br, nic 389.7 X 80-2c F3.]6 vicI-Z, 2-1, 3-1, 4-2, 5-1, br, nic 389.7 X 80-2c F321 . vicI-I, 2-1, 3-1, 4-2, 5-2, cre, nic 389.7 X 80-2c 173.26 vicI-I, 2-1, 3-1. 4-2, 5-1, br, nic 389.7 X 80-2c F339 vicI-I, 2-1, 3-1, 4-2, 5-2, br, nic 389.7 X 80-2c F4.9 vicI-I, 2-1, 3-1. 4-1, 5-2, br, MATT-2 389.7 X 80-2c F4.]3 vicl-I, 2-1, 3-1, 4-1, 5-1, br, MATT-2 389.7 X 80-2c 12.6 vicI-I, 2-2, 3-1, 4-2, 5-1, cre, nic F3.]6 X Al.13 Table 1 (cont‘d). 89 12.20 Viol-2, 2-2, 3-1, 4-2, 5-1, ere, nic F3.]6 X Al.l3 12.23 vicI-Z, 2-2, 3-1, 4-2, 5-1, br, nic F3.]6 X Al.l3 12.34 vicI-I, 2-2, 3-1. 4—2, 5-I, br, met F3.]6 X A1.l3 12.43 vicl-I, 2-2, 3-1, 4—1, 5-1, hr F3.]6 X Al.13 12.57 vicI-Z, 2-2, 3-1, 4-1, 5-1, br, met, nic F3.]6 X Al.l3 12.69 vicI-I, 2-2, 3-1, 4-1, 5-1, br, nic F3.]6 X Al.l3 Kl.9 nic' EP29 X A].10 K132 cre, nic’ EP29 X A].10 Kl.43 vicI-Z, 2-1, 3-2, 4-1, 5-1, cre EP29 X Al.10 K230 vicI-Z, 2-2, 3-2, 4-1, 5-1, or: EP29 X A1.10 K239 cref EP29 X Al.10 Ll.6 Viol-2, 2-2, 3-1, 4—2, 5-1, cre 12.23 X K230 Ll.8 Viol-2, 2-2, 3-2, 4—1, 5-1, br, nic 12.23 X K230 1.1.17 vicI-Z, 2-2, 3-2, 4—2, 5-1, cre 12.23 X K230 L139 vicI-Z, 2-2, 3-2, 4—2, 5-1, hr, nic 12.23 X K230 M 1.5 vicI-I, 2-2, 3-2, 4-1, 5-1, br, nic K230 X 12.69 Ml.6 vicI-l, 2-2, 3-2, 4-1, 5-1, are K230 X 12.69 Ml.21 vicl-Z, 2-2, 3-2, 4-1, 5-1, hr K230 X 12.69 Nl.9 vicI-I, 2-1, 3-2, 4-1, 5-1, cre F4.]3 X Kl.43 N1.19 vicI-I, 2-1, 3-2, 4-1, 5-1, br F4.]3 X Kl.43 N139 vicI-Z, 2-1, 3-2, 4-1, 5-1, hr F4.]3 X Kl.43 ' For brevity vegetative incompatibility genotypes have been written as vicI-I,2-1 rather than as vicI-I, vic2-I. ° The sexual cross is indicated from which each strain is derived. Other sources of strains are also indicated. ‘ Strains EPZ89 am! EPZ43 provider] by S. L. Anagnostakis, Connecticut Agricultural Experiment Station; strain 80-2 provided by William MacDonald, West Virginia University. ° The vc genotype for EPQ9 is incomplete. Strain EP29 is heteroallelic at additional weak vic loci relative to the other strains in this study. See the results section for an explanation. EP29 has been found to be heterokaryotic with respect to the mating type locus (Anagnostakis, 1981). ‘ The vc genotype of K1.9 is unknown, but preliminary analysis suggests that it carries alleles vicI-Z, vic2-2, vic3-I, and vic4-I. "The vc genotypes ole32 and K239 are not known, but preliminary analysis suggests that they carry alleles Viol-2, vic2-I, vic3-2, and vic4-I. 90 A maximum of eight pairings were evenly placed around the perimeter of the Petri plate which permitted the paired cre and br strains to grow in contact with each other toward the center of the plate and provided ample growth to observe the formation of orange heterokaryons. Results Heterokaryon formation. Strains 389.7 cre nic and 80-2c br were crossed and ascosporic progeny were collected. To determine vc reactions among these sexual progeny, the ascosporic cultures were paired together using the PDA vc assay. Pairings between some cream and brown progeny showed the development of wildtype orange color where the mycelia made contact, indicating the possibility that complementation of genetic markers ore and br had occurred. To determine whether the orange mycelium was a heterokaryon, ten hyphal tip subcultures were collected from orange mycelium formed at the confluence of the 389.7 cre nic parent and F2.36 br progeny (Figure 1). Colonies grown from the hyphal tips varied in color from orange to orange-brown to brown. Conidia were collected from colonies representing the three color types (Table 2). Hyphal-tipicolonies that produced orange and orange-brown mycelium yielded conidia that produced both cream and brown mycelia, demonstrating that the orange sector was composed of hyphae containing nuclei of each parental isolate, and thereby proving the heterokaryotic nature of the orange sector. This also demonstrated that the cream color mutation (ore) and the brown color mutation (br) are able to complement each other in a heterokaryon, permitting the development of wild type orange color. Figure 1. Heterokaryon formation between strains 389.7 ore and F2.36 br homoallelic at all Vic loci (A), and the prevention of heterokaryon formation between 389.7 cre and F3.4 br heteroallelic at only vicI (B). The orange sector produced between the strains in (A) is heterokaryotic and exhibits wild type color due to complementation between the ore and br mutations. 92 Table 2. Ratios of cre (cream) to br (brown) nuclear types from heterokaryons composed of vegetatively compatible strains and one putative heterokaryon from strains heteroallelic at vic5. Heterokaryon Subculture Mycelium Number of conidia ratio components“ color cream brown 389.7 + heterokaryon orange 12 97 l : 8 F2.36 #1 from PDA heterokaryon orange 61 145 1 : 2 #2 from PDA hyphal tip ] orange 57 47 l : 1 from het#1 hyphal tip 2 orange/brown l 102 l : 100 from het #l hyphal tip 3 brown 0 102 from het #1 6th subculture dark orange 0 110 from het #1 389.7 + orange sector orange 0 99 F3. 10 from PDA hyphal tip orange 106 0 6th subculture orange 0 110 " The vegetative compatibility genotypes of the homokaryotic strains are: 389.7 vicI-I, 2-1, 3-1. 4-1, 5-1; F2.36 vicl-I, 2-1, 3-1, 4-1, 5-1; F3.]0 vicI-I, 2-1, 3-1, 4-1, 5-2. 93 The differences in mycelial coloration of the hyphal tip colonies appeared to be due to a gene dosage effect brought about by varying numbers of nuclei representing each parent. Conidia were collected from two different orange heterokaryotic sectors formed between two different pairings of strains 389.7 and F2.36 (Table 2). Heterokaryotic sector number 1 produced cre to br conidia at a ratio of 1:2. Heterokaryotic sector number 2 produced ore to br conidia at a ratio of 1:8. The hyphal-tip subculture from heterokaryon number 2 that produced orange mycelium yielded ore and br conidia at a ratio of 1:] (Table 2). The orange-brown hyphal-tip subculture yielded conidia resulting in cre and br colonies at a ratio of 1:100 (Table 2). The brown colored hyphal-tip subculture only yielded br conidia. Since C. parasitica conidia are uninucleate (Puhalla and Anagnostakis, 1971), the ratio of cre to br conidia represents the nuclear ratio of ore to br nuclei within the heterokaryotic mycelium assuming the nuclei are randomly distributed during conidiogenesis. Six serial mass subcultures of heterokaryotic sector 2 were made to test the stability of the heterokaryons. All of the subcultures retained the orange color although the sixth was a darker orange. Only br conidia were collected from the sixth subculture (Table 2). Restriction of heterokaryon formation by vegetative incompatibility. To determine whether vegetative incompatibility restricted heterokaryon formation, pairs of cream and brown ascosporic progeny, representing both those that formed heterokaryons and those that did not form heterokaryons, were chosen from cross 389.7 cre nic X 80-2c hr and paired in the wood/agar vc assay. Those strains which were vegetatively compatible in the wood/agar vc assay were found to be those that formed heterokaryons on PDA, and those 94 strains where a barrage formed in the wood/agar vc assay were those that failed to form heterokaryons on PDA (Figure 1). Vegetatively compatible cre and br strains were also found to form an orange colored zone of mycelium at their point of contact in the wood/agar vc assay. Genetic analysis confirming the identity of vicI and vic2. Previous work by Anagnostakis (1980, 1982, 1988) identified and named two vic loci, vicI and vic2, using C. parasitica strains designated vc types 5, 8, 39, and 71 based on PDA vc testing. Confirmation of these vc genotypes was obtained through three sexual crosses evaluated with the wood/agar vc assay (Table 3). Strain 389.7 cre nic (vc type 5) was crossed with strain EP388 met (vc type 39). The resulting progeny segregated in a 1:1 ratio for each of the parental vegetative compatibility types indicating that the parents differed at a single vic locus as reported by Anagnostakis (1982). This locus has been previously named vicI , and EP388 met is considered to carry the allele vicI-Z. Strain 389.7 cre nic was also crossed with EP289 met (vc type 71). These progeny also segregated in a 1:1 ratio for each of the parental vegetative compatibility types indieating that EP289 also differed from 389.7 at a single vic locus. To confirm the vc genotypes of EP388 and EP289 as well as test the genotype of strain 22508 (vc type 8), strain 389.7 cre nic was crossed with strain 22508 met (Table 3). Strains 389.7 and 22508 were expected to differ from each other at both vicI and vic2. Therefore, if the vc genotype of 22508 is vicI-Z, vic2-2, then EP388 and EP289 will be compatible with the two nonparental vc types. The progeny from this cross segregated in a 1:1: 1:1 ratio consisting of the two parental types, and two nonparental types. The two 95 nonparental classes of progeny were compatible with either EP388 or EP289 confirming the postulated genotypes of Anagnostakis (1988) and demonstrating that strain EP388 and strain EP289 differ from each other at two vic loci, and that 22508 is of genotype vicI-2, vic2-2. This analysis of vicI and vic2 provided the tester strains that were subsequently used to identify new vic loci. Genetic analysis identifying two new vic loci, vic4 and vic5. Strain 389.7 cre nic was crossed with strain 80-2c br (Table 4). Initially, progeny from this cross were placed into four groups based upon the ability of cre and br progeny to form orange mycelium in heterokaryon tests on PDA. However each of these four color groups contained some pairs of are and br strains that produced abnormal thin orange mycelium with conspicuously reduced conidia production (described in detail below). Therefore, the progeny in each of these four groups were subjected to barrage tests using the wood/agar vc assay. The wood/agar vc assay demonstrated that each of the four initial groups of progeny was composed of two vegetative compatibility groups: those strains that formed morphologically abnormal orange mycelium on PDA were found to form a barrage in the wood/agar- vc assay. The progeny from cross 389.7 cre nic X 80-2c br therefore segregated into eight vegetative compatibility types: two parental vc types and six nonparental types. One of the nonparental groups was found to be compatible with the tester genotype vic1-2, vic2-I indicating that 80-2c carried the vicI-Z allele. The other nonparental compatibility genotypes were not identifiable by any other known tester strains. Therefore, it was postulated that 80-2c was heteroallelic with respect to 389.7 at three vic loci: vicI and two new vic loci, vic4 and a weak incompatibility locus designated 96 $81.8 o> cam—£383 05 9:3 0:8 203 88. began—=85 o>usowo> dong 33 Saba 382 a cab o> ”examgcg ~98“ .525 3 "ggggemaboifiewmggufimgg. .39: .2»: a 55 sea 3.3.3.: 3 car: 83 2a.. 8:80» 9.53885 33...»: been an . no fl 2 : 2 .8: «Ram X 3: use fiewm 9. cm on $5 exam X o? Eu Eewm 9. .N S BE ”mam X 9.3 use fiewm .3...on .88 m o> _ h 3 an 3 an o> 880 N-N.N.~o.~> N-N.~..~o..> ~-N.N-~u~> .~..N.~-~u.~> mambo—com 3232368 o>usomo> @886 338» 82. Eat 8:2» E-Euafioo 968%? 5 >5ch no noes—=2 .m 033. 97 .3826??? 00—003.. a3: Breastpémaegaézofi .3... sagging .3: 81%.: .3: .3: esoasafieo: .3: £58.. .3338 Eggs—coagu—afiwtcflaucaeeo “09.0% mum—35030033036833 haugoflgvofifiuoaovg a 00533 «E. :— 625 18003.. mafia—c.1008 05 ”.880 5 33:0 £0.50 a 02.— 958» :0... 89¢ «£8.» .503 05 ME»... >800 0> haw-<39; 05388338580: aspaseoeccfisa 33.5. iguaéieoeaé macefifioefis 8.8. 385832 omega 808 8 .003 £50..» 50.. 0005.? £000.: anemone =< .UBESE-Egofluggowasggeirg .5 008000 ensue 18.5: mafia—04808 00—20 $085.. £033 menace.— uo woman—00 0.83 0&5.» .50.“ 05 .«0 seem .0N.cw\v_ .Nm as gram. Jana—.2 .8233" name... senses 202.802 a :1 5e as, are 3 85.293 as use. use» as. as 103E000? 00.00033ng5 .0>§3§E%m0€5§§fi§005£giaagaeucga 4.00: .73: a 5.5 see 2.7%.: a 8%: 83 25. 8.5.80 basins! oases? been. 8". . .3095 @2 Nu a E on «a bong:— amuse as. see: .2 2 .25 .E sag—en... 8.95.8 ~-n~4.ana~.~-§: enernrnae: Neseeaqeeé: ancienrmaés E95 8 mm ON 2 he Baa—E 2 data .Nn— v.8}: . ~< QVEYNE 509$“.de N-W.~..V.~.M.~-N.Nr~.0.§ ~rm..~uv.~um..~aN.Nu~o.§ Nu%.~uV.~uM.~uN.~r~9.S .~uh.~rV.~oM.~rN.~u~u.§ gain—:00 0>380W0> .3 3.8 X 0.3 0.8 hemn 0080 88..— uoueuocmou .05 0093000» gag—Eco 9530»? 8000.50 390 E .3090...“ “—0 003352 .0 030,—. 98 vic5 (Table 4). The name vic3 was not used here because it had been used elsewhere (Rizwana and Powell, 1992; 1995), though erroneously. The status of vic3 was not known at this juncture in the study, therefore the new loci were named vic4 and vic5. Subsequent genetic analyses described below provide a proper genetic basis for vic3. The presence of a new weak incompatibility locus, designated vic5, was validated by crossing 389.7 cre nic with strain F4.9 br. Strain F4.9, an ascosporic progeny representing one of the six nonparental vc genotypes from cross 389.7 cre nic X 80-2c br, formed a narrow barrage with 389.7 in the wood/agar vc test and formed an orange colored mycelium with greatly reduced conidiation when paired with 389.7 on PDA. Fifty progeny were collected and tested with the wood/agar vc assay: 29 were compatible with parent 389.7, and 21 were compatible with parent F4.9 indicating that the parents differed at a single vic locus. Since this locus segregated independently of vicI and vic4, it was tentatively named vic5 pending additional analysis to test whether this gene was allelic with vic2 or vic3. This cross demonstrated that the weak barrage in the wood/agar vc assay and the abnormal orange sectors on PDA could be attributed to a common cause: a single weak vic locus. Cross 389.7 cre nic X 80—2c br included ascosporic progeny that were vegetatively compatible with parent 389.7. To confirm that these progeny had the same vc genotype as 389.7, one of the brown compatible progeny, F4. 13 br, was crossed with 389.7 cre nic. All of the progeny from this cross were found to be compatible with the parental vc type verifying that the two parental strains were homoallelic at all vic loci. Continuation of the other new vic allele, designated vic4-2, segregating in cross 389.7 99 cre nic X 80-2c br proceeded as follows. Strain F3. 16 br nic was found to form a weak barrage with parent 80-2c, suggesting that these two vc types were heteroallelic at vic5, but shared the new allele vic4-2. Therefore the vic genotype of F3.16 was postulated to be vicI-Z, vic2—I, vic4-2, vic5-I. To test this prediction, F3.16 br nic was crossed with the tester strain A1.13 cre met which had the genotype vicI-I, vic2-2, vic4-I, vic5-I. The progeny from this cross segregated into eight vc types as expected using the PDA heterokaryon assay (Table 5). None of the orange heterokaryons from any of the eight groups deyeloped the abnormal type of orange morphology indicating that vic5 did not segregate in this cross. Of the eight vc types of progeny, four were compatible with tester strains representing the four combinations of alleles at vicI and vic2, as expected. The presence of four additional vc types (three nonparental types and one compatible with parent F3. 16) verified that the parental strains were heteroallelic at vicI , vic2, and a third locus, vic4. The identification of the three remaining nonparental vc types from cross F3.16 br nic X Al.13 cre me: will be described below. The two remaining nonparental vc genotypes from cross 389.7 cre nic X 80-2c br were identified using the analysis of cross 389.7 cre nic X EP243. Prior analysis of strain EP243 showed that it differed from 389.7 at two vic loci, one of which caused a weak incompatibility reaction (Chapter 2). Parent EP243 and the two nonparental vc genotypes from this cross were found to be compatible with three of the nonparental vc genotypes from cross 389.7 cre nic X 8020 br. One of the nonparental vc types from 389.7 cre nic X EPZ43 was found to be compatible with strain F4.9 thereby proving that EP243 carries vic5-2. Consequently, the vc genotype of EP243, and the corresponding compatible 100 $33 EEG—833 on. unma— voumfiboov 83 gain—eon. 9580!; .963. 188:! gun—:35 mg gaseoggflufiggfiiflguoéa .TNoE 4.79.: «a as: .858 ~.N.~-~o.§ a float? :03 26a .2598» 363889 3:50»? .0395 .8...— . a 3 2 ON a 3 we con—55 Acumen 33 nudtcflmn vn.~_ ~.h.~.v.~.m.~.~.~-~o.§ ~.w.~.v.~.m.~.~.~-~u.§ .~.h.~.v.~.m.~.N.T~o.§ gm £939.50 o>m~50uo> 6.3 .5 2.9..— X as: v.8 2.7. 380 59c gap—mow a: combo—Bu basalt—So 258%: 30.5.. Emma 3 30mg.— ..e honesz .m 035—. 101 nonparental vc type from 389.7 cre nic X 80-2c br, is vicI -1 , vic2-1, vic4-2, vic5-2. The other nonparental type from cross 389.7 cre nic X EP243, and the corresponding compatible nonparental vc type from cross 389.7 cre nic X 80—2c br, is therefore vicI-I , vic2-I, vic4-2, vic5-I. The final nonparental vc genotype from cross 389.7 cre nic X 80-2c br, represented by F3. 15 br nic, was identified beeause of the abnormal orange sectors on PDA and weak incompatibility reaction in the wood/agar vc assay that occurred when F3.15 grew in contact with tester Al.13 cm met, indieating that these vc genotypes differed at only vic5. This fit expectations based upon the other nonparental genotypes. The precwding analysis of the vc genotype of strain 80-2c has shown that it is heteroallelic with respect to 389.7 at vial and two new loci, vic4 and vic5. Strain EP243 was also found to be heteroallelic relative to 389.7 at vic4 and vic5, and to carry the same alleles at loci vic4 and vic5 that 80-2c carries. The demonstration that gene vic5-2 is not allelic with locus vic2 will be presented below in the analysis of cross EP29 x Al.10 cre nic. Unusual mycelial interactions associated with heteroallelism at vic4 and vic5. Occasionally, ore and br strains heteroallelic at only vic4 produced irregularly shaped sectors with dark brown, slightly orange-tinged, appressed mycelium where the two strains made contact on PDA. These sectors of appressed mycelium never appeared when the strains made contact in the wood/agar vc assay. The incompatibility reaction (barrage) produced by vic5 in the wood/agar vc assay was less broad and more easily obscured by weak mycelial growth than the reactions produced by Vic], vic2, vic3, and vic4. When 102 ore and br strains heteroallelic at only vic5 grew in contact with each other on PDA, no barrage was visible and a sector of thin orange mycelium with greatly reduced conidiation developed at their confluence (Figure 2). However, orange mycelium never formed when cre and br strains heteroallelic at only vic5 made contact in the wood/agar vc assay. The abnormal sectors produced in association with vic4 and vic5 were maintained upon subculturing. To determine if the orange mycelium was heterokaryotic that appeared when strains heteroallelic at only vic5 made contact on PDA, ten hyphal tip subcultures were collected from a subculture of orange mycelium produced by the contact of strains 389.7 cre nic and F3.10 br nic (vicI-I, vic2-I, vic3-I, vic4-1, vic5-2), an ascosporic progeny from cross 389.7 cre nic X 80-2c br. The resulting colonies that grew from the hyphal tips were light orange in color, and none showed the variation from orange to brown found among the hyphal tip cultures of the vegetatively compatible cre and br strains. Conidia were collected from the original orange sector and from the fifth of a series of subcultures that retained the orange color. In both cases br conidia were collected (Table 2). In contrast, conidia collected from a representative light orange hyphal tip colony produced only cre conidia. These results suggest that the abnormal orange sectors produced between strains 389.7 and F3. 10 are heterokaryotic because both nuclear types were derived from the mycelium; however, both nuclear types were not recovered from a single hyphal tip. Identification of the remaining nonparental vc genotypes from cross F3.l6 br nic X A1.13 cre met. Three nonparental vc types from cross F3.16 br nic X Al. 13 cre met remained to be identified. One of these nonparental vc types could be identified using 103 Figure 2. An orange sector of abnormal mycelium has formed between strains 389.7 are and F3. 10 br heteroallelic at only vic5. The orange sector has thinner mycelium and less conidia than the normal mycelium of the ore and br strains. The plate has back lighting. 104 strain F326 br nic. The two remaining nonparental vc genotypes were distinguished by crossing 12.23 br nic with 10.30 cre. Four classes of ascosporic progeny resulted from this cross (Table 7). Therefore the vc genotype of 12.23 was determined to be vial-2, vic2-2, vic3—I, vic4-2, vic5-I. Cross 12.23 br nic X K2.3O cre will be discussed in more detail below in relation to the identification and characterization of vic3. Genetic analysis identifying a new locus vic3. A cross was performed between EP29 and testerstrain A1.10 cre nic (Table 6). The vc genotype of Al . 10 was known to be vicI-Z, vicZ-I, vic4-I , vic5-I (the status of vic3 was not known at the time of the cross). The ascosporic progeny from this cross segregated into eight compatibility groups plus nine progeny that were incompatible with these eight groups and four progeny that did not give clear reactions in the wood/agar vc assay. The eight compatibility groups included two parental groups, three unknown nonparental groups, and three nonparental groups compatible with tester genotypes vicI-Z, vic2-I, vic4-1, vic5-I; vicI-Z, vic2-I, vic4-I, vicS-Z; and vic1-2, vic2-2, vic4-I, vic5-1. These tester strains indicated that EP29 must carry the alleles vic2-2 and vic5-2. In addition, a br tester strain of genotype vicI-I , vic2- ], vic4-I, vic5-1 was not compatible with any are progeny indicating that vicI -1 was not present in EPZ9. The presence of eight predominant vc groups among the progeny suggested the presence of one unidentified strong incompatibility gene segregating in this cross. Therefore, two additional crosses were performed in an attempt to separate the strong Vic gene from vic5-2 and any other weak vic genes that might be present. To identify the new strong We gene present in EP29, ascosporic progeny K230 cre, derived from cross EP29 X A1.10 cre nic, was chosen as a potential carrier of the new vic 105 gene (vic3-2) because it formed a weak barrage with the parental strain EP29 and strong barrages with the other identified compatibility types from this cross. Therefore, K230 was postulated to carry the unidentified strong vic gene from parent EP29, and to be homoallelic with respect to the weak vic genes carried by this parent. The first cross tested whether the putative new gene vic3-2 in K2.30 was allelic with, or linked to, We]. This was accomplished by crossing K2.30 cre with 12.69 br nic (Table 7). Strain 12.69 has the genotype vicI-I, vic2-2, vic3-I, vic4-I, vic5-I. These two strains were hypothesized to be heteroallelic at only vicI and the new strong vic locus designated vic3. The progeny from this cross segregated into two parental and two nonparental compatibility groups in a 1: l : 1:1 ratio using the PDA heterokaryon assay, demonstrating that vic3-2 is not allelic with locus Vic] (Table 7). In order to provide additional confirmation that these four vegetative compatibility groups did not contain other segregating, weak incompatibility genes, 54 strains representing each of the four vc groups were also subjected to compatibility tests using the wood/agar vc assay. The wood/agar vc assay verified that all of the strains within each of the four groups were compatible with each other, demonstrating that no weak Vic genes segregated in this cross. The second cross tested whether or not the putative vic3-2 gene was allelic with, or linked to, vic4. Strain K230 cre, with the postulated genotype vicI-Z, vic2-2, vic3-2, vic4-I, vic5-I, was crossed with strain 12.23 br nic known to have the genotype vicI-2, vic2-2, vic3-I, vic4-2, vicS-I. (This cross was already referred to above). These two strains were hypothesized to be heteroallelic at only vic3 and vic4. The progeny from this cross segregated into two parental and two nonparental compatibility groups in a l:1:1:1 106 .315 u> 5 won: 203 Gen £959.30 23:80”? 2. «@9— 18 an. 5— Evan . .h 2.2 as... a? 26.38 ._ 3 IE 55...” . $83 3 ecu->303 2: new: 80v :3 95.8 gain—33m 25509.5 .35 .83. 05 25a- voooi .. 2:83» 3 on. 5.6.3 833 .953. 9:33.950 o>usoao> 05 3 .83 cum-am . . 7N2: 4-7.? 3 =3. .85... TNT?! .- 983.5 .83 25. .2593» Dana-9.83m 2580»? $03 no“. . aqua ow V a n— : n so hone-s. acumen. 9.8.8 _89 33.0.8 o> 25:33 5; 035.9308— aumm unfign. an a. 3— manage N— o N a o— no .3955 n—.n...~ onéu— n‘U— FWNn 1v.”— N.h.~¢.~-m..~-~.~-~o! ~.h.~.‘.N-M.N.N.N-~cB ~.w.~.V.N-M.TN.N-~03 ~.H.~.V.~.M.N.N.N-~ct .~.H.~.V.~.M.TN.N-~93 ugxuéoocoudamfiaago o£§oflo> .8: Eu 2. _< X OE 320 89¢ 9339.38 .3: use» banana—Eco 2580a? :03 am .339:— MO .3892 6 03.5. 107 .adxsdsosflagij3:393:51.-.deoaflspsiaiiistalas 3.51.8. .‘r-o 23!.uni-2.0.lguiibaaacgog8i3§3333§€3§3.§x3.a§£§22.9-95.3?!le .i8§33a§§§§§o§8> .osilsalaasauoaégggofiaovsasJuilfik2.33:2. 4.3.. 44.4.1385... ~.«.~-~§l8€383¥1§$o8- again-8033188.. £858"? mafia we 3 2 a 2 x 3.2 $.fl. me vs a. on «N X en.” Fuck— _88 09—52.: nu.~—.B.: .néd—uJEBnNM 9:23.34 «n.-3._< toNQnfi—t. ~-n.~.V.N-M.N.~.N-~93 ~-n.~.v.~.n.~.«.u.~8> ~-n.~¢.~.m.~-~.~-~ur ~.n.~¢.~.m.~.~.~.~93 ~-n.~.~.~.m.~.~.u-~8> .~-n.~¢.~.m.~.~.~-~8> GAE mango 258695 n I“. .35.. adxpccnduiaeksdxsccnfiaspag uni-28.15%. giggisaflgagz .33... 108 ratio (Table 7). The independent segregation of the putative gene vic3-2 and vic4-I demonstrated that vic3-2 is not allelic with locus vic4. A third cross was performed to identify the nonparental vc genotype of ascosporic progeny Kl.43. Strain Kl.43 cre was crossed with F4. 13 br (vicI-I, vic2-1, vic3-I, vic4- ], vic5-1) (Table 8). The progeny from this cross would have assorted into four classes if only vicI and vic3 segregated, or eight classes if vic5 also segregated. The ascosporic progeny were found to segregate into only four classes: two parental, one nonparental type compatible with vicI-Z, vic2-I, vic3-I , vic4-I, vic5-I , and one additional nonparental type. Additional verification that weak compatibility genes did not segregate in this cross was provided by subjecting all of the progeny in each of the four vegetative compatibility groups to the wood/agar vc assay. This test confirmed that all of the strains in each of the four groups were compatible with each other, and that no weak vic genes had segregated in this cross. Consequently, Kl.43 was shown to carry allele vic5-I. This cross also provided further confirmation that vicI and vic3 are distinct loci because of the independent segregation of the alleles at these loci. Cross EP29 X A1.10 cre nic also provided confirmation that the gene previously designated vic5-2 represents a new vic locus. Gene vic5-2 was found to be present in EP29 because it segregated in the nonparental progeny class compatible with the tester genotype vicI-Z, vicZ-I, vic3-I, vic4-I, vic5-2. The presence of this nonparental vc genotype, as well as the others identified above, demonstrate that vic5 segregates independently of both vic2 and vic3. This cross thereby provides the final confirmation that gene vic5-2, originally identified in cross 389.7 cre nic X 80-2c br, represents a new 109 .ugéfigzggegfixgag EcuschaSuagflquofiSvognagaoBgmgaggg .afiaggofiggg «58. banana—“Sufi o>380mo> denaéoaguofiuouggg o§u0>05£§§3u§u§ae£a .73. .29.... a 85 sec 2.29.... a Beta 83 25. 8.5.8» 55.5.9803 9.3%.... £35 a". . .3ch a. v 2 2 3 a has... 3.25.5. a 423.2 amuse? 3.35%.” ~.h.~.V.N.M.~.N.N-~c.S ~.w.~.v.~.m.7~.~-~o.$ ~.h.TV.~.M.TN.N.~u.S .~.n.~.$.~.m.~.~.~-~c.§ 3on9:— 38. 89680» banana—:8 o§w0> I! 6.8 9.3. X .3 2 .3— $20 89¢ Efimamou 35 8mg» gain—co 03.50»? 30.85% 58 E gouge we con—8:2 .w 03¢. 110 vic locus. The preceeding crosses have shown that the gene designated w'c3-2, derived from strain EP29, is not allelic with vicI, vic2, vic4, or vic5, and should therefore be given a new locus name. The new locus is here named vic3 because it is the strong, unnamed vic locus that Anagnostakis (1981) observed to be segregating in her analysis. Linkage analysis. Linkage analysis was conducted on the five vic loci, the two auxotrophic mutations, and the two pigmentation mutations using several crosses (Table 9). These nuclear genes were tested for Mendelian segregation using chi-square analysis with one degree of freedom. The color mutations cre (cream) and br (brown) were found to be tightly linked. One thousand forty four progeny were pooled from five crosses between different ore and br parents. Of these combined progeny three recombinants (0.3% recombination frequency) were found that displayed the orange wildtype coloration. Anagnostakis (1982) had crossed me and br strains but did not find any recombinant orange progeny. In the study by Anagnostakis fewer progeny were counted so that tight linkage could have been missed. Also, it is not known if the br mutation in this study is the same genetic locus as the br mutation in the study by Anagnostakis; the cre mutations in both studies are the same. The brown pigmentation mutation segregated as a single locus when crossed with the wild type orange strain EPISS (1 11 br progeny; 122 orange progeny; X2 = 0.52; P = not significant). The nicotinamide requirement mutation (nic) also segregated as a single genetic locus. The two auxotrophic markers, nicotinamide and methionine (met) requirements, were not linked to each other or to cre and br. The m'c mutation was not linked to any of the vic loci. The met mutation was not linked to vicI 111 Table 9. Linkage analysis of the loci used in this study. —-—=—===—=— ====== Progeny numbers Loci Cross Parental Recombinant Recombination Xz‘ frequency nic : met 389.7 X 22508 28 34 54.8 0.58 m'c : era 33 29 46.8 0.26 met : era 35 27 43.5 1.03 Vic] : MC 33 29 46.8 0.26 Vic] :met 33 29 46.8 0.26 vicl : era 30 32 51.6 0.06 vic2 : rate 32 30 48.4 0.06 vic2 : met 28 34 54.8 0.58 Vic} : vic2 35 27 43.5 1.03 vicZ : are combined crosses: 58 44 43.1 1.92 389.7 X 22508, 389.7 X EP289 cre : br combined crosses: 1044 3" 0.29 1035 389.7 X 80-2c, F3.16 X A1.13, K230 X 12.69, K230 X 12.23, F4.13 X Kl.43 vic4 : br 389.7 X 80-2c 84 81 49.1 0.05 vic5 : br 83 82 49.7 0.006 vic4 : air: 53 72 57.6 2.89 vic5 : m’c 64 61 48.8 0.07 vic4 : vic5 68 97 58.8 5.10"' vicl : vic4 71 94 57.0 , 3.21 Vic] : vic5 86 79 47.9 0.30 vic3 : vic4 K230 X 12.23 21 27 56.3 0.75 vicZ : vic4 [-3.16 X Al.l3 34 56 62.2 538* vic3 : m’c K230 X 12.69 31 28 47.5 0.15 vic3 : etc 45 50 52.6 0.26 Vic} : vic3 47 48 50.5 0.01 ' The null hypothesis for the chi-square test is that two loci segregate in a 1:1 ratio. ' Recombinant phenotype was orange. " Significantly different from 1:1 at P < 0.05. 112 or vic2; its linkage to the other three vic loci was not tested. Anagnostakis (1982) had previously shown that cre was not linked to vicI or to met. The present study has found that ore is also not linked to vic2, vic3, vic4, or vic5. The loci cre and vic2 demonstrated unlinked segregation using the combined progeny from crosses 389.7 cre nic X 22508 met and 389.7 cre nic X EP289 met. These loci also tested as unlinked from cross 389.7 cre nicX EP289 met alone (parental type = 19, recombinant type = 21; X2 = 0.10; P = not significant). However, in cross 389.7 cre nic X 22508 met these loci deviate from 1:1 segregation at a significance level of P < 0.05 (parental type = 39, recombinant type = 23; X2 = 4.13). The deviation from 1:1 segregation in cross 389.7 cre nic X 22508 met is considered to be the result of random variation. Chi-square analysis indicates that vicI is not linked to vic2, vic3, vic4 or vic5. ViC3 and vic4 are also not linked. However, vic4 and vic5 failed to segregate in a 1:1 ratio (P < 0.05). Loci vic2 and vic4 also did not segregate in a 1:1 ratio (P < 0.05). The apparent deviation from 1:1 segregation of these two pairs of vic loci may indicate either loose linkage or random variation such as observed with the loci cre and vic2 described above. The linkage of vic2, vic3 and vic5 to each other could not be determined from cross EP29 X Al.10 cre nic because some of the nonparental vc genotypes could not be determined. However, the independent segregation of the alleles at loci vic2, vic3, and vic5 was demonstrated by the identifiable vc genotypes of one of the parean strains and four of the nonparental classes. Although linkage analysis could not be done, it is evident from segregation ratios that these three loci are not closely linked to each other (Table 6). 113 Discussion This study sought to identify new vic loci that control vegetative cell fusions in C. parasitica, to test whether heterokaryon formation occurs in this species, and to test the effects of individual We loci upon heterokaryon formation. Identification of three new vic loci. The present study identified three new vic loci designated vic3, vic4, and vic5 in Cryphonectria parasitica. These new vic loci were detected and characterized using a combination of barrage development on chestnut phloem/cortex tissue and complementary pigmentation mutations as a nonselective assay to detect heterokaryons. Prior to this study only two Vic loci had been identified in C. parasitica by genetic analysis (Anagnostakis, 1980, 1982, 1988). The new locus vic5 was undetected in two previous studies (Huber and Fulbright, 1994; Anagnostalds, 1988) where the segregation of this locus through sexual recombination is now known to have occurred. All five vic loci have two known alleles that produce incompatibility only through allelic interactions; vegetative incompatibility resulting from nonallelic gene interactions has not been found in this species. This study has also shown that all five vic loci strictly prevent heterokaryon formation on autoclaved chestnut tissue even though each locus permits varying degrees of cytoplasmic transmission of hypoviruses and a senescence-inducing agent (chapters 4 and 5). Therefore, vegetative incompatibility in C. parasitica, as defined by barrage formation in the wood/agar vc assay, is equivalent to heterokaryon incompatibility. Anagnostakis (1988) assigned tentative genotypes to five different vc types (5, 8, 10, 39, 71) based upon sexual crosses of three of the genotypes (8, 10, 39) with vc type 5. 114 My genetic analysis concurs with that of Anagnostakis (1988) and demonstrates that vc types 5, 8, 39, and 71 represent all four combinations of the alleles at loci vicI and vic2. However, Anagnostakis also tentatively assigned to these four genotypes genes designated vic3-I, vic4-I, vic5-l, vic6—I, and vic7-I. These other five genes were assigned because vc types 5 and 10 were thought to be heteroallelic at seven Vic loci although none of these loci were actually identified. Therefore, in order to name new vic genes without creating an alternative nomenclature, I am here considering the designations for vic3, vic4, vic5, vic6, and vic7 found in Anagnostakis (1988) to be superseded by this study while retaining the identities of Vic] and vic2 because they have been genetically defined (Anagnostakis 1982, 1988). In addition, in keeping with the nomenclatural precedent set by Anagnostakis (1988), the We genes present in vc type 5 are considered to represent the " l " allele for each new locus identified. This revision does not affect the argument by Anagnostakis (1982, 1988) as to the degree of genetic differentiation between vc types 5 and 10, but does mean that vc types 5 and 10 must be considered to be heteroallelic at presently unidentified vic loci. The new locus vic3 has not been previously identified even though the allele name vic3-2 has been used in reference to the vc genotype of EP29 (Rizwana and Powell, 1992, 1995). The use of the designation vic3-2 by Rizwana and Powell was in reference to Anagnostakis (1988) where a cross between vc types 5 and 16 were stated to produce only parental vc types. Two points need to be made in reference to this statement. First, Anagnostakis (1988) did not resolve whether a vic gene reported to segregate in a cross between vc 5 and vc 16 represented a new locus, which could have been named vic3, or 115 a new allele at We]. Therefore, the use of the name vic3-2 by Rizwana and Powell (1992, 1995) in reference to the genotype of EP29 was invalid. Secondly, Anagnostakis (1981) began a genetic analysis of EP29 that demonstrated the presence of vic2-2 in that strain. This also was not incorporated by Rizwana and Powell (1992, 1995) into their vc genotype for EP29. Consequently, the report by Rizwana and Powell (1992, 1995) that vc 16 carries vicI-I and vic2-I is incorrect. My study has shown that EPZ9 (vc type 16) carries genes vial-2, vic2-2, vic3—2, vic4-I, and vic5-2. The presence of more than eight vegetative incompatibility groups in cross EP29 X A1.10 is probably due to the segregation of alleles at an additional heteroallelic Vic locus (or loci). Additional genetic analysis is needed to verify this. How many polymorphic vic loci are in the North American population of C. parasitica? The frequently quoted estimate of five to seven polymorphic loci was based upon a cross between vc types 5 and 10 where more than 100 vc types were found among the progeny (Anagnostakis, I982). Virus transmission tests between a vc 10 strain and several vc genotypes representing differences at vicI , vicZ, and vic3 indicate that none of the 32 vc genotypes representing the recombination of Vic] through vic5 could be compatible with vc 10 (chapter 4). Therefore, if vc 5 and vc 10 differ at seven Vic loci, which I believe is a reasonable estimate, then the North American population would be polymorphic at more than seven Vic loci. A further consideration is that vic5 was not readily detectable with the traditional PDA vc assay which suggests that other weak incompatibility loci may also be present thereby further inflating the number of potentially polymorphic vic loci in North America. 116 Heterokaryon formation under nonselective conditions. The utilization of independent methods (barrage development and heterokaryon formation) to identify vegetatively compatible individuals has clarified the relationship between vegetative compatibility and heterokaryosis in C. parasitica. Previous work had shown that heterokaryons could form between nutritional mutants and morphological mutants, but the heterokaryons did not form easily nor were they maintained (Puhalla and Anagnostakis, 1971). The authors concluded that heterokaryons would not readily occur in nature even between vegetatively compatible strains. Anagnostakis (1981) later found a strain that was heterokaryotic with regard to mating type. My study has shown that heterokaryons always formed between vegetatively compatible individuals without the imposition of selective growth conditions, and that these heterokaryons can be maintained through serial subculturing. Heterokaryons in Ascomycetes have ban found to be either confined to the anastomosed cells, which limits proliferation of the heterokaryon (e.g. Gibberella zeae) (Adams et al., 1987), or to be characterized by the presence of both nuclear types in the hyphal tips which permits proliferation (e.g. Neurospora crassa) (Beadle and Coonradt, 1944). Hyphal tips of C. parasitica were found to harbor both nuclear types at variable ratios and to be maintained upon serial subculturing. Interestingly, the ratios of the two nuclear types in the hyphal tips corresponded to the color of the mycelium that grew from the hyphal tips: a greater proportion of cre nuclei resulted in a lighter orange mycelium, whereas relatively more br nuclei produced a darker orange-brown mycelium. Both the brown hyphal tip subculture and the sixth serial mass transfer of heterokaryotic mycelium 117 yielded only br conidia. The collection of a brown hyphal tip subculture demonstrates that particular hyphae from a heterokaryotic mycelium may have greatly skewed nuclear ratios, or may even be composed of a single nuclear type. The finding of only br conidia from the sixth serial mass transfer may also be due to the progressive skewing of the ratio of br to are nuclei through prolonged growth. Differences in the ratios of nuclear types in hyphal tips have also been observed in Neurospora (Prout et al., 1953; Pittenger and Atwood, 1956). Pittenger and Atwood (1956) found that the hyphae at the colony front did not grow autonomously, but grew at a relatively constant rate determined by the mycelium rather than according to the nuclear ratios of individual hyphae. This example of the physiologieal integration of the mycelium could explain why the uniformly orange heterokaryotic mycelium in C. parasitica could produce hyphal tips with variable nuclear ratios and corresponding differences in color. The effects of vie genes upon heterokaryon formation. Some evidence was found to suggest that the effectiveness of individual vic genes at preventing heterokaryon formation varies, and is associated with growth substrate. Unusual mycelial interactions suggestive of heterokaryosis sometimes occurred between strains heteroallelic at only vic4 or vic5 when grown on PDA. Although conidia derived from hyphal tip subcultures taken from these abnormal orange mycelial interactions did not show both nuclear types, both cre and br conidia were collected from the abnormal orange sectors, and the orange mycelium was maintained upon subculturing suggesting that both nuclear types were present though perhaps segregated into different hyphal tips or at greatly skewed ratios. Previous studies have also noted differences in incompatibility interactions in C. parasitica 118 by describing strong and weak barrages (Anagnostakis 1988). Morphological differences in the interactions of conspecifics due to incompatibility genes have also been observed in Ophiostoma ulmi (Brasier 1984). One possible effect of growth substrate upon heterokaryosis may be that the frequency of hyphal anastomoses varies on the different substrates as has been found in the Basidiomycete Corticium vellereum (Bourchier, 1957). The unusual mycelial interactions associated with heteroallelism at vic4 and vic5 were always prevented whenever strains were also heteroallelic at Vic], vic2, or vic3. This indicates that heteroallelism at vicI, vic2, and via? is epistatic over the mycelial interactions associatw with vic4 and vic5. In addition, when strains were heteroallelic at both vic4 and vic5, the unusual mycelial interactions never occurred, indicating that vic4 and vic5 apparently have an additive effect that is similar to the other three loci. To my knowledge, evidence for effects upon heterokaryon formation due to epistasis between genes involved in vegetative incompatibility have only been reported previously by Coenen et al. (1994) and Holloway (1955). Additive effects between incompatibility genes have been reported in A. nidulans (Dales et al.,1983; Coenen et al.,1994). The weak nature of the incompatibility elicited by vic5, and to a lesser extent vic4, may be comparable to the partial-het genes described by Coenen et al. (1994), hetA described by Dales et al. (1983), and the genes affecting incompatibility described by Holloway (1955). Heteroallelism at a single partial-het locus in A. nidulans permitted some heterokaryotic growth under selective conditions, but heteroallelism at two partial-het loci prevented heterokaryotic growth. Dales et al. (1983) found that hetA in A. nidulans prevented heterokaryon formation under nonselective conditions but often permitted 119 normal appearing heterokaryons under selective conditions. Holloway (1955) described interesting interactions among four loci which limited the growth of heterokaryotic mycelia in N. crassa. He found that the formation of heterokaryons under selective growth conditions was dependent upon heteroallelism at three loci (W, X, Y) while the growth rate and the time of growth initiation of heterokaryons depended upon loci X, Y, and Z. Under natural growth conditions the partial-het genes and some of the genes described by Holloway may prevent heterokaryon formation just as vic5 prevents it on natural substrate. The function of vegetative incompatibility genes. The biological function of vegetative incompatibility genes is not presently understood. The presence of large numbers of vegetative incompatibility groups in Ascomycete species in nature, especially among sexually reproducing species, suggests that incompatibility has some selective advantage. However, hypotheses concerning the biological function of vegetative incompatibility are complieated by the many effects associated with incompatibility genes. Several hypotheses have been proposed including the prevention of somatic cell parasitism (Buss 1987), the prevention of the infectious transmission of parasitic genetic elements (Caten 1972; Nauta and Hoekstra 1994), the development of reproductive organs (Bernet 1992, and references therein), and ecological competition (Rayner et al. , 1984; Rayner 1991). The presence and consequences of heterokaryon formation in ascomycetous fungi in nature is an open question. Few studies have recovered heterokaryons from nature or have tested the effects of heterokaryosis upon fitness. Examples are known where heterokaryons are less aggressive than the component homokaryons (Grindle and Pittenger, 120 1968); other examples are known where the heterokaryon is more aggressive (e. g. Jinks, 1952a, 1952b). The ability of via loci to prevent heterokaryon formation in C. parasitica, while permitting cytoplasmic transmission of genetic elements, suggests that there may be a selective disadvantage to the formation of heterokaryons in nature. Of course, vegetative incompatibility systems do permit heterokaryon formation between genetically different individuals although these compatible individuals would share significant genetic similarities. Clearly, the determination of the biological significance of vegetative incompatibility genes must take into aceount the effects of individual incompatibility genes upon the transmission of nuclei and genetic elements, the potential for pleiotropy, and the environmental growth conditions of the mycelium. 121 LITERATURE CITED Adams, G., N. Johnson, J .F. Leslie, and LP. Hart. 1987. Heterokaryons of Gibberella zeae formed following hyphal anastomosis or protoplast fusion. Exp. Mycol. 11: 339- 353. Anagnostakis, S.L. 1977. Vegetative incompatibility in Endothia parasitica. Exp. Mycol. 1: 306-316. Anagnostakis, S.L. 1980. Notes on the genetics of Endothia parasitica. Neurospora Newslett. 27:36. Anagnostakis, S.L. 1981. A stable heterokaryon of Endothia parasitica. Mycologia 73: 570—576. Anagnostakis, S.L. 1982. Genetic Analyses of Endothia parasitica: linkage data for four single genes and three vegetative compatibility types. Genetics 102: 25-28. Anagnostakis, S.L. 1983. Conversion to curative morphology in Endothia parasitica and its restriction by vegetative compatibility. Mycologia 75: 777-780. Anagnostakis, S.L. 1987. Chestnut blight: the classical problem of an introduced pathogen. Mycologia 79: 23-37. Anagnostakis, S.L. 1988. Cryphonectria parasitica, cause of chesnut blight, pp. 123-136 in Advances in Plant Pathology, edited by 6.8. Sidhu. Academic Press, Ltd., London. Anagnostakis, S.L. and P.R. Day. 1979. Hypovirulence conversion in Endothia parasitica. Phytopathology 69: 1226-1229. Anagnostakis, S.L., B. Hau, and J. Kranz. 1986. Diversity of vegetative compatibility groups of Cryphonectria parasitica in Connecticut and Europe. Plant Disease 70: 536-538. Anwar, M.M., J.H. Croft, R.B.G. Dales. 1993. Analysis of heterokaryon incompatibility (h-c) groups R and GL provides evidence that at least eight het loci control somatic incompatibility in Aspergillus nidulans. J. Gen. Microbiol. 139: 1599-1603. Beadle, G.W. and V.L. Coonradt. 1944. Heterocaryosis in Neurospora crassa. Genetics 29: 291-308. Bernet, J. 1992. In Podospora anserina, protoplasmic incompatibility genes are involved in cell death control via multiple gene interactions. Heredity 68: 79-87. Bourchier, R.J. 1957. Variation in cultural conditions and its effect on hyphal fusion in 122 Corticium vellereum. Mycologia 49: 20-28. Brasier, C.M. 1984. Intermycelial recognition systems in Ceratocystis ulmi: their physiological properties and ecological importance, pp. 451-497 in The Ecology and Physiology of the Fungal Mycelium, edited by D.H. Jennings and A.D.M. Rayner. Cambridge University Press, Cambridge. Buck, K.W. 1986. Fungal virologyuan overview, pp. 1-84 in Fungal Wrology, edited by K.W. Buck. CRC Press, Inc., Boca Raton, Florida. Buss, L.W. 1987. The Evolution of Individuality. Princeton University Press, Princeton, New Jersey. pp. 201. Caten, C.E. 1972. Vegetative incompatibility and cytoplasmic infection in fungi. J. Gen. Microbiol. 72: 221-229. Coenen, A., F. Debets, F. Hoekstra. 1994. Additive action of partial heterokaryon incompatibility (partial—het) genes in Aspergillus nidulans. Curr. Genet. 26: 233-237. Collins, R.A. and 8.]. Saville. 1990. Independent transfer of mitochondrial chromosomes and plasmids during unstable vegetative fusion in Neurospora. Nature 345: 177-179. Dales, R.B.G., J. Moorhouse, and LB. Croft. 1983. The location and analysis of two heterokaryon-incompatibility (het) loci in strains of Aspergillus nidulans. J. Gen. Microbiol. 129: 3637-3642. Day, P.R., J.A. Dodds, J.E. Elliston, R.A. Jaynes, and S.L. Anagnostakis. 1977. Double-stranded RNA in Endothia parasitica. Phytopathology 67: 1393-1396. Debets, F., X. Yang, and A.J.F. Griffiths. 1994. Vegetative incompatibility in Neurospora: its effect on horizontal transfer of mitochondrial plasmids and senescence in natural populations. Curr. Genet. 26: 113-119. Glass, N .L. and G.A. Kuldau. 1992. Mating-type and vegetative incompatibility in filamentous ascomycetes. Annu. Rev. Phytopathol. 30: 201-224. Griffin, G.J. 1986. Chestnut blight and its control. Horticultural Reviews 8:291-336. Griffiths, A.J.F., S.R. Kraus, R. Barton, D.A. Court, and H. Bertrand. 1990. Heterokaryotic transmission of senescence plasmid DNA in Neurospora. Curr. Genet. 17: 139-145. Grindle, M. and T.H. Pittenger. 1968. Phenotypic and genetic changes during prolonged growth of Neurospora heterokaryons. Genetics 58: 337-349. 123 Handley, L. and C.E. Caten. 1973. Vegetative death: a mitochondrial mutation in Aspergillus amstelodami. Heredity 31: 136 (Abst.). Heiniger, U. and R. Rigling. 1994. Biological control of chestnut blight in Europe. Annu. Rev. Phytopathol. 32: 581-599. Holliday, R. 1956. A new method for the identification of biochemical mutants of micro- organisms. Nature 178: 987. Holloway, B.W. 1955. Genetic control of heterocaryosis in Neurospora crassa. Genetics 40: 117-129. Huber, D.H. and D.W. Fulbright. 1994. Preliminary investigations on the effect of individual vic genes upon the transmission of dsRNA in Cryphonectria parasitica, pp. 15- 19 in Proceedings of the International Chestnut Conference, edited by M.L. Double and W.L. MacDonald. West Virginia University Press, Morgantown. Jinks, J .L. 1952a. Heterokaryosis in wild Penicillium. Heredity 6: 77-87. Jinks, J .L. l952b. Heterokaryosis--a system of adaptation in wild fungi. Proc. R. Soc. London Ser. B 140: 83-99. Leslie, J.F. 1993. Fungal vegetative compatibility. Annu. Rev. Phytopathol. 31: l27-150. MacDonald, W.L. and D.W. Fulbright. 1991. Biological control of chestnut blight: use and limitations of transmissible hypovirulence. Plant Disease. 75: 656-661. Mahanti, N. and D.W. Fulbright. 1995. Detection of mitochondrial DNA transfer between strains after vegetative contact in Cryphonectria parasitica. Mol. Plant-Microbe Interact. 8: 465-467. Monteiro—Vitorello, C.B.M., J.A. Bell, D.W. Fulbright, H. Bertrand. 1995. A cytoplasmieally transmissible hypovinrlence phenotype associated with mitochondrial DNA mutations in the chestnut blight fungus Cryphonectria parasitica. Proc. Natl. Acad. Sci. USA. 92: 5935-5939. Nauta, M.J. and R.F. Hoekstra. 1994. Evolution of vegetative incompatibility in filamentous ascomycetes. I. Deterministic models. Evolution 48: 979-995. Nuss, D.L. 1992. Biological control of chestnut blight: an example of virus-mediated attenuation of fungal pathogenesis. Microbiol. Rev. 56: 561-576. Nuss, D.L. and Y. Koltin. 1990. Significance of dsRNA genetic elements in plant pathogenic fungi. Ann. Rev. Phytopathol. 28: 37-58. 124 Perkins, DD. 1988. Main features of vegetative incompatibility in Neurospora. Fungal Genet. Newslett. 35: 44-46. Perkins, DD. 1977. Neurospora Newsletter 24: 16—17. Pittenger, T.H. and KC. Atwood. 1956. Stability of nuclear proportions during growth of Neurospora heterokaryons. Genetics 41: 227-241. Pittenger, T.H. and T.G. Brawner. 1961. Genetic control of nuclear selection in Neurospora heterokaryons. Genetics 46: 1645-1663. Prout, T., C. Huebschman, H. Levene, and F .J . Ryan. 1953. The proportions of nuclear types in Neurospora heterocaryons as determined by plating conidia. Genetics 38: 518- 529. Puhalla, J .E. and S.L. Anagnostakis. 1971. Genetics and nutritional requirements of Endothia parasitica. Phytopathology 61: 169-173. Rayner, A.D.M. 1991. the phytopathological significance of mycelial individualism. Annu. Rev. Phytopathol. 29: 305-323. Rayner, A.D.M., D. Coates, A.M. Ainsworth, T.J.H. Adams, E.N.D. Williams, and N .K. Todd. 1984. The biological consequences of the individualistic mycelium, pp. 509- 540 in The Ecology and Physiology of the Fungal Mycelium, edited by D.H. Jennings and A.D.M. Rayner. Cambridge University Press, Cambridge. Rizwana, R. and W.A. Powell. 1992. Ultraviolet light-induced instability of vegetative compatibility groups of Cryphonectria parasitica. Phytopathology 82: 1206-1211. Rizwana, R. and W.A. Powell. 1995. Ultraviolet light-induced heterokaryon formation and parasexuality in Cryphonectria parasitica. Exp. Mycol. 19: 48-60. Chapter 4 The Effects of Vegetative Incompatibility Genes upon the Horizontal Transmission of Viruses in the Chestnut Blight Fungus, Cryphonectria parasitica, are Locus Specific and Modified by Epistasis Abstract The effects of specific vegetative incompatibility (vic) loci upon the horizontal (cytoplasmic) transmission of dsRNA hypoviruses in the chestnut blight pathogen, Cryphonectria parasitica, were examined. Two characterized hypoviruses and one uncharacterized dsRNA element were used in the transmission studies. Five vic loci were examined for their effects upon transmission. Differences in the effects of these five vic loci ean be eategorized as differences in the efficiency of transmission, reciprocality, and epistatic effects. Homoallelism at all vic loci always permitted transmission between donor and recipient strains. Strains heteroallelic at vicI exhibited nonreciprocality in transmission. A vicI-2 recipient strain was infrequently infected by a vicI-l donor strain while the reciprocal situation (vicI-2 donor, vicI-I recipient) always permitted infection. Locus vic3 also showed nonreciprocality in transmission but was more leaky than Vic]. Heteroallelism at either vic4 or vic5, or both, did not prevent transmission. The effect of vic2 was subject to epistasis. Heteroallelism at vic2 prevents transmission when the allele vicI-2 is also present in the recipient. But when the allele vicI-I occurs in conjunction with the allele vic2-I in the recipient, the transmission barrier created by vic2 becomes leaky. These data suggest that vicI-I is epistatic over vic2-I when both occur in the recipient, and corresponds to the nonreciprocality due to vicI . Other genetic background effects also caused the vic2 barrier to become unilaterally leaky. The effects of heteroallelism at vicI and vic2 upon transmission are epistatic over the effects of vic4 and vic5; vicI is epistatic over vic3. Fungi are hosts to an array of viruses, plasmids, and transposable elements (Buck 1986; Kistler and Miao 1992; Nuss and Koltin 1990; Griffiths, 1995). The horizontal mobility of these parasitic elements has been demonstrated in numerous laboratory transmission studies (Anagnostakis and Day, 1979; Brasier, 1984; Griffiths et al., 1990; Collins and Saville, 1990; Debets et al., 1994). Recent work highlighting the widespread 125 126 occurrence of viruses and plasmids in natural p0pulations of fungi suggests that horizontal transmission may also be a frequent occurrence in nature. For example, two recent surveys of the distribution of plasmids in Neurospora species have shown that some plasmids have worldwide distributions and are present in several species (Arganoza et al. , 1994a; Yang and Griffiths, 1993). The horizontal transmission of transposable elements has also been demonstrated by the movement of the transposon Tad between nuclei of a heterokaryon in N. crassa (Kinsey 1990), and has been inferred from the localized distribution of the grh retroelement in a subgroup of the rice blast pathogen, Magnaporthe grisea (Dobinson et al. 1993). The most frequently observed horizontally mobile genetic elements in filamentous fungi are viruses and double-stranded (ds) RNA elements. Viruses and dsRNA elements are common in fungi and have been shown to be infectious in nature in several species, notably, Ophiostoma ulmi (Brasier 1983) and Cryphonectria parasitica (Van Alfen et al. 1975). Concerted attention on C. parasitica, in particular, has uncovered a rich diversity of dsRNA viruses (Nuss 1992; Enebak et al., 1994; Polashock and Hillman, 1994; Chung et al., 1994; Smart and Fulbright, 1996), some of which have significant effects upon host fitness (Elliston 1985). Among the important factors defining host/parasite relationships is mode of transmission. Mode of transmission may determine the extent of the infectious spread of a parasite as well as the coevolutionary direction of the host/ parasite relationship (May and Anderson - 1983). Although little information exists as to the impact of most fungal parasitic genetic elements upon their hosts, the spread of dsRNA viruses in populations of the pathogens 0. ulmi and C. parasitica is thought to have had significant effects upon 127 fungal virulence (Brasier 1990). The transmission of viruses through fungal populations occurs through both horizontal (infectious) and vertical (intergenerational) routes, although vertical transmission is generally restricted to asexual spores rather than to meiotic offspring (Buck 1986). The horizontal transmission of viruses in filamentous fungi is dependent upon direct cytoplasmic contact from hyphal fusions between mycelia because extracellular infection is not possible (Buck 1986). The viability of intermycelial vegetative cell fusions in Ascomycetes is controlled by incompatibility systems which effectively mediate the formation of heterokaryons and provide a barrier to free parasite movement. Such systems in Ascomycetes can discriminate between conspecifics due to the presence of incompatibility loci (Glass and Kuldau 1992; Leslie 1993). Compatibility requires homoallelism at each of the incompatibility loci. When one or more of the compatibility loci of two individuals are heteroallelic, an incompatible reaction occurs which generally terminates the hyphal fusion by killing the fused cells and sometimes neighboring cells (Garnjobst and Wilson 1956; Newhouse and MacDonald 1991). One of the consequences of these vegetative cell compatibility systems is to maintain the genetic and physiological integrity of the individual (genotype) while permitting viable cell fusions to occur between some, though not all, close relatives. Vegetative incompatibility systems have been attributed to be a cellular defense mechanism for resisting infection by parasites and other suppressive cytoplasmic genetic elements (Caten 1972; Debets et al. 1994; Nauta and Hoekstra 1994). Cryphonectria parasitica (Murr.) Barn, the chestnut blight pathogen, is responsible for the chestnut blight pandemic which spread through the Castanea dentata (American 128 chestnut) population in eastern North America, and the Castanea sativa population in Europe (reviewed in Anagnostakis 1987; Fulbright et al. 1988; Griffin 1986; Heiniger and Rigling 1994). This pathogen/host system has been of particular interest because of the appearance of less virulent (hypovirulent) forms of the fungus in the North American and European populations which have acted as a natural biological control. The most frequent cause for the hypovirulent phenotypes of the fungus has been found to be novel dsRNA viruses which have recently been recognized as a new family, the Hypoviridae (Nuss 1992; Hillman et al. 1995). Although these viruses are horizontally transmissible by means of intraspecific hyphal fusions (anastomoses), their localized distribution in North America suggests that barriers to infection are present (MacDonald and Fulbright 1991). Recently, other infectious agents requiring intermycelial hyphal fusions for transmission, and unassociated with detectable dsRNA, have been found in nature and are also capable of causing transmissible hypovirulence (Fulbright 1985; Mahanti et a1. 1993; Huber et al. 1994). What are the rules governing the horizontal transmission of cytoplasmic genetic elements in fungi? The picture of horizontal transmission in C. parasitica that has emerged from studies of natural isolates has been complex and confusing; the presence of discernible rules governing transmission has been obscured by curious exceptions to apparent patterns. Virus transmission has been found to be prevented, permitted, limited to varying degrees, or to occur in nontransitive steps between different vegetative incompatibility genotypes (e.g. Anagnostakis 1983; Kuhlman et al. 1984), suggesting that individual incompatibility genes have different effects upon horizontal transmission. 129 This study addressed the question of how specific vegetative incompatibility (vic) genes affect the horizontal cytoplasmic transnrission of hypoviruses in C. parasitica. The effects of heteroallelism at five vic loci as well as interactions between these genes were examined A for their influence upon transmission. Individual loci were found to vary considerably in their effect upon transmission, and to be influenced by vic alleles at other loci. This study presents the first evidence of epistatic effects between specific vegetative incompatibility genes on virus transmission. Epistatic interactions were found to either decrease or increase the efficiency of cytoplasmic transmission. MATERIALS AND METHODS C. parasitica straim and culture conditions: The strains used in this study and their sources are listed in Table 1. Two characterized dsRNA hypoviruses and one uncharacterized dsRNA genetic element were utilized in the horizontal transmission tests. Hypovirus CHVl-713 is the most extensively characterized of the Cryphonectria hypoviruses (Nuss, 1992). CHV 1-713 was transferred sequentially into the transmission tester strains by the methods described below. This virus was first transmitted from Ep713 (source strain) to strain Ep78, and then from Ep78 to strain P1.9, and finally to strain 389.7. Hypovirus CHV3-GH2 was originally obtained from a nonlethal canker on a chestnut tree in Grand Haven, Michigan (Fulbright et al., 1983). Hypovirus CHV3-GH2 was transferred from strain GH2 to Ep289. The uncharacterized dsRNA-80-2 was obtained from strain 80-2 which was collected in West Virginia by William MacDonald (West Virginia University). This dsRNA may be of Italian origin as C. parasitica isolates 130 containing Italian dsRNAs were released into field plots for studies in West Virginia (William MacDonald, personal communication). Subcultures of the strains used in the transmission tests were taken from cultures grown on Petri plates containing potato dextrose agar (PDA, Difco). A single plate from which subcultures were drawn for the transmission tests was sometimes used for up to three months stored at room temperature. This storage or longer term storage on PDA plates at 4° C did not cause loss of dsRNA. Transmission tests: Tests for transmission of virus were conducted by placing a subculture of a virus-containing donor strain about 1 cm away from a subculture of a virus-free strain on PDA near one edge of a 100 X 15 mm Petri plate. Stacks of three to five Petri plates were kept under cool white flourescent lights at room temperature (about 25 ° C) and were rotated by placing the top plate on the bottom of the stack every other day while the cultures were growing. The donor and recipient strains grew across the Petri plates, making contact as they grew. Occasionally paired strains would grow close together but not make contact. When this occurred, the trial was rejected and repeated. Each of the three viruses used in the horizontal transmission tests produced a distinct, visible phenotype in the fungal strains which carried them. The transmission of a virus into a recipient strain was detectable by a distinct change in morphology of the recipient. The presence of hypoviruses in each of the donor strains was verified by dsRNA extraction using established protocols (Morris and Dodds, 1979; Fulbright et al., 1983). Each strain was also analyzed for the presence of dsRNA prior to infection to establish that unknown dsRNAs were not present. The reliability of morphological changes as an indicator of 131 hypovirus presence was thereby verified for every strain. In order to consider a virus transmission test as negative (no transmission), donor and recipient strains were required to grow in contact on the surface of the PDA for a distance of 2/3 of the plate diameter. That is, if transmission did not occur between strains that grew in contact with each other for a distance less than 2/3 plate diameter (because one strain grew faster than the other) this transmission test was rejected as invalid and repeated. No attempt was made to specify the amount of time that the strains had been in contact before transmission was observed. Hypovirus transmission tests were also conducted on live chestnut tissue. The method used was adapted from the Cryphonectria virulence assay developed by Lee et al.(1992). Chestnut stems about 2 cm in diameter were cut into lengths 4 to 5 cm long. Each section was split longitudinally and the bark/phloem outer layer was peeled off the underlying secondary xylem. This test was limited to stems collected during the growing season as the outer layer would become strongly attached to the secondary xylem during late fall and winter and be impossible to remove intact. The transmission test consisted of placing 5 mm3 plugs of fungal subcultures of actively growing donor and recipient strains in contact (side by side) in the center of the inside of the bark. The live bark tissue with the fungus was placed into sterile Petri dishes with a damp piece of Whatman filter paper and sealed with parafilm. The cultures were kept in the dark at room temperature (about 25 ° C) for one week. Hypovirus transmission into a recipient strain was observed by the dramatic reduction in the growth of the recipient as compared to the growth rate of controls of uninfected recipient strains on live chestnut tissue. Table 1. Strains used in this study. 132 Strain Genetic markers‘ Vegetative Hypovirus] Source‘ incompatibility dsRNA genotype” 389.7 cre, nic, MATI-I vial-1,24 ,3-1 ,4-I,5-I Mutagenesis of EP389 389.7(713) CHVl-7l3 this study 389.7(GH2) CHV3-GH2 this study 389.7(80—2) 80-2 this study F2.36 br cross 389.7 X 80-2c F2.36(713) CHVl—7l3 this study F2.36(GH2) CHV3-GH2 this study P2.36(80-2) 80—2 this study EP388 met, MATT-2 vicI-2,2-l ,3-1 ,4-1 ,5-1 ATCC 38979 EP388(713) CHV1-7l3 this study EP388(GH2) CHV3-GH2 this study EP388(SO-2) 80-2 this study F2.l7 br cross 389.7 X 80-2c F2.17(713) CHV1-7l3 this study F2.17(80-2) 80—2 this study BP289 met, MATT-2 vicl-l,2-2,3-I,4-l,5-I Conn. Agri. Exp. EP289(713) CHVl-7l3 this study EP289(GH2) CHV3-GH2 this study EP289(80-2) 80-2 this study 12.43 br cross F3.16 X Al.l3 J2.43(713) CHVl—7l3 this study 1243(80-2) 80-2 this study 22508 met. MA T1-2 vicl-2,2-2,3-l ,4-1 .5—1 ATCC 22508 22508013) CHVl—713 this study 22508(GH2) CHV3-GH2 this study 22508(80-2) 80-2 this study 11.27 br, met cross P316 X Al.l3 11.27013) CHV1-713 this study Jl.27(80-2) A 80—2 this study Nl.25 br vicI-I,2-1,3-2,4-l,5-I cross F4.l3 X Kl.43 N1.25(80-2) 80—2 this study Table l (cont'd). M1.5 M15013) M1.5(80-2) N 1.39 Nl.39(80-2) N1.36 N1.36(80-2) M1.14 Ml .l4(80-2) P3.26 P326013) P3.26(GH2) P3.16 P316013) P3.16(80-2) 12.31 12.31013) 12.23 12.23013) 12.23(80-2) P3.10 P310013) P3.10(GH2) P115 P115013) P3.15(80—2) P339 P339013) P3.39(80-2) P2.2 P2.2(713) 80-2c 80-2c(80-2) br, nic br CT! br, nic br, nic br, nic br, nic br, nit: br, nic br, nic br, nic br, nic hr 133 vicl-I,2-2,3-2,4-I.5-I vicl-2,2-I,3-2,4-l,5-I vicI-2,2-2,3-2,4-I,5-I vicI-I,2-I,3-I,4-2,5-I vicI-2,2-I,3-I,4-2,5-I vicl-I,2-2,3-I,4-2,5-I vicI-2,2-2,3-I,4-2,5-I vicI-I,2-I,3-l,4-l,5-2 vicI-Z,2-I,3-I,4—l,5-2 vicI-I,2-I,3-I,4-2,5-2 vhf-2.24.31.42.54 CHV1-713 80-2 80-2 80-2 80-2 CHV1-713 CHV3-GH2 CHV1-713 80-2 CHVl-7l3 CHV1-713 80-2 CHVl-713 CHV3-GH2 CHV1-713 80—2 CHV1-7l3 80-2 CHVl-7l3 80-2 cross K230 X 12.69 this study this study cross P4.13 X [(1.43 this study cross P4.13 X Kl.43 this study cross K230 X 12.69 this study cross 389.7 X 80-2c this study this study cnoss 389.7 X 80-2c this study this study cross P3.16 X A1.l3 this study cross P3.16 X A1.13 this study this study cross 389.7 X 80-2c this study this study cross 389.7 X 80-2c this study this study cross 389.7 X 80-2c this study this study cross 389.7 X 80-2c this study conidium of strain 80—2 this study Table l (cont’d). EP155 MATI-Z unknown ATCC 38755 EP713 MATT-2 unknown CHV1-713 ATCC 52571 EP2001 met, MATT-2 unknown ATCC 60589 EP2001013) CHVl-7l3 this study GH2 unknown CHV3-GH2 MSU collection' EP78 unknown ATCC 38752 P1.9 flown cross EP155 X 80-2c 134 ' Each genetic marker is also present in the hypovirus-infected derivative of each strain. " For brevity vegetative incompatibility genotypes have been written as vial-1.24 rather than as vicI-I. vic2-I. Each vc genotype applies to each strain in the column beneath the printed genotype until a new printed vc genotype appears. ‘ The crosses, mutagenesis of EP389, and derivation of strain 80-2c were described in Chapter 3. ‘ Strain provided by S. L. Anagnostakis, Connecticut Agricultural Experiment Station. ' Strain provided by Dennis W. Fulbright. Michigan State University collection. 135 Results Hypovirus effects upon fungal morphology, and the detection of transmission. Fungal strains were infected with two characterized hypoviruses, CHVl-7l3 and CHV3- GH2, and one uncharacterized dsRNA element, 80-2. Each of the infectious elements produced a characteristic morphological change in the host mycelium (Figures 1 and 2). The uncharacterized dsRNA 80-2 causes a loss of pigmentation and a reduction of conidiation in all infected strains. Mycelia] growth vigor on PDA was not obviously impaired by infection, although rarely a subculture from an infected strain would show an abnormal morphology and grow poorly. Strain J2.43(80-2) was consistently more debilitated as evidenced by growth cessation prior to reaching the Petri plate margin. Hypovirus CHV3-GH2 eauses the infected mycelium on PDA to grow appressed with fewer aerial hyphae and reduced mycelial vigor so that the growth front does not reach the margin of the Petri dish. Neither pigmentation nor conidiation were obviously reduced in infected strains. All of the strains infected with CHV3-GH2 exhibited similar effects upon growth morphology. Hypovirus CHVl-7l3 causes a loss of pigmentation, reduction of conidiation, and reduced growth vigor (see also Nuss, 1992). The effects upon mycelial growth could be variable both within a single strain and between strains although the range of variability within a single strain was itself characteristic of this hypovirus. For example, the mycelial growth front generally would stop before reaching the edge of the Petri plate although in some subcultures of a particular strain growth would proceed to the edge. In some strains, CHV 1-713 caused a more severe debilitation characterized as dense mycelium with an 136 Figure 1. Nonreciprocal transmission of 80-2 dsRNA between donor and recipient strains caused by heteroallelism at vicI. The strain on the left of each plate is the donor (infected with dsRNA); the strain on the right is the recipient (dsRNA-free). The recipient has become infected in the pairing in the right plate but not in the left plate. The left plate contains strains F2.36(80-2) and EP388. The right plate contains strains EP388(80-2) and F2.36. I37 Figure 2. Phenotypic changes in strain F2.36 caused by hypovirus infection. A. uninfected strain F2.36; B. strain F2.36(713) infected with CHVl-713; C. strain F2.36(GH2) infected with CHV3-GH2; D. strain F2.36(80-2) infected with dsRNA 80-2. 138 Figure 3. Banding patterns of dsRNA isolated from the infected strains of C. parasitica shown in Figure 2 electrophoresed in a 5% polyacrylamide gel and stained with ethidium bromide. Lanes: 1, F2.36; 2, F2.36(713); 3, F2.36(GH2); lane 4, F2.36(80—2). Sizes of dsRNA molecules present in strain F2.36(GH2) are indicated in kilobase pairs (kb). 139 irregular growth front that did not extend beyond about half the plate diameter. In other strains, particular subcultures of infected mycelium would grow with either the less dense, vigorous growth form or the dense, slow growing, more debilitated growth form. Similar variability in the effects of dsRNA upon growth morphology has been described by Anagnostakis (1981). Because the transmission assay required strains to grow in contact for a distance equal to two-thirds of the Petri plate diameter (about 6 cm), the more extremely debilitated strains or subcultures of strains were not used in transmission tests because they could not meet this growth requirement. The observation of hypovirus/dsRNA transmission between mycelia was dependent upon the morphological change in the recipient mycelium accompanying infection. All three dsRNAs transferred laterally through the recipient mycelium more rapidly than the mycelium grew (see also Martin and Van Alfen 1991). The donor and recipient mycelia were positioned on the medium surface so that they were not in immediate contact when growth commenced, thereby permitting the recipient to produce a portion of morphologically normal growth before dsRNA infection was possible through mycelial contact. Therefore, when infection occurred, the recipient mycelium exhibited a characteristic and easily observed laterally spreading morphological change that clearly differed from normal growth (Figure l). The time between contact of the growing strains and the visible detection of virus transmission could be quite variable. Compatible strains would normally pemrit transmission as soon as vegetative contact was made, whereas incompatible strains could permit transmission any time during which the actively growing mycelia were in contact 140 which might last up to ten days. Every strain used as a donor was also tested for dsRNA after initial infection to verify that the morphological change was caused by the presence of the hypovirus. Figure 2 shows the morphologieal changes in the fungus caused by each hypovirus, and Figure 3 shows the hypoviral genomes as they appear on a polyacrylamide gel. Oceasionally, the morphology of the recipient strain was ambiguous. In these cases the recipient was either subcultured for further observation of its morphology or a dsRNA extraction was performed to verify hypovirus infection. The effect of genetic background upon transmission when all vic loci are homoallelic. Comparisons of the effects of different genes upon phenotype are best considered in isogenic strains. Unfortunately, the long reproductive cycle of C. parasitica in culture precluded the feasibility of creating near-isogenic strains through backcrossing for this study. Instead, the effects of individual vic genes were examined both among some strains which shared several unlinked genetic markers as well as among strains with quite different genetic backgrounds to test whether the observed effects upon transmission were genotype specific. A number of the strains used are progeny from cross 389.7 X 80 2c or descendants from strain 389.7 (Table 1). Where possible, strains were used that carried the genetic markers hr and nic. These two markers are not linked to loci vicI , vic2, vic3, vic4, or vic5 (chapter 3). Hypovirus/dsRNA transmission was found to be unimpeded in all reciprocal tests between strains homoallelic with respect to several combinations of the alleles of the five vic loci, indicating that the genetic background did not have a discernible effect (Table 2). Of course, genes whose effect would be to reduce the transmission barrier caused by vic loci would remain unobserved in this test. 141 Table 2. Transmission of dsRNAs where donor am! recipied strains are homoallelic at all vic loci (=vegetatively compatible).' Transmission of dsRN A Donor Recipient vic’I 2 3 4 5 strain vic I 2 3 4 5 strain CHV1-7l3 CHV3-GH2 80-2 I I I I I 389.7 I I I I I F2.36 8/8 10/10 8/8 I I I I I F2.36 I I I I I 389.7 10/10 10/10 6/6 2 I I I I EP388 2 I I I I P2.17 10/10 7/7 8/8 2 I I I I P2.l7 2 I I I I EP388 5/5 NT‘ 6/6 I 2 I I I EP289 I 2 I I I 12.43 9/9 8/8 8/8 I 2 I I I 12.43 I 2 I I I EPZ89 8/8 NT 5/5 2 2 I I I 22508 2 2 I I I 11.27 8/8 8/8 8/8 2 2 I I I 11.27 2 2 I I I 22508 5/5 NT 6/6 I I I 2 2 EP243 I I I 2 2 F139 NT NT 8/8 I I I 2 2 P339 1 I I 2 2 EP243 NT NT 6/6 ‘ Virus transmission is given as the munber oftimes that the virus was detected in the recipient mycelium per number of times that the donor and recipient strains were grown together. Reciprocal pairings of donor and recipient strains are grouped as couplets in rows. hForbrevity viclociaredesignated only by their rarmber as a cohnnn heading with alleles for each locus listed inthecolumnbeneaththeappropriatelocus. Onlytwoalleles,designatedas Ior2,areknownforeachvic locus. ‘ NT = not tested. 142 The effects of vic4 and vic5 upon transmission. Table 3 shows reciprocal transmission tests using strains which are heteroallelic at vic4, or vic5, or both loci. To test for possible epistatic effects due to other vic alleles, strains with alternate combinations of alleles at vicI and vic2 were examined. Neither vic4 nor vic5 were found to impede the horizontal transmission of hypovirus/dsRNA. Simultaneous heteroallelism at both vic4 and vic5 also showed no diminution of transmission indicating that no additive effect occurred. The potential additive effects of vic4 and vic5 were also tested in conjunction with heteroallelism at vicI and vic2 (Tables 4, 5, and 6), and will be described below. The effect of via] upon transmission. The effects of heteroallelism at vicI upon horizontal transmission were tested using reciprocal pairings of thirteen different combinations of donor and recipient strains (Table 4). Reciprocal testing showed that the effects of vicI depended upon which allele of vicI was present in the donor strain and which was in the recipient (Table 4). When vic1-2 was in the recipient strain and vicI-I was in the donor strain, transmission of hypoviruses/dsRNA was prevented or greatly hindered. Most frequently, the infection of the vicI-Z strain did not occur at all. However, when the recipient carried vicI-I and the donor carried vicI-2, infection occurred in every test. Therefore, vicI has a nonreciprocal or unidirectional effect upon the horizontal transmission of hypoviruses. The potential epistatic effects of other vic genes upon heteroallelism at vicI were also tested by incorporating different vic alleles into the donor and recipient strains. First, the effects of heteroallelism at vic] were examined when donor and recipient strains were homoallelic for vic2-2 (pairing EP289 and 22508) to test for potential epistatic effects 143 Table 3. Transmission of dsRNAs where the donor and recipient strains are heteroallelic at vegetative incompatibility loci vic4 and vie5.‘ g a 1 Donor Recipient transmission of dsRNA vic’I 2 3 4 5 strain vic I 2 3 4 5 strain CHVl-713 CHV3-GH2 80-2 I I I I I 389.7 I I I 2 I F326 8/8 8/8 6/6 I I I 2 1 P326 I I I I I 389.7 6/6 5/5 NT‘ 2 I I I I EP388 2 I I 2 I F316 10/10 NT NT 2 I I 2 I F316 2 I I I I EP388 10/10 NT NT I 2 I I I EP289 I 2 I 2 I 12.31 10/10 NT NT I 2 I 2 I 12.31 I 2 I I I EP289 9/9 NT NT 2 2 I I I 22508 2 2 I 2 I 12.23 10/10 NT NT 2 2 I 2 I 12.23 2 2 I I I 22508 10/10 NT NT I I I I I 389.7 I I I I 2 F310 10/10 10/10 NT I I I I 2 F310 I I I I 1 389.7 8/8 5/5 NT 2 I I I I EP388 2 I I I 2 F315 10/10 NT NT 21112 F315 ZIIII EP388 7/7 NT NT I I I I I 389.7 I I I 2 2 F339 8/8 6/6 6/6 I I I 2 2 F339 I I I I I 389.7 9/9 6/6 6/6 2 I I I I EP388 2 I I 2 2 F22 10/10 NT NT 2 I I 2 2 P22 2 I I I I EP388 10/10 NT NT ' Virus transmission is given as the number of times that the virus was detected in the recipient mycelium per number of times that the donor and recipient strains were grown together. Reciprocal pairings of donor and recipient strainsare grouped as couplets in rows. ' For brevity vic loci are designated only by their number as a column heading with alleles for each locus listed in the column beneath the appropriate locus. Only two alleles, designated as I or 2, are known for each vic locus. ‘ NT = not tested. 144 Table 4. Transmission of dsRNAs where donor and recipient strains are heteroallelic at vegetative incompatibility locus vicI. The presence of additive and epistatic effects between vie genes was tested by varying the alleles present at vic4 and vie5.‘ Donor Recipient Transmission of dsRNA vie“) 2 3 4 5 strain vie I 2 3 4 5 strain CHV1-713 CHV3-GH2 80-2 I I I I I 389.7 2 I I I I BP388 0/8 0/9 2/15 2 I I I I EP388 I I I I I 389.7 10/10 8/8 16/16 I I I I I F2.36 2 I I I I EP388 0110 NT‘ NT 2 I I I I EP388 I I I I I F2.36 10/10 NT NT I I I I I 389.7 2 I I I I F2.17 0/10 NT NT 2 I I I I F217 1 I I I I 389.7 10/10 NT NT I I I I I P236 2 I I I I F2.17 0/10 NT NT 2 I I I I P2.l7 I I I I I P236 10/10 NT NT I 2 I I I EP289 2 2 I I I 22508 0/10 0/13 1/12 2 2 I I I 22508 I 2 I I I EP289 10/10 9/9 15/15 I I I I I 389.7 2 I I 2 I F316 1/8 NT NT 2 I I 2 I P316 I I I I I 389.7 10/10 NT NT I 2 I I I EP289 2 2 I 2 I 12.23 2/9 NT NT 2 2 I 2 I 12.23 I 2 I I I EP289 10/10 NT NT I I I 2 I P326 2 I I I I EP388 00 NT NT 2 I I I I EP388 I I I 2 I P326 10/10 NT NT I 2 I 2 I 12.31 2 2 I I I 22508 0/6 NT NT 2 2 I I I 22508 I 2 I 2 I 12.31 10/10 NT NT 145 Table4 (cont’d). I I I I I 389.7 2 I I I 2 F315 3/10 NT NT 2 I I I 2 F315 I I I I I 389.7 5/5 NT NT I I I I 2 F310 2 I I I I EP388 0/10 NT NT 2 I I I I EP388 I I I I 2 F310 10/10 NT NT 1 I I I I 389.7 2 I I 2 2 F22 10 NT NT 2 I I 2 2 F32 I I I I I 389.7 10/10 NT NT 2 I I I I EP388 I I I 2 2 P339 10l10 NT NT 1 I I 2 2 F339 2 I I I I EP388 00‘ NT 0/8 ' Virus transmission is given as the number of times that the virus was detected in the recipient mycelium per number of times that the donor and recipient mycelia were grown together. Reciprocal pairings of donor and recipient strains are grouped as couplets in rows. ’ For breviy vie loci are designated only by their number as a column heading with alleles for each locus listed in the column beneath the appropriate locus. Only two alleles, designated as I or 2, are known for each vie locus. ‘ NT = not tested ‘ Donor and recipient strains did not grow in contact for the minimum required distance because the donor grew poorly. 146 associated with allele vic2-2 (Table 4). This alteration in vc genotype had no effect upon transmission. Secondly, the possibility that heteroallelism at either vic4 or vic5, or both loci, in conjunction with vicI would decrease the transmission associated with vie] was also tested (Table 4). No additive effects due to the combined activity of vic], vie4 and vic5 could be found whereby the receptivity of the vicI—I strain was diminished, nor was the nonreceptivity of the vic]-2 strain obviously relaxed although each of the four recipient strains carrying vic4-2 or vic5-2 were infrequently infected. The effect of vie2 upon transmission. To determine the effects of heteroallelism at vieZ upon transmission, the first test used donor and recipient strains homoallelic for vicI-2 to preclude possible effects due to the nonreciprocal receptivity associated with vicI-I . Hypovirus transmission between strains EP388 and 22508 was found to be virtually prevented by vic2 (Table 5). Transmission was then examined between strains 389.7 and EP2 89 which were homoallelic for vie] -I to test whether vicI—I would have any effect upon the vie2 transmission barrier (Table 5). In this ease, the vic2 barrier became unidirectionally less restrictive where the vicI-I , vicZ-I recipient (389.7) became much more receptive to hypovirus infection. That is, when the recipient strain carried the allele vie2—2 and the donor carried vic2-I , transmission was prevented or occurred infrequently irrespective of whether both strains were homoallelic for vicI-I or vie] -2. But when the recipient strain carried vie] -1 along with vic2-I , the recipient became more receptive to virus infection. This demonstrates that locus vie2 prevents horizontal transmission in a conditional manner whereby increased cytoplasmic transmission results from the epistasis of vicI—I over vic2-I when both occur in the recipient. 147 The epistatic nature of the relationship between vicI-I and vic2-1 was demonstrated by pairing strains 389.7 and EP388 (Table 4). If vicI-I is epistatic over vic2-I, then vic2-I should not have a reciprocal effect upon vicI-I whereby the receptivity of vie] -] (when vie] is heteroallelic) is diminished. This pairing shows that the presence of vic2-I in the recipient does not diminish the receptivity of vie] -I . Therefore the relationship between these genes can be considered epistatic. The effect of heteroallelism at vic2 was also tested where the genetic background consisted of strains homoallelic for vicI-I, and heteroallelic for vic4, or vic5, or both loci. Three pairs of strains represent these genotypes: EP289 paired with F326, F310, and EP243 (Table 5). In all three of these pairs transmission occurred with a prominent nonreciprocal bias where the recipient that carried vicI-I and vic2-I was frequently infected. Heteroallelism at vic4, vic5, or both loci did not eliminate the epistasis of vieI-I over vie2-I. The epistasis of vie] -] over vic2-I was further tested using strains that were simultaneously heteroallelic at both vie] and vic2 (Table 6). Transmission was found to be prevented in reciprocal tests with strains EP289 and EP388, but transmission occurred unidirectionally between strains 389.7 and 22508 when the recipient carried both vicI-I and vic2-I (Fable 6). Concurrent heteroallelism at vic4 and vic5 also did not diminish viral transmission below levels associated with vie] and vicZ, nor was the unidirectional transmission associated with the epistasis of vicI-I over vic2-I prevented (Table 6). In summary, every test for the epistasis of vie] -I over vie2-1 in recipients showed the same relaxation of the vic2 transmission barrier. 148 Table 5. Transmission of dsRNAs where donor and recipient strains are heteroallelic at vegetative incompatibility locus vie2. The presence of additive and epistatic effects between vie genes was tested by varying the alleles at vie], vic4, and vic5.“ Donor Recipient Transmission of dsRNA vie‘] 2 3 4 5 strain vie I 2 3 4 5 strain CHV1-713 CHV3-GH2 80-2 I I I I I 389.7 I 2 I I I EP289 0/8 1/9 0/13 I 2 I I I EP289 I I I I I 389.7 5/11 6/9 13/21 2 I I I I EP388 2 2 I I I 22508 1/10 Ol8 0/13 2 2 I I I 22508 2 I I I I EP388 0/8 0/10 0/12 I I I I I 389.7 I 2 I 2 I 12.31 4/9 NT‘ NT I 2 I 2 I 12.31 I I I I I 389.7 3/8 NT NT I I I 2 I F326 I 2 I I I BP289 0/8 NT NT I 2 I I I EP289 I I I 2 I F326 7/9 NT NT I 2 I I I EP289 I I I I 2 F310 7/8 NT NT IIIIZ F310 1211] EP289 1/8 NT NT 2 I I I I EP388 2 2 I 2 I 12.23 6/14 NT 8/19 2 2 I 2 I 12.23 2 I I I I EP388 0/4 NT 0/21 2] IZI P316 22] I I 22508 0/9 NT 0/10 2 2 I I I 22508 2 I I 2 I F316 6/14 NT 9/10 2 I I I 2 F315 2 2 I I I 22508 0/12 NT 0/10 2 2 I I I 22508 2 I I I 2 F315 30 NT 9/10 I 2 I I I EP289 I I I 2 2 EP243 NT NT 8/12 I I I 2 2 EP243 I 2 I I I EP289 NT NT 1/13 gr .~ an.) 11-0.2... 149 Table 5 (cont’d) 2 I I 2 2 80-2c 2 2 I I I 22508 NT NT 0/9 22111 22508 21122 80—2c NT NT 10 ' Virus transmission is given as the number of times that the virus was detected in the recipient mycelium per number of times that the donor and recipient mycelia were grown together. Reciprocal pairings of donor and recipient strains are grouped as couplets in rows. ' For brevity vie loci are designated only by their number as a column heading with alleles for each locus listed in the column beneath the appropriate locus. Only two alleles, designated as I or 2, are known for each vie locus. ‘ NT = not tested. 150 Table 6. Transmission of dsRNAs where donor and recipient strains are heteroallelic at both vegetative incompatibility loci vie] and vicZ. The presence of additive or epistatic effects between vie genes was also tested by incorporating heteroallelism at vie4 and vie5.‘ Donor Recipient transmission of dsRNA vie’I 2 3 4 5 strain vie I 2 3 4 5 strain CHV 1-713 CHV3~GH2 80-2 2 I I I 1 EP388 I 2 I I I EP289 0/10 0/11 0/12 I 2 I I I EP289 2 I 1 I I EP388 0/10 0/11 0/13 I I 1 1 1 389.7 2 2 I I I 22508 0/8 0/8 0/13 2 2 I I I 22508 I I I I 1 389.7 7/8 5/8 5/15 2 I I I I EP388 I 2 I 2 I 12.31 3/11 N'I‘ NT I 2 I 2 I 12.31 2 I 1 1 I EP388 00 NT NT 2112] F316 1211] EP289 0/8 NT NT 1 2 I I 1 EP289 2 I I 2 1 F316 0/10 NT NT 2 I I I 2 F315 1 2 I I I EP289 0/8 NT NT I 2 I I I EP289 2 I I I 2 F315 0/8 NT NT 21122 80—2c 1211] EP289 NT NT 00 I 2 I I I EP289 2 I I 2 2 80-2c NT NT 0/6 I I I I 1 389.7 2 2 I 2 I 12.23 0/9 NT NT 2 2 1 2 I 12.23 I 1 1 I 1 389.7 20 NT NT I I I 2 I F326 2 2 I 1 I 22508 0/10 NT NT 2 2 I 1 I 22508 I I I 2 1 F326 4/9 NT NT 11112 P310 22111 22508 0/8 NT NT 22 1 I I 22508 I I I I 2 F310 3/9 NT NT 151 Table 6 (cont’d) I I I 2 2 BP243 2 2 I 1 I 22508 NT NT 0/13 2 2 1 1 I 22508 I 1 I 2 2 EP243 NT NT 5/16 ‘ Virus transmission is given as the number of times that the virus was detected in the recipient mycelium per number of times that the donor and recipient mycelia were grown together. Reciprocal pairings of donor and recipient strains are grouped as couplets in rows. " For brevity vie loci are designated only by their number as a column heading with alleles for each locus listed in the column beneath the appropriate locus. Only two alleles, designated as I or 2, are known for each vie locus. ° NT = not tested. 152 Other genetic background effects upon the transmission barrier caused by vic2. In the course of testing the epistasis of vicI-I over vic2-1, the genotypes were further modified by incorporating heteroallelism at the weak incompatibility loci vic4 and vic5 in conjunction with heteroallelism at vie] and vic2. This revealed additional genetic background effects that also resulted in the unidirectional reduction of the transmission barrier caused by vic2. First, strains were paired that were homoallelic for vicI—2, heteroallelic for vic2, and heteroallelic for either vic4 or vic5: EPBSS was paired with 12.23, and 22508 was paired with both F3.16 and F3.15 (Table 5). In these three reciprocal pairings transmission was still inhibited when vic1-2 and vicZ-I occurred together in the recipient, but the reciprocal pairings resulted in an increased level of transmission relative to the other tests of vic2 described above. This higher receptivity of strains 12.23, F3.15 and F3.16 was associated with the presence of either vic4-2 or vie5-2. It is possible that these genes are responsible for the unidirectional relaxation of the vic2 transmission barrier. To further test the putative effects of vic4-2 and vic5-2 upon heteroallelism at vic2, three additional tests were conducted. Strains 389.7 and 12.31 were paired, and again the transmission barrier caused by vic2 was relaxed in the recipient carrying vic4-2 (Table 5). Strains Ep388 and 12.31 were also tested where both vie] and vic2 were heteroallelic (Table 6). Here, a small reduction of the vie2 transmission barrier may be present but is not clear. lastly, strains 22508 and 80-2c were used to test the effects of simultaneous heteroallelism at vic2, vic4 and vicS (Table 5). This test resulted in the inhibition of transmission in reciprocal tests rather than a unidirectional relaxation of the vic2 barrier. 153 Modifications of transmission due to these gene associations require further testing. The effect of vie3 upon transmission, and modifications due to epistasis. Heteroallelism at vie3 was found to cause nonreciprocal transmission (Table 7). This analysis also indicated that locus vie] may exhibit two distinct epistatic effects upon vie3. In one case, allele vicI-I appears to reduce the transmission barrier caused by vie3 when both viel-I and vie3-I are present in a recipient. In the other case, the unidirectional inhibition due to heteroallelism at vie] was epistatic over the unidirectional transmission caused by vie3. Some evidence was also found that suggests that the unidirectional transmission of vic3 may be epistatic over vie2 in an allele specific manner. Not all of the combinations of the alleles at vie], vie2, and vie3 were tested, nor were multiple genetic backgrounds with each vc genotype tested. As with the previous vie loci, the effects of vie3 were first tested with homoallelism at the other vie loci. Since vie] -1 was known from previous studies to have an epistatic influence upon another vie gene, the first tests of vie3 were performed in pairs of strains homoallelic for vie1-2. Transmission tests between the pairs of strains 22508/Ml.14 and EP388/N 1.39 showed that vie3 is also associated with nonreciprocal transmission (Table 7). Recipients with allele vie3-2 were infected by vie3-I donors in most tests while the reciprocal relationship (recipient vie3-1, donor vie3-2) resulted in much less or no transmission. The potential epistatic modification of the vie3 transmission barrier by either allele vicI-I or heteroallelism at vie] was tested as follows. First, the effect of vie] on vie3 was examined by testing the effects of both vie] alleles upon vie3 alleles. The barrier to 154 Table 7. Transmission of dsRNAs where donor and recipient strains are heteroallelic at vie3. The presence of additive and epistatic effects between vie genes was tested by varying the alleles at vie] and vie2.‘ - u Donor Recipient Transmission of dsRN A vie’] 2 3 4 5 strain vie I 2 3 4 5 strain CHV1-713 80-2 2 I I I 1 EP388 2 1 2 I I N139 NT‘ 10/10 2 1 2 I I N139 2 I 1 I I BP388 NT 0/10 1 I I I 1 389.7 I I 2 I I N1.25 5/6 13/13 11211 N1.25 1111] 389.7 NT‘ 8/16 I 2 1 1 I EP289 I 2 2 I 1 M13 9/10 NT 1221] MLS 12111 EP289 1/7 NT 22] 1 I 22508 2221 I M1.14 10/10 17/17 22211 Ml.l4 22111 22508 NT“ 0/9 I 1 1 I 1 389.7 1 2 2 I I MLS NT‘ 2/12 1 2 2 I 1 M15 1 I I I 1 389.7 0/5 1/14 2211] 22508 2121] N139 NT 5/9 2121 1 N139 22] I I 22508 NT 0/10 I 2 I 1 I EP289 I I 2 I I N1.25 1/3 1/4 11211 N1.25 1211] EP289 0/6 0/8 I 1 I I 1 389.7 2 1 2 I I N139 NT‘ 1/13 2 1 2 1 I N139 1 I 1 I 1 389.7 NT 13/14 I I 1 I 1 389.7 2 I 2 I I N136 NT 3/8 2 I 2 I I N136 I I I I I 389.7 NT 6/6 155 Table 7 (cont‘d) I I I 1 1 389.7 2 2 2 I I M1.l4 NT 0/10 2 2 2 I 1 M1.l4 I I I I I 389.7 NT 3110 ' Virus transmission is given as the number of times that the virus was detected in the recipient mycelium per number of times that the donor and recipient strains were grown together. Reciprocal pairings of donor and recipient strains are grouped as couplets in rows. “ For brevhy vie loci are designated only by their number as a column heading with alleles for each locus listed in the column beneath the appropriate locus. Only two alleles, designated as I or 2, are known for each vie locus. ‘ NT = not tested. ‘ Transmission tests were attempted but the donor strain grew too poorly to meet minimum assay growth requirements. 156 transmission caused by vic3 was found to be reduced when the recipient carried alleles vie3-1 and vieI-I, and did not require heteroallelism at vie] (pair 389.7/N1.25; Table 7). This indicates that vie1-I may be epistatic over allele vie3-I in a similar manner to vie2-I . Another test of this putative epistatic relationship was conducted using strains EPZB9 and M15 which were homoallelic for vie2-2 (Table 7). This test, however, did not show the same reduction in the transmission barrier. Secondly, the effect of heteroallelism at vie] upon heteroallelism at vie3 was tested by using strains where the unidirectional transmission orientations of vie] and vie3 were opposed (pairings 389.7/N 1.39 and 389.7/N 1.36) (Table 7). In both tests transmission was found to have a unidirectional bias where strain 389.7 was more frequently infected. This indicated that the unidirectional inhibition of vie] was epistatic over the unidirectional transmission of vie3. To determine whether this epistatic relationship was due to epistasis between specific alleles (vieI-2 and vie3-2) in the recipient or due to the masking of the heteroallelic vic3 reaction by heteroallelism at vie], strains EP388 and N139 were paired in reciprocal transmission tests (Table 7). Transmission was again found to be unidirectional, demonstrating that the presence of vie] -2 and vie3-2 in the recipient was insufficient to stop transmission, and consequently showing that the elimination of receptivity in the vie3-2 recipient required heteroallelism at vie] . Is the receptivity of allele vie3-2 epistatic over vie2-I in a similar manner to the epistasis of vicI-I over vie2-I? A reciprocal transmission test between strains 22508 and N 1.39 provided evidence that the unidirectional transmission of vie3 is epistatic over the transmission barrier caused by vie2 (Table 7). Evidence that this epistatic interaction is 157 allele specific was provided by pairing strain M1.5 with 389.7 (Table 7). An additional test of allele specificity using strains Ep289 and N1.25 was inconclusive because of an insufficient number of successful transmission tests (Fable 7). Further testing is needed to establish the presence of epistatic interactions between the alleles of vie2 and vie3. Prevention of transmission due to undescribed incompatibility genes. The transmission barrier imposed by the incompatibility genotype of strain EP2001 (vc type 10) was evaluated with several tester strains representing differences at each of the five vie loci (Table 8). Strain EP2001 represents the most divergent vegetative compatibility genotype known relative to vc type 5 (vieI-I, vieZ-I, vie3-1, vie4-I, vie5-1) (Anagnostakis, 1982). In addition, EP713 was tested as a donor and recipient with EP2001. Vegetative compatibility type 40 (EP713) is reported to differ from vc type 10 (EP2001) at a single vie locus (Anagnostakis, 1988), but it is not known which vie alleles EP2001 shares with any of the tester strains. Hypovirus transmission occurred freely between EP713 and EP2001 indicating that EP2001 was competent as both a donor and a recipient. However, none of the seven tester strains was capable of infecting EPZOOl. Six of seven tester strains could not be infected by EP2001(713), but 389.7 was infected (Table 8). Based upon the transmission profiles representing most combinations of the alleles of the strong vie loci, vieI-vie3, this unusual result may also be due to epistatic interactions between unidentified vie genes. Horizontal transmission efficiencies of different dsRNA genomes. Two characterized dsRNA hypoviruses, CHV1-713 and CHV3-GH2, and one uncharacterized dsRNA element, 80—2, were used in transmission tests to see if different cytOplasmic 158 Table 8. Transmission of dsRNAs where the vie gene differences between vegetatively incompatible donor and recipient strains are not known.‘ Donor Recipient Transmission of dsRNA vie"1 2 3 4 5 strain vie I 2 3 4 5 strain CHV1-713 80-2 1 I I 1 1 389.7 unknown EP2001 0/5 0/8 unknown EP2001 I 1 1 I 1 389.7 8/8 NT‘ 2 I I I I EP388 unknown EP2001 0/5 0/8 unknown EP2001 2 I I I I EP388 0!? NT 1 2 I I I EP289 unknown EP2001 0/5 on unknown EP2001 I 2 I I 1 EP289 0/8 NT 2 2 I I I 22508 unknown EP2001 0/7 on unknown EP2001 2 2 I I I 22508 0/8 NT 2 2 2 1 I M1.14 unknown EP2001 0/‘7 NT unknown EP2001 2 2 2 I I M1.14 0/5 NT 2 I 1 2 2 80-2c unknown EP2001 NT 0/8 unknown. 292001 2 r 1 2 2 arm NT NT unknown EP'II3 unknown EP2001 5/5 NT unknown EP2001 unknown EP713 5/5 NT ' Virus transmission is given as the number of times that the virus was detected in the recipient mycelium per number of times that the donor and recipient strains were grown together. Reciprocal pairings of donor and recipient strains are grouped as couplets in rows. ' Rrr brevity vie loci are designated only by their number as a column heading with alleles for each locus listed in the column beneath the appropriate locus. Only two alleles, designated as I or 2, are known for each vie locus. ‘ NT = not tested. 159 Table 9. Transmission of hypovirus CHVl-7l3 between donor and recipient strains conducted on live chestnut tissue.‘ Donor Recipient transmission of vrrus vie” I 2345 strain vie] 2345 strain CHV1-713 I I I I 1 389.7 I I I I 1 F2.36 4/4 1 I I I 1 389.7 2 1 I I 1 F2.17 1/6c 21111 F2.17 11111 389.7 6/6 1 I I 1 1 389.7 I 2 I 1 1 12.43 +" I 1 I I 1 389.7 I I 2 I 1 N1.25 2/3 I 1 I I 1 389.7 1 I I I 2 F3.10 4/4 I I I I 1 389.7 1 I I 2 1 F326 4/4 1 I 1 I 1 389.7 I I I 2 2 F339 4/4 ‘ Virus transmission is given as the number of times that the virus was detected in the recipient mycelium per number of times that the donor and recipient strains were grown together. See materials and methods for an explanation of procedures. ' For brevity vie loci are designated only by their number as a column heading with alleles for each locus listed in the column beneath the appropriate locus. Only two alleles, designated as I or 2, are known for each vie locus. ° This single record of infection is uncertain because of the presence of contaminants on the wood. Two other possible infections of F2.17 were also observed, but are not recorded because of more severe contamination in those trials. ‘ One pairing between 389.7(713) and 12.43 showed infection of 2.43 upon subculturing although contaminants were present. Other pairings were also done, but contaminants obscured the results. 160 elements transmit with different efficiencies. In most tests all three cytoplasmic elements were transmitted with equivalent efficiency. Because CHV1-713 causes a more severe debilitation in the fungus, transmission tests sometimes had to be repeated several times in order to obtain pairings that met the growth standards set for this study. This only had to be done when the donor strain differed from the recipient at vie] or vie2 because transmission between compatible strains or those heteroallelic at vie4 and vie5 occurred rapidly so that the distance of growth contact before infection was not significant. Horizontal transmission on live chestnut tissue. Because the effects of vegetative incompatibility upon heterokaryon formation were found to vary depending upon whether strains were grown on PDA or chestnut tissue (Chapter 3), hypovirus transmission was also tested on live chestnut tissue using a limited number of vc genotypes (Table 9). Viral transmission was unimpeded between strains homoallelic at all vie loci. Heteroallelism at vie], vie4, and vie5 were also found to have similar effects upon transmission when strains were grown on live chestnut compared to those grown on PDA. Transmission was found to occur when vie2 was heteroallelic although this represents only a single observation because other tests were destroyed by contaminants. vie3 was found to limit, but not prevent, transmission. Additional testing on live chestnut tissue is necessary to determine the efficiency of transmission through the vie2 and vie3 incompatibility barriers. Ideally, these transmission tests should also be conducted by infecting live trees with these fungal strains so that mycelial interactions can be observed under natural conditions. 161 Discussion The effects of five vie loci upon the horizontal cytoplasmic transmission of hypoviruses in C. parasitica have been found to be quite variable. The different types of effects can be categorized as differences in the efficiency of the transmission barrier, presence or absence of reciproeality in transmission, and modifications in transmission due to epistasis between vie genes. Individual vie genes have different effects upon horizontal transmission. This study found that the effectiveness of vegetative incompatibility polymorphisms in C. parasitica as a barrier to horizontal transmission is quite variable due to the unique effects of individual vie genes (summarized in Table 10). Heteroallelism between conspecifics at loci vie4 and vie5, individually and in concert, did not prevent hypovirus and dsRNA element transmission. In contrast, heteroallelism at vie], vie2, and vie3 can preVent the transmission of these cytoplasmic parasites although their effectiveness is dependent upon either reciprocality (vieI and vie3) or upon epistasis by other vie genes (vie2 and vic3). The permissiveness to hypovirus transmission of each of the five vie loci, and the epistatic modifications of transmission at two of the vie loci, stand in sharp contrast to the restrictiveness of the same loci with regard to horizontal nuclear transmission. Heterokaryosis was previously found to be strictly prevented by each of the vie loci when mycelia were grown on chestnut tissue. On an agar substrate vie], vie2, and vie3 were also found to prevent heterokaryon formation although vie4, and especially vie5, were associated with abnormal mycelial interactions suggestive of limited nuclear exchange (Chapter 3). In contrast, vie4 and vie5 always permitted hypovirus transmission on both 162 Table 10. Summary of the effects of each vie locus upon the horizontal (cytoplasmic) transmission of hypovinrses in Cryphonectria parasitica. r — E vie locus‘ Effect upon transmission” Exceptions (due to epistasis or genetic background) vie] nonreciprocal none vieI-I - vie1-2 vie2 reciprocal inhibition 1. Unidirectional transmission vie2-I - I - vie2-2 where allele vieI-I is epistatic over allele vieZ-I in the recipient: .dnnnr. recipient I-2,2-2 - I-I,2-I I-I,2-2 - 1-1,2-I 2. Unidirectional transmission associated with vie4-2 and vie5-2 in recipient. 3. Unidirectional transmission where vie3—2 may be epistatic over vie2-1.° vic3 nonreciprocal l. Heteroallelism at vieI is epistatic vie3-I -° vie3-2 over heteroallelism at vie3: I-I.3-I °- I-2,3-2 2. Allele vieI-I may be epistatic over allele vie3-I.° vie4 no inhibition none vie4-I ~ vie4-2 vie5 no inhibition none vie5-I ~ vie5-2 ‘ Each vie locus has two known alleles designated I or 2, and written as vieI-I and vie]- 2, etc. " Arrows indicate direction of hypovirus transmission. -I~ indicates no (or little) transmission in either direction. ° Requires further testing. 163 PDA and live chestnut tissue. Both the study of hypovirus transmission and heterokaryosis are congruent, though, in so far as they demonstrate that vie4 and vie5 are less restrictive to cytoplasmic exchange than vie], vie2 and vie3. Interestingly, while simultaneous heteroallelism at vic4 and vie5 showed an additive effect that prevented the formation of abnormal mycelial interaction zones, they did not cumulatively inhibit hypovirus transmission. The differences between viral and nuclear transmission for all five vie loci may indicate that small cytoplasmic elements are able to breach some incompatibility barriers that entirely prevent the horizontal movement of the larger, and less abundant nuclei. Strains of C. parasitica previously have been found whose ability to infect several different vegetative incompatibility genotypes was considered novel (Kuhlman et al. 1984; Kuhlman and Bhattacharyya 1984). At least two types of genetic explanations could account for this behavior. Suppressor mutants have been found in Neurospora crassa (Arganoza et al. 1994b) and in Podospora anserina (Bernet 1992b) that suppress specific vegetative incompatibility genes. It is possible that some C. parasitica strains that can transmit dsRNA to multiple vc types contain similar suppressor mutations. Alternatively, those strains capable of receiving dsRNA from a broad donor may simply differ from the donor at loci such as vie4 and vie5 which cause incompatibility but do not stop cytoplasmic transmission of dsRNA. For example, the present genetic analysis of the vc genotype vie1-2, vie2-2, vie3-1, vie4-I, vie5-I has shown that it could infect at least 20 of the 32 vc groups derived from the recombination of alleles at vie] through vic5, and possibly more. This genotype could certainly be referred to as having broad infection capabilities, 164 but its transmission behavior is determined by the particular complement of vie alleles that it and the recipient strains possess, and not by suppressor mutations or other unique characteristics. Therefore, without specific genetic evidence for the suppression of incompatibility, the capacity to broadly infect need only be considered a function of the particular vie alleles present in a population. What effects will vie gene polymorphisms have upon virus transmission in natural populations of C. parasitica? This study has shown that the effects of vegetative incompatibility upon horizontal transmission will depend upon both the particular loci that are polymorphic in a population as well as the complement of vie alleles present in each individual. C. parasitica populations polymorphic at only loci such as vie4 and vie5 would be expected to experience much horizontal transmission. Polymorphisms at other loci such as vie2 may pose a significant barrier to transmission although epistasis may mitigate the restrictiveness of particular loci such as vie2. Nonreciprocal transmission may also become important depending upon which genotypes are acting as donors and recipients in a population. The nonreciprocal transmission found between genotypes vc 5 and vc 10, which probably differ at seven vie loci, likely represents an outstanding example of the effects of epistasis, and well illustrate the difficulty in predicting transmission efficiencies without prior knowledge of the genotypes involved. Field studies are needed to test whether these vie genes limit virus transmission in a similar manner in natural populations (Double, 1982). Other studies have also found that individual vegetative incompatibility genes in Ascomycetes have different effects upon the horizontal transmission of cytoplasmic agents 165 and nuclei. Handley and Caten (1973) observed that heat and herB have different capacities to hinder transmission of the cytoplasmic vgd mutation in Aspergillus amstelodami. In Ophiostoma ulmi, a single incompatibility locus was found to limit the transmission of the d-factor more than several other loci (Brasier 1984). Recently, Debets et al. (1994) found that her loci in N. crassa can have different effects upon the cytoplasmic transmission of plasmids and senescence. Coenen et al. (1994) and Dales et al. (1983) have found differences in the effects of vegetative incompatibility loci upon heterokaryon formation. N onreciprocal transmission caused by vie loci. Unexpectedly, nonreciprocal (unilateral) horizontal transmission was found to be produced by vie] and vie3 and in conjunction with some epistatic modifications of vie2. The pronounced nonreciprocality eaused by vie] and vie3 was always associated with particular alleles at these loci and was never found to be reversed. The inhibitory effects of vie2 were also found to be decreased nonreciprocally due to the epistasis of vie] -1 over vie2-I and to other genetic background effects that are probably attributable to the presence of alleles vie4-2 and vie5-2. Liu and Milgroom (1996) have also found nonreciprocal transmission. The present study is the first demonstration of nonreciprocal horizontal transmission associated with particular vegetative incompatibility loci in an Ascomycete. Other types of unilateral effects associated with incompatibility reactions between the somatic cells of fungi and plasmodial protists are known. The Ascomycete Podospora anserina exhibits unilateral cell death with respect to a nonallelic interaction between particular alleles of the incompatibility loci r and v (Bernet 1992a and references therein). 166 Bernet has suggested that the r locus produces a transfusable hormone-type signal while v produces its receptor. Directional cell death responses also occur in some somatic cell reactions between incompatible plasmodia of the cellular slime mold Didymium (Clark and Collins 1973). It is not known whether similar mechanisms underlie these processes. Evidence that epistasis between vie genes modifies cytoplasmic transmission. This study has provided evidence for three different types of epistatic interactions between vie genes. The first type of epistasis occurred between specific alleles at different vie loci and decreased the barrier to cytoplasmic transmission caused by one of the loci. The clearest example of this type of epistasis was the effect of the vie] -1 allele upon the vie2-I allele where the vie2 transmission barrier was reduced when both alleles occurred in the recipient. No other epistatic relationships were found among the other combinations of vie] and vie2 alleles. It is also undoubtedly significant that this epistatic relationship was associated with the unidirectional transmission caused by vie] and that it was present whether or not vie] was heteroallelic; that is, epistasis did not depend upon the presence of heteroallelism at both vieI and vie2. Therefore, this interaction cannot be described as an additive effect produced by two separate incompatibility reactions, but rather is evidence for the epistatic modification of one vie gene by another. The second type of epistatic relationship occurred when the effect of heteroallelism at one vie locus was epistatic over the effect of heteroallelism at a different vie locus. The barriers to transmission eaused by vie], vie2, and vie3 were epistatic over the uninhibited transmission associated with vie4 and vie5. Interestingly, although both vie] and vie3 were found to cause unidirectional transmission, the effects of vie] were epistatic over vie3, 167 indieating that there are differences in the action of these two phenotypically similar loci. The nonreciprocal transmission observed between vc types 5 and 10 is particularly striking because these two genotypes probably differ at seven vie loci (Anagnostakis 1982), and may also be due to epistasis. Another example of epistasis between vegetative incompatibility loci is in 0. ulmi where transmission inhibition of the w locus is epistatic over the less restrictive barriers of several other incompatibility loci (Brasier 1984). Less conclusive evidence was found for a third type of epistatic relationship. The vie2 incompatibility barrier was also relaxed unilaterally by other genetic background effects associated with the presence of alleles vie4-2 and vie5-2 when they occurred in the recipient. In this case, the apparent epistatic association was between alleles vie4-2 or vie5-2 and locus vie2, where the vie2 transmission barrier was reduced irrespective of which vie2 allele was present in the recipient. That all modifications of transmission resulted in an unusual unilateral relaxation of the transmission barrier suggests that a similar cause may be responsible for the effect in each case. If so, the other unilateral alterations of the vie2 barrier may be due to the presence of alleles vie4-2 and vie5—2 which were always present when the effects were observed. Although vie4 and vie5 do not prevent cytoplasmic transmission, the effects of these loci may be similar to that of vie] but undetected by the virus transmission assay. However, this putative epistatic relationship requires further testing. The epistasis of vie] -I over vie2-1 is quite different from the nonallelic incompatibility found in Podospora anserina because the nonallelic interactions in P. anserina cause an incompatibility reaction whereas epistasis in C. parasitica does not. In fact, the epistasis 168 between vie genes in C. parasitica actually acted to reduce the effectiveness of the transmission barrier caused by the vie2 and vie3 allelic incompatibility reactions. Significantly, this study has demonstrated that epistatic interactions can increase cytoplasmic transmission even when the number of heteroallelic vie loci increases. This is the first evidence for epistasis between vegetative incompatibility genes that decreased the barrier to cytoplasmic transmission imposed by a particular vegetative incompatibility locus in an Ascomycete. Several authors have provided evidence that the transmission barrier imposed by a vegetative incompatibility reaction in C. parasitica becomes incrementally greater with each additional heteroallelic incompatibility locus; that is, the effects of vegetative incompatibility loci are additive (Anagnostakis and Waggoner 1981; Liu and Milgroom, 1996). The recent work by Liu and Milgroom is especially significant because they examined a series of strains fiom nature that differed by 0, l, 2, and more than 2 vie loci, and found that generally the frequency of transmission between unknown genotypes was successively decreased by increasing the number of heteroallelic vie loci. However, because the vc genotypes of their strains were unknown it is not possible to strictly attribute the differences in transmission to the successive addition of new genes rather than the individual effects of particular genes. For example, some of their vc genotypes that differed at two vie loci did not exhibit decreases in transmission relative to other single vie gene differences, indicating that transmission is influenced by particular genes as well as by additive effects among the genes. In contrast, I have found that heteroallelism at a single vie locus (vieI or vie2) can be 169 more prohibitive to transmission than a particular four locus difference. Although untested, certain five locus differences should also be less prohibitive than vie] and vie2. It is important to note that these permissive four or five gene differences include heteroallelism at vie] and vic2. In other words, because of the combined effects of nonreciproeality and epistasis, transmissibility even depends upon the total complement of vie alleles the donor and recipient carry (vc genotype) and not simply on heteroallelism at a certain locus. It is likely that the strains examined by Liu and Milgroom (1996) which may demonstrate additive effects represent different vie genes than those studied here. An important corollary arising from the transmission genetics is that genetic relatedness between vc genotypes cannot be directly inferred from the degree of cytoplasmic transfer because of epistasis and the widely different effects of individual vie loci. The implications of cytoplasmic transmission variability for vie gene action. The cytoplasmic transmission of viruses has proven to be a sensitive assay for the subtle differences in the effects of individual vie genes. Although the mechanism of vie gene action is not known, it has been suggested that the gene products produced by allelic vie genes may form either homomultimeric or heteromultimeric complexes that function to trigger cell death through either the induction of incompatibility or through the blocking of inhibitors of the reaction (Arganoza et al. 1994b; Begereut et al. 1994). It is especially intriguing that every genetic alteration of the incompatibility reaction resulted in a nonreciprocal (unilateral) effect upon hypovirus transmission. The epistasis between particular vie alleles suggests that these genes, or gene products, are interacting with each other or with a common site of action in an antagonistic manner. The phenomena 170 described here are reminiscent of the effects of some of the suppressor mutations recently studied by Arganoza et al. (1994b) that affect one allele but not both alleles at a het locus, although in C. parasitica, of course, suppression is associated with epistasis between vie genes. The nonreciproeal transmission and epistasis between vie genes found in this study indicate that the interactions of the vie genes (vie gene products) are more complex than previously recognized. Presumably, viral transmission is prevented by the death of the heterokaryotic cells and the concomitant plugging of the septa] pores that form a boundary for the cell death reaction (Garnjobst and Wilson 1956; Newhouse and MacDonald 1991). Conceivably, the cytoplasmic transmission of viruses could be limited by the rate of the cell death reaction that terminates the cytoplasmic pathway, by the extent of cell death, or by some other effect upon the cytoplasm that affects viral particle movement. Variability in the rate and extent of cell death have been observed in N. crassa (Garnjobst and Wilson 1956). As mentioned above, unilateral cell death has been found in nonallelic incompatibility interactions in P. anserina (Labarere et al. 1974). I have observed barrages in C. parasitica produced by vie] on PDA to be more extensive in the vie1-2 mycelium suggesting that the unilateral cyt0plasmic transmission may be due to a unilateral cell death response. Since nothing is known about the molecular interactions of the vie gene products in allelic incompatibility, or how this interaction is translated into cell death, it is difficult to speculate on how the allelic interactions can be epistatically modified as observed in this study. However, any model of allelic incompatibility must take into account the following 171 phenomena. First, vie loci have been found to have three basic effects upon cytoplasmic transmission: reciprocal inhibition, nonreciprocal (unilateral) inhibition, and no inhibition. Secondly, incompatibility caused by some vie loci can be modified by epistasis between specific vie alleles as well as between heteroallelic vie loci. Lastly, every genetic modification of transmission occurred when the respective genes were present in the same nucleus, and did not occur when the same genes were placed together in the transient heterokaryotic state that follows hyphal fusions between incompatible strains. Transmission efficiences of different viruses. Do different types of cytoplasmic genetic elements transfer with different efficiencies through hyphal fusions or through vegetative cell incompatibility reactions? The two characterized hypoviruses, CHV1-713 and CHV3-GH2, and the uncharacterized dsRNA-80-2 transferred with comparable efficiencies through the vegetative incompatibility barriers imposed by the five vie loci. A limited examination of the transmission profile of a non—dsRNA-induced senescence phenotype in C. parasitica has shown that it transmits cytoplasmically with similar efficiency to the hypoviruses (Chapter 5). Primary function of vegetative incompatibility genes. What is the primary function of vegetative incompatibility genes? Is their function to restrict the infectivity of parasitic cytoplasmic elements? What other possible primary functions might these genes have? Several hypotheses have been proposed. The most widely considered hypothesis is that vegetative-incompatibility functions as a protective mechanism to prevent an individual (genet) from infection by cytoplasmically restricted parasitic genetic elements, including viruses, plasmids, and mutant mitochondria (Caten 1972; Nauta and Hoekstra 1994). A 172 second hypothesis, similar to the first, considers vegetative incompatibility to protect against cellular parasitism which could occur if a less fit nuclear genotype proliferates in a cytoplasm along with a more fit genotype (Hartl et al. 1975; Buss 1987). A third hypothesis regards vegetative incompatibility to be the pleiotropic effect of mutant genes whose normal function lies elsewhere, such as in development (Bernet 1992a, and references therein). Other proposed functions for incompatibility genes are the maintenance of the developmental and/or physiological independence of the mycelium (Rayner and Coates 1987) and ecologieal competition (Rayner et al. , 1984; Rayner, 1991). I have shown in chapter 3 that all five vie loci in C. parasitica effectively prevent heterokaryosis, while the present study has shown that none of the five loci totally prevent viral transmission, and two of the loci, vic4 and vie5, do not appear to provide any hindrance to transmission. Therefore, while particular vegetative incompatibility polymorphisms have been found to prevent horizontal viral transmission, no single vie locus has been found to prevent transmission irrespective of genetic background. These results suggest that the effects of vegetative incompatibility genes are more important for maintaining the nuclear integrity of the individual (genet). Considerations of the evolutionary origin and ecological impact of vegetative incompatibility must take into account these highly variable effects. The variety of biological effects associated with vegetative incompatibility genes has made discernment of a primary function difficult and well illustrates the inherent problems in identifying adaptations (Gould and Lewontin 1979). However, an important distinction can be made between primary function and associated effect (sensu Williams 1966). If 173 molecular studies show that vegetative incompatibility genes do have a primary or normal function in some process other than protecting the individual from cytoplasmic invasion, self protection may still be of great ecological and evolutionary importance. Indwd, the abundance of cytoplasmic genetic elements in fungi, and their restriction to intracellular transmission routes, implicates vegetative incompatibility systems as important determinants of the coevolution of these host-parasite relationships. 174 Literature Cited Anagnostakis, S.L. 1981. Stability of double-stranded RNA components of Endothia parasitica through transfer and subculture. Exp. Mycol. 5: 236-242. Anagnostakis, S.L. 1982. Genetic analyses of Endothia parasitica: linkage data for four single genes and three vegetative compatibility types. Genetics 102: 25-28. Anagnostakis, S.L. 1983. Conversion to curative morphology in Endothia parasitica and restriction by vegetative incompatibility. Mycologia 75: 777-780. Anagnostakis, S.L. 1987. Chestnut blight: the classical problem of an introduced pathogen. Mycologia 79: 23-37. Anagnostakis, S.L. 1988. Cophoneenia parasitica, cause of chestnut blight, pp. 123-136 in Advances in plant pathology, edited by G.S. Sidhu. Academic Press, Ltd., London. Anagnostakis, S.L. and P.R. Day. 1979. Hypovirulence conversion in Endothia parasitica. Phytopathology 69: 1226—1229. Anagnostakis, S.L. and P.E. Waggoner. 1981. Hypovirulence, vegetative incompatibility, and the growth of cankers of chestnut blight. Phytopathology 71: 1198-1202. Arganoza, M.T., J. Min, 2. Hu, and R.A. Akins. 1994a. Distribution of seven homology groups of mitochondrial plasmids in Neurospora: evidence for widespread mobility between species in nature. Curr. Genet. 26: 62-73. Arganoza, M.T., J. Ohmberger, J. Min, R.A. Akins. 1994b. Suppressor mutants of Neurospora crassa that tolerate allelic differences at single or at multiple heterokaryon incompatibility loci. Genetics 137:731-742. Begueret, 1., B. Turcq, and C. Clave. 1994. Vegetative incompatibility in filamentous fungi: het genes begin to talk. Trends Genet. 10: 441-446. Bernet, J . 1992a. In Podospora anserina, protoplasmic incompatibility genes are involved in cell death control via multiple gene interactions. Heredity 68: 79-87. Bernet, J. 1992b. A gene suppressing the allelic incompatibility specified by genes at five different loci in Podospora anserina. J. Gen. Microbiol. 138: 2567-2574. Brasier, C.M. 1983. A cytoplasmically transmitted disease of Ceratoeystis ulmi. Nature 305: 220. Brasier, C.M. 1984. Intermycelial recognition systems in Ceratoeystis ulmi: their 175 physiological properties and ecological importance, pp. 451-498 in The Ecology and Physiology of the Fungal Mycelium, edited by D.H. Jennings and A.D.M. Rayner. Cambridge University Press, Cambridge. Brasier, C.M. 1990. The unexpected element: mycovirus involvement in the outcome of two recent pandemics, Dutch elm disease and chestnut blight, pp. 289-307 in Pests, Pathogens and Plant Communities, edited by U. Burdon and S.R. Leather. Blackwell Scientific Publications, Oxford. Buck, K.W. 1986. Fungal virology--an overview, pp.2-84 in Fungal Virology, edited by K.W. Buck. CRC Press, Boca Raton, Florida. Buss, L.W. 1987. The Evolution of Individuality. Princeton University Press, Princeton, New Jersey. pp. 201. Caten, C.E. 1972. Vegetative incompatibility and cytoplasmic infection in fungi. J. Gen. Microbiol. 72: 221-229. Chung, P.-H., P.J. Bedker, and 8.1. Hillman. 1994. Diversity of Cryphonectria parasitica hypovirulence-associated double-stranded RNAs within a chestnut population in New Jersey. Phytopathology 84: 984-990. Clark, J. and OR. Collins. 1973. Directional cytotoxic reactions between incompatible plasmodia of Didymium iridis. Genetics 73: 247-257. Coenen, A., F. Debets, R. Hoekstra. 1994. Additive action of partial heterokaryon incompatibility (partial-het) genes in Aspergillus nidulans. Curr. Genet. 26:233-237. Collins, R.A. and B.J. Saville. 1990. Independent transfer of mitochondrial chromosomes and plasmids during unstable vegetative fusion in Neurospora. Nature 345: 177-179. Dales, R.B.G., J. Moorhouse, J.H. Croft. 1983. The location and analysis of two heterokaryon-incompatibility (het) loci in strains of Aspergillus nidulans. J. Gen. Microbiol. 129: 3637-3642. Debets, F., X. Yang, and A.J.F. Griffiths. 1994. Vegetative incompatibility in Neurospora: its effect on horizontal transfer of mitochondrial plasmids and senescence in natural populations. Curr. Genet. 26:113-119. Dobinson, K.F., R.E. Harris, J.E. Hamer. 1993. Grasshopper, a long terminal repeat (LTR) retroelement in the phytopathogenic fungus Magnaporthe grisea. Mol. Plant- Microbe Interact. 6: 114-126. Double, M.L. 1982. The ability of hypovirulent isolates and mixtures of hypovirulent 176 isolates to control artificially established virulent cankers, pp. 145-152 in Proceedings of the USDA Forest Service American Chestnut Cooperators Meeting, edited by H.C. Smith and W.L. MacDonald. West Virginia University Press, Morgantown, WV. Elliston, J.E. 1985. Characteristics of dsRNA-free and dsRNA-containing strains of Endothia parasitica in relation to hypovirulence. Phytopathology 75:170-173. Enebak, S.A., B.I. Hillman, W.L. MacDonald. 1994. A hypovirulent isolate of Cryphonectria parasitica with multiple, genetically unique dsRNA segments. Mol. Plant- Microbe Interact. 7: 590-595. Fulbright, D.W. 1985. A cytoplasmic hypovirulent strain of Endothia parasitica without double-stranded RNA (dsRNA). Phytopathology 75: 1328 (Abst). Fulbright,iD.W., H.W. Weidlich, K.Z. Haufler, c.s. Thomas, and C.P. Paul. 1983. Chestnut blight and recovering American chestnut trees in Michigan. Can. J. Bot. 61: 3164-3171. Fulbright, D.W., C.P. Paul, and S.W. Garrod. 1988. Hypovirulence: a natural control of chestnut blight, pp. 121-139 in Biocontrol of Plant Diseases Vol. 11, edited by K.G. Mukerji and K.L. Garg. CRC Press, Inc., Boca Raton, Florida. Gamjobst, L. and J.F. Wilson. 1956. Heterokaryosis and protoplasmic incompatibility in Neurospora crassa. Proc. Natl. Acad. Sci. USA 42: 613—618. Glass, N .L. and A.G. Kuldau. 1992. Mating-type and vegetative incompatibility in filamentous ascomycetes. Annu. Rev. Phytopathol. 30: 201-224. Gould, SJ. and R.C. Lewontin. 1979. The spandrels of San Marco and the Panglossian paradigm: A critique of the adaptationist programme. Proc. R. Soc. London B 205: 581- 598. Griffin, G.J. 1986. Chestnut blight and its control. Hort. Rev. 8: 291-335. Griffiths, A.J.F. 1995. Natural plasmids of filamentous fungi. Microbiol. Rev. 59: 673- 685. Griffiths, A.J.F., S.R. Kraus, R. Barton, D.A. Court, and H. Bertrand. 1990. Heterokaryotic transmission of senescence plasmid DNA in Neurospora. Curr. Genet. 17:139-145. Handley, L. and C.E. Caten. 1973. Vegetative death: a mitochondrial mutation in Aspergillus amstelodami. Heredity 31: 136 (Abst). 177 Hartl, D.R., E.R. Dempster, and S.W. Brown. 1975. Adaptive significance of vegetative incompatibility in Neurospora crassa. Genetics 81: 553-559. Heiniger, U. and D. Rigling. 1994. Biologieal control of chestnut blight in Europe. Annu. Rev. Phytopathol. 32: 581-599. Hillman, B.I., D.W. Fulbright, D.L. Nuss, N.K. Van Alfen. 1995. Hypoviridae, pp. 261- 264 in Virus Taxonomy, edited by F.A. Murphy, C.M. Fauquet, D.H.L. Bishop, S.A. Ghabrial, A.W. Jarvis, G.P. Martelli, M.P. Mayo, and MD. Summers. Springer-Verlag, New York. Huber, D.H., D.W. Fulbright, H. Bertrand, C.M. Vitorello, J. Bell, and B. Shaw. 1994. Seneseence-like phenotypes in dsRNA-free hypovirulent strains of Cryphonectria parasitica. Phytopathology 84: 1063 (Abst). Kinsey, J .A. 1990. Tad, a LINE-like transposable element of Neurospora, can transpose between nuclei in heterokaryons. Genetics 126: 317-323. Kistler, H.C. and V.P.W. Miao. 1992. New modes of genetic change in filamentous fungi. Annu. Rev. Phytopathol. 30: 131-152. Kuhlman, E.G. and H. Bhattacharyya. 1984. Vegetative compatibility and hypovirulence conversion among naturally occurring isolates of Cryphonectria parasitica. Phytopathology 74: 659-664. Kuhlman, E.G., H. Bhattacharyya, B.L. Nash, M.L. Double, W.L. MacDonald. 1984. Identifying hypovirulent isolates of Cryphonectria parasitica with broad conversion capacity. Phytopathology 74: 676-682. Labarere, J ., J. Begueret, and J. Bernet. 1974. Incompatibility in Podospora anserina: comparative properties of the antagonistic cytoplasmic factors of a nonallelic system. J. Bacteriol. 120: 854-860. lee, J.K., T.A. Tattar, P.M. Berman, and M.S. Mount. 1992. A rapid method for testing the virulence of Cryphonectria parasitica using excised bark and wood of American chestnut. Phytopathology 82: 1454-1456. Leslie, J .F. 1993. Fungal vegetative compatibility. Annu. Rev. Phytopathol. 31: 127-150. Liu, Y.-C. and M.G. Milgroom. 1996. Correlation between hypovirus transmission and the number of vegetative incompatibility (vie) genes different among isolates from a natural population of Cryphonectria parasitica. Phytopathology 86: 79-86. MacDonald, W.L. and D.W. Fulbright. 1991. Biological control of chestnut blight: use 178 and limitation of transmissible hypovirulence. Plant Disease 75: 656-661. Mahanti, N., H. Bertrand, C.B. Monteiro-Vitorello, and D.W. Fulbright. 1993. Elevated mitochondrial alternative oxidase activity in dsRNA-free, hypovirulent isolates of Cryphonectria parasitica. Physio]. Mol. Plant Pathol. 42: 455-463. Martin, R.M. and N.K. Van Alfen. 1991. The movement of viral-like RNA between colonies of Cryphonectria parasitica. Mol. Plant-Microbe Interact. 4: 507-511. May, R.M. and R.M. Anderson. 1983. Parasite-host coevolution, pp. 186-206 in Coevolution, edited by DJ. Futuyma and M. Slatkin. Sinauer, Sunderland, Mass. Morris, T.J., and J .A. Dodds. 1979. Isolation and analysis of double-stranded RNA from virus-infected plant and fungal tissue. Phytopathology 69:854-858. Nauta, M.J. and R.F. Hoekstra. 1994. Evolution of vegetative incompatibility in filamentous ascomycetes. I. deterministic models. Evolution 48: 979-995. Newhouse, LR. and W.L. MacDonald. 1991. The ultrastructure of hyphal anastomoses between vegetatively compatible and incompatible virulent and hypovirulent isolates of Cryphonectria parasitica. Can. J. Bot. 69: 602-614. Nuss, D.L. 1992. Biological control of chestnut blight: an example of virus-mediated attenuation of fungal pathogenesis. Microbiol. Rev. 56: 561-576. Nuss, D.L. and Y. Koltin. 1990. Significance of dsRNA genetic elements in plant pathogenic fungi. Annu. Rev. Phytopathol. 28: 37-58. Polashock, LL and BI. Hillman. 1994. A small mitochondrial double-stranded (ds) RNA element associated with a hypovirulent strain of the chestnut blight fungus and ancestrally related to yeast cytoplasmic T and W dsRNAs. Proc. Natl. Acad. Sci. USA 91:8680-8684. Rayner, A.D.M. 1991. The phytopathological significance of mycelial individualism. Annu. Rev. Phytopathol. 29: 305-323. Rayner, A.D.M. and D. Coates. 1987. Regulation of mycelial organisation and responses, pp. 115-136 in Evolutionary Biology of the Fungi, edited by A.D.M. Rayner, C.M. Brasier, and D. Moore. Cambridge University Press, Cambridge. Rayner, A.D.M., D. Coates, A.M. Ainsworth, T.J.H. Adams, E.N.D. Williams and N .K. Todd. 1984. The biological consequences of the individualistic mycelium, pp. 509- 540 in The Ecology and Physiology of the Fungal Mycelium, edited by D.H. Jennings and A.D.M. Rayner. Cambridge University Press, Cambridge. 179 Smart, CD. and D.W. Fulbright. 1996. Molecular biology of fungal diseases, pp. 57-70 in Molecular Biology of the Biological Control of Pests and Diseases of Plants, edited by M. Gunasekaran and DJ. Weber. CRC Press, Inc., Boca Raton, Florida. Van Alfen, N.K., R.A. Jaynes, S.L. Anagnostakis, and P.R. Day. 1975. Chestnut blight: biological control by transmissible hypovirulence in Endothia parasitica. Science 189: 890-891. Williams, G.C. 1966. Adaptation and Natural Selection. Princeton University Press, Princeton, New Jersey. Yang, X., and A.J.F. Griffiths. 1993. Plasmid diversity in senescent and nonsenescent strains of Neurospora. Mol. Gen. Genet. 237:177-186. Chapter 5 The Transmission and Phenotypic Effects of a Seneseence Syndrome in Cryphonectria parasitica Abstract A severely debilitating senescence phenotype has been identified in Cryphonectria parasitica strain KFC9 collected in southwestern Michigan. The syndrome includes several of the phenotypic characteristics of senescence found in other filamentous Ascomycetes. Mycelium shows a progressive loss of growth potential such that subcultures taken from increasingly distal regions of a colony show increasingly debilitated growth and morphology. The senescence agent is horizontally (cytoplasmically) transmissible to other mycelia, and causes a rapid degeneration in the recipient mycelium. The effects of vie loci upon transmission were found to be similar to their effects upon hypovirus transmission. Transmission occurs between strains homoallelic at all vie loci and between strains heteroallelic at only vie4. Loci vie2 and vie] (only vie] -2 was tested as recipient) prevent transmission; vie3 and vie5 were not tested. Senescing mycelia exhibited a pronounced decline in virulence (aggressiveness). Conidia from senescing mycelia exhibited varying degrees of senescence ranging from normal growth to death soon after germination. Serrescence was characterized by elevated levels of respiration via the alternative oxidase. Collectively, these characteristics are indicative of mitochondrial dysfunction and are similar to the suppressiveness quality of the Neurospora senescence syndrome. Diminished aggressiveness in the chestnut blight fungus (Cryphonectria parasitica), termed hypovirulence, has been found to be caused by cytoplasmically transmissible genetic elements. The best characterized of these elements are double-stranded RNA viruses (hypoviruses) that have been collected from strains of C. parasitica from healing chestnut eankers in numerous loeations in Europe and eastern North America (MacDonald and Fulbright, 1991; Nuss, 1992; Heiniger and Rigling, 1994). A second type of hypovirulence phenotype has also been discovered in fungal strains from healing chestnut eankers in Michigan that are not associated with dsRNA (Fulbright, 1985; Mahanti et al., 180 181 1993; Huber et al., 1994). These preliminary studies of the dsRNA-free type of hypovirulence have shown that it is comparable, phenotypically, to senescence in Neurospora, Podospora, and Aspergillus (reviewed in Griffiths, 1992). The eausal agent of this new, senescence-like form of hypovirulence is not yet known. Physiological characterization of several dsRNA-free hypovirulent strains by Mahanti et al. (1993) found that the hypovirulence was associated with elevated levels of cyanide- resistant respiration, which is an indication of mitochondrial energy metabolism dysfunction. Extensive work in Neurospora has shown that its senescence syndrome includes the induction of high levels of cyanide-resistant respiration through an alternative oxidase following from any of several different types of mitochodrial DNA mutations (reviewed in Griffiths, 1992). To test whether mitochondrial DNA mutations are capable of reducing virulence and eliciting senescence in C. parasitica, Monteiro-Vitorello et al. (1995) induced mutations in the mitochondrial chromosome. They found that induced mitochondrial mutations could reduce virulence, thereby providing additional evidence that the new senescence-like form of hypovirulence could have a mitochondrial etiology. However, the nature of the naturally occurring type of senescence in C. parasitica is still not known, nor is its potential for cytoplasmic transmission through vegetative incompatibility barriers understood. Hypovinrlerrt strains of C. parasitica which lack detectable levels of dsRN A have been found in healing cankers of American chestnut trees in the Kellogg Forest near the Michigan State University Kellogg Biological Station (Fulbright, 1985; Mahanti et al. , 1993). This study presents a preliminary characterization of another dsRNA-free strain 182 from this site that exhibited severe senescence-like characteristics. The hypovirulence phenotype of this mutant strain has been found to be associated with elevated levels of the alternative oxidase, is cytoplasmically transmissible, and produces a progressive degeneration of the mycelium that culminates in death. These characteristics are suggestive of the suppressive mitochondrial mutations of Neurospora although the molecular basis of this syndrome is not yet known. However, the nature of this mutant phenotype indieates that it can be termed senescent in keeping with the use of this term in other fungi (Griffiths, 1992). A preliminary report of this work has been published (Huber et al., 1994). MATERIALS AND METHODS C. parasitica grains and culture conditiom: The strains of Cryphonectria parasitica used in this study and their sources are listed in Table 1. Strain KFC9 was collected from healing cankers on American chestnut trees from the Kellogg Forest, Michigan State University. All of the strains used in this study were grown on potato dextrose agar (PDA; Difco Laboratories) at room temperature (about 25° C) under cool white flourescent lights (34 watt). Endothia complete medium (Puhalla and Anagnostakis, 1971) was used to grow mycelia prior to the respiration assay. Conidia were collected from strains infected with the senescence agent by scraping a small amount of conidiating mycelium from the surface of a PDA plate and placing it in sterile water. After a serial dilution, the conidia were spread onto PDA plates. When the conidia had germinated (at about 24 hours) they were individually cut out of the PDA 183 Table 1. Strains used in this study. Strain Vegetative incompatibility Seneseence agent Sourceb seaotype' KFC9 unknown present Kellogg Forest, MI 389.7 vicI-I ,2—1 ,3-1 ,4-I,5-I none this study (chap. 2) EPB88 vieI-2,2-I ,3-1 ,4-1 ,5-1 none ATCC 38979 22508 vieI-2,2-2,3-I ,4-1 ,5-1 none ATCC 22508 A1.18 none 389.7 X 22508 EP289 vieI-I,2-2,3-I ,4—1,5-I none Conn. Agri. Exp. Sta.c EP289(KFC9) KFC9 source infection of EP289 A1.13 none 389.7 X 22508 C13 none 389.7 X EPZS9 C1.20 none 389.7 X EP289 Cl.20(KFC9) KFC9 source infection of €1.20 C2.8 none 389.7 X EP289 C2. 10 none 389.7 X EP289 F3.2 vieI-1,2-1,3-1,4-2,5-I none 389.7 X 80-2c F3.13 vieI-2,2-1,3-I,4-2,5-I none 389.7 X 80-2c 12.31 vieI-1,2-2,3-1,4-2,5-1 none F3.16 X A1.13 1231(KFC9) KFC9 source infection of J23] EPISS unknown none ATCC 38755 EP2001 unknown none ATCC 605 89 ‘ For brevity, vegetative incompatibility genotypes have been written as vie] -1 ,2—1 rather than as vieI-I, vie2-I. " The sexual crosses from which certain strains were derived are described in chapter 3 ° Strain provided by S. L. Anagnostakis, Connecticut Agricultural Experiment Station. 184 under a dissecting scope (40X) and placed onto new PDA plates. Strain KFC9 was tested for the presence of double stranded RNA (dsRNA) using the protocol of Morris and Dodds (1979) as modified by Fulbright et al. (1983). Transmission assay: Tests for the horizontal (cytoplasmic) transmission of the senescence-inducing agent were conducted by placing a 5 mm3 subculture of a senescing donor strain on PDA near one edge of 100 X 15 mm Petri plate. The prospective recipient strain was placed on the PDA about 1 cm from the donor. The two cultures made contact with each other as they grew across the Petri plate. Because the senescent strains grew much less vigorously than the wild type strains, the senescent donor strain was allowed to grow for five to seven days before the recipient strain was added to the medium so that it could attain a diameter of about 2 cm. Care was taken to observe continued mycelial growth in the senescent culture following contact between the two mycelia in order to be certain that the senescent strain was still viable. The infection of the recipient strain with the senescence agent was easily observed as a severe change in morphology of the recipient. Virulence assays: The virulence assay used follows that described by Lee et al. (1992). Chestnut stems about 2 cm in diameter were cut into sections about 5 cm long. Each section was bisected longitudinally and the outer layer of bark was peeled away from the underlying secondary xylem. A mycelial plug (5 mm’) cut from cultures grown on PDA was placed on the inside center of each bark tissue piece with the mycelial side down. The cultures were then incubated in petri dishes on top of damp Whatman filter paper, sealed by Parafilm, and stored in the dark for five days at 25° C. Only the outer 185 tissue layer consisting of living bark/cortex/phloem was used for the virulence assays. The underlying’secondary xylem was not used for the virulence assays as described by Lee et al. (1992) because the fungus did not grow well on this tissue. The mycelial plugs taken from PDA cultures to inoculate the chestnut tissue were taken a few millimeters behind the leading edge of cultures that had been growing for one to two weeks. This zone consisted of mycelium which showed a more advanced development of senescence compared to the center of the colony. Alternative oxidase respiration assay: The procedure used for the alternative oxidase assay follows that of Mahanti et al. (1993) and Monteiro-Vitorello et al. (1995). These procedures were adapted from Lambowitz and Slayman (1971). Strains which were tested for the presence of alternative oxidase were first grown on PDA. Small plugs (5 mm’) of PDA with mycelium from near the actively growing margin of a colony were placed into a stationary culture of 50 ml of Endothia complete broth (Puhalla and Anagnostakis, 1971) in a 125 ml Erhlenmeyer flask for two or more days. The number of mycelial plugs added to a flask and the number of days given for growth depended upon the rate of growth of the particular strain. Nonsenescent strains were grown in stationary culture for two days using three or four mycelial plugs per flask. Senescent strains were grown in stationary culture for five or six days using five or six mycelial plugs per flask. After this time period, both senescent and nonsenescent cultures had roughly comparable amounts of growth (measurements of the mass of the cultures were not taken). The liquid cultures were then shaken at 200 rpm overnight. The mycelium was homogenized with a Biospec Products, Inc. Tissue Tearer homogenizer and placed into 6 ml of Vogel's medium 186 (Vogel, 1956). Three ml of the Vogel's/mycelial slurry were added to a chamber of YSI 5300 biological oxygen monitor. Glucose (200 pl of 20%) was also added to the chamber. The mycelial slurry was aerated for 20-30 seconds with an aquarium pump to saturate the solution. Oxygen consumption was measured by inserting a Clark electrode into the reaction chamber maintained at 25° C and mixed continuously with a magnetic stirring bar. The contributions to respiration provided by the cytochrome chain and the alternative oxidase pathways were measured using inhibitors of each of the two pathways. Potassium cyanide (KCN) was used as the inhibitor of the cytochrome chain, and salicylhydroxamic acid (SHAM) was used as an inhibitor of the alternative oxidase pathway. KCN was added to the mycelial slurry in the respiration chamber to a final concentration of 1.0 mM fiom a 0.1 M solution in IOmM Tris-HCl, 5 mM EDTA, pH 7.2. SHAM was added to the reaction chamber to a final concentration of 4.16 mM from a 50 mg ml‘1 solution in 95 % ethanol. The inhibitors were added in succession to the reaction chambers. The order of the inhibitors was reversed for each of the 3 ml aliquots of the Vogel's solution/mycelial slurry. The second inhibitor was not added until the oxygen consumption became approximately linear following the addition of the first inhibitor. RESULTS The senescence phenotype. Strain KFC9 showed a progressive degeneration of the mycelium as it grew on PDA (Figure 1). A subculture of the mycelium taken from the center of a senescing culture produced normal appearing mycelium as it began to grow. 187 Figure 1. Growth phenotypes of (A) strain EP289 and (B) strain EP289(KFC9). Strain EP289(KFC9) was infected with the senescence agent from strain KFC9 and exhibits the senescence growth form. 188 Then as the mycelium advanced it lost aerial growth and became progressively thinner. Before reaching the edge of the Petri dish, the mycelium became very thin and only grew within, rather than on the surface of, the medium. Growth ceased before the edge of the petri dish was reached. Those senescent strains with aerial mycelium at the center of the colony produced exceptional amounts of conidia, visible as globules on top of the mycelium at the center of the colony, but no conidia in the region where growth became very thin and ceased. Subcultures from a senescent mycelium grown on PDA were found to resume growth according to the stage of senescence present in the sampled region of mycelium. That is, subcultures taken from the center of the senescent culture where the mycelium appeared normal exhibited the same pattern of growth and degeneracy as the parental culture, while subcultures taken progressively closer to the growth front showed increasingly advanced stages of senescence. Mycelium subcultured from the leading edge of the colony exhibited extremely poor growth often requiring a microscope to observe. Conidia collected from KFC9 showed different degrees of senescence. The type of growth observed from conidia ranged from normal mycelium to thin, appressed growth to only germination. The range of degeneracy exhibited by the conidia was similar to the levels of senescence shown by subcultures taken from senescent mycelia as described above. For example, some conidia produced a small amount of growth (anywhere from 3cm or less maximum colony diameter) that never included normal mycelial growth. Conidia collected from a strain converted by KFC9, EP289(KFC9), also showed the presence of senescence in some of the resulting colonies, thereby demonstrating that the effect of the senescence agent upon conidia is also transmitted between strains. 189 Strain EP289(KFC9), which exhibits pronounced senescence, was subjected to dsRNA extraction using established protocols. No dsRNA was found in EP289(KFC9). Serial transmission of the senescence phenotype. Transmission of the senescence phenotype between donor and recipient mycelia resembled the transmission of the hypovirulent phenotype eaused by hypovirus infection in C. parasitica. The leading edge of the recipient mycelium became very debilitated with appressed, thin mycelium morphologically similar to the degenerative phenotype of the leading edge of the donor mycelium. This degeneration then spread laterally through the leading edge of the recipient mycelium in the same manner as hypovirus infection. Transmission of the senescence phenotype was accomplished serially thereby demonstrating that infected recipients can also act as donors. The serial transmission of the senescence agent was conducted in this order: KFC9 to EP289 to A1.13 to J2.31 (Table 2). EP289 was also used as donor with several other strains (see below). The effects of vegetative incompatibility loci upon transnksion of senescence. The cytoplasmic transmissibility of the senescence phenotype was tested by growing strain KFC9 on PDA plates along with potential recipient strains so that their hyphae would have the opportunity to anastomose. KFC9 was paired with six tester strains which have known vegetative compatibility genotypes relative to each other (389.7, EP388, EP289, 22508, F3.2, and F3.13) and to two strains with unknown vc genotypes (EP155 and EP2001) (Table 2). Strains EPISS and EP2001 are unable to transmit or to receive dsRNA from these four tester strains. These eight genotypes were chosen because they represent a broad range of vegetative incompatibility genotypes thereby increasing the chances that one 190 Table 2. Transmission of the senescence-inducing agent between donor and recipient strains that differ at several vegetative incompatibility (vie) loci. Donor Recipient Transmission of senesence vie‘12345 strain vie12345 strain agent unknown KFC9 I I I I 1 389.7 - unknown KFC9 2 1 I I I EP388 - unknown KFC9 I 2 I I I EP289 + unknown KFC9 2 2 I I I 22508 - unknown KFC9 I I I 2 1 F3.2 - unknown KFC9 2 I I 2 I F3.13 - unknown KFC9 unknown EP155 - unknown KFC9 unknown EP2001 - 12111 EP289 1211] C13 + 12111 EP289 1211] C1.20 + 12111 EP289 1211] C23 + 12111 EP289 12111 C2.10 + 12111 EP289 12111 A1.13 + 12111 EP289 22111 Al.l8 - 12111 A1.13 12121 12.31 + ' For brevity, vie loci are designated only by their number as a column heading, rather than as vie], vie2, etc. The alleles for each locus that are present in a particular strain are listed in the column beneath the appropriate vie locus. Only two alleles, designated as I or 2, are known for each vie locus. 191 would be a competent recipient of the senescence agent in KFC9. Transmission of the senescence phenotype only occurred into strain EP289. Strains 389.7, EP388, EP289 and 22508 represent all the combinations of the alleles at vie] and vie2. Therefore, although the vegetative incompatibility genotype of KFC9 is not known, this transmission pattern suggests that KFC9 carries the alleles vie] -2 and vic2-2 since hypovirus transmission would be limited in this same manner if the unknown genotype was so constituted. However, since KFC9 is vegetatively incompatible with EP289 these strains may only differ at weak incompatibility loci, or alternatively epistatic effects or unidirectional loci could also be involved. Transmission of the senescence agent occurred from donor strain EP289(KFC9) into five other recipient strains which are homoallelic at all vie loci (Table 2). The senescence agent freely moved between vegetatively compatible strains in contrast to the lack of transmission associated with the previous incompatibility tests. The effect of vie4 upon transmission was tested by pairing the donor Al.13(KFC9) with recipient strain J231. This test demonstrated that transmission of the senescence agent is not noticeably inhibited by vegetative incompatibility caused by vie4. These limited tests indicated that transmission of the senescence agent is affected by vie genes similarly to hypoviruses. Respiration, senescence and reduced virulence. To determine whether the agent causing the transmissible senescence in C. parasitica might be causing mitochondrial dysfunction such as occurs in Neurospora, respiration was assayed. Mitochondrial dysfunction in fungi, such as cytochrome deficient mutants of Neurospora, has been found to induce the activity of an alternative terminal oxidase (Lambowitz and Zannoni, 1978). 192 The determination of the presence of the two pathways and the contribution of each to total respiratory activity can be accomplished by using inhibitors specific for each pathway. Electron transport through the cytochrome electron transport system can be inhibited at the cytochrome oxidase step by KCN which blocks the terminal electron acceptor. The pathway to the alternative oxidase can be blocked by SHAM. Therefore, mycelia with normal levels of cyanide-sensitive respiration will show a dramatic reduction in respiration when exposed to KCN, but not to SHAM. Conversely, respiration due primarily to the cyanide-insensitive alternative oxidase will be little affected by KCN, but greatly affected by SHAM. The percent of total respiration attributable to the alternative oxidase in wildtype strains EP289, C130, and 12.31 was in the range of about 10 to 25% (Table 3). This is consistent with previous work which took baseline readings of alternative oxidase in nonsenescent C. parasitica (Mahanti et al., 1993; Monteiro-Vitorello et al., 1995). The field-collected senescent strain KFC9 was found to have 78% of total respiration due to the alternative oxidase. Each of the three strains infected with the senescence agent showed significantly elevated levels of cyanide-insensitive respiration relative to the isogenic uninfected progenitors (Table 3). EP289 showed 23 % of respiration that was cyanide insensitive, whereas EPZS9(KFC9) showed 60% of its respiration due to the cyanide-insensitive alternative oxidase. Strain C1.20 showed a larger change in respiration: C1.20 had 14% of total respiration attributable to the alternative oxidase, while C1.20(KFC9) had 61% so attributable. Virulence of the wildtype and senescent strains was measured as growth on live 193 Table 3. Alternative oxidase as percent of total respiration, virulence, and senescence phenotypes of strains that have been cytoplasmically infected with the senescence inducing agent from strain KFC9. Strains Phenotype Virulence Alt. Oxd. as % (cm’) of total resp. KFC9 8 ND 78 i 10.4 EP289 N 4.17 ;|: 0.16 23 :1: 3.1 EP289(KFC9) S 0.23 :1; 0.08 60 :t; 7.0 C1.20 N 7.01 :1; 0.13 14* C1.20(KFC9) S 1.29 i 0.49 61* 1231 N 6.36 :1: 0.42 12* J2.31(KFC9) S 0.29 :1; 0.15 ND S, senescent growth; N, normal growth. ND, no data. Values are means 1 SE. *, percent alternative oxidase based upon one sample. Virulence was determined by growth on live chestnut bark as described in the materials and methods. 194 chestnut tissue. The virulence of each of the strains was dramatically reduced when they were infected with the senescence agent (Table 3). In fact, the growth of EP289(KFC9) and J231(KFC9) was extremely debilitated and may not have continued if the inoculations on the live chestnut tissue were observed for a longer time. DISCUSSION Filamentous fungi capable of producing anastomoses with genotypically different mycelia are inherently vulnerable to infection by cytoplasmically borne genetic agents. Infectious cytoplasmic agents have been found in several plant pathogenic fungi that reduce their aggressiveness (Buck, 1986), in particular, hypoviruses in C. parasitica (Van Alfen et al. 1975, MacDonald and Fulbright, 1991) and d-factor in Ophiostoma ulmi (Brasier 1983). Recently, hypovirulent strains of C. parasitica have been found that lack detectable levels of dsRNA viruses (Fulbright, 1985; Mahanti et al., 1993; Huber et al., 1994). This study has shown that one type of nonviral hypovirulence occurring in strains from nature eauses infectiously transmissible senescence. The senescence phenotype was found to be transmissible through hyphal fusions, includes a respiratory deficiency characterized by high levels of respiration via an alternative oxidase, and exhibits a progressive degeneration during vegetative growth. These characteristics are similar to the transmissible senescence syndrome in Neurospora and implicate mitochondrial dysfunction in C. parasitica senescence. In Neurospora, senescence has been found to result from energy metabolism deficiencies that are due to mitochondrial dysfunction. Mitochondrial dysfunction can 195 occur as the result of several naturally occurring mutations in mitochondrial DNA that inhibit mitochondrial protein synthesis, including deletions (Bertrand et al. , 1980), point mutations (Mannella and Lambowitz, 1978), and the integration of plasmids into the mitochondrial chromosome (Bertrand et al., 1986; Akins et al. , 1986). The disruption of mitochondrial protein synthesis impairs the function of the cytochrome-mediated electron transport chain which thereby reduces oxidative phosphorylation, and consequently growth. One of the diagnostic physiological symptoms of senescence in Neurospora is the induction of high levels of respiration via an alternative oxidase rather than cytochrome oxidase (Griffiths, 1992; Bertrand, 1983; Lambowitz and Zannoni, 1978). Cellular respiration in some fungi and plants has been found to consist of the conventional cytochrome-mediated electron transport pathway plus an alternate branch (beginning at ubiquinone) where electrons can be transferred to an alternative terminal oxidase. The induction of high levels of respiration through the alternative oxidase in Neurospora occurs when the cytochrome system is not functioning properly which can occur due to the disruption .of mitochondrial protein synthesis by naturally occurring mutations or by the use of chemical inhibitors. Under normal physiological conditions, less than 10% of the total respiration in Neurospora is contributed by the alternative oxidase. However, in senescent mutants the alternative oxidase can contribute the majority of respiration. An outstanding characteristic of transmissible senescence in C. parasitica is the rapidity of degeneration of the recipient mycelium, recalling the suppressiveness quality of the Neurospora senescence syndrome. Suppressiveness in Neurospora refers to the 196 replacement of normal mitochondria by mutant mitochondria resulting in degeneration of the mycelium. Several models have been suggested to explain this unusual phenomenon in the filamentous fungi (reviewed in Bertrand, 1995). One model suggests that mutant mitochondrial chromosomes are able to replicate more efficiently than wild type chromosomes due to their smaller size and more efficient origins of replication (Blanc and Dujon, 1980; Almasan and Mishra, 1988). However, this model does not adequately account for the induction of suppressiveness by point mutations and insertions (Bertrand et al. 1986). Recently, an alternative model has been presented by Bertrand (1995) that more cogently accounts for the unusual characteristics of suppressiveness. Bertrand notes that suppressiveness is associated with those mitochondrial mutations that inhibit electron transport and thereby adversely affect oxidative phosphorylation. These senescing fungal strains have been found to have increased numbers of mutant mitochondria. It also has been shown that the number of mitochondria will increase in a cell as a result of the chemical inhibition of oxidative phosphorylation. Taken together, this evidence suggests that mitochondria with inhibited oxidative phosphorylation are induced to proliferate more rapidly than those with normal oxidative phosphorylation. This response has been named the OXPHOS stress response by Bertrand. The senescence model presented by Bertrand therefore considers the suppressive quality of senescence in Neurospora to be the result of the natural proliferation of mitochondria resulting from induction of the OXPHOS stress response in individual mitochondria. The induction of high levels of alternative oxidase in the senescent strains of C. parasitica suggests that a similar replacement of 197 metabolically normal mitochondria by metabolically abnormal mitochondria may be occurring in this species also. Cytoplasmically transmissible agents capable of causing mitochondrial dysfunction have been found in several filamentous Ascomycetes. The diseased state of 0. ulmi occurs after infection of mitochondria with dsRNA elements (d-factor) and is characterized by the loss of cytochrome aa, (Rogers et al. , 1987). Another diseased state in 0. ulmi has been found that is not associated with dsRNA, but is infectiously transmissible, and causes the generation of plasmids derived from the mitochondrial DNA of the recipient strain (Charter et al. , 1993). In Neurospora, the linear plasmids, kalilo and maranhar, are cytoplasmically transmissible and integrate into the mitochondrial genome, thereby causing the induction of senescence (Bertrand et al., 1986; Griffiths, 1992). The cytoplasmic mutation vgd in Aspergillus amstelodami is also transmissible through hyphal fusions and may affectmitochondrial function although the molecular basis of its effect has not been elucidated (Caten, 1972). Several similar mutants termed ragged (rgd) have also been identified in A. amstelodami that show the generation of head-to—tail concatamers of specific regions of mitochondrial DNA (Lazarus et al., 1980). The horizontal mobility of the senescence phenotype has also been shown in Podospora (Marcou and Schecroun, 1959). Seneseence in Podospora is associated with the appearance of circular DNA molecules termed senDNA that are composed of head-to-tail multimers of specific regions of the mitochondrial chromosome (reviewed in Griffiths, 1992). Cytoplasmic transmission in fungi is not limited to parasitic genetic elements. The transmission of mitochondrial chromosomes has also been shown to occur in C. parasitica 198 and several other fungi. Gobbi et al. (1990) reported that the mitochondrial chromosome of C. parasitica would not transfer between strains through hyphal fusions. However, Mahanti and Fulbright (1995) and Monteiro-Vitorello et al. ( 1995) have shown that the mitochondrial chromosome is capable of lateral transfer under laboratory conditions. Whether such transmission through hyphal fusions is a regular occurrence between compatible strains of C. parasitica in nature is not known. Transmission of the mitochondrial chromosome of Neurospora has been found to occur even during incompatible vegetative cell fusions when strains were placed under nutrient selection for heterokaryotic growth (Collins and Saville, 1990). Again the significance of this observation for populations growing under natural conditions has not been studied. In the Basidiomycetes, recent studies of cytoplasmic mixing following mating have shown that heteroplasmons form in a restricted region of hyphal fusions between the parents under laboratory conditions (Baptista-Ferreira et al. , 1983; Hintz et al., 1988; May and Taylor, 1988; Smith et al., 1990). However, Smith et al. (1990) did not find evidence for heteroplasmy in natural populations of Annillaria even though heteroplasmy occurred in laboratory matings suggesting that natural conditions may be more restrictive of cytoplasmic exchange. Different types of cytoplasmic genetic elements exist in filamentous fungi. Are all of these genetic elements infectiously transmitted between strains with the same efficiency? Viral transmission in C. parasitica has been shown to be highly responsive to specific vie genes (chapter 4). Two vie genes whose effects upon the horizontal transmission of hypoviruses are known were tested for their effects upon the transmission of the 199 senescence agent in C. parasitica. Each of the vie loci examined were found to have the same effect upon senescence transmission as they had upon hypovirus transmission: transmission occurred between strains homoallelic at all vie loci and between strains heteroallelic at only vie4; transmission was prevented by heteroallelism at vie] (recipient vie] -2). The effects of vic3 and vie5 upon transmission, as well as the reciprocality of vie], were not tested. To my knowledge, this is the first time that the efficiency of transmission of viruses has been compared to nonviral cyt0plasmic agents. The presence of these various instances of horizontally transmissible diseases that specifically affect mitochondria indicates that the populations of mitochondria in coenocytic fungal cells are susceptible to infectious agents. Caten (1972) suggested that the purpose of vegetative incompatibility systems was to limit cytoplasmic infection. The original suggestion by Caten (1972) was applied to the vgd cytoplasmic mutation in Aspergillus which is presumed to affect the mitochondria. Caten's work and subsequent studies demonstrated that the transmission of the vgd mutation could be limited by vegetative incompatibility. The efficiency of cytoplasmic transmission of the vgd mutation was found to vary according to whether strains were heteroallelic at heat or hetB (Handley and Caten, 1973), or whether strains were heteroallelic at one or two het genes (Caten, 1972). Recently, the effects of vegetative incompatibility upon the transmission of the senescence-inducing linear plasmids in Neurospora has been examined. Using nonselective conditions, Debets et al. (1994) found that while cytoplasmic transmission of kalilo and two other plasmids occurred between strains of N. crassa that differed at hetD and hetE, transmission was usually prevented by hetC. With the application of forcing conditions, 200 however, the kalilo plasmid has been transmitted between different species of Neurospora (Griffiths et al., 1990). The infectious cytoplasmic spread of these agents through populations in nature has not been studied. The extent of the spread of the senescence agent in C. parasitica populations is also not known. Several American chestnut trees in the vicinity of the KFC9 collection site have healing cankers which may be due to infection with the senescence agent, but this has not been investigated. The cytoplasmically transmissible senescence phenotype in C. parasitica is of particular interest as a potential biocontrol agent. The cytoplasmically infectious hypoviruses have functioned effectively as a biocontrol agent in European C. parasitica populations and loeally in North American populations. It is possible that the senescence agent may also prove to be effective in this manner. Presently, only a few healing cankers can be attributed to this agent so its potential for population level infection is not known. The severity of the disease may be a limitation in its spread. However, the abundance of conidia produced by diseased mycelia under laboratory conditions and their (unproven) potential for sexual inheritance may significantly improve their transmission through populations, and hence, effectiveness as a biocontrol. 201 LITERATURE CITED Akins, R.A., R.L. Kelley, and A.M. Iambowitz. 1986. Mitochondrial plasmids of Neurospora: integration into mitochondrial DNA and evidence for reverse transcription in mitochondria. Cell 47: 505-516. Almasan, A. and N .C. Mishra. 1988. Molecular characterization of the mitochondrial DNA of a new stopper mutant, ER-3, of Neurospora crassa. Genetics 120: 935-945. Baptista-Ferreira, J.L.C., A. Economou and L.A. Casselton. 1983. Mitochondrial genetics of Coprinus: recombination of mitochondrial genomes. Curr. Genet. 7: 405-407. Bertrand, H. 1995. Seneseence is coupled to induction of an oxidative phosphorylation stress response by mitochondrial DNA mutations in Neurospora. Can. J. Bot. 73(Suppl. 1): 8198-8204. Bertrand, H. 1983. Aging and senescence in fungi, pp. 233-251 in Modern aging research, Vol. 38. Intervention in the aging process. Part B. Basic research and preclinical screening, edited by W. Regelson and P.M. Sinex. Alan R. Liss, Inc., New York. Bertrand, H. and A.J.F. Griffiths. 1989. Linear plasmids that integrate into mitochondrial DNA in Neurospora. Genome 31: 155-159. Bertrand, H., A.J.F. Griffiths, D.A. Court, and C.K. Cheng. 1986. An extrachromosomal plasmid is the etiological precursor of kalDN A insertion sequences in the mitochondrial chromosome of senescent Neurospora. Cell 47 : 829-837. Bertrand, H., R.A. Collins, L.L. Stohl, R. Goewert, and A.M. Iambowitz. 1980. Deletion mutants of Neurospora crassa mitochondrial DNA and their relationship to the ”stopper” growth phenotype. Proc. Natl. Acad. Sci. USA 77: 6032-6036. Blane, H. and B. Dujon.1980. Replicator regions of the yeast mitochondrial DNA responsible for suppressiveness. Proc. Natl. Acad. Sci. USA 77: 3942-3946. Brasier, C.M. 1983. A cytoplasmically transmitted disease of Ceratoeystis ulmi. Nature 305: 220. Buck, K.W. 1986. Fungal virology—an overview, pp. 2-84 in Fungal virology, edited by K.W. Buck. CRC Press, Boca Raton, Florida. Caten, C.E. I972. Vegetative incompatibility and cytoplasmic infection in fungi. J. Gen. Microbiol. 72: 221-229. Charter, N.W., K.W. Buck, and C.M. Brasier. 1993. De-novo generation of 202 mitochondrial DNA plasmids following cytoplasmic transmission of a degenerative disease in Ophiostoma novo-ulmi. Curr. Genet. 24: 505-514. Collins, R.A. and B.J. Saville. 1990. Independent transfer of mitochondrial chromosomes and plasmids during unstable vegetative fusion in Neurospora. Nature 345: 177-179. Debets, F., X. Yang, and A.J.F. Griffiths. 1994. Vegetative incompatibility in Neurospora: its effect on horizontal transfer of mitochondrial plasmids and senescence in natural populations. Curr. Genet. 26:113-119. Fulbright, D.W. 1985. A cytoplasmic hypovirulent strain of Endothia parasitica without double-stranded RNA (dsRNA). Phytopathology 75:1328 (Abst.). Fulbright, D.W., C.P. Paul, and S.W. Garrod. 1988. Hypovirulence: a natural control of chestnut blight, pp. 121-139 in Biocontrol of Plant Diseases Vol. H, edited by K.G. Mukerji and K.L. Garg. CRC Press,Inc., Boca Raton, Florida. Fulbright,.D.W., H.W. Weidlich, K.Z. Haufler, C.S. Thomas, and C.P. Paul. 1983. Chestnut blight and recovering American chestnut trees in Michigan. Can. J. Bot. 61: 3164-3171. Gobbi, E., Y. Wang, R.M. Martin, W.A. Powell, and N.K. Van Alfen. 1990. Mitochondrial DNA of Cryphonectria parasitica: Lack of migration between vegetatively compatible strains. Mol. Plant-Microbe Interact. 3:66-71. Griffiths, A.J.F. 1992. Fungal senescence. Annu. Rev. Genet. 26: 351-372. Griffiths, A.J.F., S.R. Kraus, R. Barton, D.A. Court, and H. Bertrand. 1990. Heterokaryotic transmission of senescence plasmid DNA in Neurospora. Curr. Genet. 17:139-145. Handley, L. and C.E. Caten. 1973. Vegetative death: a mitochondrial mutation in Aspergillus amstelodami. Heredity 31:136 (Abst.). Hintz, W.E.A., J.B. Anderson, and P.A. Horgen. 1988. Nuclear migration and mitochondrial inheritance in the mushroom Agarieus bitorquis. Genetics 119: 35-41. Huber, D.H., D.W. Fulbright, H. Bertrand, C.M. Vitorello, J. Bell, and B. Shaw. 1994. Seneseence-like phenotypes in dsRNA-free hypovirulent strains of Cryphonectria parasitica. Phytopathology 84:1063 (Abst.). Lambowitz, A.M. and CW. Slayman. 1971. Cyanide-resistant respiration in Neurospora crassa. J. Bacteriol. 108: 1087-1096. 203 lambowitz, A.M. and D. Zannoni. 1978. Cyanide-insensitive respiration in Neurospora. Genetic and biophysical approaches, pp. 283-291 in Plant Mitochondria, edited by G.Ducet and C. lance. Elsevier/North-Holland Biomedical Press, Amsterdam. Lazarus, C.M., A.J. Earl, G. Turner, and H. Kuntzel. 1980. Amplification of a mitochondrial DNA sequence in the cytoplasmically inherited ‘ragged’ mutant of Aspergillus amstelodami. Eur. J. Biochem. 106: 633-641. Lee, J.K., T.A. Tattar, P.M. Berman, and M.S. Mount. 1992. A rapid method for testing the virulence of Cryphonectria parasitica using excised bark and wood of American chestnut. Phytopathology 82: 1454-1456. MacDonald, W.L. and D.W. Fulbright. 1991. Biological control of chestnut blight: use and limitation of transmissible hypovirulence. Plant Disease 75: 656-661. Mahanti, N. and D.W. Fulbright. 1995. Detection of mitochondrial DNA transfer between strains after vegetative contact in Cryphonectria parasitica. Mol. Plant Microbe Interact. 8: 465-467. Mahanti, N., H. Bertrand, C. Monteiro-Vitorello, and D.W. Fulbright. 1993. Elevated mitochondrial alternative oxidase activity in dsRNA-free, hypovirulent isolates of Cryphonectria parasitica. Physiological and Molecular Plant Pathology 42: 455—463. Mannella, C.A. and A.M. Lambowitz. 1978. Interaction of wild-type and poky mitochondrial DNA in heterokaryons of Neurospora. Biochem. Biophys. Res. Commun. 80: 673-679. Marcou, D. and J. Schecroun. 1959. La senescence chez Podospora pourrait etre due a des particules cytoplasmiques infectantes. Compte rendu de l'Acad des sciences 248: 280- 283. May, G. and J .W. Taylor. 1988. Patterns of mating and mitochondrial DNA inheritance in the agaric basidiomycete Coprinus cinereus. Genetics 118: 213-220. Monteiro-Vitorello, C.B., J .A. Bell, D.W. Fulbright, H. Bertrand. 1995. A cytoplasmieally transmissible hypovinrlence phenotype associated with mitochondrial DNA mutations in the chestnut blight fungus Cryphonectria parasitica. Proc. Natl. Acad. Sci. USA 92: 5935-5939. Morris, T.J., and J .A. Dodds. 1979. Isolation and analysis of double-stranded RNA from virus-infected plant and fungal tissue. Phytopathology 69:854-858. Nuss, D.L. 1992. Biological control of chestnut blight: an example of virus-mediated attenuation of fungal pathogenesis. Microbiol. Rev. 56: 561-576. 204 Puhalla, J .E. and S.L. Anagnostakis. 1971. Genetics and nutritional requirements of Endothia parasitica. Phytopathology 61: 169-173. Rogers, H.J., K. Buck, and C.M. Brasier. 1987. A mitochondrial target for double- stranded RNA in diseased isolates of the fungus that causes Dutch elm disease. Nature 329: 558-560. Smith, M.L., L.C. Duehesne, J.N. Bruhn, and J.B. Anderson. 1990. Mitochondrial genetics in a natural population of the plant pathogen Armillaria. Genetics 126: 575-582. Van Alfen, N.K., R.A. Jaynes, S.L. Anagnostakis, and P.R. Day. 1975. Chestnut blight: biological control by transmissible hypovirulence in Endothia parasitica. Science 189: 890-891. Vogel, HI. 1956. A convenient growth medium for Neurospora (Medium N). Microb. Genet. Bull. 13: 42-43. "Illlttltlltlllllllf