Recombination, and chromosomes, in a changing environment


The distribution of favourable alleles among individuals within a species of most eukaryotic organisms is a collaborative process (sexual reproduction), with rare exceptions, for example, the infamous bdelloid rotifers (Mark Welch et al., 2004). Shuffling of allelic combinations is constrained by their arrangement in a few large DNA molecules per nucleus; the lowest number in plants being two (in the Compositae species Brachycome dichromosomatica and Haplopappus gracilis, and the grasses Colpodium versicolor and Zingeria biebersteiniana Paradoxically, even the ciliated protozoans, which have genes on many small DNA molecules in their macronuclear genomes (Hamilton et al., 2005), do not take this as an opportunity for independent assortment of a large set of allelic variants, but use the conventional genome structure of their micronuclei for meiotic division. The arrangement of allelic forms of genes along the DNA molecule within a chromosome can be transmitted intact through meiosis, but the formation of balanced gametes, with a single copy of each chromosome, normally requires chiasmata that also exchange allelic forms between parental chromatids (Wijnker et al., 2012). Chiasma formation seems to be required for proper chromosome disjunction (but Drosophila has found a way round this, Subramanian & Bickel, 2009). Alleles on distinct chromosomes usually segregate independently and so recombine freely. Alleles carried on one chromosome are recombined only by the action of chiasmata; they form a single genetic linkage group except when there is an obligate chiasma at a specific location, as for the pseudoautosomal region on our sex chromosomes (Otto et al., 2011). Chiasma formation therefore seems to be primarily about the proper segregation of whole chromosomes. Despite its importance in genetics, recombination within a linkage group could be considered as a side issue; nondisjunction has an obvious selective disadvantage, so recombination is the result of a process that has a selective advantage.

‘… their general conclusion causes us to think about what strategies we need to deploy for our self-interest in food security.’

Recombination is of great practical importance in breeding (Martinez-Perez & Moore, 2008) and the pattern of allele re-assortment within populations is the central issue underlying the genetic basis of evolutionary change (Lewontin, 1970). But is more recombination better than less, or does that depend on circumstances?

In this issue of New Phytologist, Escudero et al. (pp. 237–247) address this question directly, making elegant use of the holocentric chromosomes of the Cyperaceae (sedges). Holocentric chromosomes occur in diverse groups of organisms (e.g. nematodes, butterflies, sedges); they behave as though the whole chromosome is a centromere. Species with holocentric chromosomes are notably resistant to the consequences of chromosome breakage or fusion; for these chromosomes the risks from ectopic recombination (between nonhomologous chromosomes as might be common at very high recombination rates) seem not to pose a serious problem. Genera with holocentric chromosomes typically have a wide range of chromosome number per species, presumably because they are tolerant of chromosome fragmentation or fusion.

Interestingly the Poales, which include the grasses that are directly or indirectly our major source of food, have several taxa in which holocentric chromosomes occur. Chromosome number in grass (Poaceae) genomes ranges from two to 40 for diploid species ( and comparative genomics in grasses reveal chromosome evolution through frequent nested chromosome insertions into centromeric regions (Luo et al., 2009).

Chromosome number can be used as a proxy for effective recombination rate per meiosis in comparative studies, so genera containing species with holocentric chromosomes are ideally suited to such analysis because of their recent diversification and range of chromosome number. Escudero et al. have compared habitat preference and plant morphological traits with chromosome number in Carex to test the idea that environmental stability may permit high recombination rates.

It is bad form to say how a story ends. Although Escudero et al. have studied wild species not used in agriculture, suffice to say that, faced with the potential for rapid and extreme change in our environment (in which we breed our cultivated species), their general conclusion causes us to think about what strategies we need to deploy for our self-interest in food security.


The authors thank Julie Hofer for useful discussions on the text.