Sex in Fungi: Molecular Determindtio/tJlrd I- t uLurr'nrty IntPltiltiu't'

Edited bv Joscph Heitman et al.O 2007 ASM Press, \X/ashirrgton, D.(i.

Timothy Y. James 19Analysis of Matittg-tpe LocusOrganrzation and Synteny inMushroom Fungi:Beyond Model Species

The ability of a fungal individual to mate or outcrosswith another individual is dependent on its mating type.The mating type of an individual is determined by thephenotypic expression of its mating-type locus (MAT)or loci, and only individrials with different rnating typesare compatible. Species that have two MAT loci are

termed tetrapolar, and those that have only a singleMAT locus are bipolar. MAT Ioci correspond with re-gions of the genome that may be as small as a singlegene (e.g., Cocbliobolus heterostrophus 181)) or as largeas 0.5 Mbp (e.g.,IJstilago hordei [50]). As discussed indetail in this volume, the genes that function in deter-mining mating type in both Ascomycota and Basid-iomycota encode either transcription factors or phero-mone receptors and their pheromone ligands (12, 32).MAT loci encode both genes whose expression directlydetermines mating-type specificity as well as those thatdo not determine the mating type of a cell but are

nonetheless mating-type specific. For example, the gene

mtA-2 rn I'Jeurospora crassa is specific to the A mating-type allele of the MAT locus, functions in ascospore de-

velopment, but is not utilized in mating-type determina-tion (21). Likewise, the a2 allele of the Ustilago maydisMAT-a locus harbors two genes (lga2 and rga2) that are

specific to a single mating type but are involved in mito-chondrial fusion rather than mating-type control (8,

8 3 ) . In this review, genes that are in or near the MAT Io'cus but do not function in detern-rining mating-typespecificity are referred to as non-mating-type MAZlinked genes.

The Basidiomycota appear to be divided into three

major iineages: rusts (Urediniomycetes), smuts (Ustilagi-

nomycetes), and mushroom-like fungi (Hymenomycetes,

including homobasidiomycetes and ;elly fungi; Fig. 19.1).Most of the described basidiomycete species are homoba-sidiomycetes (42), and these species are common compo-nents of soil ecosystems (33, 61). The basidiomycetes are

unique among fungi in having species with tetrapolarmating systems. Mushroom fungi are further distinctfrom other basidiomycete species because there may be

numerous (up to hundreds) mating-type alleles at bothof the MAT Iocr. Most of the proliferation of matingtypes is due to the manner in which the mushroomMAT locus is composed of multiple, redundant sublocithat can display recombination distances as high as 16

centimorgans (cM) (66). There is a great body of litera-ture on the mating systems and mating-type numberand distribution in mushrooms, in part because of the

Timothy Y. James, Department of Biology, Dr.rke University, Durham, NC 27708.

317

318

Coprinopsis cinerea

Coprinopsis scobicola

Co p ri n e I I u s d i s sem i n atu s

Pholiota nameko

Hypholoma fasciculare

Schizophyllum commune

Mycena sp. A

Pleurotus djamor

Laccaria bicolor

Bestorol,rycETES: THE Musnnoous

Agaricales

Ascomycota

To3oo!D@.

so3ooOJ

IDqt

2.CL

6'3oo9)

origin of tetrapolarmating system?

a reversion to a bipolarmating system

utility of the mating-type locus to serve as a highly poly-morphic marker (60, 82). Among homobasidiornycetes,an estimated 10o/o are homothallic (non-outcrossing), 25to 35o/o are bipolar, and 55 to 65'% are tetrapolar (68,88). From an evolutionary perspective. maring-systemswitches in fungi are fascinating and none more so thanthe former genus Coprinus, in which homothallic, bipo-laq and tetrapolar species interdigitate along the speciesphylogeny (7, 34). The molecular genetic bases for thesemating-system switches have been less tractable than thatobserved in Ascomycetes (91), in part due to the com-plerity of the homobasidiomycete mating genes (4.1 ).

Coprinopsis cinerea (- Coprinus cinereus) and Schizo-phyllum commune are both model systems for studyingmating ger-retics in mushrooms (10, 66). C. cineres andS. commune are ercellent model systems because theymate readily and can fruit directly on a petri dish of sim-

polyporoidclade

Trichaotum biforme

- hymenochaetoid j

' cladeC ryptococc u s 11 gsfs 1rn 4p I

- Tremel lomycetidae

Ustilaginomycetes/smuts

Microbotryum visls6sum

- UrediniomyceteArusts

Neurospora crassa

Saccharomyces ce revisiae

Pneumocystis carinii

'::;::7n**"*,,,)

Figure 19.1 Phylogeny of the Basidiomycota. The cladogram depicts the current knowledge of the rela-tionships arnong the fungi discussed in the text. I)ata are derived from references 30, 59, and 79, the Treeof Life ril/eb project (http:iitolweb.org/tree/phylogenv.html), and the mor \/eb project (http://mor.clarku.edu/).

ple nutrient agar. C. cinerea has the added advantagesof producing asexual propagules (oidia) and meiotictetrads that can be isolated (44). Both species have been

advanced as model systems by the development of auxo-trophic mutants, MAT rnutants, and genetic transfor-mation (85).

During the 1990s the MAT-A and MAT-B loci of C.

cinerea and S. commune were cloned. Cloning of the

MAT loci was accomplished by various means (see chap-ters 16 and 17 by Stankis and Specht and by Casseitonand I(ties), but the determination that the cloned regionactually carried MAT genes relied on transformation olthe genes into a suitable strain followed by screening for a

mating-compatible phenotype (27, 64, 76). The mush-room MAT-A locus encodes two types of dissimilarhomeodomain proteins (HD1 and HD2 [48]). Dimeriza-tion between heteroallelic HD1 and HD2 proteins forms

linkage of mipto MAT-A?

19. MAT rN MusuRoou FuNcr: BEyoND MolEl SpsctEs

a transcription factor that activates a MAT-A-specific de-

velopmental pathway. The mushroom MAT-B locus en-

codes pheromone receptors and their lipopeptide phero-mone ligands (64, 87). As with MAT-A, heteroalleliccombinations of pheromones and pheromone receptorsin mated cells cause the activation of the MAT-B-specifi,cdevelopmental pathway via a putative mitogen-activatedprotein kinase signaling cascade triggered by the G-protein-coupled pheromone receptors (10).

Cloning of these MAT loci revealed not only thatthey encode a large number of alleles but also that thealleles were highly dissimilar in sequence. The alleles

were so different that they generally failed to cross-hybridize in Southern hybridizations (64, 77). This alsoholds at the amino acid level; for example, the MAT-Ahomeodomain proteins of S. commune are only 42 to54o/o identical between alleles (78).

When trying to study the MAT genes of mushroomfungi other than C. cinerea and S. commune) there are

three major obstacles. First, the loci are likely to be

complex and may comprise several genes. Second, theloci display high sequence variation among MAT alleles

and cannot be cloned by heterologous hybridization us-ing probes from C. cinerea or S. commwne. Lastly, theabsence of a transformation system in which to test thecloned fragments prevents definitive proof that an iso-lated DNA region carries MAT. One way of gettingaround these obstacles is to target not the mating-typegenes themselves but the more slowly evolving genes

that they are tightly linked to, provided that synteny (orconserved gene order) has been maintained among thespecies of interest. This approach allows the investiga-tor to isolate MAT loci and to determine homology ofthe isolated genes using syntenic arguments rather thanby genetic transformation. This review focuses on howMAT genes may be cloned in nonmodel mushroomspecies.

SYNTENY AND MATA IN MUSHROOMSIn order that non-mating-type MAT-linked genes can be

targeted as a proxy for MAT genes, it is essential thatthe linkage of non-mating-type MAT-linked genes toMAT be conserved among species. Because of the highamounts of genome rearrangements among closely re-lated species (80), there was no reason, a priori, to as-

sume that conserved gene order would hold near theMAT loci. Classical genetic studies, however, indicatedearly on that some markers might display highly con-served linkage to the MAT-A locus (66). Specifically, theearliest mapping studies of the MAT-A chromosomal re-gions in C. cinerea and S. commwne indicated that they

319

were syntenic and very tightly linked to loci conferringpara-aminobenzoic acid (pab) and adenine (ade) auxo-trophy in mutants (66). The pab locus mapped to (1cM from MAT-A in both species, and the MA?A genes

in S. commune were first cloned by a chromosomalwalk from pab, ultrmately found to be -50 kbp fromMAT-A (27). After MAT-A loci from C. cinerea and S.

commune had been cloned, it was further apparentthat they shared the presence of a metalloendopepti-dase encoded immediately adjacent to MAT-A (1.3,78).The metalloendopeptidase gene (mip) was not part ofMAT-A as delimited by transformation assays (27). }4ipis a mitochondrial matrix-localized enzyme that func-tions in the cleavage of the leader peptides of precursorproteins targeted to the mitochondrial matrix or innermembrane (37). The first use of non-mating-type MAT-linked genes in positional cloning of MAT genes was ac-

complished when Kiies et aI. (47) used a heterologousmip probe from C. cinerea to isolate the MAZA locusfrom the related Coprinopsis scobicola (: Coprinus bi-lanatws). After recovering cosmid clones containing theC. scobicola mip gene, Kiies et al. were able to delimitthe M,|T-A locus to an -15-kbp region adjacent to mipthrough the use of genetic transformation in C. cinereaand C. scobicola host strains.

In order to develop mip as a marker for MAT-A inmushrooms, conservation of the genetic linkage be-

tween the two loci was explored. Because the mip gene

of C. cinerea did not appear to hybridize to genomicDNA of other mushroom genera (U. Kiies, personalcommunication), an approach based on PCR amplifica-tion of mip was attempted (39). The mip gene was suc-

cessfully amplified from over 30 species of mushroomsthroughout the diversity of homobasidiomycetes. Thelinkage relationships in several species were analyzed bystudying cosegregation of mip and MAT-A in progenyarrays of single spore isolates obtained from single fruit-ing bodies (Table 19.1). The result of the cosegregationstudies, as well as information derived from the se-

quencing of complete genomes, is that linkage of mip toMAT-A is completely conserved throughout homoba-sidiomycetes. Furthermore, in all known cases, the mipgene is directly adjacent and less than 1 kbp from the 3'end of one of the mating-type genes.

How deep into evolutionary history was linkage be-

tween MAT-A and mip established? More data on hetero-basidiomycetes, including rust and smut fungi, would be

useful to address this question. The b mating-type locus(MAT-U) of U. maydis, which is homologous to MA?Ain mushrooms, displays no linkage to mip, nor does thesingle MAT locus of the pathogenic yeast Cryptococcusneoformans (25, 5 1 ) (Table 1 9. 1 ). Synteny between m ip

BastuolaycETEs: THE MusHRoorlls

z.t-rc

o9

= :^-

x;., u o

.Z-"-i-! c e o. --=s

a- ! ! ad_l!!F 4

c; J- tr>=JF>T e &D!!

l-er r !

>aebi3<$$<ara..q

o

Ec

I

-fu.

<5

,>tr-C

ai

:i!

=-3^.E

:9=

*c:P^tr^dJ

?i

lu

o-

aooc

a@--

.g+€O

+-

i9

ii< ';

<!9*gO

; e:i ^-

et-o:

f,E u

:i9,;

F:'

ad-

7-; a=

L=ico

c!

\O

Y1 o=,ouuf;'p' t j € : 'i-,=.= 6 6

-a__L o.a _ j g t.., l€<- ; o @= -I, a + L q-3.:F_--.-;-a>o L r- ^.1 o o (l o o. o o r-o^aff<-v.-

trE': 9.,1 = vllouv-LoF;6 6 u ; 6 i '<-:=..=-Ytsa?> =O Y. u ! - - t - -\ r - a= ! = - = - a =_ 5 5 f o 6 96 =-if '--7 : E 7 :

= i = =,=

jt-o,l -:6V 6 d Y a 6 g 6 E.E YFIE u YoF o p cn . d d .: sD eri c!iotr -=tr ; I E E g E E- E"EE- y.4c i a;: rt bO : o! !O = oO !n; il nfrv,; Eq; e^ U z e_ V, i U X%y. FO *u

= "'u o o u a oJ o I e o o e d d d 3 3*d 3

+^.+^.+++^.+

++++++++++++t

=.!3-aa>>Gir-aFdu ; = >43 p h p I E g p E €-o o c =:rc

-o ; E ; -o -o -o € 6=-6 h i !E5 d o 3 t 6 6 d a 7€ d A * o_o _o i *a = -a ,o -o -- G

9x;ri=;!!!!!!!!!

d I'i rl d d d 6 .3 d d 6r ? jilr o C O o O I C O o C(€ O- :LC O- o- O_ o C- 5 a O_ C_ 1a I 6 9E g I g E 6 = t - - 6O- , aS a L -.o#yE##+e:f.EeLEg

O-o'. .;a

^ , a , 6 O_l:a u u Y u u u u q u u o u--,

q q u X q u j o uY u UY u u or..9 q Y u ? o i 9 o ? 9 9 u,, ?'! u 9-: ? o ? u lt.lt,F^.j^_41.^_>r.^;>;>JE-F+ E+ F+ a , = + ='F+ F ? =s =

u i i Fs 79i t.= J.= J.= r =

u,= ..=.:.= J.=.!.= J,=.!.= r.= J.= : >=

E P; aE t :s; lE t= aE ?Z uE ?=- tE agE 5 E3 A 3 6 3 a-o g l. X 6-X a 3 nX E E aX; X h F E.= S= o!= oD= bD o t = o!;

=u boZ

=7 on;

=7 bo; cb7 : o!=:-1-=-Lt=:---L-

€ ;€ iEi ii Si€ as i 5 6-5 i 545 ;E iEE=3IIJI

: S sc q

.l* r*SS"is I S 1 ii-\ S". ES:3."3--6SS\S* -s d * {t; ESSEs!;!SSSI*i:s s .! s EEr tsis"F,iFssPi;*iI F ! E ssS Fst.ssd\dc.d;i\Fqi \ s 5 _s_: r-r_iUUUUIiJJ>

O

OoO&

O

O

e

o

oz

tm-r=<.t:U;

:<_t.rel:

=\$E

bo

E0F]

EO

bo

Gz

O'dd(,

']J

a6(,

320

OOO>-E

(t'o

d

rAo\-\s

(n

a.

o0bD6

.l

q.o.GF

19. MAT rN MusnRoolr FuNcr: BEyoND Morgl Spscrss

and MAT is generally absent in Ascomycota but existsin the genome of N. crdssa. Linkage in this species isover 100 kbp and quite possibly a statistical artifact.Further studies of Tremeilales, Auriculariales, and otherheterobasidiomycete genomes will be necessary to de-

termine when the linkage relationship was established.Interestingly, the genome sequence of U. maydis showsrhat mip is syntenic (at a distance of - 150 kbp) with theMAT-I locus that encodes the pheromone and phero-mone receptors (homologous to MAT-B of mushroomfungi).

The tight linkage between mip and MAT-A was alsoused to study the MAT loci o{ Coprinellus dissemindtusand Pleurotus djamor (40, 41). The entire MAT-A Iociand surrounding DNA regions were sequenced, thus al-lowing an assessment of conservation of gene order forthe genomic region (Fig. 19.2). These gene order com-parisons reveal synteny to be widespread at. the MAT-Alocus. Besides the mip gene, 11 additional genes have

conserved linkage to MAT-A in C. cinered, C. dissemina-tus, Phanerochaete chrysosporium, and P. djamctr. Thefunctions of 8 of the 11 additional conserved genes are

uncertain (chp1 -5 [conserved hypothetical proteir-rs],

ypl109, and (3-fg [Ap-flanking genel), but the function ofthree of them (sec61, glgen, and glydh) can be speculated

321

upon based on homology to genes in other organisms.sec61 encodes the gamma subunit of a translocase in-volved in moving proteins from the cytoplasm to endo-plasmic reticulum. glgen encodes an enzyme putativelyfunctioning in lipopolysaccharide and glycogen synthesis.glydh encodes a putative glycine decarboxylating enzyme.

The MAT-A loci of C. cinerea and S. commune ltaveboth been divided into two subloci (Acv and Ap) by finemapping studies. A major difference between these twospecies was that the Aa and A0 subloci are very close inC. cinerea (-7 kbp) whereas rn S. commwne the Aa andA0 subloci are far apart. There are no non-nrating-typeMAT-linked genes harbored in the intervening regionbetween the Aa and AA subunits of C. cinerea-hencethe previous name of "homologous hole" for this re-gion. In C. cinerea, a gene termed the B-fg was identifiedas a gene of unknown function on the other border ofthe MAT-A iocus (48). This gene is among the 12 genes

that show conserved linkage to MAT-A among the fourhomobasidiomycetes studied to date (Fig. 19.2).

In contrast to all other mushroom species for whichdata are available, in S. commune MAT-linked genes are

housed between Aa and AB subloci, including milt andpabl (66). This arrangement of Aa and AB sublocicould be erplained by two large inversions rnoving the

Pleurotus djamor

Coprinopsis cinerea

Coprinell u s d isse minalu s

Phaneroch a ete ch rysospo riu m

Figure19.2 Schenraticshowingconseneclgeneorderofthechromosomal regionsurroundingMAT-Ain n'rushroom fungi. Arrows indicate genes and their direction of transcription. Vertical lines connect ho-

nrcrlogous genes betwcen species. Genes r.vith vcrtical stripes or horizontal stripes represent MAT-Ahomeodonrain-t,vpe-1 (HD1) geres and homeodomain-type-2 (HD2) genes, respectively (14). Genes ingra,v are found in all four species, and genes in black arc restricted to a single species. This map includes

a number of additional genes for P. djamor that r'vere rnissed in the pr:evior-rs annotation of the MAT-A re-

gion (40). chp,gene encoding a conserved hypothetical protcin.

glqert

--.

322

genes normally flanking the subloci into the region be-tween the two subloci. The general rarity of non-mating-type MAT-linked genes berween MAT subloci of mush-rooms may reiate ro the theoretical prediction thatincreases in recombination between MAT subloci (thusgenerating recombinant MAT alleles at a higher fre-quency) will increase the chances of mating occurringamong siblings and thereby reduce a species' outbreed-ing potential (74).ln S. commune, a gene termed X wasidentified as the other MAT-Acx flanking gene (56). Dis-ruption of X caused no obvious phenotype in develop-ment (56). Gene X may be specific to S. commune, asthere are no clear homologues of X detectable in othermushroom genomes.

MAT-B ALSO DISPLAYS CONSERVEI)GENE ORDERThe MAT-B locus of mushroom fungi encodes smalllipopeptide pheromones and G-protein-coupled phero-mone receptors that have seven transmembrane-spanningdomains (homologues of yeast STE3 a-factor receptor).In C. cinerea, the MAT-B locus comprises three sub-groups with redundanr function (29). Each subgroup en-codes 1 to 3 pheromone genes and a homoallelic recep-tor; the whole MAIB locus of C. cinerea spans -25 kbp,and ncr non-mating-type MAT-Iinked genes are encodedin the regions berween the subgroups (29).

The sequencing of the first genome of a homobasid-iomycete fungus (P h aner o cb a et e ch ry s o sp or ium) r ev ealedhomologues of the MAT proteins of mushrooms (57)and allowed an erploration of synteny at the MAT-B 1o-cus. Five genes encoding pheromone receptors (homo-logues of STE3) were derected in the (. chrysosporiumgenome. Three of these were found clustered into an-12-kbp region (57), similar to the organization ofMAT-B rn C. cinerea. Further investigation of the MAIB-like genomic region in P. chrysr.tsporium demon-strated very close linkage to the gene sre20 (38). Link-age to ste20 was striking, as it is also linked to the STESreceptor homologues of both Cryptococcus neoformans(-51) and Pneumocystis cdrinii (75). Thus, a conservediinkage between MAT-B and ste20 was demonsrraredfor a wide diversity of organisms including both As-comycota and Basidiomycota. Ste20 is a p21-activatedkinase (PAI() required for mating in budding yeasr (65).The closely related yeast protein CIa4 is also a PAI( andis involved in budding and cytokinesis in yeasr (15).Cla4 differs from Ste20 in possessing a pleckstrin ho-moiogy domain, a region possibly involved in protein-protein interactions. The PAI( genes linked to STE3 inCryptococcus neoformans and Pneumocystis carinii en-

BesnroltycETES: THE Mt.rsHRooHls

code proteins with a pleckstrin homology domain andare phylogenetically more closely related to clal of Sac-

cbdromyces cereuiside and U. maydis (38, 53). There-fore, for the remainder of the chapter I refer to theMAT-B linked PAK as cla4.

The data from Cryptococcus and Pneumocystis M,ATloci (or STE3-encoding regions) provided compelling evi-dence that these species have a cluster of genes all func-tioning in the process of maring (I7,75). Both Crypto-coccus and Pneumocystis MAT loci also encode ste12, a

key transcriptional activator of genes in the pheromoneresponse pathway in Saccharomyces (3). The observationof pheromone receptors linked to genes that may be in-volved in the same developmental pathways suggested a

possible coregulated cluster of genes that might be ex-pected to show evolutionarily conserved gene order.Thus, the potential for cloning MAT-B from mushroomsby exploiting synreny with cldl was invesrigated. Thelinkage between MAT-B and cla4 in additional mush-room species was explored by analyzing the cosegrega-tion of the two loci among progeny of a fruired dikaryon.These data revealed linkage between the two loci in bothof the agarics Pleurotus djamor (40) and Schizophyllumcommune (Fig. 19.3), but the loci were unlinked in thepolyporoid Irpex ldcteus (Table 19.1).

By use of a positional cloning approach, the MAT-Blocus of Pleurotus djamor was partiaily cloned by iden-tifying cosmids from a genomic library that containedthe cla4 gene (40). cla4 was determined to be -29 kbpdistant from a STE3-like pheromone receptor of the p.

djamor MAT-B locus, as opposed to the -6-kbp dis-tance observed in the Phdnerochaete chrysosporiumgenome. A search of the Coprinctpsis cinerea genomedemonstrates that cla4 is -95 kbp from MAT-B.In theheterobasidiomycetes Cryptococcus neoformans and U.maydis, cla4 is -5 and -65 kbp, respectively, from the

B1MP r---------*---tl

400bp-300bp*

20obp-

100bp-

Figure 19.3 Cosegregation ctf clal and MAT-B in Schizopht,lluntcommune. Shown is an a€larose gel (2%) of clal antplicons clig,-stedr'vith MseI. In lane P is the parental dikaryon (Gu1'.21.2) that is het-erozygous at the clal locus. The other lanes show the progenr. utGtry.21 .2. Lanes Bl are of the BJ mating t,vpe and posscss a cbl allelethat lacks an Msel cut sirc. Lanes 82 are of the Il2 mating type andpossess the Msel cut site that digests the 297-bp amplicon into rwofragments of 148 and 149 bp. Lanes marked M contain DNA nrarker

19. MAT rN Musunoott FuNct: BL,YoND Monll Spscrls

STE3 homologue in these species (53). These datademonstrate conserved synteny of cla4 and MAT-B inthe basidiomycetes, though this linkage is clearly looserthan that of mip and MAT-A and there is at least one

example in which synteny has been disrupted (Irpex lac-teus).In ascomycetes other than Pneumocystis carinii,this tight linkage is not observed (based on a scan ofseveral available genomes), though the genes are on thesame chromosome in the ascomycetes Magndportbegrised and Neurospora crassa (at. nearly 1 Mbp dis-tance ).

Comparisons of the genomic regions surrounding theMAT-B loci of the homobasidiomycetes (same species as

that shown in Fig. 19.2) revealed no additional genes

with conserved synteny to MAT-B shared by all fourspecies. However, the MAT-B sequences available forPleurotus diamor and Coprinellus disseminatus arc m:uch

shorter than the available MAT-A sequences. Halsall andcolleagues (29) identified a gene encoding a putarive

transporter of the major facilitator family (mfs1) directlyflanking the MAT-B locus of Cctprinopsis cinerea. Inter-estingly, alle1es of mfsl dilfered by as much as 40ok attheDNA level. A homologue of C. cinerea n/s is syntenicwith the MAT-B locus of Phanerochaete chrysosporium,but the intervening distance is nearly 500 kbp. In gen-

eral, comparative genomics of the MAT-B region sug-

gests that rearrangements happen at a much greater fre-quency than those of the MAT-A region (data notshown).

APPROACHES TO CLONING MATUSING CONSERVED GENE ORDERAs opposed to the manner in which MAT genes werecloned in the model species, cloning MAT genes in.non-model species lacks the advantage of directly testir-rg thefunction of the cloned genes by transformation of theDNA into a mating-compatible host strain. This distinc-tion means that care has to be given when assigning acloned gene to the MAT locus. One strategy for cloningMAT using linked genes begins with amplifying theflanking genes (mip and clal), using PCR with degener-

ate primers (39,40). PCR amplification of these flank-ing genes is expected to be much easier than direct am-plification of the MAT genes themselves. Primers for themip gene have amplified a diversity of homobasidio-mycetes, but this gene dispiays more sequence variationthan cld4. Primers used successfully to amplify cla4 are

STE2O-1F (5'-GTNATGGART\TYATGGARGG-3')and STE20-2R (5'-ACNACTTCAGGNGCCATCCA-3' ).

These primers work readily or-r homobasidiomycetes butofter-r amplify both PAI(s (cla4 and ste20).

12.7

After sequences of mip or cla4 are obtained in the or-ganism of interest, the fragment can be used to probe a ge-

nomic library. If genomic libraries are constructed usingvectors capable of replicating a large DNA fragment (e.g.,

cosmids, bacterial artificial chromosomes, or phages),

clones hybridizing to mip are very likely to contain all orpart of the MAT-A locus. Alternatively, PCR-based ap-proaches such as inverse PCR (62) or thermal asymmet-ric interlaced PCR (54) may be used to amplify and se-

quence the DNA regions immediately adjacent to mip.In the case of cla4, a strategy such as inverse PCR is un-likely to be fruitful given the larger physical distances

typically observed between the gene and MAT-B.Given that sequences homologous to MAT genes can

be obtained for the species of interest, further effort may

be required to demonstrate that the sequenced region ac-

tually harbors MAT.Ideally, it is desirable to show thatthe MAT hon-rologues in nonmodel species cosegregate

with MAT as determined by interstrain matings. Forsome species, such as ectomycorrhizal taxa for which sin-

gle spore isolates cannot be obtained due to lack of spore

germination (e.g., Russuld and Boletzs spp.), this may

not be possible. Since the MAT genes are highly polymor-phic in DNA sequence in all species for which they have

been investigated (58, 69, 78), demonstration that theputative MAT genes are highly polymorphic in the

species of interest can also provide evidence that they are

indeed MAT genes. Finally, heterologous expression ofthe cloned MAT homologues, when transformed into C.

cinerea or S. commune, may be used to confirm theirfunction (41,47). Although MAT genes of most mush-room species are unlikely to interact with those of themodei host species, by transformation of two alleles

into a single host strain or through mating of strainstransformed with different alleles, the interactions be-

tween the gene products may be tested. This assay relies

on comrron MAT-A and MAT-B downstream targetsbetween the host and nonmodel species, which may notbe valid for all rnushroom species.

CLONING MAT USING DIRECT PCRAMPLIFICATIONAlthough most of this review has focused on how con-

served gene order can be used to clone MAT from non-

model species, a more direct option is through PCR am-

plification of the MAT homologues then-rselves. The STE3

MAT-B homologues display some relatively conserved

amino acids among the seven transmembrane alpha-helices (amino terminus) that can be targeted by degener-

ate primers (1,, 40,41). These primers have successfully

amplified STE3 homologues from a wide range of

4--

321

mushrooms, from the wood ear fungus Auricularia poly-tricbd (GenBank accession number Ay226009\ to rhe a;hbolete Gyrodon merulioides (GenBank accession number4Y226018). Phylogenetic analyses of the MAIB recep_tors suggest two divergent groups of STE-l-like phero_mone receptors in homobasidiomycetes (1, 40, 69), andthe PCR-amplified receprors group among the ones thathave been demonstrated to have true MAT-determiningfunction (data not shown). However, data suggest thaiSTES-like genes are plentiful in mushroom fungi, andnon-mating-type,specific pheromone receptor-encodinggenes have been detecte d in Coprinellus dissemindtus(41), Phanerochaete chrysosporium (57), and even Co-prinopsis cinered (T. Y. James, unpublished observations).For example, ar least four STEJ homologues have beendetected tn C. dissemindtus, mapping to at least two sep-arate locations distinct from the MAT locus. Sirnilarly, inP. chrysosporium five STE3 homologues are present andfound in rhree separate regions of the genome (57). Un-dersranding rhe lunction of these non-maring-rype-specific pheromone receptors is an exciting prospect thatmay reveal developmental pathways specific to mush-room fungi.

PCR amplification of the homeodomain transcrip-tion factors of MAT-A has been heroically accomplishedwith the mushroorn Pboliota nameko (1). Here the au-thors targeted the few conserved amino acids of thehomeodomain region (HD1) by degenerate pCR andisolated the entire HD1 gene from this species bygenome walking using cassette-mediated pCR. A similarapproach was aiso used to amplify and sequence a pher-omone receptor homologue of the MAT-B locr:s. Aimr etal. (1) went further and amplified homologues of bothpabl and adcS in P. nameko and used cosegregerionanalyses to show that the horneodomain-encoJing gene(hox1) was linked to both pabl and adeS and.,Lor.-over, segregated 1:1 with the MAT locus in this bipolarmtrshroom species. The isolated pheromone recepror(rcbl), however, was nor linked to MAT. Attempts toamplify homeodornain-encoding genes in other mush-rooms have been unsuccessful (James, unpublished).

THE ORGANIZATION OP METIN HOMOBASIDIOMYCETESThe ancestor of the homobasidiomycetes is inferred tobe tetrapolar (31). In terrapolar species, the homeo-domain proteir-is encoded by the MAT-A locus in amonokaryon ar:e unable to heterodimerjze to form anactive transcription regulator. In compatible matingsbringing together nuclei with different A mating typesinto a single dikaryotic cell, the protein products of ihe

BesnronycETES: THE MusnRool,ts

two heteroallelic MAT-A loci are able to form the activeheterodimer. It is unclear how MAT loci are organizedin homothallic species, but the srructure of MAT rn het-erothallic species suggests that recombination events be-tween different MAT alleles could bring togerher com-patible gene products into a single MAT allele capableof turning on dikaryotic development without the needfor a mating partner. Recombination within a singleMAT-A haplotype has been demonsrrared to create a

self-compatible MAT allele by the fusion of HDj anclHD2 genes from different subunits of the Coprinopsiscinerea M,4T,A locus (46). This manner by which one ofthe two MATloci can mutate to a self-compatible alleleled Raper (66) to predict that this process led to the for-mation of bipolar species from tetrapolar species. Healso suggested that homothallic species should originatefrom bipolar ancestors more readily than tetrapolar an-cestors (i.e., bipolarity is a transition state from terrapo-lar to homothallic). Data on a few species (Coprineilusdissemindtus 1411, Pboliota nameko [1|, and phane-rochdete chrysosporium [38]) now suggesr that bipolarmushroom species originate through the loss of rnating-type-determining function of the pheromones or recep-tors of MAT-B (Fig. 19.1). These data agree with thefacts that all previous arremprs ro murare the MAT lociof mushrooms resulted only in self-compatible MAT at-leles, rather than novel alleles (66), and that the self-compatible murarions observed at the MAT-B loci of C.cinerea and Schizophyllum commune occurred throughsingle point murarions (24,63).

Classical genetic studies determined that the MATloci are composed of multiple, tightly linked, and r:edun-dant subloci (78, 67) and that recombinarion betweenthese subloci, in part, generates the huge number ofmating types. Molecular genetic studies have shownthat the different subloci harbor the same genes; there-fore, the diversity of mushroom mating types can alsohe explained by rhe hyperveriabiliry of rhe genes. parric-ulariy in their specificity-determining regions 1i, SSy.The number of subloci thus far observed at MAT loctranges from one (e.g., MAT-A of p. chrysosporium) tothree (e.g., MAT-B of C. cinerea); it is likely that specieshaving four or more subloci wili be found (45). Thesesubloci can be close together (Fig. 19.2) or further apart(1 to 16 cM for S. commune MAT-A), though the closearrangement seems to be more common. Although pre_vious observations suggested that the bipolar MAT lo-cus may be indivisible into subloci (66), it is now appar-ent that this is not universaliy the case (j, 4j).Conversely, tetrapolar species such as pleurotus djamormay have only a single MAT-A sublocus. The S. com-mune MAT-A locus appears to be differenr from MAT-A

19. MAT rN MusuRoolr FuNct: BEYoND Monn Spsctls

loci of other mushroom species because genes are en-

coded between the MAT-Aa and MAT-Ap subloci, pre-

sumably due to large chromosomal inversions. There is

good evidence in S. commune that the physical distance

between subloci at MAT-B may be much closer than the

mapping distance suggests and that the products of the

subloci may actually be able to activate each other (23).

Future research will be needed to explain the observa-

tion that different MAT alleles of S. commune generate

recombinant mating types at different frequencies.

WHY HAS LINKAGE BEEN CONSERVEDBETWEEN MAT AND OTHER GENES?

Conserved gene order is generally observed to disappear

or decay as organisms diverge (80). Nonetheless, some

regions of the genome, such as the X chromosome ofvertebrates, display extensive synteny. even among

puffer fish and humans, which diverged over 400 mil-lion years ago (28). Other regions of the genome have

undergone repid and extensive rearrangemettrs. in eu-

karyotes often as small inversions (20). \fhy have some

genes, but not others, remained syntenic for such an ex-

tended period of time?There are at least five hypotheses for the observation

of conserved gene order near the MAT loci of basid-

iomycetes. The first and most obvious hypothesis is thatthe gene linkages are just due to chance or historical ac-

cident. This postulates that the observed gene order is

due to a historical genome rearrangement that was unre-

lated to the functional coding ability of the genes. Since

then, the maintenance of the gene order between species

is a probabilistic function related to time of species diver-gence and distance between genes. There are a few argu-

ments against this simple explanation, however. One is

that the degree of synteny or at the least frequency ofgenome rearrangement at MAT-A differs from that at

MAT-B, suggesting that the process is not random. Acounterargument to this point is that different regions

of the genome experience different levels of gene re-

arrangement, i.e., teiomeric and centromeric portions ofa chromosome appear to be more dynamic than other

portions (20). A second argument against the "historical

accident" hypothesis is that the linkage between mip and

MAT-A has been maintained despite a small inversionnear the MAT locus that switched whether the HD1 or

HD2 MAT gene was proximate to mip (Fig. 1 9.2). Lastly,

the observed synteny between cla4 and MAT-B has been

conserved for a very long stretch of time (-400-million-year estimated divergence between Pneumocystls and

Phanerochaere [5]), despite the rapid rate at which syn-

teny has been observed to decline in fungi (22).

Another possible hypothesis for the conserved synteny

between MAT and other genes is that the genes form a

coregulated cluster. Groups of bacterial and eukaryoticgenes that display conserved gene order have been

Jemonstrated to physically interact or be coexpressed

(1,6,36). The clustering of genes that interact with each

other suggests that their transcription factors may be dis-

tributed heterogenously over the genome/nucleus or thatthe genomic region may be coregulated by general mech-

anisms such as methylation or histone modification(36). Genes such as mip and the homeodomain MAT-Agenes may be so tightly linked that they share regulatory

regions. Severing these gene linkages by inversions or

translocetions could disrupt proper expression of one or

both genes.

A third hypothesis is that the flanking non-mating-type

MAllinked genes are undergoing coevolution with the

MAT genes. In order for coevolution to occur' the genome

region would have to experience recombination suppres-

sion such that alleles at one gene could be correlated withthe alleies at the linked genes. In this scenario, cophy-

logeny of alleles at the non-mating-type MA?linked genes

and the MAT genes is expected; in other words, the gene

tree for the two genes should show an identical branching

order among alleles. This is observed at ser-determining

loci where recombination is suppressed over large regions'

For example, at the self-incompatibility locus of the

angiosperm Brassica, the dispensable gene SLG shows

coevolution, cophylogenn and even gene conversion

with the self-incompatibility specificity-determininggene SRK (71). Similarly, in the large MAT locus of

Cryptococcus neoformans, cophylogeny is observed be-

tween the specificity-determining genes (e.g., STE3) and

the genes more recently recruited to the MAT locus (e.g.,

ZNFl t25l).In contrast, in homobasidiomycetes, recom-

bination appears to occur very frequently outside the ac-

tual specificity-deterrnining genes of the MAT locus (41,

55 ). Recombination between mip and MAT- Aa in S ch izo-

phyltum commune is also observed such that the two loci

do not show strict cophylogeny (James, unpublished)'

This recombination disrupts the linkage disequilibriumneeded to create coevolved complexes of alleles at physi-

cally linked genes.

A fourth hypothesis is that recombination is sup-

pressed at the MAT region and this has a depressing ef-

?..t o.r the frequency of genon-re rearrangements as well.

This does not seem to be the case, however, as recombi-

nation at the MAT loci appears to be normal relative to

other regions of the genome (23,49,55). In addition,the high number of small inversions' gene duplications,

and deletions observed (Fig' 19.2) (51) suggests that re-

arrangements are frequent near the MAT loci.

i2.t

326

A final hypothesis for linkage between MAT andother genes is related to how t.he MAT loci originare.Hurst and Hamilton (35) hypothesized that MAT locioriginate as a mechanism to minimize cytoplasmic genewarfare between fusing gametes. They envisioned a sce-nario in which the fusion of isogamous gametes createsa conflict between mates for the control of mitochon-drial genorne inheritance. In the first step of this three-step rnodel, a mutant mitochondrial gene arises that candestr:oy the mitochondrion of a mating partner. These de-stroyer initochondrial genotypes can reduce the fitness ofzygotes, particuiarly when two destroyer genotypesfuse. But, nonetheless, the destroyer phenotype is verylikely to reach fixation. In the second step, a nucleargene arises that suppresses the destroyer phenotype ofits mitochondrion, thus partially alleviating the fitnessloss due to destroyer mitochondria. This "suppressor"gene can be shown to result in a stable polymorphism ofsuppressor and nonsuppressor alleles. In the final step, a"choosy" gene arises which allows the cell to preferen-tially mate with a cell of the opposite suppressor pheno-type. These suppressor/nonsuppressor matings have thehighest fitness if the destroyer mitochondrial genorype isat fixation. The choosy gene may show a preference formating with a suppressor or nonsuppressor genotype,and this scenario is mechanistically the easiest to imag-ine. Selection will then favor tight linkage between thechoosy gene and the suppressor gene and lead to theformation of a MAT locus and uniparental mitochon-drial inheritance.

Hurst and Hamilton (35) envisioned this origin ofsexes to occur when both parental nuclear and cytoplas-mic genotypes were combined in a zygote, such as thefusion of mating yeast cells. The case of the MAT-a lct-cus of the basidiomycete yeast Ustilago maydis (homol-ogous to MAT-B in mushrooms and encoding phero-mones and pheromone receptor genes) provides aninteresting situation in which the Hurst and Hamiltonmodel may be tested. In this species, genes that appearto be involved in mitochondriai morphology and fusion(lga2 and rga2) are found in only one of the two MATalleles (a2) of the MAT-I locus (8). When overex-pressed, /ga2 causes both nritochondriai fragmentationand mitochondrial DNA degradation (8). In this sce-nario, the lga2 and rga2 genes are suppressors of selfishmitochondria, and the pheromone/receptor gerres arethe choosy genes that help cells choose the proper part-r-rer by signaling with lipopeptide pheromones. This ob-servation provides a compelling case for how a MAT 1o-cus can maintain genes that do not control mating-typespecificity but actually control other processes such as

mitochondrial fusion and inheritance. In this example,

BesmrovycF,TEs: THE MusuRoorvls

selection to mediate nuclear/cytoplasmic conflict couldhave been the reason the pheromone/receptors became a

MAT locus or could have merely tightened the linkagebetween the pheromone/receptors controlling mate pref-erence and the lga2lrga2 genes controlling mitochon-drial morphology.

Additionai support for an interaction between MATand mitochondrial inheritance in basidiomycetous yeasrs

comes from the data on Cryptococcus neoformans. Inthis species with two mating types (a and a), laboratorycrosses demonstrated that progeny almost exclusively in-herit mitochondria from the MATa parent (90). Further-rnore, Yan et al. (89) demonstrated that the homeo-domain protein SXIIcv encoded by Cryptococcus MATcontrols the process of mitochondrial inheritance bydemonstrating biparental inheritance of mitochondria inSXIIa deletion mutants. Since mitochondria from theMATct parent are rarely inherited, the MATa locus mayencode a "suppressor" alleie or gene that is missing fromMATa.

In mushrooms, the non-mating-type MAl:linked gene

most consistently linked to MAT is mip, a gene that aiso

functions in the mitochondrion. In fact, a iarge number ofgenes that function in the mitochondrion are also linked toone or the other MAT locr in basidiomycetes (Table 19.2).

One hypothesis for why this could occur is that the controlof the sexual cycle (by MAT) is coregulated with the con-trol of mitochondrial inheritance. ln Sdccharomyces cere-

uisiae, deletion of mip causes loss of functional mitochon-drial genomes as well as severe defects in respiration (9).

These data lead to the specuiation that mip could be a sup-pressor locus of selfish mitochondrial genomes (70) andcould control inheritance in heteroplasmic cells that resultfrom mating. Hurst and Hamilton have argued that theirtheory does not apply to basidiomycetes since these fungiexchange nuclei but not cytoplasrn following cell fusion. Itnow appears that mitochondria in mushroom species may,

on occasion, be biparentally inherited, and recombinantmitochondria have now been detected (4, 72). Further-more, reconciliation of their theory and the observationof mitochondrion-targeted genes linked to MAT locr inmushrooms is straightforward since the tetrapolar MATloci of mushrooms and basidiomycetous yeasts are homol-ogous and likely derived from a common origin.

Conservation of linkage between MAT and non-mating-type MAllinked genes has also been observedin Ascomycetes. The two genes sla2 (encoding an acrirl-binding protein involved in cytoskeleton assembiy) andapn2 (encoding a DNA lyase) flank the MAT locus ina large number of species including both hemiasco-mycetes and euascomycetes (11, 26, 84). Remarkably,the gene sla2 also shows tight linkage to the U. mdydis

19. MAT rN MusHRool,t FuNcr: BEYoND MonEl Spscrss

'fablet9.2 Non-mating-type MAT-linked genes in basidiomycetes that encode proteins targeted to mitochondria"

-JZ /

TaronLinked tcr

MAT-Aor MAT-B?

Function References(s)( lene

ETFl

Glycine dehydrogenase

@tydh)Iga2

Mitocl-rondrial intermediatepeptidase (miP)

Mitochondrial ribosomesmall subunit component(rPs19)

rga2

RP041

)rp1109

Cryp to co ccus n eoformans

H omob asidiomycetes

LI stilagct m ay di s ; SP or i s or iutn

reilianumHomobasidiomycetes

Pleurotus djdmor

IJ s til a go may dis ; Sp or isorium

reilianumCrl,pto cct ccus neoformansHomobasidiomycetes

Mitochondrial electrontransport

Anrino acid transportand metabolism

Mitochondrial morphologyand fusion

N-terminal processing of nuclear

encoded proteins targeted tothe mitochondrial matrix orinner membrane

Protein translation

Mitochondrial morphology and

fusionMitochondrial RNA polyn'rerase

Possible role in ubiquinonebiosynthesis

A+Bb

A

B

A

t)

A+BA

51

8, 73

37

8,73

51

,,ln this tablc, MA?A refers to thc homeoclomain protein encoding MAT region,.and MAT-B refers to the pheronone/rcceptor gene Jocus

,ii.;;;iil;;r ths bip.la, species cont:rins honologues of both MAT-,A and MAT-B genes of homobasicliomvcetes'.-bo..d on subce llular predictions using thc softu,are Wot-f pSOnl (http://wol{psort.seq.cbrc.jp/).

MAT-| locus encoding homeodomain proteins. A func-

tional link between sld2 and mating is suggested by

these data but yet to be established. it is worth noting

that sla2 and clal have been demonstrated to interact in

yeast based on yeast two-hybrid interactions (19).

CONCLUSIONSGenomics tries to find higher-order meaning in the orga-

nization of genes within genomes. In this review, I dis-

cussed how mating-type genes can be cloned from non-

model species and attempted to synthesize what is known

about the organization of the MAT loci and neighboring

genomic regions in mushroom fungi. Synteny appears to

be more conserved at MAT-A than MAT-B. MAT genes

in mushrooms can be cloned by both direct PCR ampli-

fication and positional cloning using non-maring-type

MA?linked genes by probing a large insert genomic li-brary. Proper determination of whether a cloned gene is

a mating-type gene is nontrivial but can utilize syntenic

arguments and population genetics, as well as transfor-mation into heterologous hosts. MAT loci show high

gene order conservation, but the significance of this ob-

servation is unclear. Among the leading candidates for

the conserved gene order are coregulation and remedia-

tion of nuclear/cytopiasmic genome conflict.

Model species are indispensable for understanding

M,\T at the molecular level and provide all of the infor-

mation on how MAT genes function. However, in order

to fully unravel the marvelous mysteries of mushroom

mating we will need to explore MAT across the phylo-

genetic diversity of mushrooms. Observations, such as

,rr.rshroo- species having whorled (multiple) clamp

connections per septum in both homokaryotic and het-

erokaryotic mycelium (e.g., Stereum hirsutum) and het-

erothallic species lacking clamp connections altogether(e.g., Phanerochdete chrysosporium), give an indication

that a simplified mushroom maring system with MAT-

A/B-controlled developmental pathways may not hold

for the whole goup. Further, the organization of MATloci in homothallic species of mushrooms is unknown'As oniy a fraction of the edible and medicinal mush-

room species can be readily fruited, access to MAT loci

may provide information useful for breeding, produc-

tion, or genetic engineering (43). Despite the previous

notion tiat Schizophyllum commune was a member ofthe Aphyllophorales, molecular systematics suggests

that it is related to fleshy mushrooms (Agaricales), the

order to which Coprinus sensu lato also belongs (6)'

Thus, the molecular knowledge of MAT in mushrooms

is primariiy limited to one order' As we move deeper

towards the base of the mushroom phylogeny' it is

.d--

expected that the same genes will be used in MAT d.eter-mination; howeveq novel gene arrangements and func_tions are likely to be uncovered.

I am indebted to (Jrsula Kiies for her countless discussions onmating type in mushrooms. I thank her and Greg Bonito forcomments on the manuscript.

References1. Aimi, T., R. Yoshida, M. Ishikawa, D. p. Bao. and y. Ki_

tamofo.200s. Idenrilicarion and linkage mrpprng of rhegenes for the putative homeodomain p'rotein (tr"irl1

^iithe, pr,rtative pheromone receptor protein t o,rroiot""(rcb1) rn a bipolar basidiomycete , pholiota nameko. Cirr.Genet. 18:184-1.94.

2. Badrane, H., and a.ynr. 1.999. The divergence_homoge_nization duality in rhe evolution of the b"1 m"til,g ty"p.gene of Coprinus cinereus. Mol. Biol. Euot. 16:97SJ9g6.

3. Banuett, F. 1998. Signalling in the yeasts: an informa_tional cascade with links tu ih. fil"-.rrrous fungi. mirai.bbl. Mol. Biol. Reu.62:249-274.

4. Baptistaferreira, J. L. C., A. Economou, and L. A. Cas_selton. 1983. Mitochondrial genetics ' ol Copririi_

recombination of mitochondriaf genomes. Curr. Genet.7:405-407.

5. Berbee, M. L., and J. W. Taylor. 1993. Dating the evolu_tionary radiations of the true fungi. Can. J. Bot. 71:III4_1127.

6. Bodensteiner, P., M. -B]r-rder, J. M. Moncalvo, R. Agerer,

and D. S. Hibbett. 2004. pirylogeneric rela;ionshifis oicyphelloid homobasidiomy..t"s. "Mol.

p hy lctgenet. -Euctl.

33:501-515.7. Boidin, J. 197"1. Nuclear behavior in the mycelium and

the evolution of the Basidiomyceres, p. U,e_t+S. InR.H.Petersen (ed.), Euolution in the Higher Basidiomycetes.The University of Tennessee press, K"noxville.

8. Bortfeld, M., K. Auffarth, R. Kahmann, and C. W. Basse.

i00-4 The Ustilago maydis a2 mating_typ. lo.r, g.,",.,Iga2 tnd rga2 compromise parhogeniciiy i'n rhe ab,erice o[rhe-mirochorrdrirl p32 familt piorein Mrbl. plant Ccll16:2233-2248.

9. Branda, S. S., and G. Isaya. 1995. prediction ancl identifi_cation of new natural substrates of the yeast mitochon_drial inrermediate peptidase. J. Biot. Ct en . ZZOTZZS1|_27 37 l.

10. Brown, A. J., and L. A. Casselton, 2001. Mating in mush_rooms: increasing the chances but prolonging-the affair.T-rends Genet. 17 :393-400

11. Butler, G., C. Kenny, A. Fagan, C. Kurischko, C. Gail_lardin, and K. H. Wolfe. 200?. Evolutior, oith. MAT lo_cus and ir, Ho endonuclease irr yeasr species. ir"r. Nnit.Acad. Sci. USA 101:1632-1637.

12. Casselton, L. A. 2002. Mate recognition in fungi. Hered_ity 88:142-147.

13. Casselton, L. A., R._N. Asante-Owusu, A. H. Banham,C

^S. Kingsnorth, U. Kiies, S. F. O'Shea, and E. H. pardo.

1995. Mating type of ,sexual development in Coprinuscinereus. Can. .1. Bot. 7 3:5266_5272.

BesntouycETES: THE MusnRoovs

14. Casselton,,L. A., and N. S. Olesnicky. 199g. Moleculargenerics of mating recognition in basidiomycete fungi.MicrobioL. Mol. Biol. Rei. 62:5 s_70.

15. Cvrikovi, F., C. !e- Jirgilio, D. Manser, J. R. pringle,and K. Nasmyrh. j 995. Ste20_like protein'liinas.. nr."..jquired for normal localization of cell growth

"rrd fcr. cy

tokinesis in budding yeast. Genes Deu."9:Ig17_1g30.16. Dandekar, T., B. Snel, M. Huynen, and p. Bork. 199g.

Conservation of gene order: a fingerprint of proteins thatphysrcally rnreracr. Trends Biocbem. Sci. 23:324_329.

17. Davidson, R. C., C-. !. Nicholls, G. M. Cox, J. R. perfect,and J. Heitman. 2003. A MAp kinas. ."r.nj. ."-;;;;;of cell type specific and non-specific elements controlsmating and differentiation of the fungal pathogen Crypto_co ccus n e oformans. M o l. M i cr o b i o l.- 49i+ eg _4 g.s .

18. Day, P. R. i 9d0. The strucrure of the A mating-type locustn Coprinus lagoltus. Genetics 45:641_-651.

19. Drees, B. L., B. Sundin, E.Brazeau,J. p. Caviston, G. C.9!.1, W. Guo, K. G. Kozminski," n4. W f*, J. J.Moskow, A. Tong, L. R. Schenkman, A. l,l.fi."ri!,P. Brennwald, M..Lo.ngtine, E. Bi, C. ihan, p. Novick,C. Boone, J. R. pringle, T. N. Davis, S. Fieldi, a"d O. C.Drubin. 2001. A prorein.inreraction map for cell pol"riifdevelopmenr. J. Celt Biol. 154:S4q_57 I.

20. Eichler, E. E., and D. Sankoff. 2003. Structural dynamics ofeukaryofic chromosome evolution. Science 30I:j%_797.

21. Ferreira, A. V. 8., S. Saupe, and N. L. Glass. 1996.Tran_scriptional analysis of the mtA idiomorph ol Neurospiiacrassa identtfres two genes in addition to mtA-1. Mol.G en. Genet. 250:7 67 -77 4.

22. Fischer, G., E. P. C. Ro.cha, F. Brunet, M. Vergassola, andB. Dujon. 2005. Highiy variable rates of ?.""-. r.._arrangements between hemiascomycetous yeit lineages.PLoS Genet.2:e32.

23. Fowler, T.J., M. F. Mitton, E. I. Rees, and C. A. Raper.2004. Crossi|g the bourrdary between the Ba and Bpmating-rype loci in

^Schizophyllum commune. FungalGenet. Biol. 4l:89-f01

24. Fowler, T.J., M. F. Mitton, L. J. Vaillancourt, and C. A.Raper. 2001. Changes in mate iecognition th;or;h olr;;-ations of pheromones and receptois in the muj'tisexualmu-shroom lungus Schizophyllui commune. Genetics 15g:1 491_1 503.

25. Fraser, J. A., S. Diezmann, R. L. Subaran, A. Allen, K. B.Lengeler, F. S. Dietrich, and J. Heitman. 2004. Conver_gent evolution of chromosomal sex-determining regionsin the animal and fungal kingdorns. pLoS Biol.-2,2-Z+S_2255.

26. Fraser, J. A., and 1.. Heitman. 2006. Ser, MAT, and the:rq"tigl of-fungal virulence, p. 13_33. In f. Heitman,S. G Filler, J. E. Edwards, Jr.,'and a. p. lf".fr.LL t.J.i,Molec.ular Priaciples of Fungai pathogenesis. ASM press,\Tashington, DC.

27. Giasson, !.,. C. A, Specht, C. Milgrim, C. p. Novotny andR. C. Ullrich. 198q. Clorring anJcornparrson of Aa mar_ing-type alleles of the basidi,omycere Schizophylt"; ;;;mune. Mol. Gen. Genet. 21,8:72_77.

28. Griitzner, F., H. R. Crollius, G.. Liitiens, O. Jaillon, J.Weissenbach, H. H. Ropers, and T. 'Uait.

Z.OOZ. fou,

43.

44

45.

46.

29.

30.

19. MAT rN MusHRool,t FuNcI: BEYoND Monll Spscrss

hundred million years of conserved synteny of human Xpand Xq genes on three Tetrdodon chromosomes. GenomeRes. 1,2:131.6-1.322.

Halsall, J. R., M. J. Milner, and L. A. Casselton. 2000.Three subfamilies of pheromone and receptor genes gen-

erate multiple B mating specificities in the mushroomCoprinus cinereus. Genetics 154:1115-1123.Hibbett, D. S., and M. Binder. 2002. Evolution of com-plex fruiting-body morphologies in homobasidioml'gs1s5.Proc. R. Soc. Lond. B 269:1963-1'969.

31. Hibbett, D. S., and M. J. Donoghue' 200'l . Analysis t,fcharacter correlations among wood decay mechanisms,mating systems, and substrate ranges in homobasidio-mycetes. Syst. Biol. 50:215-242.

32. Hiscock, S. J., and U. Kiies. 1999. Cellular and molecularmechanisms of sexual incompatibility in plants and fungi.Int. Reu. Cytctl. 193:165-295.

33. Hogberg, P., A. Nordgren, N. Buchmann, A. F. S. Taylor,A. Ekblad, M. N. Hogberg, G. Nyberg, M. Ottosson-Lofvenius, and D. J. Read. 2001. Large-scale forestgirdling shows that current photosynthesis drives soil res-

piration. N at ur e 411:7 89 -7 92.

34. Hopple, J. S., Jr., and R. Vilgalys. 1999. Phylogenetic re-lationsl.rips in the mushroom genus Coprinus and dark-spored allies based on sequence data from the nucleargene coding for the l:'rrge ribosomal subunit RNA: diver-gent domains, outgroups, and monophyly. Mol. Phylo-genet. Euol. 13:1'-1'9.

3.5. Hurst, L. D., and W. D. Hamilton. 1'992. Cytoplasmic fu-sion and the nature of sexes. Proc. R. Soc. Lond. B 247:r89-194.

36. Hurst, L. D., C. Pal, and M. J. Lercher.2004. The evolu-tionary d-vnamics of eukaryotic gene order. Nat. Reu.

Genet. 52299-310.

37. Isaya, G., W. R. Sakati, R. A. Rollins, G. P. Shen, L. C.Hanson, R. C. Ullrich, and C. P' Novotny. 1995. Mam-malian mitochondrial intermediate peptidase-structure-function analysis of a new homolog from Schizophyllumcontmune and relationship to thimet oligopeptidases.Genomics 28:450-461.

38. James, T. Y. 2003. The Euolution of Mating-Type Genes

in the Mushroctm Fungi (Homobasidictmycetes) . D;keUniversitv. Durham, NC.

39. James, T. Y., U. Kiies, S. A' Rehner, and R. Vilgalys.2004. Evolution of the gene encoding mitochondrial in-termediate peptidase and its cosegregation with the An-rating-type locus of mushroom fungi. Fungal Genet.Biol. 1l:381.-390.

40. James, T. Y., S. R. Liou, and R. Vilgalys. 2004. The ge-

netic structure and diversity of the A and B mating-typegenes from the tr:opical oyster mllshrctom, Pleurotusdj amor. Fungal G enet. Biol. 41:8 1 3-825.

41. .fames, T. Y., P. Srivilai, U. Kiies, and R. Vilgalys' 2006.Evolution of the bipolar mating system of the mushroomCoprinellus disseminatus from its tetrapolar ancestors in-volves loss of mating-t1,pe-specific pheromone receptorfunction. G enetics 17 2:1 877 -1 89 1.

42. Kirk, P. M., P. F. Cannon, J. C. David, and J. A. Stalpers(ed.). 2001. Ainsworth 6 Bisby's Dictionary of the

329

Fungt, 9th ed. CAB International, Wallingford. IlnitedKingdom.Kothe, E. 2001. Mating-type genes for basidiomycetestrain improvement in mushroom farming. Appl. Micrtt-biol. B iote chno l. 5 6:602-612.Kiies, U. 2000. Life history and developmental processes

in the basidiomycete Coprinus cinereus. Microbiol. Mol.B iol. Reu. 64:3 I 6-3.53.

Kiies, U., and L. A. Casselton. 1993. The origin of multi-ple mating types in mushrooms. .1. Cell Sci. 104:227-230.

Kiies, U., B. Grittgens, R. Stratmann, W. V.J. Richardson,S. F. O'Shea, and L. A. Casselton. 1994. A chimerichomeodomain protein causes self-compatibility and con-stitutive sexual development in the mushroom Coprinuscinereus. EMBO l. 13:40 54-40 59.

47. I(iies, U., T. Y. James, R. Vilgalys, and M. P. Challen.2001. The chromosomal region containing pab-1 , mip,and the A mating type locus of the secondarily horrrothal-lic homobasidiomycete Coprinus bilanatus. Curc. Genet.39:16-24.

48. Kiies, U., A. M. Tymon, W.V.J. Richardson, G. May'P. T. Gieser, and L. A. Casselton. 1994. A rnerrng-tvpefactors of Coprinus cinereus have variable numbers ofspecificity genes encoding two classes of homeodomainproteins. Mol. Gen. Genet. 245:'45-52.

49. Larraya, L. M., G. Perez, E. Ritter, A. G. Pisabarro, andL. Ramirez. 2000. Genetic linkage map of the edible ba-sidiomycete Pleurotus ostredtLts. Altpl. Enuiron- Micro-biol. 66:5290-5300.

-50. Lee, N., G. Bakkcren, K. Wong, J. E. Sherwood, and J' W.I(ronstad. 1999. The mating-tvpe and pathogenicitl' locus

of the fungus Ustildgo hordei spans a 500-kb region.Proc. Narl. Acad. Sci. U.SA 96:15026-15031.

51. Lengeler, K. 8., D. S. Fox, J. A. Fraser, A. Allen, I(' For-rester, F. Dietrich, and J. Heitman, 2002. The meting ti pe

locus of Cryptococctts neoformans: a step in the evolutionof sex chrorrosomes. Eukaryot. Cell 7:704-71'8-

52. Lengeler, K. 8., P. Wang, G. M. Cox, J' R. Perfect' and

J. Heitman. 2000. Identification of the IdATa Inating-typelocus of Cryptococcus nec:formans reveals a serotype AMAIa strain thought to have been extirrct. Proc' Natl.Acad. Sci. USA 97:/'4455-14460.

53. Leveleki, L., M. Mahlert, B. Sandrock, and M' Bijlkcr.2004. The PAK family kinase Cla4 is required for bud-ding and morphogenesis in Llstilago maydis. Mol. Micro-biol. 542396-406.

54. Liu, Y.-G., and R. F. Whittier. 1995 ' Thermal asymmetricinterlaced PCR: auton-ratable amplification ar-rd seqllenc-ing of insert end fragments from P1 and YAC clones forchronrosome walking. Genomics 25l.67 4-681'.

55. Lukens, L., H. Yicun, and G' May. 1996. Correlation ofgenetic ancl physical maps at the A mating-type locus ofC op r inu s ciner eu s. G en eti c s'l'4 I :1 47 1' -1 47 7'

56. Marion, A. L., K. A. Bartholomew, J. Wu, H. L. Yang,C. P. Novotnn and R. C. Ullrich. 1996'The Act mating-type locus o{ Schizophyllum commune: structure andfunction of gene X. Cttrr. Genet. 29:143-149.

-57. Martinez, D., L. F. Larrondo, N. Putnam, M. D' S. Gelpke,K. Huang, J. Chapman, K. G. Helfenbein, P. Ramaiya, J. C.

330

Detter, F. Larimer, p.. M. Coutinho, B. Henrissat, R.Berka, D. Cullen, and D. Rokhsar. )oO+. C.noro. .._quence of the lignocellulose degrading f"gr- iion,r^o^c h-qe1 e

_gh r y s ct s1) o r ium rtrnir,, {p 78.'N at. B i o t e c h n o L) ).Aq <--7.|n

58. May, G., F. Shaw_, H. Badrane, and X. Vekemans. 1999.Tl.re signature of balancirg- selecrion: fr"gnl _;ii", c;_patrbilrty gene evolutior.r.l,rnr. Nott. iiia. Sri. iie-Si,9 172_9177.

59. Moncalvo, J.-M., $ Vllgalys, S. A. Redhead, J. E. John_1on, T. Y. James, M. C. Aime, V. Hofstetter,-S. V.id.rirr,E. Iarsen, T. J. Baroni,,R. G. Thorn, S. 1u*Urrorr, ft.Cl6menqon, and O. K. Miller, Jt.20Ot:Orr. hrrrdred'"rrJsevenreen clades of euagarics. Mol. phylogenet. Euctl. 23:357_400.

50. Murphy, J. F., and.O. If. Mi!e.r, Jr. 1997. Diversity andlocal distribution ef 63sing alleles"in Moror*irltr,, ironi-cutus and Collybia,submzda (Basidiomycetes, Agaricales).Can. J. Bot. 75:8_17.

61. O'Brien, ]{: E., J. L. parrent, J. A. Jackson, J. M. Mon_i1"?: .ild *; Vilgalys. 20 05.' iungai .o_Lr,.,i,y anatysisD) targe-scrle \equencing of environrnental

'.ampler.Appl. Enuiron. Microb. 71-:554,1_5550-'

62. Ochman, H., A. S. Gerber, and D. L. Hartl. 19gg. Geneticapplications of an inverse polymerase .truiu ..r.iiorr.Cenetics 12O:621_623.

63. Olesnicky, ry. S., I J. B1own, y Honda, S. L. Dyos, S. J.Dowell, and L. A. Cisselton. 2000. Self_colp"tibi. A *.1_

l, f.r) i n

_C"pl'lnrz1 1 j r ! 3]

r^e red pt e ro m on e- recepro r s pec i _nerlre\. Genet ics 156: I 02i_l 0JJ.6,1. O'Shea, S. F., p. T. Chaure, J. R. Halsall, N. S. Olesnicky,A

-Leibbrandr, I. F. Conn-erron, u"d L. A. Casselton.1998. A large pheromone and receptor gene complex de_termines multiple {.T"jrlg type specific"ittes tn ioprinLiscinercus. Cenetics l4g: l0g l_l'090.

65. Ramer, S. W., and R. W. Davis. 1993. A domrnant trunca_tion allele identilles a gene, S.TE20, that encodes " ;;;;tive protein kinase necessary for mating ;niorrhoroiyiu

cereuisiae. proc. Natl. Acad. Sci. USA\0t452_456.66. Raper,.J. R. 1966.

Qgngtics of Sexuality in Higher Fungi.Ronrld Pre,s. New york, Ny.67. Raper, J. R., M. G. Baxrer, and A. H. Ellingboe. 1960.The genetic srru*ure of tie incomp;rrb;il;y faoors ofS^chizophyllum communei the A factcr'i F ro). Nott. eroa.Sci. USA 46:833-842.68. Raper, J. R., and A. S..Flexer. 1971. Maring sysrems andevolution of rhe Basidiomyceres, p. ti/_ial .12 R. H.

P_etersen (ed.,), Euolution ilt thc Highu Basidiomy,cetes.University of Tennessee press, I(noxiiil;.

69. Riquelme, M., M. p. Challen, L. A. Casselton, and A. J.Brown. 200.!. The origrn of multiple B ;;;; specificitiesin Coprinus cinereus.benetics rio, r roj_i ilc.

70. Riihr, H., U.-Kiies, and U. Stahl. 199g. Organelle DNA ofpJnrr:^"li fungi: inheritalce and ,...,-Bi,lurlo n. prog.Bot. 60:39-87.

71. Sato,.K., T. Nishio, R. Kimura, M. Kusaba, T. Suzuki, K.Hatakeyama. D. J. Ockendon. and y. Sriir.'U OOZ. Coevo-lutron of the S_Jocus genes SRK, StG and SpTllSCR inBrassica ctleracea and B. rapa. Genetics tiZr*gZt_g+O-

BesprontycETEs: THE MusHnoovs

72. Saville,B.J., y. Kohli, and J. B. Anderson. 199g. mtDNArecombination in a natural population. proc. Natl. Acad.Sci. USA 95:1331_1335.

73. Schirawskl, J., B. Heinze, M. Wagenknecht, and R. Kah_mann.2005. Maring rype loci of-spctrisoriu* rrilioru*,novel pattern with three a and mulripl. 1, ,p".ifi;i;;;;.Eu karyot. Cel I 4: I 3 l-_l 327.

74. Simchen, G. 1967. Genetic conrrol of recombination andthe incompatibiliry system in SchizofiyUu* ,u*ii,ri.Cenct. Res. 9: I 9 5-2 I 0.

75. Smulian, A. G., J. Sesterhenn, R. Tanaka, and M. T.Cushion.2001. The.sre3 plreromorr. i...p,o, gene ofPneumctcystis carinii is surrounded by , .l..rire, "?

;i;";itransducrion genes. G enetics 1 57:997-1002.76. Specht, C. A. 1996.Iso1ation of the Bcv and .Bp maring-rype

loci ol Schizctphyllum commune. C"ri. C))nt.28:374_379.77.

ly9c.h,t, C. A., M. M. Stankis, C. p. Novotny, and R. C.Ullrich. 1994. Mapping rhe hererogeneous i)NA ..gionthat derermines the niie A" -"ti"i_ryp" specificitiei oiSchizophyllum commune. G enetics iziizoy_lt+.78. S-t1n_kis, M.M., C._A. Specht, H. yang, L. Giasson, R. C.Ullrich, and C. p. Novotny. 1992.Th:A;mating locus of

Sc.hiz9phyllum commune encodes two dissimilar multial_leiic homeodomain proteins . proc. Natl. Acad. Sci. USA89:-lo9--l-3.

79. Stoll, S:, D. Begerow, and F. Oberwinkler. 2005. Molec_rrlar phylogen y of tJstilago, Sporisorium, and relared raxabased on combined arrrlyses ol rDNA sequence,. M),col.Res. I 09:J42-i5o.

80. Suyama, M., and p. Bork. 2001. Evolution of prokaryoticgene order: genome rearrangement, i" .t"r.ty ,.i"i.Jspecies. Trends Genet. 17:10_13.

81. Turgeon, B.G., q.Bohlmann, L. M. Ciuffetri, S. K.Christiansen, G. yang, W. Schafer, n"a-O. C. yoder.1993. Cloning and analysis of the mating_type genes fromCochliobolus heterostr:ophus. Mol. G'eil. Genet. 23g:270_284.

82. Ullrich, \..C., ryd J. R. Raper. 1974. Number and distri-Dutron ot hrpotar incompatibility factors jn Sistotremabrinkmannii. Am. Nar. f fiS,SOg_Sf S.-

--"

83. Urban, M., R. Kahmann, and M. Brilker. 1996.Thebial-lelic a mating type locus'"1 Urtitoti *iydir, ,._l,"nr, ofan additional pheromone gene iniicate evolution from an'rultiallelic ancesror. Mo l.

"G e n. C e n et. Z S O :+ t 4_420.

84. Waalwijk, C., T. van.der Lee, I. de Vries, T. Hesselink, J.Li:r_l,ilO G. .H. {. Kema. 2004. Sy,,r.nl' in rorisenicrusdnum specte\: the f umonisin gene clurter and rhe'nrat.f1C^ty_q. region as examples. nir. f . iilii pathol. 1t0:533-544.

85. W^a1ser, P.J., M. Hollenstein, M. J. Klaus, and U. Kiies.i0?l ,G:l::,i. ana.ll,is of basidioiycere funsi. p. 5e_00./r N. J. Talbot (ed.), Molecular and Cell Bi;;;y il fiimentous F_ungi: a pra,ctical Approach. Oxford Tj"#.;;;;yPress, Oxford, United Kingdo;.

86. Wang, P., C. B. Nichols, M. B. Lengeler, M. E. Cardenas,G. M. Cox, J. R. perfect,.rnd;. Heltman. )OOZ. l,tutingitype-specific and nonspecific pAK kinases play shar.ed anddiv.ergent roles in Cryptococcus ,"oforioir. Eukaryot.Cell l:257-272.

19. MAT rN MusHnoon Ful.lcr: BEYoND MooEI- Sppctps

Wendland, J., L. J. Vaillancourt, J. Hegner, K. B. Lengeler,K. J. Laddison, C. A. Specht, C. A. Raper, and E. Kothe.1995.The mating-type locus Bal ol Schizophyllum com-mune contains a pheromone receptor gene and putaqivepheromone genes. EMBO J. 14:5271-5278.Whitehouse, H. L. K. 1.949. MuJtrple allelomorph het-erothallism in the fungi. New Phytol. 48212-244.Yan,2., C. M. Hull, J. Heitman, S. Sun, and J. P. Xu'2004. SXIIa controls uniparental mitochondrial inheri-

331

tance in Cryptococcus neoformans. Curr. Biol. l4:R743-R744.Yan, 2., and J. P. Xu. 2003. Mitochondria are inheritedfrom the MAIa parent in crosses of the basidiomycete fun-g:ls Cryp t o co c cw s n e of orman s. G eneti cs 163 :13 I 5 -I 32 5 .

Yun, S.-H., M. L. Berbee, O. C. Yoder, and B. G. Tur-geon. 1,999. Evolution of the fungai self-fertile reproduc-tive life style from self-sterile ancestors. Proc. Natl. Acad.Sci. USA 96:5 592-5 597 .

90.

9t.

87.

88.

89.