J. Gener., Vol. 75, Number 3, December 1996, pp. 325-339. ((') Indian Academy of Sciences

Fungal transposable elements: generators of diversity and genetic tools

M. J. DABOUSSI Institut de G6n6tique et Microbiologie, Universit~ Paris-Sud, 91405 Orsay, France

Abstract. Representatives of several classes of transposable elements (TEs) have been characterized in a broad range of fungal species. The studies indicate flint these elements are ancient and ubiquitous components of fungal genomes. Some of these elements have been shown to actively affect gene structure and function in several ways: inactivation of gene expression upon insertion, modification of the nucleotide sequence through excision, and probably by inducing extensive chromosomal rearrangements. The ability o['TEs to generate a high degree of genetic diversity may therefore be important in the evolution of the fungal genome. TEs also have many potential applications in genetic research, including inserfional mumgenesis and population fingerprinting, as well as gene transfer within and between species. All these genetic approaches are important as tools in studies &molecular biology and evolution of fungal species, many of which lack a functional sexual cycle.

Keywords. Transposable elements; mutagenesis; filamentous fungi; genetic application.

1. Introduction

Filamentous fungi have proved to be valuable in the development of many fundamen- tal ideas: these include the one gene-one enzyme concept (Neurospora crassa), models for recombination (Ascobolus immersus), and regulation of gene expression (Aspergillus nidulans). However, there is one area where fungi, particularly filamentous fungi, were long silent; that is the area of transposable elements (TEs). It is now more than four decades since Barbara McClintock developed the concept of transposable genetic elements through genetic studies in maize. Since then, such elements have been found in a wide range of organisms from to higher (see Berg and Howe 1989 for a review) and are presumed to be ubiquitous. However, with the notable exception of Ty elements in the yeast Saccharomyces cerevisiae (Boeke 1989), very little was known until a few years ago about the incidence and properties of TEs in other fungal species. This was probably due to the fact that the laboratory strains of the best-studied ascomycetes, A. nidulans and N. crassa, appear to be devoid of active transposons. In another ascomycete, A. immersus, genetic elements with properties of classical transposons associated wih genetic instabilities were reported but not characterized at the molecular level (Decaris et al. 1978). In the past few years our picture of fungal transposons has grown considerably as a result of work with species from different phylogenetic groups and ecological situations: pathogens, and industrial and field strains (see reviews by Oliver 1992 and Dobinson and Hamer 1993). Many different classes of TEs are now known in different species of ascomycetes (listed in this review), and TEs have recently been reported in phycomycetes (Avalos et al. 1996) and basidiornycetes (Gaskell et al. 1995), opening a door of opportunity for genetic research. The aim of this paper is to review progress made in the molecular-genetic analysis of these fungal transposons. Emphasis will be on what is known about their mechanisms of transposition and their consequences to 325 326 M. J. Daboussi the host genome, the spread of transposons within and between species, and finally the potential use of transposons for genetic research.

2. Structural features of fungal transposons

Fungal transposons have been identified using a variety of strategies (figure 1) that exploit the two main characteristics of TEs: (i) they are present in multiple copies in the genome and thus can be discovered by analysing dispersed repetitive sequences, and (ii) they can move from one position to another in the genome, and therefore can be trapped through their insertion in a selected gene. Using the first strategy, different transposon- like sequences have been recognized by comparison with the known transposon sequences from other organisms, although we do not know if these transposon-like sequences are still active. The most satisfactory way to screen for active transposons is the transposon trapping approach, which is based on spontaneous inactivation of a cloned gene. This strategy is usually applied to genes whose mutant phenotypes can be positively screened. Examples of this approach are the am gene coding for glutamate dehyd- rogenase which allowed the discovery of the Tad element in N. crassa (Kinsey and Helber 1989), and mutations in the nia gene encoding nitrate reductase which can be selected through resistance to chlorate (Cove 1976). Trapping using the nia gene has been very efficient in Fusarium oxysporum (Daboussi and Langin 1994) and has been successfully applied to other fungal species, leading to identification of at least eight different mobile elements. Finally, some other elements have been found using heterologous hybridiz- ation or have been discovered fortuitously through DNA sequencing. The TEs or sets of related sequences identified in fungal genomes reflect the whole spectrum of eukaryotic transposable elements. Their general structural features are summarized in figure 1. Representatives of the two classes of transposable elements are found in fungi: those that transpose by reverse transcription of an RNA copy of the element (class I) and those that probably transpose directly through DNA copies (class II) (Finnegan 1989; McDonald 1993). Class I elements are LTR (long terminal repeats) that belong mostly to the gypsy group: Foret and skippy in F. oxysporum (Julien et al. 1992; Anaya and Roncero 1995), grh and MAGGYin Magnaporthe grisea (Dobinson et aI. 1993; Farman et aI. 1996a), CfT-1 in Cladosporium fulvum (McHale et al. 1989, 1992), Boty in Botrytis cinerea (Diolez et al. 1995), Afutl in Aspergillus fumigatus (Neuv6glise et al. 1996a) and Mars4 in Ascobolus immersus (Goyon et al. 1996). Only Mars2 of A. immersus is a copia-like element (Goyon et al. 1996). Several non-LTR retrotransposons with structural features of LINEs (long inter- spersed nuclear elements) have been characterized: Tad, an element of N. crassa, the first reported in fungi (Kinsey and Helber 1989), MGR583 in M. grisea (Hamer et. al. 1989), Palm in F. oxysporum (Mouyna et al. 1996), Mars1 in A. immersus (Goyon et al. 1996), and CgT-I in ColIetotrichum gloeosporioides (He et al. 1996). SINE (short interspersed nuclear element)-like elments have also been recog- nized: Nrsl in Nectria haematococca (Kim er aI. 1995), MGSR1 and Mg-SINE in M. grisea (Sone et al. 1993; Kachroo et aI. 1995), as well as Eg-RI and EGH in Erysiphe (Rasmussen et at. 1993; Wei et aI. 1996). Class II or DNA transposons of different types have also been isolated, most of them after transposition in the nia gene. Some of them are representatives of families Fungal transposable elements 327

Class h Retroelements

LTR-RETROTRANSPOSON8 type Example

5' LTR ~Ta~ pOt ~ ? 3' LTR Feint t~) and skippy f3; iF. e~ysporumJ gypsy grh O) and MAGGYCt.;) (M. grisee) CtT-I r futvum); Boty ~r tin~ Afu110) (A. fumJg~tus); Mars4 (.4, immersus) 5' LTR gag pol 3' LTR copia Mars2 (~ (A. immersus)

non LTFI-RETROTRANSPOSONS 5. gag? 3' Tad ~2)(N, erassa); PalmOJ iF. oxysporum) MGR583 (1)(M, grisea); Mars1 (~)(A. imrnersLts) CgT1 (IJ(C. gtoeospo[ioides)

Nrs l O) (N. haematocc~eca) MGSRI {=)and Mg,BtNEII) (M. gfises) Eg-R1 ilt(E, graminis) and EGH (g(E. hordei)

Class Ih DNA transposons

Tc I/mariner impatel2) (F.oxysporumy Ant1 (~(A. niger)

IR IR Fort {2~arid For2 (2~ (F.ox3,sporurn);Ta~ {2>(A.nigsO Pot2 (1~and MGR586 (~) (M. Grisea); Flipper {~J(B, cinerea); Puntr~ iN. crassa)

A c-like Ascot ft) and Tasce I~) (A. imrnemus); restless (lilT. Inflaturn)

Hop (a~ (F.oxysporctm) UnreLated Pc~ (P. chrysosporlum)

Unclassified Guest (S)(N, cr~a)

Figure. 1. Schematic representations of the structural features of fungal transposable ele- ments. For purposes of clarity, the overall size of the elements is not to scale. Arrows denote the short duplication (TSD) generated in the host genera~ during transposition. There are two types of class I elements. LTR-retrotransposons resemble in having long terminal direct repeats (LTR) flanking an internal domain encoding proteins analogous to the gag and pol retroviral gone products. They are divided into two groups, the gypsy and the copia types, on the basis of the order of functional domains in the pot gone (PR, protease; RT, ; RH, RNAase H; 1N, integrase) and the presence or not of a third ORF in a position similar to that of the viral env gone but with no structural or functional homology to it. The non-LTR-retrotransposons lack terminal repeats and carry a poly(A) taft at their 3' ends. They fall into two families similar to the long and short interspersed nuclear elements described in mammals and referred to as LINEs and SINEs respectively. LINE-like elements commonly possess two long ORFs, with similarities to ~]c~,]as well as RT and RH genes, and are often deleted in their 5' ends. SINE-like elements are short elements wl~ich contain an internal RNA polymerase Ill promoter with a bipartite structure represented by the A and B boxes and which rely on RT for mobilization but do not themselves encode the enzyme. Class II elements share the same basic design of short inverLed terminal repeats (IR) bordering one or more ORFs, encoding transposase, an enzyme required for their own transposition. Some very short elements with IR and potential to form a hairpin structure are difficult to classify at present. Numbers within brackets indicate the strategy used to identify the transposable element: (1) isolated by cloning band of repetitive DNA; (2) isolated after transposition within the nia gone, with the exception of Tad, which was transposed into the am gone, and Ascot, which preexisted as an insertion in the b2 gone; (3) isolated by differential screening with antibodies against RT; (4) isolated using a CfT-1 heterologous probe; (5) identified through gone sequcncing. described in other organisms: impala (Langin et at. 1995) and Anti (Glayzer et al. 1995) are clearly related to Tel from Caenorhabditis elegans and maritwr from Drosophila mauvitiana, a superfamily of elements widely distributed from insects to mamrnals 328 M. J. Daboussi

(Radice et al. 1994; Robertson 1995). Ascot and Tasco (Colot et al. 1995; C Goyon, personal conamunication) as well as restless (Kempken and Ktick 1996) are elements of the hAT transposon family such as the maize activator Ac (Calvi et al. 1991). Others are members of new families described so far only in fungi. This is the case for the Fotl-like elements including For1 from F. oxysporum (Daboussi et al. 1992), Pot2 and MGR586 from M. grisea (Kachroo et al. 1994; Farman et al. 1996b), Punt from N. crassa (Margolin 1995), Vader/Tan from A. niger (Amutan et al. 1996; NyyssSnen et al. 1996), and Flipper from B. cinerea (Levis et al. 1996). Other, less-characterized transposable elements are known which may represent new families: Hop in F. oxysporum (Daboussi and Langin 1994) and Pcel fl'om Phanaerochaete chrysos- porium (Gaskell et al. 1995). In addition, an element from N. crassa, called Guest (Yeadon and Catcheside 1995), has characteristics of the unclassified Tourist family described in many cereal grasses (Bureau and Wessler 1994). Interestingly, as seen in figure 1, some fungal species are very rich niches for these elements. This is best illustrated by the plant pathogens F. oxysporum and M. grisea, in which an assortment of the major types of TEs has been found in the same strain, and representing a substantial fraction of their genome. The presence of many types of TEs in the fungal genome raises several questions: How do they function? How do they interact with their host? How do they evolve? From where do they originate?

3. Expression and transposition of transposable elements

Despite the wide distribution of TEs, relatively little is known about the mechanism and regulation of transposition. This is probably due to the fact that many TEs appear to be remnants of transposons and because in some sexual strains mechanisms exist that inactivate repeated DNA sequences. Although some fungal TEs contain most of the genetic information necessary for their own transposition, demonstration of transposition has been obtained only in a few cases. For retroelements that transpose through a mechanism involving an RNA inter- mediate, direct evidence for active transposition has been provided for the Tad LINE- like element of N. crassa and the fosbury (homologous to the MAGGY dement) of M. grisea (Shull and Hamer 1996), which were isolated after their insertion in cloned genes (figure 1). For other elements, indirect evidence for activity has been provided by the structural integrity of cloned elements, the presence of -like particles associated with reverse transcriptase activity, the increase in copy number, and the absence of polymorphism between the 5' and 3' LTRs, suggesting that some elements have probably been active in recent times. This is the case for the retrotransposons of the gypsy group, CfT-I from C.fulvum (McHale et al. 1989, 1992), skippy from F. oxysporum (Anaya and Roncero 1995), and grh and MAGGYfrom M. grisea (Dobinson et al. 1993; Farman et al. 1996a). Transposition by reverse transcription has been demonstrated only in the case of Tad. Like other LINEs, Tad is thought to transpose through reinsertion of the product of reverse transcription of an RNA copy of the element. This has been demonstrated by Kinsey (1993) using the strategy defined by Boeke et al. (1985) for Ty retrotransposition, that is the introduction of an intronic sequence within the element. The different steps are summarized in figure 2A. An artificial for Neurospora has been constructed and placed directly within the coding sequence of the active TadI-t Fungal transposable elements 329

(A) Retrotransposition: e, g. Tad (B) Cut-and-paste transposition: e. g. impala

ntroduct oft of an artif c al intron R R R R wt M nto Tad1-1

ntrodpct on of the Tadl-li in an am stra n

Construction of a forced heterokaryen

Serial transfers on minimal medium

Selection of lye" strain 'V' ?~

Ted tranposes between nuclei via an RNA intermediate since the Impale tranpoaee by excision of a copy from a donor site intron has been removed from transposed copies of Tad1-1/ to a recipient site {replicated or unrepilosted)

Figure 2. Schematic representation of transposition mechanism for two types of fungal transposons. (A) Transposition of the LINE-like element Tad fl'om N. crassa via an RNA intermediate. The introduced intron is represented by a triangle. The in the nucleus carrying the marked T,d element and the am chromosomal mutation in a forced heterokaryon are represented in grey. The chromosomes in the other nucleus, free of Tout elements and marked with the nuclear lys mutation, are represented in black. (B) Transposition of the hnpala element fl'om F. oxysporum through a DNA copy. Upper, Southern blot analysis of wild type (wt), mutant (M) and revertant (R) strains showing the reinsertion of the impala copy inserted within the a~faD gone (white star) to new gcnomic positions (black star). Lower, model of transposition which explains the gain (1), the loss (3), or the movement (2) of an impala copy, depending on replication and location of the recipient site relative to the niaD donor site (black boxes). element. Transformants containing the Tad element, marked wi~h the intron, were placed in forced heterokaryons with strains lacking Tad elements. In these hetero- karyons, nuclei are forced to exist in a common cytoplasm without fusion. These experiments have clearly shown that the nuclei of naive nuclear type rapidly acquired Tad elements. Examination of the transposed Tad elements showed that the intron was precisely spliced during transposition. This confirmed that Tad transposes via an RNA intermediate and that it can invade a genome free of elements. For the second class of TEs, assumed to transpose directly through DNA copies, active transposition has been deduced in most cases through their mobility (see figure 1), or in the case of restless through transcriptional expression and alternative mRNA splicing (Kempken and Ki,ick 1996). These elements can also excise and the variation in copy number and genomic positions associated with transposition events can be followed easily when elements are present in the genome in low copy number (exemplified by impala in F. oxysporum). Analysis of revertants from the original nia unstable allele indicates that impala transposes via excision of a copy from a donor site and is frequently reinserted elsewhere in the genome (figure 2B). Like other class II 330 M.J. Daboussi elements, this element can also increase in copy number (one additional copy in the mutant relative to wild type). Maize AciDs elements do so by transposing during chromosomal DNA synthesis and moving from replicated to unreplicated DNA (Fedoroff 1989; Dash and Peterson 1994). The same could be true for impala elements. Although the precise mechanism by which these fungal DNA transposons transpose is not known, different models of transposition can be drawn from the analysis of nucleotide changes, the 'footprints', left upon excision (see section 4.2).

4. Mutational effects of transposable elements

Transposable elements are agents of genome restructuring and mutagenesis in all organisms (Berg and Howe 1989). The ability of fungal TEs to induce mutations depends on their intrinsic capability of transposing within their host genome. Some of them have been shown to alter the host genome in several ways. They can promote changes in gene expression, in gene sequence, and probably in chromosome structure.

4.1 Mod!fication of' 9erie expression

In most cases insertion of a TE within or adjacent to a gene creates a null phenotype because the element blocks transcription of nearby genes or alters the pattern of transcription. In N. crassa, one transposition event placing the Tad element in a position upstream of the transcription start sites of am creates an unstable allele. Reversion depends on DNA methylation within and upstream of Tad, indicating that am expression is controlled epigenetically by the methylation state of the Tad element (Cambareri et al. 1996). Alteration of transcription of the target gene has been demonstrated in F. oxysporum with mutants resulting from the insertion of Fott elements in an intron of the niaD gene (Daboussi and Langin 1994; Deschamps et aI., submitted). In the mutants different transcripts, all shorter than the wild-type tran- script, are observed. These truncated transcripts are all composed of sequences fi'om the element and the target gene, indicating that FotI contains polyadenylation signals, termination signals and sequences that can be used as alternative promoters. In view of these findings, we can imagine that some insertions of Fotl can impose new temporal and spatial patterns ofgene expression that may have profound effects on the evolution of the host genome.

4,2 Modification of gene sequence

Transposable elements that excise often generate an altered DNA sequence at the donor site, the so-called 'footprint'. The length and sequence of the footprint depends on the element but also on the constraints of the selection for revertants. In the case of For1 elements inserted in an intron of the niaD gene, the footprint consists often of the same three or four base pairs, likely the duplication TA plus one or two nucleotides from one or the other end of For1 (Daboussi et al. 1992). When Foil is inserted in niaD exons, excisions leading to a functional protein are infrequent and correspond either to precise excision or to a 3-bp footprint leading to one amino acid more in the protein. Different models have been proposed to explain the nature of footprints. One is the gap FungaI transposable elements 331 repair model proposed by Engels for the excision of the in Drosophila (Engels etal. 1990) and demonstrated to be applicable for Tcl excisions in C. elegans as well (Plasterk 1991; Plasterk and Groenen 1992). In this model excision leaves a double- strand break which is repaired using tb_e homologous chromosome or the sister chromatid as a template. In some cases repair can be interrupted, resulting in the generation of small footprints containing only traces of the element. Given the similarities of the Fotl and Tcl systems, For1 is assumed to transpose via a gap repair mechanism using the sister chromatid as a template in somatic cells.

4.3 Modification of chromosome structure

Transposable elements have been found at the origin of numerous types of chromo- some rearrangements such as deletions, inversions and translocations in yeast, Drosophila and man (Rouyer etal. 1987; Boeke 1989; Montgomery et al. 1991; Sheen et al. 1993). Many of these chromosomal rearrangements are the consequence of the action of cellular homologous recombination upon families of repeated sequences which are widely scattered around the genome. Karyotypes of fungi have been shown to be quite variable between strains in several plant and human pathogens (reviewed by Skinner etal. 1991; Kistler and Miao 1992). Karyotypic variation results fl'om length differences in separated chromosome-sized DNA molecules and has been shown to involve translocation, deletion and duplication (Skinner etal. 1991; Kistler etal. 1996). Although most of these rearrangements appear to be neutral variation (Kistler and Miao 1992; Talbot etal. 1993), some of them may account for altered phenotype (Tzeng et at. 1992; Magee 1993). The precise rnechanisms for the observed karyotypic variation are currently not understood. However, con- sidering the numerous families of TEs in some strains, exemplified by F. oxysporum and M. grisea, each family with many copies closely conserved (Daboussi and Langin 1994), we favour the idea that interchromosomal and intrachromosomal ectopic exchanges between these elements and the subsequent chromosomal rearrangements could account for the exceptional karyotypic variation observed. Some situations are indica- tive of involvement of repeat sequences in chromosomal rearrangements. In Candida translocation seems to occur at hot spots characterized by the RSPI (Iwaguchi etal. 1992). In M. grisea the instability of the Bufand PWL2 loci is assumed to result from recombination of repetitive sequences leading to deletion of genomic segments including these loci (Valent and Chumley 1994). In Y. oxysporum analysis of the chromosomal distribution of the different families of TEs and reconstitution of the physical map ofa polymorphic chromosome by contigs are being performed to provide information on which repeated elements are involved and how frequently (Davi6re etal. 1996).

5. Silencing processes as control of transposon activity

It is now clear that fungal TEs are agents of genomic changes that could have many deleterious effects either directly by disrupting or modifying the activity of genes upon insertion or indirectly by providing substrates for numerous types of chromosome rearrangements. However, in some fungal species their activity can be affected by silencing processes. 332 M. .l. Daboussi

The processes called RIP (repeated-induced point mutation) in N. crassa (Selker and Stevens 1985; Selker et al. 1993) and MIP (methylation induced premeiotically) in A. immersus (Goyon and Faugeron 1989; Rhounim et al. 1992) inactivate duplicated sequences, linked or unlinked, artificial or natural, at a specific period between fertilization and karyogamy. This inactivation is associated with heavy cytosine methylation of the duplicated sequences. Although RIP and MIP share many features, they differ in their outcomes (Rossignol and Faugeron 1994). Genes silenced in A. immersus can recover expression whereas in N. crassa they cannot because cytosine methylation is accompanied by polarized transition mutations in which a:c pairs are replaced by A:T pairs. RIP and MIP have been envisioned as processes ensuring genome stability; firstly because they might control the mobility of transposons by inactivating theln and hence limiting the number of repeats, and secondly because RIP causes rapid divergence of repeated sequences which may prevent gross chromosome rearrangements (Kricker et al. 1992; Rossignol and Faugeron 1995). Consistent with this interpretation, only relics, but no active Tad transposon has been detected in the laboratory strain of N. crassa (Kinsey et al. 1994); in A. immersus the only mobile element identified, Ascot, has escaped MIP probably because of its very short size (Color et al. 1995). So RIP and MIP may be defence mechanisms aimed at TEs but they may not occur in all fungi since many types of active transposons have been identified. However, one should keep in mind that in mast of the species in which transposons have been found meiosis never occurs or occurs infrequently in nature. Consequently repeated sequences can be safe from RIP and MIP, if these mechanisms exist, as long as sexual reproduction does not occur in the species.

6. Vertical versus horizontal transmission

The data reported here indicate that each of the major groups of TEs can be found in fungi. However, questions dealing with how they evolved and how they spread within and between species have to be answered. Analysis of the distribution of some elements in fungal species and in some cases extensive sequence analysis have helped to cast some light on these questions. Two different mechanisms of genetic transfer are fiequently claimed to explain the distribution of TEs arnong species: (i) vertical transmision--like most genes, transposable elements can be transmitted vertically fi-om parent to offspring--, and (ii) horizontal transfer, which is defined as the transmission of genetic rnaterial by nonsexual means across taxonomic boundaries (for a review see Flavel11992 and Lobe et al. 1995). Some examples of these two contrasting, but not mutually exclusive, mechanisms are presented here.

6.1 Vertical transmission

Retrotransposons of the gypsy class have been found in the different ffmgal species analysed. The ubiquity of this class of retrotransposons throughout fungal species may be explained by assuming that the common ancestor of fungi also had retrotransposons and that during subsequent speciation events retrotransposons were transmitted vertically along with other components of the genome. These vertical relationships are revealed through phylogenetic analyses of their reverse transcriptase (figure3). The majority of fungal LTR retrotransposons are grouped together and form a Fungal transposable elements 333

Ingi I Cin4 F L1

copra~ Tyl del IGFG7

-- 17.6 gypsy Ty3 Mag Micropia DIRS1 MOMLV eao

Figure 3. Phylogenetic relationships between retroelements from fungi (shaded) and from different organisms by using amino acid sequences of their reverse transeriptase proteins. Sequences were obtained from GenBank, and from Xiong and Eickbush (1990) and Springer and Britten (1993). Delineation of boundaries for the reverse transcriptase protein correspond to that used by Xiong and Eickbush (1990). Sequence data were analysed by PAUP. monophyletic clade that is composed exclusively of fungal transposons, a pattern consistent with vertical inheritance.

6.2 Horizontal transmission

Some situations are indicative of horizontal transmission of fnngal transposons. First, the sporadic distribution of an element among species may reflect recent acquisition of this element through interspecific transfer: this is the case for Tad, which appears to be restricted to a few strains of Neurospora (Kinsey 1990), and for grh, found exclusively in a subgroup of M. grisea infecting Eleusine (Dobinson et al. 1993). Secondly, the high similarity between elements in distant species: the discontinuous distribution of For1 elements within the Fusarium genus and their extremely low level of divergence in two distant species compared to the divergence observed for nontransposable sequences suggest that this element could have been horizontally transferred through a rare interspeeific cross or by a parasitic vector of unknown origin (Daboussi and Langin 1994). TEs may also provide a mechanism for dispersing nontransposable sequences. The fact that the Ant1 element of A. niger was found to carry genomic sequences (Glayzer et al. 1995) shows how genes or parts of genes might be amplified and dispersed in 334 M. J. Daboussi genomes. If horizontal transmission of such a transposon occurs, the potential exists for host genes to be transmitted between species.

7. Transposable elements as genetic tools

Transposable elements also have great potential as tools both for increasing our understanding of biological processes and for modifying organisms for practical purposes. In plant-pathogenic fungi these elements have been exploited to trace lineages of clonal populations over broad geographical areas, aiding our understanding of population structure and epidemiology. Current interest in TEs also focusses on the development of a gene tagging system, a cloning strategy not yet available for filamentous fungi.

7.t Tracing populations with transposons

Little is known about the genetic structure of natural populations of phytopathogenic hmgi as many of them exist as composite species composed of morphologically indistinghuishable forms that infect different hosts. From an epidemiological point of view, it is important to define genetically related groups and to understand their interrelatedness. For this purpose, analysis of patterns of chromosomal distributions of TEs that transpose infrequently has proved to be very valuable in the study of the population biology of plant and human pathogens (Kistler et aI. 1991; Dobinson et al. 1993; Diolez et al. 1995; Neuv~glise et al. 1996b). For example, the MGR and grh elements from M. 9risen have provided insights into the evolution of the host-specific forms, demonstrating the clonal organization of rice-infecting populations and the possibility that new lineages of crop pathogens could emerge in independent lineages (Hamer et al. 1989; Dobinson et at. 1993). In F. oxysporum fingerprinting of isolates with the Palm element, pathogenic to oil palm tree, revealed that the recent appearance of the disease on oil palm in South America probably resulted from the introduction of an African isolate (Mouyna et al. 1996). Insights into the origin ofinvasive aspergitlosis have been provided by fingerprinting of strains of the opportunistic A.fumigatus isolated from patients and clinical environment (Neuv4glise et al. 1996b). Another potential application of TEs is that j unctions of particular insertions can be used as signatures which may provide valuable insights into the relationships between populations.

7.2 Development of a gene tagging system

Another use of TEs is the possibility to develop a transposon-based gene tagging system, a powerful method of gene isolation that does not require prior knowledge of the protein product. TEs act as insertional mutagens, and genes mutated in this way can be cloned with relative ease as sequences flanking the transposon insertion site are part of the gene of interest. This approach, successfully used in with different DNA transposons (see Balcells et aI. 1991; Walbot 1992), could be applied to fungi once autonomous copies of TEs have been cloned. The method would be of particular interest for cloning genes involved in developmental processes and pathogenicity because of the lack of knowledge of where they are expressed or when they are active. Fungal transposable elements 335

0 0 Nia'] @ [ Nia +] Inactivation of niaD Selection for impala excisions through by impala insertion restoration of nlaD expression

Determination of impala reinsertionsBy Southern blot analysis: 0 up to 70%

Screening for altered phenotype on plants: O 10% of path- mutants recovered

Figure 4. Experimental design allowing the selection of impala transposition and the subsequent mutational events. The grey box represents the ~fiaD gene interrupted (inactivated) or uninterrupted (expressed) by the insertion of an impala copy (black triangle); the hatched box corresponds to agene of interest, ['or example a gone (path) involved in the pathogenic process. Cloning of flanking regions of the reinserted impala element and their use as a probe to clone the target gone is being performed.

At the moment different class II transposons have been identified and the properties of their transposition appear pertinent when designing a transposon tagging method (exemplified by impala, figure 2B). The fact that excision of the impala copy inserted into the niaO gone is associated in the majority of the cases with its reinsertion into a new genomic position allows many independent insertions to be isolated easily (figure 4). Upon impala excisions selected through the restoration of niaD expression, a high proportion of mutants impaired in the pathogenic process have been already re- covered, indicating that this system offers the possibility of screening novel mutations. Demonstration that these mutants are transposon-induced is being carried out (un- published data). The recent identification of autonomous copies of For1 (Migheli etal. 1996) represents the first step in the development of a transposon tagging system in F. oxysporum. Although these results are prornising, much remains to be done before the application of transposon tagging to other fungal species.

8. Concluding remarks

Many types of TEs have been found in a wide range of filamentous fungi, indicating that they are ancient components of their genomes. Most of our knowledge comes from species that are not among the most popular, demonstrating that the diversity of organisms for research offers further opportunities for fundamental discoveries. A number of changes in gone structure and function have been shown to be mediated by transposable elements. These changes have the potential to influence many aspects of fungal genome evolution and might provide the flexibility for populations to adapt successfully to environmental conditions. Thus elucidation of the extent to which transposons contribute to genetic variation in nature is an important area for future research. Transposable elements have great potential as tools with which to rnanipulate fungal genomes, both for increasing our understanding of biological processes and for modifying organisms for practical purposes. Our understanding of transposons has 336 M. J. Daboussi increased dramatically in the last few years, and the future holds promise for important new insights.

Acknowledgements

I am grateful to Denise Zickler for helpful comments on the manuscript and to Fiona Kaper for critical reading of the manuscript. Our laboratory's contribution to the work on F. oxysporum transposable elements was done by Thierry Langin, Jean-Michel Daviare, Francine Deschamps, Catherine Gerlinger and Aur61ie Hua-Van, and on the fosbury element of M. grisea by Marc-Henri Lebrun.

References

Arnutan M., Nyyss~Snen E., Stubbs J., Diaz-Torres M. R. and Dunn-Coleman N. 1996 Identification and cloning of a mobile transposon from Aspergillus niger var. awamori. Curt. Goner. 29:468-473 Anaya N. and Roncero M, I. G. 1995 skippy, a retrotransposon from the fungal plant pathogen Fusarium oxysporum. Mol. Geu. Genet. 249:637-647 Avalos J., Mehta B., Obraztsova I., Prados N., Ruiz-Albert J., Holzmann K., Corrochano L. M. and Cerdi-Olmedo E. 1996 Recent advances in the molecular biology of Phycomyces. Abstract, 3rd European Conference on Fnngal , Mfinster, Germay, p. 126 Balcells L., Swinburne J. and Coapland G. 1991 Transposons as tools for the isolation of plant genes. TreMs Biotechnol. 9:31-37 Berg D. E. and Howe M. M. 1989, eds. Mobile DNA (Washington, DC: American Society for Microbiology) Boeke J. D. 1989 Transposable elements in Saccharomyces eerevisiae. In Mobile DNA (eds,) D. E. Berg and M. M. Howe (Washington, DC: American Society for Microbiology) pp. 335-374 Boeke J. D., Garfinkel D. J., Styles C. A. and Fink G. R. 1985 Ty elements transpose through an RNA intermediate. Ceil 40:491-500 Bureau T. E. and Wessler S. R. 1994 Mobile inverted-repeat elements of the Tourist family are associated with the genes of many cereal grasses. Proc. Natl. Acad. Sci. USA 91:1411-1415 Calvi B. R., Hong T. J., Findley S. D. and Gelbart W. M. 1991 Evidence for a common evolutionary origin af transposons in Drosophila and plants: hobo, Activator, and Tam3. Cell 465-471 Cambareri E. B., Foss H. M., Rountree M. R., Selker E. U. and Kinsey J. A. 1996 Epigenetic control of a transposon-inactivated gone in Neurospora is dependent on DNA methylation. Proc. Natl. Acad. Sci. USA 143:137-146 Colot V., Goyon C., Faugeron G. andRossignol J. L. 1995 Methylation of repeated DNA sequences and genome stability in Ascobolus immersus. Can. J. Bot. $221-$225 Cove D. J. 1976 Chlorate toxicity in AspergiIIus niduIans. Heredity 36:191-203 Daboussi M. J. and Langin T. 1994 Transposable elements in the fungal plant pathogen Fusarium oxysporum. Genetica 93:49-59 Daboussi M. J., Langin T. and Brygoo Y. I992 For1, a new family of fnngaI transposable elements. Mol. Geu. Genet. 232:12-16 Dash S. and Peterson P. A. 1994 Frequent loss of the En transposable element after excision and its relation to chromosome replication in rnaize (Zea mays L.). Genetics 136:653-671 Davi~re J. M., Langin T., Geflinger C. and Daboussi M. J. 1996 Chromosomal rearrangements and dispersed repetitive sequences in Fusar~um oxysporum. Abstract, 3rd European Conference on Fungal Genetics, Miinster, Germany, p. 141 Decaris B., Francou F., Lefort C. and Rizet G. 1978 Unstable ascospore color mutants ofAscobolus immersus. Mol. Gen. Goner. 162:69-81 Deschamps F., Langin T., Maurer P., Gerlinger C., Felenbok B. and Daboussi M. J. Functional analysis of the Fusarium FotI transposable element: characterization of a specific transcript and effects on target gene expression (submitted to Molecular Microbiology) Diolez A., Marchez F., Fortini D. and Brygoo Y. 1995 Bory, a long-terminal-repeat retroelement in the phytopathogenic fungus Botrytis einerea. A ppl. Environ. Mierobiol. 61: 103-108 Fungal transposable elements 337

Dobinson K. F. and Hamer J. E. 1993 The ebb and flow of a fungal genome. Trends MfcrobioI. 1: 348-352 Dobinson K. F., Harris R. E. and Hamer J. E. 1993 Grasshopper, a (LTR) retroelement in the phytopathogenic fungus Magnaporthe grisea. Mol. Plant-Microbe Interact. 6:114-126 Engels W. R., Johnson-Schlitz D. M., Eggleston W. B. and Sved J. 1990 High-frequency P element loss in Drosophila is homolog dependent. Celt 62:515-525 Farman M. L., Tosa Y., Nitta N. and Leong S. A. 199 6a Muggy, a retrotransposon in the genome of the rice blast fungus Magnaporthe grisea. Mol. Gen. Genet. 251:665-674 Furman M. L., Taura S. and Leong S. A. 1996b The Magnaporthe grisea DNA fingerprinting probe, MGR586, contains the 3' end of an inverted repeat transposon. Mol. Gen. Goner. 251:675-681 Fedoroff N. V. 1989 Maize transposable elements. In Mobile DNA (eds.) D. E. Berg and M. M. Howe (Washington, DC: American Society for Microbiology) pp. 375-411 Finnegan D. J. 1989 Eukaryotic transposable elements and genorne evolution. Trends Genet. 5: 103-107 Flavell A. J. 1992 Tyl-copia group retrotransposons and the evolution of retroelements in the eukaryotes. Genetica 86:203-214 Gaskell J., Wymelenberg A. V. and Cullen D. 1995 Structure, inheritance, and transcriptional effects of Peel, an insertional element within Phanerochaete chrysosporium lignin pcroxidase gone tip1. Prec. Nati. Acad. Sci. USA 92:7465-7469 Glayzer D. C., Roberts I. N., Archer D. B. and Oliver R. P. 1995 The isolation of Anrl, a transposable element from AsperffilIus niger. Mol. Gen. Goner. 249:432-438 Goyon C. and Faugeron G. 1989 Targeted transformation of Aseobolus immersus and de nero methylation of the resulting duplicated sequences. Mol. Ceil. Biol. 9:2818-2827 Goyon C., Rossignol J. L. and Faugeron G. 1996 Native DNA repeats and methylation in Aseobolus. Nucl. Acids Res. 24:3348--3356 Hamer J. E., Farall L., Orbach M. J., Valent B. and Chumley F. G. 1989 Host species-specific conservation of a family of repeated DNA sequences in the genome of a fnngal plant pathogen. Proc. Natl. Aead. Sci. USA 86:9981-9985 He C., Nourse J. P., Kelemu S., Irwin J. A. G. and Manners J. M. 1996 CgTI: a non-LTR retretransposon with restricted distribution in the fungal phytopathogen Colletotrichum gloeosporioides. MoI. Gen. Genet. 252:320-331 Iwaguchi S., Homma M., Chibana H. and Tanaka K. 1992 Isolation and characterization of a repeated sequence (RSP1) of Candida aIbicans. J. Gen. Microbiol. 138:1893-1900 Julien J., Poirier-Hamon S. and Brygoo Y. 1992 Foretl, a reverse transcriptase-like sequence in the filamentous fungus Fusarium oxysporum. Nucl. Acids Res. 20:3933-3937 Kachroo P., Leong S.A. and Chattoo B. B. 1994 Pot2, an inverted repeat transposon from the rice blast fnngus Magnaporrhe grisea. MoI. Gen. Goner. 245:339-348 Kachroo P., Leong S. A. and Chattoo B. B. 1995 Mg- SINE: a short interspersed nuclear element from the rice blast fungus, M agnaporthe grisea. Prec. Natl. Acad. Sci. USA 92:11125 11129 Kempken F. and Kiick U. 1996 Restless, an active Ac-like transposon from the fungus Tolypocladium inflatum: structure, expression, and alternative splicing. Mol. Cell. Biol. 16 (in press) Kim H. G., Meinhart L. W., Benny U. and Kistler H. C. 1995 Nrsl, a repetitive element linked to pisatin demethylase genes on a dispensable chromosome of Nectria haematoeocca. Mol. Plato-Microbe Interact. 4:524-531 Kinsey J. A. 1990 Restricted distribution of the Tad transposon in strains of Neurospora. Curr. Goner. 15: 271-275 Kinsey J. A. 1993 Transnuclear transposition of the Tad element of Neurospora. Prec. Nc~tl. Acad. Sci. USA 90:9384-9387 Kinsey J, A. and Helber J. 1989 Isolation era transposable element from Neurospora crassa. Prec. Natl. Acad. Sci. USA 86:1929-1933 Kinsey J. A., Garrett-Engcle P. W., Cambareri E. B. and Selker E. U. 1994 The Neurospora transposon Tad is sensitive to repeat-induced point mutation (RIP). Prec. Natl. Acad. Sci. USA 138:657-664 Kistler H. C. and Miao V. P. W. 1992 New modes of genetic change in filamentous fungi. Annu. Rev. Phytopathol. 30:131-152 Kistler H. C., Momol E. A. and Benny U. 1991 Repetitive genomic seqttences for determining relatedness among strains of Fusarium oxysporum. Phytopathology 81 : 331- 336 Kistler H. C., Benny U., Boehm E. W. A. and Katan T. 1996 Genetic duplication in F~sarium oxysporum. Curt. Genet. 28:173-176 338 M. J. Dabouxsi

Kricker M. C., Drake J. W. and Radman M. i992 Duplication-targeted DNA mcthylation and mutagenesis in the evolution of eukaryotie chromosomes. Proc. Natl. A cad. Sci. USA 89:1075-1079 Langin T., Capy P. and Daboussi M. J. 1995 The transposable element impala, a fungal member of the TcI-mariner superfamily. Mol. Gen. Genes. 246:19-28 Levis C., Fortini D. and Brygoo Y. 1996 Flipper, a bacterial-like transposable element in Botrytis cinel'ea. MoL Gen. Genet. (in press) Lohe A. R., Moriyama E. N., Lidholm D. and Hartl D. L. 1995 Horizontal transmission, vertical inactivation and stochastic loss of mariner-like transposable elements. Mol. Biol. Evol. 12:62-72 McDonald J. F. 1993 Evolution and consequences of transposable elements. Curr. Opin. denet. Dev. 3: 855-864 McHale M. T., Roberts I. N., Talbot N. and Oliver R. P. 1989 Expression of reverse transcriptase in Fulvia fi~lva. Mol. Plant-Microbe Interact. 2:165-168 McHale M. T., Roberts I. N., Noble S. M., Beaumont C., Whitehead M. P., Seth D. and Oliver R. P. I992 CIT-I : an LTR-retrotransposon in Cladosporiumfulvum, a fungal pathogen of tomato. Mol. Gen. Genet. 233:337-347 Magee P. T. 1993 Variations in chromosome size and organization in Candida albicans and Candida stellatoidea. Trends Microbiol. 1 : 338-342 MargoIin B. 1995 Punt, a RIPED For-like transposon of New'ospora crassa. Oral communication, Workshop ~Fungal transposable elements', Asilomar, USA, 1994 Migheli Q., Daboussi M. J., Gerlinger C., Langin T. and Laug6 R. 1996 Identification of autonomous copies of the FusaHum oxysporum For1 transposable element. Abstract, 3rd European Conference on Fungal Genetics, Miinster, Germany, p. 131 Montgomery E. A., Huang S. M., Langley C, H. and Judd B. H. 1991 Chromosome rearrangement by ectopic recombination in Drosophila melanogaster: genome structure and evolution. Genetics 129:1085-11398 Mouyna I., Renard J. L. and Brygoo Y. 1996 DNA polymorphism among FusaHum oxysporum f. sp. elaeidis populations from oil palm, using a repeated and dispersed sequence "palm". Curt. Gelwt. 30:174-~80 Neuv6glise C., Sarfati J., Latg6 J. P. and Paris S. 1996a Afull. a retrotransposon-Iikc clement from Aspergillus finnigatus. N~,rcl. Acids Res. 34:1428-1434 Neuv6glise C., Sarfati J, Debeaupuis J. P., Latg6 J. P. and Paris S. 1996b Molecular epidemiology of Aspergillusfumigatus with AfitrI, a retrotransposon-like element. Abstract, 3rd European Conference on Fungal Genetics, Miinster, Germany, p. 132 Nyyss~Snen E., Amutan M. and Dunn-Coleman N. 1996 Tan and Varlet-transposable elements found from A. niger var. awclmori. Abstract, 3rd European Conference on Fungal Genetics, Mtinster, Germany, p. 133 Oliver R. P. 1992 Transposons in filamentous fnngi. In Molecular biology oil fi lamentousfun~li (eds.) U. Stahl and P. Tudzynski (Proceedings of the EMBO Workshop, Berlin 1991) (Berlin: VCH) Plasterk R. H. 1991 The origin of footprints of the Tel transposon of Caenorhabditis elegans. EMBO d. 10: 1919-1925 Plasterk R. H. and Grocnen J. T. 1992 Targeted alterations of the Caenorhabditis elegans genome by transgene instructed DNA double strand break repair following Tel excision. EMBO J. 11 : 287-90 Radice A. D., Bugaj B., Fitch D. H. A. and Emmons S. 1994 Widespread occurrence of the Tel transposon family: Tel-like transposons from teleost fish. MoI. Gen. Genet. 244:606-612 Rasmussen M., Rossen L. and Giese H. 1993 SINE-like properties of a highly repetitive element in the genome of the obligate parasitic fungus Erysiphe graminis f. sp. hordei. MoL Gen. Genes. 239:298-303 Rhounim L., Rossignol J. L. and Faugeron G. 1992 Epimutation of repeated genes in Ascobolus immersus. EMBO J. 11:4451-4457 Robertson H. M. 1995 The Tel-mariner supcrfamily of transposons in . J. Insect Physiol. 41: 99-105 RossignoI J. L. and Faugeron G. 1994 Gene inactivation triggered by recognition between DNA repeats. Experientia 50:307-317 Rossignol J. L. and Faugeron G. 1995 MIP: an epigenetic gene silencing in Ascobolus immersus. In Gelre s~leneing in higher plants and related phenomem~ in other eukaryotes (cd.) P. Meyer (Berlin: Springer) 197: 179-191 Rouyer F., Simmler M. C., Page D. and Weissenbach J. 1987 A sex rearrangement in a human XX male caused by Alu-Alu recombination. Cell 51:417-425 Selker E. U. and Stevens J. N. 1985 DNA methylation at asymmetric sites is associated with numerous transition nautations. Pvoc. Natl. Aead. Sci. USA 82:8114- 8118 Selker E. U., Fritz D. Y. and Singer M. J. 1993 Dense nonsymmetrical DNA methylation resulting from repeat-induced point mutation (RIP) in Neurospora. Science 262:1724-1728 Fungal transposable elements 339

Sheen F., Lim J, K. and Simmons M. J. 1993 GenEtic instability in Drosophil~ melano,qastermediated by hobo transposable elements. Generics 133 : 315-334 Shull V. and Hamer J. E. 1996 Genetic differentiation in the rice blast fungus revealed by the distribution of the fosbury retrotransposon. FungaI Genet. Biol. 20:59-69 Springer M. S. and Britten R, J. 1993 Phylogenetic relationships of reversc transcriptase and RNase H sequences and aspects of genome structure in the gypsy group of retrotransposons. MoL Biol. Evol. 10: 1370-1379 Skinner D. Z., Budde A, D. and Leong S. A. 1991 Molecular karyotype analysis of fungi. In. More oene manipulations infuruji (eds.) J. W. BennEtt and L. L. Lasure (London Academic Press) pp. 86-103 Sone T, Suto M. and Tomita F. 1993 Host species-specific repetitive DNA sequence in the genome of Magmtporthe grisea, the rice blast fungus. Biosci. Bio~ech. Biochem. 57:1228--1230 Talbot N. J., Salch Y. P., Ma M. and Hamer J. E, 1993 Karyotype variation within elonal lineages of the rice blast fungus, Magnaporthe ~jrisea. AppI. El~viron. Microbiol, 59:585-593 TzengT., Lyngholm L. K., Ford C. F. and Bronson C. R. 1992 A restriction fragment polymorphism map and eleetrophoretic karyotype of the fungal maize pathogen Coehliobolus heterosrrophus. Genetics 130:81- 96 Valent B, and Chumley F. G. 1994 AvirulEncE genes and mechanisms of genetic instability in the rice blast fungus. In Rice blast disease (eds.) R. S. Zeigler~ S. A. Leong and P. S, Teng (Cambridge: CAB international) pp. 111-134 Walbot V. 1992 Strategies for mutagenesis and gene cloning using transposon tagging and T-DNA mutagenesis. Amau. Rev. Plant Physiol. Plant Mol. Biol. 43:49-82 Wei Y. D., Collinge D. B,, Smedegaard-Petersen V. and Thordal-Christensen H, 1996 Characterization of the transcript of a new class of retroposon-type repetitive element cloned from the powdery mildew fungus, Erysiphe 9raminis. Mol. Gem Genet. 250:477-482 Xiong Y, and Eickbush T. H. 1990 Origin and evolution of retroelements based upon their reverse transcriptase sequences. EMBO J. 9:3353-3362 Yeadon P. J. and Catcheside D, E. A. 1995 Guest: a 98 bp inverted repeat transposable element in New'ospora erassa. Mol. Gels. Genet. 247:105-109