Sex is not just an activity confined to higher animals such as mammals and not just for reproduction. The changes in offspring that result during reproduction are integral to the continuing adaptation and evolution of millions of species of plants, fungi and other organisms throughout the food chain. 

Without such genetic shuffling, life wouldn't have adapted to the planet's frequently changing climates, or to new pests, parasites and predators. Our environment is always changing and emerging pressures, such as predicted rises in temperature for example, could threaten food production systems by affecting overall yield – it has been estimated that a 1°C rise in night-time temperature could reduce rice yields by about 10%. Don't panic about your coffee, despite the doomsday scenarios you read in mainstream media, other species of coffee will be suited for whatever conditions arise, but what about other plants that have not been artificially selected by mankind for agriculture?

Controlling genetic shuffling could be our friend.  And so researchers are taking a detailed look at the fundamental molecular processes by which organisms shuffle and exchange their genes. Looking at barley as a model for cereal food crops, their findings show how genetic exchange during sex can be affected by more subtle increases in temperature than previously recognized. It's a discovery that could impact fields ranging from agricultural development to advanced plant breeding techniques to community-scale climate and ecosystem modeling. 

Plant breeders in particular have been interested in how genetic variation is controlled ever since Gregor Mendel's classic experiments with peas in the late 1860s. If you don't understand how characteristics such as seed size or growth are inherited you can't breed varieties to exhibit certain traits or not to express them at all. 

Fortunately, genetic variation through generations is a natural part of sexual reproduction in higher organisms (eukaryotes) such as fungi, plants and animals. At conception each individual receives a set of maternal and paternal chromosomes. Together this set of chromosome pairs makes up their genetic blueprint. During sexual reproduction DNA replicates on structures called chromosomes, which typically appear in pairs, and a process called meiosis ensures only one half of each chromosome pair is passed to their eggs or sperm cells. Subsequent fusion of a sperm and egg during fertilization restores a full set of chromosome pairs; this halving and re-pairing during meiosis prevents the number of chromosomes doubling each generation which would cause chaos in a cell.

One of the consequences of this system is that just as a deck of cards is split in two before being shuffled back together before a round of poker, each new generation – be it from a seed, spore or a fetus – receives a slightly different set of genes from their parents. This genetic variation arises because during meiosis, segments of the parental chromosome pairs reciprocally recombine to swap places in a process called crossing-over.

Akin to the top half of the king of spades appearing with the bottom half of the queen of hearts, the creation of new stretches of DNA in this way adds a whole new dimension of variation into the mix – as it would in a game of poker if cards recombined in such a way.

Professor Chris Franklin and colleagues from the University of Birmingham, funded by the UK's Biotechnology and Biological Sciences Research Council, have been looking at the molecular processes behind this variation for over 15 years. Franklin says that although meiosis has been studied for more than a century, it was not until the late 1980s onwards when scientists started looking at yeast that we really began to understand the underlying biochemical processes."

By then, it had been long-noted by plant breeders working that some plants didn't engage in the gene-swapping meiotic processes as others did. In the cereals in particular, certain chromosomal regions didn't crossover very much, it was assumed that these regions didn't have much in the way of genes and just contained 'junk DNA'. Dr. Robbie Waugh's work at the Scottish Crop Research Institute showed in grass chromosomes that 30-50% of genes are not in these regions.

"So then it became more of an issue because there is a very low frequency of recombination there compared to distal ends of chromosomes, potentially limiting genetic variability," says Franklin.


New genetic combinations are formed during meiosis. But regions where this happens are restricted in barley. Image: NIH

Teaming up with colleagues at the James Hutton Institute (JHI), University of Dundee and at Aberystwyth University, who had a strong background in the genetics of the grass family, Franklin says the idea was to try to determine why crossing-over is mainly confined to the ends of chromosomes.

"We thought it would be a good idea to apply the tools for we have developed to study meiosis in Arabidopsis to cereals," says Franklin, adding that barley was a good system as like most organisms their chromosomes are in pairs (diploid) rather than the complex hexaploid (three pairs) seen in bread-making wheat. As a food crop, findings in barley would likely be applicable to a host of other cereal grasses, including oat and rye for example.

Detecting meiotic proteins with fluorescent antibodies that can be seen with light microscopes, coupled with techniques that allow the isolation of chromosomes during meiosis, Franklin and colleagues, James Higgins, Sue Armstrong and Ruth Perry were able to assemble, image by image, a complete picture of chromosomal recombination and crossing over during meiosis in barley in exquisite detail.

Overall, researchers found 12-19 crossovers per cell (spread across all chromosomes) and that there were 1-3 crossovers per chromosome. But crossovers were 25 times more likely at the ends compared to the middle of chromosomes, which is very different to observations of recombination in brassica crops, such as cabbage and broccoli, and Arabidopsis. 

The story that began to emerge was that the initiation of recombination does occur throughout the length of the chromosome – but it's all a matter of timing. The process begins two-to-three hours earlier at the end of chromosome (distal) than the interstitial (middle) and proximal regions found towards the centre of the chromosomes. "Basically, it starts at the end and works to the middle."


Meiotic proteins ASY1 (red) and Zyp1 (green) in a barley nucleus. The skewed distribution of the short stretches of ZYP1 is due to recombination initiating earlier in the distal regions compared to interstitial and proximal chromosomal regions. Image: James Higgins

What might be the origins of such a system? It could be due to the nature of the physical structures at play. Chromosomes are dynamic assemblages, and cluster near the membrane of the nucleus within which they are contained. Because the ends of chromosomes (telomeres) are tethered there, that could be where recombination initiates first.

But Franklin doesn't think that's the whole reason. Another aspect of recombination timing could be related to cycles in chromatin, which are the DNA-plus-protein structures on chromosomes that control gene expression and DNA replication, as well as storage and repair. Hence, these chromatin cycles are physical changes in mechanical forces along the chromosomes that are in some way related to the biochemical transitions taking place.

"The problem is that you can't directly measure these mechanical forces," Franklin explains. "The evidence in yeast is that you get cycles of [chromatin] expansion and contraction, and that the maximum expansion is coincident with some of the most important biochemical transitions in the recombination process." Among these important biochemical transitions are the way that double-strand breaks in DNA can be repaired, so they end up as non-crossover regions.
Cycle paths

Having found that the chromatin cycles are important in crossing over, the researchers then asked themselves if they could perturb them and affect the distribution of recombination events to get genetic mixing along more of the chromosome. To jolt the system, Franklin and colleagues used temperature because it's been known for a long time that extra heat during meiosis can affect the process but most studies used large increases in temperature, which completely disrupt meiosis. Only one late-1950's study in Tradescantia, a non-crop species, had shown that a subtle increase in temperature has an effect on crossover distribution.

This proved to be the case in barley. But to the team's surprise a subtle increase in temperature of around 5°C had no effect on the cycles and length of meiotic process. "What did change was the timing of DNA replication before meiosis," says Franklin. Typically, gene-rich regions in the end regions are replicated earlier and DNA in the middle and centre of chromosomes is replicated later. "We found with increased temperature that the differentiation of when DNA was replicated was blurred. In a sense we've found the underlying basis to the earlier observation in Tradescantia, and as a consequence we see more crossovers in interstitial regions," he says.

How much more is hard to quantify as there is natural variation in crossing over frequency from chromosome to chromosome. On chromosome 5 for example, there is a 3.5-fold increase in cross over number in the interstitial areas. It's also noteworthy that increased temperature caused a significant drop in the average number of crossover events per cell, from 14.8 to 13.5. So higher temperatures could cause fewer recombination events to occur, which could slow reactions to environmental change, balanced by more crossovers in areas where it does not usually occur.

And the subtlety of the temperature increase is pivotal. When meiosis fails, plants become infertile. Even an increase from 25°C to 30°C causes a significant reduction in fertility (ref 5). "There's a potential environmental and food issue here in that you have climate change and with median temperature increases you could see a reduction in fertility because of a failure in meiosis," says Franklin.

In the short term, the work could help plant breeders to adapt new varieties. For instance, a technique called map-based cloning uses recombination events to move genetic markers to where a target gene might be. But if there are no (or few) crossing over events you can't infiltrate that part of the chromosome to pin it down.


Temperature effect on chiasma (crossover) position in barley meiosis chromosomes (blue rings) at 22°C (left) and 30°C (right). In the diagram maternal and paternal chromosomes are shown in different shades of blue; crossovers are the black crosses and the centromeres are represented by triangles. At 30°C more crossovers are observed in the interstitial regions of the chromosomes making it possible to separate desirable from undesirable traits during plant breeding. Image: James Higgins

Similarly, when breeders make new varieties they often bring in desirable genes along with 'junk DNA', so subsequent cross-breeds are needed to filter out all but the wanted traits. But if there are no crossovers in that region you can't breed them out. "This is called linkage drag – where you drag in a whole lot of stuff you don't want," says Franklin. "With more recombination you could break up this unwanted material. And the beauty of using heat-temperature is that it's dead simple and requires no major facilities."

The next phase of the research is to use a range of temperatures rather than just the 30°C employed and to see how the different chromosomes respond. "We don't know what the rules are and that's what we really want to find out," says Franklin. His team's microscopy work will also be complemented by collaboration with the JHI, and their expertise in genetic studies may yet pin down the rules of sex, reproduction and chromosomal recombination in many important food plants.