INTRODUCTION

Top Abstract Introduction Materials & Methods Results Discussion References

The life cycle of a species has an impact on the structure of its populations. In the basidiomycete Agaricus bisporus (Lange) Imbach, the button mushroom, the life cycle varies with the wild population examined (1, 3, 5, 13). A. bisporus has a unifactorial sexual incompatibility system (24) and an amphithallic life cycle; i.e., it is both pseudohomothallic and heterothallic. Of the approximately 500 species of agarics (22), 9% are considered amphithallic. Individual spores can be heterokaryotic (n+n) or homokaryotic (n) (20, 23).

In three field populations (Alberta, coastal California, and France) and the cultivated strains, all of which belong to A. bisporus var. bisporus, most basidia are bisporic and produce the heterokaryotic spores that characterize a pseudohomothallic (or secondarily homothallic) (28) life cycle (Fig. 1). In this life cycle, each of the two spores borne on a basidium usually receives two nonsister postmeiotic nuclei carrying different mating type alleles and germinates to produce heterokaryotic fertile mycelium (18, 27).

View larger version (41K): [in this window] [in a new window]   FIG. 1.   Schematic representation of the life cycles of A. bisporus. Heterokaryotic spores (n + n) receive two compatible meiotic products. Homokaryotic spores from heterothallic strains give rise to mycelia that are usually sterile but sometimes fertile in fruiting tests (3). During plasmogamy and following anastomosis, two homokaryotic and sexually compatible primary mycelia give rise to a heterokaryotic secondary mycelium. The three percentages given for the type of basidia are the mean percentages of basidia that produce, respectively, four or more spores, three spores, and one or two spores (5).

Agaricus bisporus var. burnettii Kerrigan & Callac was recently described on the basis of specimens found in the Sonoran Desert of California (3). Most of the basidia of this variety are tetrasporic, and this organism has a predominantly heterothallic life cycle (16). Each of the four spores borne on each basidium receives one meiotic product and germinates to form a homokaryotic mycelium. Plasmogamy between two sexually compatible homokaryons gives rise to a fertile heterokaryotic mycelium. A. bisporus var. bisporus and A. bisporus var. burnettii are completely interfertile. The number of spores per basidium is primarily determined by the BSN locus linked to the mating type locus (MAT) (33) on chromosome I. The Bsn-t allele (tetrasporic basidia) is dominant with respect to the Bsn-b allele (bisporic basidia) with variable penetrance (4, 11, 12). In the chromosomal region near BSN, recombination was greater in a Bsn-t/b heterozygote (4) than in a Bsn-b/b homozygote (18), suggesting that recombination might be greater in the A. bisporus var. burnettii population.

By introducing Bsn-t into heterokaryons, highly recombined haploid progeny could be obtained from crosses between bisporic and tetrasporic strains. Unfortunately, one component of the heterokaryon must be from the Sonoran Desert population, which limits the genetic background available for the crosses. Recently, we identified two tetrasporic strains obtained in France (2, 5). Our objectives in this study were (i) to determine if these French isolates carried Bsn-t alleles that conferred the tetrasporic phenotype and heterothallism, (ii) to determine the heritability of these traits from the French isolates, and (iii) to determine recombination rates and other properties of hybrids carrying the French Bsn-t alleles.

	MATERIALS AND METHODS

Top Abstract Introduction Materials & Methods Results Discussion References

Strains. A total of 250 wild specimens of A. bisporus were collected from 67 different sites in France, and two tetrasporic isolates, Bs 261 and Bs 423, were detected (5). These two strains were isolated by tissue culture of fruiting bodies collected under Cupressus macrocarpa at two coastal sites that were 230 km apart; Bs 261 was collected at Dinard by François Callac and P. Callac in 1992, and Bs 423 was collected at Olonne-sur-mer by Jacques Guinberteau and P. Callac in 1994. U 1 is a bisporic cultivar, and JB 3 is a tetrasporic isolate from the Sonoran Desert of California. Homokaryotic strains U 1-2 and JB 3-128 carrying the Mat-1 and Mat-5 alleles (16) were used for mating tests and hybridization. Two single-spore isolates (SSIs) of Bs 261, Bs 261-20, and Bs 261-32 have been described previously as Bs 261-9 and Bs 261-13 (2).

Fructification. Fruiting tests were performed at 15°C and a relative humidity of 90%. Wild isolates were grown in wood trays containing 15 kg of pasteurized compost that had been inoculated with a grain spawn culture. Hybrids and homokaryons were grown in plastic trays containing either 100 or 500 g of pasteurized compost that had been inoculated with cultures from four petri dishes containing compost agar medium. At least two trays were prepared for each strain.

Basidial spore number variables. The numbers of spores per basidium, which varied from one to six, were determined by light microscopy (11). Basidia having one or two spores were placed in the bisporic class (BSC), basidia having three spores were placed in the three-spore class (3SC) and basidia having four or more spores were placed in the tetrasporic class (TSC). The average number of spores per basidium (ASN) (11, 17) was calculated as follows: ASN = [2(BSC) + 3(3SC) + 4(TSC)]/(BSC + 3SC + TSC). The theoretical fraction of homokaryotic spores (FHS) was calculated by assuming that two-, three-, and four-spore basidia produced 2, 1, and 0 heterokaryotic spores, respectively, and 0, 2, and 4 homokaryotic spores, respectively (Fig. 1). Therefore, FHS = [2(3SC) + 4(TSC)]/[2(BSC) + 3(3SC) + 4(TSC)]. Algebraic substitution showed that FHS = 2  (4/ASN).

Graphic representation. For a given strain, if two of the four values described above (BSC, TSC, 3SC, and ASN) are known, the two remaining values can be calculated. Therefore, it is possible to represent a given strain by one point in a [BSC, TSC] system of axes, as well as in a [3SC, ASN] system of axes (5, 11). In the graphic representation used here, the two systems of axes are combined. Thus, for any strain, the values for the four variables can be represented by a single point in the triangular part of the graph.

Protoplasts. Protoplasts were released from heterokaryons by using Glucanex (Novo Ferment, Swiss SA) to digest cell walls, and homokaryons were recovered by the method of Kerrigan et al. (16).

Hybridization and pedigree of the hybrids. Spore germination and mating tests were conducted on compost agar medium as described by Kerrigan et al. (16). The presumed heterokaryotic fluffy mycelium was removed at the junction line between two homokaryons and subcultured. When necessary, heterokaryosis was confirmed by demonstrating heterozygosity at at least one marker. The pedigrees of the hybrids are summarized in Fig. 2. Except for the hybrid Bs 261-32 × JB 3-128, which resulted from a cross between two tetrasporic isolates (this hybrid was called the TT hybrid [for tetrasporic × tetrasporic]), all of the hybrids resulted from crosses with the U 1-2 homokaryotic tester, which carries the recessive Bsn-b allele.

View larger version (25K): [in this window] [in a new window]   FIG. 2.   Pedigrees of the hybrids. (A) First-generation hybrids derived from crosses between U 1-2 (a homokaryotic isolate produced via protoplasting from bisporic strain U 1) and SSIs from Bs 261. [T], tetrasporic; [B], bisporic. (B) Second-generation hybrids derived from crosses between U 1-2 and SSIs from the TT hybrid (a hybrid between the tetrasporic Bs 261 and JB 3 strains). (C) Hybrids derived from crosses between U 1-2 and either Bs 423 or Bs 423-2 (an SSI from Bs 423). Presumed BSN genotypes are indicated.

Alloenzyme markers. Homo- or heteroallelism and segregation in the offspring of Bs 261 were analyzed at the esterase 1 (EST1) and phosphoglucomutase (PGM) loci (6, 16, 29). The genetic nomenclature used is the nomenclature of Kerrigan et al. (15).

Molecular markers. Three sequence-characterized amplified-region (SCAR) (26) markers (PR6, PR2, and PR5) were analyzed in the offspring of Bs 261 and the offspring of the TT hybrid. On a previous map, PR6-MAT, PR2, PR5, and BSN were located in that order (PR6 and MAT are completely linked) on chromosome I (4). DNA extraction and PCR were performed as previously described (11). The primer sequences used and the restriction endonucleases used (RsaI and HaeIII) have been described by Callac et al. (4). After restriction, digestion fragments were separated by electrophoresis in 1.2% agarose. Six codominant alleles were identified previously (4). We identified three new alleles, Pr6-4, Pr5-5, and Pr5-7; these alleles gave bands at approximately 830 bp, at approximately 220, 210, 180, 65, and 55 bp, and at approximately 270, 210, 180, and 65 bp, respectively, following digestion by HaeIII.

Methods used to analyze the ploidy of the SSI offspring. We evaluated the proportions of homokaryons in the offspring of Bs 261 and Bs 423. Two methods described by Kerrigan et al. (16) were used, a multilocus genotype test and a mating test. Both tests were used for the offspring of Bs 261, but for the offspring of Bs 423 the genotype test was not informative because the parental strain was not heterozygous.

	RESULTS

Top Abstract Introduction Materials & Methods Results Discussion References

Characterization of Bs 261. Bs 261 produced brown, morphologically typical fruiting bodies and was clearly tetrasporic (BSC = 0.11, 3SC = 23.22, TSC = 76.67, ASN = 3.77). Its genotype is shown in Table 1. Eight of the 10 alleles are common; the exceptions are Pgm-3 and Est1-1, which are rare in the French population and have frequencies of <5% (15). Among the French field isolates, Bs 261 was the only isolate carrying both of these rare alleles.

Ploidy of Bs 261 offspring. A total of 128 SSIs were prepared, from a sporeprint of a Bs 261 sporocarp, and the level of spore germination was approximately 12%. The ploidy of 40 randomly selected SSIs was determined. In multilocus genotype tests, a single SSI, Bs 261-20, was heterozygous at all five loci, while the 39 other SSIs were homozygous at the same loci (Table 1). Heterozygosity at all five loci is clear evidence that Bs 261-20 is heterokaryotic. For the 39 other SSIs, PGM, EST1, and PR6 segregated independently in the three pairwise combinations (contingency ² tests). The probability of simultaneously losing parental heterozygosity at these three markers was low, so these progeny were considered homokaryotic. EST1 segregated in a 1:1 manner (17:22) in this progeny, but segregations at the other four loci were skewed against the alleles from nucleus B of Bs 261 (see below).

Distribution of spore number data for first-generation hybrids. Heterokaryons were subcultured from the 36 positive matings between SSIs of Bs 261 and U 1-2 (Fig. 2A). All 36 heterokaryons fruited, and the numbers of spores per basidium were determined. Two classes of heterokaryons were found (Fig. 3). The ASN of the 23 bisporic [B] hybrids (ASN < 3) ranged from 2.1 to 2.5, and the average ASN was 2.2 ± 0.1 (95% confidence interval). For the 13 tetrasporic hybrid progeny [T] (ASN > 3), the ASN ranged from 3.3 to 3.8, and the average ASN was 3.6 ± 0.1 (95% confidence interval). If basidial spore number is controlled by a single BSN locus, then Bs 261 is Bsn t/b, the 23 [B] hybrids are Bsn-b/b, and the 13 [T] are Bsn-t/b (Table 1). Segregation is not significantly different from 1:1.

View larger version (31K): [in this window] [in a new window]   FIG. 3.   Graphic representation of the distribution of the four basidial spore number variables and frequency distribution of ASN for 36 first-generation hybrids (Bs 261-x × U 1-2). The ASN values for progenitors Bs 261 and U 1 are 3.77 and 2.12, respectively. The bisporic and tetrasporic groups lie below and above, respectively, the ASN = 3.0 point. BSC, 3SC, and TSC are the percentages of basidia bearing, respectively, one or two spores, three spores, and four or more spores.

Recombination rate in the segregating offspring of Bs 261. In previous progeny, MAT, PR6, PR2, PR5, and BSN were linked on a segment of chromosome I (4, 11, 18). MAT could not be analyzed in the segregating offspring of Bs 261 because only 40% (51 of 128) of the confrontations gave positive mating interactions when eight SSIs were paired with 16 standard tester strains (12). The results of mating tests between SSIs of Bs 261 were either negative or ambiguous.

Test for allelism at BSN. To determine if the BSN locus responsible for the tetrasporic phenotype was the same in the French and Sonoran Desert strains, we crossed JB 3-128 with Bs 261-32 (Fig. 2B). The resulting TT hybrid gave highly tetrasporic sporocarps (ASN = 3.96). Sixty SSIs (TT-x) were obtained, and 52 of these SSIs formed heterokaryons with U 1-2 (Fig. 2B). These heterokaryons all fruited and were tetrasporic (average ASN, 3.62). These data suggest that the French and Sonoran Desert strains carry the Bsn-t allele at the same locus.

Recombination rate in the segregating offspring of the TT hybrid. JB 3-128 × Bs 261-32 was heterozygous for all three SCAR markers on chromosome I. Genotypes at the three SCAR markers were determined for 42 homokaryotic SSIs (Table 1). The proportion of recombinant homokaryons in the TT offspring (20 of 42 homokaryons) was significantly higher than the proportion of recombinants from Bs 261 (2 of 39 homokaryons) (Table 1).

Haploid fruiting. We tested 17 SSIs from Bs 261 (including the single heterokaryotic SSI) for haploid fruiting. Sporocarp caps were used for basidial counts, and sporocarp stems were used to obtain tissue for allozyme and/or SCAR identification. As expected, the single heterokaryon, Bs 261-20, gave tetrasporic sporocarps (ASN = 3.86). Of the 16 homokaryotic SSIs tested, 9 produced fruiting bodies with the same genotype as the homokaryon from which they were derived. Fructification was generally poor and late. The ASN values for the fruiting bodies of these homokaryotic SSIs ranged from 2.0 to 3.2. No correlation was detected between the ASN values of the homokaryons and the BSN genotypes (based on an analysis of the first-generation hybrids).

Characterization of Bs 423 and analysis of its progeny. Bs 423 normally fruited when it was cultivated under standard conditions on compost and produced sporocarps with particularly smooth light brown (putty) caps that were darker in the center and lighter on the edges. The stems were covered by adpressed and transversally arranged fibrillose scales. These traits occur infrequently in A. bisporus. Bs 423 was clearly tetrasporic (BSC = 2, 3SC = 24, TSC = 74, ASN = 3.71). The genotype of Bs 423 (Table 1) is similar to the genotype of the deduced haploid component B of Bs 261. Bs 423 was the only wild French isolate among the 90 tested to have only a single allele at all five markers, and none of the 90 isolates carried all five alleles of Bs 423.

	DISCUSSION

Top Abstract Introduction Materials & Methods Results Discussion References

Life cycle of the two tetrasporic strains. Eight characteristics of the two strains studied are compared in Table 2. Bs 261 was heterokaryotic. The observed fraction of homokaryons among its offspring, 0.975 (39 of 40 offspring), agrees with the theoretical fraction of homokaryotic spores (0.939). Therefore, its life cycle is predominantly heterothallic. Mycelia of Bs 423 and of its SSIs not only had the same haploid genotypes at all of the loci analyzed but also produced similar tetrasporic sporocarps. These data are consistent with a homothallic (or primarily homothallic [9]) life cycle (Fig. 1). Homokaryotic status should be confirmed by cytological examination at the time that spores are formed, as was done for the homothallic tetrasporic species Agaricus campestris (19).

Control of the basidial spore number. The analysis of the offspring of Bs 261 indicated that this strain is heterozygous at BSN and that Bsn-t, the allele for tetrasporism, is dominant. The Bsn-t allele of Bs 261 is probably the same as the Bsn-t allele in tetrasporic strains of A. bisporus var. burnettii. We do not think that Bs 261 belongs to A. bisporus var. burnettii because preliminary unpublished data have shown that Bs 261 is genotypically different. The tetrasporic trait from Bs 423 was dominant in hybrids with the bisporic strain U 1. Bs 423 appears to be haploid and has the same nuclear genotype at five markers as the nucleus of Bs 261 that carries Bsn-t does. This genotypic similarity and the phenotypic similarity in the basidial spore number trait suggest that these two strains carry the same Bsn-t allele.

Correlation between recombination rates and BSN genotypes. On the basis of analyses of four homokaryotic progenies, it appeared that the rate of recombination in chromosome I depended upon the BSN genotype. No recombination was observed among the progeny of a bisporic (A. bisporus var. bisporus × A. bisporus var. bisporus) Bsn-b/b hybrid (18). Rates of recombination were significantly higher for an intervarietal (A. bisporus var. bisporus × A. bisporus var. burnettii) Bsn-t/b hybrid (4) and even higher for Bs 261-32 × JB 3-128, which is Bsn-t/t. These data indicate that an incompletely dominant allele(s) for high recombination, possibly Bsn-t, occurs in heterothallic wild strains. This finding is consistent with the hypothesis that high and low rates of recombination are adaptative for heterothallic and pseudohomothallic isolates, respectively (4, 18). The pseudohomothallic propagation of A. bisporus var. bisporus through successive generations of heterokaryotic spores may permit deleterious alleles to accumulate. A low rate of recombination maintains a high level of heterozygosity and, by complementation, high viability and fitness among most offspring. Such suppression of recombination is unnecessary in heterothallic varieties in which deleterious alleles are subjected to selection during the homokaryotic phase. The variation in recombination observed on chromosome I needs to be confirmed in other portions of the genome.

Haploid fruiting. Haploid fruiting has previously been sporadically observed in culture (3, 7). For Bs 261, only a fraction (9 of 16) of the homokaryotic SSIs tested fruited, and those SSIs produced variable and sometimes feeble sporocarps. A greater number of progeny and loci must be examined to identify genetic determinant(s) for homokaryotic fruiting, as has been done in other basidiomycetes (8, 10, 30). The homokaryotic fruiting bodies of the SSIs of Bs 261 ranged from clearly bisporic to slightly tetrasporic even though about one-half of them carried the Bsn-t allele. In contrast, Bs 423 and all of its SSIs produced tetrasporic, presumably haploid, fruiting bodies. Homokaryotic fruiting bodies that produce predominantly bisporic, mitotic basidia have been found in Polyporus ciliatus and Agrocybe aegerita, while homokaryotic fruiting bodies that produce predominantly tetrasporic, meiotic basidia have been found in Coprinus cinereus (= Coprinus macrorhizus) (25) and Agrocybe aegerita (21). Cytogenetic studies are necessary to understand the basidial events in homokaryotic fruiting bodies of A. bisporus.

Relationships between the two tetrasporic isolates. Bs 423 appears to have the same genotype as one of the two nuclei of Bs 261. This similarity is unlikely to have occurred by chance because the genetic variability in French populations is high (32) and because two of the shared alleles, Pgm-3 and Bsn-t, are individually rare. Bs 423 is the first field isolate that appears to be homothallic. We did not observe allelic segregation at MAT among the progeny of Bs 423, and mating tests performed with Bs 423 SSIs generally failed in pairings with typical A. bisporus isolates. As a result of this apparent partial intersterility and of its ability to fruit at the haploid level, Bs 423 may belong to a homokaryotic "clonal" lineage (genet) that can persist in the field for an extended period of time. Under this scenario, Bs 261 could be a hybrid between a member of this genet and a homokaryon from a bisporic strain. This hypothesis is supported by the Bsn-t/b genotype of Bs 261 and by the cooccurrence of Pgm-3, Est1-3, and Bsn-t. We do not know of any report of such a cross under field conditions among basidiomycetes; however, restricted hybridization between homothallic and heterothallic strains of Sistotrema brinkmani, an aggregate of biological species, has been forced with nutritional markers. (31).

Use of the tetrasporic isolates in breeding work. Bsn-t alleles, high recombination rates, and haploid fruiting could be used in breeding work. While the tetrasporic French isolates appear to be incompletely interfertile with bisporic strains, hybridizations were performed, which showed that the tetrasporic trait is transmissible and therefore can potentially be used in breeding work. Without this trait, isolation of 50 homokaryons could require 6 months of effort (14). The Bsn-t allele from California strains has already been introduced into bisporic strains by backcrossing (11). To exploit heterosis in American-European hybrids, both American and European tetrasporic strains could be used to produce highly recombined homokaryons. Haploid fruiting could help to select homokaryotic offspring and could be used to study clonal variability and the genetics of fruiting body morphogenesis or susceptibility to diseases.