The B chromosomes in Brachycome.

This review presents a historical account of studies of B chromosomes in the genus Brachycome Cass. (synonym: Brachyscome) from the earliest cytological investigations carried out in the late 1960s though to the most recent molecular analyses. Molecular analyses provide insights into the origin and evolution of the B chromosomes (Bs) of Brachycome dichromosomatica, a species which has Bs of two different sizes. The larger Bs are somatically stable whereas the smaller, or micro, Bs are somatically unstable. Both B types contain clusters of ribosomal RNA genes that have been shown unequivocally to be inactive in the case of the larger Bs. The large Bs carry a family of tandem repeat sequences (Bd49) that are located mainly at the centromere. Multiple copies of sequences related to this repeat are present on the A chromosomes (As) of related species, whereas only a few copies exist in the A chromosomes of B. dichromosomatica. The micro Bs share DNA sequences with the As and the larger Bs, and they also have B-specific repeats (Bdm29 and Bdm54). In some cases repeat sequences on the micro Bs have been shown to occur as clusters on the A chromosomes in a proportion of individuals within a population. It is clear that none of these B types originated by simple excision of segments from the A chromosomes.

The genomic complexity of micro B chromosomes of Brachycome dichromosomatica.

A major sequence component of the micro B chromosome of Brachycome dichromosomatica (2 n=4) is the tandem repeat Bdm29, which was found by in situ hybridisation to be distributed along the entire length of the chromosome. A high copy number of this sequence does not occur as a regular feature of the A chromosomes in this species but it was found in infrequent individuals in two wild populations that were analysed. In these instances Bdm29 is localised within heterochromatic, polymorphic segments on the long arm of chromosome 1. The origin of the micro B chromosomes was investigated by determining whether they are related to this A chromosome polymorphism by simple excision and/or integration. Results obtained by using Bdm29, together with a newly isolated repeat sequence, Bdm54, and a number of other sequences known to occur on the micro B chromosome, as probes in in situ hybridisation and Southern analysis demonstrated that the formation of micro B chromosomes is a complex multistep process. The observation that the genomic organisation of the micro B chromosome is unlike anything found on the A chromosomes precludes their origin by simple excision and also indicates that micro Bs do not integrate directly into the A complement to form polymorphic heterochomatic segments.

Cloning and characterisation of polymorphic heterochromatic segments of Brachycome dichromosomatica.

After selective enrichment and differential hybridisation of Cot-1 DNA fractions of plants with and without polymorphic heterochromatic segments, a repetitive sequence (called Bds1) specific to the polymorphic chromosome segments of Brachycome dichromosomatica (Brachyscome dichromosomatica) was isolated. A single repeat unit of Bds1 is 92 bp long and is organised in tandem arrays at three different polymorphic segment sites on the chromosomes of cytodeme A2. Although all three sites showed extensive polymorphism between plants, the karyotypes of all analysed mitotic root cells were stable within a single plant. Electron microscopy revealed heavily condensed chromatin structures at the most obvious polymorphic site. The mechanisms that generate and maintain the observed chromosome structure polymorphisms are discussed.

Ribosomal RNA genes specific to the B chromosomes in Brachycome dichromosomatica are not transcribed in leaf tissue.

Ribosomal RNA genes are present near the end of the short arm and, to a lesser extent, near the centromere of the B chromosomes of some populations of Brachycome dichromosomatica. The internal transcribed spacer (ITS2) was amplified by PCR from total leaf DNA using primers within the conserved regions encoding the 5.8S and 25S stable rRNA species. Comparison of PCR amplified ITS2 sequences from several individual plants without B chromosomes with corresponding sequences derived from microdissected B chromosomes revealed two consistent differences between the rDNA of A and B chromosomes. One of these differences produced an SfcI restriction site that was present only in the ITS2 of the B-chromosome rDNA. Amplification by PCR of ITS2 from total genomic DNA from plants with and without B chromosomes showed an additive relationship between the amount of PCR product containing the SfcI site and the number of B chromosomes present. Quantitative analysis indicated that the proportion of total nuclear rDNA present on a single B chromosome varied between 2 and 4% in different A chromosome backgrounds. Similar experiments, with appropriate positive and negative controls, using reverse transcriptase PCR of the equivalent region within the 40S precursor rRNA, suggested that the B-chromosome rDNA was not transcribed. Similarly, PCR of reverse transcribed total RNA from plants containing B chromosomes using primers specific for the B chromosome ITS2 was unable to detect a transcript from the B chromosome.

Differences of histone H4 acetylation and replication timing between A and B chromosomes of brachycome dichromosomatica.

Differences are demonstrated between A (transcriptionally active) and B (transcriptionally inactive) chromosomes that are characterized by a different level of histone H4 acetylation and a different timing of DNA replication. These differences between the chromatin of A and B chromosomes were found after immunolabelling of chromsomes of Brachycome dichromosomatica with antibodies specific for different acetylated forms (lysine 5, 8, 12 and 16) of histone H4. In contrast to the A chromosomes, which are labelled brightly in their entirety, the transcriptionally inactive B chromosomes are faintly labelled with antibodies against H4Ac5 and H4Ac8. No such difference between the chromosomes is found after immunostaining with the other antibodies H4Ac12 and H4Ac16. Analysis of DNA replication timing in root-tip meristems suggests that B chromosomes are labelled late in S-phase compared with A chromosomes. After C-banding the B chromosome appeared to have a similar amount of heterochromatin to the A chromosomes.

A repetitive DNA sequence common to the different B chromosomes of the genus Brachycome.

Dot-like micro B chromosomes of Brachycome dichromosomatica were analysed for their sequence composition. Southern hybridization patterns of a total micro B probe to genomic DNA from plants with and without micro Bs demonstrated that the micro Bs shared sequences with the A chromosomes. In addition to telomere, rDNA and common A and B chromosome sequences, a new B-specific, highly methylated tandem repeat (Bdm29) was detected. After in situ hybridization with Bdm29 the entire micro B chromosome was labelled and clustering of the condensed micro Bs could be observed at interphase. A high number of Bdm29-like sequences were also found in the larger B chromosomes of B. dichromosomatica and in other Bs within the genus Brachycome.

The molecular organisation of a B chromosome tandem repeat sequence from Brachycome dichromosomatica.

A high copy, tandemly repeated, sequence (Bd49) specific to the B chromosome and located near the centromere in Brachycome dichromosomatica was used to identify lambda genomic clones from DNA of a 3B plant. Only one clone of those analysed was composed entirely of a tandem array of the B-specific repeat unit. In other clones, the Bd49 repeats were linked to, or interspersed with, sequences that are repetitious and distributed elsewhere on the A and B chromosomes. One such repetitious flanking sequence has similarity to retrotransposon sequences and a second is similar to chloroplast DNA sequences. Of the four separate junctions analysed of Bd49-like sequence with flanking sequence, three were associated with the same A/T-rich region in Bd49 and the fourth was close to a 25 bp imperfect dyadic sequence. No novel B-specific sequences were detected within the genomic clones.

Organisation and origin of a B chromosome centromeric sequence from Brachycome dichromosomatica.

Brachycome dichromosomatica is an Australian native daisy that has two pairs of A chromosomes and up to three B chromosomes in some populations. A putative B-specific tandem repeat DNA sequence (Bd49) was isolated previously. Here we describe further characterisation of this sequence and investigate its possible origin. Southern analysis showed that all individual B chromosomes examined have highly methylated tandem repeats of Bd49 but differences in banding pattern for distinct B isolates suggested that the sequence is in a state of flux. Using in situ hybridisation, the sequence was shown to be located at the centromeric region of the B chromosome. Southern analysis of genomic DNA with Bd49 demonstrated that multiple copies of the sequence exist in the genomes of B. eriogona, B. ciliaris, B. segmentosa and B. multifida (none of which have B chromosomes) whereas other species tested (including 0B plants of B. dichromosomatica and 0B and +B B. curvicarpa and B. dentata) have few or no copies. Genomic clones and Bd49-like sequences derived by the polymerase chain reaction (PCR) were obtained from five species but determination of phylogenetic relationships within the genus and inference as to the possible origin of the B chromosome were problematic because of extensive intragenomic heterogeneity of the sequences.

Ribosomal RNA genes and the B chromosome of Brachycome dichromosomatica.

Fluorescence in situ hybridization (FISH) with biotinylated rDNA revealed the presence of an rRNA gene cluster on both the A and B chromosomes of Brachycome dichromosomatica, an Australian native ephemeral plant of the arid regions of south-eastern Australia. This species contains only two pairs of A chromosomes and up to three B chromosomes. The regular attachment of the B chromosome to a nucleolus suggests that these ribosomal RNA genes are transcribed. Southern hybridization of DNA from 0B and +B plants digested with a variety of restriction enzymes indicates that the rRNA genes on the A and B chromosomes are the same in sequence and methylation status.

Quantitatively determined self-incompatibility. 5. Detection of multi-locus systems.

Multi-locus self-incompatibility systems may be distinguished from single-locus systems by reciprocal differences in backcrosses and between crossed progeny of individual clearly compatible crosses. Such crosses are extremely laborious, so other methods have been suggested. In this note, it is shown that the coefficient of crossability is not a useful discriminant of self-incompatibility, as indeed should be expected from the properties of multi-locus systems, and that linkage methods are also unlikely to be successful. Until more self-incompatibility genes have had their sequences characterised, there is no substitute for the traditional genetical methods.

Why are there so few homomorphic multi-locus sporophytically determined self-incompatibility systems?

Homomorphic multi-locus sporophytically determined self-incompatibility systems are much rarer than multi-locus gametophytic systems. This note examines some of the possible reasons for this disparity and concludes that, while each additional locus in a gametophytic system allows increased crossing among related plants as well as a lower mutation rate to maintain a given level of variability, the same conclusion cannot be drawn for sporophytic systems.

Quantitatively determined self-incompatibility. IV. Pollination and seed set in Borago officinalis.

The outcomes of sequential double pollination, mixed double pollination and single pollination are compared. Single pollination leads to lower seed set than double pollination. Systematic differences between female genotypes are shown to be possible. It is also shown that failure to set seed is generally due to pre-zygotic maternal factors.

A sequence specific to B chromosomes of Brachycome dichromosomatica.

Supernumerary B chromosomes represent one of many causes of numerical chromosome variation that exist in higher plants and animals. Sequences of DNA unique to B chromosomes of Brachycome dichromosomatica were enriched prior to cloning and resultant clones hybridizing only to plants containing B chromosomes were further investigated. Sequences of DNA that were characterised include members of a family of 176-bp tandem repeats that are specific to the B chromosomes of B. dichromosomatica, an annual Australian native plant species with only two pairs of A chromosomes and up to three dispensable B chromosomes. Sequence analysis of these six related clones indicated that some regions of the sequence are more highly conserved than others or, alternatively, that some adenine residues at the NdeII site are methylated. The repeat is homologous to DNA from Brachycome ciliaris var. languinosa but not to DNA from other related taxa growing in the vicinity of the B. dichromosomatica populations.

Quantitatively determined self-incompatibility : 3. Genetical variability in Borago officinalis.

It is shown by simulation that a hypothetical multilocus, quantitatively determined self-incompatibility system, whether gametophytic or sporophytic, should maintain variability in small populations at a higher level than would panmixia. Studies of more than 20 isozyme loci show that borage has almost no variability.

The heritability of fitness: some single gene models.

Because directional selection exhausts additive-genetic variance, it is frequently claimed that the heritability of fitness should be very close to zero. However, mutation-selection balance generates a certain amount of additive-genetic variance, so that even parent-offspring measures of heritability may be greater than zero at equilibrium. Intra-generation heritability may also be non-zero, providing the potentials for genetic change following environmental change.

Quantitatively determined self-incompatibility : 2. Outcrossing inBorago officinaux.

It has been claimed that Borage (Borago officInalis L.) has a multifactorial self-incompatibility system. Such systems may have a high level of ineffective pollination, and we show that this is the case in borage. The ranking of seed set from highest to lowest is as follows: bee-pollination; natural pollination in the absence of bees; artificial cross-pollination between unrelated plants; artificial cross-pollination between related plants; artificial self-pollination. In diallel crosses, significant parental effects were detected but no consistent patterns of seed set, which suggest a simple self-incompatibility system, were detected. The level of outcrossing with natural pollination was very variable but greater than 50%. Thus, there appears to be no straightforward self-incompatibility system in borage.

Quantitatively determined self-incompatibility : 1. Theoretical considerations.

It has been reported that incomplete self-incompatibility could be determined in Borago officinalis by many genes. Simple ten-gene models for such enforced cross-fertilization have been developed and their properties examined by computer simulation. Mutation rates necessary to maintain a given level of variability in small populations are high, as already determined theoretically for oligogenic self-incompatibility systems. However, the extent of ineffective pollination is very much greater in the ten-gene system. This finding may be verifiable in borage if it is indeed self-incompatible.

Stability of self-incompatibility systems.

Multi-locus self-incompatibility systems offer few obvious adaptive advantages to the species possessing them. However, the gametophytic system's independent gene action allows the separate genes in a two gene system to behave as if they were individually not involved in a systematic disruption of panmixia. Under such circumstances, fixation of one of the two genes is readily obtained if an allele possesses a selective advantage. The resulting single gene system (the classic Nicotiana system) is then resistant to disruption, except by genes which allow selfing, which rapidly reach fixation.

Linkage disequilibrium and gametophytic self-incompatibility.

The approach to linkage equilibrium of a locus linked to the locus determining gametophytic self-incompatibility (S) is considered. For the simplest case of three alleles at the S locus and two at the linked locus it is necessary to consider 3 measures of linkage disequilibrium. These are found to approach their equilibrium value of zero in one of three ways: 1) steadily declining to zero; 2) oscillating as decline proceeds; 3) a combination: 2) followed by 1). Linkage equilibrium may be established before genotype frequencies reach their expectation under random crossing. Earlier studies (Li 1951; Moran 1962) of the approach to S allele equilibrium have been based on the assumption that all types of pollen take part in fertilizations equally frequently. Such an assumption leads to simpler expressions for changes in S gene frequencies but is extremely unrealistic and, in particular, leads to a different rate of approach to equilibrium from the more comprehensive model. It is shown that even in the absence of selection it is not possible to predict the equilibrium gene frequency of a linked locus until S allele equilibrium is reached. This frequency may be either higher or lower than that calculated from a gene count in the starting genotype pool. However, these two gene frequencies may stabilize long before linkage equilibrium is achieved. An examination of selection against one genotype at the linked locus is undertaken. If linkage is complete, lethality can be less effective at reducing the gene frequency than is less intense selection (in only a few generations of selection). Here too linkage equilibrium may be established with selection still effective in bringing about a decline in gene frequency. An examination of the analysis and conclusions of Rasmuson (1980) shows that because these were based on the inadequate formulae previously discussed and exclude phenomena discussed above, they are misleading. The possibility of a gametophytic self-incompatibility system providing a sufficient condition for the sheltering of lethals in the absence of the condition of complete linkage to the S locus (r=0) is shown to be unlikely.

Fluctuations in heteromorphic self-incompatibility systems.

The genetic determination of heterostyly is briefly reviewed and experimental data are presented on the inheritance of tristyly in Oxalis compressa. It is shown that the model appropriate to Lythrum salicaria, of two diallelic loci, is inadequate to explain segregation patterns found in O. compressa, especially the reversal of dominance of the short phenotype. An extended model having an additional allele at the short locus, and a separate modifier gene, is presented and its deterministic genotype frequency dynamics examined numerically. It is shown that fixation of these extra genes or alleles is unlikely. Problems of testing the extended model are considered.