Journal of Experimental Botany, Vol. 54, No. 380, pp. 11-23,
January 1, 2003
© 2003 Oxford University Press
Meiosis, recombination and chromosomes: a review of gene isolation and fluorescent in situ hybridization data in plants
Received 1 July 2002; Accepted 17 September 2002
1 Department of Biology, University of Leicester, University Road, Leicester LE1 7RH, UK
1 Fax: +44 (0)116 252 3330. E-mail: TS32{at}le.ac.uk
 |
Abstract |
|---|
 Top Abstract IntroductionMeiotic genes and gene...RecombinationSynaptonemal complexHomologous chromosome pairingHomology recognitionChromosome morphologyRepeated DNA sequencesSites of recombination and...ConclusionsReferences |
|---|
Evidence is now increasing that many functions and processes of meiotic genes are similar in yeast and higher eukaryotes. However, there are significant differences and, most notably, yeast has considerably higher recombination frequencies than higher eukaryotes, different cross-over interference and possibly more than one pathway for recombination, one late and one early. Other significant events are the timing of double-strand breaks (induced by Spo11) that could be either cause or consequence of homologous chromosome synapsis and SC formation depending on the organisms, yeast plants and mammals versus Drosophila melanogaster and Caenorhabditis elegans. Many plant homologues and heterologues to meiotic genes of yeast and other organisms have now been isolated, in particular in Arabidopsis thaliana, showing that overall recombination genes are very conserved while synaptonemal complex and cohesion proteins are not. In addition to the importance of unravelling the meiotic processes by gene discovery, this review discusses the significance of chromatin packaging, genome organization, and distribution of specific repeated DNA sequences for homologous chromosome cognition and pairing, and the distribution of recombination events along the chromosomes.
Key words: Centromere, chiasmata, cross-overs, genome organization, homologous chromosome pairing, repeated DNA sequences, synaptonemal complex.
|
Introduction |
|---|
|
TopAbstractIntroductionMeiotic genes and gene...RecombinationSynaptonemal complexHomologous chromosome pairingHomology recognitionChromosome morphologyRepeated DNA sequencesSites of recombination and...ConclusionsReferences |
|---|
Meiosis, the unique and essential part of the life cycle of all sexually reproducing organisms, is the process by which a diploid cell of the sporophyte gives rise to haploid cells which develop further to the gametophyte and gametes (for review see John, 1990). It involves two divisions that are linked together without any further DNA replication. While the second division resembles mitosis and segregates sister chromatids, the first division is unique involving the pairing of homologous chromosomes and their subsequent segregation (see Armstrong and Jones, 2003). This process is accompanied by the normally random disjunction of parental chromosomes and recombination of chromosomes and genes so the haploid cells have different and new combinations from those of the organism’s parents giving rise to much of nature’s diversity. Between different organisms, there are many remarkably conserved features of meiosis while at the same time showing striking differences; however, all involve the following most critical events of meiosis: (1) initiation; (2) cognition and pairing; (3) synapsis and the synaptonemal complex; (4) recombination; (5) cohesion and segregation; and (6) tetrad formation and gamete development.
This review discusses some aspects of these important events and shows how gene discovery, in combination with cell biological observations, can lead to a comprehensive picture of meiosis.
|
Meiotic genes and gene isolation |
|---|
|
TopAbstractIntroductionMeiotic genes and gene...RecombinationSynaptonemal complexHomologous chromosome pairingHomology recognitionChromosome morphologyRepeated DNA sequencesSites of recombination and...ConclusionsReferences |
|---|
Meiotic gene discovery has been led by studies in budding yeast, Saccharomyces cerevisiae, over the past decades, but with the use of insertional mutagenesis and the availability of whole genome DNA sequences, knock-out mutants and differential screening, many homologues to yeast genes have now been identified in other organisms including Schizosaccharomyces pombe, Caenorhabditis elegans, Drosophila melanogaster, Arabidopsis thaliana, mouse, and man. About 200 genes specific to meiosis and gamete formation had been identified by classical methods by 2000 (Primig et al., 2000). Probably as many as 1500 genes show altered gene expression as analysed by microarrays (Chu et al., 1998; Primig et al., 2000; Andrews et al., 2000; Reinke et al., 2000) estimating the total of core genes specific for meiosis at 300 and those specific for sporulation/gametogenesis at 600.
Microarray analyses study the expression of many thousands of genes simultaneously and semi-quantitatively, giving information not only about genes that are induced during certain processes, developmental stages or disease, but also those that are repressed. However, in many cases only a global picture results and only genes with significant expression changes can be identified. Microarrays with all known yeast ESTs have been used to compare expression profiles between different yeast strains, growth conditions and times during sporulation (Chu et al., 1998; Primig et al., 2000): a large category of genes identified in the meiotically induced yeast cells were stress and metabolism-related genes because of the response to growth medium changes. Nevertheless, distinct temporal patterns of induction could be identified, and the transcript profiles correlated with the distribution of defined meiotic promoter elements and with the time of known gene functions.
Microarray hybridization generates huge amounts of data about expression profiles, and with its comparative and high throughput and exploitation capabilities this approach has revolutionized genomic analysis, but it rarely assigns function to individual, hitherto unknown, ESTs or genes, and finding these functions will be a massively difficult task. Rabitsch et al. (2001) have recently deleted 301 unknown ORFs in yeast that were preferentially expressed in meiotic cells according to published microarray gene expression data (Chu et al., 1998; Primig et al., 2000): of these, 33 genes showed a meiotic phenotype (three no replication, eight chromosome mis-segregation, four formation of abnormal asci, 15 no spore and ascus formation), 15 genes were essential for vegetative growth, while 253 genes showed no apparent phenotype illustrating the problem at hand.
|
Recombination |
|---|
|
TopAbstractIntroductionMeiotic genes and gene...RecombinationSynaptonemal complexHomologous chromosome pairingHomology recognitionChromosome morphologyRepeated DNA sequencesSites of recombination and...ConclusionsReferences |
|---|
Many genes have now been identified that are important for meiotic recombination and establishing their function, interaction and timing enables the building of a comprehensive picture of the meiotic recombination pathway (for a review see Lichten, 2001; Keeney, 2001; Villeneuve and Hillers, 2001). In general, recombination genes are highly conserved and direct homologues are found in diverse organisms, although some variation of function is possible; Table 1 lists plant homologues of some of the most important genes and gives a brief summary of their function. More detailed descriptions of these plant genes and comparisons of their function can be found in recent review articles by Bhatt et al. (2001) and Anderson and Stack (2002).
|
There is now no doubt that recombination is initiated by the enzyme Spo11 which generates double-strand breaks (DSBs) in yeast and possibly in most if not all organisms (see Lichten, 2001; Keeney, 2001). This is evident from the isolation of SPO11 homologues in most organisms that have genome projects or are commonly used to study meiosis: for example, A. thaliana (Table 1), mouse, man, D. melanogaster, and C. elegans. Borde et al. (2000) have shown elegantly that DNA replication is needed for DSBs and suggest that this causal link functions as a safety check to ensure that breakage is not induced before sister chromatids are available for repair, in case corresponding sequences on homologous chromosomes cannot be found. Further, DNA replication in pre-meiotic S-phase is important for Rec-8 dependent processes and the establishment of meiotic cohesin complexes (Watanabe et al., 2001).
The Spo11 protein has homology to Top6A, the catalytic subunit of an archaebacterial type 2 topoisomerase (Keeney, 2001; Villeneuve and Hillers, 2001) and is linked covalently to the 5' ends of the double-stand breaks forming a Spo11p-DNA intermediate. Spo11p is then removed and 3'OH single strand tails are formed by the Mre11–Rad50–XrS2 protein complex (Lichten, 2001). The action of Spo11 is possibly controlled region-by-region and tied to the establishment of the higher-order chromosome structure needed for DSBs (Borde et al., 2000; Baudat and Keeney, 2001). In A. thaliana two further Spo11 genes are found, AtSPO11-2 and -3 (Table 1) that interact with a subunit B of archaebacterial-type topoisomerase 6 found in plants, but not in other eukaryotes (Hartung and Puchta, 2001).
Despite the high sequence homology between Spo11 genes from different species, two distinct patterns of function are emerging (for review see Lichten, 2001; Villeneuve and Hillers, 2001). DSBs are found in mutants without SC formation in yeast, and spo11 mutants do not have SCs, indicating that DSBs are not only initiated before SC formation but are required for homology search and SC formation. This is also the case in mouse and Arabidopsis. By contrast, in D. melanogaster and C. elegans pairing centres initiate synapsis and double-strand breaks follow, indicating that DSBs are not needed for SC formation in these species. These differences raise the issue as to whether flies and nematode worms have evolved recombination mechanisms that differ substantially from those of yeast, mouse and man.
The recombination pathway in yeast continues as the generated single strand ends invade homologous sequences and the missing strands are resynthesized (Keeney, 2001). Homologues and orthologues of the yeast genes DMC1, RAD51, RAD 54, and MND1, and of the mammalian genes RPA and ATM, involved in these processes have been identified in plants (Table 1) with high conservation in possible gene structure and function. Dmc1, the meiotic homologue of Rad51, is important in early recombination events in eukaryotes by converting double-strand breaks into hybrid joint molecules. Dmc1 and Rad51 are homologous to RecA, the major protein that catalyses homologous pairing and DNA strand exchange in prokaryotes (Bishop, 1994). It has been isolated as cDNA and genomic clones from many plants (Table 1) and is a large gene, 8–11 kb with 12–15 exons in A. thaliana and H. vulgare (Klimyuk et al., 2000; Klimyuk and Jones, 1997) and shows not only meiotic, but also mitotic expression in several plants, possibly with differential splicing (Shimazu et al., 2001).
Rad51 proteins were found to form discrete nuclear foci from early zygotene to pachytene, to co-localize with lateral element proteins in yeast, mouse, rat, human, and lily meiosis (Barlow et al., 1997; Bishop, 1994; Terasawa et al., 1995) and are components of early recombination nodules (Anderson et al., 1997; Tarsounas et al., 1999). Rad51 has also been proposed to have a role in the homology search phase of chromosome pairing (Franklin et al., 1999). Using antibodies against Rad51, in early meiotic prophase nuclei from rye (Fig. 1a) distinct foci are seen that are localized near the DAPI positive subtelomeric heterochromatin indicating that the initiation of recombination and DNA sequence homology testing is concentrated near the ends of rye chromosomes.
|
Once DSBs have been repaired, double Holliday junctions form and can be resolved as either reciprocal recombination (cross-over) or gene conversion events (non-cross-over). It is now speculated that yeast has two pathways of recombination (Gilbertson and Stahl, 1996; Borts et al., 2000; Allers and Lichten, 2001; Hunter and Kleckner, 2001; Villeneuve and Hillers, 2001): cross-overs are formed by resolution of Holliday junction intermediates, dependent on Dmc1, while non-cross-overs are resolved by a second early mechanism of synthesis-dependent strand annealing. MSh4/5 mediated cross-over resolution generates interference when Type II intermediates are programmed in an SC-dependent manner.
In summary, many genes responsible for recombination have been identified and their interaction and timing are now well known. Most genes are highly conserved and direct homologues have been found in different organisms increasing the likelihood of evolutionarily conserved recombination machinery. However, some variation of function and multiple pathways seem possible.
|
Synaptonemal complex |
|---|
|
TopAbstractIntroductionMeiotic genes and gene...RecombinationSynaptonemal complexHomologous chromosome pairingHomology recognitionChromosome morphologyRepeated DNA sequencesSites of recombination and...ConclusionsReferences |
|---|
The SC is a tripartite proteinaceous structure that forms between homologous chromosomes as they synapse during zygotene (for reviews see von Wettstein et al., 1984; Gillies, 1985; Moens and Pearlman, 1988). Before pairing at leptotene, each single chromosome develops a proteinaceous axial core, or axial element, along its entire length to which the two sister chromatids are attached in a series of loops. When homologous chromosomes synapse these axial elements come together to become the lateral elements of the SC. Perpendicular, thin transverse filaments traverse the gap between the lateral elements. The width of the gap varies slightly between different species, usually being somewhere between 100–300 nm. Lying parallel between the lateral elements is the central element. This, together with the lateral elements, makes up the tripartite structure of the synaptonemal complex.
Early nodules (ENs), darkly stained ellipsoid bodies along the central element (Fig. 1b) are found at many axial element convergence sites where homologous chromosomes initiate synapsis (Albini and Jones, 1987; Anderson and Stack, 2002). ENs are thought to be involved in recognition and alignment of homologous chromosomes and in early events of recombination as they are the sites of recombination-related and homology searching proteins, such as Rad51-p/Dmc1-p, Rap-p and Atm-p (Anderson et al., 1997; Tarsounas et al., 1999; Anderson and Stack, 2002; Table 1). Most ENs are lost from the SCs by mid-pachytene and only those that are involved in reciprocal recombination events remain. They are now called late nodules (LNs) and have been shown to correspond in number and locations to cross-over events.
The synaptonemal complex is found in most eukaryotic organisms and shows remarkable morphological and structural conservation (Moens and Pearlman, 1988). Because of this universality and because some organisms and many mutants lacking SC formation do not have recombination, the SC has long been viewed as being central and essential for initiating and mediating recombination. However, recombination has now been shown to be initiated well before the formation of the SCs (see above) and there are a few cases where reciprocal recombination occurs without the presence of an SC, for example, S. pombe and Aspergillus nidulans. Accordingly, the importance of the SC has changed to reflect its role in the maturation of cross-over events into chiasmata, cross-over interference and chromatin cohesion, all events that happen later in meiotic prophase (Moens, 1994; Heyting, 1996; Zickler and Kleckner, 1999). Despite the conserved nature of the synaptonemal complex, only few components and genes have been isolated: in yeast, Zip1 and Red1, are associated with the central element and transversal filaments, and Hop1, is associated with the lateral elements (Zickler and Kleckner, 1999) corresponding to the mammalian proteins Scp1, Scp2 and Scp3 (Heyting, 1996). In plants, Asy1 and 2 show some similarities to Hop1, but might have different specificities (Caryl et al., 2000; Armstrong et al., 2002; Table 1). Generally, SC proteins show remarkably low conservation and sequence homology; anti-bodies against SC components of related species have shown limited and variable cross-reaction, and no cross-reactions have been reported between vertebrates, invertebrates, plants, and fungi (Moens, 1993). Similarly, cohesin proteins, responsible for the cohesion between sister chromatids and essential for the bipolar attachment of bivalents to the spindle that promotes proper disjunction (Nasmyth, 2001), have generally shown low levels of conservation.
|
Homologous chromosome pairing |
|---|
|
TopAbstractIntroductionMeiotic genes and gene...RecombinationSynaptonemal complexHomologous chromosome pairingHomology recognitionChromosome morphologyRepeated DNA sequencesSites of recombination and...ConclusionsReferences |
|---|
Fluorescent in situ hybridization allows direct observation of the behaviour of individual chromosomes during interphase (Lichter et al., 1988; Schwarzacher et al., 1989, 1992a); and the technology has recently been applied to studies of early meiotic prophase from yeast to plants and mammals, starting to assemble a comprehensive although contrasting picture of chromosome behaviour immediately before and at meiotic prophase. Comparing different models, it is becoming clear that they are not necessarily contradictory, but that different organisms exploit different mechanisms and that pairing is a multi-step and often multi-path process. A consequence is that compensation occurs when one route is blocked, making mutant approaches to gene discovery difficult.
In budding yeast, it has been observed that some homologues are paired, possibly via multiple interstitial interactions involving unstable side-by-side joints between intact DNA duplexes. These interactions occur at multiple sites along each chromosome pair in premeiotic and probably also in vegetative cells, implying that homologues align prior to synapsis (Loidl et al., 1994; Roeder, 1997; Zickler and Kleckner, 1998). Work on spo11 mutants in yeast now has shown that double-strand breaks are required for homology search and SC formation (Allers and Lichten, 2001; and see above), but whether DSBs actually promote homology search and the exact timing are still debated.
In man, Scherthan et al. (1996) postulated movements of centromeres and then telomeres to the nuclear envelope and subsequent bouquet formation as conserved motifs of the pairing process. At the onset of meiotic prophase, at leptotene, the compact and separate chromosome territories developed into long thin threads. Subsequently, telomeres moved towards the clustered site and produced numerous encounters among the now elongated chromosomes that are suggested to contribute to homology testing at exposed pairing sites. Dawe et al. (1994) and Bass et al. (1997) demonstrated that the homologous chromosomes of maize, similar to mouse and man, are apart when entering meiosis, but undergo a dramatic structural reorganization prior to synapsis at zygotene. Telomeres are localized peripherally, and cluster de novo before the initiation of pairing.
In the large cereal genomes, such as wheat (17 000 Mbp), homologous chromosomes associate during the interphase before leptotene. Total genomic DNA from rye, used as a probe for in situ hybridization identified the rye chromosome arm in a wheat–rye translocation line (T5AS·5RL) at meiotic prophase and the preceding interphase (Schwarzacher, 1997): three stages of pairing were identified. First, cognition occurs during the interphase before leptotene bringing the homologous chromosome domains into close proximity, apparently starting at the centromeres. Secondly, the chromosomes are organized into the meiotic chromosome ‘threads’, as recognized in conventionally stained preparations, and the homologous sequences align during leptotene and early zygotene. The third step of pairing, is the synapsis of the homologous chromosomes and the formation of the tripartite SC during zygotene after further alignment of homologous sequences. In many cases, the telomeric region was among the first to synapse, while the middle of the rye chromosome arm synapsed last, often with several small interstitial non-paired sites, confirming earlier studies using SC surface spreading in rye, wheat and other plants (Albini and Jones, 1987; Gillies, 1985; Jenkins and White, 1990).
Aragon-Alcaide et al. (1997) and Mikhailova et al. (1998) have also demonstrated that homologous chromosomes associate before leptotene and that centromeres are clustered at that stage. In lines with deletions of the homologous paring gene Ph1, centromere morphology is altered and centromeres fail to associate indicating that Ph1 influences chromatin organization at the interphase before meiosis (Martinez-Perez et al., 1999; Mikhailova et al., 1998; and see also below).
|
Homology recognition |
|---|
|
TopAbstractIntroductionMeiotic genes and gene...RecombinationSynaptonemal complexHomologous chromosome pairingHomology recognitionChromosome morphologyRepeated DNA sequencesSites of recombination and...ConclusionsReferences |
|---|
While pairing models are now well established, homologous recognition and testing mechanisms are less clear. For yeast, random search models have been considered (Loidl, 1998) and have some attraction. However, the applicability of any single-cell yeast model to most multicellular animals and plants is unknown. In particular, direct scaling-up of homology search processes that would work with the 14 Mbp genome of yeast to even the 100 Mbp genomes of A. thaliana or C. elegans, let alone the 3000 Mbp of humans or 17 000 Mbp of wheat, may not be possible.
Premeiotic and mitotic association of homologues is proposed in budding (see above) and fission yeast (Kohli, 1994), and such associations seem to be a universal condition in somatic Diptera cells (Hiraoka et al., 1993). Three-dimensional reconstructions of human cells (Leitch et al., 1994) and cereal plant species (Heslop-Harrison et al., 1988) were not able to find evidence for homologous association of chromosomes, even at the last mitotic division before meiosis (Bennett, 1984; Schwarzacher et al., 1992b). Chromosome painting in human spermatogonia (meiotic stem cells) revealed compact, largely mutually exclusive chromosome territories that did not show homologue association (Armstrong et al., 1994; Cheng and Gartler, 1994; Scherthan et al., 1996).
|
Chromosome morphology |
|---|
|
TopAbstractIntroductionMeiotic genes and gene...RecombinationSynaptonemal complexHomologous chromosome pairingHomology recognitionChromosome morphologyRepeated DNA sequencesSites of recombination and...ConclusionsReferences |
|---|
While homologous synapsis at zygotene is dependent on direct DNA homology check mechanisms, it is likely that chromosome recognition and initial pairing cannot be achieved by DNA–DNA interactions alone (Sybenga, 1999). Chromosome morphology, specific sequence distribution, proteins bound to DNA, and the resulting chromatin condensation patterns are likely to be involved directly or indirectly in chromosome homology recognition. For example, repeated sequences could influence recognition through their secondary folding structures, protein binding sites, and the condensation patterns that give a chromosome a characteristic shape. Specific coiling patterns with apparent denser and weaker zones, presumably reflecting more or less condensed chromatin were observed in the homologous chromosome domains of wheat. When the homologous domains associate, these condensation patterns were also aligned (Schwarzacher, 1997) and the specialized prezygotene chromosome morphology observed in maize may facilitate homology recognition (Dawe et al., 1994). Karpen et al. (1996) have analysed pairing of achiasmatic chromosomes in D. melanogaster and proposed that DNA and protein structures inherent to heterochromatin could produce a self complementary chromosome ‘landscape’ that ensures partner recognition and alignment by ‘best fit’ mechanisms. Cook (1997) argued that each chromosome in a haploid set has a unique array of transcription units strung along its length and that, therefore, chromatin fibres will be folded into a unique array of loops and that only homologues share similar arrays.
|
Repeated DNA sequences |
|---|
|
TopAbstractIntroductionMeiotic genes and gene...RecombinationSynaptonemal complexHomologous chromosome pairingHomology recognitionChromosome morphologyRepeated DNA sequencesSites of recombination and...ConclusionsReferences |
|---|
Repeated sequences play an important part in several aspects of meiosis, both structurally and functionally. They might be responsible or at least aid the first recognition of homologous chromosomes when they search for pairing partners and synapse at early meiosis. This could be achieved at the DNA level itself (Roeder, 1997) through chromosome specific sequence motifs or chromosome specific patterns of several motifs.
Telomeres
Knowledge of the importance of telomeres for synapsis comes from studies that, in the absence of homology in the distal regions of chromosome arms, very long homologous segments may remain unrecognized in meiosis (Lukaszewski, 1997). Bouquet formation heavily depends on telomere movement and attachment to the nuclear envelope followed by de novo clustering of telomeres before the initiation of pairing (Scherthan et al., 1996; Dawe et al., 1994; Bass et al., 1997).
Subtelomeric repeats with chromosome-specific distribution would be essential for recognition of homologous chromosome ends. The sequence at the telomere is highly conserved, and consists of a short repeat, similar to (TTTAGGG)n in most plants and (TTAGGG)n in mammals, tandem arrays many hundreds of units long at the physical ends of chromosomes of most eukaryotes (for review see Fuchs et al., 1995; Zakian, 1995). In plants, the enzyme adding the telomeric sequence, telomerase, is apparently active in all tissues. The length of arrays of telomeric repeats is species-specific, ranging from 2–5 kb in Arabidopsis thaliana (Richards and Ausubel, 1988), through 12–15 kb in cereals (Schwarzacher and Heslop-Harrison, 1991), up to 60–160 kb in tobacco (Fajkus et al., 1995). The number of copies of the repeat differs between chromosome arms of the karyotype (Schwarzacher and Heslop-Harrison, 1991) and varying from cell-to-cell and tissue-to-tissue (Kilian et al., 1995). It will be interesting to know about the activity of telomerase and telomere length changes through meiocyte mother cell development in plants.
Adjacent to the simple telomere repeats most organisms have more complex, often chromosome-specific, proximal sequences (Richards et al., 1993) and various interacting proteins (Biessmann and Mason, 1994). Analysis of these sequences on rye chromosomes shows that they are able to evolve in copy number rapidly (Alkhimova et al., 1999), and may be part of a complex chromosome end structure (Vershinin et al., 1995; Fig. 1c).
Centromeres
Centromeres at premeiotic interphase through to pachytene and anaphase I have a more diffuse structure in hexaploid wheat, exhibiting high homoeologous pairing compared to low homeologous pairing wheat (Aragon-Alcaide et al., 1997; Mikhailova et al., 1998). Suzuki et al. (1997) found a specific antiserum that did not stain centromeres during mitotic division in somatic cells of lily, but stained centromeres during the meiotic division (male and female) and postulate that the meiosis-specific centromere protein is associated with conversion of a mitotic to a meiotic chromosome, that meiosis is regulated by modification of the structure of chromosomes and particular centromeres, and that a meiosis-specific centromere protein is required for the meiosis-specific behaviour of the centromere. In wheat, it has been noted that the association of homologous domains apparently starts predominantly at the centromeres (Schwarzacher, 1997; Aragon-Alcaide et al., 1997; Mikhailova et al., 1998). Hence cognition of homology could be mediated by proteins or sequences associated with the wheat centromeres. However, there is no evidence that centromeres have a specific role in chromosome pairing in A. thaliana (Armstrong et al., 2001).
Eukaryotic centromeres and kinetochores are responsible for sister chromatid cohesion, the correct alignment of chromosomes on the metaphase plate, the attraction of microtubules and organization of the spindle assembly, all leading to the proper segregation of chromosome during anaphase of division (for review see Earnshaw, 1994; Allshire, 1997). The functional centromere of budding yeast is attributed to a 125 bp sequence carrying three characteristic ‘centromere DNA elements’, CDEI, CDEII and CDEIII (Clarke, 1990). However, the budding yeast and even fission yeast, with more complex and longer centromeric sequences, do not seem a particularly good model for understanding the centromeres in the plant or animal kingdoms. Despite their highly conserved function and detailed knowledge of the proteins involved, the DNA sequences at the centromeres of higher eukaryotes remain poorly understood and no universal features have been found.
In many species, the centromere is associated with blocks of heterochromatin containing highly repetitive DNA sequences which include tandem satellite repeats and retroelement-like components, representing a considerable fraction of the genomic DNA. In humans, many, but not all, authors regard the tandemly repeated 
-satellite sequences (0.3% of the human genome) as playing a key role in centromere function and chromosome segregation (Tyler-Smith et al., 1998). At each human centromere, the tandemly arranged 170 bp units of the -satellites form large arrays of megabase pair length showing a chromosome-specific subpattern of sequence variants and characteristic multimers of slightly variant units (Willard, 1985). Similar chromosome-specific variants have been identified in the centromeric minor satellite of the mouse (Kipling et al., 1994). In Arabidopsis, a major 180 bp satellite sequence, some 3% of the genome, is located at the centromeres of all five chromosome pairs (Maluszynska and Heslop-Harrison, 1991; Murata et al., 1994; Fig. 1d), although several other repetitive DNA sequences, including retroelements and degenerate telomeric motifs, have also been located in this region (Brandes et al., 1997; Fransz et al., 2000). Retrotransposon-like elements have also been found in cereal and beet centromeric repeats and many centromeric repeats include regions with homology to the 17 bp mammalian CENP-B binding motifs, although not always recognized in the source reference (Aragon-Alcaide et al., 1996; Gindullis et al., 2001; Murata, 2002).
Detailed sequence analysis of the units of a 180 bp long tandem repeat motif in Arabidopsis (Heslop-Harrison et al., 1999) identified not only conserved regions including the CENP-B and CEDII binding site motifs, but several regions with alternate base pair changes. Using specific primers for these regions, PCR or primed in situ hybridization amplified sequences on some chromosomes more than others, indicating that chromosome-specific variants are present (Heslop-Harrison et al., 1999; Fig.1e). Such specific variants at centromeres or other pairing sites would be a requisite for recognition of chromosomes using repetitive sequence motifs. Different proteins associated with DNA, including histone H3 and CENP-A or others could be involved.
Simple sequence repeats
Microsatellites or simple sequence repeats (SSRs), runs of repetitions of single nucleotides or motifs up to about 5 bp long, are ubiquitous elements of eukaryotic genomes (Tautz and Renz, 1984). However, the genomic organization of different microsatellite sequences varies widely, with implications for amplification and dispersion mechanisms, and hence evolution, and their utility for mapping (Schlötterer, 2000). Schmidt and Heslop-Harrison (1996) in beet and Gortner et al. (1998) in chickpea, used different SSRs as probes for in situ hybridization and found, apart from the dispersed overall signal, motif-specific patterns of distribution with site-specific enrichment or depletion of some motifs at centromeric or intercalary positions. In wheat and rye, several SSR motifs cluster in similar sites in the two species, although conventional staining systems give very different chromosome bands, suggesting that the pattern was established before their evolutionary separation (Cuadrado and Schwarzacher, 1998; Cuadrado et al., 2000). In Fig. 1f, in situ hybridization with the motif AAC to chromosomes of rye is shown. With many intercalary sites, AAC cluster distribution is diagnostic, but notably different and rarely overlaps with the long tandemly repeated sequences of the subtelomeric heterochromatin.
During meiotic prophase, each single chromosome along its entire length develops the synaptonemal complex (see above), to which the two sister chromatids are attached in a series of loops. The average size of the chromatin loops attached to the SC are species-specific (Moens and Pearlman, 1988; Zickler and Kleckner, 1999) and it has been postulated that one mechanism for the regulation of the loop size is the existence of specialized DNA sequences that associate with the meiotic chromosome core. DNA sequences isolated from purified, DNAse-treated rat SCs did not contain sequences that are unique to chromosome cores, but proved to be notably different from random genomic fragments and contained an excess of simple sequence repeats (mostly the dinucleotide motif GT or, on the complementary DNA strand, CA) and retroelement-related repetitive sequences (LINE and SINE elements, long and short interspersed sequences; Pearlman et al., 1992). Fluorescent in situ hybridization to mouse SCs has shown that unique sequences, the mouse minor satellite DNA sequence and some other tandemly repeated sequences are mainly found in the chromatin loops, while the signal from the telomere sequence does not come from the loops (Moens and Pearlman, 1993). Similarly, in humans, telomeric sequences were seen tightly associated with the SCs while centromeric alpha-satellites and classical satellites were all found to form loops that are associated with the SC only at their base (Barlow and Hulten, 1996).
In rye, two tandemly repeated sequences, the 18S–25S rDNA and a 120 bp repeat from rye heterochromatin are closely associated with the bivalent axes, corresponding to the SC, and also located in the surrounding chromatin loops (Albini and Schwarzacher, 1992). The relative lengths of the bivalent axes covered with signal from the 120 bp repeat appears to be less than expected from somatic metaphases (compare Fig. 1f and g), supporting the speculation that heterochromatin is under-represented in the SC length (Stack, 1984), although perhaps only with respect to somatic metaphase. In contrast to the tight packing and close association of these classes of long tandem repeats, studies using simple sequence repeats as probes for fluorescent in situ hybridization showed that very little signal is associated with the SC, but is mainly found in the chromatin loops (Fig. 1h).
|
Sites of recombination and cross-over |
|---|
|
TopAbstractIntroductionMeiotic genes and gene...RecombinationSynaptonemal complexHomologous chromosome pairingHomology recognitionChromosome morphologyRepeated DNA sequencesSites of recombination and...ConclusionsReferences |
|---|
In organisms with conventional meiosis, at least one cross-over is needed per bivalent to guarantee proper alignment of bivalents on the equatorial plate of the first meiotic division and subsequent proper disjunction of homologous chromosomes. Interference increases the distance between two or more cross-overs on the same chromosome and is possibly promoted by SCs (see also above), as interference was found to be absent in fungi and yeast mutants without SCs (Moens, 1994). To some extent chiasma frequency and distribution is determined genetically, as differences between male and female (Lagercrantz and Lydiate, 1995) and between strains (Sall et al., 1993) have been found in the same species. Chromatin, chromosome and genome structure, perhaps under genetic control, play a part in determining how many and where chiasma happen. When comparing recombination frequency in different organism, Anderson and Stack (2002) have shown that it correlates better with total pachytene SC lengths and gene number than with total genome size, except for yeast that has a remarkable higher recombination frequency: an average of 100 cross-overs per 10 Mbp and 29 µm of SC (2n=32), compared to 10 cross-overs per 180 Mb and 145 µm SC of Arabidposis (2n=10) and 40 cross-overs per 17 330 Mbp and 1475 µm SC of wheat (2n=42) are reported in their table.
In general, chiasmata are more frequent in euchromatin that contains actively transcribed genes, while they are reduced in heterochromatin containing repeated sequences. DSBs in yeast are preferentially located in promotor regions of genes and correlate with DNase sensitivity (Wu and Lichten, 1995) and their non-random distribution define large, 39–105 kb, chromosomal domains that are either hot or cold spots (Baudat and Nicholas, 1997). Overall, recombination hot-spots correspond to regions with high CG contents and a certain transcriptional profile, while cold spots are found near telomeres and centromeres. (Klein et al., 1996; Gerton et al., 2000). In Arabidopsis, Fransz et al. (2000) described a hot spot of recombination that correlated with a region of low chromatin condensation while the core centromeric region has suppressed recombination (Haupt et al., 2001). In maize, recombination hot-spots in a 140 kb multigenic regions were located in gene as well as non-gene regions, but the retrotransposons present in the region were recombinationally inert (Yao et al., 2002).
The physical distances between genes and markers along chromosomes correlate poorly with genetic map distances, particularly in the large cereal genomes (Heslop-Harrison, 1991; Lukaszewski and Curtis, 1993; Schwarzacher, 1996), but also in Arabidopsis (Lin et al., 1999; Mayer et al., 1999) and humans (Dunham et al., 1999). In wheat and rye, many genes and RFLP markers are clustered near the ends of some chromosome arms, while genetically they are far apart, indicating that genetic recombination frequency is high near the telomeres, but rare towards centromeres. In situ hybridization of cloned probes to meiotic metaphase I preparations in rye (Fig. 1i) shows that chiasmata are very close to the subtelomeric heterochromatin. The chiasmata which are visualized to occur in different segments of the rye chromosome 1R, distally or proximal of the rDNA site (Schwarzacher, 1996) show close correlation with genetic map distances described by Devos and Gale (1993). In tomato, chiasmata were found almost exclusively in more distal, rather subterminal chromosome segments in Lycopersicum esculentumxL. peruvianum back crosses (Parokonny et al., 1997) and FestucaxLolium hybrids (King et al., 1998). In early meiotic prophase nuclei from rye, Rad51 foci are seen that are localized near the telomeres (Fig. 1a) indicating that already the initiation of recombination is concentrated near the ends of rye chromosomes.
In the physical model of a rye chromosome (Fig. 1j left), blocks of simple sequence repeats are located near the centromere, long tandemly repeated sequences make up the subtelomeric heterochromatin. Dispersed repetitive sequences, related to retroelements are distributed over most of the chromosome arms (Moore et al., 1991), and genes are enriched distally close to the subtelomeric repetitive sequences. During interphase, the subtelomeric repetitive sequences do not decondense and form tight clusters, called chromocentres (Fig. 1j, right). The gene-rich distal chromosome regions are very loose, indicated by the weak DAPI staining between the chromocentres and in contrast to the relatively more condensed, more intensely DAPI-stained, centromeric regions. Similarly, at pachytene, the distal chromosome segment is more decondensed. (Fig. 1k).
|
Conclusions |
|---|
|
TopAbstractIntroductionMeiotic genes and gene...RecombinationSynaptonemal complexHomologous chromosome pairingHomology recognitionChromosome morphologyRepeated DNA sequencesSites of recombination and...ConclusionsReferences |
|---|
While great advances have been made in unravelling the mechanism of recombination by isolating the genes involved and analysing their functions and interactions, important questions in meiosis remain unanswered by looking at meiotic genes alone. In particular, the nature and mechanism of homologous chromosome recognition and pairing and the distribution of recombination sites are influenced by local and regional chromatin conformation. Further influences on meiotic processes stem from the enormous differences in genome sizes of different organisms that vary greatly between yeast with 14 Mbp per haploid genome, C. elegans with 100 Mbp, mammals with 3000 Mbp and plants ranging between 150 Mbp for Arabidopsis to well over 25 000 Mbp for lily or pine. Genome size has considerable consequences for chromosome and genome organization, the packing of DNA into chromatin and the distribution of genes and repeated DNA sequences that, in turn, influence pairing mechanism and recombination site choice.
The tools are now available to explore meiotic gene discovery and to combine the molecular sequence data with high throughput expression analysis and studies of the organization and behaviour of chromatin and chromosomes to build a comprehensive structural, temporal and functional model of meiosis. Its application to plant research will be of fundamental and applied significance, allowing the understanding and exploitation of the variation nature has provided by inventing recombination.
|
Acknowledgements |
|---|
I would like to thank Pat Heslop-Harrison and Josef Loidl for fruitful discussions and Shaobin Wu and Tanja Garkoucha for technical assistance. The following are acknowledged for collaboration: Angelines Cuadrado University Alcala, Madrid, Spain (Fig. 1f–h); Umesh Lavania, Lucknow, India (Fig. 1c); Chris Gillies, University Sidney, Australia (Fig. 1i); Lorrie Anderson, University Fort Collins, Colorado, USA (Fig. 1a, and sharing her in-press Current Genomics Article). Figures 1c and f–i are reproduced from Schwarzacher and Heslop-Harrison (2000) with permission from the publisher. This work was partially supported by BBSRC and CREST of JST (Japan Science and Technology Corporation).
|
References |
|---|
|
TopAbstractIntroductionMeiotic genes and gene...RecombinationSynaptonemal complexHomologous chromosome pairingHomology recognitionChromosome morphologyRepeated DNA sequencesSites of recombination and...ConclusionsReferences |
|---|
Albini SM, Schwarzacher T. 1992. In situ localization of two repetitive DNA sequences to surface-spread pachytene chromosomes of rye. Genome 35, 551–559.
Albini SM, Jones GH. 1987. Synaptonemal complex spreading in Allium cepa and Allium fistulosum. I. The initiation and sequence of pairing. Chromosoma 95, 324–338.[CrossRef]
Alkhimova AG, Heslop-Harrison JS, Shchapova AI, Vershinin AV. 1999. Rye chromosome variability in wheat–rye addition and substitution lines. Chromosome Research 7, 205–212.[CrossRef][ISI][Medline]
Allers T, Lichten M. 2001. Differential timing and control of non-cross-over and cross-over recombination during meiosis. Cell 106, 47–57.[CrossRef][ISI][Medline]
Allshire RC. 1997. Centromeres, checkpoints, and chromatid cohesion. Current Opinion in Genetics and Development 7, 264–273.[CrossRef][ISI][Medline]
Anderson LK, Offenberg HH, Verkuijlen WMHC, Heyting C. 1997. RecA-like proteins are components of early meiotic nodules in lily. Proceedings of the National Academy of Sciences, USA 94, 6868–6873.
Anderson LK, Stack SM. 2002. Meiotic recombination in plants. Current Genomics (in press).
Andrews J, Bouffard GG, Cheadle C, Lü J, Becker KG, Oliver B. 2000. Gene discovery using computational and microarray analysis of transcription in the Drosophila melanogaster testis. Genome Research 10, 2030–2043.
Aragon-Alcaide L, Miller T, Schwarzacher T, Reader S, Moore G. 1996. A cereal centromeric sequence. Chromosoma 105, 261–268.[ISI][Medline]
Aragon-Alcaide L, Reader S, Miller T, Moore G. 1997. Centromeric behaviour in wheat with high and low homoeologous chromosomal pairing. Chromosoma 106, 327–333.[CrossRef][ISI][Medline]
Armstrong SJ, Caryl AP, Jones GH, Franklin FCH. 2002. Asy1 protein, required from chromosome synapsis, localizes to axis-associated chromatin in Arabidopsis and Brassica. Journal of Cell Science (in press).
Armstrong SJ, Franklin FCH, Jones GH. 2001. Nucleolus-associated telomere clustering and pairing precede meiotic chromosome synapsis in Arabidopsis thaliana. Journal of Cell Science 114, 4207–4217.
Armstrong SJ, Jones GH. 2003. Meiotic cytology and chromosome behaviour in wild-type Arabidopsis thaliana. Journal of Experimental Botany 54, 000–000.
Armstrong SJ, Kirkham AJ, Hulten MA. 1994. XY Chromosome behaviour in the germ-line of the human male: a FISH analysis of spatial orientation, chromatin condensation and pairing. Chromosome Research 2, 445–452.[CrossRef][Medline]
Bai XF, Peirson BN, Dong FG, Xue C, Makaroff CA. 1999. Isolation and characterization of SYN1, a RAD21-like gene essential for meiosis in Arabidopsis. The Plant Cell 11, 417–430.
Barlow AL, Benson FE, West SC, Hulten MA. 1997. Distribution of the Rad51 recombinase in human and mouse spermatocytes. EMBO Journal 16, 5207–5215.[CrossRef][ISI][Medline]
Barlow AL, Hulten MA. 1996. Combined immunocytogenetic and molecular cytogenetic analysis of meiosis I human spematocytes. Chromosome Research 4, 562–573.[CrossRef][ISI][Medline]
Bass HW, Marshall WF, Sedat JW, Argard DA, Cande WZ. 1997. Telomeres cluster de novo before the initiation of synapsis: a three-dimensional spatial analysis of telomere position before and during meiotic prophase. Journal of Cell Biology 137, 5–18.
Baudat F, Keeney S. 2001. Meiotic recombination: making and breaking go hand in hand. Current Biology 11, R45-R48.
Baudat F, Nicolas A. 1997. Clustering of meiotic double-strand breaks on yeast chromosome III. Proceedings of the National Academy of Sciences, USA 94, 5213–5218.
Bennett MD. 1984. Premeiotic events and meiotic chromosome pairing. Symposium of the Society for Experimental Biology 38, 87–121.
Bhatt AM, Canales C, Dickinson HG. 2001. Plant meiosis: the means to 1N. Trends in Plant Science 6, 114- 121.
Bhatt AM, Lister C, Page T, Fransz P, Findlay K, Jones GH, Dickinson HG, Dean C. 1999. The DIF1 gene of Arabidopsis is required for meiotic chromosome segregation and belongs to the REC8/RAD21 cohesin gene family. The Plant Journal 19, 463–472.[CrossRef][ISI][Medline]
Biessmann H, Mason JM. 1994. Telomeric repeat sequences. Chromosoma 103, 154–161.[ISI][Medline]
Bishop DK. 1994. RecA homologs Dmc1 and Rad51 interact to form multiple nuclear complexes prior to meiotic chromosome synapsis. Cell 79, 1081–1092.[CrossRef][ISI][Medline]
Borde V, Goldman ASH, Lichten M. 2000. Direct coupling between meiotic DNA replication and recombination initiation. Science 290, 806–809.
Borts RH, Chambers SR, Abdullah MFF. 2000. The many faces of mismatch repair in meiosis. Mutation Research 451, 129–150.[ISI][Medline]
Brandes A, Thompson H, Dean C, Heslop-Harrison JS. 1997. Multiple repetitive DNA sequences in the paracentromeric regions of Arabidopsis thaliana L. Chromosome Research 5, 238–246.[CrossRef][ISI][Medline]
Caryl AP, Armstrong SJ, Jones GH, Franklin FCH. 2000. A homologue of the yeast HOP1 gene is inactivated in the Arabidopsis meiotic mutant asy1. Chromosoma 109, 62–71.[CrossRef][ISI][Medline]
Cheng EY, Gartler SM. 1994. A fluorescent in situ hybridization analysis of X chromosome pairing in early human female meiosis. Human Genetics 94, 389–394.[ISI][Medline]
Chu S, DeRisi J, Eisen M, Mulholland J, Botstein D, Brown PO, Herskowitz I. 1998. The transcriptional program of sporulation in budding yeast. Science 282, 699–705.
Clarke L. 1990. Centromeres of budding and fission yeasts. Trends in Genetics 6, 150–154.[CrossRef][ISI][Medline]
Cook PR. 1997. The transcriptional basis of chromosome pairing. Journal of Cell Science 110, 1033–1040.[Abstract]
Couteau F, Belzile F, Horlow C, Grandjean O, Vezon D, Doutriaux MP. 1999. Random chromosome segregation without meiotic arrest in both male and female meiocytes of a dmc1 mutant of Arabidopsis. The Plant Cell 11, 1623–1634.
Cuadrado A, Schwarzacher T. 1998. The chromosomal organization of simple sequence repeats in wheat and rye genomes. Chromosoma 107, 587–594.[CrossRef][ISI][Medline]
Cuadrado A, Schwarzacher T, Jouve N. 2000. Use of synthetic oligonucleotides as cytological markers to study heterochromatin distribution in wheat. Theoretical and Applied Genetics 101, 711–717.[CrossRef]
Culligan KM, Hays JB. 1997. DNA mismatch repair in plants—an Arabidopsis thaliana gene that predicts a protein belonging to the MSH2 subfamily of eukaryotic MutS homologs. Plant Physiology 115, 833–839.[Abstract]
Culligan KM, Hays JB. 2000. Arabidopsis MutS homolog, AtMSH2, AtMSH3, AtMSH6 and a novel AtMSH7, form three distinct protein heterodimers with different specificities for mismatched DNA. The Plant Cell 12, 991–1002.
Dawe RK, Sedat JW, Agard DA, Cande WZ. 1994. Meiotic chromosome pairing in maize is associated with a novel chromatin organization. Cell 76, 901–912.[CrossRef][ISI][Medline]
Devos K, Gale M. 1993. The genetic maps of wheat and their potential in plant breeding. Outlook on Agriculture 22, 93–99.
Ding ZJ, Wang T, Chong K, Bai SN. 2001. Isolation and characterization of OsDMC1, the rice homologue of the yeast DMC1 gene essential for meiosis. Sexual Plant Reproduction 13, 285–288.[CrossRef][ISI]
Dong F, Cai X, Markaroff CA. 2001. Cloning and characterization of two Arabidopsis genes that belong to the AD21/REC8 family of chromosome cohesin proteins. Gene 271, 99–108.[CrossRef][ISI][Medline]
Dunham I, Shimizu N, Roe BA, et al. 1999. The DNA sequence of human chromosome 22. Nature 402, 489–495.[CrossRef][Medline]
Earnshaw WC. 1994. Structure and molecular biology of the kinetochore. In: Hyams JS, Lloyd CW, eds. Microtubules. New York: Wiley-Liss, 393–412.
Fajkus J, Kovarik A, Kralovics R, Bezdek M. 1995. Organization of telomeric and subtelomeric chromatin in the higher plant Nicotiana tabacum. Molecular and General Genetics 247, 633–638.
Franklin AE, McElver J, Sunjevaric I, Rothstein R, Bowen B, Cande WZ. 1999. Three-dimensional microscopy of Rad51 recombination protein during meiotic prophase. The Plant Cell 11, 809–824.
Fransz PF, Armstrong S, de Jong JH, Parnell LD, van Drunen G, Dean C, Zabel P, Bisseling T, Jones GH. 2000. Integrated cytogenetic map of chromosome arm 4S of A. thaliana: structural organization of heterochromatic knob and centromere region. Cell 100, 367–376.[CrossRef]