UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Homothallism in the Sordariaceae : Mating-type loci in selected species of Neurospora, Anixiella, and… Beatty, Nicholas P. 1993

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_1993_spring_beatty_nicholas.pdf [ 9.08MB ]
JSON: 831-1.0086070.json
JSON-LD: 831-1.0086070-ld.json
RDF/XML (Pretty): 831-1.0086070-rdf.xml
RDF/JSON: 831-1.0086070-rdf.json
Turtle: 831-1.0086070-turtle.txt
N-Triples: 831-1.0086070-rdf-ntriples.txt
Original Record: 831-1.0086070-source.json
Full Text

Full Text

HOMOTHALLISM IN THE SORDARIACEAE:MATING-TYPE LOCI IN SELECTED SPECIES OFNEUROSPORA, ANIXIELLA, AND GELASINOSPORAbyNICHOLAS P. BEATTYB.A., Williams College, 1988A THESIS SUBMITTED IN PARTIAL FULFILLMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIESDepartment of BotanyWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAApril 1993© Nicholas P. Beatty, 1993In presenting this thesis in partial fulfilment of the requirements for an advanceddegree at the University of British Columbia, I agree that the Library shall make itfreely available for reference and study. I further agree that permission for extensivecopying of this thesis for scholarly purposes may be granted by the head of mydepartment or by his or her representatives. It is understood that copying orpublication of this thesis for financial gain shall not be allowed without my writtenpermission.(SignatureDepartment of BotanyThe University of British Cort(mbiaVancouver, CanadaDate Aff^■ 113DE-6 (2/88)ABSTRACTThe mating-type genes in Neurospora crassa, called "idiomorphs" and designatedA and a, control entry into the sexual cycle and aspects of meiosis. In heterothallicspecies, nuclei with unlike mating-types must come together before karyogamy andmeiosis can occur. Homothallic species, however, can enter the sexual cyclespontaneously and without a mating event per se. Two homothallic classes exist inNeurospora, those with only the A-specific sequence (the majority) and those with boththe A-specific and a-specific sequence (Neurospora terricola). Neurospora terricola anda number of homothallic Gelasinospora and Anixiella species are self-fertile but harborboth A and a idiomorphs in a single nucleus. Of what use are mating-types to organismsthat do not mate?A molecular analysis of the mating-type loci of selected Sordariaceae homothallicshas investigated the degree to which the homothallic A and a genes are similar to those ofheterothallic and A-type homothallic species of Neurospora. Mapping of the N.terricola,Gelasinospora, and Anixiella mating-type loci has demonstrated that the majority of theN. crassa A and A sequences are conserved in these homothallics with the exception ofapproximately 1.1 kb of A absent in N.terricola and approximately 600bp of a absent inall homothallics examined. Further analysis was made of the sequences that flank themating-type genes themselves to further define the locus. The sequence that in N.crassaflanks both A and a idiomorphs to the left is present in the genomes of all homothallicspecies examined. In most of the homothallics examined, this flank sequence iscontiguous with the a idiomoiph. A similar right flank sequence is absent in Sordariaceaehomothallics. The ORF portions of the N.terricola A and a genes were amplified by PCRand sequenced. A sequence comparison of N.crassa and N.terricola A ORFs hasItdemonstrated an average of 89% DNA identity. Functional analysis has shown that theN.terricola ORF A confers function (ability to mate with strains of opposite mating-typeand ability to form perithecia) when transformed into N.crassa sterile mutant spheroplasts.TABLE OF CONTENTSABSTRACT^TABLE OF CONTENTS^ ivLIST OF TABLES viLIST OF FIGURES^ viiACKNOWLEDGMENTS ixGENERAL INTRODUCTION^ 1Neurospora mating strategies  3Neurospora life cycle  8Life cycle differences between hetero- , homo- , andpseudohomothallics^  11Designation of species in the Sordariaceae^  12N. crassa mating-type locus  15PART ONEMAPPING THE HOMOTHALLIC MATING-TYPE LOCI^ 19INTRODUCTION I^  19MATERIALS AND METHODS I^ 20List of strains^  20Growth conditions  22Isolation of DNAs 22Digestion, electrophoresis and capillary blotting^  23N.terricola^  23Other Sordariaceae homothallics  23Probes and hybridizations  24RESULTS I^ 27Determining conservation of homothallic mating-type loci^ 27A idiomorph^ 27a idiomorph 29Analysis of A and a linkage in homothallic Sordariaceae^ 31A / a linkage in N.terricola^  31A / a linkage in other Sordariaceae homothallics  32Conservation in N.terricola of N.crassa left and right flanksequences^  32Conservation in Anixiella and Gelasinospora of N.crassa leftand right flank sequences^  33tifDISCUSSION I^ 48Left flank region^  48Right flank region 49N. terricola A / a linkage data suggest a novel mating-type locus^ 50Evolution of homothallic mating-types^  51A mechanism for the evolution of homothallic mating-type loci^ 52PART TWOSEOUENCE AND FUNCTIONALITY ANALYSIS OF N.terricola ORF A ^55INTRODUCTION II ^ 55MATERIALS AND METHODS II^ 58PCR^ 58Cloning N. terricola ORF A 59Purification of PCR fragments^ 59Blunt end ligation of A 60Transformation^  60Selection  61Cloning N. terricola ORF a 61Sequencing pNT A^ 62Testing functionality of N. terricola ORF A^ 62Testing "escape" from homothallism in N.terricola  64RESULTS II^  65Sequence analysis of N.terricola, N.africana, and N.crassa A ORFs^ 65Functionality of N. terricola ORF A^ 65"Escape" from homothallism in N. terricola  69DISCUSSION II^ 71Monophylogeny Versus Polyphylogeny in N.crassa, N.africana, andN.terricola  71A case against polyphylogeny ^  73Evolution in terms of functionality at the mating-type locus^ 73Evolution in Gelasinospora and Anixiella^  75GENERAL DISCUSSION^ 76Evolution of homo- and heterothallism^  76Evolutionary models  78Future Work^ 82REFERENCES 84LIST OF TABLESTable 1^Neurospora species and their origins.^ 7Table 2^Mating-type and ascospore morphology of 21homothallic isolates.Table 3^Summary of hybridization results.^ 38Table 4^Results of cotransformation of Am56 spheroplasts^67with pbC1 and pNTA DNAs.41.LIST OF FIGURESFigure 1^N.crassa mating-type idiomorphs and their functions.^2Figure 2^Neurospora mating strategies.^ 4Figure 3^Neurospora life cycle.^ 10Figure 4^Ascospore ornamentation in Neurospora.^ 13Figure 5^Ascospore morphology and ornamentation in the^14Sordariaceae.Figure 6^N. crassa A clones and probes derived from them.^25Figure 7^N.crassa a clones and probes derived from them. 26Figure 8^Results of N.terri cola genomic DNA hybridizations to^28N.crassa A probes.Figure 9^Results of N.terricola genomic DNA hybridizations to^30N.crassa a probes.Figure 10^Summary of linkage between flank sequences and^34idiomorphs in Sordariaceae homothallics.Figure 11^Detail of N.crassa right flank region and its conservation^35in N.terricola.Figure 12^N.terricola A and a idiomorphs are not closely linked (1).^39Figure 13^N.terricola A and a idiomorphs are not closely linked (2).^40Figure 14^Sordariaceae homothallics, with the exception of N.terricola,^41contain the entire N.crassa A idiomorph; all contain a leftflank sequence.Figure 15^Sordariaceae homothallics lack a portion of the^42N.crassa a idiom orph.Figure 16^Clarifying the border of the N.terricola a idiomorph.^43Figure 17^N.crassa left flank sequences are continuous with the^44a idiomorph in N.terricola.YltlFigure 18^Idiomorph and left flank linkage in Sordariaceae^45homothallics.Figure 19^Idiomorph linked N. crassa right flank sequences are^46not present in homothallic Sordariaceae. More distalright flank sequences are, however, conserved.Figure 20^Distal right flank sequences are present in N. terricola^47but not clearly linked to an idiomorph.Figure 21^Mechanistic model for the evolution of homothallic^54Neurospora species.Figure 22^Strategy to test functionality of N.terricola ORF A.^57Figure 23^Comparison of DNA sequences between N. crassa mt A-1,^66N. terricola ORF A, and N. africana ORF A.Figure 24^Results from cotransformations of Am56 spheroplasts with^68pbC1 (selection) and mating-type DNAs.Figure 25^Three cladistic trees for evolution in the genus Neurospora.^79ACKNOWLEDGMENTSMy thanks to Rajgopal Subramaniam for his technical expertise and general support,Myron Smith and Gretchen Kuldau for timely advice and brainstorming, Louise Glass forpatience, and Jody Holmes for everything else.i xGENERAL INTRODUCTIONThis work is a two part investigation of the nature of mating and mating-type loci inhomothallic fungi of the Sordariaceae. The first part describes the organization of mating-type loci of a Neurospora homothallic, Neurospora terricola, and of several homothallicsin two closely related genera, Anixiella and Gelasinospora. These mating-type loci arecompared to the well described mating-type locus of the heterothallic Neurospora crassaand the homothallic Neurospora africana. The second part of this investigation is a detailedanalysis of one portion of the mating-type locus of N.terricola, a portion which in N.crassacontains ORFs encoding functional polypeptides, mt A-1 and mt a-1, responsible forregulatory control over mating and meiosis. This latter analysis employs the sequencing ofthe ORFs of the N.terricola A and a idiomoThs, tests of their functionality, and acomparison of these sequences to those of N.crassa mt A-1 and mt a-1 and N.africana mtA-1. All the information is then utilized in a general discussion of heterothallism andhomothallism, their evolution, and common elements in eukaryotic control of mating events.Neurospora crassa is a filamentous ascomycete that, until the emergence of yeastbiology in the last two decades, was a preferred organism for the study of genetics. Beadleand Tatum's now famous work on the first nutritional mutants (6) was done withNeurospora and a rich period of genetic research followed, much of it based onNeurospora techniques. Currently, a resurgence of interest in filamentous ascomycetes hasfocused in part on mating, particularly the classical and molecular nature of sexualreproduction. In a number of fungal taxa, reproductive functions have been associated withspecific loci and particular genes have been identified that are now known to have mating-specific functions. Some of these genes have been cloned and sequenced and a large bodyof research has had as its focus their manipulation in mutational and functional analyses.Neurospora genes that function specifically in mating and sexual cycle events havebeen cloned and sequenced and we now know that Neurospora, like many fungi, has twoleft flank right flank'LA idiomorph 5301 bymeiotic functions mt A-1left flank^ right flanka idiomorph 3235 bymt a-1^meiotic functions?Figure 1 N.crassa mating-type idiomorphs and their functions.N. crassa idiomorphs are entirely dissimilar. A and a each encode a single polypeptidethat is implicated in mating, meiotic, and vegetative incompatibility functions. Asecond region in A has been associated with meiotic functions and the productionof fertile ascospores. A similar region in a has not been confirmed. The sequencesflanking the idiomorphs are identical in strains of opposite mating-type.43mating-types, designated A and a [Figure 1] (31). A and a sequences have been termedidiomorphs (51) in N.crassa to reflect the fact that the sequences are dissimilar, thoughresident at the same locus; a single N.crassa individual harbors a A or a idiomorph but notboth. A and a idiomorphs have several functional regions. One such region of A and Aencodes regulatory polypeptides, mt A-1 (26) and mt a-1 (88), that are thought to activatedownstream genes, presumably the many genes that effect sexually specific events of the lifecycle such as mating. Mt A-1 and mt a-1 also function vegetatively to prevent theformation of mixed mating-type heterokaryons. A x a constitutes a successful cross in thesexual cycle but a A + a union of vegetative mycelia results in a reaction known asheterokaryon incompatibility; contact between hyphae of opposite mating-types results inseverely inhibited growth or even cell death at the point of contact (27). The interaction ofmt A-1 and mt a-1 is but one factor in vegetative compatibility between individuals. A hostof het genes define compatibility groups so that individuals with different het constitutionsare incompatible (27). Heterokaryon incompatibility is also affected by a locus, tol, which isunlinked to the mating-type locus. The wild-type tol (for "tolerant") allele contributes tothe incompatibility between strains with opposite-mating-types during vegetative growth. Amutant form of the allele allows the formation of mixed mating-type heterokaryons bysuppressing A/a incompatibility in the sexual cycle (60).Neurospora mating strategies In the genus Neurospora, there are at least three distinct reproductive strategies, notall of which involve a mating event per se [Figure 2]. One is heterothallic mating, thestrategy of N. crassa described above. Heterothallic individuals are invariably of a singlemating-type, containing either a A idiomorph or a a idiomorph but not both. The fourheterothallic species of Neurospora all have one mating-type idiomorph per nucleus andindividual nuclei in multinucleate fertilizing agents (vegetative spores or hyphae) are all ofthe same mating-type (31). Heterothallic reproduction is truly sexual; individuals withheterothallichomothallicsexual reproductionandmeiosispseudohomothallicFigure 2 Neurospora mating strategies.Heterothallic mating requires the union of individuals with opposite mating-types. Pseudohomothallics have multinucleate spores which typically containnuclei with both mating-types. N.tetrasperma, the only knownNeurosporapseudohomothallic, is self-fertile in >75% of reproductive events. Outcrossingis possible with the estimated 10% to 20% of ascospores that are self-sterile.Constitutive homothallics include those with only A-specific sequences (themajority) and N.terric ola which contains both mating-type sequences in singlenuclei.opposite mating-types must interact to initiate the sexual cycle. The second and thirdNeurospora reproductive strategies are forms of homothallism or self-fertility. Homothallicindividuals are capable of undergoing sexual reproduction and meiosis without a matingevent in the traditional sense (76). A vegetative colony can spontaneously develop sexualfruiting bodies that produce fertile meiotic progeny in a manner that is macroscopicallyindistinguishable from sexual reproduction in heterothallic species (76). At the nuclear andmolecular level, however, there is considerable variation in the mating-type constitution ofwhat are generally considered homothallic species.The pseudohomothallic class, represented in Neurospora by the single speciesN.tetrasperma, is functionally homothallic but has a reproductive mechanism that isperhaps more accurately considered heterothallic. Each sexual spore or ascospore normallycontains haploid nuclei of both mating-types, A and a, and produces a self-fertile culture. Amajority of cultures from single vegetative spores (conidia) are also self-fertile because mostconidia are multinucleate and contain nuclei of both mating-types (49,72). The majority ofN.tetrasperma reproductive events are "homothallic" since the presence of A and a mating-types in single propagules precludes any need for mating. It is presumed that N. tetraspermaidiomorphs encode functional mt A-1 and mt a-1 polypeptides and that these polypeptidesinteract in a manner similar to that outlined for N.crassa. Such a process would suggest aheterothallic mating mechanism in an otherwise pseudohomothallic species but no directevidence exists to support this. Nevertheless, certain aspects of the N.tetrasperma life cycleare clearly heterothallic in nature. Approximately 10% of ascospores and 20% of conidiaare self-sterile. Cultures derived from these propagules are homokaryotic and cross-fertilewith strains of opposite mating-type whether self-sterile themselves or binucleate with bothmating-types (72). N.tetrasperma is predominantly inbred but a significant proportion ofmeioses result from these outcrossings involving self-sterile individuals.The N.tetrasperma A and a idiomorphs, while functional during the sexual cycle, donot elicit vegetative heterokaryon incompatibility in N.tetrasperma (72). The sameidiomorphs, when transformed into mutant (Am or am mutants that have lost heterokaryonincompatibility) N.crassa strains, do cause a vegetative incompatibility response whenopposite mating-types are associated in forced heterokaryons. The 1V.tetraspermaidiomorphs are thus capable of eliciting the incompatibility phenotype (in N.crassa) yet Aand a nuclei are not incompatible when they occupy the same N.tetrasperma conidium. Aplausible explanation for this fact might be the constitutive expression in N.tetrasperma of ato/-like factor which suppresses incompatibility.N.tetrasperma also differs from heterothallic Neurospora species by lacking thespecialized female hyphae known as trichogynes (72). In heterothallic species, these hyphaeare attracted to and grow toward pheromones produced by opposite mating-type conidia(9). Contact between male conidia and female trichogynes, thought to be regulated by mtJ. and mt a-1, initiates mating and leads ultimately to karyogamy and meiosis (30).N.tetrasperma bypasses such a mating event. Presumably the interaction of A and a nuclei,still controlled by mt A-1 and mt a-1, occurs within the self-fertile spore, obviating the needfor macroscopic mating structures such as trichogynes.True Neurospora homothallics, and there are five species of this type, have nucleiwith only one mating-type configuration. Selfing appears to be obligate in at least onespecies, N.terricola (Beatty, unpublished results); self-sterile ascospores are not observedas they are in N.tetrasperma (72). Two homothallic classes have been reported inNeurospora, a class in which individuals contain only the A mating-type gene and a class inwhich both mating-types are present in single nuclei [Table 11(29). The former classincludes four species while the latter class is represented by a single species, N.terricola.N.tetrasperma and N.terricola both typically harbor A and a mating-types in single cells(ascospores in N.terricola; ascospores and conidia in N.tetrasperma) and both are self-fertile. Only N.terricola, however, is an obligate homothallic; the pseudohomothallicN.tetrasperma has the ability to outcross.Table 1. Neurospora species and their origins.All heterothallic and A-type homothallics have been isolated from temperate climates.N.terricola and N.tetrasperma are the only Neurospora isolates from temperate climates.Each of these species may represent a monophyletic group (see Part Two Discussion).Heterothallic species^Climate in which first isolatedNeurospora crassa tropicalNeurospora discreta^ tropicalNeurospora intermedia tropicalNeurospora sitophila^ tropicalPseudohomothallic speciesNeurospora tetrasperma^temperateHomothallic speciesNeurospora africana^ tropicalNeurospora dodgei tropicalNeurospora galapogosensis^tropicalNeurospora lineolata tropicalNeurospora terricola^ temperate8Homothallic species have access to relatively little genetic diversity when comparedwith outbreeding heterothallic or pseudohomothallic populations but there is no evidencethat homothallism is any less successful a means for reproduction (68). Of what use, then,are mating-types to organisms that are self-fertile? What is the nature of the mating-typegenes of homothallics? Are they similar to mating-type genes of heterothallic species; arethey functional or is their presence residual, reflecting prior evolutionary events? And whatdegree of functionality, if any, is present in homothallics? Do mt A-1 and mt a-1 have onlymeiotic functions or might they be active in as yet unidentified mating events? These aresome of the questions to which I addressed my research. I chose to examine N.terricola inparticular detail because of its unique status in the Neurospora genus. In addition tomolecular identity at the mating-type locus, N.terricola is also unique in severalmorphological and cytogenetic characteristics (32,76). Considerable work has been done todescribe the evolution of N.tetrasperma (16,49,56,72,90) but comparatively little researchhas had as its focus the molecular genetics and evolution of obligate homothallics. Recentinvestigations into the N.africana mating-type locus (Glass, unpublished results) have begunto increase understanding of A-type homothallics and my work on A / a -type homothallics,including representatives from genera closely related to Neurospora, should lead toward abetter understanding of the evolution of homothallism in filamentous ascomycetes.Neurospora life cycleNeurospora heterothallics are familiar colonizers of the sites of recent fires. Charredwood is a preferred substrate though there are many reports of growth on decayingvegetation and on food products, particularly bread (14,68,69,83). All known Neurosporaspecies grow vegetatively as coenocytic, branched hyphae. These hyphae are clonal andmultinucleate and have genetic complements that reflect exactly those of their parents.Vegetative colonies are self-perpetuating when sufficient nutrients are available and hyphaeof heterothallic species form two types of vegetative spores, micro- and macro-conidia.These, upon germination, can initiate new mycelia though microconidia seem to serveprimarily as fertilizing agents to initiate sexual reproduction (68). Vegetative reproductioncontinues until nutrient depletion, particularly nitrogen limitation, induces the beginning ofthe sexual phase of the life cycle {Figure 3]. Heterothallic Neurospora species can growvegetatively for extended periods of time without forming sexual structures; conidia areproduced in generational succession as long as nitrogen, inorganic salts and a carbon sourceare available (69). In contrast, homothallic species do not form conidia. Instead, hyphalgrowth which usually occurs in soil rather than on exposed substrates, proceeds only for afew days before the spontaneous formation of proto- sexual structures or protoperithecia(76). These same structures differentiate from heterothallic mycelia but only when nitrogenavailability is limited.In heterothallic species, the sexual phase of the life cycle begins when the femaleprotoperithecium is fertilized, which induces further differentiation into the mature sexualstructure or perithecium. Fertilization takes place between a male conidium or hyphal tipand a specialized hypha of protoperithecial origin, the trichogyne. Regardless of the celltypes of the fertilizing agents, mating can only occur between individuals with differentmating-types. One function of the mating-type locus in Neurospora is presumably toregulate the production of pheromones. The female trichogyne detects and grows towardconcentrations of these pheromones whereupon contact is made with the male conidium (9).The male nucleus, presumably under its own genetic control (95), migrates down thetrichogyne and into the interior of the perithecium. One nucleus then enters the ascogoniumand becomes associated with the female nucleus. A series of synchronous nuclear divisionsgives rise to a cluster of dikaryotic ascogenous hyphae in which the original male and femalenuclei are juxtaposed (73). Karyogamy then takes place in the penultimate cell of onehypha followed immediately by meiosis and a single mitotic division to produce eighthaploid sexual progeny, the ascospores. The eight ascospores are arranged linearly inindividual asci which fill the interior of the now flask shaped perithecium. A forcibleperitheciummeiosis 0 OSC U Snuclear fusion■ascus initialmacroconidium•^• .ascospore^ua int nh ue cdjeatemycelium/Amature^ c;)ascus microconidiumt 0trichogyneprotoperitheciummicroconidium,macroconidiumor myceliumascogenous hyphaFigure 3 Life cycle of Neurospora crassa.Figure adapted from Fincham, Day, and Radford (1979) .tidischarge of ascospores ensures their dispersal and heat-induced germination gives rise tonew vegetative growth to complete the life cycle.Life cycle differences between hetero- homo- and pseudohomothallics Neurospora homothallics lack micro- and macroconidia and trichogynes (29,36,76).The distinctive orange color observed in heterothallic vegetative colonies is due to anabundance of carotenoids in the conidia; as homothallics lack conidia, their mycelia arerelatively colorless. The absence of conidia also means that there is no direct evidence forthe action of pheromones in these species, the action of which has been observed directly inN.crassa (9). Otherwise, the life cycles of heterothallics and homothallics areindistinguishable macroscopically. Both form identical hyphae, protoperithecia, andperithecia; only ascospore ornamentation differs (3,74). Variations have also beendocumented in the spindle forming structures and nucleoli of homothallics (76),pseudohomothallics (12), and heterothallics (74,76). Homothallic Neurospora species allhave spindle-forming structures known as polar caps (described below) that are unknown inheterothallics (76).The cytological picture for ascogenous hypha and ascus initiation is identical in bothhomothallic and heterothallic species (76). The pseudohomothallic N.tetrasperma differsonly in the behavior of its nuclei in the ascus. Two nuclear divisions occur (one meiotic andone mitotic) prior to ascospore delimitation, but only four ascospores are normally delimited(12,72). Each of the four ascospores encloses two nuclei, one of each mating-type. Anensuing division makes N.tetrasperma ascospores four-nucleate (12,76). Homothallic andheterothallic Neurospora species all have identical nuclear divisions but delimit only eight-spored asci (76).The chromosome number is seven for all Neurospora species (62,76,87). Severalcharacters, however, are unique to N.terricola, including:i) hemispherical nucleolus-- These are spherical in all other Neurosporaspecies, heterothallic and homothallic (76).ii) presence of polar caps-- most prominent in N.terri cola though present inother homothallics. These are nucleus-associated structures, distinguishablefrom nucleoli, which may play a role in spindle formation. Their presence hasnot been confirmed in heterothallics (76).morphology of ascospores-- N.terricola has ovoid ascospores in contrastto the spindle-shaped ascospores of other homothallic species. N.terricolaascospores are also unique in having a single germinal pore, the fixedperforation of the epispore of mature ascospores; all other Neurosporaspecies have bipolar ascospores (3,32,76).Designation of species in the SordariaceaeHeterothallic species designations are based primarily on crossing abilities withestablished reference strains. For the most part, inter-specific mating does not yield fertileprogeny. There are exceptions, however, such as the relatively high percentage of viableoffspring that result from crosses between N.crassa and N.intermedia (67). Homothallicspecies do not lend themselves so easily to systematics since they do not mate. Ascosporemorphology and ornamentation is the alternative character upon which species designationsare made. Species of the genus Neurospora are distinguished from those of closely relatedgenera principally by their longitudinally ribbed ascospores. These are characterized by thepresence of ridges or "ribs" alternating with "intercostal veins" (44). Reticulation amongsthomothallics varies from the deeply furrowed veins of N.dodgei to the shallow rib-veintopography of N.lineolata. Size and shape also varies from the relatively large, oblongascospores of N.galapogosensis to the smaller and more ovoid ascospores of N.terricola(3) [Figure 4]. A final character is the presence of single versus double germ pores. Asdescribed previously, only N.terricola has a single germinal pore; all other homothallicshave two germinal pores (3,76). These various features and ornamentations, though subtlein some cases, are consistent and have constituted the established basis for delimitingspecies. Inspections by Austin et.al . (3) of greater than 100 ascospores of each of theFigure 4 SEM views of ascospores of five homothallic species ofNeurospora. 1. N.dodgei; 2. N.lineolata; 3. N.terricola; 4. N.africana;5. N.galapogosensis. N.terricola ascospores are smaller and more ovoid thanthose observed in other Neurospora homothallics. Figure adapted from Austinet.al. (1974).V'S20 pm6•Figure 5 Sordariaceae ascospore ornamentation. 1. N.crassa: longitudinal veins areunbranched; 2. Gelasinospora species: large ovoid ascospores show prominent round pits.3. A.sublineata: veins are highly developed, branched, and reticulated. 4. Neurosporaspecies: veins unbranched and less reticulated. 5 and 6. Gelasinospora species: very smallascospores with pits. 7. Gelasinospora species: mature ascospores of this strain are-huge andhave no visible pits (immature spores show many small pits). From Glass etat (1990).tsNeurospora homothallics revealed no significant variations from the characters described.Any further confirmation of a Neurospora homothallic taxonomy will presumably have tocome from an examination of sequence and other molecular data.Genus and species identifications of other Sordariaceae homothallics are based onascospore and fruiting body morphology. Gelasinospora species have pitted ascospores[Figure 5] (23) while Anixiella species are characterized by their ascocarps which arespherical as compared to the flask-shaped ascocarps (perithecia) of Gelasinospora andNeurospora. In addition, the ascocarp of A.sublineolata lacks an ostiole, the perithecialopening through which ascospores are forcibly discharged in Gelasinospora andNeurospora species (23,29).There are numerous indications that Gelasinospora and Anixiella are very closelyaligned with Neurospora. Spindle formation and much of the meiotic progression inGelasinospora calospora (43) is indistinguishable from that in N.crassa  (74,76). Thechromosome number is seven in G. calospora (43) as it is in all homothallic species ofNeurospora (76). At least one Anixiella species, A.sublineolata, has ribbed ascospores (23)that are nearly indistinguishable from those of certain Neurospora species (3) [Figure 5].Ascospores in the two genera are similar enough to have prompted one worker to reclassifyA.sublineolata as a Neurospora species (97). Finally, as will become clear, there is strikingevidence of alignment at the molecular level, namely a conservation in selectedGelasinospora and Anixiella species of Neurospora mating-type idiomorphs and flanksequences.N.crassa mating-type locus N.crassa individuals are haploid and contain only one mating-type sequence, A or a,that is present at the mating-type locus. The A and a sequences are entirely dissimilar anddo not, therefore, fit the traditional definition of alleles. Their dissimilarity has prompted theterm "idiomorphs" by the researchers who cloned them (26,88). As idiomorphs, A and gILare defined as unique 5301 and 3235 bp sequences, respectively, that in opposite mating-types occupy the same locus on linkage group I [Figure 11.While the mating-type idiomorphs are unique, sequences bordering the idiomolphs(flank sequences) are nearly identical in A and a strains (26,88) [Figure 1] . It is these flanksequences to the left and right of the idiomorphs that really define the mating-type locus perse. At least one of the N.crassa flanks is conserved in the homothallic N.africana (L.Glass,personal communication; N.Beatty, unpublished results) and portions of both right and leftflanks may define the mating-type locus in numerous taxa of the Sordariaceae (T.Randall,personal communication; N.Beatty, unpublished results). Sequence data from the rightflank has revealed a single ORF that contains a DNA motif common to fungal pheromonesand the region containing this ORF appears to be species-specific. Other portions of theright and left flanks, however, are conserved in heterothallic and homothallic species ofNeurospora, Gelasinospora, and Anixiella (T.Randall, personal communication; N.Beatty,unpublished results).Functional analyses of the N.crassa mating-type locus have associated at least threedistinct functions with specific portions of the A and a idiomorphs [Figure 1]. A contains atleast two functional segments, one of which acts in both the vegetative and sexual phases ofthe life-cycle. The A idiomorph contains a single ORF that encodes a polypeptide, mt A-1,of 288 amino acids (26). Similarly, the a idiomorph contains an ORF that encodes apolypeptide, mt a-1, of 382 amino acids (88). These polypeptides control entry into thesexual cycle; mutants in which the A or a ORF has been disrupted cannot mate (26,33). MtJ. and mt a-1 also act vegetatively to prevent the formation of mixed mating-typeheterokaryons. In a reaction known as heterokaryon incompatibility, juxtaposition ofvegetative mycelia with opposite mating-types results in severely limited growth or celldeath at the point of contact (27). This reaction is dependent on intact copies of bothmtA-1 and mt a-1 and presumably these polypeptides interact differently in vegetativegrowth than they do in the sexual cycle. Otherwise, A and a nuclei could not be associatedin the ascogenous dikaryon without eliciting an incompatibility reaction.A and a functions have been determined by deletional (26,88) and mutational(13,28,33,34) analyses. The former has been employed to associate mating andheterokaryon incompatibility functions with specific regions of the mating-type locus. Thelatter analysis has led to the generation of mating-type mutants which have helped toelucidate the specific activities of mt A-1 and mt a-1 and other functional regions of theidiomorphs (28). Griffiths and DeLange (34) and Griffiths (33) generated a class ofmutants, Am and am, with mutations in mt A-1 and mt a-1 respectively. In functionalitytests, these mutants were found to be sterile and heterokaryon compatible supporting theidea that mt A-1 and mt a-1 function in mating and vegetative incompatibility.Subsequently, other A mating-type mutants have been generated by repeat induced point(RIP) mutation (28). RIP operates on any duplicated sequence in Neurospora by alteringboth copies and rendering them dysfunctional. Alterations come in the form of G-C to A-Ttransition mutations and arise only during mitotic divisions of haploid nuclei in ascogenoushyphae (80,81). The result is that RIP can be used to generate mutants simply byintroducing an extra copy of the sequence one wishes altered. Both the native andintroduced copies are changed and selection for transformants yields functional mutants.RIP-generated mutations (28) in regions (other than mt A-1) of the A idiomorph haverevealed a novel functional region [Figure 1]. Strains with these mutations aremorphologically indistinguishable from the wild type in early perithecial development butare greatly diminished in their ability to produce fully developed perithecia and ascospores.This same region, while required for perithecial development and certain meiotic functions,is apparently not necessary for vegetative incompatibility (28).A region of the a idiomorph with similar function has been assessed by deletionalanalysis (Staben and Yanofsky 1990). N.crassa sterile mutants ami, am33 , and am3°(33,34) were transformed with various deletional portions of the a idiomorph and assessedwith regard to mating identity, heterokaryon incompatibility and perithecium maturationfunction. In this manner, mating and incompatibility functions were associated with the mta-1. A second region of the idiomoiph, not including the ORF, was found to function in thematuration of perithecia and the generation of ascospores (88). [Note: The secondfunctional region is now in question; C.Staben, unpublished results; L.Glass, personalcommunication.] Both the A and a idiomorphs, then, contain ORFs which function inmating and opposite mating-type vegetative incompatibility; both also harbor additionalregions that are required for perithecium development and certain meiotic functions.19PART ONEMAPPING THE HOMOTHALLIC MATING-TYPE LOCIINTRODUCTION IHeterothallic Neurospora mating-type loci contain a single idiomoiph, A or a, andconserved flanking sequences that are identical in both A and a strains. The idiomorphscode for mating, meiotic, and vegetative incompatibility functions and flank sequences havebeen implicated in the production of sexual pheromones (T.Randall, unpublished results).Heterothallic sexual reproduction requires a mating event between strains withopposite mating-types. Mt A-1 and mt a-1 maintain control over cellular recognition eventsduring vegetative and sexual growth. The role played by these polypeptides in self-fertilizing or homothallic species, though, is less clear. Especially in A/A-type homothallicssuch as N.terricola, the existence of mating-type genes is difficult to reconcile.Homothallics may require sexual reproduction in order to utilize meiotic repair mechanismsor to produce desiccation resistant ascospores. Mt A-1 and mt a-1 have important meioticfunctions in homothallic species and yet the mating and vegetative incompatibility functionsassociated with mt A-1 and mt a-1 seem extraneous.A mating-type locus, defined by an idiomoiph and flank sequences, is well conservedin heterothallic Neurospora species including N.crassa, N.intermedia, and N.sitophila(L.Glass, personal communication). The A-type homothallic N.africana contains an Aidiomorph that is >90% similar to the N.crassa idiomorph (L.Glass, unpublished results).Little is known, however, about the conservation in N.africana of N.crassa flank sequences.Even less is known about the mating-type loci of other Neurospora homothallics.Preliminary work has demonstrated that the N.crassa A idiomorph is highly similar inrepresentative heterothallic (27) and homothallic (29) Neurospora species. NumerousGelasinospora and Anixiella species also contain similar A sequences. A / a -typehomothallic species of Neurospora, Gelasinospora and Anixiella contain A and asequences, both of which have a high degree of homology to the corresponding N. crassaidiomorphs (29). It is not clear, however, whether these species harbor both A and amating-types in their entirety and whether both mating-types occupy a single locus.The analysis of the mating-type loci of A/a-type Sordariaceae homothallics presentedhere has as its aim the formulation of a tenable evolutionary history of homo- andheterothallism. Arguments can be made for a history in which heterothallism evolved intohomothallism. The presence in homothallic species of elaborate mating-type loci isreconcilable if they evolved first in heterothallic species and have been subsequentlyinherited by homothallics. Alternatively, it is counter-intuitive that homothallic specieswould develop mating and incompatibility functions they do not require. It is unclearwhether A and A sequences function in A / a -type homothallics but a detailed picture of themating-type locus(i) of selected homothallics may suggest mechanistic models to supportthe concept that a heterothallic Neurospora (or other Sordariaceae) species is ancestral toall homothallic species. The loss of mating-type-specific sequences, for example, cansupply a directionality to cladistic trees and allow an evolutionary history to be constructedfor Neurospora species. Coupled with information about the functionality in homothallicspecies of A and a idiomorphs (part II of this thesis), cladistic trees can then be expanded toaddress the evolution not only of species but of mating strategies in general.MATERIALS AND METHODS IList of strains The strains employed in the analysis of mating-type locus conservation and linkage arelisted in Table 2.Table 2. Mating-type and ascospore morphology of Sordariaceae homothallicisolates. Neurospora species, heterothallic and homothallic, are known to have bi-polarascospores. N.terricola is the single exception and has only a single germ pore. N.terricolais also unique in being the only Neurospora species isolated from a temperate region.FGSC refers to Fungal Genetics Stock Center: Department of Microbiology, University ofKansas Medical Center.species mating-type(s)source wherecollectedascosporemorphologyAnixiella sublineolata A and a FGSC 5508 Japan intercostal veins andribs; two .erm poresGelasinospora calsopora A and a FGSC 6535 FrenchCongopitted ascosporesGelasinospora reticulospora A and a FGSC 6537 Colombia pitted ascosporesGelasinospora 142-1 A and a R.L.Metzenber.YucatanPeninsulapitted ascosporesGelasinospora 143-4 A and a R.L.MetzenbergYucatanPeninsulapitted ascosporesGelasinospora S23 A and a D.D. Perkins India pitted ascosporesNeurospora africana A FGSC 1740 Nigeria intercostal veins andribs; two germ poresNeurospora crassa A A FGSC 987 Louisiana intercostal veins andribs; two germ poresNeurospora crassa a a FGSC 988 Louisiana intercostal veins andribs; two germ poresNeurospora terricola A and a FGSC 1889 Wisconsin intercostal veins andribs; one germ poreZ12:1-Growth conditions Homothallic strains were grown in liquid media cultures containing lx ModifiedWestergaard's salts [0.1% 10103 , 0.08% KH2PO4 (anhydrous), 0.02%K2HPO4(anhydrous), 0.05% MgSO4-7H20, biotin (24tg/mL), trace elements + NaCl +CaC12] and 2% fructose.Cultures were kept stationary at room temperature (-25°C) for seven to ten days oruntil perithecia were clearly visible. Optimal mycelial yield was obtained after the onset offruiting but before the forcible discharge of ascospores. The time required for the formationof perithecia varied from species to species and was generally shortest (4-5 days) forAnixiella cultures and longest (8-10 days) for N.africana and N.terricola cultures.Heterothallic strains were cultured in liquid media containing lx Vogel's salts and 2%sucrose. Mycelia were allowed to grow for six to eight days until conidiation was wellestablished. Both homo- and heterothallic mycelia were harvested by vacuum filtration ontoWhatman 1 filter paper disks. Hyphal mats were then washed twice with 0.9% NaCl andtransferred to sterile tubes.Isolation of DNAs After desiccation, hyphae were pulverized by vortexing with a glass rod. Then thefollowing protocol (modification of the method of Berlin and Yanofsky, 1989) wasperformed to isolate DNAs of both homo- and heterothallic strains:1. transfer 3504, pulverized hyphal powder to 1.511L Eppendorf tube2. suspend sample in 40011L of salt-detergent solution (4mg/mL sodiumdeoxycholate, 10mg/mL polyoxyethylene cetyl ether, 2M NaCl)3. vortex and incubate at room temperature for 20 minutes4. spin at 12,000rpm for 10 minutes5. collect supernatant and transfer to fresh tube6. add 1:4 vol:vol supernatant:TCA/Et0H [TCA/Et0H = 4.5M Na-TC- in an ice bath, add concentrated NaOH until pH reaches neutrality]7. mix gently by inversion and incubate on ice for 20 minutes8. pellet nucleic acids by microfuging for 30 seconds2:59. wash pellet with 70% Et0H, dry in speed vac, and re suspend in 3004,10mg/mL RNAse10. incubate at 50°C for one hour,11. add 1:1 sample:phenol/CC14, vortex, spin, take supernatant12. add NH40Ac to 0.3M; then add 2.5 volumes 95% Et0H13. mix well by inversion, incubate 10 minutes room temperature, spin, wasldry, resuspend Digestion, electrophoresis and capillary blottingN.terricolaN.terri cola genomic DNA was digested with one of two groups of restrictionenzymes. Group I incorporated the following single and double digestions: EcoRV,HindIII, EcoRV/HindIII, EcoRI, jJI, çRIIBglIJ,PstI, BamHI, PstI/BamHI. Group IIincorporated single digests with more rare cutters: BssHII, KspI, MluI, Sad, ScaI, SmaLSphI, XbaI. Digestions were performed with 51.tg DNA and 20u enzyme for three hours at37°C.Other Sordariaceae homothallics Genomic DNAs from the remaining homothallics were digested with restrictionendonucleases organized into Groups III EcoRV; IV Psfl; and V BamHI, HindIII, SacI,subjected to gel electrophoresis, and transferred to nylon membranes in the mannerdescribed for N.terricola. Membranes were hybridized to the same N.crassa probes utilizedin the mapping of the N.terricola idiomorphs and flank regions.Digested DNAs were electrophoresed in 0.8% agarose gels at 35 volts for 14-18hours. Gels were stained and photographed to visualize the efficiency of digestions. Thefollowing denaturations (77) were then performed to facilitate DNA transfer to membranes:1. acid depurinate in 0.25M HC1 for 10 minutes2. alkali denature in 1.5M NaC1/ 0.5M NaOH 2 x 15 minutes3. neutralize in 1.0M Tris-Cl pH 7.4 / 1.5M NaC1 2 x 15 minutes2 - 4Transfer (77) to Nylon membranes (Amersham Corp.) was performed by capillary transferwith 10X SSC. Blotting was effectively complete after 12 hours and gels weretransilluminated with UV light to confirm the absence of fluorescent EtBr (indicating acomplete transfer). Membranes were baked at 65°C for two hours and stored in plasticbags.Probes and hybridizations Idiomorph and flank region probes were constructed from cloned portions of theN.crassa A and A mating-type loci. This was done by analyzing consensus maps of N.crassaclones and digesting with restriction enzymes to yield the desired fragments. Digests wereconfirmed by electrophoresis and target bands were purified by excising them from the geland extracting the DNA (GeneClean; Bio 101 San Diego, CA). Membranes werehybridized to a variety of probes representing idiomorph and flanking regions of theN.crassa mating-type locus. A graphic representation of N.crassa clones and the probesconstructed from them can be found in Figures 6 and 7. Membranes were incubated in pre-hybridization fluid (77) for a minimum of three hours. During this incubation, probes werelabeled by the random priming method (T7 Quick Prime kit; Pharmacia Corp.) with [a3211dCTP (Amersham Corp.), gravity filtered through a Sephadex G-50 column, andmeasured in a scintillation counter. The target range for labeled probes was a specificactivity of 8x106 to 1.5x107 counts per minute. Probes were denatured by boiling for fiveminutes and placed on ice. Pre-hybridization fluid was replaced with 15 mL of hybridizationfluid (77), labeled probe was added, and membranes were incubated at 65° for 12 to 18hours. Membranes were then subjected to the following washes: 1% SDS, 0.1x SSC 15'RT; 0.1%SDS, 0.1x SSC 30' 60° (2 times). After three to seven day exposures to KodakX-OMAT film, restriction fragment size patterns were compared to each other and analyzedwith respect to band sizes predicted from established maps of N.crassa.n2..A.-140...A idiomorphcosmid G16C10 pMT 2.519Ir^SphI^Ilir11kbHi ndIII iBamHI SphIPstI pMT-1501.0 kb1100pMT 6.8187_5SphIEcoRVSphIEcoRVBamHI1000800SacII^PvuII^HindIII^SphI6001000PvuII Pvull450PvuII^SacII1800Figure 6 N.crassa A clones and probes derived from them.The G16 C10 cosmid, pMT 6.818, pMT -150, and pMT 2.519 clones wereprovided by N.L.Glass. Consensus N.crassa A sequences (27) were used tochoose restriction sites and to construct probes based on digestion at thosesites. Probes are drawn so that they lie beneath the approximate region inN.crassa from which they were derived.lit a- 1a idiomorphpCSN15SadI400Sac I^ pCSN5^EcoRIEcoRV^NcoI600EcoRV^pCSN4^BamHI2000BamHI^BamHI900Figure 7 N.crassa a clones and probes derived from them.The pCSN15, pCSN5, and pCSN4 clones were provided by N.L.Glass.N.crassa consensus a sequences (88) were used to choose restriction sitesand to construct probes based on digestion at those sites. The pCSN4 plasmidwas used in its entirety as a probe. Probes are drawn so that they lie beneaththe approximate region in N.crassa from which they were derived.110RESULTS IDetermining conservation of homothallic mating-type lociA idiomorphDNA from each of the homothallic isolates was digested with a Group IV enzyme(PstI) and hybridized to selected N.crassa A probes. A HindIII/SphI probe representingthe N.crassa mt A-1 and two central idiomorph probes, PvuIl/SacII and PvuIl/Pvull[Figure 6], hybridized strongly to N.terricola, Gelasinospora, and Anixiella DNA. Basedon these strong hybridizations between N.crassa and homothallic DNAs, it is clear that themajority of the N.crassa A idiomorph is conserved in the homothallic species examined[Figures 8,12,14].The distal portion of the idiomorph, opposite mt A-1, is more variable. Though mostof the homothallic species tested contain the N.crassa A idiomorph in its entirety, at leastone homothallic species is missing a significant segment of the idiomorph. ABamH1/EcoRV 800 by probe [Figure 6], derived from the end of the idiomorph oppositemt A-1, hybridized strongly to all homothallic species except N.terricola [Figure 14].Strong hybridization was apparent in all lanes except that with N.terricola DNA [Figure 14,lane g], indicating that a portion of the A idiomorph at least as large as that spanned by the800 by probe is missing from the N.terricola genome. A probe, Pvull/PvulI [Figure 6],generated from a sequence slightly internal to the BamHI/EcoRV probe, hybridized well toN. terricola DNA, indicating that the N. crassa region spanned by this probe is present in theN. terricola genome. It is clear, then, that the border in N.terricola between A-specificsequence and "flank"sequence lies in the 400bp intervening region between theBamHI/EcoRV and PvuIl/PvuII probes. Thus, N.terricola lacks approximately 1.0 kb ofthe N.crassa A idiomorph; the N.crassa idiomorph is 5301 by while the N. terricolaidiomorph is approximately 4300 bp. All other homothallic species tested have Aidiomorphs that are similar in size to that of N.crassa.2:1 N.crassa A idiomorphmt A-1left flank right flankA 5301 bpHind!!! PstIEcoRV BamHI EcoRVSphISac!!PvuIIEPvuIIHind!!!Sac!!hybridizeN.terricok genomic DNAFigure 8 Results of N.terricola genomic DNA hybridizations toN.crassa A probes. Probes generated from N.crassa clones were hybridizedto N.terricola genomic DNA. Shaded boxes represent those probes that hy-bridized. The unshaded boxes represent probes that did not hybridize to N.terri-cola though it did hybridize strongly to Gelasinospora and Anixiella DNAs. TheHindIII/PstI left flank fragment is included as a reference. In N.terricola, thisflank sequence is contiguous with the a idiomorph. (See Figure 9). In G.reticulo-spora and G.S23 , the same flank sequence may be contiguous with the a idiomorph,the A idiomorph or both.2._a idiomoiphThe N.crassa a idiomorph was found to be largely conserved in each homothallicspecies examined. Three N.crassa a idiomorph probes, pCSN4 (entire), pCSN4EcoRV/NI and pCSN15 EcoRI/SacII [Figure 7] were hybridized to homothallic DNAsdigested with the Group IV enzyme (PstI). A pCSN4 probe, which represents the majorityof the N.crassa a idiomorph, hybridized strongly to all homothallic DNAs [Figure 15: 1]. ApCSN4 EcoRV/NcoI probe, corresponding to mt a-1, also hybridized well (data notpresented as a photograph) to homothallic DNAs. The pCSN15 EcoRI/SacII probe, fromthe end of the idiomoiph opposite the ORF, did not hybridize to any homothallic DNAs[Figure 15: 2]. The weak hybridization in lanes c, d, and i of Figure 15 (probe 2) can beattributed to a small amount of probe contamination, possibly with A idiomorph sequences.The difference in intensity between the A conrol DNA hybridization (lane j) and the weakhybridizations in neighboring lanes, however, makes clear the fact that the N.crassa rightflank sequence is absent or highly diverged in Sordariaceae homothallics.The pCSN15 EcoRUSacII probe represents a portion of the N.crassa a idiomorph,extending from the EcoRI site to the right flank border, that is absent from the genomes ofall the homothallic species examined. The missing portion includes the EcoRI/SacIIfragment itself as well as the remaining 200bp that separate the Sacll site from the border ofthe idiomorph [Figures 7, 9]. The a idiomorphs of the homothallic species tested aremissing approximately 700 bp of the N.crassa a idiomoiph. The homothallic a idiomorphsmeasure approximately 2500 bp in length while the N.crassa idiomorph is 3235 bp inlength.To clarify the N.terri cola a idiomorph border, Group I digested genomic DNA washybridized to a third N.crassa a probe, pCSN4 BamHI/BamHI, representing a regioninternal to the EcoRI/Sacll fragment which is not present in N.terri cola [Figures 7,9].Given the absence in N.terricola of sequences represented by the pCSN15 ^probe, the intent was to determine the approximate region where conservation of the31)N.crassa a idiomorphleft flank mt a-1.4111111—.■—right flanka 3235 byHindIII PstI SphI EcoRVEcoRV NcoIBamHIEcoRI SacHBamHIhybridize1N.terricola genomicFigure 9 Results of N.terricola genomic DNA hybridizations toN.crassa a probes. Probes generated from N.crassa clones were hybridized toN.terricola genomic DNA. Shaded boxes represent those probes that hybridizedand unshaded boxes those that did not hybridize. Flank probes are identical tothose in the Figure 8. In most homothallic species tested, left flank sequences arecontiguous with the a idiomorph [Figure 10]. Right flank sequences are absent inall cases.31N.crassa a idiomorph resumes. Hybridization yielded a characteristic a pattern similar tothat observed from hybridization to the pCSN4 probe [see Figure 12: 2]. Sequencesrepresented by the BamHI/BamHI (900bp) probe, then, are contiguous with the a ORF andpart of the N.terricola a idiomorph. The N.terricola a idiomorph border must lie betweenthe BamHI site and the EcoRI site.Summarizing a idiomorph data, approximately 700bp of the N.crassa a idiomorph ismissing from the genomes of all Sordariaceae homothallics tested. In the remainder of thehomothallic isolates, the a idiomorph is well conserved and, based on the intensity ofhybridizations, appears to be highly similar to the N.crassa a idiomorph.Analysis of A and a linkage in homothallic SordariaceaeA / a linkage in N.terricolaNeurospora terricola genomic DNA, when digested with the Group I restrictionenzymes, yielded fragments ranging from 0.5 to 12 kb in length. These fragments weresubjected to gel electrophoresis, transferred to nylon membranes, and membranes werehybridized to N.crassa A idiomorph, a idiomorph, and flank region probes. The RFLPpattern [Figure 12: 1] observed after N.terricola hybridization to a N.crassa HindIII/SphImt A-1 probe [Figure 6] was entirely dissimilar to that [Figure 12: 2] generated from aN.crassa NcoI/EcoRV mt a-1 probe [Figure 7]. The fact that N.terricola hybridizations toN.crassa mt A-1 and mt a-1 probes each produced a unique pattern indicates that thelocation of each N.terricola A and a idiomorph is also unique. (The ORF is consideredintegral to the idiomorph so that the ORF locus also describes the idiomorph locus). Basedon analysis of band sizes, no single N.terricola fragment hybridized to both A and aN.crassa probes. The A and A idiomorphs in N.terricola must be separated by a region atleast as sizable as the largest unique band observable from the autoradiographs, in this casesome 10kb.To elucidate further the linkage between N.terri cola A and a idiomorphs, Group IIenzymes were employed. The genomic fragments (3-20 kb) produced by these less frequentcutters were larger on average than those (0.4- 10kb) produced by the Group I enzymes.Of eight Group II enzymes employed, three gave similarly sized bands in hybridizations toboth A and a N.crassa idiomorphs. BssHII, MluI and SmaI digests each yielded bands ofsimilar size in A and a hybridizations [Figure 13: lanes a, c, fl. Two interpretations of thesedata are possible. The N.terri cola A and a idiomorphs may be separated by an interveningsequence greater than 10kb but smaller than 18kb in length; the similar size bands mayrepresent the same fragment and indicate linkage between the A and A idiomornhs.Alternatively, the similar sizes of MluI and SmaI bands in A and A hybridizations,respectively, may be random and insignificant. In either case, size differences of largebands are difficult to resolve in low percentage (0.8%) agarose gels. Pulsed field gelelectrophoresis (CHEF) could likely resolve the linkage between A and a and theassociation of homothallic mating-type idiomorphs with specific linkage groups.A / a linkage in other Sordariaceae homothallics Gelasinospora and Anixiella DNAs were digested with Group V enzymes, subjectedto gel electrophoresis, and transferred to nylon membranes. Membranes were hybridized topMT -150 (G-2) and pCSN4 1V.crassa probes [Figures 6, 71 representing mt A-1 and mt a-1 sequences, respectively. Autoradiograph band patterns [Figure 18: 2 versus 3] that resultedfrom hybridizations to these probes were entirely dissimilar. As with N.terricola, uniqueband patterns were equated with unique positions for the A and a idiomorphs. The twoidiomorphs must be separated by a distance at least as great as the largest dissimilar band, inthis case 12kb.Conservation in N.terricola of N.crassa left and right flank sequences When a HindIII/PstI 1V.crassa left flank probe [Figure 61 was hybridized toN.terri cola genomic DNA , a RFLP pattern [Figure 17: 11 was observed that exactlymatches the pattern [Figure 17: 2] generated when hybridization was made to the N.crassapCSN4 [Figure 7] probe. Such a matching pattern could only be produced if the left flankand a idiomorph sequences are contiguous in N.terricola. It is not clear, however, how theN.terricola a idiomorph is oriented and whether the conserved N.crassa flank sequenceborders to the left or the right. No bands were observed [Figure 19: 1] when aSphI/EcoRV N.crassa right flank sequence [Figure 6] was used as a probe indicating thatthis sequence is unique and not conserved in N.terricola.Additional hybridizations were performed to investigate the possibility that someportion of the N.crassa right flank might be conserved in N.terricola. N.terricola genomicDNA digested with Group I enzymes was hybridized to a large (12kb) SphI/SphI fragment[Figure 11] of the N.crassa cosmid G16 C10. This fragment encompasses the 1.0kbSphI/EcoRV right flank probe that did not hybridize to N.terricola. The probe also includesan additional 11kb of N.crassa flank sequence that extends further out from the (A or a)idiomorph. The intent was to hybridize this large flank probe to N.terricola genomic DNAand to look for a RFLP pattern that matches the pattern produced by hybridizations to A ora idiomorph probes. Such a match would indicate some conservation of N.crassa rightflank sequences in N.terricola and the linkage of these sequences to an idiomorph. Theresult obtained [Figure 20] was a strong hybridization but a RFLP pattern dissimilar to bothA and a idiomorph patterns. The presence of N. crassa right flank sequences was confirmedbut no linkage was established. CHEF gel analysis could likely resolve whether the rightflank sequences conserved in N.terricola are linked to the A or a idiomorph (or not linkedto either idiomorph).Conservation in Anixiella and Gelasinospora of N.crassa left and right flank sequences Homothallic DNAs digested with Group 111 and Group IV enzymes were hybridizedto a pMT 2.519 HindIII/PstI probe [Figure 6] that represents the N.crassa left flank region.Each of the homothallic DNAs hybridized strongly to this probe [Figure 14: 2]. Thus aA 999A. sublin.G. 142-1G. 143-4G. calosp.(G. retic.)(G. S23)N. terricola999HeterothallicIN. crassaType A homothallicN. africanaType A / a homothallicFigure 10 Summary of linkage between flank sequences and idiomorphs inSordariaceae homothallics. The N.crassa left flank region is similar in Sordariaceaehetero- and homothallics.The left flank is contiguous with the A idiomorph of N.africana.The same flank is contiguous with the A idiomorph in 6 of 8 AIR-type homothallics tested.Idiomorph/left flank linkage has not been conclusively determined for G.reticulosporaand G.S23. The N.crassa right flank is absent or diverged in all homothallics examined.left flank right flankA or a idiomorph3.pheromone ORFcosmid G16 C10SphI^EcoRVI 1.2 kb ISphI SphIN.terricola genomicFigure 11 Detail of N.crassa right flank region and its conservationin N.terricola. Probes generated from an /V. crassa cosmid were hybridized toN.terricola genomic DNA. The shaded box represents a probe that hybridizedstrongly, the unshaded box a probe thatdid not hybridize. The region containingthe pheromone ORF is absent or diverged in N.terricola,Gelasinospora, andAnixiella species. This region may be species-specific in the Sordariaceae. Distalto this region, N.crassa sequences are again present in N.terricola, suggesting thatthe mating-type region as a whole may be conserved. Figure is not to scale.3Lsignificant portion of the N.crassa left flank region is present in the genomes of each of thehomothallic species tested. The single digests from groups III and IV, however, yieldedinsufficient data to resolve the question of whether the homothallic left flank sequence islinked to the A or a idiomorph.To resolve this linkage question, A. sublineata, G.calospora, G.reticulospora, G.142-1, G.143-4, and G.S23 DNAs were digested with Group V enzymes. Membranes werehybridized to pMT -150(G-2), pMT 2.519 PstI/HindIII [Figure 6] and pCSN4 [Figure 7]N.crassa probes, representing mt A-1, left flank, and mt a-1 sequences respectively.A.sublineolata, G.calospora, and G.143-4 each yielded similar RFLP patterns when probedwith the N.crassa left flank and mt a-1 probes [Figure 18: 1 versus 2]. This fact suggeststhat the left flank sequence is contiguous with the A idiomorph in these species. Linkage isless clear for the remaining Sordariaceae homothallics. 1 of 3 G. reticulospora bands is thesame in A idiomorph and left flank hybridizations while 0 of 3 match in a idiomorph and leftflank hybridizations. Precisely the same situation was observed in G.S23 hybridizations.Presumably due to insufficient DNA, G.142-1 hybridizations were too weak [Figure 18:lanes j-1] to make any conclusions about linkage. Summarizing these data, the left flankregion of N.crassa is present in all Sordariaceae homothallics examined. This flank iscontiguous with the a idiomorph in A.sublineolata, G.calospora, and G.143-4. The sameflank region may be contiguous with the A idiomorph in G.reticulospora and G.S23 butsuch a conclusion is equivocal.Similar analyses were made of the right flank region. An N.crassa pMT 6.818SphIacQRV right flank probe [Figures 6] was hybridized to homothallic DNAs digestedwith Group III and Group IV enzymes. No hybridization was observed to any of the eight[Table 2] homothallic DNAs; six repeat hybridizations [Figure 19: 1] with DNA from eachstrain yielded no observable bands. Thus all Sordariaceae homothallics tested, includingN.terricola, lack the N.crassa right flank sequence encompassed by the SphI/EcoRV rightflank probe. Most of the Anixiella and Gelasinospora species tested do harbor a distal rightflank sequence from N.crassa, however. Hybridizations of homothallic DNAs to a G16C10 Sphl/Sphi probe [Figure 111 showed the presence of this sequence in all species tested[Figure 19: 2] with the exception of G.142-I and G.S23 [lanes d and fl.239a^b^cde^f^ghiFigure 12 N.terricola A and a idiomorphs are not closely linked.N.terricola genomic DNA was probed with 1. pMT 6.818 (entire) and 2. pCSN4.[See figures 7 and 8]. Lanes are digests with a)EcoRV, b)HindIII,c)EcoRV/HindIII,^d)EcoRI, e)BgIII, f)EcoRI/BglII, g)PstI, h)BamHI,i)PstI/BamHI. Band patterns are entirely dissimilar indicating in N.terricola that A- andspecific sequences are^separated (unlinked) by at least 12kb (lane d).Table 3. Results of hybridizations of homothallic genomic DNA to N.crassaidiomorph and flank region probes. Hybridizations, when positive, were strong in allcases. Failed hybridizations (N.terricola genomic to BamH1/EcoRV fragment and allisolate genomic DNAs to SphI/EcoRV right flank fragment) were repeated three or moretimes to confirm the absence of those sequences from genomes. See Figures 6-10 forgraphic representation of probes and hybridizations.N.crassa size mating• region of hybridization hybridization toprobe type idiomorph to N.terricola Anixiella, and(locus) GelasinosporaHindIII/SphI 600 A mt A-1 yes yesPvuII/SacII 1800 A middle yes yesPvuIl/Pvull 450 A middle yes yesPvull/SacII 1000 A middle yes yesBamHI/EcoRV 800 A oppositemt A-1 no yesEcoRV/NcoI 600 a mt a-1 yes yesBamHI/BamHI 900 a middle yes yespCSN4 2000 a middle plusmrLAJ„yes yesEcoRUSacII 400 a oppositemt a-1 no noHindIII/PstI 1100 ---- left flank yes yesSphliEcoRV 1000 ---- right flank no noG16 C10 12kb ---- right flank yes yes *SphUSphI* with exception of G.142-1 and G.S233 61012•■■••■•■•Figure 13 N.terricola A and a idiomorphs are not closely linked.N.terricola genomic DNA was digested with rare cutters and probed with 1. pMT6.818 (entire) and 2. pCSN4. I See figures 6 and 7]. Lanes are digests witha)BssHII, b)KspI, c)MluI, d)SacII, e)ScaI, f)SmaI, g)Sphl, h)XbaI. Bandpatterns are largely dissimilar indicating in N.terricola that A- and a-specificsequences are separated (unlinked) by at least 16kb (lane d). Lanes a, c and f havebands that are similar in size in digests 1 and 2, a result which may or may not besignificant in terms of linkage (see Materials and Methods I: Analysis of A/alinkage in N.terricola).a^b^c^d^e^f^g^h^1Figure 14 Sordariaceae homothallics, with the exception of N.terricola,contain the entire N.crassa A idiomorph; all contain a left flank sequence.Homothallic DNA was digested with PstI and probed with 1. pMT6.818 EcoRV /BamHI (800bp) and 2. pMT2.519 HindIII / PstI (1100bp). [Figures 6 and 8].Probe 1 represents an internal A-specific sequence that is absent (lane g) inN.terricola. Probe 2 represents the N.crassa left flank region which is present in allspecies tested. Similar band patterns in 1 and 2 suggest that the left flank is linked tothe A idiomorph in a number of the homothallic species but such a conclusion is basedon the comparison of a single band in most cases. Lanes a, b, and g are partial digestsand show multiple bands (unreliable for linkage analysis) [see Figure 18]. Lanes areas follows: a)A.sublineata, b)G.reticulospora, c)G.143-4, d)G.142-1, e)G.calospora,f)G.S-23, g)N.terricola,h)N.africana,ON.crassa A, j)N.crassa a.423abcde^f^ghiiFigure 15 Sordariaceae homothallics lack a portion of the N.crassa aidiomorph. Homothallic DNA was digested with PstI and probed with 1. pCSN4and 2. pCSN15 EcoRI / SacII (400bp) [Figures 7 and 19]. Probe 1 represents themajority of the a idiomorph including mt a-1. Probe 2 represents a segment of thea idiomorph from the end opposite mt a- 1. This region is diverged or absent in allhomothallics tested. Faint bands visible in 2 are probably due to slight probecontamination, probably with A idiomorph sequences. Lanes are as follows:a)A. sublineata, b)G.reticulospora, c)G.143-4, d)G.142-1, e)G.calospora, f)G.S-23 , g)N.terricola,h)N.africana,i)N.crassa A, j)N.crassa a.a^b^c^d^e^f^g^h^i^j^k6Figure 16 Clarifying the border of the N.terricola a idiomorph.N.terricola genomic DNA was digested with Group I enzymes and probed with aBamHI/BamHI (900bp) fragment representing a central portion of the N.crassa aidiomorph [see Figures 7 and 19]. Hybridization yielded a characteristic a patternsimilar to that observed from hybridization to an ORF a (pCSN4) probe [seeFigure 12: 2]. Sequences represented by the BamHI/BamHI (900bp) probe, then,are contiguous with the a ORF and part of the N.terricola a idiomorph. TheN. terricola a idiomorph border must lie between the BamHI site and the EcoRIsite [Figures 7,9,15] so that at least 600bp of the N.crassa a idiomorph are missingfrom the N.terricola genome. Lanes are digests with a)EcoRV, b)HindIII,c)EcoRV/HindIII, d)EcoRI, e)BglII, f)EcoRI/BglII, g)PstI, h)BamHI,i)PstI/BamHI, j)N.crassa A, k)N.crassa a. Lanes h and i contain insufficient DNAfor visible hybridization.2.9abcde^f^ghij^k^IFigure 17 N.crassa left flank sequences are contiguous with the aidiomorph in N.terricola. N.terricola genomic DNA was probed with 1.pMT2.519 Pstl / HindIll (1100bp) and 2. pCSN4 [Figures 6 and 8]. The bandpatterns match exactly indicating that the idiomorph and flank sequences arecontiguous in N.terricola. Lanes were digests with a)EcoRV, b)HindIII,c)EcoRV/HindlII, d)EcoRI, e)BgIII, f)EcoRI/B2III, g)PstI, h)BamHI,OPstl/BamHI, j)N.africana, k)N.crassa a, 1)N.crassaA.4 ID129IPabcdef ghijk lmnopqrs t2.9 • VA"'129MEP211■NPANI■IMI93 6.52.74111.Figure 18 Idiomorph and left flank linkage in Sordariaceae homothallics.Homothallic genomic DNA, digested with Group V enzymes, was probed with 1.pMT2.519 HindlII/PstI (1 100bp), 2. pCSN4, and 3. pMT- 150(G-2) [Figures 6-91. Threedigests (consecutive lanes) were performed with each species. Enzymes were BamHI,HindIII, Sad. Lanes were: a-c)A.sublineata, d-f)G.calospora, g-i)G.retiettlospora, j-1)G.142- 1, m-o)G.143 -4, p-r)G. S-23 , s)kb ladder, t)N.crassa a, u)N.crassa A. Bandpatterns are similar with probes 1 and 2 for all species with exception of G.retieulosporaG.S23. G.142-1 is inconclusive. Similar band patterns indicate contiguousness betweenflank sequence and idiomorph. Lanes without bands probably have insufficient DNA.46 -0 141^41310^GIPNIP0'a4 r,Figure 19 Idiomorph linked N.crassa right flank sequences are notpresent in homothallic Sordariaceae. More distal right flank sequences are,however, conserved. Homothallic DNAs were digested with PstI and probedwith 1. pMT 6.818 SphI/EcoRV (1.0kb) and 2. G16 C10 SphI/SphI (12kb).Lanes are as follows: a)A.sublineata, b)G.reticulospora, c)G.143-4, d)G.142-1,e)G.calospora, f)G.S-23, g)N.terricola,h)N.africana,i)N.crassa A, j)N.crassa a.The right flank sequence that is contiguous with an idiomorph in N.crassa is absentin homothallic species. The more distal N.crassa flank sequence is present in mostSordariaceae homothallics but not clearly linked to an idiomorph. Exceptions arelanes d and f which represent species that appear to lack the N.crassa distal rightsequence.12841a^b c de^f^g^h^ijklFigure 20 Distal right flank sequences are present in N.terricola but notclearly linked to an idiomorph. N.terricola genomic DNA was digested withGroup I enzymes and probed with the N.crassa G 16C 10 SphI/SphI probe (Figure11) representing a distal portion of the right flank region of N.crassa. Lanes aredigests with a)EcoRV, b)HindIII, c)EcoRV/HindIII, d)EcoRI, e)BglII,f)EcoR1/1301II, g)Pstl, ^ i)PstUBamHI, j)N.africana/ EcoRV,k)N.crassa a, 1)N.crassa A. The band pattern does not match that produced whenN.crassa A or a internal sequences were used as probes. Rather, the pattern seenhere is novel and indicates only that the N.crassa sequence is present in theN.terricola genome; Nothing can be determined with regard to the location of thesequence or its linkage to an idiomorph. The large number of bands in control(genomic) DNA lanes can be attributed to the size of the probe (-12kb) and itshybridization to sequences other than mating-type.DISCUSSION IThe well described mating-type locus of N.crassa is comprised of idiomorphs, Aor a, and sequences that flank the idiomorph to the left and right. The idiomorphs aredissimilar while the flank sequences are identical between mating-types. A N.crassahaploid nucleus invariably has single copies of idiomorph and flank sequences, occupyingthe same locus in A and a individuals.It was this researcher's prediction that characteristics of the tightly packagedmating-type locus of N. crassa might be conserved in Sordariaceae homothallics, enablingthe creation of mechanistic models to explain the evolution of one or more forms ofhomothallism. Though the A and a idiomorphs differ in N.crassa, the similarity betweenflank sequences helps define a mating-type locus as such.Left flank regionThe N.crassa left flank region is well conserved in the Sordariaceae. Every speciesexamined, hetero- and homothallic, has a sequence highly similar to the left flank region ofN.crassa. An idiomorph, A or A, and a conserved left flank sequence support the concept ofa single mating-type locus in both heterothallic and homothallic species. N.crassa A and aidiomorphs are both bordered by an identical left flank sequence. The same flank sequenceis contiguous with the a idiomorph in four of the six homothallic species examined. Theremaining two homothallics (G.reticulospora and G.S23) each contain a sequence similar tothe N.crassa left flank sequence but linkage is yet undetermined.It is interesting that the N.crassa left flank region is contiguous with the a idiomorphin most of the homothallic species tested. Such a result suggests that the two sequencesmay have acted as a unit in the evolution of a homothallic mating-type locus. It is alsopossible that A/a-type homothallics have resulted from a single mechanistic event in which aheterothallic ancestor underwent an unequal crossover at the mating-type locus. Acrossover in the left flank regions of A and a individuals might have occurred so as toisolate the A idiomorph from its left flank while leaving a left flank sequence connected with49the a idiomorph. The fact that a present day heterothallic (N.crassa) contains the same leftflank sequence in both A and a strains suggests a functional requirement for that region.The left flank present in A/a-type homothallics (and contiguous with the idiomorph) maybe present as an artifact while in heterothallics the same sequence has been retained forfunction in both A and a strains.Right flank regionThe N.crassa right flank region, represented by the 1.0 kb SphI/EcoRV probe[Figures 6, 7, 101 did not hybridize to genomic DNAs of any Sordariaceae homothallicexamined. This same N.crassa right flank sequence is not conserved in other heterothallicNeurosporas either (N.L.Glass, personal communication). This N.crassa flank region maybe absent in other species or substantially diverged but appears, in any case, to be highlyvariable in the Sordariaceae. This interpretation is supported by the recent work of TomRandall (unpublished results) who has sequenced a portion of the right flank region fromN.crassa. He has identified an ORF that contains a motif common to fungal pheromones.Furthermore, he has attempted hybridizations between various Sordariaceae homothallicgenomic DNAs and the N.crassa right flank region that contains the pheromone ORF. Inno instance did he observe hybridization between these DNAs indicating that this region ishighly divergent between species. Hybridizations were also performed between theN.crassa right flank region and genomic DNAs of heterothallic Neurospora species.Heterothallic DNAs also failed to hybridize to portions of the IV.crassa right flank,suggesting an extensive variability for this region in the Sordariaceae.Given these results, it is tempting to speculate that the right flank region is species-specific in the Sordariaceae and codes for distinct pheromones unique to individual taxa.Alternatively, heterothallics alone may have pheromone ORFs and functional pheromones.Homothallic species may simply have variable right flank sequences that are presentresidually and reflect prior evolutionary events. There is no direct evidence (isolation ofactual pheromone, for instance) that any homothallic species possess a pheromone ORF andthere is certainly no plausible reason why homothallics should have any need ofpheromones since they have no conidia, no trichogynes, and no observable matingmechanisms.N. terricola A / a linkage data suggest a novel homothallic mating-type locus While the conservation of left flanking sequences in heterothallic species ofNeurospora suggests a single heterothallic mating-type locus, the idea of a well definedhomothallic mating-type locus is not strongly supported by the data that describes thelinkage between N.terricola A and a idiomorphs. The A and a idiomorphs are not closelylinked in N.terricola. Yet the similar sizes (18kb) of N.terricola MluI and SmaI bands inhybridizations to N.crassa A and a idiomorph probes suggest that the N.terricolaidiomorphs may not be completely un- linked. Idiomorphs separated by 18 kb cannotstrictly be said to occupy the same locus. Indeed, preliminary CHEF gel analysis (MyronSmith, unpublished results) of whole N.terricola chromosomes indicate that A and aidiomorphs may be resident on different linkage groups altogether.The Neurospora homothallic mating-type locus clearly differs from heterothallic lociin the constitution of the right flank region. This is true for both A-type (N.africana) andA/a-type (N.terricola) homothallics. The (lack of close) linkage between the idiomorphs ofN.terricola cannot, however, be interpreted as reasonable grounds on which to distinguishthe homothallic from the heterothallic mating-type locus. No other Neurospora species canbe described in terms of idiomorph linkage since N.terricola alone contains bothidiomorphs. A Sordariaceae A/4-type homothallic mating-type locus is, however, areasonable concept. N.terricola is similar to selected Gelasinospora and Anixiella A/a-typehomothallics in harboring idiomorphs that appear to be unlinked. Homothallics in thesegenera are also alike in the constitution of their left flank regions. Four of sixGelasinospora and Anixiella species examined, like N.terricola, contain N.crassa left flanksequences that are contiguous with the a idiomorph. These data strongly suggest aconserved mating-type locus in Sordariaceae A/il-type homothallic species and arequirement of linkage between the left flank and the a idiomorph for functionalhomothallism. CHEF gel analysis of the linkage between A and a idiomorphs must becompleted, however, before the question of a homothallic mating-type locus and itsevolution can be resolved.Evolution of homothallic mating-types Idiomorph conservation and linkage data support the hypothesis that N.terricola andN.africana (a representative A-type homothallic) each evolved, monophyletically, from aheterothallic ancestor distinct from the ancestor that gave rise to N.crassa. Such distinctevolutions could account for the lack of hybridization between N.crassa right flanksequences and 1V.terricola and 1V.africana genomic DNAs. The conservation throughoutthe Neurospora genus of a single mating-type locus defined by left and right flank sequenceswould suggest a close evolutionary relatedness between N.africana and N.terricola. Such aclose relationship, however, is not supported by other data, namely the linkage between leftflank and idiomorph sequences in these species. Left flank sequences are contiguous withthe a idiomorph in N.terricola and the A idiomorph in N.africana. A linear relationshipbetween 1V.terricola and N.africana, one in which one species evolved directly into theother, is not likely since such an event would require the left flank sequence to be separatedfrom one idiomorph and re-linked to the opposite idiomorph. On the other hand, distinctevolutions from a common ancestor could easily associate a left flank sequence with eitheridiomorph. The fact that N.terricola lacks some 1.0 kb of the N.crassa A idiomorph alsodistinguishes N.terricola from N.africana which contains the A idiomorph in its entirety. Itis unlikely that N.africana would have evolved from N.terricola to regain the lost 1.0 kbfragment. N.africana must have preceded N.terricola evolutionarily if both species arectmonophyletic. Alternatively, each species itself may be monophyletic in which case noconclusions can be drawn about when the two species evolved.A mechanism for the evolution of homothallic mating-type lociGiven that N.terricola does contain some N.crassa right flank sequences, it is likelythat N.terricola and N.crassa descended from a species ancestral to all heterothallicNeurospora species. A mechanism can be proposed to account for the presence of both Aand a idiomorphs in single nuclei of N.terricola. The hypothesis posits the nuclearjuxtaposition of A and a idiomorphs as a result of meiotic non-disjunction. In a mutantmeiotic event, haploid genomes containing A and a could have in participated in cross-overevents but failed to dissociate. Instead of eight haploid meiotic progeny (the ascospores)resulting from normal disjunction, the result of such a mutant meiosis would be fourascospores, each harboring A and a idiomorphs. These idiomorphs could be resident in thesame nucleus or separate nuclei depending on the precise nature of the aberrant meiosis. Inthe preferred model [Figure 21], the idiomorphs are contiguous with no interveningsequences between them. Interestingly, the tetrasporic product of this mutant meiosis is anindividual much like N. tetrasperma, a species proposed as an evolutionary intermediate inthe General Discussion of this thesis. Raju (1978) has proposed that a single mutationcould result in the conversion of a tetrasporic state (N.tetrasperma) to an octasporic state(N.terricola et.al .) in Neurospora. Finally, the N.terricola A and a idiomorphs could havebeen separated (even to different linkage groups) by translocation events. The N.terricolaleft flank sequence could have been translocated as a unit with the a idiomorph, leaving theA idiomorph with a novel (undescribed) left flanking region. Figure 21 summarizes some ofthese mechanistic events.N.africana and the other A-type Neurospora homothallics can be reconciled easily asspecies in which the a idiomorph was lost altogether. The A idiomorph and contiguous leftflank region remain at the mating-type locus. Distal portions of the right flank may or maySLnot be present in these species and they may or may not have regions encoding pheromoneORFs. As comparatively little work has been done to describe Gelasinospora and Anixiellaheterothallic loci, it is not reasonable to extend the N.terri cola model to A/a-typehomothallics in these genera without reservation. Nevertheless, given the degree ofconservation between the left flank and idiomorph regions of these groups and those ofN.crassa, application of the N.terricola model is not altogether untenable.N.terricolaA-type homothallic(N.africana)pseudohomothallic :N.tetraspermaloss of a^I A )^ samelocus aIlr meiotic non-disjunctiona^I A translocation of aLG I ^LG ILG ILG I ?A LG IV  aIlr loss of amating-types^loci / linkage5-4heterothallic ancestorwith A and amating-typesLG IA samelocusor LG I ^ aFigure 21 Mechanistic model for the evolution of homothallic Neurosporaspecies. The mating-type locus of an ancestral heterothallic is essentially unchangedin existing heterothallics such as N.crassa. Cytogenetic events bring A and a nucleitogether into single ascospores or conidia and give rise to N.tetrasperma. Evolution topseudohomothallism is monophyletic. Meiotic non-disjunction brings A and A idio-morphs (and some flank sequences) together at the mating-type locus, yielding A/a-type homothallism such as that in N.terricola. Alternatively, non-homologous integrationcould put a on a different linkage group (IV). A-type homothallics, represented by N.afri-cana, arise after the loss of-specific sequences. These homothallics may diverge (greyarrows) from N.terricola or directly from the ancestral heterothallic.PART TWOSEQUENCE AND FUNCTIONALITY ANALYSIS OF N.TERRICOM ORF AINTRODUCTION IIThe presence of mating-type sequences in homothallic genomes is puzzling since self-fertile species have no mechanism for mating. It is likely that evolution in the Sordariaceaeproceeded from heterothallism to homothallism. If this is true, homothallic species maycontain mating-type sequences that are residual and no longer functional. Alternatively,homothallics may employ some but not all the functions associated with heterothallicNeurospora mating-type loci.N.crassa mating-type idiomorphs are known to be required for meiotic functions aswell as mating events. Homothallic Neurosporas may well require functional A and aidiomorphs for meiosis and the production of viable ascospores. Yet N.africana containsonly A-specific sequences and is capable of a robust sexual cycle. The a idiomorph may beentirely absent so that the A idiomorph alone is sufficient for sexual reproduction.Alternatively, N.af'ricana may contain an undescribed polypeptide (a diverged mt a-1 or anovel protein) that performs functions similar to those furnished by the a idiomorph; thisputative polypeptide may not be similar enough to N.crassa a to be detected by DNAhybridizations. N.africana A is functional when transformed into N.crassa sterile mutantsand a functional copy is required for the formation of perithecia (N.L.Glass, personalcommunication). A similar result is expected for N.terricola A. Based on DNAhybridizations, both the A and a idiomorphs are highly conserved from N.crassa and A islikely to be functional throughout the genus. Idiomorph and flank region similarity suggestsa close evolutionary relationship between Neurospora hetero- and homothallics.While it is relatively easy to hypothesize molecular genetic mechanisms that couldaccount for an evolution from heterothallic to homothallic species, mating-type idiomorphfunctionality is harder to interpret. If, for instance, the N.terricola a idiomorph is functionalSS1.6and necessary for sexual reproduction, then this species is likely to be monophyletic and notlinearly related to A-type homothallics. Other Neurospora homothallics are unlikely to havequickly gained reproductive independence from an ancestral species in which the aidiomoiph was functional. Alternatively, if the N.terricola a idiomorph is non-functional, aA-type homothallic such as N.africana could easily have lost a sequences and descendedfrom N.terricola in a single mutational event. A-type and a-type homothallics could belinearly related and represent members of a single phylogenetic line from a heterothallicancestral species.Evolutionary questions such as these can be made more tenable with specificsequence and functionality data. Part two of this thesis examines the sequence andfunctionality of A in N.terricola. Sequence data from the predicted ORF region of theidiomorph is compared with similar data from N.crassa (26,88) and N.africana (N.L.Glass,unpublished results). Comparisons of mt A-1 sequences are made between N.crassa,N.africana, and N.terricola. These comparisons are used in conjunction with data fromPart I of this thesis to speculate on the evolution of hetero- and homothallism inNeurospora. A preferred evolutionary model is proposed which positions heterothallicspecies ancestral to homothallics and pseudohomothallic species as intermediates.(Arguments will be presented to support an evolutionarily transitional status forpseudohomothallism). A final analysis speculates on evolutionary relationships in the wholeSordariaceae family.N.terricolaORF A pBC 1N. crassasterile mutantco-transformN.terricola genomic■111■•■■■■MII■115110■■•■PCRI cross5 "1a mating-type testerFigure 22 Strategy to test functionality of N.terricola ORF A .N.crassa oligonucleotide primers were used to amplify the N. terricola A ORF.Blunt ligation into a Bluescribe vector yielded an A clone which was cotransformedwith a benomyl resistance gene into sterile mutant spheroplasts. Cotransformantswere crossed with an opposite mating-type (a) reference strain and functionality wasassessed by observing for the development of perithecia.MATERIALS AND METHODS IIPCRA strategy for testing the functionality of N.terricola ORF A is presentedschematically in Figure 22. Oligonucleotide primers were chosen from consensus 1V.crassaA and a sequences (26,88) to span the entire ORFs of mt A-1 and mt a-1. 24 mer and25mer A primers were synthesized (Oligonucleotide Synthesis Laboratory, UBC) while19mer and 24mer a primers were generously donated by C. Staben.primer sequence^ mating-type1778^5' CCACCTTCACCCAAACTTCCCACC 3'^A3194 5' GGGTTACTGGAAGATGAGGTACCAT 3' AY694 5' GAGGTGATATCCTTGGTGACCGGG 3' aY759 5' GAATGCTATTCAGGGCCGG 3' aThe following reaction conditions were employed:A^aTemplate DNA^10Ong^10OngdNTPs^10mM 10mMMgC12 2.0mM^2.0mMPrimers 50g.tM^5011MTaq buffer*^lx lxTaq polymerase*^5u^5u*(Perkin Elmer-Cetus)ddH20^to 10011L^to 1001AL PCR template DNAs were N.terri cola and N.crassa genomic, the latter used as a control.DNAs were isolated in the same manner described in Materials and Methods of Part I ofthis thesis. Samples underwent the following cycle in a Perkin Elmer (model 480)thermocycler:1. 95°^90s2. 55°^80s3. 72°^180s4. repeat steps 1-3 5 times5. 95°^90s6. 55°^120s7. 72°^120s8. repeat step 5-7 20 times9. 95°^120s10. 72°^480sAmplification was confirmed by subjecting a fraction of the reaction mix to electrophoresisand visualizing the appropriately sized band. In this manner, the predicted 1.1 and 1.6 kbbands were seen in the A and a amplifications respectively. A second confirmation wasperformed by Southern hybridization analysis. A and a PCR products were subjected to gelelectrophoresis, transferred to Nylon membranes (Amersham Corp.), and hybridized toN.crassa A (pMT -150 (G-2)) and A (pCSN4) probes [Figures 6, 7] according to themethod described in Materials and Methods I of this thesis. Strong hybridizations wereobserved in both cases indicating that N.terricola mating-type sequences had beenamplified.Cloning N.terricola ORF A1. Purification of PCR fragments N.terricola A and a PCR products were electrophoresed and re-confirmed byvisualizing the appropriate sized bands. Slits were cut into the gel just distal to the PCRbands and the bands electroeluted into DE-81 (Stratagene Inc.) paper inserted into the slits.The paper was removed and centrifuged to spin off excess buffer. DNA, which binds to theDE-81 paper, was then extracted by the following method:1. place DE-81 paper in 0.8mL eppendorf tube nested in 1.5mL tube, theformer pierced at the bottom to allow flow through2. extract DNA by adding 201.th of 1.0M NaCl-saturated urea to DE-81paper3. incubate at 60° for 3 minutes4, spin at 12,000g for 3' to collect eluent5. repeat extraction two additional times6. add combined eluents to prepared 1.0 mL syringe Sephadex G-50column7. spin 6000g for 10' to collect eluent (only DNA passes through; ureabinds to the column)2. Blunt end ligation of APurified A PCR DNA was given blunt ends by the following reaction:PCR fragment^10OngdNTPs^20mMT4 buffer* 1 x^*(Pharmacia)T4 polymerase*^15uddH20^to 201.1LBlunt-ended DNA was then purified again by adding the entire T4 reaction mix to a 1.0mLsyringe Sephadex G-50 column and spinning at 6000g for 10 minutes.Blues cribe M13 (Stratagene) was chosen as the cloning vector and linearized withHindi. Vector DNA was then extracted once with 2:1:1 DNA(aq.):phenol:CHC13 andonce with 1:1 DNA(aq.):CHC13. Vector DNA was precipitated and re-suspended in TE.The following ligation reaction was then performed:A DNA from T4 reaction^10Ongvector DNA^500ngligation buffer* lxDNA ligase* 10u *(Stratagene)ddH20^ to 251ILincubate overnight at 16°C6 03. TransformationDH5-a E.coli cells were transformed by adding the entire 25pt ligation reaction to100RL of cells and placing on ice for one hour. Cells were then placed at 42° for threeminutes, iced for three additional minutes, and added to 0.9mL of LB broth. Afterincubation at 37° for one hour, 1004 of cells were plated (LB + 100pg/mL ampicillin)with 2Opt of X-gal and incubated overnight at 37°.4. Selection Putative transformants were visualized as white colonies in a background of bluecolonies after a 12-18 hour incubation. The bluescript vector contains a lac-Z gene for theproduction of 13-galactosidase which, in the presence of X-gal, constitutively produces ablue pigment when transcribed. The lac-Z promoter and lac-Z gene are separated by amultiple cloning site which ordinarily does not interfere with transcription. When afragment is ligated into the multiple cloning site, however, transcription is disrupted, no bluepigment is produced, and white colonies are formed. White colonies, then, representputative recipients of the insert or clones.In this manner, potential N.terricola A clones were screened. White colonies werepicked and cultured overnight in LB + ampicillin and DNA was extracted by alkaline lysis [amodification of the method of VoHiner and Yanofsky (95) by R.Subramaniam]. Enzymeswere chosen from those that cut in the multiple cloning site of the vector. By digestingputative clones with restriction enzymes that cleave on both ends of the polylinker, theinsert could be cut out of the vector. This fragment and the vector could then be visualizedas distinct bands on a gel, confirming success in producing a clone. The A clone was named"pNTA" (for "N.terricola A").Cloning N.terricola ORF aBlunt cloning was not successful with the PCR-amplified N.terricola ORF a, even afterrepeated attempts.Sequencing pNT ASequencing work of the N.terri cola A ORF (from pNT A) was performed by RajgopalSubramaniam and N. Louise Glass.Testing functionality of N.terricola ORF AThe strategy for testing functionality involved cotransformation of N.crassa sterile-mutant strains with a plasmid containing an antibiotic resistance gene p1:1C1 (for selection)and the N.terricola A ORF [Figure 22]. The pi3C1 plasmid contains an altered Neurospora13-tubulin gene that confers resistance to the antibiotic benomyl. Sterile mutant cells (seebelow) cotransformed with pf3C1 and an N.crassa mt A-1 were employed as positivecontrols. Cells that receive both genes (cotransformants) form colonies that will mate withtester strains of opposite mating-type. These colonies are not fertile and do not produceascospores; they do, however, form reproductive structures (perithecia) and in this waysignal the action of a functional gene. Cotransformations with pl3C1 and N.terricola DNAare expected to produce colonies that will mate and fruit in a similar manner if theN.terricola DNA includes a functional ORF.The following materials were used for cotransformations:AN.terri cola DNA^pNTAantibiotic resistance gene pl3C1recipient strain^Am56Transformants were selected by growth on media containing 0.514/mL benomyl.Recipient cells were N.crassa spheroplasts (prepared by N.L.Glass) of Am56 strains (33)that harbor a single frame shift mutation in mt A-1, effectively rendering them sterile.Growth on media with benomyl ensures that all transformant colonies have received thebenomyl resistance gene but there is no direct evidence that spheroplasts have also received4 2_N.terricola mating-type DNA. Rather, cotransformation (the receipt of both DNAs) ispredicted to occur if the ratio of DNA copies (pf3C1:pNTA) is in the range of 1:2 to 1:2.5.Confirmation of the receipt of pNTA can be confirmed if it is unequivocally shown to confermating ability to the sterile Am56 cells. Alternatively, receipt of mating-type DNA can beconfirmed by isolating DNA from the cotransformant colony and hybridizing the DNA toN.terri cola mating-type DNA.Cotransformations were performed according to the following protocol (96):1. aliquot 50111., spheroplasts into eppendorf tubes on ice.2. add l[tg pf3C1 DNA + 2.51.tg pNTA DNA; incubate 30' on ice.3. add 1.0mL 40% PEG / 500mM MOPS pH 8.0 / 50mM CaCl24. incubate 30' RT5. add cells to prepared 8.0mL aliquots of regeneration top agar6. add entire mixture to prepared plates (bottom agar + 0.514/mL benomyl)7. incubate at RT until cotransformant colonies become visible (3-5 days) Putative pNTA / p13C1 cotransformant colonies were allowed to grow until vegetativetissue breached the surface of the agar. Sterile disks of Whatman 3 filter paper were appliedto the surface of the agar so as to cover all visible colonies and colonies were allowed togrow into and inoculate the disks (approximately 24 hours). The filter paper was thenremoved and transferred to prepared plates with established growth of an opposite mating-type tester strain, in this case a fluffy (fl) mutant of A (FGSC strain 4347). The fl mutant isan aconidial female receptor strain with enhanced fertility; fl strains have the phenotype ofcolonial growth in dense mats that facilitates use in mating-type test crosses. Inoculatedfilter paper disks were allowed to fertilize a fl A plate whereupon reproductive structures(perithecia) would become visible in two to five days.Control cotransfonnations were performed concurrently with N.terri cola mating-typecotransformations. All procedures were identical to those described for N.terricola butwith transforming DNA substitutions or omissions. Positive controls werecotransformations with N.crassa PMT-150(G-2) mt A-1. Negative controls weretitransformations only with pi3C1 DNA; transformation without mating-type DNA wasexpected to result in transformant colonies but no perithecial development upon crossing toan opposite mating-type strain.Testing "escape" from homothallism in N.terricolaOne hundred individual ascospores were picked, heat shocked (60°C, 30 minutes),and allowed to germinate on minimal media (modified Westergaard's salts + 2% fructose).Individual slants were observed first for germination and mycelial growth and then for thedevelopment of perithecia. Perithecia were in turn observed to see if they shot ascospores.Escapes were considered to be those slants that demonstrated vegetative growth but nodevelopment of perithecia.RESULTS IISequence analysis of N.terricola, N.africana, and N.crassa A ORFs Results from nucleotide sequence comparisons of N.crassa, N.africana, andN.terricola A ORFs are presented in Figure 23. The carboxy and amino terminal regions ofall three sequences are quite similar: 91% similarity in the amino terminal region and 86-87% similarity in the carboxy terminal region for all three species. Intron sequences areconsiderably more variable. The N.terricola intron is 86% similar to the N.crassa intronwhile the N.africana intron is only 77% similar to the N.crassa intron. The N.terricola andN.africana introns are 90% similar to one another.Functionality of N.terricola ORF AResults of functionality tests are presented in Table 4 and Figure 24. N. terricolaORF A was considered functional if cotransformation into N.crassa spheroplasts resulted inthe growth of colonies that could mate with an opposite mating-type tester strain (seeMaterials and Methods II). Positive mating was assayed by observing for the developmentof brown or black perithecia at the sites of cotransformant colonies. Negative controls weretransformations with pl3C1 [selection] DNA and no mating-type DNA. Positive controlswere cotransformations with pMT-150(G-2) [N.crassa A] or pCSN4 [N.crassa a] andCl DNA. Six replicate negative control transformations with pf2C1 only yielded a total of14 colonies, none of which mated with A or 2 mating-type tester strains. Six replicatepositive control cotransformations with pr3C1 and pMT-150(G-2) yielded a total of 78colonies, 48 of which mated with a tester strains. Nine replicate cotransformations withpNTA [N.terricola A] and prdC1 DNA yielded a total of 128 colonies, 40 of which matedwith a tester strains [Table 4]. No difference was visually discernible between peritheciafrom positive control and N.terricola cotransformations [Figure 24]; black perithecia wereobserved for both control and experimental cotransformations but ascospores were notN.terricolaN.crassaN.africanaintron 91%iii&86% 87%i&introna'91% 77%i&l'86%introna'96%!&90%^92%i&^i& I ammo ^1^ carboxy IN.terricola intronFigure 23 Graphic representation of DNA sequence similaritybetween the A ORFs of N.crassa,N.terricola, and N.africana.Percentages are percent similarities between DNA sequences in the regionsindicated.ChTable 4 Results of cotransformations of Am56 spheroplasts with pl3C1 andN.terricola A mating-type DNAs.transformingDNA(s)replicates (co)transformantcoloniescolonies thatmatedMT testerstrainpi3C1 6 14 0 fl aP3C1pMT-150(G-2)6 78 48 fl ap pc1pNTA9 128 40 fl aC,1GSFigure 24 Results from cotransformations of Am56 spheroplasts withpl3C1 (selection) and mating-type DNAs. 1. Transformation with no DNA 2.Transformation with pi3C1 only 3. Cotransformation with pCSN4 (N,crassa mtA-1) and p33C1 4. Cotransformation with pNTA (N.terricola ORF A) and pi3C I.In each case, (co)transformant colonies were crossed with a tester strain of theopposite mating-type (fl a). The development of prominent black perithecia wasinterpreted as a positive mating reaction. No difference is observable between theperithecia resulting from positive control (pCSN4) and experimental (pNTA)cotransformations. The N.terricola ORF A, therefore, confers full functionality(mating ability) on the recipient sterile spheroplasts.produced in either case. Ascospore formation was not expected since N.crassa sterilemutants have been characterized by the irrecoverable loss of certain meiotic functionsincluding the ability to generate ascospores. Receipt of transforming mating-type DNAdoes not restore the ability of Am56 mutants to form ascospores (N.L. Glass, personalcommunication). Ascospore formation aside, mating was interpreted to be equallyvigorous in control and experimental cotransformations and the N.terricola A ORF wasconfirmed to confer full functionality to the recipient mutant spheroplasts. Though the AORF has not been proved to function in N.terricola, cotransformation results suggest that Acould function in N.terricola.As a negative control, a second set of matings were attempted betweencotransformant colonies and fl strains of the same (A) mating-type. In this case, the same128 putative cotransformant colonies (described previously) did not mate or yieldperithecia. The N.terricola A ORF, then, acts authentically as a mating-type gene andinteracts only with strains of opposite mating-type."Escape" from homothallism in N.terricolaA test was devised to determine whether homothallism is stable in N.terricola.Experimentally, cultures of N.terricola are invariably homothallic; mycelial growth of asingle culture always leads spontaneously to the formation of perithecia and the forcibledischarge of viable ascospores. Convenience dictates that cultures be grown from multipleinocula, ascospores or vegetative hyphae. The possibility exists that cultures initiated in thismanner may be the result of crossing between propagules that contain an A or a idiomorphbut not both. It is unclear whether N.terricola can lose a mating-type and becomefunctionally heterothallic. Even if the N.terri cola mating-type idiomorphs are notfunctional, the question of homothallic stability remains. Is N.terricola constitutivelyhomothallic or can individuals lose the ability to self?9An examination of colonies grown from single propagules suggests that N.terricolamating could rarely if ever be occurring as the result of crossing between functionallyheterothallic individuals. 100 individual N.terricola ascospores were picked and heatshocked to induce germination. Of these, 87 germinated and grew into normal mycelialcolonies. After 4-7 days of growth, all 87 colonies formed visible brown or blackperithecia. Each of these 87 went on to shoot ascospores. As a result, it was concludedthat N. terricola is constitutively homothallic; in no instance was functional heterothallismobserved.10I fDISCUSSION HThe strong hybridizations between N.crassa and N.terricola mating-type DNAssuggest a high degree of similarity. This fact was utilized in formulating a strategy toascertain whether the A ORF of N.terricola can confer mating activity to sterile mutants ofN.crassa. A PCR amplification of N.terricola A and a ORFs was predicted to be possiblewith the use of N.crassa oligonucleotide primers. Homology between the genomes ofN.crassa and N.terricola means that N.crassa primers could effectively bind N.terricolagenomic templates and allow a high yield PCR amplification. The successful amplificationby PCR of the N.terricola ORFs A and a supports the predicted similarity betweenN.terri cola and N.crassa and provides a tool with which to address the question ofidiomorph functionality in N.terricola. (A discussion of functionality results will bepresented later). Two approaches to the functionality problem were possible withN.terricola PCR products; the ORF fragments could be cloned and sequenced or sequenceddirectly. Subsequently, PCR product or whole ORF containing plasmids could betransformed into N.crassa sterile mutants.Cloning was chosen as the preferable option but came to fruition only with theN.terricola A ORF . Blunt cloning is difficult even under ideal conditions and perhaps therapid cloning of A was effected in part by luck. Repeated attempts to do a similar ligationof a into the same vector were unsuccessful despite countless strategies and technicalmodifications. Failure to clone a, while frustrating, cannot be construed as biologicallysignificant as ORF a appears to have just as much similarity as ORF A to the correspondingN.crassa sequence. Nevertheless, cloning difficulties did not preclude sequencing andfunctionality tests with the a PCR product.Monophylogeny Versus Polyphylogeny in N.crassa. N.africana, and N.terricolaSequence analyses [Figure 23] of N.crassa, N.africana, and N.terricola A ORFsdemonstrate that the intron DNA sequence of N.terricola is quite similar (86%) to that of1Z.N.crassa. The N.africana intron is less similar (77%) to the N.crassa intron but the twohomothallic introns (N.terricola and N.africana) are very similar (90%). In the followinganalyses, it is assumed that intron sequence similarity is a direct measure of temporalevolution, dissimilar sequences having diverged earlier than similar sequences. With thatassumption, the A intron sequence data immediately suggest several possible relationshipsbetween the species. N. terricola and N. africana are closely aligned and more similar toeach other than either is to N.crassa. Comparing N.terricola and N.africana, the former isevolutionarily closer to (its intron sequence is more similar to) N. crassa. If heterothallism isancestral, N.africana, N.terricola, and N.crassa may be members of one phyletic group inwhich a heterothallic type, as typified by N.crassa, arose first, followed by the juxtapositionof A and a idiomorphs in a single nucleus and the evolution of N.terricola. Finally, the lossof A sequences marked the emergence of N.africana.Monophylogeny could also be described by evolutionary events in whichhomothallism arose twice. In such a history, a heterothallic ancestor (different fromN.crassa) evolved into N.terricola first, maintaining the presence of both A and aidiomorphs in a single individual. The separation of idiomorphs into distinct individualscould have given rise to N.crassa followed by cytogenetic events that allowed the aindividual to become self-fertile (N. africana). Though mechanistically feasible, this doubleevolution of homothallism is counter-intuitive. It is unlikely that the heterothallic ability toconidiate would be lost in the evolution of N. terricola only to be gained again in N. crassaand lost a second time in N.africana. Such an oscillation between conidial heterothallismand aconidial homothallism is difficult to reconcile unless selective pressures wereparticularly strong. There is, however, no evidence in Neurospora to suggest that onemating strategy is significantly more successful than the other (68).13A case against polyphylogenyA second interpretation of the A intron sequence data would be that N.africana andN.terricola each represent a monophyletic evolution from a heterothallic Neurosporaancestor as typified by N.crassa. [It is assumed that N.crassa has evolved directly, and withrelatively few changes, from the heterothallic ancestor]. In this history, N.africana andN.terricola represent individual divergences from a continuous heterothallic lineage andeach has membership in a distinct homothallic lineage. Such a history is stronglycontradicted by the intron sequence data, however. The high degree of similarity (90%)between the introns of N.terricola and N.africana strongly refute the hypothesis that thesespecies diverged long ago. Rather, intron similarity between them suggests that they aremembers of a single phyletic group and have diverged only recently.All these assignations to phyletic groups and evolutionary models must be consideredwith the caveats that a) the supporting or refuting data come from a single locus and b)sequence data were derived from single isolates of each species. Only single isolates areavailable for N.terricola and N.africana. In positing hypotheses for the evolution ofNeurospora (and other Sordariaceae) mating strategies, we are likely merely scratching thesurface of what is a multi-faceted phenomenon. Realistically, homo- and heterothallism mayhave evolved numerous times in Neurospora. Many transitional species must now be lostand we may never be able to describe the entire evolutionary history of Neurospora. At thesame time, in the analysis of present day hetero- and homothallic species, we may have awindow on one-time evolutionary events that reflect the only instances in whichhomothallism has evolved in Neurospora.Evolution in terms of functionality at the mating-type locus The evolution of Neurospora homothallism can be considered in the context offunctionality at the mating-type locus. Presumably, the A and a idiomorphs of an ancestralheterothallic had both mating and meiotic functions as they do in the present day N.crassa.IN. intermediaN.sitophilaN. tetraspermaIIIIIN. crassaN. africanaN. terricolaN.tetraspermaN.crassaN. africanaN.terricolaN.tetraspermaFigure 25Three cladistic trees for evolution in the genus Neurospora. All trees are unrooted. Tree I isadapted fromTaylor and Natvig (1989) and reflects nuclear and mtDNA RFLP data that showN.tetrasperma and N.sitophila to be monophyletic. N.tetrasperma RFLP patterns are the mostdissimilar to N.crassa, indicating an early divergence. Trees II and III are possible histories forthe evolution of homothallic species, based primarily on N.crassa, N.africana, and N.terricolasequence data. Greater sequence similarity is interpreted to indicate a more recent divergence.Tree III is more feasible than tree II since N.africana mating-type sequences are closer to N.crassathan are those of N.terricola.The fact that the N.terricola A ORF confers functionality in N.crassa sterile mutants is to beexpected if N.terri cola inherited essentially intact (and functional) idiomorphs directly froma heterothallic ancestor as typified by N.crassa. The N.terricola A ORE, however, confersmating ability to N.crassa sterile mutants but does not confer meiotic ability. One mightexpect the reverse situation where a homothallic ORE would confer meiotic functions(useful in producing resistant ascospores for propagation) but not the irrelevant matingfunctions. It is necessary to point out here that N.crassa mating-type DNA (pMT -150 0-2), transformed into the same sterile mutants, also confers only mating functions. Theinability of transformed Am56 mutants to produce ascospores probably reflects uniqueproperties of the mutants themselves and not the wild type functionality of the N.terricolaand N.crassa A ORFs. Nonetheless, the fact that the N.terricola A ORE is largelyconserved from N.crassa and can indeed function in N.crassa mutants suggests thatevolution has proceeded from heterothallism to homothallism. Why would a homothallicorganism that does not mate develop a polypeptide that can function in elaborate matingprocesses? A more plausible history describes the presence of functional mating-types inN.terri cola as residual, the polypeptide retained for its meiotic functions.An obvious question is whether A (and A) function in N.terricola and whether theirfunction is meiotic, mating-related, or both. [This question could be answered by thegeneration of N.terricola sterile mutants and their transformation with N.terricola wild-typeA and a ORFs]. An even more intriguing question is whether N.terri cola sterile mutants,transformed with A and a ORFs, could be taken through a cross (A transformant x atransformant) and produce fertile meiotic progeny. If so, the argument could be made thatNeurospora homothallism is not fundamentally different from heterothallism at least at thelevel of mechanisms and molecular interactions at the mating-type locus. N.terricolahomothallism may retain all the molecular requirements for heterothallism (two mating-types that can function) but not require a heterothallic mating strategy. Rather the loss (orrepression) of certain genes (conidiation genes that contribute to chemotropic/ pheromonal15-mating interactions, for instance) may have coincided with the evolution of species that areself-reproductive. Finally, such an evolution may have reflected selective pressureswhereupon environmental conditions were unfavorable for vigorous vegetative growth andconidiation and mates were hard to come by. The predominance of homothallicSordariaceae in soil suggests the occupation of a niche where fitness is related to the abilityto self (independence from mating), slower growth, and the production of hardy ascospores.Evolution in Gelasinospora and AnixiellaThe degree to which the A and a idiomorphs of Neurospora are conserved inGelasinospora and Anixiella suggests an evolutionary relatedness between these genera.Though no idiomorph and flank sequence data are available for species of Gelasinosporaand Anixiella, it is compelling nonetheless to consider the evolution of homothallism in thewhole Sordariaceae family and the possibility that the root heterothallic species we haveconsidered may be ancestral not only to Neurospora homothallics but to Gelasinospora andAnixiella homothallics as well. An ancestral Sordariaceae heterothallic may havedeveloped different ascospore ornamentations independently from the evolution of mating-types which have remained largely conserved in Neurospora, Gelasinospora and Anixiella.In Gelasinospora and Anixiella, A/a-types constitute the majority of homothallics while inNeurospora, A-types are more common. This fact suggests that homothallism has evolvedseparately in Neurospora and the other two genera though possibly from the same (or asimilar) ancestor. It is possible that A/a-type and A-type homothallism are significantlydifferent forms of a reproductive strategy that in the former requires two functionalidiomorphs and in the latter only one. This interpretation contrasts that in which thepresence of an a idiomorph in A/a-type homothallics is residual (reflecting priorevolutionary events) and the a idiomorph is not functional. In either case, the conservationof A and a idiomorphs and (left) flank sequences seems to be more important than thearrangement of these elements at the mating-type locus(i). The fact that the left flank regionis conserved in all the A/a-type homothallics examined suggests a functional role for thissequence. At the same time, the variability in that sequence's linkage to an idiomorph(linked to a in G.calospora et.al.; inconclusive in G.reticulospora or G.142-1, for instance)suggests that an A/a-type homothallic locus may have evolved more than once, even withina genus.GENERAL DISCUSSIONEvolution of homo- and heterothallismInnumerable models can be formulated to explain the evolution of homothallism andheterothallism. A key question is whether heterothallism preceded homothallism or thereverse. This writer favors the hypothesis that heterothallism evolved first followed by oneor several events giving rise to homothallic populations. Intuitively, it seems improbablethat self-fertile populations would undergo the relatively expensive evolution of mating-types and reproductive structures without strong selective pressure. Clearly there areselective advantages conferred by aspects of sexual reproduction in Neurospora but arethese advantageous enough to envision the evolution of heterothallism from homothallicpopulations? Alternatively and more probably, heterothallism arose first in order to meetthe need for what are now considered the numerous advantages to sexual reproduction.These advantages might include: i) ability to outcross --use of meiosis for geneticrecombination and creation of new and potentially advantageous gene combinations ii) useof meiosis for repair of DNA copied during cell division iii) production of ascospores whichare resistant to desiccation and therefore effective as propagules over time and distance(68,69). If heterothallism evolved to meet one or more of these needs, it is easy to see howthe same advantages could be useful to homothallic descendants. Advantages conferred byoutcrossing are irrelevant in self-fertilizing homothallics but meiosis for repair or theproduction of hardy ascospores could be valuable to homothallics.1 1Several approaches have been taken in the attempt to develop an evolutionary historyfor Neurospora. Some workers have used mitochondrial DNA (mtDNA) length mutationsand random fragment hybridization analysis of nuclear DNA to describe the relationshipsbetween heterothallic species in the genus (56,90,91). Though no such analyses haveincluded constitutively homothallic species, data from homothallic species can neverthelessbe merged with heterothallic cladistic trees.Random fragment length polymorphisms in nuclear DNA have been analyzed inheterothallic (56) and homothallic (29) groupings of Neurospora species. Natvig et.al . (56)have detected the existence of distinct clusters of heterothallic isolates that have commonRFLP patterns when digested with the same restriction enzyme(s). Analysis of theseclusters lends credence to the hypothesis that N.tetrasperma and N.sitophila are eachmonophyletic. Conversely, N.crassa and N.intermedia cannot form monophyletic groupsbased on comparison of RFLP patterns. Instead they constitute a phyletic branch tothemselves. In support of these hypotheses, these same data support established speciesconcepts based on mating. N.tetrasperma and N.sitophila are not capable of crossing withother Neurospora species. N.crassa and N.intermedia, however, produce a significantpercentage of fertile ascospores when crossed (67).N.tetrasperma is believed to have diverged first from an ancestral Neurosporaheterothallic because its nuclear RFLP patterns are the most dissimilar to patterns fromother Neurospora species. N.sitophila is believed to represent the second oldest divergencewhile the remaining Neurospora heterothallics form a third phylogenetic group thatdiverged more recently. [This temporal analysis makes no accommodation for the evolutionof true homothallic species]. This hypothesis has since been affirmed by mtDNA lengthmutation data from the same species (91). The clusters that emerge from mtDNA analysesare consistent with those obtained from nuclear RFLP analyses. Cladistic trees forheterothallics, though, are unrooted and not based on specific knowledge of ancestralNeurospora species. For this reason, it is impossible to determine which divergences78occurred earliest. Relationships are translated to histories on the basis of characters andtheir similarities between species. Groups possessing the most character diversity arethought to represent the oldest divisions. Indeed, the greatest RFLP variability is seenamong isolates of N.crassa and N.intermedia. N.tetrasperma is intermediate andN.sitophila has the least intra-specific diversity.Similar data are available for the analysis of evolutionary relationships betweenhomothallic species of Neurospora. Glass et.al . (29) compared RFLP patterns ofNeurospora and other Sordariaceae isolates. Digests of genomic DNAs were hybridized torandom N.crassa cosmid probes and to N.crassa mating-type probes. Several groupingsemerged from the data. Within the Neurospora genus, all the A-type homothallics clusteredin a single group; their RFLP patterns were very similar. N.terricola yielded unique patternswhen probed with random and mating-type probes. This suggests that A-type and A/a-types represent distinct monophyletic groups. The lack of variability within the A-typecluster also engenders the possibility that members of this grouping are actually a singlespecies (29).Evolutionary models Nuclear RFLP (56) and mtDNA length mutation (90,91) data for heterothallicspecies can be considered in the context of homothallic RFLP (29) data and the sequencedata presented in this thesis. A cladistic tree can then be proposed that considers theevolution of both homothallic and heterothallic Neurospora species [Figure 25]. In thetrees presented, N.tetrasperma, N.terricola, and A-type homothallics may each representmonophyletic groups and mark separate divergences into functional homothallism.Alternatively, there may have been fewer than three distinct evolutions to functionalhomothallism; at least two groups of homothallics may be linearly related, one representingan intermediate form ancestral to the other. This latter scenario is not likely because ofcytogenetic and molecular differences between groups. N.tetrasperma is thought to beBOmonophyletic because its RFLP patterns cluster and because it is the only Neurosporaspecies that is four spored. Similar arguments can be made for the monophylogeny ofN.terricola. N.terricola is unique in possessing both mating-types in a single nucleus.RFLP patterns are dissimilar to those of any other homothallic grouping (29) and onlyN.terricola has ascospores with a single germ pore (3,32). Finally, N.africana RFLPpatterns cluster with those of the other A-type homothallics; no similarity is observed topatterns from any other cluster.Evolutionary histories based on RFLP analyses may seem counter-intuitive whenmore variability is observed within a species (N.crassa) than within entire groups (A-typehomothallics) (56). In this case, however, differences between heterothallic and homothallicreproduction must be considered. The fact that homothallic populations are almost entirelyinbreeding means that there is little opportunity for genetic variability to arise. Because ofdifferences in genetic input, homothallic and heterothallic speciations cannot be temporallygrouped on the basis of RFLP patterns.Nauta and Hoekstra (57,58) have applied a population genetic model to weigh theconditions under which heterothallism might have evolved into homothallism and vice versa.The model is applied to theoretical populations in which there are varying extents of selfingand variably deleterious effects realized by selfing. In all the scenarios, the conditions forhomothallism to invade heterothallic populations are much more easily realized thanconditions for heterothallism to invade homothallic populations. The fact that mosthomothallic Sordariaceae do not form conidia is significant. This means that outcrossingcan occur only by occasional mycelial contact and only then if there is no conflict invegetative incompatibility types. Selfing for homothallics then approaches 100%. Severeinbreeding depression could drive the evolution of heterothallism from homothallicpopulations if outcrossing conferred a substantial fitness advantage over selfing. In the caseof haploid organisms such as Neurospora, however, selfing is not especially deleterious.Neurospora inbreeding is actually a form of intragametophytic selfing. Such selfing doesnot imply any recombination and is effectively equivalent to asexual reproduction. Thedisadvantages to inbreeding common to diploids reflect the combination of deleteriousrecessive alleles and these disadvantages do not apply to haploid, intragametophytic selfers.Indeed, populations from nature show homothallism to be no less successful a reproductivestrategy than heterothallism (68).The Nauta and Hoekstra model also predicts that when evolution from hetero- tohomothallism (or the reverse) occurs, one should also expect to find polymorphicpopulations in which both systems exist in the same species. Interestingly, most of theknown polymorphic species also produce conidia. Notable examples include Nectriahaematococca, Gibberella zeae, and Glomerella cingulata (58,68). Perhaps these speciesare transitional and represent intermediates in the evolution from hetero- to homothallism.The fact that Sordariaceae homothallics are neither polymorphic nor conidial may meanthat an evolutionary event is complete and that intermediate species, now extinct or not yetisolated, had lesser fitness. Nauta and Hoekstra's model considers the question ofintermediates. Since intragametophytic selfing is not significantly deleterious for haploids,homothallism is expected to eclipse heterothallism in natural populations. In their words,"the only explanation for the existence of heterothallic Sordariaceae seems to be that insome cases the fitness threshold for intermediate stages is too high" (58). The evolutionfrom hetero- to homothallism is not likely to occur suddenly and in a single mutationalevent. Rather, intermediate polymorphic species are predicted. In many cases thesepolymorphic strains are stable (G. zeae, N. haematococca, G.cingulata) but significantnumbers are not stable and presumably remain homothallic.Conidial homothallics are not observed in the Sordariaceae. Nevertheless, the lack ofconidia in Sordariaceae homothallics does not preclude the possibility that lost intermediatespecies were both conidial and homothallic. Indeed, the pseudohomothallic N.tetraspermadoes form conidia (72) suggesting an intermediate status for this species in the evolutionfrom hetero- to homothallism. [N.tetrasperma's evolution was previously considered onlyBI-with respect to heterothallism]. It is easy to identify N.tetrasperma characters that fit theidea of intermediacy. Sexual reproduction can occur by inbreeding or by outcrossing (72).Both mating-types are present in single ascospores and conidia, bestowing functionalhomothallism. At the same time, the fact that A and a mating-types occupy separate nucleiimplies a close alignment with heterothallic mating mechanisms.If indeed conidia and microconidia serve in Neurospora for fertilization in addition topropagation, then the loss of conidiation in true homothallics is not surprising. N.terricolaand the A-type species can be thought of as constitutive homothallics because they have lostthe ability to outcross; they have no need for conidia as fertilizing agents. At the same time,conidia themselves are very successful as propagules. Perhaps the evolution ofhomothallism is coincidental with the loss of ability to conidiate. Homothallics would thenbe disadvantaged as compared to conidiating heterothallics but the ability to self and theindependence from mating could compensate. Theoretically, it should still be possible forN.terricola to evolve to heterothallism by the unlikely restoration of conidiation function. Asimilar evolution in N.africana is even more unlikely because it would also require theevolution of a functional a idiomorph. Alternatively, N.africana may already harbor asecond idiomorph that functions like a but is not detectable by hybridization to N.crassa a.Future WorkThe mating-type locus of N.crassa is largely conserved in homothallic members of theSordariaceae. The majority of the N.crassa A and a idiomorphs and the left flankingsequence are present in the genomes of A/a-type homothallic species of Neurospora,Gelasinospora, and Anixiella. The idiomorphs are not closely linked in any of these generathough the a idiomorph is linked to the left flank sequence in most cases. The right flanksequence is diverged or absent in the same species. While the conservation of theidiomorphs themselves is well established, their linkage is not well described. Reasonablemechanistic models for the evolution of a homothallic mating-type locus will require themapping of A and a idiomorphs to a specific locus(i). This result could be obtained byCHEF gel analysis and the localization of idiomorphs to linkage groups. Presumably, theassociation of A and a idiomorphs in single nuclei could then be attributed to crossover(idiomorphs on same linkage group) or translocation (idiomorphs on different linkagegroups) events.The N.terri cola A ORF confers functionality when transformed into N.crassa sterilemutants. A similar result is expected for the N.terricola a ORF but awaits confirmation. Ofeven more interest would be the functionality in N.terricola of these ORFs. It is easy toconceive of an evolution in which the idiomorphs of a heterothallic ancestor are presentresidually in descendant homothallic species; their functionality upon transformation intoN.crassa sterile mutants could also be residual. A key question is whether an N.terricolaindividual with a mutation in the ORF of A or a (or both) can still enter the sexual cycle andgenerate viable meiotic progeny. If sexual functions were lost, would the transformation ofthis mutant with an N.terricola mating-type ORF(s) restore functionality and/or ascosporeviability? The answers to these questions would hinge on the development of an N.terri colasterile mutant, the generation of competent spheroplasts of that mutant, and its successfultransformation by Neurospora DNA.REFERENCES1. Akins, R.A. and Lambowitz, A.L. (1985) General method for cloning Neurosporacrassa nuclear genes by complementation of mutants. Molecular and CellularBiology 5: 2272-2278.2. Austin, B. and Tyler, B.M. (1990) Strategies for high-efficiency cotransformation ofNeurospora crassa. Experimental Mycology 14: 9-17.3. Austin, W.L., Frederick, L. and Roth, I.L. (1974) Scanning electron microscopestudies on ascospores of homothallic species of Neurospora. Mycologia 66: 130-138.4. Banuett, F. and Herskowitz, I. (1989) Different a alleles of Ustilago maydis arenecessary for maintenance of filamentous growth but not for meiosis.Proc.Natl.Acad.Sci. USA 86: 5878-5882.5. Barrett, J. (1987). Molecular variation and evolution. from Evolutionary Biology ofthe Fungi eds. A.D.M. Rayner, C.M. Brasier and D. Moore. Symposium of theBritish Mycological Society. pp. 83-94.6. Beadle, G.W. and Tatum, E.L. (1941) Genetic control of biochemical reactions inNeurospora. Proc.Natl.Acad.Sci.USA 27: 499-506.7. Berlin, V. and Yanofsky, C. (1985) Protein changes during the asexual cycle ofNeurospora crassa. Molecular and Cellular Biology Vol. 5, No. 4: 839-848.8. Berlin, V. and Yanofsky, C. (1985) Isolation and characterization of genesdifferentially expressed during conidiation of Neurospora crassa. Molecular andCellular Biology Vol. 5, No. 4: 849-855.9. Bistis, G.N. (1981) Chemotropic interactions between trichogynes and conidia ofopposite mating-type in Neurospora crassa. Mycologia 73: 959-975.10. Cain, R.F. (1950) Studies of coprophilus ascomycetes. Canadian Journal of ResearchVol. 28 Sec. C: 566-576.11. Cavalier-Smith, T. (1987) The origin of fungi and pseudofungi. from EvolutionaryBiology of the Fungi eds. A.D.M. Rayner, C.M. Brasier and D. Moore. Symposium ofthe British Mycological Society. pp. 340-353.12. Colson, B. (1934) The cytology and morphology of N.tetrasperma Dodge.Ann. Bot. (London) 48: 211-224.13. DeLange, A.M. and Griffiths, A.J.F. (1975) Escape from mating-type incompatibilityin bisexual (A + a) Neurospora heterokaryons. Can.J.Genet.Cytol. 17: 441-449.14. Dodge, B.O. (1935) The mechanics of sexual reproduction in Neurospora.Mycologia 27: 418-438.15. Dutta, S.K. (1976) DNA homologies among heterothallic species of Neurospora.Mycologia 68: 388-401.16. Dutta, S.K., Sheikh, I., Choppala, J., Aulakh, G.S., and Nelson, W.H. (1976) DNAhomologies among homothallic, pseudo-homothallic, and heterothallic species ofNeurospora. Molec.Gen.Genet. 147: 325-330.17. Ebert, P.R., Anderson, A., Bernatzky, R., Altschuler, M. and Clarke, A.E. (1989)Genetic polymorphism of self-incompatibility in flowering plants. Cell 56: 255-262.18. Esser, C. (1971) Breeding systems in fungi and their significance for geneticrecombination. Molecular and General Genetics 110: 86-100.19. Fincham, J., Day, P., and Radford, A. (1979) Fungal Genetics. Blackwell ScientificPublications, London.20. Fox, R.D. and Wong, G.J. (1989) Homothallism and heterothallism in Tremellafuciformis. Canadian Journal of Botany 68: 107-111.21. Frederick, L., Uecker, F.A. and Benjamin, C.R. (1969) A new species of Neurosporafrom soil of West Pakistan. Mycologia 61: 1077-1084.22. Frohman, M.A., Dush, M.K. and Martin, G.R. (1988) Rapid production of full lengthcDNAs from rare transcripts: amplification using a single gene-specificoligonucleotide primer. Proc.Natl.Acad.Sci. USA 85:8998-9002.23. Furuya, K. and Udagawa, S. (1976) New species of Gelasinospora and Anixiella.Trans. Mycol. Soc. Japan 17: 313-320.24. Garnjobst, L. and Wilson, J.F. (1956) Heterocaryosis and protoplasmicincompatibility in Neurospora crassa. Proc.Natl.Acad.Sci. USA 42: 613-618.25. Giasson, L. Specht, C.A., Milgrim, C., Novotny, C.P. and Ullrich, R.C. (1989)Cloning and comparison of A a mating-type alleles of the basidiomyceteSchizophyllum commune. Mol.Gen.Genet. 218: 72-77.26. Glass, N.L., Grotelueschen, J. and Metzenberg, R.L. (1990) Neurospora crassa Amating-type region. Proc.Natl.Acad.Sci. USA 87: 4912-4916.27. Glass, N.L. and Kuldau, G.A. (1992) Mating type and vegetative incompatibility infilamentous ascomycetes. Annu.Rev.Phytopathol. 30: 201-224.gs28. Glass, N.L. and Lee, L. (1992) Isolation of Neurospora crassa mating type mutantsby repeat induced point (RIP) mutation. Genetics 132: 125-133.29. Glass, N.L., Metzenberg, R.L. and Raju, N.B. (1990) Homothallic Sordariaceae fromnature: The absence of strains containing only the a mating type sequence.Experimental Mycology 14: 274-289.30. Glass, N.L. and Staben, C. (1990) Genetic control of mating in Neurospora crassa.Developmental Biology 1: 177-184.31. Glass, N.L., Vollmer, S.J., Staben, C., Grotelueschen, J., Metzenberg, R.L. andYanofsky, C. (1988) DNAs of the two mating-type alleles of Neurospora crassa arehighly dissimilar. Science 241: 570-573.32. Gochenaur, S.E. and Backus, M.P. (1962) A new species of Neurospora fromWisconsin lowland soil. Mycologia 54: 555-562.33. Griffiths, A.J.F. (1982) Null mutants of the A and a mating type alleles of Neurosporacrassa. Can.J.Genet.Cytol. 24: 167-176.34. Griffiths, A.J.F. and DeLange, A.M. (1978) Mutations of the a mating-type gene inNeurospora crassa. Genetics 88: 239-254.35. Holliday, R. (1987) The inheritance of epigenetic defects. Science 238:163 - 170.36. Howe, H.B. and Page, J.E. (1964) Nonconidiation in the new homothallic species,Neurospora terricola. Neurospora Newsletter 4: 7.37. Jensen, R., Sprague, G.F. and Herskowitz, I. (1983) Regulation of yeast mating-typeinterconversion: Feedback control of HO gene expression by the mating-type locus.Proc.Natl.Acad.Sci USA 80: 3035-3039.38. Kelly, M., Burke, J., Smith, M., Klar, A. and Beach, D. (1988) Four mating-typegenes control sexual differentiation in the fission yeast. EMBO Journal Vo1.7 No.5:1537-1545.39. Kitada, K. and Hishinuma, F. (1988) Evidence for preferential multiplication of theinternal unit in tandem repeats of the mating factor a genes in Saccharomyces yeasts.Current Genetics 13: 1-5.40. Klar, A.J.S., Strathern, J.N. and Hicks, J.B. (1981) A position-effect for genetransposition: State of expression of yeast mating-type genes affects their ability toswitch. Cell 25: 517-524.8G41. Koopman, P., Gubbay, J., Vivian, N., Goodfellow, P. and Lovell-Badge, R. (1991)Male development of chromosomally female mice transgenic for Sry.Nature 351: 117-121.42. Kronstad, J.W. and Loeng, S.A. (1989) Isolation of two alleles of the b locus ofUstilago maydis. Proc.Natl.Acad.Sci. USA 86: 978-982.43. Lu, B.C. (1967) The course of meiosis and centriole behaviour during ascusdevelopment of the ascomycete Gelasinospora calospora. Chromosoma 22: 210-226.44. Mahoney, D.P., Huang, L.H. and Backus, M.P. (1969) New homothallic Neurosporasfrom tropical soils. Mycologia 61: 264-272.45. McClintock, B. (1947) Cytogenetic studies of maize and Neurospora. Carnegie Inst.Washington Yearbook 46: 146-152.46. McGill, C., Shafer, B. and Strathern, J. (1989) Coconversion of flanking sequenceswith homothallic switching. Cell 57: 459-467.47. McLaren, A. (1991) The making of male mice. Nature 351: 96.48. Metzenberg, R.L. (1991) Benefactor's lecture: The impact of molecular biology onmycology. Mycol.Res. 95: 9-13.49. Metzenberg, R.L. and Ahlgren, S.K. (1973) Behaviour of Neurospora tetraspermamating-type genes introgressed into N.crassa. Can.J.Genet.Cytol. 15: 571-576.50. Metzenberg, R.L., Stevens, J.N., Selker, E.U. and Morzycka-Wroblewska, E. (1985)Identification and chromosomal distribution of 5S rRNA genes in Neurospora crassa.Proc.Natl.Acad.Sci. USA 82: 2067-2071.51. Metzenberg, R.L. and Glass, N.L. (1990) Mating type and mating strategies inNeurospora. Bioessays Vol.12, No.2: 53-59.52. Miller, A.M. and Nasmyth, K.A. (1984) Role of DNA replication in the repression ofsilent mating type loci in yeast. Nature 312: 247-251.53. Mishra, N.C. (1971) Heterokaryosis in Neurospora sitophila. Genetics 67: 55-59.54. Nasmyth, K.A. (1982) The regulation of yeast mating-type chromatin structure bySIR: An action at a distance affecting both transcription and transposition. Cell 30:567-578.55. Nasmyth, K. and Shore, D. (1987) Transcriptional regulation in the yeast life cycle.Science 237: 1162-1170.al8856. Natvig, D.O., Jackson, D.A., and Taylor, J.W. (1987) Random fragment hybridizationanalysis of evolution in the genus Neurospora: The status of four-spored strains.Evolution 41(5): 1003-1021.57. Nauta, M.J. and Hoekstra, R.F. (1992) Evolution of reproductive systems infilamentous ascomycetes. I. Evolution of mating types. Heredity 68: 405-410.58. Nauta, M.J. and Hoekstra, R.F. (1992) Evolution of reproductive systems infilamentous ascomycetes. II. Evolution of hermaphroditism and other reproductivestrategies. Heredity 68: 537-546.59. Nelson, A.C. and Backus, M.P. (1968) Ascocarp development in two homothallicNeurosporas. Mycologia 60: 16-28.60. Newmeyer, D. (1970) A suppresser of the heterokaryon incompatibility associatedwith mating-type in Neurospora crassa. Can J.Genet.Cytol. 12: 914-926.61. Olive, L.S. (1959) On the evolution of heterothallism in fungi. Am.Nat. 42: 233-251.62. Orbach, M.J., Vollbrath, D., Davis, R.W. and Yanofsky, C. (1988) An electrophoretickaryotype of Neurospora crassa. Mol.and Cell.Biol. Vol. 8 No. 4: 1469-1473.63. Perkins, D.D. (1977) Behaviour of Neurospora sitophila mating-type alleles inheterozygous duplications after introgression into Neurospora crassa. ExperimentalMycology 1: 166-172.64. Perkins, D.D. (1991) In Praise of Diversity. from More gene manipulations in fungi.Academic Press Inc. pp. 3-26.65. Perkins, D.D. (1987) Mating-type switching in filamentous ascomycetes.Genetics 115: 215-216.66. Perkins, D.D. (1992) Neurospora: The organism behind the molecular revolution.Genetics 130: 687-701.67. Perkins, D.D. and Raju, N.B. (1986) Neurospora discreta, a new heterothallic speciesdefined by its crossing behaviour. Experimental Mycology 10: 323-338.68. Perkins, D.D. and Turner, B.C. (1988) Neurospora from natural populations:Toward the population biology of a haploid eukaryote. Experimental Mycology 12:91-131.69. Perkins, D.D., Turner, B.C. and Barry, E.G. (1976) Strains of Neurospora collectedfrom nature. Evolution 30: 281-313.70. Ptashne, M. (1986) Gene regulation by proteins acting nearby and at a distance.Nature 322:697 -701.71. Ptashne, M. (1988) How eukaryotic transcriptional activators work.Nature 335:683-689.72. Raju, N.B. (1992) Functional heterothallism resulting from homokaryotic conidia andascospores in Neurospora tetrasperma. Mycological Research 96: 103-116.73. Raju, N.B. (1992) Genetic control of the sexual cycle in Neurospora.Mycological Research 96: 241-262.74. Raju, N.B. (1980) Meiosis and ascospore genesis in Neurospora.European Journal of Cell Biology 23: 208-223.75. Raju, N.B. (1979) A four spored mutant of Neurospora crassa.Exp.Mycol. 3: 270-280.76. Raju, N.B. (1978) Meiotic nuclear behaviour and ascospore formation in fivehomothallic species of Neurospora. Canadian Journal of Botany 56: 754-763.77. Sambrook, M., Fritsch, D., and Maniotis, R. (1989) Molecular Cloning andLaboratory Manual. Cold Spring Harbor Laboratory Press, Second edition.78. Schiestl, R. and Wintersberger, U. (1983) Induction of mating-type interconversion ina heterothallic strain of Saccharomyces cereviseae by DNA damaging agents.Molecular and General Genetics 191: 59-65.79. Schmidt, H. (1987) Strains of Schizosaccharomyces pombe with a disrupted swilgene still show some mating-type switching. Molecular and General Genetics 210:485-489.80. Selker, E.U. (1990) Premeiotic instability of repeated sequences in Neurosporacrassa. Annu.Review.Genet. 24: 579-613.81. Selker, E.U. and Garrett, P.W. (1988) DNA sequence duplications trigger geneinactivation in Neurospora crassa. Proc.Natl.Acad.Sci. USA 85: 6870-6874.82. Selker, E.U., Yanofsky, C., Driftmier, K. and Metzenberg, R.L. (1981) Dispersed 5SRNA genes in N.crassa: strucure, expression and evolution. Cell 24: 819-828.83. Shear, C.L. and Dodge, B.O. (1927) Life histories and heterothallism of the redbread-mold fungi of the Monilia sitophila group. Journal of Agricultural ResearchVol.34 No.11: 1019- 1042.84. Siliciano, P.G. and Tatchell, K. (1984) Transcription and regulatory signals at themating type locus in yeast. Cell 37: 969-978.85. Simonet, J. and Zickler, D. (1978) Genes involved in caryogamy and meiosis inPodospora anserina. Molecular and General Genetics 162: 237-242.86. Sinclair, A.H. et.al . (1990) A gene from the human sex-determining region encodes aprotein with homology to a conserved DNA-binding motif. Nature 346: 240-244.87. Singleton, J.R. (1953) Chromosome morphology and the chromosome cycle in theascus of Neurospora crassa. Am. J. Bot. 40: 124-144.88. Staben, C. and Yanofsky, C. (1990) Neurospora crassa a mating-type region.Proc.Natl.Acad.Sci.USA 87: 4917-4921.89. Struhl, K. (1983) The new yeast genetics. Nature 305: 391-397.90. Taylor, J.W., Smolich, B.D., and May, G. (1986) Evolution and mitochondrial DNA inNeurospora crassa. Evolution 40(4): 716-739.91. Taylor, J.W. and Natvig, D.O. (1989) Mitochondrial DNA and evolution ofheterothallic and pseudohomothallic Neurospora species. Mycological Research 93:257-272.92. Tse, W.E. and Forget, B.G. (1990) Reverse transcription and direct amplification ofcellular RNA transcripts by Taq polymerase. Gene 88:293-296.93. Turcq, B., Denayrolles, M. and Begueret, J. (1990) Isolation of the two allelicincompatibility genes s and S of the fungus Podospora anserina.Current Genetics 17: 297-303.94. Turner, G. and Ballance, D.J. (1987) Structural variation and expression of fungalchromosomal genes. from Evolutionary Biology of the Fungi eds. A.D.M. Rayner,C.M. Brasier and D. Moore. Symposium of the British Mycological Society. pp.17-31.95. Vigfusson, N.V. and Weijer, J. (1972) Sexuality in Neurospora crassa.Gen.Res.,Camb. 19: 205-211.96. Vollmer, S.J. and Yanofsky, C. (1986) Efficient cloning of genes of Neurosporacrassa. Proc.Natl.Acad.Sci. USA 83: 4869-4873.97. von Arx, J.A. (1981) The Genera of Fungi Sporulating in Pure Culture,3rd ed. J. Cramer, Vaduz.90


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items