Mutational Analysis of the [Het-s] Prion Analog of Podospora anserina: A Short N-Terminal Peptide Allows Prion Propagation

Corresponding author: Sven J. Saupe, Laboratoire de Génétique Moléculaire des Champignons, Institut de Biochimie et de Génétique Cellulaire, Centre National de la Recherche Scientifique, Unité propre de recherche 9026, 1 rue Camille St. Saëns, 33077 Bordeaux Cedex, France., sven.saupe{at}ibgc.u-bordeaux2.fr (E-mail)

Communicating editor: R. H. DAVIS

	ABSTRACT

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

The het-s locus is one of nine known het (heterokaryon incompatibility) loci of the fungus Podospora anserina. This locus exists as two wild-type alleles, het-s and het-S, which encode 289 amino acid proteins differing at 13 amino acid positions. The het-s and het-S alleles are incompatible as their coexpression in the same cytoplasm causes a characteristic cell death reaction. We have proposed that the HET-s protein is a prion analog. Strains of the het-s genotype exist in two phenotypic states, the neutral [Het-s*] and the active [Het-s] phenotype. The [Het-s] phenotype is infectious and is transmitted to [Het-s*] strains through cytoplasmic contact. het-s and het-S were associated in a single haploid nucleus to generate a self-incompatible strain that displays a restricted and abnormal growth. In the present article we report the molecular characterization of a collection of mutants that restore the ability of this self-incompatible strain to grow. We also describe the functional analysis of a series of deletion constructs and site-directed mutants. Together, these analyses define positions critical for reactivity and allele specificity. We show that a 112-amino-acid-long N-terminal peptide of HET-s retains [Het-s] activity. Moreover, expression of a mutant het-s allele truncated at position 26 is sufficient to allow propagation of the [Het-s] prion analog.

A prion can be defined as an infectious protein: an altered form of a normal cellular protein capable of converting the normal form into the altered one (PRUSINER 1998 ). The prion concept was proposed to explain the properties of the infectious agent causing spongiform encephalopathies, a class of neurodegenerative diseases in mammals. Numerous data support the "protein only" hypothesis that states that transmission of these fatal diseases is caused by an aberrantly folded form of the prion protein (Prp), but that this process does not require an associated nucleic acid (for a review see HORWICH and WEISSMAN 1997 ).

With the development of the prion concept, the properties of two non-Mendelian elements in yeast were reevaluated in light of this "protein only" theory. It was proposed that the [URE3] and [PSI+] genetic elements are yeast prions (WICKNER 1994 ). Presence of the [URE3] element mimics mutations in the chromosomal URE2 gene. [URE3] strains, like ure2 mutants, can use ureidosuccinate in the presence of ammonium. But unlike classical ure2 mutations, [URE3] shows a non-Mendelian inheritance. [URE3] strains can be reversibly cured, transmission of [URE3] requires the presence of a functional chromosomal copy of the URE2 gene, and overexpression of the URE2 gene increases the frequency of de novo [URE3] appearance. By analogy with the prion model, these features have led to the hypothesis that the Ure2p protein can adopt an altered and transmissible form in [URE3] strains. Similarly, the [PSI+] element, which causes an increased suppression of nonsense mutations, was proposed to correspond to the prion form of the translation termination factor Sup35 (WICKNER 1994 ). The similarity between these yeast proteins and mammalian prions was futher reinforced by biochemical analyses. It was shown that the Sup35p protein aggregates in [PSI+] strains in vivo (PATINO et al. 1996 ; PAUSHKIN et al. 1996 ) and that [PSI+] propagation is controlled by the Hsp104 chaperone (CHERNOFF et al. 1995 ). The prion-like conversion of purified Sup35p protein can be obtained in vitro and is catalyzed by extracts from [PSI+] strains (PAUSHKIN et al. 1997 ). Finally, both Sup35p and ure2p proteins were shown to aggregate in amyloids in vitro (GLOVER et al. 1997 ; KING et al. 1997 ; TAYLOR et al. 1999 ; THUAL et al. 1999 ). These results show that the infectious propagation of a protein conformation could have broad biological implications (for review see LINDQUIST 1997 ; KUSHNIROV and TER-AVANESYAN 1998 ). We have previously proposed that the HET-s protein encoded by the het-s heterokaryon incompatibility locus of Podospora anserina is a prion protein fitting the definition of an infectious protein (COUSTOU et al. 1997 ).

The het-s locus is one of nine het (heterokaryon incompatibility) loci known to restrict heterokaryon formation in the filamentous ascomycetes Podospora anserina (BEGUERET et al. 1994 ). A het locus is defined as a locus in which heteroallelism cannot be tolerated in a heterokaryon. Similar to other het loci, two alternate antagonistic wild-type alleles, het-s and het-S, exist at the het-s locus. Strains bearing the alternate het-s and het-S alleles are incompatible; i.e., coexpression of het-s and het-S in the same cell triggers a lytic reaction leading to cell death. Confrontation of het-s and het-S strains on solid medium results in the accumulation of dead cells between the strains and leads to the formation of an abnormal contact, visible at the macroscopic level and named barrage (RIZET 1952 ). The particularity of the het-s locus lies in the fact that strains bearing the het-s allele can display two alternate phenotypes: the reactive [Het-s] phenotype, ([Het-s] strains are incompatible with [Het-S] strains) and the neutral [Het-s*] phenotype ([Het-s*] are compatible with both [Het-S] and [Het-s] strains). Strains can switch from the neutral [Het-s*] to the reactive [Het-s] phenotype either spontaneously at a very low frequency or invariably after a cytoplasmic contact with a [Het-s] strain. Because filamentous fungi spontaneously produce hyphal anastomoses (cell fusions), this cytoplasmic contact can readily be obtained by confrontation of the strains on solid medium.

Overexpression of the HET-s protein strongly increases the frequency of the [Het-s*] to [Het-s] transition (COUSTOU et al. 1997 ). het-s cells can be reversibly cured of the [Het-s] phenotype when protoplasts are prepared from a [Het-s] mycelium (BELCOUR 1976 ). Molecular analyses have shown that the protein encoded by het-s is present in [Het-s*] strains and that therefore the neutral [Het-s*] phenotype does not result from the lack of expression of the het-s gene but from a post-translational modification of the protein (COUSTOU et al. 1997 ). This modification does not alter the electrophoretic mobility of the protein in SDS-PAGE. We proposed, by analogy with the prion model, that HET-s* and HET-s are two conformers of the same protein. In this model, the het-s allele leads to the expression of the nonreactive HET-s* form, which can be converted into the reactive HET-s form through physical interactions.

Little is known of the mechanism that leads to cell death as a consequence of the coexpression of het-s and het-S. The incompatibility reaction is thought to be induced by the formation of poison complexes formed between the proteins encoded by the het-s and het-S alleles. It is not known whether het-s possesses a cellular function in addition to its role in triggering incompatibility-associated cell death. The het-s sequence does not display any significant similarity with known proteins, and the gene is not essential. The HET-s and HET-S proteins are both 289 amino acids in length and differ by 13 residues (TURCQ et al. 1991 ). Two of these polymorphic positions located in the N terminus of the protein were found to be crucial for specificity (DELEU et al. 1993 ). The single histidine to proline substitution at position 33 in HET-S leads to a strain with all phenotypic characteristics of [Het-s] strains: the mutant strain is compatible with a [Het-s] strain and incompatible with a [Het-S] strain. Therefore a single amino acid difference is sufficient to trigger incompatibility. To obtain the reciprocal specificity switch, two substitutions are required (D23A and P33H). As the reactivity of the [Het-S] allele is not associated with the existence of a metastable nonreactive state, these amino acid substitutions affect both specificity and the prion properties of HET-s.

We decided to use a genetic approach to try to identify other regions or positions in the sequence that are required for the reactivity and the prion-like metastable properties of the proteins encoded at the het-s locus. The two antagonistic alleles were brought together in the same haploid nucleus by transformation. Strains that coexpress the two antagonistic het-s alleles display very restricted growth and are termed self-incompatible. We used such strains to screen for intra- or extragenic suppressor mutations that restore normal or close to normal growth rates (Figure 1). Characterization of extragenic suppressors will be described elsewhere. Herein we describe molecular and functional characterization of a collection of 17 mutants in het-s or het-S that escaped from self-incompatibility. As a complementary approach, deletion mutants of het-s were constructed and tested to determine if a minimal prion domain could be defined. Finally, because of the crucial role of the amino acid at position 33, systematic amino acid replacements were produced at this position.

View larger version (67K): In this window In a new window Download PPT slide   Figure 1. Selection of mutant sectors escaping self-incompatibility. (A) Compares the growth phenotype of a wild-type strain and of a het-s/het-S self-incompatible strain submitted to UV mutagenesis. The strains were grown for 4 days on cornmeal medium. Note the emergence of three mutant sectors recovering wild-type growth. (B) The strategy of selection and characterization of the mutant sectors is depicted. The self-incompatible strain contains the het-S and the overexpressed pGPD-het-s alleles. After UV mutagenesis, emergence of sectors recovering normal growth occurs. This can either result from inactivation of one of the incompatible het-S or het-s alleles (in the example shown here, het-S is inactivated) or can be due to a mutation in an extragenic suppressor noted A. To characterize the mutational event, first the activity of the het-s and het-S alleles is tested in a confrontation test (barrage), next extragenic and intragenic suppressors are distinguished by genetic analyses, and finally mutant het-s and het-S alleles are cloned and sequenced after PCR amplification.

	MATERIALS AND METHODS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Strains and culture conditions:
P. anserina is a filamentous ascomycete. The mycelium has a coenocytic structure; cells are multinucleate and have incomplete cell septa. This organization provides an original situation compared to other organisms in which prion proteins were described. The transition of HET-s* to HET-s occurring at one point will propagate to all the mycelium and will not be limited to a single cell as in yeast. This amplification of the transition may favor the detection of rare events.

The life cycle and methods for genetic analysis have been described (ESSER 1974 ). Vegetative compatibility between strains can easily be determined by confrontation on solid cornmeal agar medium. Incompatibility results in the formation of a dense line or barrage in the region of contact between strains. All strains used in this study are isogenic except for the het-s and the mating-type loci. Three wild-type alleles have been identified at the het-s locus, the two reactive and mutually incompatible alleles het-s and het-S and a neutral allele het-s^x in which a 46-bp tandem duplication leads to the truncation of the het-s open reading frame (ORF) at position 183 (DELEU et al. 1993 ). The DNA sequences of all three alleles have been reported (TURCQ et al. 1991 ; DELEU et al. 1993 ). Finally, het-s°, a strain in which the het-s locus was inactivated by gene replacement, was also used (TURCQ et al. 1991 ). Neutral [Het-s*] strains were obtained in the offspring of crosses between het-s and het-S strains in which the het-s strain was used as the male parent. The ability to trigger the [Het-s*] to [Het-s] transition (converting activity) is assayed by confronting the strain to be tested with a [Het-s*] strain on solid medium. An implant from the [Het-s*] tester is then sampled and reinoculated on solid medium to test if it has gained the ability to produce the barrage reaction with the [Het-S] tester strain.

P. anserina protoplast formation and transformation were carried out as described previously (BERGES and BARREAU 1989 ).

UV mutagenesis:
Pieces of mycelium of the self-incompatible strain were inoculated on cellophane discs laid on cornmeal agar medium and grown for 72 hr at 18°. Mycelia were then irradiated for 20 sec with a UV light source (250–310 nm) located at 20 cm. Plates were incubated 24 hr at 18° in the dark, then individual cellophane discs were transferred to fresh cornmeal agar medium and incubated for 2–4 days at 18° to allow the growth of sectors of mycelium that display a growth rate higher than that of the parental strain. Such sectors were observed on about one-third of the individual cellophane discs.

DNA constructs:
The pGPD-het-s vector for overexpression of het-s was constructed by inserting the 1.2-kb NcoI-XbaI fragment containing the het-s ORF downstream of the G3PD promotor of Aspergillus nidulans in the pAN8.1 vector (PUNT et al. 1988 ).

The p1004-s vector was constructed by inserting the 2.6-kb HindIII-KpnI het-s fragment into the pCB1004 vector (CARROL et al. 1994 ). The p1004-s(1-253) vector was constructed by inserting a 1.8-kb HindIII-BamHI containing the promotor and the 253 first codons of the het-s ORF into the pCB1004 vector. The p1004-GPD-s(1-253) was constructed by inserting a 2.5-kb PstI-BamHI fragment from pGPD-het-s containing the G3PD promotor and 253 first codons of the het-s ORF into pCB1004. The p1004-s(1-229) vector was constructed by coligating a HindIII-EcoRV fragment containing the het-s promotor and the 229 first codons of the het-s ORF, and a VspI-KpnI fragment containing the het-s terminator and the TAA stop codon and blunted at the VspI site into the pCB1004 vector. The p1004-s(82-121) vector was constructed by deleting the 114-bp PvuI fragment from the het-s ORF from p1004-s. The p1004-s(30-289) vector was constructed by coligating a PvuII-KpnI fragment containing the 259 last codons of the het-s ORF and a HindIII-NcoI fragment containing the het-s promotor and initiator ATG and blunted at the NcoI site into the pCB1004 vector. The p1004-s(1-112) vector was constructed by coligating a EcoRI-NcoI fragment from pGPD-het-s containing the G3PD promotor and a NcoI-HindIII fragment obtained by restriction on a PCR fragment achieved by amplification on the pGPD-het-s vector with a primer located in the G3PD promotor (5'TCCATACTCCATCCTTCCCATCC3') and a primer introducing a stop codon at codon position 113 (5'GGCAAGCTTACACCACCAGATCCTGCTGG3') into the pBC1004 vector. The 1004-s(86-289) vector was constructed by coligating a EcoRI-NcoI fragment from pGPD-het-s containing the G3PD promotor and a NcoI-HindIII fragment obtained by restriction of a PCR fragment amplified from the pGPD-het-s vector using a primer introducing an initiator ATG codon at the codon position 85 (5'TCGCCATGGAAATCTTGCTTCTCTTCG3') and a primer located downstream of the STOP codon (5'CCTGGAAAGAAGCTTGATGCCTTTC3'). The p1004-s(1-25) vector was constructed by coligating a HindIII-NcoI fragment containing the het-s promotor and a NcoI-BglII fragment obtained by restriction on a PCR fragment amplified from the p1004-s vector with a primer located in the het-s promotor (5'GCCGCACAGGAACATCAAGCTTCGC3') and a primer introducing a STOP codon at position 26 (5'GGAGATCTAAAAACAGTCAACGCAG3') into the pCB1004.

PCR and DNA sequencing:
Mutant het-s or het-S alleles were amplified by PCR. Genomic DNA was prepared as described (LECELLIER and SILAR 1994 ). The het-s and het-S ORFs were amplifed as a single PCR fragment containing the entire ORF. Primers were 5' GCCGCACAGGAACATCAAGCTTCGC 3' and 5' CCTGGAAAGAAGCTTGATGCCTTTC 3' (for het-S) and 5' TCCATACTCCATCCTTCCCATCC 3' and 5' CCTGGAAAGAAGCTTGATGCCTTTC 3' (for het-s). PCR fragments were cloned in the pBluescript SK vector (Stratagene, La Jolla, CA) and sequenced at Genome Express (Grenoble, France).

Protein extraction and detection:
Proteins were extracted from protoplasts by osmotic lysis in 50 mM Na₂HPO₄ pH 8 at a concentration of 2.10⁸ cells per ml. Cellular debris was removed from protein extracts by centrifugation for 5 min at 5000 x g. Protein separation was performed by SDS-PAGE on a Biorad MiniProtean II apparatus according to instructions provided by the manufacturer. Immunoblotting was carried out as previously described (DELEU et al. 1993 ).

	RESULTS

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

Construction of a self-incompatible het-s het-S strain—effect of expression level:
Strains that contain both het-s and het-S alleles have been obtained by transformation. When het-s was introduced into a het-S strain, the transformants displayed a sublethal phenotype. The restricted and abnormal growth is the result of the coexpression of the two incompatible genes. This phenotype was found to be very unstable. Sectors recovering normal growth emerged spontaneously at a high frequency. Barrage tests revealed that these strains displayed only the [Het-S] phenotype (barrage with a [Het-s] strain but not with a het-S strain). However, the presence of a functional het-s allele could be demonstrated by analyzing crosses between the sectors and the het-s^x strain containing the inactive allele. Both het-s and het-S strains were recovered in the progeny from these crosses. So emergence of these sectors was not due to a mutation in het-s but likely resulted from the expression of the [Het-s*] phenotype, this phenotype being strongly favored in a [Het-S] context.

It has been shown that overexpression of het-s strongly increases the frequency of spontaneous [Het-s*] to [Het-s] transition (COUSTOU et al. 1997 ). Therefore, we have examined the effect of the overexpression of het-s in strains that contain the het-S gene. The het-s° strain was transformed with the pGPD-het-s vector containing the het-s ORF downstream of the A. nidulans G3PD promotor (COUSTOU et al. 1997 ). A transformant (pGPD-het-s) containing a single ectopic copy of the transgene was recovered after Southern blot analysis and it was verified by Western blotting that this strain did overexpress the HET-s protein. This pGPD-het-s strain was then crossed with the het-S strain. One-fourth of the strains in the progeny from this cross displayed a sublethal phenotype (Figure 1). These strains display a restricted radial growth, and they are characterized by a ragged colony edge and the absence of aerial filaments. When these sublethal strains were crossed with the neutral het-s^x strain both [Het-s] and [Het-S] progeny were recovered, indicating that these sublethal strains contained the two antagonistic genes: the resident het-S allele and the ectopic overexpressed het-s allele. The sublethal phenotype of this het-S pGPD-het-s strain is much more stable than that for the het-S het-s strain.

To test the effect of the simultaneous overexpression of het-s and het-S, the pGPD-het-s transformant was crossed with a strain in which the het-S allele is also overexpressed and located in an ectopic position (pGPD-het-S). One-fourth of the progeny from this cross correspond to strains that overexpress both het-s and het-S. Surprisingly, these strains did not display the abnormal phenotype of self-incompatible strains but they produced a barrage reaction in confrontations with both the [Het-s] and the [Het-S] tester. Therefore, the het-s and het-S alleles are reactive in these strains. The fact that the strains in which both het-s and het-S are overexpressed are not self-incompatible suggests that the relative amounts of the HET-s and HET-S proteins might be critical in inducing self-incompatibility.

The het-S pGPD-het-s strain was chosen to screen for mutations that allow the strain to escape self-incompatibility. Stability of its phenotype was tested at different temperatures ranging from 18° to 32° and on different growth media. Spontaneous emergence of sectors resulting from expression of the [Het-s*] phenotype was found to be minimal at 18° on cornmeal agar medium.

Selection and genetic characterization of mutant sectors:
A total of 500 self-incompatible mycelia were UV-mutagenized and incubated at 18° to allow emergence of sectors that grow better than the parental strain. Forty-eight sectors that displayed an increased growth rate were recovered (Figure 1). To distinguish sectors corresponding to mutations from those resulting from expression of the neutral [Het-s*] state, the sectors were confronted with a [Het-s] strain to trigger the [Het-s*] to [Het-s] transition. For 4 sectors, the contact with the [Het-s] strain induced the recovery of the sublethal phenotype, suggesting that they were not actual mutants but merely corresponded to strains expressing the [Het-s*] phenotype. Therefore, these sectors were not analyzed further. The 44 remaining sectors were tested for the barrage reaction with the [Het-s] and [Het-S] strains to determine if the mutant retained the [Het-s] or the [Het-S] phenotype. All the sectors were then crossed to the neutral het-s^x strain to determine if the mutation affected het-S, pGPD-het-s, or an extragenic suppressor. If the escape from self-incompatibility is due to an extragenic suppressor, then the pGPD-het-s and het-S alleles are intact and therefore self-incompatible offspring will be recovered in the progeny. They will represent one-eighth of the offspring if the suppressor gene is not linked to het-S or pGPD-het-s.

Alternatively, if the escape is due to inactivation of either the ectopic pGPD-het-s or the resident het-S allele, no self-incompatible progeny will be recovered from this cross. If het-S is the affected allele and pGPD-het-s is still active, one-half of the progeny will produce a barrage reaction with a [Het-S] strain and vice versa; if pGPD-het-s is inactivated and het-S is intact, one-half of the progeny will produce a barrage with a [Het-s] strain. Of 44 analyzed sectors, 27 gave self-incompatible offsprings and were considered to result from mutations in extragenic suppressors. Genetic and molecular characterization of these suppressors will be reported elsewhere. Based on this genetic analysis, among the 17 remaining sectors, 14 were affected for het-S and the other 3 for pGPD-het-s. Strains that contained these mutations in a compatible genetic background (in the absence of the antagonistic wild-type allele) were isolated. The reactivity of strains containing these different mutations was assayed in barrage tests and their capacity to induce the [Het-s*] to [Het-s] transition (converting activity) was also tested. To identify the nature and the position of the mutation, the mutant alleles were amplified by PCR, subcloned, and sequenced.

Description of the 17 selected mutants:
Figure 2 gives the nature and position of the mutations found in pGPD-het-s or het-S. Seven mutations introduced premature stop codons in the ORF. The remaining 12 corresponded to single amino acid substitutions. The 17 mutants were classified according to their phenotype.

View larger version (57K): In this window In a new window Download PPT slide   Figure 2. Seventeen het-s and het-S mutants classified according to their phenotype. At the top (wild type), the products of het-s and het-S ORFs are shown. Position and nature of the 13 polymorphic amino acid positions in which HET-s and HET-S differ are given. Protein products of mutant het-s and het-S alleles are grouped by phenotypic class (see text). The nature and position of the amino acid substitution are given. Premature stop codons are noted by arrowheads.

Null mutants: Nine mutants behaved as null mutants; they are not reactive in incompatibility as judged on barrage tests and cannot induce the [Het-s*] to [Het-s] transition. All but one were found to be affected in the het-S allele.

Four mutations resulted in the appearance of a premature stop codon at positions 44, 55, and 79, respectively. The het-S L79* mutation was found in two independent sectors. The severely truncated polypeptides that would be expressed from these mutant alleles were not detected on Western blots (data not shown), suggesting that these products are unstable and rapidly targeted for degradation.

The five remaining mutants of this class corresponded to single amino acid replacements. These replacements were distributed over nearly the entire sequence from position 43 to 266. In four of these mutant strains, the protein was undetectable on Western blots (het-S L89P, het-s T266P, het-S L43P, and het-S P120S), suggesting that the altered proteins are also unstable and rapidly degraded. It is noteworthy that all of these substitutions involve proline residues. The proline residue has a very low -helix propensity, especially when located in the middle of helices (BLABER et al. 1993 ). Residues L43 and L89 are predicted to be part of -helices (predator algorithm at IBCP; FRISHMAN and ARGOS 1996 ). The replacement of the leucine residue (an amino acid with a high helix propensity; PACE and SCHOLTZ 1998 ) by a proline residue is predicted to disrupt this putative -helix and could therefore greatly alter the overall structure of the HET-S protein. Conversely, the HET-S protein could be detected in the het-S L52S mutant, suggesting that the mutant protein is stable. This might indicate that the affected residue is directly involved in the reactivity during the incompatibility reaction.

The het-s T266P missense mutant is the only het-s mutant of this class. Its phenotype could correspond either to a loss of reactivity of the protein or to a loss of its ability to undergo the [Het-s*] to [Het-s] transition. It is possible that for some null mutants the altered protein is blocked in a nonreactive state similar to the HET-s* state.

Nonreactive mutants retaining [Het-s*] to [Het-s] converting activity: Two mutants, het-S E86K and het-s E26*, were found to have lost reactivity in incompatibility but after confrontation with a [Het-s] strain, these two strains were found to be able to induce the [Het-s*] to [Het-s] transition. These strains display a new phenotype different from that of any known mutant or wild-type allele. The capacity to induce the [Het-s*] to [Het-s] transition had never been observed in the absence of reactivity in incompatibility.

The het-S E86K mutant strain has gained the ability to induce the [Het-s*] to [Het-s] transition, a property that the het-S wild-type strain does not possess.

The second mutant of this class contains a premature stop codon at position 26 in the het-s allele. Two alternate hypotheses can explain the properties of this allele. First, it is possible that, as reported for the [URE3] yeast prion (MASISON et al. 1997 ), this short N-terminal peptide is sufficient to propagate the prion analog. However, it cannot be excluded that occasional read-through of the stop codon in this mutant can lead to the production of a very low amount of normal protein, sufficient to induce the transition but not sufficient to trigger the incompatibility reaction. Due to the autocatalytic mechanism of the [Het-s*] to [Het-s] conversion, it is likely that only a very low amount of HET-s protein is required to initiate the process. To test these hypotheses, a het-s allele truncated at amino acid position 26 (het-s 1-25) was constructed and expressed in a het-s° strain. Transformants were tested in the barrage reaction with a [het-S] strain and for converting activity. Transformants were neutral in the incompatibility reaction and did not spontaneously display the [Het-s*] to [Het-s] converting activity. However, after confrontation with a wild-type [Het-s] strain, the het-s(1-25) transformants acquired the [Het-s*] to [Het-s] converting activity; i.e., they were able to convert a [Het-s*] strain to [Het-s] phenotype after cytoplasmic mixing. Thus, the converting activity of the het-s E26* mutant appears to result from the expression of the truncated polypeptide rather than from occasional read-through of the stop codon at position 26.

These results imply that expression of this short peptide encompassing only the first 25 amino acids of HET-s is sufficient to induce the propagation of the [Het-s] phenotype.

Mutants affected in their specificity: Two mutants of het-S that have switched to the [Het-s] specificity have been obtained (het-S F25S and het-S Y38H). These mutant alleles have gained all the properties of wild-type het-s alleles. Strains bearing these alleles exhibit the het-s specificity in a barrage test; i.e., they are incompatible with a het-S strain and they can display alternate [Het-s*] and [Het-s] phenotypes. When they display the [Het-s] phenotype, they can induce the transition of [Het-s*] strains to the [Het-s] phenotype. Furthermore, as observed for wild-type [Het-s] strains, the amount of protein detected in these mutants on Western blots is at least fivefold lower than in wild-type het-S strains (not shown). So these het-S mutants display all properties of the wild-type het-s allele.

Mutants in which the specificity was altered have been previously described (DELEU et al. 1993 ). Construction of het-s/het-S chimeric alleles and site-specific mutagenesis revealed that amino acids present at positions 23 and 33 are essential in defining specificity in incompatibility. The location of mutations leading to alternate specificity at positions 25 and 38 strongly supports the hypothesis that the N-terminal domain of the protein is essential for specificity in incompatibility. It should be noted that these mutations not only modify specificity from [Het-S] to [Het-s] but also permit the expression of the nonreactive [Het-s*] phenotype. Thus, this N-terminal region also appears crucial to support expression of the alternate [Het-s*] and [Het-s] phenotypes.

The third mutant found to be affected for specificity is the het-S P175S mutant. Strains containing this mutant allele display a complex phenotype. When this mutant strain is subcultured, three phenotypic classes can be distinguished. A number of subcultures display the [Het-s] phenotype, others the [Het-S] phenotype. The remaining strains, defining a third phenotypic class, exhibit a growth defect (slow growth and a decreased density of the filaments) and produce a barrage reaction with both the [Het-s] and the [Het-S] testers. After cytoplasmic contact with the [Het-s] strain, the phenotype of the mutant is stabilized under the [Het-s] phenotype but upon further subcultures the phenotypic heterogeneity described above is recovered. The growth defect observed in subcultures producing the barrage reaction with both testers can be interpreted as the result of the coexpression in the same mycelium of the two antagonistic specificities. This metastable phenotype is similar to the phenotype resulting from mutations in het-s at amino acid positions 23 and 33. This phenotype is designated [Het-S^S] (DELEU et al. 1993 ). The het-s P33H and het-S A23D alleles allow expression of the alternate [Het-s] and [Het-S] specificities from the same nucleotide sequence. But once acquired, the [Het-s] phenotype of these two mutants is vegetatively stable whereas the het-S P175S mutant strain is able to switch back from [Het-s] to [Het-S] during vegetative propagation.

The het-S P175S mutation affects specificity but the mutation does not lead to a simple specificity switch; nevertheless, it allows expression of a [Het-s] specificity from a mutant of the het-S allele. This indicates that specificity is not determined solely by the N-terminal part of the protein.

Mutants that display a novel metastable phenotype, [Het-S*]: Two mutants in het-S were found to display a neutral phenotype that could spontaneously switch to [Het-S]. In one mutant, the mutation leads to a substitution of a lysine residue for an arginine at position 201. The other mutation results in a premature stop codon at position 240. This novel unstable phenotype was designated [Het-S*] by similarity with the behavior of [Het-s*]. Spontaneous [Het-S*] to [Het-S] switching was much more frequent than the [Het-s*] to [Het-s] transition. Unlike [Het-s*], the [Het-S*] state cannot be propagated for long periods. Therefore, it was not possible to determine whether confrontation with a [Het-S] or a [Het-s] strain significantly increases the frequency of the [Het-S*] to [Het-S] transition.

Mutation affecting vegetative growth in a compatible genetic background: A mutation of the het-s allele leading to the replacement of an arginine residue by a glycine at position 165 allowed escape from self-incompatibility but did not result in a complete restoration of the growth. This mutant grows slowly, and the mycelium is unpigmented and colonial. When this mutation was recovered in a compatible genetic background (in the absence of the antagonistic allele), the mutant strain displayed the same growth alteration as the original mutant sector. It could not be determined if this allele displays the [Het-s] or the [Het-S] specificity in barrage tests because the altered growth of the mutant strain made the barrage test inconclusive. This strain did not promote the [Het-s*] to [Het-s] transition. Thus, it appears that this amino acid substitution results in the loss of the [Het-s*] to [Het-s] converting activity. It is, however, possible that the growth defect of this strain alters the cytoplasmic contact with the [Het-s*] tester, thus limiting the diffusion of HET-s protein and therefore preventing the [Het-s*] to [Het-s] transition. The position of the mutation, amino acid 165, is one of the 13 polymorphic positions in which HET-s and HET-S differ. This mutant het-s allele encodes a glycine at position 165, the amino acid that is present at this position in the wild-type het-S allele. As it has been proposed that the incompatibility reaction is triggered by HET-s/HET-S heteromultimers (COUSTOU et al. 1997 ), it is possible that homomultimers of this mutant protein resemble HET-s/HET-S heteromultimers. The observed growth defect could then correspond to an attenuated form of the incompatibility reaction.

Properties of het-s deletion constructs:
We have constructed a series of deletions in the het-s gene to identify regions of the protein that are essential for reactivity in incompatibility and to induce the [Het-s*] to [Het-s] transition. The deleted alleles were introduced by transformation into the het-s° strain. Six truncated het-s alleles were constructed, het-s (1-253), het-s (1-229), het-s (82-121), het-s (30-289), het-s (1-112), and het-s (85-289). A number of truncated alleles [het-s (1-229), het-s (82-121), and het-s (30-289)] were expressed from the het-s promoter while others were cloned downstream of the strong G3PD promotor of A. nidulans [het-s (1-112) and het-s (85-289)]. The het-s (1-253) allele was expressed from both promoters (Figure 3).

View larger version (39K): In this window In a new window Download PPT slide   Figure 3. Activity of het-s deletion constructs. The structure of the different deletion constructs is given with the results of the assay for [Het-s*] to [Het-s] converting activity and barrage reaction with [Het-S]. The promotor used to express the deletion construct is given (either the native promotor, noted het-s, or the strong A. nidulans G3PD promotor, noted GPD). The assays have been performed either before or after cytoplasmic contact (confrontation on solid medium) with a [Het-s] strain.

Transformants were tested for the barrage reaction with a het-S strain either directly or after a confrontation with the [Het-s] strain. Transformants were also tested for their ability to induce the [Het-s*] to [Het-s] transition.

The het-s (1-253), het-s (1-229), and het-s (30-289) alleles did not retain reactivity in the barrage test and did not induce the [Het-s*] to [Het-s] transition. The het-s (30-289) and het-s (1-229) alleles were found to cause a slight growth alteration. This could lead to a decreased frequency of fusions between filaments and reduce cytoplasmic mixing. Such an alteration could then prevent not only the induction of the [Het-s*] to [Het-s] transition but also the macroscopic expression of the incompatibility reaction (formation of barrages).

The het-s (82-121) allele was found to have retained the ability to trigger the [Het-s*] to [Het-s] transition; however, in the barrage test, the reactivity of this mutant in incompatibility was weak. Different subcultures of the same transformant could either present a weak barrage or no barrage reaction at all when tested against the het-S strain. The region from position 82 to 121, therefore, is not required for triggering the [Het-s*] to [Het-s] transition but seems to be essential for full reactivity in incompatibility.

Transformants expressing the het-s (1-112) or the het-s (86-289) constructs produce a barrage reaction with [Het-S] strains and also possess the [Het-s*] to [Het-s] converting activity. This occurred only if the transformants had been previously confronted with a [Het-s] strain. The two deletion products overlap in a 27-amino-acid region. This region is absent in the het-s (82-121) allele product, which has retained converting activity. This region then seems to be dispensable for converting activity. Therefore, it appears that the reactivity of the HET-s protein can be determined by two different regions of the protein. There is an apparent contradiction in the fact that larger deletion construct regions have lost reactivity. For instance, the het-s (1-253) construct includes the N-terminal 112 region (1–112) and the het-s (30-289) construct includes the C-terminal region (85–289), but both lack reactivity. This could be due to the fact that the corresponding peptides are unstable and rapidly degraded. This is supported by the fact that the proteins expressed from the het-s (1-229) and het-s (30-289) alleles were not detected on Western blots (not shown). Alternatively, the partially truncated C-terminal or N-terminal regions of these peptides might prevent the interaction with HET-S or, even if formed, the heterodimer might not be active, i.e., might not be able to trigger the incompatibility reaction.

The polypeptide that is expressed from the het-s (1-112) allele is the shortest fragment of the HET-s protein defined in this study that retains activity in incompatibility.

Consequences of amino acid replacements at position 33:
Previous studies of het-s specificity have demonstrated the essential role of the N-terminal region of the protein in specifying the differences between het-s and het-S. A mutation in het-S that leads to the replacement of the histidine residue at position 33 by a proline, the amino acid present at this position in HET-s, yields an allele that displays all the properties of the wild-type het-s allele. Therefore, this position participates both in defining the specificity in incompatibility and in expressing the metastable [Het-s*] phenotype (DELEU et al. 1993 ).

We wanted to determine if the observed specificity switch could be obtained with residues other than proline. We have therefore introduced random amino acid replacements at position 33 by site-specific mutagenesis. Mutants in which codon 33 is deleted have also been produced. Mutant alleles were used to transform a het-s° strain and the properties of the transformants were studied. Twenty-four mutant alleles were obtained, 12 in each allele. All the mutants were still reactive in incompatibility; no null mutant was recovered. They could be classified into three phenotypic classes (Table 1). Thirteen alleles behaved as wild-type het-s. These included het-s and het-S mutants in which the amino acid position 33 has been deleted (het-s 33 and het-S 33). Six mutants all derived from het-S behaved as wild-type het-S alleles. A group of 5 mutants defines a third phenotypic class designated [Het-S^s]. The primary transformants exhibit the [Het-S] phenotype but, either spontaneously or after contact with the [Het-s] strain, they switch to the [Het-s] phenotype. When the transition was induced by a contact with the [Het-s] strain, transition was not systematic but the frequency of the transition increased with the length of the confrontation with the tester. This could be due to the presence of the barrage that may limit the diffusion of the HET-s protein (the confronted strains are originally incompatible). This phenotype had previously been described for the het-s P33H and het-S A23D mutant alleles (DELEU et al. 1993 ) and for the het-S P175S described here above.

Out of the 13 het-s mutants, 10 expressed the [Het-s] specificity. For any given phenotypic group, amino acids with quite diverse physico-chemical properties can be found at position 33. However, in all polypeptides conferring the [Het-s] specificity, the residues found at position 33 were apolar with the exception of threonine. It should be noticed that appearance of the [Het-s] specificity was always associated with existence of a metastable phenotype; no mutants directly expressing the [Het-s] specificity, without displaying the [Het-s*] state, were obtained. As previously observed by DELEU et al. 1993 , some amino acid substitutions gave identical specificities, whether or not they were introduced in a het-s or a het-S allele. For example, het-s P33L and het-s P33I displayed the same specificity as het-S H33L and het-S H33I. In these examples, the properties of the het-s allele seem to be controlled only by polymorphism at position 33 and unaffected by the 12 other polymorphic positions. However, other substitutions gave opposite specificity, depending on the allele in which they were introduced. For example, het-s P33A and het-S H33A have opposite specificities, indicating that other amino acid differences participate in the specificity of the encoded polypeptides. These results also provide novel examples of single amino acid differences that are sufficient to lead to incompatible interactions. For example, wild-type het-S and het-S H33L are incompatible while the protein they encode only differs by one amino acid.

From these results it appears that the [Het-s] specificity can be obtained with various amino acids at position 33 and not only with proline. Thus, the properties of the HET-s protein are not based on the unique structural properties of the proline residue. This rules out the possibility that the [Het-s*] to [Het-s] transition is due to proline isomerization at that position.

	DISCUSSION

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

We have proposed that the HET-s protein encoded by the het-s heterokaryon incompatibility gene behaves as a prion, i.e., as an infectious protein. Strains containing this gene can display two distinct phenotypes. They can either be neutral or reactive in incompatibility. A transition from the nonreactive state to the reactive one can occur spontaneously or can be induced by cytoplasmic contact with a reactive strain. Previous results strongly suggest that the [Het-s*] to [Het-s] transition involves a heritable post-translational modification of the protein. We have proposed that this phenotypic transition corresponds to a structural modification of the protein. In this hypothesis HET-s* and HET-s are two conformers of the same protein and HET-s is capable of converting HET-s* into the reactive HET-s.

In an effort to progress in the functional dissection of this protein we have isolated mutations allowing escape from self-incompatibility resulting from the coexpression of het-s and het-S in haploid strains and constructed various mutant alleles of het-s. Amino acid positions critical for reactivity and specificity have been identified. Moreover, alleles conferring a new metastable phenotype designated [Het-S*] have been obtained. A 112-amino-acid-long N-terminal peptide of HET-s appears sufficient for full [Het-s] activity. Moreover, we found that expression of a mutated allele encoding only the first 25 amino acids of the het-s ORF is sufficient to induce the [Het-s*] to [Het-s] transition.

Escape from self-incompatibility: bias toward selection of het-S mutants:
Isolation of mutations allowing escape from self-incompatibility has proven to be a powerful tool in the genetic dissection of vegetative incompatibility in filamentous fungi. The genetic screen is based on growth restoration of a strain that grows poorly as the result of the coexpression of incompatible het genes. This self-incompatible strain can either be a forced heterokaryon, a partial diploid, or a haploid offspring from a cross in which nonallelic het genes are segregating. This approach has been used, for example, to isolate mutations in mod A, a suppressor of nonallelic incompatibility in P. anserina (BELCOUR and BERNET 1969 ), mutations that suppress incompatibility associated with the mating-type locus or the het-6 locus in Neuropora crassa (NEWMEYER 1970 ; GRIFFITHS 1982 ; SMITH et al. 1996 ). Molecular characterization of the mutational events has been shown to range from large deletions (SMITH et al. 1996 ) to point mutations inactivating the incompatible alleles (SAUPE et al. 1996 ). This article is the first to report the isolation of mutations that suppress self-incompatibility in which the incompatible alleles were brought together in the same nucleus by transformation. Transformation in filamentous fungi occurs mainly by ectopic integration and the stability of the transgene was sufficient to allow efficient isolation of suppressor mutations. All mutational events that we have characterized were found to correspond to single base substitutions; no deletion or insertion events were detected. Consistent with the spectrum of UV-light-induced mutations (KUNZ et al. 1987 ), the wide majority of the detected mutations were G:C to A:T transitions and, with one exception, occurred within dipyrimidine sequences. In one example, however, the het-S R165G mutant, the mutation corresponds to the replacement of an arginine codon by the glycine codon present in the het-s allele, suggesting that this mutant allele might have resulted from a gene conversion involving the two alleles.

A strong bias for the selection of het-S rather than het-s mutations was observed; among 17 selected mutants, 14 were het-S mutants. If one assumes that mutation rates are similar in the het-s and het-S sequences, two types of hypotheses can be proposed to explain this bias.

First, this bias might be due to the overexpression of the het-s allele in the strain used for selection. Expression from the strong G3PD promotor of A. niger results at least in a 20-fold increase in the amount of HET-s. The high amount of the HET-s protein might prevent to some extent the emergence of het-s mutant sectors because the turnover of the overexpressed protein is so slow. It is also possible that a number of "leaky" het-s mutations are compensated by the overexpression and therefore do not allow the escape of the mutant sector.

A second hypothesis is that fewer het-s mutations allow escape from self-incompatibility. It is possible that while numerous mutations can lead to total inactivation of het-S, only a few mutational events inactivate het-s. In the case of mutations altering specificity, it might be easier to convert a het-S allele to the [Het-s] specificity than the opposite. This is suggested both by the previous specificity studies (DELEU et al. 1993 ) and the amino acid replacements at position 33 presented here. Finally, it is possible that a number of mutations in het-s have a toxic effect as suggested, for example, by the phenotype of the het-s R165G mutation or that of some het-s deletion mutants.

[Het-S*], a novel metastable phenotype:
One novel result obtained in this work is the characterization of two mutants leading to the expression of a metastable phenotype designated [Het-S*]. The het-S K201R and het-S Q240* mutants spontaneously express this neutral [Het-S*] phenotype but they can switch to the active [Het-S] state. Three het-s-associated metastable phenotypes can now be distinguished. Wild-type het-s strains switch from [Het-s*] to [Het-s]; some mutants, the het-s P33H mutant, for instance, switch from [Het-S] to [Het-s], and the het-S K201R and Q240* mutants switch from [Het-S*] to [Het-S]. This locus then provides numerous examples of alternate phenotypes that can be expressed from the same nucleotide sequence.

In the prion hypothesis, the transition from HET-S* to HET-S in het-S K201R and het-S Q240* mutants would correspond to a conformational transition. It is possible that this putative HET-S* inactive form also exists in wild-type het-S strains but that the transition to HET-S form is too efficient to allow detection of a neutral [Het-S*] phenotype. The selected mutation, by reducing the rate of this transition, would permit the expression of the [HET-S*] phenotype.

The N-terminal parts of the HET-s and the HET-S proteins control their specificity in incompatibility:
Previous studies emphasized the importance of the N-terminal end of the protein in the control of the specificity in the incompatibility (DELEU et al. 1993 ). Polymorphism at positions 23 and 33 was found to be responsible for the difference of specificity between het-s and het-S. We have characterized two additional positions, 25 and 38, located in the same region of the protein at which amino acid substitution led to a specificity switch from [Het-S] to [Het-s]. This confirms the importance of this region in the control of the specificity of the two alleles. No mutants switching from [Het-s] to [Het-S] were obtained. A number of replacements at codon position 33 in het-S can produce alleles that display the [Het-s] specificity. These results show that several single amino acid substitutions at different positions in the N-terminal end of HET-S produce a protein that displays the [Het-s] specificity. This suggests that, from an evolutionary point of view, heterokaryon incompatibility can be obtained by various minor mutational events at this locus. Emergence of incompatible alleles at this locus can therefore be considered to be relatively likely.

Most striking is the fact that deletion of codon position 33 results in alleles with the [Het-s] specificity. Thus, although the nature of the codon present at that position determines specificity, its deletion does not cause appearance of a null allele but of an allele with the [Het-s] specificity. Similarly, numerous random replacements at position 33 produce mutants of the [Het-s] specificity but no null mutant is obtained. These random mutations are likely to alter the structure of the N-terminal part of HET-S. One could propose that the failure for this region to fold in a defined secondary structure, existing in HET-S, produces proteins that display the [Het-s] specificity by default. In the prion hypothesis, these mutant HET-S polypeptides would become competent for HET-s-directed folding because they are unable to acquire a defined structure in that region.

The N-terminal part of HET-s allows propagation of the [Het-s] prion analog:
For Ure2p and Sup35p, it was possible to delimit a prion domain that corresponds to the N-terminal part of the protein (TER-AVANESYAN et al. 1994 ; MASISON and WICKNER 1995 ). A similar delimitation was obtained in the case of HET-s. We have shown that a 112-amino-acid-long N-terminal peptide of HET-s possesses reactivity in incompatibility and converting activity. It also appeared that the C-terminal part (position 85 to 289) of HET-s can display these two properties. The fact that the C-terminal part can participate in the expression of [Het-s] specificity is illustrated by the het-S P175S mutant allele. This mutant is affected in this C-terminal region and gains [Het-s] reactivity and converting activity. The incompatibility reaction is thought to be induced by the formation of poison complexes formed between the protein encoded by het-s and het-S alleles. The HET-s and HET-S proteins are able to form dimers in the yeast two-hybrid system (COUSTOU et al. 1997 ). Thus the [Het-s] reactivity in incompatibility should be understood as the ability to form heteromultimers with HET-S, which are abnormal and therefore poisonous. The results presented here suggest that a HET-s/HET-S interaction involving only the N-terminal or the C-terminal parts of HET-s is sufficient to trigger incompatibility. Furthermore, strains that express the N-terminal or C-terminal part of HET-s can switch from [Het-s*] to [Het-s]. Therefore both regions are able to transmit the proposed prion-like conversion, and the putative conformational modifications involve the two regions of the protein. There thus appears to be a functional redundancy in HET-s regarding [Het-s] activity. It has recently been shown that the ure2p protein contains two distinct prion-inducing domains (MADDELEIN and WICKNER 1999 ). The redundancy in the prion-inducing function is then a common feature in HET-s and Ure2p. In the same study, it was noted that the existence of the second prion-inducing domain could be unmasked after deletion of certain regions of the protein. This is reminiscent of the results presented here showing that deleted alleles larger than the reactive het-s (1-112) allele were lacking [Het-s] activity.

The crucial role of the N-terminal region in the conversion is emphasized not only by the mutants with modified specificity affected in this region, but also by the fact that the het-s E26* mutant, a strain that expresses only the 25 N-terminal amino acids of HET-s, possesses the ability to induce the [Het-s*] to [Het-s] transition. This converting activity is not acquired spontaneously but only after cytoplasmic mixing with a [Het-s] strain. This apparently indicates that the 25-amino-acid-long N-terminal peptide is modified by the HET-s protein from the [Het-s] tester and is then capable of transmitting this modification. Among the peptides that appear capable of inducing prion propagation, this 25-amino-acid-long polypeptide is the shortest described for any of the known systems.

The [Het-s*] to [Het-s] transition involves a post-translational modification of the protein that does not affect its electrophoretic mobility. This modification could either be a covalent modification or, in the prion hypothesis, be conformational. We are currently favoring the hypothesis that this transition corresponds to a transmissible conformational modification. The fact that a mutant strain expressing only a 25-amino-acid-long peptide retains the ability to propagate [Het-s] is difficult to reconcile with a model of an enzymatic covalent modification of HET-s. Confirmation of the prion hypothesis for [Het-s] propagation requires the purification of the het-s-encoded proteins and their structural characterization. The different mutants with altered properties and reactive deletion constructs described herein will be valuable tools for the acquisition and interpretation of future structural data. This will allow to precise the mechanism of the infectious [Het-s*] to [Het-s] transition and to determine to what extent it resembles the propagation of other known prions or prion analogs.

	ACKNOWLEDGMENTS

We express our gratitude to Martine Sabourin for her valuable technical assistance in the genetic and molecular analyses. Brice Roux and Magalie Verdier helped in the construction of the deletion alleles.

This work was supported by grants from the Centre National de la Recherche Scientifique and from the Comité Interministériel du programme de recherche sur les ESST et les prions (ACC n°1). V.C. is funded by the Ministère de l'Education Nationale, de l'Enseignement Supérieur et de la Recherche.

Manuscript received June 18, 1999; Accepted for publication August 16, 1999.

	LITERATURE CITED

TOP ABSTRACT MATERIALS AND METHODS RESULTS DISCUSSION LITERATURE CITED

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

BELCOUR, L., 1976  Loss of a cytoplasmic determinant through formation of protoplasts in Podospora.. Neurospora Newsl. 23:26-27.

BELCOUR, L. and J. BERNET, 1969  Sur la mise en évidence d'un gène dont la mutation supprime spécifiquement certaines manifestations d'incompatibilité chez le Podospora anserina.. C. R. Acad. Sci. Paris 269:712-714.

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].

BLABER, M., X. ZHANG, and B. W. MATTHEWS, 1993  Structural basis of amino acid helix propensity. Science 260:1637-1640[Abstract/Free Full Text].

CARROL, A. M., J. A. SWEIGARD, and B. VALENT-CENTRAL, 1994  Improved vectors for selecting resistance to hygromycin. Fungal Genet. Newsl. 41:22.

CHERNOFF, Y. O., S. L. LINDQUIST, B. ONO, S. G. INGE-VECHTOMOV, and S. W. LIEBMAN, 1995  Role of the chaperone protein Hsp104 in propagation of the yeast prion-like factor. Science 268:880-884. [psi+][Abstract/Free Full Text].

COUSTOU, V., C. DELEU, S. SAUPE, and J. BEGUERET, 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 64:9773-9778.

DELEU, C., C. CLAVE, and J. BEGUERET, 1993  A single amino acid difference is sufficient to elicit vegetative incompatibility in the fungus Podospora anserina.. Genetics 135:45-52[Abstract].

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

FRISHMAN, D. and P. ARGOS, 1996  Incorporation of non-local interactions in protein secondary structure prediction from the amino acid sequence. Protein Eng. 9:133-142[Abstract/Free Full Text].

GLOVER, J. R., A. S. KOWAL, E. C. SCHIRMER, M. M. PATINO, and J. LIU et al., 1997  Self-seeded fibers formed by Sup35, the protein determinant of [PSI+], a heritable prion-like factor of S. cerevisiae.. Cell 89:811-819[Medline].

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

HORWICH, A. L. and J. S. WEISSMAN, 1997  Deadly conformations-protein misfolding in prion disease. Cell 89:499-510[Medline].

KING, C., P. TITTMANN, H. GROSS, R. GEBERT, and M. AEBI et al., 1997  Prion-inducing domain 2-114 of yeast Sup35 protein transforms in vitro into amyloid-like filaments. Proc. Natl. Acad. Sci. USA 94:6618-6622[Abstract/Free Full Text].

KUNZ, B. A., M. K. PIERCE, J. R. MIS, and C. N. GIROUX, 1987  DNA sequence analysis of the mutational specificity of u.v. light in SUP4-o gene of yeast. Mutagenesis 2:445-453[Abstract/Free Full Text].

KUSHNIROV, V. V. and M. D. TER-AVANESYAN, 1998  Structure and replication of yeast prions. Cell 94:13-16[Medline].

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

LINDQUIST, S., 1997  Mad cows meet psi-chotic yeast: the expansion of the prion hypothesis. Cell 89:495-498[Medline].

MADDELEIN, M. L. and R. B. WICKNER, 1999  Two prion-inducing of Ure2p are nonoverlapping. Mol. Cell. Biol. 19:4516-4524[Abstract/Free Full Text].

MASISON, D. C. and R. WICKNER, 1995  Prion-inducing domain of yeast Ure2p and protease resistance of Ure2p in prion-containing cells. Science 270:93-95[Abstract/Free Full Text].

MASISON, D. C., M. L. MADDELEIN, and R. WICKNER, 1997  The prion model for [URE3] of yeast: spontaneous generation and requirements for propagation. Proc. Natl. Acad. Sci. USA 94:12503-12508[Abstract/Free Full Text].

NEWMEYER, D., 1970  A suppressor of heterokaryon-incompatibility associated with mating type in Neurospora crassa.. Can. J. Genet. Cytol. 12:9914-9926.

PACE, C. N. and J. M. SCHOLTZ, 1998  A helix propensity scale based on experimental studies of peptides and proteins. Biophys. J. 75:422-427[Medline].

PATINO, M. M., J. LIU, J. R. GLOVER, and S. LINDQUIST, 1996  Support for the prion hypothesis for inheritance of a phenotypic trait in yeast. Science 273:622-626[Abstract].

PAUSHKIN, S. V., V. V. KUSHNIROV, V. N. SMIRNOV, and M. D. TER-AVANESYAN, 1996  Propagation of the yeast prion-like [psi+] determinant is mediated by oligomerization of the SUP35-encoded polypeptide chain release factor. EMBO J. 15:3127-3134[Medline].

PAUSHKIN, S. V., V. V. KUSHNIROV, V. N. SMIRNOV, and M. D. TER-AVANESYAN, 1997  In vitro propagation of the prion-like state of yeast Sup35 protein. Science 18:381-383.

PRUSINER, S. B., 1998  Prions. Proc. Natl. Acad. Sci. USA 95:13363-13383[Abstract/Free Full Text].

PUNT, P. J., M. A. DINGEMANSE, B. J. M. JACOBS-MEIJSING, P. H. POUWELS, and C. A. M. J. J. VAN DEN HONDEL, 1988  Isolation and characterization of the glyceraldehyde-3-phosphate dehydrogenase gene of Aspergillus nidulans. Gene 69:49-57[Medline].

RIZET, G., 1952  Les phénomènes de barrage chez Podospora anserina, I. Analyse génétique des barrages entre souches S et s. Rev. Cytol. Biol. Veg. 13:51-93.

SAUPE, S., L. STENBERG, K. T. SHIU, A. J. F. GRIFFITHS, 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].

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].

TAYLOR, K. L., N. CHENG, R. W. WILLIAMS, A. C. STEVEN, and R. B. WICKNER, 1999  Prion domain initiation of amyloid formation in vitro from native ure2p. Science 283:1339-1343[Abstract/Free Full Text].

TER-AVANESYAN, M. D., A. R. DAGKESAMANKAYA, V. V. KUSHNIROV, and V. N. SMIRNOV, 1994  The SUP35 omnipotent suppressor gene is involved in the maintenance of the non-mendelian determinant [psi+] in the yeast Saccharomyces cerevisiae.. Genetics 137:671-676[Abstract].

THUAL, C., A. A. KOMAR, L. BOUSSET, E. FERNANDEZ-BELLOT, and C. CULLIN et al., 1999  Structural characterization of the Saccharomyces cerevisiae prion-like protein Ure2. J. Biol. Chem. 274:13666-13674[Abstract/Free Full Text].

TURCQ, B., C. DELEU, M. DENAYROLLES, and J. BEGUERET, 1991  Two allelic genes responsible for vegetative incompatibility in the fungus Podospora anserina are not essential for cell viability. Mol. Gen. Genet. 288:265-269.

WICKNER, R., 1994  as an altered URE2 protein: evidence for a prion analog in Saccharomyces cerevisiae.. Science 264:566-569. [URE3][Abstract/Free Full Text].

This article has been cited by other articles:

				R. B. Wickner, H. K. Edskes, B. T. Roberts, U. Baxa, M. M. Pierce, E. D. Ross, and A. Brachmann Prions: proteins as genes and infectious entities Genes & Dev., March 1, 2004; 18(5): 470 - 485. [Full Text] [PDF]

				H. J. P. Dalstra, K. Swart, A. J. M. Debets, S. J. Saupe, and R. F. Hoekstra Sexual transmission of the [Het-s] prion leads to meiotic drive in Podospora anserina PNAS, May 27, 2003; 100(11): 6616 - 6621. [Abstract] [Full Text] [PDF]

				Q. Xiang and N. L. Glass Identification of vib-1, a Locus Involved in Vegetative Incompatibility Mediated by het-c in Neurospora crassa Genetics, September 1, 2002; 162(1): 89 - 101. [Abstract] [Full Text] [PDF]

				S. W. Liebman Progress toward an ultimate proof of the prion hypothesis PNAS, July 9, 2002; 99(14): 9098 - 9100. [Full Text] [PDF]

				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]