The oscillation of an organism between haploid and diploid stages is the very essence of sexual reproduction. Thus the ability to accurately pair and separate homologous chromosomes during meiosis is essential for any sexually reproducing organism. Reporting in Nature, Graham Moore and colleagues at the John Innes Centre, Norwich, UK, have explored how this fidelity is maintained.

Pairing of homologous chromosomes is difficult enough in a diploid organism, but many plants are polyploid with, in effect, several genomes sharing each cell. Pairing of chromosomes occurs mainly through the centromeres, so in wheat, a hexaploid, each centromere has five very similar potential partners, only one of which is its true homologue. One might expect that this complexity would result in meiosis taking longer in polyploid species compared with diploids — but in fact it is faster. This is probably because, while homologous chromosomes come together only at the onset of meiotic prophase in diploids, polyploids pair their chromosomes long before meiosis begins.

In 1958, the agricultural geneticist Sir Ralph Riley identified the PH1 locus in wheat to be essential for correct homologous chromosome pairing and recombination. As a result of his work, a single mutant carrying a deletion of this locus has been used in wheat breeding ever since. Its activity has remained controversial, however, and even its composition is unclear.

Moore and colleagues used confocal microscopy to investigate PH1 function after the pairing of chromosomes in developing xylem cells from wheat. They were particularly concerned with a stage at the beginning of meiosis when a 'telomere bouquet' forms. Here, all of the telomeres (red in the figure) come together at one side of the dividing cell, whereas the centromeres (green) congregate at the other. It was difficult to distinguish paired from single centromeres, but in the absence of the PH1 locus there were more centromere pairs and fewer singlets than in its presence. In wheat/rye hybrid plants, the chromosomes from the different parents could be identified, allowing the authors to deduce that there was less mispairing of centromeres in the presence of PH1.

The molecular mechanism by which the PH1 locus weeds out incorrect centromere pairings before meiosis begins — so that only correct pairings are carried into cell division — remains to be worked out, as does identification of the responsible gene or genes within the PH1 locus. However, it is clear that PH1 is more divorce lawyer than matchmaker, breaking up unsuitable pairings so perfect couples can meet.