Why apomixis is genetic gold

They seek it here, they seek it there, but it remains hidden in the genetic thickets of flowering and seed development. A place in history awaits its finder.

It's a gene that could take plant breeding into a new dimension in the 21st century -- one in which breeders will create elite varieties with all of the advantages of F1 hybrids, but whose seeds will yield genetic replicas of the female parent plants.

The gene -- or genes -- endow some plants with the ability to produce seeds and fruits without being pollinated, a trait called apomixis. Many species have independently evolved the trait as a fallback strategy when birds, bees and breezes fail.

Research teams in Europe, Asia, South America, the US and Australasia are searching for genes that will allow molecular geneticists to develop apomictic crops, in which genes for desirable traits could be 'stacked' and kept together between generations, permanently insulated against the vagaries of sexual reproduction.

Before World War II, most farmers in developed nations saved seed from their favourite varieties -- they distrusted hybrids because they did not "come true to seed".

Because recombination in self-pollinated F1 plants re-assorts the genes from the original, inbred parental lines, each F2 seedling carries a unique mix of their best and worst traits.

Seed companies eventually convinced them that it would be more profitable to buy F1 hybrid seed each year, to take advantage of the greater vigour, higher yield and disease resistance of F1 hybrids.

Apomixis promises the best of both worlds: the genetic stability of traditional varieties, combined with the superior attributes of F1 hybrids. Molecular geneticists could also add transgenes for novel traits like 'designer' virus-resistance, enhanced nutritional properties, or health-promoting compounds such as antioxidants.

Progress

Dr Anna Koltunow, of CSIRO Plant Industry's Horticulture Unit in Adelaide, described her team's progress towards understanding the genetic mechanisms of apomixis at an international symposium held in Canberra last month to honour the division's retiring chief, Dr Jim Peacock.

Koltunow's team uses a dandelion look-alike, Hieracium, as a model for apomixis. It is collaborating with Dr Ross Bicknell's research group at the Crop and Food Research Institute in Christchurch, New Zealand.

Koltunow says some Hieracium biotypes reproduce sexually, others apomictically -- crosses between them yield both sexually-reproducing and apomictic progeny, suggesting the 'choice' of reproductive mode is under relatively simple genetic control.

Her team has shown that the two pathways share common regulatory elements. Selected marker genes involved in forming reproductive structures in sexually reproducing biotypes exhibit the same patterns of activity in apomictic types.

Apomixis simply appropriates the ancient mechanisms of sexual reproduction, but deviates from the typical scheme in dispensing with two key, early checkpoints: meiosis and fertilisation.

The seeds will develop as genetic replicas -- clones -- of the female parent. There is no male contribution of genes from a pollen parent.

That's an important finding, says Koltunow, because if apomixis simply recruits most of the genes normally involved in forming the plant's female structures and seeds, it should be possible to engineer the same trick into sexually reproducing crop species.

"In our apomictic plants, apomixis begins after the sexual process has started, and then the sexual process degenerates," she says. "Our marker-gene work reveals a changed pattern of gene expression in early reproductive structures when we compare sexual and apomictic plants.

"Interestingly, in the apomict we saw a change in the expression pattern of a particular gene that may, or may not, be involved in apomixis.

"The marker gene we used was a modified Polycomb-like gene originally identified in Arabidopsis. We fused it with the GUS marker gene, put it into sexual and apomictic plants and observed a beautiful pattern shift that was clearly different in apomictic plants at the start of the process.

"The original Arabidopsis gene is involved in later reproductive events in the early stages of seed initiation -- these sorts of polycomb genes may have been recruited and used in a different way in apomixis, to circumvent the normal sexual reproductive process.

"In a sexually reproducing biotype, the gene is expressed in three out of four cells formed after meiosis that are destined to degenerate. But it is not expressed in the cell that eventually develops into the female reproductive structure-or embryo sac. In apomictic biotypes it is expressed in all four cells but is absent from an enlarging cell positioned directly behind them that is destined to initiate apomictic reproduction."

Starting-point

Koltunow suspects that, in sexually reproducing Hieracium biotypes, a similar gene transmits a signal that normally instructs the target cells to degenerate, so that only one of the four cells carries on with sexual reproduction. But in apomictic types the signal is transmitted to all cells involved in sexual reproduction once the cell that initiates apomixis has begun to form, so reproductive development is diverted into the apomictic pathway.

"We're at the starting-point, when the success or failure of either process is at stake," she says. "The really interesting question is: what are the interactions between the two processes at initiation, and how can we use the information to control which one survives. How can we create plants with 100 per cent apomictic seed production?"

Koltunow's team wants to know if the apomictic process involves a genetic 'memory' effect, inherent to the initiating cell's identity, so that once the process starts it is programmed to continue inexorably to completion.

She says sexual development is not a completely autonomous, all-or-nothing process. Rather, it unfolds sequentially, passing through a series of developmental checkpoints. If something goes wrong, one of these checks halts the process. Perhaps slip-ups in these checkpoint controls allowed a switch into a backup route -- the apomixis pathway.

Koltunow says that because sexual reproduction is geared to maximising seed production, any ovule that fails to produce a seed represents a waste of resources. So natural selection may have created a back-up route.

In Canberra, Jim Peacock, Liz Dennis and Abed Chaudhury have already shown that perturbing members of the Polycomb-like complex of genes that act late in Arabidopsis ovule development causes strange things to happen. Disruption of any one of three particular Polycomb-like genes results in seeds that begin development of nutritive endosperm tissue without fertilisation, but no embryo develops and the seed is not viable.

In apomictic Hieracium seeds, the embryo and the endosperm tissue develop autonomously, and simultaneously, without fertilisation. Normally, neither embryo nor endosperm develops if fertilisation fails, as the Polycomb group genes are involved in blocking the expression of the cascades of genes required for both embryo and endosperm developmental pathways until fertilisation lifts the blockade.

Yet in apomictic plants, both pathways remain active, and coupled, in the absence of fertilisation. Koltunow and her colleagues believe apomixis somehow misregulates the genes that would normally block both endosperm and embryo development, so they proceed normally even though fertilisation has not occurred.

It's a challenging but exciting area of research, says Koltunow. "Students love it because they usually work only on sexually reproducing plants, and it's inconceivable to them that you can get seeds without fertilisation. Our project opens their eyes to the diversity of reproductive mechanisms in plants."