HET-E and HET-D Belong to a New Subfamily of WD40 Proteins Involved in Vegetative Incompatibility Specificity in the Fungus Podospora anserina

Corresponding author: Béatrice Turcq, Institut de Biochimie et de Génétique Cellulaires, CNRS UMR 5095, 1 rue Camille Saint-Saëns, 33077 Bordeaux Cedex, France., beatrice.turcq{at}ibgc.u-bordeaux2.fr (E-mail)

Communicating editor: R. H. DAVIS

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Vegetative incompatibility, which is very common in filamentous fungi, prevents a viable heterokaryotic cell from being formed by the fusion of filaments from two different wild-type strains. Such incompatibility is always the consequence of at least one genetic difference in specific genes (het genes). In Podospora anserina, alleles of the het-e and het-d loci control heterokaryon viability through genetic interactions with alleles of the unlinked het-c locus. The het-d2^Y gene was isolated and shown to have strong similarity with the previously described het-e1^A gene. Like the HET-E protein, the HET-D putative protein displayed a GTP-binding domain and seemed to require a minimal number of 11 WD40 repeats to be active in incompatibility. Apart from incompatibility specificity, no other function could be identified by disrupting the het-d gene. Sequence comparison of different het-e alleles suggested that het-e specificity is determined by the sequence of the WD40 repeat domain. In particular, the amino acids present on the upper face of the predicted ß-propeller structure defined by this domain may confer the incompatible interaction specificity.

IN filamentous fungi, anastomoses between hyphal filaments occur frequently and produce heterokaryotic cells in which two different genomes coexist. Heterokaryotic cell viability is controlled by specific het loci involved in vegetative or heterokaryon incompatibility (ESSER and BLAICH 1973 ; GLASS and KULDAU 1992 ; BEGUERET et al. 1994 ; LOUBRADOU and TURCQ 2000 ; SAUPE 2000 ). In Podospora anserina, a single genetic difference between the two genomes at one het locus is sufficient to impair the viability of the resulting heterokaryotic cell (RIZET 1952 ; BERNET 1965 ). A lytic process rapidly destroys the heterokaryotic cell (BEISSON-SCHECROUN 1962 ). Accumulation of dead cells between two incompatible strains leads to formation of an abnormal contact zone called "barrage" (RIZET 1952 ).

The number of het loci is generally high: 11 in Neurospora crassa (PERKINS et al. 1982 ), 9 in P. anserina (BERNET 1967 ), 8 in Aspergillus nidulans (GRINDLE 1963 ) and 6 or 7 in Cryphonectria parasitica (ANAGNOSTAKIS 1977 ; CORTESI and MILGROOM 1998 ). In P. anserina, two different types of incompatibility systems have been described. If incompatibility results from the coexpression of two antagonistic alleles from a single het locus, this is called an allelic incompatibility system; if, conversely, incompatibility results from the coexpression of two antagonistic alleles from two separate het loci, this is known as a nonallelic incompatibility system (BERNET 1965 ).

Characterization of het genes is the first step toward understanding the phenomenon of vegetative incompatibility. Several genes involved in vegetative incompatibility have been characterized in fungi. In N. crassa, the mating-type alleles (mat a-1 and mat A-1), which also control vegetative incompatibility, have been studied extensively (GLASS et al. 1988 , GLASS et al. 1990 ; STABEN and YANOFSKY 1990 ): the het-C encoded protein displays a glycine-rich domain (SAUPE et al. 1996 ), the het-6 gene product presents sequence similarity with P. anserina HET-E (SMITH et al. 2000 ), and the un-24 gene encodes the large subunit of type I ribonucleotide reductase (SMITH et al. 2000 ). In P. anserina, the allelic incompatibility het-s gene encodes a prion-like protein (TURCQ et al. 1990 ; COUSTOU et al. 1997 ). The het-c gene involved in both nonallelic systems het-c/het-e and het-c/het-d encodes a protein similar to the glycolipid transfer protein, and inactivation of this gene leads to abnormal ascospore formation (SAUPE et al. 1994 ). The het-e gene encodes a protein that displays a GTP-binding site and a WD40 repeat domain. Previous studies have demonstrated that the P. anserina het-e allele reactivity in incompatibility is dependent on both GTP binding by the P-loop domain and a minimal number of 10 WD40 repeats (SAUPE et al. 1995 ; ESPAGNE et al. 1997 ). Investigation of different het-e alleles, which were reactive in incompatibility, showed that the number of WD40 repeats does not correlate with allele specificity.

The WD40 repeat, first identified in the ß-subunit of a G-protein, is a degenerate sequence repeat of 40–43 amino acids in length (FONG et al. 1986 ). Heterotrimeric G-protein structure has revealed that the seven WD40 repeats of the Gß subunit are organized in a circular structure of seven ß-sheets, forming the blades of a ß-propeller structure around a central pore (LAMBRIGHT et al. 1996 ; SONDEK et al. 1996 ). Upper, lower, and circumferential faces are potential binding surfaces (SMITH et al. 1999 ). Each ß-sheet is composed of four antiparallel strands. The same structure has been described for the C-terminal domain of TUP1, which also displays seven WD40 repeats (SPRAGUE et al. 2000 ). In addition, the biochemical properties of four other WD40 repeat proteins, with repeats ranging from five to seven, suggest that all WD40 repeats fold into a similar tertiary structure (GARCIA-HIGUERA et al. 1996 ). The ß-propeller structure was also described in non-WD40-repeat proteins (SMITH et al. 1999 ). The highest-known number of blades in a propeller structure is eight and the highest-known number of WD40 repeats is 16. For a protein containing more than eight WD40 repeats, it remains unclear whether the protein forms two small propellers or a large one (SMITH et al. 1999 ). The potential role of HET-E WD40 repeats in a biological function and the mechanism of regulating vegetative incompatibility both remain unclear. Inactivation of the het-e gene does not lead to any particular phenotype other than the incompatibility phenotype.

Southern blot analyses on genomic DNA have suggested the existence of a sequence similar to het-e (ESPAGNE et al. 1997 ). It was possible that the het-d gene, the other het-c antagonist, could be a functional homolog of het-e. To increase our understanding of het genes in vegetative incompatibility, we undertook het-d characterization. As for HET-E, domains characteristic of both - and ß-subunits of a heterotrimeric G-protein (ESPAGNE et al. 1997 ), were also found in HET-D putative protein. All the het-d alleles reactive in incompatibility had a minimal number of 11 full-length WD40 repeats.

To identify the domains potentially involved in the vegetative incompatibility specificity, we compared the amino acid sequence of different alleles of het-e rather than of het-d alleles, because first, four different allele specificities have been described for het-e alleles and only three for het-d alleles, and second, because only mutant alleles reactive or nonreactive in incompatibility are available for het-e. The results showed a crucial role of the WD40 repeat sequence on the vegetative incompatibility specificity.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains and growth conditions:
P. anserina is a heterothallic ascomycete. Its life cycle and general methods for genetic analyses have been described (RIZET and ENGELMAN 1949 ; ESSER 1974 ). Vegetative incompatibility can be determined by strain confrontation on solid corn-meal agar. Incompatibility results in the formation of a barrage, i.e., a dense unpigmented line in the contact region between two mycelia of antagonistic strains (RIZET 1952 ).

The two nonallelic incompatibility systems het-c/het-e and het-c/het-d are characterized by multiple alleles present at the het-c, het-e, and het-d loci in wild-type isolates. All strains used in this study were isogenic to wild-type isolate s (RIZET 1952 ) but with different alleles at het-c, het-e, and het-d loci. Among 17 wild-type strains studied by BERNET 1967 , four classes of het-c alleles, four classes of het-e alleles, and three classes of het-d alleles have been described. An allele is designated by the class number and the name of the wild-type isolate. For example, the allele het-d2 of the Y strain is denoted het-d2^Y. If a strain contains a neutral allele at any of the three loci, this locus is not mentioned in the strain designation.

DNA analysis:
Standard techniques were used for DNA cloning, restriction enzyme digestion, Southern blot analysis, and DNA sequencing (SAMBROOK et al. 1989 ). Genomic DNA was prepared with the rapid petri-dish-grown mycelia method (LECELLIER and SILAR 1994 ). For the library construction, genomic DNA was prepared as described (JAVERZAT et al. 1993 ). For het-d sequencing, exonuclease III sequential deletions were performed using an ExoIII/mung bean nuclease deletion kit (Stratagene, La Jolla, CA) as recommended by the supplier. To determine the number of WD40 repeats in different het-d alleles, genomic DNA was digested by BamHI and SphI and probed with the fragment SphI-XhoI [nucleotide (nt) 2578–2925] located upstream from the repeat region.

DNA library construction:
The DNA library was constructed into the pMOcosX cosmid (ORBACH 1994 ) using the XL kit (Stratagene). Partially SalI-restricted DNA fragments from the P. anserina Y strain were size selected by pulsed field gel electrophoresis performed in 0.5x TBE at 9° using a contour-clamped homogeneous electric field apparatus (LKB, Piscataway, NJ) on 1% low-melting-point agarose gel (SeaPlaque agarose, FMC, Rockland, ME). Samples migrated for 16 hr in an electric field of 180 V with a 5-sec switching time. Gels were then stained for 1 hr in a 0.2-µg/ml ethidium bromide solution and washed for 30 min in water. The DNA was visualized by UV fluorescence. DNA oligomers (between 48.5 and 485 kbp) were used as size markers. DNA fragments between 25 and 50 kbp were isolated from the gel and agarose eliminated by GELase treatment (GELase, Epicentre, Madison, WI). The pMOcosX vector was restricted by XbaI to generate vector arms with cos extremities, treated with alkaline phosphatase, and then restricted with XhoI. Ligation was performed for 16 hr at 26° in 10 µl of ligation buffer using 2 µg of purified genomic fragments, 4 µg of vector arms, and 5 units of T4 DNA ligase (Life Technologies). The ligation mix was then used for in vitro packaging (Gigapack II Gold kit, Stratagene) and transfection into Escherichia coli DH5, as described by the manufacturer.

Protoplasts were prepared and transformed as previously described (BERGES and BARREAU 1989 ). The pMOcosX vector contained the bacterial hygromycin resistance gene hph as a selectable marker (ORBACH 1994 ), and transformants were screened on hygromycin B at 100 µg/ml.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Cloning of the incompatible het-d2^Y gene by phenotypic expression:
To clone the het-d2^Y gene, a genomic cosmid library from the Y wild-type strain was divided into 16 pools of 192 cosmids. DNA prepared from each pool was used to transform a recipient strain containing a het-c allele null in incompatibility. Hygromycin-resistant transformants were tested for incompatibility with the strain containing the het-c4^M allele antagonistic to the het-d2^Y allele by screening for barrage formation with this tester strain. Three strains, arising from transformation with the same DNA pool, showed a barrage reaction: they had gained the het-d2^Y phenotype. The cosmid conferring this phenotype was isolated from the cosmid pool after three rounds of sib-selection (AKINS and LAMBOWITZ 1985 ). The functional het-d2^Y gene was then subcloned on a 5.2-kbp SalI-XbaI fragment, which was then sequenced.

The het-d2^Y gene encodes a protein showing high similarity with HET-E1^A protein:
Sequence analysis revealed that the 5.2-kbp fragment contained an open reading frame (ORF) of 4187 bp interrupted by a putative intron of 56 bp, according to consensus sequences for filamentous fungi splicing sites (BALLANCE 1991 ; BRUCHEZ et al. 1993 ). This putative intron splits the open reading frame into two exons, each of them encoding a distinct, characteristic domain. Such a structure has already been reported for the het-e1^A gene (SAUPE et al. 1995 ). Although these two genes show 66.5% nucleotide identity, some minor differences were observed. The intron is located at a position corresponding to the 751st amino acid for the predicted het-d2^Y polypeptide and to the 761st amino acid for the protein encoded by het-e1^A gene. The het-d^Y intron is 56 bp, whereas the het-e1^A intron is only 49 bp; in addition, the splicing site sequences are different for the two introns (data not shown).

The het-d2^Y gene encodes a 1376-amino-acid putative polypeptide with a predicted molecular mass of 152 kD, displaying high similarity (53% identity, 71% similarity) with the HET-E1^A protein (SAUPE et al. 1995 ; ESPAGNE et al. 1997 ). Sequence analyses show that a GTP-binding site (SARASTE et al. 1990 ; BOURNE et al. 1991 ) and a WD40 domain (VAN DER VOORN and PLOEGH 1992 ) are found in N-terminal and C-terminal regions, respectively (Fig 1). This structure is identical to that of the HET-E1^A protein (SAUPE et al. 1995 ). Positions of the GTP-binding-site P-loop and G2, G-3, and G-4 sequences are highly conserved between HET-D2^Y and HET-E1^A, but only the P-loop sequences are identical (Fig 1B). The HET-D2^Y WD40 domain is very similar to the HET-E1^A WD40 domain, although HET-D2^Y and HET-E1^A do not share any identical repeats. Equally, the HET-D2^Y domain is longer, as it contains 12 WD40 repeats (11 full-length repeats and the first 30 amino acids of a 12 repeat) instead of 10 for HET-E1^A (Fig 2A). Out of the 42 amino acids of a repeat, 31 are conserved within the 11 full-length repeats. Moreover, some repeats (repeats 3 and 7 and repeats 5 and 10) are identical (Fig 2A). The repeats differ from each other by, at the most, 8 amino acids out of 42. Where any two random repeats are compared, sequence identity is very strong (>80%). The same situation was previously reported for HET-E1^A protein (SAUPE et al. 1995 ). These findings contrast with the situation for other known WD40 proteins, where identity between any two repeats never exceeds 20–30% (SAUPE et al. 1995 ). The HET-D2^Y WD40 repeat consensus sequence and its predicted secondary structure are reported in Fig 2B. In the first 32 N-terminal amino acids of the repeats, 4 of the 11 polymorphic positions located there are highly polymorphic, with 3–5 different amino acids being observed for a given position (Fig 2A and Fig B). One highly polymorphic position (amino acid 11) is in the loop between the d ß-strand and the a ß-strand; another (amino acid 13) is in first position on the a ß-strand; the remaining two positions (amino acids 29 and 31) are in the turn between the b ß-strand and the c ß-strand.

View larger version (47K): In this window In a new window Download PPT slide   Figure 1. Amino acid sequence of HET-D2Y. (A) Schematic representation of HET-D2Y polypeptide. (B) Alignment of HET-D2Y (D, top line: accession no. AF323585) and HET-E1A (E, bottom line: accession no. L28125) polypeptides. The HET-E1A sequence corresponds to the revised sequence. Periods indicate identity and dashes indicate gaps. The large arrow indicates position of the intron for each sequence. The three sequences described by SMITH et al. (2000) defining the three blocks contained in the conserved domain are in white uppercase letters inside dark boxes. Regions of the GTP-binding domain conserved among G-proteins (P-loop, G2, G3, and G4) are underlined (BOURNE et al. 1991 ). Two G2 sequences are possible for HET-E1A polypeptide (SAUPE et al. 1995 ). The shaded boxes in the C terminus indicate the WD40 repeats.

View larger version (57K): In this window In a new window Download PPT slide   Figure 2. Sequence of the WD40 repeats of the HET-D2Y polypeptide. (A) Alignment of the WD40 repeats of the HET-D2Y protein. The positions that are conserved are lightly shaded. Uppercase letters in white indicate the positions that differ from the consensus sequence. They are indicated by a gray background (polymorphic position) or by a black background (highly polymorphic). Spaces have been introduced to separate the different structural blocks. (B) Consensus sequence for the repeats is given with the predicted secondary structure of a repeat schematized above (___, a ß-strand; $$$, a loop; ???, a turn). The letters d, a, b, and c indicate the ß-strands d, a, b, and c, respectively. The numbers indicate the amino acid positions in a repeat. Variable positions in uppercase letters are indicated by a gray background (polymorphic) or by a black background (highly polymorphic). Every possible amino acid found at these positions is reported below, arranged in order of frequency.

A BLAST search in databanks with the N-terminal part of the HET-D polypeptide identified only two polypeptides homologous to HET-D2^Y. The first polypeptide is HET-E1^A; the second is an open reading frame translation product, Q9P654, issued from the German Neurospora genome-sequencing project (accession no. AL356324). This Q9P654 predicted polypeptide is annotated as related to ß-transducin-like protein because of its similarity to HET-E1^A. In fact, only the 242 N-terminal amino acids of this 605-amino-acid-long polypeptide show similarities to HET-D2^Y (53 and 70% of identity and similarity, respectively) and to HET-E1^A N-terminal. These identities are restricted to the blocks corresponding to conserved regions previously described in P. anserina HET-E, in N. crassa HET-6, and in N. crassa TOL proteins involved in vegetative incompatibility (SMITH et al. 2000 ). This conserved region of 150 amino acids displays three blocks of 17, 36, and 10 amino acids, respectively. These blocks, believed to represent an incompatibility domain, are highly conserved between the three HET-D2^Y, HET-E1^A, and Q9P654 polypeptides (>50, 33, and 80% similarity, respectively; Fig 3). Outside this region, both HET-D and HET-E polypeptides are completely unrelated to N. crassa incompatibility polypeptides.

View larger version (18K): In this window In a new window Download PPT slide   Figure 3. Alignment of conserved blocks of HET-D2Y (accession no. AF323585) and HET-E1A (accession no. L28125) predicted proteins from P. anserina and Q9P564 ORF (accession no. AL356324) resembling HET-E1A, HET-6OR (accession no. AF206700), HET-6PA (accession no. AF208542), and TOL (accession no. AF085183) predicted proteins from N. crassa. HET-D2Y, HET-E1A, and Q9P564 sequences were aligned using Clustal with default setting and conserved block aligned to the previously described region (SMITH et al. 2000 ). Amino acids present in one of the two P. anserina polypeptides and conserved in the other predicted proteins are indicated by shading. A star indicates that the amino acids at this position are identical in all sequences, a colon represents a conservative substitution, and a period a semiconservative substitution. The consensus sequence is called "cons." Uppercase letters show that the amino acid at this position is identical in all sequences and lowercase letters show that the amino acid at this position is conserved in at least half of the six sequences.

The number of WD40 repeats in different wild-type het-d alleles:
Our previous results showed that HET-E reactivity in incompatibility depends on two functional elements: a WD40 domain with at least 10 repeats and a functional GTP-binding domain (SAUPE et al. 1995 ; ESPAGNE et al. 1997 ). Different wild-type het-d alleles were then examined to determine the number of WD40 repeats by estimating the size of the region encompassing the repeats (Table 1). Two classes of wild-type het-d alleles were analyzed: those reactive in incompatibility (het-d1^A, het-d2^F, and het-d2^Y) and those nonreactive in incompatibility (het-d3 alleles from B, D, E, H, M, s, U, V, W, X, and Z strains). Southern experiments revealed the presence of a 2100-bp fragment for the three active het-d alleles. All three polypeptides encoded by het-d1^A, het-d2^F, and het-d2^Y active alleles display 12 WD40 repeats (11 full-length and 1 truncated repeat). For the het-d3 neutral alleles, the size of the hybridizing fragments ranged from 1340 bp (het-d3^W allele) to 2225 bp (het-d3^s allele), corresponding to an estimated number of 6 and 13 WD40 repeats, respectively (Table 1). All het-d alleles containing <11 WD40 repeats are neutral alleles. As observed for het-e alleles (ESPAGNE et al. 1997 ), a minimum number of repeats seems to be necessary for an incompatibility reaction; this number is 10 for HET-E1^A and may be 11 for HET-D2^Y. However, the number of repeats does not seem to be the only determinant of activity in incompatibility, since five het-d3 alleles also contain 11 repeats and are neutral. Moreover, the three alleles reactive in incompatibility encode a polypeptide with the same predicted number of repeats but with a different incompatibility spectrum; thus the number of repeats is not specific for a given incompatible interaction.

Sequence differences and het-e allele specificity:
Allele specificity was analyzed on het-e rather than on het-d since the number of wild-type het-e alleles reactive in incompatibility is higher. Moreover, only mutant alleles still reactive in incompatibility are available for het-e. To determine which domains of the HET-E polypeptide are specific to the incompatibility spectrum, four het-e genes have been sequenced. One, het-e2^C, is reactive in incompatibility. Another, also reactive in incompatibility, het-e1^A, was resequenced and five differences with the sequence previously published by SAUPE et al. 1995 were found. Only three of these differences (Ala₈₅₁ Pro, Arg₈₉₅-Glu Gly-Gln, and Gly₁₀₁₀ Asn) modify the protein sequence in the WD40 domain. The third, het-e4^s, is null in incompatibility. The fourth, het-e2^C-4, is a het-e2^C mutant allele still reactive in incompatibility. This mutant allele, which was obtained by UV mutagenesis and previously described by ESPAGNE et al. 1997 , has lost incompatibility with the het-c3 allele but has retained incompatibility with het-c1 and het-c4 antagonistic alleles. It should be noted that positions and sequences of the intron were identical in all four alleles.

Outside the WD40 domain, seven polymorphic positions were found (Table 2). Two of these polymorphic positions are specific to the null het-e4^s allele: one leads to an insertion of three amino acids after the amino acid 1285 in the predicted polypeptide; the other is a single nucleotide change that modifies the amino acid 482 (Pro₄₈₂ Ser₄₈₂). What distinguishes the het-e4^s allele from all the sequenced het-e alleles is that it is not reactive in incompatibility. Although the inactivity is probably due to the low number of repeats (three repeats), these two positions could be important for reactivity in incompatibility. The five other polymorphic positions concern the het-e1^A allele compared to the het-e2^C and het-e4^s alleles. Two positions correspond to synonymous substitutions at amino acids 543 and 691. The three other positions are nonsynonymous substitutions that modify amino acids 693, 1309, and 1342 (Table 2). These three amino acid modifications could be responsible for the differences between the het-e1 and het-e2 phenotypes. To test this hypothesis, other different wild-type het-e alleles that are reactive in incompatibility (het-e1^H, het-e1^M, and het-e3^F) have been sequenced at these polymorphic positions (Table 2). Amino acids present at positions 693, 1309, and 1342 are identical in het-e1^M, het-e2^C, and het-e3^F alleles. Since these three alleles do not display the same incompatibility spectrum, these three polymorphic positions are not responsible for het-e allele specificity. These polymorphic positions lie in the region that shows less conservation between HET-E1^A and HET-D2^Y polypeptides. Altogether, it appears that amino acid sequence outside of the WD40 domain is not responsible for the specificity of the incompatibility spectrum.

The incompatibility specificity should be due to difference in the WD40 domain. SAUPE et al. 1995 and ESPAGNE et al. 1997 have already established that the number of WD40 repeats was not correlated with the incompatibility specificity. So, the difference in the het-e1 and het-e2 incompatibility spectrum should result from difference in the amino acid sequence of the WD40 domain. Wild-type het-e1^A and het-e2^C and the mutant het-e2^C-4 WD40 domains were sequenced (Fig 4). Analysis of the sequences confirmed our previous observations that the repeat number is 10 for these alleles (SAUPE et al. 1995 ; ESPAGNE et al. 1997 ). Only three repeats (1, 2, and 5) display the same sequence for the encoded polypeptides (Fig 4). A total of 26 polymorphic positions were found between the HET-E1^A and HET-E2^C WD40 domains (Fig 4). There are three positions in the turn following the c ß-strand (amino acids 38 and 39) and one (amino acid 1) in the d ß-strand (Fig 4). The 22 other changes are located exclusively within three specific regions of the WD40 repeats. Six polymorphic positions are found in a region corresponding to the loop between the d and a ß-strands (amino acids 10 and 11). Seven are located at the first position of the a ß-strand (amino acid 13) and one in the a ß-strand (amino acid 16). Eight occur in the turn between b and c ß-strands (amino acids 29 and 31). These polymorphic sites between the het-e1^A and het-e2^C alleles have already been reported (SAUPE et al. 1995 ) as variable positions between the different HET-E1^A WD40 repeats (Fig 4A). These positions correspond to the highly polymorphic positions also observed between the different WD40 repeats of the HET-D2^Y (Fig 2A). It can be observed that highly polymorphic regions are located at corresponding positions in the same structural element for the WD40 repeats of the HET-E1^A and HET-D2^Y polypeptides, suggesting that the assortment of amino acids present at these critical positions is important to determine the specificity (Fig 2A and Fig 4A).

View larger version (53K): In this window In a new window Download PPT slide   Figure 4. Amino acid sequence comparison of the WD40 repeat domain of the HET-E1A (A in sequence), HET-E2C (C in sequence), and the mutant HET-E2C-4 (C4 in sequence) polypeptides. (A) Consensus sequence for HET-E1A repeats based on a frequency of at least 0.4 for a conserved amino acid (aa) at each position. The number indicates the aa position in a repeat. Polymorphic positions and highly polymorphic positions are indicated by gray and black backgrounds, respectively. Above the consensus sequence, the predicted secondary structure is schematized (__, for a ß-strand; $$$, for a loop; ???, for a turn). The letters d, a, b, and c indicate the ß-strands d, a, b, and c, respectively. (B) Alignment of wild-type and mutant HET-E WD40 repeats. Amino acid positions in the sequence are indicated at the right end of each repetition. For HET-E2C and HET-E2C-4, only amino acids that differ from those of HET-E1A are given. Polymorphic positions between HET-E1A and HET-E2C are indicated by shading and differences between HET-E2C and HET-E2C-4 are indicated by a black background.

Amino acid sequence comparison between HET-E2^C-4 and HET-E2^C sequences shows five differences in the WD40 repeated domain. These changes are located within the sixth and seventh repeats (Fig 4B). Strikingly, these five mutations modify amino acids at positions corresponding to the highly polymorphic positions in HET-E2^C and HET-E1^A repeats. Amazingly, four of these mutations change the amino acid present in the HET-E2^C-4 sequence into an amino acid identical to the one present in the HET-E1^A sequence (positions 1055, 1073, 1095, and 1097). For the fifth mutation (position 1071), an amino acid, different from the one in HET-E1^A and in HET-E2^C, is present. The het-e2^C-4 gene is a het-e2^C mutant allele that has lost incompatibility with the het-c3 allele but has retained incompatibility with the het-c1 and het-c4 antagonistic alleles (ESPAGNE et al. 1997 ). Our results suggest that amino acids present at the polymorphic positions in the sixth and the seventh repeats are important for the incompatible interaction with the het-c3 allele.

Disruption of the het-d2^Y allele:
het-d2^Y gene inactivation was performed to investigate its potential function besides incompatibility. The het-d2^Y gene, cloned on a 6-kbp SmaI-XbaI fragment, was disrupted (Fig 5A). The two SalI fragments (1000 and 1547 bp) encompassing the N-terminal conserved region and the GTP-binding domain were replaced by the ura5 gene (TURCQ and BEGUERET 1987 ). The resulting plasmid, pSD, was used to transform a recipient strain carrying the het-d2^F allele, which is reactive in incompatibility, as well as the ura 5-6 mutation. To screen transformed strains that have lost the het-d2^F specificity, 2500 prototrophic transformants were confronted with a tester strain containing the antagonistic het-c4^M allele. Four transformants displaying a neutral incompatibility phenotype were obtained. Genomic DNA from these four transformants and from the recipient strain was then restricted with SalI and XhoI and subsequently submitted to Southern blotting, using the 696-bp HindIII fragment as a probe (Fig 5). In the disrupted locus, only the 768-bp SalI-XhoI fragment was expected. The DNA of the recipient strain (Fig 5B, lane D) contains both the 768-bp SalI-XhoI and the 1547-bp SalI fragments. These two fragments were also detected in the DNA of three transformants (Fig 5B, lanes 1–3). These transformants probably resulted from complex integration events: insertion by a single crossing over of the pSsD plasmid probably duplicated the het-d locus; meanwhile, the resident het-d gene was inactivated by an unknown mechanism. As expected for a perfect gene replacement, the fourth transformant has lost the 1547-bp SalI fragment and retained the 768-bp SalI-XhoI fragment (Fig 5B, lane 4). By probing with the 1000- and 1547-bp SalI fragments, we confirmed that the two SalI fragments, which were replaced by the ura5 gene in the pSD plasmid, were not detected in the genomic DNA of this transformant (Fig 5C). This disrupted transformant was called D and was retained for further studies.

View larger version (36K): In this window In a new window Download PPT slide   Figure 5. Molecular analysis of strains transformed with the pSD (het-d) disruption vector. (A) Physical map of the 6-kbp SmaI-XbaI fragment containing the het-d2Y gene (at the top) and of the pSD disruption construct (at the bottom). The shaded, solid, and open boxes indicate the DNA encoding the WD40 domain, the intron, and the ura5 gene, respectively. The arrow in the open box gives the direction of gene transcription. Sizes of the fragments represented by double arrows are given in base pairs. (B) Southern blot analysis of the genomic DNA from the recipient strain (lane D) and from four transformed strains (lanes 1–4). (C) Southern blot analysis of the genomic DNA from the recipient strain (lane D) and from the D knockout mutant strain. Genomic DNAs were digested by SalI and XhoI and probed with a 32P-labeled 696-bp HindIII fragment (B) and with 32P-labeled 1000- and 1547-bp SalI fragments (C). The sizes of the fragments are given in base pairs.

To determine whether, in addition to the loss of incompatibility, the inactivation of het-d had any other effect on fungus biology, the phenotype of the D strain was investigated under different conditions. The mutant strain phenotype was found to be similar to the wild type during its vegetative and sexual phases. Double-mutant strains containing both disrupted het-c and het-d loci were then constructed by crossing single-mutant strains. The phenotype of these strains was identical to that of single het-c mutants. They displayed abnormal ascospore production but the rate of aborted asci was not different from that described for crosses between single het-c mutant strains (SAUPE et al. 1994 ).

The same results were obtained with mutant strains containing either a single disrupted het-e locus or the two disrupted het-c and het-e loci. The lack of any detectable phenotype in strains containing either a disrupted het-e or a disrupted het-d locus may be due to the complementation of one gene by the other one, since they are homologous. To test this hypothesis, double- and triple-mutant strains containing either disrupted het-e and het-d loci or disrupted het-e, het-d, and het-c loci were then constructed. No effect was observed. This result suggests that het-e and het-d genes may act only in vegetative incompatibility.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The het-d locus is one of three incompatibility loci involved in the nonallelic incompatibility systems het-c/het-e and het-c/het-d. The het-c and het-e loci have been previously characterized (SAUPE et al. 1994 , SAUPE et al. 1995 ). In this study, we reported the characterization of the vegetative incompatibility het-d2^Y gene, which encodes a polypeptide of 1376 amino acids. As suspected (ESPAGNE et al. 1997 ), sequence comparison indicates that the HET-D2^Y polypeptide shows 71% similarity with the HET-E1^A protein (SAUPE et al. 1995 ). Like the HET-E protein, the HET-D putative protein exhibits a putative GTP-binding domain in the N-terminal part (SARASTE et al. 1990 ) and a WD40 repeat domain in the C-terminal part (VAN DER VOORN and PLOEGH 1992 ). Moreover, our results pointed out the presence of a conserved putative incompatibility domain in the N-terminal part. This domain was also described in a N. crassa incompatibility protein (SMITH et al. 2000 ). Comparison of this highly conserved domain with sequences present in the databank reveals a sequence issued from the Neurospora sequencing project on chromosome II. It would be interesting to know if this sequence matches 1 of the 11 N. crassa incompatibility loci. Mutational analyses on this domain should be performed to confirm the involvement of this sequence in the incompatibility activity. The boxes identified in this domain may define a family of proteins specialized in incompatibility (SMITH et al. 2000 ). The het-e and het-d gene products define two members of a subfamily of incompatibility polypeptides exhibiting, in addition to the incompatibility specific domain, two other domains: a GTP-binding domain and a WD40 repeat domain. The HET-E and HET-D proteins are the only known proteins showing both characteristics of the - and ß-subunits of G-proteins.

Except for vegetative incompatibility, no other function in the P. anserina life cycle was identified for the het-e and het-d genes under laboratory conditions. This result may be either because the two gene products act exclusively in incompatibility or because another gene complements the loss of function caused by the inactivation of the het-e and het-d genes.

We previously reported that the P-loop domain and the WD40 repeat number of the HET-E protein are essential for incompatible activity (SAUPE et al. 1995 ; ESPAGNE et al. 1997 ). The P-loop sequence is conserved in the HET-D polypeptide. Depending on the het-e or het-d active or inactive allele, the number of WD40 repeats ranges from 3 to 12 (ESPAGNE et al. 1997 ) and 6 to 13 in HET-E and HET-D polypeptides, respectively. Analyses of the different het-d alleles show that all those that are reactive in incompatibility contain 11 full-length WD40 repeats and a 12th truncated repeat. The number of WD40 repeats seems to be very important in determining incompatible interaction. Of the 42 amino acids, 33 are conserved within the 10 HET-E1^A repeats (SAUPE et al. 1995 ) and 31 out of the 42 amino acids are conserved within the 11 HET-D2^Y repeats. Whereas the similarity between the repeats of HET-E1^A and HET-D2^Y polypeptides is very strong and the number of WD40 repeats is variable, this is not the case for all other known WD40 proteins.

Concerted evolution can occur in two different cases: in multiple copies of a given gene family and/or in tandemly repeated sequences of a single gene (SMITH 1976 ). These two different cases of concerted evolution can apply to the het-e and het-d genes.

Multiple copies of a given gene family can undergo concerted evolution so that the sequences of all gene copies are very similar within a given species, although they normally diverge between different species. The primary driving force for concerted evolution is intrachromosomal; interchromosomal genetic exchanges are much rarer. The het-e and het-d genes could have arisen by gene duplication. STEWART and CULLEN 1999 showed that duplicated genes are often clustered. Recent results indicate that the het-d and het-e genes are not found in the same chromosome (C. BARREAU, personal communication). Moreover, the amino acid conservation between the HET-E and HET-D WD40 repeats is not always linked to codon conservation: 11 conserved amino acids do not use the same codon in the HET-E and HET-D WD40 repeats (data not shown). As the het-d and het-e genes are not in the same chromosome and do not seem to be subjected to concerted evolution, the het-d and het-e genes seem to be orthologs, having been subjected to parallel evolution.

Concerted evolution of tandemly repeated sequences is due to gene conversion (DOVER 1982 ) and unequal crossing over (SMITH 1976 ). Due to unequal crossing over, the number of repeats is variable among individuals as has been shown for the polyubiquitin alleles (BAKER and BOARD 1989 ). SAUPE 2000 proposed that the het-e WD40 domain evolved in a concerted way. The high degree of similarity with the HET-D WD40 coding sequences suggests that het-d repeats are also subjected to concerted evolution. The variable number of repeats found in the different het-e and het-d alleles corroborates this suggestion. Concerted evolution seems not to be the only mechanism to explain the het-e and het-d repeat evolution. In fact, five positions are highly polymorphic between the WD40 repeats in the HET-E and HET-D polypeptides. These polymorphic positions are located in the loop between the d and a strands and the turn between the b and c strands. Evolution seems to have proceeded in two stages: homogenized concerted evolution followed by specific mutational evolution at specific polymorphic positions.

The crystal structure of the Gß protein shows that the seven WD40 repeats fold into a seven-blade ß-propeller (SONDEK et al. 1996 ). The seven-blade ß-propeller has also been demonstrated for the C-terminal domain of Tup1 (SPRAGUE et al. 2000 ). The Tup1-Ssn6 complex regulates the expression of several sets of genes in the yeast Saccharomyces cerevisiae. The ß-propeller structure looks like a torus with a central tunnel. In the case of a highly symmetrical structure like the HET-E and HET-D WD40 repeats, the central tunnel must be quite circular. In WD40 repeat propellers, a hallmark of blades is the hydrogen-bonded tetrad comprising Trp-30, Thr/Ser-20, His-2, and Asp-24 (WALL et al. 1995 ; SONDEK et al. 1996 ; SPRAGUE et al. 2000 ). In the Gß subunit and in Tup1, not all members of the tetrad are present in every motif. In the HET-E and HET-D polypeptides, all four members of the hydrogen-bonded structural tetrad are found in each repeat, suggesting that more than seven blades are involved in the HET-E and HET-D putative ß-propeller structure.

The function of the WD40 repeats appears to be the organized binding of many proteins, either simultaneously or sequentially (NEER and SMITH 2000 ). The G subunit is tightly bound to the bottom surface of the Gß propeller. The G subunit is bound to the Gß central tunnel (NEER and SMITH 2000 ). The Gß structure interacts with >20 proteins (CLAPHAM and NEER 1997 ). These interacting proteins bind either to the top surface near the central tunnel or to the side of the torus (NEER and SMITH 2000 ). Given the conservation of the WD40 repeats, it is very likely that all proteins containing multiple WD40 repeats will form a propeller structure. On the basis that the HET-E and HET-D polypeptides, even with more than seven repeats, take a ß-propeller structure, the polymorphic positions found in the different repeats will all be located at the top of the ß-propeller. Our results show that amino acids present at the polymorphic positions in the sixth and seventh repeats are essential for the incompatible interaction between the HET-E2 and HET-C3 proteins. BEGUERET et al. 1994 proposed the poison complex model to explain the vegetative incompatibility reaction. If the two alleles are compatible, the HET-E (or HET-D) and HET-C proteins will form a viable complex. If the two alleles are incompatible, the HET-E (or HET-D) and HET-C proteins will form a poison complex that will be lethal for the cell. Although a physical interaction between HET-E (or HET-D) and HET-C has not been demonstrated yet, it is possible that the HET-C proteins interact at the top of the HET-E ß-propeller. The polymorphism located at the top of the ß-propeller structure should be maintained to allow binding of HET-C protein from different het-c alleles.

Different hypotheses referring to the biological significance of vegetative incompatibilty have been proposed (GLASS and KULDAU 1992 ; BEGUERET et al. 1994 ; WORRALL 1997 ; SAUPE 2000 ). There are two major conflicting theories. According to the first hypothesis, vegetative incompatibility limits heterokaryosis to prevent horizontal transmission of viruses or other deleterious organelles. If this is the case, het gene evolution should favor maintaining the occurrence of vegetative incompatibility. In the second hypothesis, however, vegetative incompatibility is considered an accident of evolution. For instance, neutral polymorphism in het genes creates a variant with lethal consequences for the heterokaryotic cell when it is associated with its incompatible antagonist. In the "accident" hypothesis, there is no biological reason to maintain polymorphism selectively among het genes. Our results, however, demonstrate that the fungi selectively maintain specific polymorphism at the top of the ß-propeller structure, ruling out the accident hypothesis in this case. Vegetative incompatibility seems, instead, to be necessary in the P. anserina life cycle.

	FOOTNOTES

Sequence data from this article have been deposited with the EMBL/GenBank Data Libraries under accession nos. AF323585, AF323582, and AF323583.
¹ Present address: Institut de Recherche sur la biologie de l'insecte, UPRESA 6035, Université F. Rabelais, 37200 Tours, France.
² Present address: School of Biological Sciences, University of Exeter, Exeter EX4 4QG, England.

	ACKNOWLEDGMENTS

E.E. was supported by a fellowship from the Ministère de la Recherche.

Manuscript received November 5, 2001; Accepted for publication February 8, 2002.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

AKINS, R. A. and A. M. LAMBOWITZ, 1985  General method for cloning Neurospora crassa nuclear genes by complementation of mutants. Mol. Cell. Biol. 5:2272-2278[Abstract/Free Full Text].

ANAGNOSTAKIS, S. L., 1977  Vegetative incompatibility in Endothia parasitica.. Exp. Mycol. 1:306-316.

BAKER, R. T. and P. G. BOARD, 1989  Unequal crossover generates variation in ubiquitin coding unit number at the human UbC polyubiquitin locus. Am. J. Hum. Genet. 44:534-542[Medline].

BALLANCE, D. J., 1991 Transformation systems for filamentous fungi and an overview of fungal gene structure, pp. 1–29 in Molecular Industrial Mycology, edited by S. A. LEONG and R. M. BERKA. Marcel Dekker, New York.

BEGUERET, J., B. TURCQ, and C. CLAVE, 1994  Vegetative incompatibility in filamentous fungi: het genes begin to talk. Trends Genet. 10:441-446[Medline].

BEISSON-SCHECROUN, J., 1962  Incompatibilité cellulaire et interactions nucléo-cytoplasmiques dans les phénomènes de "barrage" chez le Podospora anserina.. Ann. Genet. 4:3-50.

BERGES, T. and C. BARREAU, 1989  Heat shock at an elevated temperature improves transformation efficiency of protoplasts from Podospora anserina.. J. Gen. Microbiol. 135:601-604[Medline].

BERNET, J., 1965  Mode d'action des gènes de "barrage" et relation entre l'incompatibilité cellulaire et l'incompatibilité sexuelle chez Podospora anserina. Ann. Sci. Nat. Bot. Ser. 12 6:611-768.

BERNET, J., 1967  Les systèmes d'incompatibilité chez le Podospora anserina.. C. R. Acad. Sci. Paris 265:1330-1333.

BOURNE, H. R., D. A. SANDERS, and F. MCCORMICK, 1991  The GTPase superfamily: conserved structure and molecular mechanism. Nature 349:117-127[Medline].

BRUCHEZ, J. J. P., J. EBERLE, and V. E. RUSSO, 1993  Regulatory sequences in the transcription of Neurospora crassa genes: CAAT box, TATA box, introns, poly(A) tail formation sequences. Fungal Genet. Newsl. 40:89-96.

CLAPHAM, D. E. and E. J. NEER, 1997  G protein beta gamma subunits. Annu. Rev. Pharmacol. Toxicol. 37:167-203[Medline].

CORTESI, P. and M. G. MILGROOM, 1998  Genetics of vegetative incompatibility in Cryphonectria parasitica.. Appl. Environ. Microbiol. 64:2988-2994[Abstract/Free Full Text].

COUSTOU, V., C. DELEU, S. SAUPE, and J. BÉGUERET, 1997  The protein product of the het-s heterokaryon incompatibility gene of the fungus Podospora anserina behaves as a prion analog. Proc. Natl. Acad. Sci. USA 94:9773-9778[Abstract/Free Full Text].

DOVER, G., 1982  Molecular drive: a cohesive mode of species evolution. Nature 299:111-117[Medline].

ESPAGNE, E., P. BALHADÈRE, J. BEGUERET, and B. TURCQ, 1997  Reactivity in vegetative incompatibility of the HET-E protein of the fungus Podospora anserina is dependent on GTP-binding activity and a WD40 repeated domain. Mol. Gen. Genet. 256:620-627[Medline].

ESSER, K., 1974 Podospora anserina, pp. 531–551 in Handbook of Genetics, edited by R. C. KING. Plenum Press, New York.

ESSER, K. and R. BLAICH, 1973  Heterogenic incompatibility in plants and animals. Adv. Genet. 17:107-152.

FONG, H. K., J. B. HURLEY, R. S. HOPKINS, R. MIAKE-LYE, and M. S. JOHNSON et al., 1986  Repetitive segmental structure of the transducin ß subunit: homology with the CDC4 gene and identification of related mRNAs. Proc. Natl. Acad. Sci. USA 83:2162-2166[Abstract/Free Full Text].

GARCIA-HIGUERA, I., J. FENOGLIO, Y. LI, C. LEWIS, and M. P. PANCHENKO et al., 1996  Folding of proteins with WD-repeats: comparison of six members of the WD-repeat superfamily to the G protein ß subunit. Biochemistry 35:13985-13994[Medline].

GLASS, N. L. and G. A. KULDAU, 1992  Mating type and vegetative incompatibility in filamentous ascomycetes. Annu. Rev. Phytopathol. 30:201-224.

GLASS, N. L., S. J. VOLLMER, C. STABEN, J. GROTELUESCHEN, and R. L. METZENBERG et al., 1988  DNAs of the two mating-type alleles of Neurospora crassa are highly dissimilar. Science 241:570-573[Abstract/Free Full Text].

GLASS, N. L., J. GROTELUESCHEN, and R. L. METZENBERG, 1990  Neurospora crassa A mating-type region. Proc. Natl. Acad. Sci. USA 87:4912-4916[Abstract/Free Full Text].

GRINDLE, M., 1963  Heterocaryon incompatibility of unrelated strains in the Aspergillus nidulans group. Heredity 18:191-204[Medline].

JAVERZAT, J. P., C. JACQUIER, and C. BARREAU, 1993  Assignment of linkage groups to the electrophoretically-separated chromosomes of the fungus Podospora anserina.. Curr. Genet. 24:219-222[Medline].

LAMBRIGHT, D. G., J. SONDEK, A. BOHM, N. P. SKIBA, and H. E. HAMM et al., 1996  The 2.0 Å crystal structure of a heterotrimeric G protein. Nature 379:311-319[Medline].

LECELLIER, G. and P. SILAR, 1994  Rapid methods for nucleic acids extraction from Petri dish-grown mycelia. Curr. Genet. 25:122-123[Medline].

LOUBRADOU, G. and B. TURCQ, 2000  Vegetative incompatibility in filamentous fungi: a roundabout way of understanding the phenomenon. Res. Microbiol. 151:239-245[Medline].

NEER, E. J. and T. F. SMITH, 2000  A groovy new structure. Proc. Natl. Acad. Sci. USA 97:960-962[Free Full Text].

ORBACH, M. J., 1994  A cosmid with a HyR marker for fungal library construction and screening. Gene 150:159-162[Medline].

PERKINS, D. D., A. RADFORD, D. NEWMEYER, and M. BJORKMAN, 1982  Chromosomal loci of Neurospora crassa.. Microbiol. Rev. 46:426-570[Free Full Text].

RIZET, G., 1952  Les phénomènes de barrages chez Podospora anserina: analyse génétique des barrages entre les souches s et S. Rev. Cytol. Biol. Vég. 13:51-92.

RIZET, G. and C. ENGELMAN, 1949  Contribution à l'étude génétique d'un ascomycète tetrasporé: Podospora anserina. Rev. Cytol. Biol. Vég. 11:201-304.

SAMBROOK, J., E. F. FRITSCH and T. MANIATIS, 1989 Molecular Cloning: A Laboratory Manual, Ed. 2. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.

SARASTE, M., P. R. SIBBALD, and A. WITTINGHOFER, 1990  The P-loop, a common motif in ATP- and GTP-binding proteins. Trends Biochem. Sci. 15:430-434[Medline].

SAUPE, S. J., 2000  Molecular genetics of heterokaryon incompatibility in filamentous ascomycetes. Microbiol. Mol. Biol. Rev. 64:489-502[Abstract/Free Full Text].

SAUPE, S., C. DESCAMPS, B. TURCQ, and J. BÉGUERET, 1994  Inactivation of the Podospora anserina vegetative incompatibility locus het-c, whose product resembles a glycolipid transfer protein, drastically impairs ascospore production. Proc. Natl. Acad. Sci. USA 91:5927-5931[Abstract/Free Full Text].

SAUPE, S., B. TURCQ, and J. BÉGUERET, 1995  A gene responsible for vegetative incompatibility in the fungus Podospora anserina encodes a protein with a GTP-binding motif and Gß homologous domain. Gene 162:135-139[Medline].

SAUPE, S. J., G. A. KULDAU, M. L. SMITH, and N. L. GLASS, 1996  The product of the het-C heterokaryon incompatibility gene of Neurospora crassa has characteristics of a glycine-rich cell wall protein. Genetics 143:1589-1600[Abstract].

SMITH, G. P., 1976  Evolution of repeated DNA sequences by unequal crossover. Science 191:528-535[Abstract/Free Full Text].

SMITH, M. L., O. C. MICALI, S. P. HUBBARD, N. MIR-RASHED, and D. J. JACOBSON et al., 2000  Vegetative incompatibility in the het-6 region of Neurospora crassa is mediated by two linked genes. Genetics 155:1095-1104[Abstract/Free Full Text].

SMITH, T. F., C. GAITATZES, K. SAXENA, and E. J. NEER, 1999  The WD repeat: a common architecture for diverse functions. Trends Biochem. Sci. 24:181-185[Medline].

SONDEK, J., A. BOHM, D. G. LAMBRIGHT, H. E. HAMM, and P. B. SIGLER, 1996  Crystal structure of a G-protein ß dimer at 2.1A resolution. Nature 379:369-374[Medline].

SPRAGUE, E. R., M. J. REDD, A. D. JOHNSON, and C. WOLBERGER, 2000  Structure of the C-terminal domain of Tup1, a corepressor of transcription in yeast. EMBO J. 19:3016-3027[Medline].

STABEN, C. and C. YANOFSKY, 1990  Neurospora crassa a mating-type region. Proc. Natl. Acad. Sci. USA 87:4917-4921[Abstract/Free Full Text].

STEWART, P. and D. CULLEN, 1999  Organization and differential regulation of a cluster of lignin peroxidase genes of Phanerochaete chrysosporium.. J. Bacteriol. 181:3427-3432[Abstract/Free Full Text].

TURCQ, B. and J. BEGUERET, 1987  The ura5 gene of the filamentous fungus Podospora anserina: nucleotide sequence and expression in transformed strains. Gene 53:201-209[Medline].

TURCQ, B., M. DENAYROLLES, and J. BEGUERET, 1990  Isolation of the two allelic incompatibility genes s and S of the fungus Podospora anserina.. Curr. Genet. 17:297-303.

VAN DER VOORN, L. and H. L. PLOEGH, 1992  The WD-40 repeat. FEBS Lett. 307:131-134[Medline].

WALL, M. A., D. E. COLEMAN, E. LEE, J. A. INIGUEZ-LLUHI, and B. A. POSNER et al., 1995  The structure of the G protein heterotrimer G_i1ß₁₂. Cell 83:1047-1058[Medline].

WORRALL, J. J., 1997  Somatic incompatibility in basidiomycetes. Mycologia 89:24-36.

This article has been cited by other articles:

				M. Paoletti and C. Clave The Fungus-Specific HET Domain Mediates Programmed Cell Death in Podospora anserina Eukaryot. Cell, November 1, 2007; 6(11): 2001 - 2008. [Abstract] [Full Text] [PDF]

				K. Pal, A.D. van Diepeningen, J. Varga, R.F. Hoekstra, P.S. Dyer, and A.J.M. Debets Sexual and vegetative compatibility genes in the aspergilli Stud Mycol, January 1, 2007; 59(1): 19 - 30. [Abstract] [Full Text] [PDF]

				K. Dementhon, G. Iyer, and N. L. Glass VIB-1 Is Required for Expression of Genes Necessary for Programmed Cell Death in Neurospora crassa Eukaryot. Cell, December 1, 2006; 5(12): 2161 - 2173. [Abstract] [Full Text] [PDF]