Characterization of mat A-2, mat A-3 and matA Mating-Type Mutants of Neurospora crassa

Corresponding author: N. Louise Glass, Rm. 238 Wesbrook, The Biotechnology Laboratory, The University of British Columbia, Vancouver, BC V6T 1Z3, Canada, glass{at}unixg.ubc.ca (E-mail).

Communicating editor: R. H. DAVIS

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The mating-type locus of Neurospora crassa regulates mating identity and entry into the sexual cycle. The mat A idiomorph encodes three genes, mat A-1, mat A-2, and mat A-3. Mutations in mat A-1 result in strains that have lost mating identity and vegetative incompatibility with mat a strains. A strain containing mutations in both mat A-2 and mat A-3 is able to mate, but forms few ascospores. In this study, we describe the isolation and characterization of a mutant deleted for mat (matA), as well as mutants in either mat A-2 or mat A-3. The matA strain is morphologically wild type during vegetative growth, but it is sterile and heterokaryon compatible with both mat A and mat a strains. The mat A-2 and mat A-3 mutants are also normal during vegetative growth, mate as a mat A strain, and produce abundant biparental asci in crosses with mat a, and are thus indistinguishable from a wild-type mat A strain. These data and the fact that the mat A-2 mat A-3 double mutant makes few asci with ascospores indicate that MAT A-2 and MAT A-3 are redundant and may function in the same pathway. Analysis of the expression of two genes (sdv-1 and sdv-4) in the various mat mutants suggests that the mat A polypeptides function in concert to regulate the expression of some sexual development genes.

MANY fungal mating-type genes encode products with characteristics of transcription factors (reviewed in HERSKOWITZ 1989 ; GLASS and NELSON 1994 ; KOTHE 1996 ). Mating type in Saccharomyces cerevisiae is the best-characterized fungal mating system, and activation, repression, and protein-protein interactions have been demonstrated biochemically and genetically in that system. Mating-type genes in yeast regulate the fusion of haploid cells of the opposite mating type, as well as the entry of the diploid cell into meiosis (reviewed in JOHNSON 1995 ; HERSKOWITZ 1989 ; NASMYTH and SHORE 1987 ). In filamentous fungi, mating-type sequences also regulate the fusion of reproductive structures of two genetically distinct individuals, which results in the production of meiotic progeny.

The filamentous ascomycete fungus, Neurospora crassa, has two mating-types, A and a, which regulate entry into the sexual cycle (reviewed in GLASS and NELSON 1994 ). The mating-types are determined by dissimilar DNA sequences at the mating-type locus (mat)⁴ and have been termed idiomorphs (METZENBERG and GLASS 1990 ). In mat a individuals, a single gene, mat a-1, is required for mating identity, postfertilization functions, and for vegetative incompatibility with mat A strains (GRIFFITHS and DELANGE 1978 ; CHANG and STABEN 1994 ). MAT a-1 has an HMG domain and has been shown to bind to DNA by in vitro assays (STABEN and YANOFSKY 1990 ; PHILLEY and STABEN 1994 ). The mat A idiomorph contains three genes, mat A-1, mat A-2, and mat A-3 (FERREIRA et al. 1996 ). The mat A-1 gene is required for mat A mating identity, postfertilization functions, and vegetative incompatibility with mat a strains (GLASS et al. 1990 ; SAUPE et al. 1996 ). MAT A-1 has similarity to the S. cerevisiae mating-type transcriptional activator MAT1 (SPRAGUE et al. 1983 ). MAT A-2 shows ~20% identity with the mating-type polypeptide SMR1 of the pseudohomothallic ascomycete Podospora anserina (FERREIRA et al. 1996 ). SMR1 is one of the three genes (FMR1, SMR1, and SMR2) encoded by the mat- mating-type locus of P. anserina (DEBUCHY et al. 1993 ). MAT A-3 has an HMG domain and shows ~20% identity with both MAT a-1 and P. anserina SMR2 (DEBUCHY et al. 1993 ; FERREIRA et al. 1996 ).

A mutant, A^IIRIP, which contains mutations in both mat A-2 and mat A-3, normally mates as a mat A strain, but produces very few asci with ascospores (GLASS and LEE 1992 ; FERREIRA et al. 1996 ). The effect of individual mutations in mat A-2 and mat A-3 was not determined. Mutation of the P. anserina SMR1 or SMR2 genes also results in strains that mate qualitatively normally, but that produce a reduced number of progeny (DEBUCHY et al. 1993 ). Cytological analyses of reproductive structures from crosses between P. anserina mat+ and SMR1 or SMR2 mutants show haploid meioses and uniparental progeny (ZICKLER et al. 1995 ); uniparental spores are of the mutant SMR1 or SMR2 genotypes only.

The products of the mating-type locus of N. crassa and P. anserina are thought to act as master regulators of both mating and nuclear identity during sexual development. However, no direct target genes of any of the mating type products have been identified. NELSON and METZENBERG 1992 isolated genes (sdv) that are expressed early in sexual development. A functional mat A-1 was required for the expression of at least seven of the sdv genes.

In this study, we describe the construction and phenotypes of a mating type deletion strain, as well as mat A-2- and mat A-3-specific mutants. We show that the mating-type genes are not essential for N. crassa, and that mutations in mat A-2 or mat A-3, unlike their P. anserina counterparts, do not dramatically affect sexual development nor do they result in the production of uniparental asci. We also analyzed the expression of two sdv genes in the various mat mutants. Both genetic data and sdv gene analyses suggest that in N. crassa, the regulation of sexual development may involve the formation of complexes of the mating-type products that activate or repress genes in different pathways.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

N. crassa and Escherichia coli strains:
N. crassa strains used in this study and their relevant genotypes are listed in Table 1. E. coli XL1-blue (Stratagene, La Jolla, CA) was used for all DNA manipulations involving the construction of the mating- type deletion strain, and E. coli DH5 (BRL, Burlington, Ontario, Canada) was used for all other manipulations.

Construction of plasmids for making the matA strain:
The wild-type Schizophyllum commune ADE5 gene encodes aminoimidazole riboside-5'-phosphate synthase (ALIC et al. 1990 ) and is capable of complementing ad-2 mutants of N. crassa, albeit weakly (our unpublished observations). ADE5 was used as a selectable marker for gene replacement at the mating- type locus. The gene replacement plasmid (p2ADE-X) was constructed as follows: the 5.3-kbp A idiomorph and its left and right flanking regions (4 and 2 kbp, respectively) were cloned into pBR322 to give p2A4. The 5.7-kbp NdeI-Nsi I fragment (Figure 1), which contains mat A, was replaced with the 0.6-kbp multiple cloning site region from pGEM3Z(-) (Promega, Madison, WI) to give p2LAC3Z-4DA. The 3.8-kbp KpnI-BamHI fragment from pADE5-2g (ALIC et al. 1990 ), which contains the ADE5 gene of S. commune, was subcloned into the multiple cloning region of p2LAC3Z-4DA to give p2ADE-X.

View larger version (28K): In this window In a new window Download PPT slide   Figure 1. —Construction of mating-type deletion strain (matA; RLM 44-02 deposited as FGSC 8292) by transformation of strain RLM 41-10 with plasmid p2ADE-X. LGI, N. crassa linkage group I.

Construction of plasmids for generating mat A-2 and mat A-3 repeat-induced point mutants:
Plasmid pGq1310 (constructed by J. GROTELUESCHEN, Department of Biomolecular Chemistry, University of Wisconsin) contains a 1.3-kb Bcl I/PvuII mat A fragment (mat A-2) in a qa-2⁺ vector. Plasmid pAL2 contains a 2.5-kb EcoRV mat A fragment (mat A-3) interrupted by a 1.3-kbp BamHI fragment containing the hygromycin phosphotransferase gene (hph) inserted into the BamHI site in the centromere distal flank region of the mat A locus (constructed by S. SAUPE, Department of Botany and the Biotechnology Laboratory, University of British Columbia). Neither the pGq1310 nor the pAL2 plasmids contain the 5' ends of mat A-2 or mat A-3 ORFs.

N. crassa transformation and repeat-induced point mutation analysis:
N. crassa spheroplasts were prepared from germinated conidia by the method of AKINS and LAMBOWITZ 1985 , and they were transformed as described by VOLLMER and YANOFSKY 1986 . Transformants were normally grown for 3–4 days at 30°. For screening of matA strains, transformants were kept on plates in subdued light at 30° for 10 days and then moved to bright light at room temperature until all or nearly all transformants formed conidia. Mating-type tests were performed on the medium of WESTERGAARD and MITCHELL 1947 .

Plasmids pGq1310 and pAL2 were introduced into RLM 52-22 (qa-2; aro-9 A) or NLG R2-39 (thr-2 het-6PA A) strains, respectively, for isolation of the mat A-2 and mat A-3 mutants. Transformants were randomly chosen (~25 from each transformation) and grown under selective conditions. Homokaryons were isolated and crossed to wild-type mat a (FGSC 532) or a temperature-sensitive strain, un-3 a (NLG R1-09); mat A progeny were chosen for further screening as described in GLASS and LEE 1992 . Genomic DNA from the transformants and from first- and second-generation progeny was isolated (RAEDER and BRODA 1985 ) and digested with two isoschizomers, MboI (C-methylation insensitive) and Sau3AI (C-methylation sensitive). After gel electrophoresis, DNA was transferred to nylon membranes (Nytran; Schleicher & Schuell, Keene, NH). Membranes were probed with ³²P-labeled mat A-2, mat A-3, or qa-2–specific fragments, and were then exposed to X-ray films. DNA sequence analysis was performed using an automated sequencing procedure (Applied Biosystems, Inc., Mississauga, Ontario, Canada) at the Nucleic Acid-Protein Service Unit at Biotechnology Laboratory of the University of British Columbia (Vancouver, British Columbia, Canada).

Phenotypic characterization of mutants:
Photographs of perithecial squashes from mutant and wild-type crosses were taken at x10. The number of ascospores was quantified in the following manner: at 8 and 11 days after fertilization, 10 perithecia were picked from a crossing plate and transferred to an Eppendorf tube containing 25 µl of water. Perithecia were lightly squashed in the tube using a fitted plastic pestle. The number of ascospores present in a sample of the supernatant was counted using a hemacytometer. The ascospores present in five samples from each cross were counted. Statistical analyses were applied using analysis of variance and Student's t-test.

Heterokaryon tests:
For heterokaryon tests, 25 µl of conidial suspensions containing approximately the same number of conidia (4 x 10⁶/ml) from each auxotrophic strain were mixed and inoculated onto minimal medium (VOGEL 1964 ) in race tubes (DAVIS and DESERRES 1970 ). Linear growth of compatible and incompatible controls plus the experimental heterokaryons were measured over time. All heterokaryon tests were performed in triplicate.

RNA analyses:
N. crassa strains for RNA analyses were grown as described by NELSON and METZENBERG 1992 . Mycelial mats (grown for 3–4 days) were isolated by filtration and stored frozen at -70°. RNA isolation, reverse transcription (RT)-PCR, and Northern blots were performed as described (FERREIRA et al. 1996 ), except that for Northern analyses, 3 µg of polyA⁺ RNA was loaded per lane. Hybridization of membranes (Zeta-Probe; Bio-Rad, Richmond, CA) was carried out according to the manufacturer's suggestions. After a 24-h exposure to X-ray film, membranes were exposed to a Phosphorscreen (Molecular Dynamics, Sunnyvale, CA). The screen was then scanned in a Phosphorimager scanner after 18–24 h of exposure and analyzed using Imagequant software (Molecular Dynamics). Radiation emitted from the blots was quantified for sdv-specific transcripts and for the constitutive gene controls. Controls included the N. crassa crp-1 gene for the small ribosomal subunit protein, which produces two transcripts (KREADER and HECKMAN 1987 ), and/or the ß-tubulin gene (tub-1) (ORBACH et al. 1986 ). For subsequent analyses with other sdv genes, radioactivity on blots was allowed to decay or was stripped away by boiling the blot for 20 min in 0.1x SSC/0.5% SDS.

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Deletion of the A mating type locus:
The general strategy for removal of the mating type sequences was analogous to that used by CHANG and STABEN 1994 for replacement of one mating type by another. Their procedure took advantage of heterokaryon incompatibility between the resident mat A mating type and the introduced mat a mating-type gene to counterselect the very large majority of ectopic integrants that contained both mating types. Integrants in which the resident mating-type gene had been replaced were spared the incompatibility reaction and could grow. Because we wished to replace the mat A mating type genes with nonmating-type sequences (the ADE5 gene of S. commune), we needed to introduce an additional step. Spheroplasts of RLM 41-10 (thi-4 ad-2; lys-1 A) were transformed to (partial) adenine independence with p2ADE-X, which contains the S. commune ADE5 gene flanked by the flanking sequences of N. crassa mat A (see MATERIALS AND METHODS). About 3600 transformants were selected on 18 plates containing lysine plus thiamine, and the ~200 colonies per plate were allowed to conidiate. The conidia from each plate were suspended in minimal medium (VOGEL 1964 ), on which they were, of course, unable to grow. They could, however, be rescued by forming lysine- and thiamine-independent heterokaryons with a second strain, RLM 73-12 (ad-2; am132 inl inv mei-2 a). At least in principle, this nutritional helper strain of the mat a mating-type should allow survival and growth of only those transformants in which mat A had been deleted or otherwise inactivated. A comparable number of conidia of RLM 73-12 was therefore added to each of the 18 pools of conidia from the transformants. After 2 wk at 30°, one of the 18 tubes had formed a vigorous heterokaryon and had conidiated. Mating-type tests of this strain showed that it mated as a mat a strain, but not as a mat A strain, suggesting that it was a heterokaryon carrying the desired deletion. Conidia of this strain were streaked onto lysine plus thiamine plates, incubated at room temperature overnight, and 34 germinated conidia were picked to nonselective medium. Of the resulting cultures, 28 proved to require lysine and thiamine, and they were infertile with both mating types. Genomic DNA from four of these strains was tested on a Southern blot with several probes. As expected, this revealed that the NdeI-NsiI fragment, which contains the A idiomorph, was absent, and that the KpnI-BamHI fragment, which contains the ADE5 gene of S. commune, was present instead (data not shown). Figure 1 summarizes the double-crossover replacement event that is consistent with the data from blotting. One of these strains (matA) was used in the remaining work and is deposited in the Fungal Genetics Stock Center as FGSC 8292. We refer to this strain as matA to reflect the fact that it is a deletion of the mat A idiomorph; differences are present in the flanking regions of the mat locus between mat a and mat A strains (RANDALL and METZENBERG 1995 ).

The matA strain was phenotypically identical to wild type during vegetative growth, and it produced female reproductive structures (protoperithecia) and conidia. The matA strain, however, was sterile and did not mate with either mat A or mat a strains, but it formed a heterokaryon with either mat A or mat a strains, a phenotype identical to that described for mat A-1 mutants (GRIFFITHS 1982 ). The capacity of the matA strain to mate and produce perithecia as either an A or a strain was restored by the introduction of ectopic copies of either mat A or mat a sequences, respectively, but ascospores were not produced (our unpublished results).

Isolation of mat A-2 and mat A-3 mutants:
To determine the phenotype of mat A-2- and mat A-3-specific mutants, we chose to use a method termed repeat-induced point (RIP) mutation, which is a natural mutagenic process in N. crassa that involves duplicated sequences (SELKER 1990 ). RIP occurs when transformants containing duplicated sequences introduced by the transformation process are taken through a cross; both introduced and resident copies suffer GC to AT transition mutations and typically show methylation of C residues in the duplicated segments (SELKER and GARRETT 1988 ; CAMBARERI et al. 1989 ). This strategy was adopted primarily because of the low frequency of gene replacement events at the mating-type locus (text above and unpublished results). The mat A-2 plasmid (pGq1310) was introduced into RLM 52-22 (qa-2; aro-9 A) (see MATERIALS AND METHODS), and two independent homokaryotic transformants (G11 and G12) were chosen. G11 and G12 were crossed to FGSC 532 (OR a), and more than 200 mat A progeny resulting from the two crosses were analyzed. These progeny displayed wild-type morphology during vegetative growth. In crosses with mat a strains, all the progeny produced perithecia and ascospores and were indistinguishable from wild-type crosses.

To get a preliminary indication of whether RIP had occurred in the mat A progeny of G11 and G12 transformants without giving an obvious phenotype, genomic DNA was analyzed for DNA methylation by digestion with the isoschizomers Sau3AI and MboI. When a mat A-2 probe was hybridized to genomic DNA from random mat A progeny from the G11 and G12 crosses, methylation was observed in a few mat A progeny (data not shown).

Based on DNA methylation and RFLP analyses, a number of G11 and G12 mat A progeny were selected for DNA sequence analysis of the resident mat A-2 sequences. One progeny of G11, NLG R5-38 (mat A-2^m1), had 27 bp changes in mat A-2, resulting in eight amino acid substitutions and two stop codons in the mat A-2 ORF (Figure 2A). The first stop codon resulted in a truncated MAT A-2 of 195 amino acids. About 65% of the mutations occurred in 5'CpA 3' dinucleotides, which is typical of RIP events (GRAYBURN and SELKER 1989 ) (data not shown). All mutations were of the same polarity, resulting in C to T changes on the mat A-2 noncoding strand.

View larger version (15K): In this window In a new window Download PPT slide   Figure 2. —MAT A-2 and MAT A-3 wild-type and mutant proteins. (A) Changes in MAT A-2m1 as compared to the MAT A-2 wild-type protein. A stop codon appears at position 196, and there are amino acid substitutions at positions 62 (cysteine to tyrosine), 93 (aspartic acid to asparagine), and 96 (arginine to lysine). Shaded area represents conserved region between MAT A-2 and SMR1. (B) Changes in MAT A-3m1 and MAT A-3m2 as compared to MAT A-3. MAT A-3m1 is 83 amino acids long, and it has a threonine at position 47 (instead of alanine) and isoleucine in place of methionine (positions 48 and 80). MAT A-3m2 has a stop codon at position 112 and a leucine replacing a proline in position 102 of the polypeptide. The HMG box is stippled.

To isolate mat A-3-specific mutants, we introduced pAL2, which contains mat A-3, into NLG R2-39 (thr-2 het-6^PA A) spheroplasts. Two homokaryotic mat A-3 transformants (L22 and L25) were subjected to RIP by crossing to NLG R1-09 (un-3 a). As with the mat A progeny from the mat A-2 RIP crosses (see above), 200 mat A progeny from the L22 and L25 crosses were examined, and none of them showed a mutant phenotype during either vegetative growth or sexual reproduction. A number of L22 and L25 first- and second-generation mat A progeny were subjected to RFLP, DNA methylation, and DNA sequence analyses (data not shown). DNA sequence analysis of mat A-3 in one L22 progeny, NLG R4-66 (mat A-3^m1), showed 28 mutations in the first 600 bp that resulted in three amino acid substitutions and a stop codon at position 84 (Figure 2B). Fifty-seven percent of the changes occurred in 5'CpA3' dinucleotides (data not shown). DNA sequence analysis of mat A-3 in a second progeny of L22, NLG R5-39 (mat A-3^m2), revealed 16 mutations, two of which resulted in stop codons. The first stop codon would result in a truncated polypeptide of 111 amino acids (Figure 2B). Both the mat A-3^m1 and the mat A-3^m2 alleles have stop codons before the region encoding the HMG domain.

Mating-type specific cDNAs are absent in the matA-2 ^m1, mat A-3 ^m1, and A^IIRIP mutants:
Because the mat A-2^m1 and mat A-3^m1 mutants still produce abundant ascospores (see below), the extremely low fertility seen in the A^IIRIP mutant (GLASS and LEE 1992 ) was suspected of being caused by absence of both mat A-2 and mat A-3 function. The DNA sequencing of the mat A-2 and mat A-3 ORFs from A^IIRIP (NLG R4-16) showed mutations in mat A-3 (mat A-3^m3) that lead to only three amino acid substitutions, while mat A-2 was heavily mutated (mat A-2^m3) with a stop codon at amino acid 130 in MAT A-2 plus additional amino acid substitutions (FERREIRA et al. 1996 ). We therefore sequenced the promoter region between mat A-2^m3 and mat A-3^m3, and we found mutations in possible transcriptional regulatory regions of both mat A-2 and mat A-3 (Figure 3).

View larger version (70K): In this window In a new window Download PPT slide   Figure 3. —DNA sequence changes in the promoter region of mat A-2 and mat A-3 in NLG R4-16 (mat A-2m3 mat A-3m3). Nucleotide base changes are shown above promoter sequences. Most of the mutations were polarized G to A transitions in one strand; only two C to T mutations appear on the same strand. Transcriptional start sites and relevant motifs and repeats are shown as described in FERREIRA et al. 1996 . The DNA sequence between the proposed start codons of mat A-2 and mat A-3 is given. The arrows (in gray for mat A-2 and black for mat A-3) indicate the positions of the proximal (p) and distal (d) transcriptional start sites. The putative CAAT box sequences are shown in bold type. The stippled boxes show the repeated sequences, which are possibly important for transcription of mat A-2 and mat A-3.

To determine if mat A-2 and mat A-3 transcripts were produced in the mating-type mutants, a series of RT-PCR experiments was performed. RNA from WT A, NLG R5-38 (mat A-2^m1), NLG R4-66 (mat A-3^m1), and NLG R4-16 (mat A-2^m3 mat A-3^m3) strains grown in crossing medium was isolated and subjected to RT-PCR using mating-type-specific primers (Figure 4). Transcripts corresponding to mat A-2 were absent in cDNA preparations from NLG R5-38 (mat A-2^m1) and transcripts corresponding to mat A-3 were absent in cDNA preparations from NLG R4-66 (mat A-3^m1). Both mat A-2 and mat A-3 transcripts were absent from cDNA preparations from NLG R4-16 (mat A-2^m3 mat A-3^m3). In contrast, mat A-1 transcripts were present in all three mutants (Figure 4). These data indicated that the NLG R5-38 and NLG R4-66 strains were null for mat A-2 or mat A-3, respectively, and that the mutant phenotype associated with the NLG R4-16 was caused by the absence of functional mat A-2 and mat A-3.

View larger version (86K): In this window In a new window Download PPT slide   Figure 4. —Amplification of cDNA of mating-type genes in the FGSC 2489 (A), NLG R5-38 (A-2m1), NLG R4-66 (A-3m1), and NLG R4-16 (AIIRIP) mutants by RT-PCR. Genomic DNA from FGSC 2489 (gen) was included in all experiments as a control for the amplification reaction and for size comparison with cDNA. PCR negative control (neg) is shown on the left of the DNA size marker (M). (A) Amplification products (~0.9 kb cDNA, 1.2 kb genomic) using primers specific for mat A-2. (B) Amplification products (~0.95 kb cDNA, 1.1 kb genomic) for mat A-3. A nonspecific 0.6-kb band in mat A-3 cDNA amplifications was present. (C) Amplification of mat A-1 cDNA used as control for the reverse transcriptase reaction (~0.8 kb). Some cDNA preparations contained traces of genomic DNA that were also amplified by PCR.

Phenotypic analysis of mat mutants:
MAT A-2 and MAT A-3 each show ~20% amino acid identity to the P. anserina SMR1 and SMR2 mating-type peptides, respectively (FERREIRA et al. 1996 ). Crosses between P. anserina SMR1 or SMR2 mutants and wild type result in haploid meioses, a decrease in the number of meiotic progeny, and the occurrence of uniparental progeny of the SMR1 or SMR2 mutant genotype (ZICKLER et al. 1995 ). To determine if MAT A-2 and MAT A-3 function in a similar manner, crosses with the mat A-2^m1, mat A-3^m1, and mat A-2^m3 mat A-3^m3 mutants were monitored for the production of ascospores and for the occurrence of uniparental asci.

In crosses of the NLG R5-38 (mat A-2^m1) or the NLG R4-66 (mat A-3^m1) mutants and wild-type a strains (FGSC 532 and FGSC 2226), the number of ascospores produced at both 8 and 11 days after fertilization did not differ significantly from the number of progeny produced in wild-type crosses (data not shown). In contrast, crosses between wild-type a strains and NLG R4-16 (mat A-2^m3 mat A-3^m3) produce only a few asci per perithecium. These few asci contain eight black ascospores (GLASS and LEE 1992 ). To test for the production of uniparental asci, NLG R5-38, NLG R4-66, NLG R5-39, and NLG R4-16 strains were crossed to R a (FGSC 2088). The R (round-spore) mutation is ascus dominant (MITCHELL 1966 ). Thus, in crosses with an R mutant, all eight ascospores (4A and 4a) in each ascus are round. If uniparental asci of the mat A-2^m1, mat A-3^m1, mat A-3^m2, or mat A-2^m3 mat A-3^m3 genotypes were produced in a cross with R a, all ascospores within an ascus would be spindle shaped. More than 50 perithecia from each cross were observed cytologically, and only round ascospores were seen. A perithecial squash from a cross of mat A-3^m2 X R a is shown in Figure 5B. Figure 5C shows one of the few perithecia that contained asci and ascospores from mat A-2^m3 mat A-3^m3 X R a. Thus, mutations in mat A-2 or mat A-3, or in both mat A-2 and mat A-3, did not confer uniparental inheritance in N. crassa.

View larger version (80K): In this window In a new window Download PPT slide   Figure 5. —Perithecial squashes of crosses of mat A, NLG R5-39 (mat A-3m2 thr-2), and NLG R4-16 (mat A-2m3 mat A-3m3) to R a. (A) A rosette of asci with spindle-shaped ascospores from a wild-type cross. (B) A rosette 8 days after fertilization in a cross between mat A-3m2 and R a that shows production of round spores. (C) A cross between mat A-2m3 mat A-3m3 and R a. Only a small number of perithecia from this cross produce a few asci that bear eight ascospores. Arrows indicate round ascospores from asci.

mat A-2 and mat A-3 are not involved in vegetative incompatibility: The mat A-1 gene confers both mating identity and mating-type-associated vegetative incompatibility. All of the known mat A-1 mutants (with one exception) have lost their mating and heterokaryon incompatibility functions concomitantly (GRIFFITHS 1982 ; SAUPE et al. 1996 ). To determine if mutations in mat A-2 or mat A-3 affected mating-type-associated incompatibility, we assayed the growth of the mat A-2^m1 and mat A-3^m1 mutants as forced heterokaryons with an a strain. A heterokaryon between NLG R4-67 (pyr-4 mat A-3^m1) and NLG R2-54 (qa-2; aro-9 a) grew at a similar rate and had a phenotype similar to that of an incompatible heterokaryon between RLM 52-22 (qa-2; aro-9 A) and RLM 57-30 (pyr-4; cyh-1 a) (data not shown). A heterokaryon between NLG R5-40 (ad-3B mat A-2^m1) and NLG R2-54 (qa-2; aro-9 a) was inhibited and grew at a rate similar to an incompatible heterokaryon between RLM 52-22 (qa-2; aro-9 A) and RLM 57-30 (pyr-4; cyh-1 a) (data not shown). Thus, mutations in mat A-2 and mat A-3 do not appreciably affect heterokaryon incompatibility mediated by mat A-1 and mat a-1.

Expression of sexual development (sdv) genes is altered in the mating type mutants:
Several sdv genes isolated by subtractive hybridization require a functional mat A-1 for expression and were therefore potential target genes for the mating type proteins (NELSON and METZENBERG 1992 ). We analyzed the expression of a number of sdv genes by Northern analyses in mat A-1^m92 (mat A-1^RIP; GLASS and LEE 1992 ), mat a-1^m1 (GRIFFITHS 1982 ), mat A-2^m3 mat A-3^m3 (GLASS and LEE 1992 ), and the mat A-2^m1, mat A-3^m1, and matA mutants. In initial RNA analyses, variable expression of some sdv genes was observed during time course experiments in various mat mutants although the requirement for mat A-1 for expression was consistent (data not shown). The expression pattern of sdv-1 and sdv-4 in these strains, however, did not vary materially when analyzed from different RNA preparations during time course experiments, and these genes were therefore analyzed further.

Figure 6A shows analysis of sdv-1 expression. Although the absence of functional MAT A-1 (mat A-1^m92) in strain NLG R4-50 resulted in a dramatic decrease in transcript levels of sdv-1, transcripts for sdv-1 were observed in the matA strain, i.e., in the absence of MAT A-1, MAT A-2, and MAT A-3. These observations indicated that either MAT A-2 or MAT A-3 or both repress the expression of sdv-1 when mat A-1 is inactivated by RIP. The transcript level of sdv-1 in both NLG R5-38 (mat A-2^m1) and NLG R4-66 (mat A-3^m1) strains was higher than in a wild-type mat A strain (about twofold). In NLG R4-16 (mat A-2^m3 mat A-3^m3), the transcript level of sdv-1 was more than threefold higher than in a mat A strain, consistent with our hypothesis that MAT A-2 and MAT A-3 modulate the expression of sdv-1 in conjunction with MAT A-1. Because similar levels of sdv-1 transcript were observed in FGSC 532 (mat a) and FGSC 4565 (mat a-1^m1), a functional MAT a-1 is not required for the expression of sdv-1 in a mat a context. Transcript levels of sdv-1 were typically about threefold higher in wild-type a strains than in A strains, indicating that factors in the mat a and mat a-1^m1 strains function to express sdv-1 in the absence of mat A-1. Transcripts from sdv-1 were also detected in RNA from perithecia and from a mat A/mat a duplication strain (PS D25) grown in crossing medium.

View larger version (62K): In this window In a new window Download PPT slide   Figure 6. —Expression of sdv-1 and sdv-4 in mat A, mat a, and various mating-type mutants grown in crossing media (WESTERGAARD and MITCHELL 1947 ). Results are presented as the ratio of radioactivity emitted by hybridization to probes from the sdv-1 or sdv-4 genes compared to that emitted by hybridization to probes from the constitutive genes tub-1 (ORBACH et al. 1986 ) and/or crp-1 (KREADER and HECKMAN 1987 ). All comparative analyses were done relative to the ratio obtained with RNA of wild-type A control (normalized to 1). Standard deviation bars that were calculated from at least three separate experiments show variability. (A) Expression levels of sdv-1 in the various strains. (B) Expression levels of sdv-4 in the various strains. A, FGSC 2489; AIRIP, NLG R4-50; A-2m1, NLG R5-38; A-3m1, NLG R4-66; AIIRIP, NLG R4-16; a, FGSC 532; am1, FGSC 4565; mat, RLM 44-02; D25, mat A/mat a partial duplication strain PS D25 (see Table 1); per, perithecial RNA from a wild-type cross; VG, culture grown in vegetative medium (VOGEL 1964 ).

The analysis of sdv-4 transcripts is shown in Figure 6B. Similar to the case with sdv-1, sdv-4 transcripts were undetectable from RNA isolated from the NLG R4-50 (mat A-1^m92), indicating that a functional mat A-1 is required for sdv-4 expression. However, sdv-4 transcripts were observed in RLM 44-02 (matA) at a level approximately sevenfold higher than that observed in a wild-type mat A strain. Thus, as with sdv-1, either MAT A-2 or MAT A-3 (or both) must repress the expression of sdv-4 in the absence of MAT A-1. Transcript levels of sdv-4 were similar to those of a mat A strain in the NLG R5-39 (mat A-2^m1), NLG R4-66 (mat A-3^m1), and NLG R4-16 (mat A-2^m3 mat A-3^m3) strains, unlike what was observed in the matA strain. Expression levels of sdv-4 in FGSC 4565 (mat a-1^m1) were elevated as compared to a wild-type mat a, but they were similar to those of the RLM 44-02 (matA), indicating the presence of factors in these mutants that increase sdv-4 transcript levels in the absence of MAT a-1 or MAT A-1, MAT A-2, and MAT A-3. Interestingly, sdv-4 was expressed in RLM 44-02 (matA) grown in vegetative medium at levels similar to those found in a mat A strain (FGSC 2489) grown in crossing medium. In contrast, the expression of sdv-4 is completely suppressed in a wild-type mat A strain grown in vegetative medium (NELSON and METZENBERG 1992 ; our unpublished results) and was indeed a prerequisite for its isolation.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The matA strain described in this study was morphologically similar to the wild type in vegetative characteristics, but it did not mate as either a mat A or mat a strain, even though female and male reproductive structures were formed. In this respect, the N. crassa matA strain resembles mating type deletion mutants from other ascomycete species, such as P. anserina and Cochliobolus heterostrophus, i.e., mat strains are morphologically indistinguishable from the wild type during vegetative growth, but they are incapable of mating with strains of either mating type (COPPIN et al. 1993 ; WIRSEL et al. 1996 ). Fertility is restored to the mat mutants of P. anserina and C. heterostrophus by the introduction of ectopic copies of mating type sequences. In N. crassa, matA transformants containing either the a or A idiomorph in an ectopic position will mate as a female or a male, but they fail to produce ascospores (our unpublished results). The reason for the failure of ectopic copies to fully complement a mating type deletion in N. crassa is unknown. Apparently mating-type regulation of sporulation in N. crassa differs from that in P. anserina and C. heterostrophus.

As with the N. crassa matA mutant, the mat A-2^m1, mat A-3^m1, or mat A-2^m3 mat A-3^m3 mutants do not have an altered vegetative growth phenotype. All of the mat A mutants produce a large number of conidia and form normal protoperithecia when grown in crossing medium (nitrogen-limiting conditions). Unlike the matA mutant, however, the mat A-2, mat A-3, and mat A-2 mat A-3 mutants mate as a mat A strain and are vegetatively incompatible with mat a strains; these functions are attributable to mat A-1 (GLASS et al. 1990 ; STABEN and YANOFSKY 1990 ). The fact that the mat A-2^m1, mat A-3^m1, and mat A-2^m3 mat A-3^m3 mutants retain mat A-1 function, that all three contain a transcript for mat A-1, and that RIP of mat A-1 during the sexual cycle does not affect ascosporogenesis (GLASS and LEE 1992 ) indicates that mat A-1 is functional in all three mutants. The mat A-2^m1 and mat A-3^m1 mutants were shown by DNA sequence analysis to contain stop codons in the mat A-2 and mat A-3 ORFs, respectively, and transcripts of the mutated genes were undetectable in RT-PCR experiments. The absence of the transcripts could be caused by either lack of transcription or mRNA instability. In the S. cerevisiae phosphoglycerol kinase gene, a nonsense codon present within the first two thirds of the coding region led to mRNA instability and degradation, a phenomenon termed nonsense-codon-mediated mRNA degradation (PELTZ et al. 1994 ). The promoter region of mat A-2 and mat A-3 in the mat A-2^m3 mat A-3^m3 mutant showed mutations in putative transcriptional regulatory regions of both genes, and transcripts for mat A-2 and mat A-3 were absent in the mat A-2^m3 mat A-3^m3 mutant.

In contrast to the case with mat A-2 and mat A-3 single mutants, the mat A-2^m3 mat A-3^m3 mutant displays reduced fertility in crosses with mat a strains. Similar to mat A-2 or mat A-3 mutants, however, the few asci that are produced in crosses between mat A-2^m3 mat A-3^m3 and mat a segregate mating-type 1:1 (GLASS and LEE 1992 ) and form biparental asci. Our results are in contrast to those observed with P. anserina mutants in SMR1 (mat A-2 homolog) and SMR2 (mat A-3 homolog). Crosses between a mat+ strain and an SMR1 or SMR2 mutant result in the production of uniparental progeny of only SMR1 or SMR2 mutant genotype, respectively (ZICKLER et al. 1995 ). Interestingly, crosses between SMR1 mutants and an FPR1 mutant (mat a-1 homolog) result in a restoration of predominantly biparental asci, indicating that SMR1 is not the only factor involved in compartmentation of opposite mating-type nuclei in the crozier (ZICKLER et al. 1995 ). It is also possible that mat A-2 and mat A-3 are involved in the recognition of opposite mating-type nuclei during sexual development. Because the mat A-2^m1, mat A-3^m1, and mat A-2^m3 mat A-3^m3 mutants form only biparental asci in crosses to mat a, however, factors other than MAT A-2 and MAT A-3 must be sufficient for compartmentation of opposite mating-type nuclei in the crozier. The fact that the few progeny of crosses with the mat A-2^m3 mat A-3^m3 mutant segregate normally for mating-type indicates that it is not an occasional spore that can survive from one ascus, but rather, it is an occasional successful compartmentation of opposite mating-type nuclei into a crozier (GLASS and LEE 1992 ). An alternative hypothesis to the one above is that MAT A-2 and MAT A-3 are involved in orchestrating the multiple mitotic divisions of opposite mating-type nuclei in the ascogenous hyphae before the migration of opposite mating-type nuclei into the crozier (for review of asci development, see RAJU 1992 ). In the mat A-2^m3 mat A-3^m3 mutant, this process—and not compartmentation of opposite mating-type nuclei into the crozier—may be impeded.

Our results suggest that mat A-1 and mat a-1 are the critical factors for both mating and sexual development in N. crassa; mat A-2 and mat A-3 increase the efficiency of the process but are not essential for the production of ascospores. Mating-type genes homologous to mat A-1 and mat a-1 have been found in many filamentous ascomycetes (ARIE et al. 1997 ). mat A-2 and mat A-3 homologs, however, are absent or nonfunctional in other ascomycetes such as C. heterostrophus and related species, as well as in Neurospora terricola (TURGEON et al. 1993 , TURGEON et al. 1995 ; T. VELLANI and L. GLASS, unpublished results), suggesting that other means of increasing ascospore production have evolved in these fungi.

Direct target genes of the N. crassa, P. anserina, or C. heterostrophus mating-type polypeptides have not been reported. NELSON and METZENBERG 1992 isolated a number of putative target genes of mat A-1 by subtractive hybridization. Mutational analysis of two of the sdv genes (asd-1 and asd-2) resulted in recessive mutants that displayed blocks during sexual development. Mutational analysis of the remaining sdv genes by RIP did not result in mutants with a detectable vegetative or sexual phenotype (NELSON and METZENBERG 1992 ). Additional analyses showed that the expression patterns of some of the sdv genes are erratic (NELSON et al. 1997B ; this study). The two genes analyzed in this study, sdv-1 and sdv-4, however, are promising candidates as target genes for regulation by the mating-type polypeptides. Analyzing genes that are preferentially expressed during sexual development may also identify additional mating-type target genes. As part of the N. crassa genome project, several clones specific for conidial, mycelial, and sexual stages have been identified (NELSON et al. 1997A ). In addition, several genes from N. crassa that appear to be transcriptional regulators of vegetative and sexual development have been isolated recently (ARAMAYO et al. 1996 ; BALLARIO et al. 1996 ; YAMASHIRO et al. 1996 ; LINDEN and MACINO 1997 ). Expression studies using strains with mutations at these genes may also be of help in identifying additional regulators of sdv gene expression.

In the analysis of sdv gene expression in the N. crassa mat mutants, it was shown that both sdv-1 and sdv-4 required MAT A-1 for expression, but both sdv-1 and sdv-4 transcripts were observed in the matA strain, which lacks MAT A-1, MAT A-2, and MAT A-3. These data indicate that MAT A-2 and/or MAT A-3 (or the products they regulate) repress the expression of sdv-1 and sdv-4 unless MAT A-1 is also present. A simplistic model consistent with our data is that MAT A-1, MAT A-2, and MAT A-3 form a complex that regulates the expression of sdv-1 and sdv-4 in a mat A context under crossing conditions. In S. cerevisiae, direct protein-protein interactions of the mating-type polypeptides MAT2 and MATa1 have been described (DRANGINIS 1990 ); the a1/2 heterodimer signals that karyogamy has occurred and is involved in the regulation of entry into meiosis (GOUTTE and JOHNSON 1988 ). The mating type polypeptides of the P. anserina mat- idiomorph, FMR1 and SMR2, showed an interaction in assays using the yeast two-hybrid system (ZICKLER et al. 1995 ; R. DEBUCHY, personal communication). In N. crassa, interactions may also occur between polypeptides from the same mating type, i.e., MAT A-1, MAT A-2, and MAT A-3, and may be important in maintaining opposite mating-type nuclear identity before karyogamy.

	FOOTNOTES

¹ Current address: Departamento de Microbiologia, Universidade Federal de Minas Gerais, Belo Horizonte, MG, 31270-910, Brazil.
² Current address: Millennium Pharmaceuticals, Inc., Cambridge, MA, 02139.
³ Department of Biological Sciences, Stanford University, Stanford, CA, 94305-5020.
⁴ The designation for the mating-type locus for Neurospora crassa has recently been changed from mt to mat, in accordance with the designation of other ascomycete mating type loci (GLASS and STABEN 1997 ).

	ACKNOWLEDGMENTS

This work was supported by grants from the Human Frontiers of Science Program and National Science and Engineering Research Council to N.L.G. and by United States Public Health Service grant GM08995 to R.L.M. We thank Dr. NAMBOORI RAJU for cytological observations and advice, Dr. SVEN SAUPE for plasmid pAL2, JEFF HALL for plasmid pGq1310 and for help with some transformations, Dr. BOB ULLRICH for advice on the ADE5 gene of S. commune, MARGARET ALIC and Dr. MIKE GOLD for furnishing the plasmid p2ADE4, and Drs. TOM RANDALL and RODOLFO ARAMAYO for both hands-on help and useful discussion. We also thank Dr. RICHARD TODD for critically editing the manuscript. A.V.-B.F. was supported during part of this work by a scholarship from Coordenadoria de Aperfeiçoamento de Pessoal de Nível Superior, Ministry of Education, Brazil.

Manuscript received August 21, 1997; Accepted for publication December 1, 1997.

	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].

ALIC, M., E. K. CLARK, J. R. KORNEGAY, and M. H. GOLD, 1990  Transformation of Phanerochaete chrysosporium and Neurospora crassa with adenine biosynthetic genes from Schizophyllum commune.. Curr. Genet. 17:305-311.

ARAMAYO, R., Y. PELEG, R. ADDISON, and R. METZENBERG, 1996  Asm-1⁺, a Neurospora crassa gene related to transcriptional regulators of fungal development. Genetics 144:991-1003[Abstract].

ARIE, T., S. K. CHRISTIANSEN, O. C. YODER, and B. G. TURGEON, 1997  Efficient cloning of ascomycete mating type genes by PCR amplification of the conserved MAT HMG box. Fungal Genet. Biol. 21:118-130[Medline].

BALLARIO, P., P. VITTORIOSO, A. MAGRELLI, C. TALORA, and A. CABIBBO et al., 1996  White collar-1, a central regulator of blue light responses in Neurospora, is a zinc finger protein. EMBO J. 15:1650-1657[Medline].

CAMBARERI, E. B., B. C. JENSEN, E. SCHABTACH, and E. U. SELKER, 1989  Repeat-induced G-C to A-T mutations in Neurospora. Science 244:1571-1575[Abstract/Free Full Text].

CHANG, S. and C. STABEN, 1994  Directed replacement of mt a-1 effects a mating-type switch in Neurospora crassa.. Genetics 138:75-81[Abstract].

COPPIN, E., S. ARNAISE, V. CONTAMINE, and M. PICARD, 1993  Deletion of the mating-type sequences in Podospora anserina abolishes mating without affecting vegetative functions and sexual differentiation. Mol. Gen. Genet. 241:409-414[Medline].

DAVIS, R. H. and F. J. DESERRES, 1970  Genetic and microbial research techniques for Neurospora crassa.. Methods Enzymol. 17A:79-143.

DEBUCHY, R., S. ARNAISE, and G. LECELLIER, 1993  The mat- allele of Podospora anserina contains three regulatory genes required for the development of fertilized female organs. Mol. Gen. Genet. 241:667-673[Medline].

DRANGINIS, A. M., 1990  Binding of yeast a1 and 2 as a heterodimer to the operator DNA of a haploid-specific gene. Nature 347:682-685[Medline].

FERREIRA, A. V. B., S. SAUPE, and N. L. GLASS, 1996  Transcriptional analysis of the mt A idiomorph of Neurospora crassa identifies two genes in addition to mt A-1.. Mol. Gen. Genet. 250:767-774[Medline].

GLASS, N. L. and L. LEE, 1992  Isolation of Neurospora crassa A mating type mutants by repeat induced point (RIP) mutation. Genetics 132:125-133[Abstract].

GLASS, N. L., and M. A. NELSON, 1994 Mating-type genes in mycelial ascomycetes, pp. 295–306 in The Mycota I: Growth, Differentiation and Sexuality, edited by J. D. H. WESSELS and F. MEINHARDT. Springer-Verlag, Berlin.

GLASS, N. L. and C. STABEN, 1997  Neurospora mating-type symbol mt revised to mat.. Fungal Genet. Newsl. 44:64.

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].

GOUTTE, C. and A. D. JOHNSON, 1988  a1 protein alters the DNA binding specificity of 2 repressor. Cell 52:875-882[Medline].

GRAYBURN, W. S. and E. U. SELKER, 1989  A natural case of RIP: degeneration of DNA sequence in an ancestral tandem duplication. Mol. Cell Biol. 9:4416-4421[Abstract/Free Full Text].

GRIFFITHS, A. J. F., 1982  Null mutants of the A and a mating-type alleles of Neurospora crassa.. Can. J. Genet. Cytol. 24:167-176.

GRIFFITHS, A. J. F. and A. M. DELANGE, 1978  Mutations of the a mating type in Neurospora crassa.. Genetics 88:239-254[Abstract/Free Full Text].

HERSKOWITZ, I., 1989  A regulatory hierarchy for cell specialization in yeast. Nature 342:749-757[Medline].

JOHNSON, A. D., 1995  Molecular mechanisms of cell-type determination in budding yeast. Curr. Opin. Genet. Dev. 5:552-558[Medline].

KOTHE, E., 1996  Tetrapolar fungal mating types: sex by the thousands. FEMS Microbiol. Rev. 18:65-87[Medline].

KREADER, C. A. and J. A. HECKMAN, 1987  Isolation and characterization of a Neurospora crassa ribosomal gene homologous to cyh-2 of yeast. Nucleic Acids Res. 15:9027-9041[Abstract/Free Full Text].

LINDEN, H. and G. MACINO, 1997  White collar-2, a partner in blue-light signal transduction, controlling expression of light-regulated genes in Neurospora crassa.. EMBO J. 16:98-109[Medline].

METZENBERG, R. L. and N. L. GLASS, 1990  Mating type and mating strategies in Neurospora.. BioEssays 12:53-59[Medline].

MITCHELL, M. B., 1966  A round spore character in Neurospora crassa.. Neurospora Newsl. 10:6.

NASMYTH, K. A. and D. SHORE, 1987  Transcriptional control of yeast life cycle. Science 237:1102-1170[Free Full Text].

NELSON, M. A., S. KANG, E. BRAUN, M. CRAWFORD, and P. L. DOLAN et al., 1997a  E xpressed sequences from conidial, mycelial and sexual stages of Neurospora crassa.. Fungal Genet. Biol. 21:348-363[Medline].

NELSON, M. A., S. T. MERINO, and R. L. METZENBERG, 1997b  A putative rhamnogalacturonase required for sexual development of Neurospora crassa.. Genetics 146:531-540[Abstract].

NELSON, M. A. and R. L. METZENBERG, 1992  Sexual development genes of Neurospora crassa.. Genetics 132:149-162[Abstract].

ORBACH, M. J., E. B. PORRO, and C. YANOFSKY, 1986  Cloning and characterization of the gene for ß-tubulin from a benomyl-resistant mutant of Neurospora crassa and its use as a dominant selectable marker. Mol. Cell. Biol. 6:2452-2461[Abstract/Free Full Text].

PELTZ, S. W., F. HE, E. WELCH, and A. JACOBSON, 1994  Nonsense-mediated mRNA decay in yeast. Prog. Nucleic Acid Res. Mol. Biol. 47:271-325[Medline].

PERKINS, D. D. and E. G. BARRY, 1977  The cytogenetics of Neurospora.. Adv. Genet. 19:133-285[Medline].

PHILLEY, M. L. and C. STABEN, 1994  Functional analyses of the Neurospora crassa MT a-1 mating type polypeptide. Genetics 137:715[Abstract].

RAEDER, U. and P. BRODA, 1985  Rapid preparation of DNA from filamentous fungi. Lett. Appl. Microbiol. 1:17-20.

RAJU, N. B., 1992  Genetic control of the sexual cycle in Neurospora.. Mycol. Res. 96:240.

RANDALL, T. A. and R. L. METZENBERG, 1995  Species-specific and mating-type specific DNA regions adjacent to the mating type idiomorphs in the genus Neurospora. Genetics 141:119-136[Abstract].

SAUPE, S., L. STENBERG, K.-T. SHIU, and N. L. GLASS, 1996  The molecular nature of mutations in the mt A-1 gene of the Neurospora crassa A idiomorph and their relation to mating-type function. Mol. Gen. Genet. 250:115-122[Medline].

SELKER, E. U., 1990  Premeiotic instability of repeated sequences in Neurospora crassa.. Annu. Rev. Genet. 24:579-613[Medline].

SELKER, E. U. and P. W. GARRETT, 1988  DNA sequence duplications trigger gene inactivation in Neurospora crassa.. Proc. Natl. Acad. Sci. USA 85:6870-6874[Abstract/Free Full Text].

SMITH, M. L., C. J. YANG, R. L. METZENBERG, and N. L. GLASS, 1996  Escape from het-6 incompatibility in Neurospora crassa partial diploids involves preferential deletion within the ectopic segment. Genetics 144:523-531[Abstract].

SPRAGUE, G. F. J., R. JENSEN, and I. HERSKOWITZ, 1983  Control of yeast cell type by the mating type locus: positive regulation of the a-specific STE3 gene by the MAT1 product. Cell 32:409-415[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].

TURGEON, B. G., H. BOHLMANN, L. M. CIUFFETTI, S. K. CHRISTIANSEN, and G. YANG et al., 1993  Cloning and analysis of the mating type genes from Cochliobolus heterostrophus.. Mol. Gen. Genet. 238:270-284[Medline].

TURGEON, B. G., A. SHARON, S. WIRSEL, K. YAMAGUCHI, and S. K. CHRISTIANSEN et al., 1995  Structure and function of mating type genes in Cochliobolus spp. and asexual fungi. Can. J. Bot. 73:S778-S783.

VOGEL, H. J., 1964  Distribution of lysine pathways among fungi: evolutionary implications. Am. Nat. 98:435-446.

VOLLMER, S. J. and C. YANOFSKY, 1986  Efficient cloning of genes of Neurospora crassa.. Proc. Natl. Acad. Sci. USA 34:573-577.

WESTERGAARD, M. and H. K. MITCHELL, 1947  Neurospora V. A synthetic medium favoring sexual reproduction. Amer. J. Bot. 34:573-577.

WIRSEL, S., B. G. TURGEON, and O. C. YODER, 1996  Deletion of the Cochliobolus heterostrophus mating-type (MAT ) locus promotes the function of MAT transgenes. Curr. Genet. 29:241-249[Medline].

YAMASHIRO, C. T., D. J. EBBOLE, B.-U. LEE, R. E. BROWN, and C. BOULAND et al., 1996  Characterization of rco-1 of Neurospora crassa, a pleiotropic gene affecting growth and development that encodes a homolog of Tup1 of Saccharomyces cerevisiae.. Mol. Cell. Biol. 16:6218-6228[Abstract/Free Full Text].

ZICKLER, D., S. ARNAISE, and M. PICARD, 1995  Altered mating-type identity in the fungus Podospora anserina leads to selfish nuclei, uniparental progeny, and haploid meiosis. Genetics 140:493-503[Abstract].

This article has been cited by other articles:

				E. U. Selker Robert L. Metzenberg, June 11, 1930-July 15, 2007: Geneticist Extraordinaire and "Model Human" Genetics, February 1, 2008; 178(2): 611 - 619. [Full Text] [PDF]

				K.A. Aronstein, K.D. Murray, J.H. de Leon, X. Qin, and G.M. Weinstock High mobility group (HMG-box) genes in the honeybee fungal pathogen Ascosphaera apis Mycologia, July 1, 2007; 99(4): 553 - 561. [Abstract] [Full Text] [PDF]

				H. V. Colot, G. Park, G. E. Turner, C. Ringelberg, C. M. Crew, L. Litvinkova, R. L. Weiss, K. A. Borkovich, and J. C. Dunlap A high-throughput gene knockout procedure for Neurospora reveals functions for multiple transcription factors PNAS, July 5, 2006; 103(27): 10352 - 10357. [Abstract] [Full Text] [PDF]

				N. Nolting and S. Poggeler A MADS Box Protein Interacts with a Mating-Type Protein and Is Required for Fruiting Body Development in the Homothallic Ascomycete Sordaria macrospora Eukaryot. Cell, July 1, 2006; 5(7): 1043 - 1056. [Abstract] [Full Text] [PDF]

				S. L. Panwar, M. Legrand, D. Dignard, M. Whiteway, and Paul. T. Magee MF{alpha}1, the Gene Encoding the {alpha} Mating Pheromone of Candida albicans Eukaryot. Cell, December 1, 2003; 2(6): 1350 - 1360. [Abstract] [Full Text] [PDF]

				P. K. T. Shiu and R. L. Metzenberg Meiotic Silencing by Unpaired DNA: Properties, Regulation and Suppression Genetics, August 1, 2002; 161(4): 1483 - 1495. [Abstract] [Full Text] [PDF]

				S. Arnaise, D. Zickler, S. Le Bilcot, C. Poisier, and R. Debuchy Mutations in Mating-Type Genes of the Heterothallic Fungus Podospora anserina Lead to Self-Fertility Genetics, October 1, 2001; 159(2): 545 - 556. [Abstract] [Full Text] [PDF]

				M. Duarte and A. Videira Respiratory Chain Complex I Is Essential for Sexual Development in Neurospora and Binding of Iron Sulfur Clusters Are Required for Enzyme Assembly Genetics, October 1, 2000; 156(2): 607 - 615. [Abstract] [Full Text]

				S. J. Saupe Molecular Genetics of Heterokaryon Incompatibility in Filamentous Ascomycetes Microbiol. Mol. Biol. Rev., September 1, 2000; 64(3): 489 - 502. [Abstract] [Full Text] [PDF]

				P. K. T. Shiu and N. L. Glass Molecular Characterization of tol, a Mediator of Mating-Type-Associated Vegetative Incompatibility in Neurospora crassa Genetics, February 1, 1999; 151(2): 545 - 555. [Abstract] [Full Text]