SEARCH

SEARCH BY CITATION

Keywords:

  • Agaricus bisporus;
  • Breeding system;
  • Genetic linkage;
  • Map expansion

Abstract

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

A previous map of the genome of a hybrid strain which had European parents belonging to the secondarily homothallic fungus Agaricusbisporus var. bisporus appeared to be unusually compact, with a particularly recombophobic segment in the central part of chromosome I. A new map of this segment was constructed based on allelic segregations among 103 homokaryotic offspring of an A. bisporus hybrid between a European parent of the var. bisporus and a Californian parent of the heterothallic var. burnettii. Markers completely linked on the previous map were distributed along 28 cM in the new map. These results suggest that the greater recombination rate could be correlated with the outbreeding behaviour of the var. burnettii.


1Introduction

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

Physical and sexually heritable relationships among genetic loci can be affected by altered recombinational behaviour as well as by processes of genomic reorganization. Chromosomal inversions affect gene order within linkage groups and recombination frequencies between pairs of loci which bracket breakpoints. Translocations can disrupt prior linkage relationships while creating novel ones. Changes in local or general recombination behaviours can result in map expansion or contraction.

A cross (AG 93b) of the mushroom Agaricus bisporus (Lange) Imbach was found to have 13 electrophoretically polymorphic chromosomes, of which at least nine were marked with multiple, linked markers [1]. The map of this genome appeared to be unusually compact (approx. 49–96 kbp per cM). Several regions of little or no recombination were noted, particularly in central regions of the chromosomes. The most substantial of these occurred on chromosome I (length approx. 4.0 Mbp), and included five markers for which no recombination was observed among 52 meiotic products. Subsequently, the mating type (MAT) locus was also mapped in AG 93b to this non-recombining cluster of markers [2]. This cross appears to be entirely of European ancestry [1, 3, 4] and belongs to the bisporic var. bisporus.

Further mapping efforts have been hampered by the difficulty of obtaining haploid meiotic offspring from this species. In a bisporic sporocarp of A. bisporus most of the spores are heterokaryotic and typically receive two non-sister postmeiotic nuclei which complement each other at the mating type locus and therefore produce fertile heterokaryons [1]. The 52 homokaryotic offspring of AG 93b required about 6 person months of effort to isolate [5]. We recently reported on the existence of tetrasporic wild strains from the Sonoran desert of California which belong to the var burnettii Kerrigan and Callac [6]. These strains and the first generation of intervarietal hybrid strains between them and bisporic strains of var. bisporus produce predominantly homokaryotic offspring [7, 8]. In the present study 103 homokaryotic offspring of a cross (JB 3-83×U 1-7) between a bisporic, European var. bisporus strain (U 1) and a tetrasporic strain of var. burnettii (JB 3) were scored for five markers which map to the segment of chromosome I that was non-recombining in AG 93b [1, 2]. We report here that linkage was conserved between the two maps, but that crossing over was much more frequent in JB 3-83×U 1-7. Moreover, the map presented here includes two linked loci controlling reproductive traits: the MAT locus [2], and the basidial spore number (BSN) locus [8].

2Materials and methods

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

2.1Hybrid strain JB 3-83×U 1-7

The intervarietal hybrid JB 3-83×U 1-7 was previously constructed by crossing homokaryons JB 3-83 and U 1-7 which issued, respectively, from the heterokaryons JB 3 and U 1 [6]. The MAT genotypes of JB 3-83, U 1-7, and JB 3-83×U 1-7 were previously determined as being Mat-6, Mat-2, and Mat-2/6, respectively [7]. The basidial spore number trait is primarily determined by the BSN locus [8] linked to the MAT locus on chromosome I. It was previously shown that the hybrid (JB 3-83×U 1-7) had received a dominant allele (Bsn-t) from its tetrasporic parent, and a recessive allele (Bsn-b) from its bisporic parent. Therefore, this hybrid was phenotypically predominantly tetrasporic and had the BSN-t/b genotype [8].

2.2Homokaryotic offspring of JB 3-83×U 1-7

Among 200 single spore isolates of the hybrid JB 3-83×U 1-7, chosen at random, 153 were homokaryotic according to multiple tests (multilocus allozyme genotype tests, mating tests, and mycelial growth tests) [7]. To determine the BSN genotypes, the homokaryons had to be crossed with a compatible homokaryotic tester. However, 40 of them which gave reactions with this tester, that we interpreted as false negatives, could not be used. Among the 113 remaining homokaryons, 103 (randomly chosen) were previously used to analyse allelic segregation at the MAT and BSN loci [8]. In that study, for four homokaryons, the BSN genotype could not be determined with at least 95% confidence. Among the 99 remaining homokaryons, the parental genotypes [Mat-6, Bsn-t] and [Mat-2, Bsn-b] were inherited by 46 and 37 homokaryotic offspring, respectively. The non-parental genotypes [Mat-2, Bsn-t] and [Mat-6, Bsn-b] were respectively inherited by 7 and 9 homokaryons. The proportion of recombinants was therefore 16/99, or 0.16%. Stastistical tests have shown strong evidence for genetic linkage of MAT and BSN[8].

2.3Molecular markers

Segregation at four sequence characterized amplified region (SCAR) [9] markers was analysed in the sample of 103 homokaryons. For three of the markers (PR5, PR6, and PR7), the primers were derived from the sequences of the restriction fragment length polymorphism (RFLP) probes of three loci previously localized on chromosome I[1]. Sequences of these primers, provided by J. Anderson (University of Toronto), are given in Table 1. For the remaining one (PR2), the primers were derived from the sequence of the RFLP probe Sd19 obtained at INRA-CTC from our genomic DNA library constructed with EcoRI fragments of U1-7 (unpublished data). DNA extraction and PCR reactions were conducted as previously described [8]. Restriction enzymes and codominant alleles of the two homokaryotic nuclear constituents of the hybrid strain JB 3-83×U 1-7 are given in Table 1. Segregation data were analyzed by using MAPMAKER V 3.0 [10] with a LOD threshold of 3.0 for linkage.

Table 1.  Primer sequences, restriction enzymes, and DNA fragment sizes of four SCAR markers
SCAR lociRFLP locibPairs of primer sequences 5′-3′Restriction enzymesDNA fragment sizesa
Allele 1 (U 1-7)Allele 2 (JB 3-83)    
  1. aApproximate sizes in bp.bPrimers of the SCAR markers were derived from the sequences of the RFLP probes of these loci.

PR2Sd19GAAGGAAAGTGGGATGATCAARsaI430+360+90520+360
  CATCGGTCTGCCAACAATACA   
PR5P1N148ACAGCAGCGATCGAGTATGGHaeIII280+210+220+210+
  AGATGCTCGTCATGATCACG 150+55150+60+55
PR6P1N150CAATCTCAAGCTTGCCTGGHaeIII630+200550+280
  AGGTGACATGTCAGAAGCGC   
PR7P1N17TTACCACAAGCGGATACTTGGAluI435+420435+300+120
  TTATGCCGAGGAAGTTGGC   

2.4Allozyme marker

Segregation among a sample of 27 of the 103 homokaryons was analysed at the carboxypeptidase locus PEP2[1, 11]. These homokaryons were selected to represent each of the genotypic groups resulting from segregation at the SCAR loci (see Table 2). Homokaryons grown in PDY broth for about 3 weeks were harvested by filtration and frozen in liquid nitrogen until used. Samples were ground, centrifuged in 1.5 ml tubes at 14.000 rpm, and the supernatant removed for electrophoresis, which was carried out on 1.5 mm thick vertical polyacrylamide slab gels of 5% concentration, bis=2.75% T, in a discontinuous Tris-glycine buffer system, on a Hoeffer SE 600 apparatus at 200 V for 3.5 h. Staining for carboxypeptidase activity employed leucyl-leucyl-leucine as a substrate [1, 6]. The linked locus PEP1[1] was homoallelic and thus uninformative in this cross.

Table 2.  Genotypes of 103 homokaryotic single spore isolates from JB 3-83×U 1-7
GenotypesNumbers of homokaryonsNumber of crossovers
PEP2BSN PR5 PR2 PR6MAT PR7  
  1. Dashes indicate missing data. For BSN, four data are missing (due to intermediate values for average spore number), and for PEP2, parenthetical values are the numbers of tested homokaryons (total of 27). For MAT and the other markers, all the 103 genotypes are determined. The two first genotypes are equivalent to the JB 3-83 and U 1-7 homokaryotic midparents. Markers are in the same order as on the most likely map and X indicates the presumed locations of the crossovers corresponding to this map.

3 (5)t 2 2 26 243 (=JB 3-83)0
4 (2)b 1 1 12 135 (=U 1-7)0
4 (1) 1 1 12 110
 1 1 12 110
4 (7)bX2 2 26 271
4 (2)X2 2 26 221
3 (4)tX1 1 12 151
4 (1)b 1X2 26 211
3 (1)t 2X1 12 111
b 1 1 12X221
3 (1)t 2 2 26X121
3 (1)tX1 1X26 212
4 (1)b 1 1X26X112
3 (1)t 2 2X12X212

3Results

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

Genotypes of the 103 homokaryons are given in Table 2. Mendelian segregation ratios were observed for all the SCAR loci, BSN, and MAT; segregation at the latter locus (46:57) gave the greatest χ2 value of 1.18. All markers were found to constitute a single linkage group, which corresponds to a contiguous segment of chromosome I (Fig. 1), [1]. Among the 103 homokaryons, 23 (or 22%) were recombinant in this segment. Among the 80 remaining ones, 43 (54%) had the JB 3-83 midparental genotype, and 37 (46%, including two incomplete genotypes) had the U 1-7 midparental genotype. For all the 103 homokaryons, MAT and PR6 cosegregated completely. In a sample of 24 homokaryons for which both BSN and PEP2 genotypes were determined, these two markers also cosegregated completely.

image

Figure 1. Partial genetic map of chromosome I of A. bisporus. This map was generated from the analysis of the offspring of JB 3-83×U 1-7. Genetic distances are in cM. The log-likelihood is −69.64.

Download figure to PowerPoint

In the analysis of 52 homokaryotic offspring from AG 93b [1], no crossing over was observed between the markers PEP2, P1N148 (corresponding to PR5), P1N150 (corresponding to PR6), P1N17 (corresponding to PR7), and P1N31. The hypothesis of equality of the proportions of recombinant homokaryons in the two offspring (observed proportions: 23/103 here, and 0/52 for the previous one) was tested by using the hypergeometric distribution: P=3.23 10−5<α/2=0.025. Therefore, assuming that the two analysed samples are representative of the two sets of offspring, the hypothesis of equality of the two proportions must be rejected.

Using MAPMAKER with a LOD threshold of 3.0, linkage between all the loci was confirmed. The most likely genetic map is presented in Fig. 1: about 29 cM separate PEP2 from PR7 and a total of 26 crossovers were observed. However, one can note in Table 2 that, for this map, the three crossovers between PR2 and PR6/MAT were all associated with a second crossover in a flanking segment. Such a phenomenon has been previously observed [1, 8] and implies that single crossovers could be lethal. Because a great deficit of single crossovers could be envisaged, the second best map given by MAPMAKER, although having a relative log-likelihood of −2.8 and a greater number of double crossovers (5), must not be completely excluded. In this map, the positions PR6 and PR7 are inverted, about 31 cM separate PEP2 from PR6 and a total of 28 crossovers were observed.

4Discussion

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

The correspondence of linkage in maps of chromosome I based on two distantly related pedigrees furnishes a preliminary indication that the organization of the nuclear genome may be conservative over most or all lineages of A. bisporus. However, because gene order could not be determined in the previous map, linkage is here simply defined as `joint segregation'. The European and California Desert populations are genotypically very divergent [3, 4, 6], which we interpret to indicate the occurrence of an ancient isolating process or event. Both populations are genotypically diverse, exhibiting no obvious sign of a founder effect. Genomic organization is therefore more highly conserved than is the primary structure of the DNA.

The map expansion in JB 3-83×U 1-7 is very distinct compared to what was observed in AG 93b. We should not overgeneralize based on these two maps, however, some testable hypotheses may be noted. First, there may be a greater overall rate of crossing over in the heterothallic var. burnettii strains relative to the European var. bisporus strains. This is consistent with preliminary observations on other strains ([5] and unpublished data), however, many more data are needed. If correct, it would imply that recombination is favored in association with the heterothallic, outbreeding behaviour of var. burnettii[7, 8]. This supports a recent hypothesis in which the suppression of recombination would be adaptative in the secondarily homothallic, inbreeding life cycle of var. bisporus[1]. For this variety, through successive generations via heterokaryotic spores, deleterious alleles may be accumulated. A low rate of recombination maintains high levels of heterozygosity, and therefore, by complementation, high viability and fitness, among most offspring. Such a suppression of recombination is not required for the heterothallic var. burnettii for which most loci are exposed to selection during the homokaryotic phase of each outbred generation. In this hypothesis the expanded map of the intervarietal hybrid would suggest that normal frequencies of crossing over are a dominant trait from the var. burnettii parent.

There were indications of structural heterogeneity in chromosome I of AG 93b [1, 2], which might render recombination in this region lethal for some or all offspring of the AG 93b heterokaryon. Our observations on crossovers in the PR2-PR6/MAT-PR7 segment may also indicate a similar but more localized area of structural heterogeneity. An inversion in a segment including PR6/MAT or PR7 could explain such a deficit of single crossovers which, in the inverted segment, would give rise to inviable recombinants, while double crossovers would give rise to viable progeny [12]. In any case, it is not yet possible to state whether the map expansion we have observed in the more outbreeding-oriented pedigree is segment specific or genome wide.

The fact that in both pedigrees MAT was, or appeared likely to be, within or near a non-recombining segment (of chromosome I) may not be coincidental; maintenance of heterozygosity at MAT in heterokaryotic offspring, however common or uncommon they may be, should be adaptative and could be maintained by such a mechanism, although lethality would seem a drastic penalty to pay for the loss of an allele at this locus.

We note that we presently have no data on the 40 homokaryons for which the BSN genotype could not be determined in the initial analysis (due to unsuccessful mating, see Section 2). This tendency might be associated with the inheritance of a recombinant genotype which, hypothetically, might be underrepresented in the 67% of homokaryotic offspring which we characterized. We should also note the possibility that AG 93b might not have been representative of var. bisporus. One of the homokaryons used to construct this hybrid had been treated with X-radiation [13].

The diversity that may exist in the genome of A. bisporus will become apparent as more strains are mapped. For the var. bisporus, besides the European population (more particularly and recently investigated in France [14]), there exist at least three other geographically isolated populations, including two from North America [4, 6]. The difficulty of obtaining homokaryotic offspring from this variety will delay these studies. However, the multilocus allozyme genotypic tests previously used to select homokaryons among offspring [1, 5, 7] can be now favourably replaced by the use of a SCAR marker (PR6) tightly linked to MAT.

In other respects, with sufficient codominant markers, it will become practical to determine recombination rates among heterokaryotic offspring of this secondarily homothallic variety, incidentally mapping the centromere of this or other chromosomes. The SCAR markers described here, developed at the University of Toronto and at INRA-CTC, are suitable for such studies. If frequencies of crossing over are comparable between heterokaryotic and homokaryotic offspring (but see [1, 10, 15]) then it will be feasible to survey linkage relationships broadly over the species.

Acknowledgements

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References

The contributions of S. Rakotonirainy (IGM, University Paris XI Orsay), S. Granit (INRA), M. Barrey (CTC), A.J. Velcko Jr. (Sylvan), and J. Anderson (University of Toronto) are greatly appreciated. In France, this research was supported by INRA and CTC under joint contract. Financial support from the Ministère de l'Agriculture is also acknowledged.

References

  1. Top of page
  2. Abstract
  3. 1Introduction
  4. 2Materials and methods
  5. 3Results
  6. 4Discussion
  7. Acknowledgements
  8. References
  • 1
    Kerrigan, R.W., Royer, J.C., Baller, L.M., Kohli, Y., Horgen, P.A. and Anderson, J.B. (1993) Meiotic behavior and linkage relationships in the secondarily homothallic fungus Agaricus bisporus. Genetics 133, 225236.
  • 2
    Xu, J., Kerrigan, R.W., Horgen P.A. and Anderson, J.B. (1993) Localization of the mating-type gene in Agaricus bisporus. Appl. Env. Microbiol. 59, 30443049.
  • 3
    Kerrigan, R.W., Carvalho, D.B., Horgen P.A. and Anderson, J.B. (1995) Indigenous and introduced populations of Agaricus bisporus, the cultivated button mushroom, in eastern and western Canada: implications for population biology, resource management, and conservation of genetic diversity. Can. J. Bot. 73, 19251938.
  • 4
    Kerrigan, R.W., Billette, C., Callac, P. and Velcko, A.J., Jr. (1996) A summary of allelic diversity and geographical distribution at six allozyme loci of Agaricus bisporus. In: Mushroom Biology and Mushroom Products. Proceedings of the Second International Conference (Royse, D.J., Ed.), pp. 25–35. University Park, PA.
  • 5
    Kerrigan, R.W., Baller, L.M., Horgen P.A. and Anderson, J.B. (1992) Strategies for the efficient recovery of Agaricus bisporus homokaryons. Mycologia 84, 575579.
  • 6
    Callac, P., Billette, C., Imbernon, M. and Kerrigan, R.W. (1993) Morphological, genetic, and interfertility analyses reveal a novel, tetrasporic variety of Agaricus bisporus from the Sonoran Desert of California. Mycologia 85, 335351.
  • 7
    Kerrigan, R.W., Imbernon, M., Callac, P., Billette, C. and Olivier, J.M. (1994) The heterothallic life cycle of Agaricus bisporus var. burnettii, and the inheritance of its tetrasporic trait. Exp. Mycol. 18, 193210.
  • 8
    Imbernon M., Callac, P., Gasqui, P., Kerrigan, R.W. and Velcko, A.J., Jr. (1996) BSN, the primary determinant of basidial spore number and reproductive mode in Agaricus bisporus, maps to chromosome I. Mycologia 88, 749761.
  • 9
    Paran, I. and Michelmore, R.W. (1993) Development of reliable PCR-based markers linked to downy mildew resistance genes in lettuce. Theor. Appl. Genet. 85, 985993.
  • 10
    Lander, E.S., Green P., Abrahamson, J., Barlow, A., Daly, M.J., Lincoln, S.E. and Newburg, L. (1987) MAPMAKER: an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 136, 14351446.
  • 11
    Spear, M.C., Royse, D.J. and May, B. (1983) Atypical meiosis and joint segregation of biochemical loci in Agaricus brunnescens. J. Hered. 74, 417420.
  • 12
    Sinnot, E.W., Dunn, L.C. and Dobzhansky, T. (1958) Principles of Genetics, 499 pp. McGraw-Hill, New York.
  • 13
    Raper, C.A., Raper, J.R. and Miller, R.E. (1972) Genetic analysis of the life cycle of Agaricus bisporus. Mycologia 64, 10881117.
  • 14
    Callac P. (1995) Breeding of edible fungi with emphasis on the variability among French genetic resources of Agaricus bisporus. Can. J. Bot. 73 (Suppl. 1), S980–S986.
  • 15
    Kerrigan, R.W., Velcko, A.J., Jr. and Spear, M.C. (1995) Linkage mapping of two loci controlling reproductive traits in the secondarily homothallic agaric basidiomycete Agaricus bisporus. Mush. Sci. 14, 2128.