The phoenix factor

Compounds called karrikins, present in bushfire smoke, induce mass germination of seeds shed by Australia’s fire-adapted plants in the wake of fire. University of Western Australia chemist Dr Mark Waters has traced the story of karrikins right back to the first simple plants to colonise the land, more than 430 million years ago.

Vistas of millions of seedlings erupting from blackened, scorched earth in the wake of a bushfire are all too familiar to people living in south-eastern Australia, the most fire-prone region on the planet. In fact, many of Australia’s highly combustible, dry-leaved plants have evolved a dependence on episodic wildfires to reproduce.

For a long time, the intense heat of bushfires was thought to be the agent that initiated mass germination — it was just a matter of adding water to the ash bed. But in 1990, researchers in South Africa made a surprising discovery: the chief stimulus for mass germination of seedlings in that country’s similarly fire-tolerant fynbos heathlands is not heat, but the dense smoke given off by burning plants and leaf litter.

In 1995, South African researchers showed that a dilute aqueous solution of smoke induced germination of a rare succulent species, Audounia capita. The seeds of many other fynbos species, renowned for being very difficult to germinate, sprouted rapidly and almost simultaneously after treatment with a dilute smoke-water solution.

Smoke-water treatments have revolutionised the conservation of rare flora by making it possible to germinate deeply dormant seeds of rare and common plant species, under controlled conditions or in situ in the wild, from the dormant soil seed bank.

Smoke from the bushfires that ravage Australia’s typical species-rich vegetation contains as many as 3000 different combustion products. More than a decade ago, Dr Kingsley Dixon’s research group at Perth’s Kings Park and Botanic Garden embarked on a needle-in-the-haystack search for the mysterious agent in smoke that, even at extremely dilute concentrations, breaks seed dormancy in a wide variety of fire-tolerant species.

It took several years, but in 2004, Dixon and his colleague Dr Gavin Flematti, a chemist at The University of Western Australia (UWA), announced that they had isolated and identified a class of butenolide molecules that are extremely potent activators of germination. They called the compounds ‘karrikins’ — karrik meaning ‘smoke’ in the language of south-west Western Australia’s Noongar people.

But the crucial question remained unanswered: how do karrikins stimulate dormant seed embryos to germinate?

Research by UK-born molecular biologist Dr Mark Waters, of the ARC Centre of Excellence in Plant Energy Biology at UWA, has provided the answer. In 2012, he discovered a likely receptor protein for karrikins, called KAI2. He published his finding in the journal Development.

Dr Waters, who was honoured with this year’s Peter Goldacre Award from the Australian Society of Plant Scientists, will describe his findings in the Peter Goldacre Memorial Lecture at the upcoming ComBio 2015 research conference in Melbourne.

Dr Mark Waters.

Since Flematti et al identified the first karrikin molecule, KAR1, it has become clear that the karrikin response is widespread among the world’s land plants and is not restricted to fire-responsive species. Dr Waters believes karrikins have co-opted elements of a much earlier signalling network involved in plant development — from as far back as bryophytes like mosses and liverworts, which grow in moist environments and rarely experience fire.

The fact that proteins responsive to karrikin signalling exist in bryophytes suggests to Dr Waters that the signalling originated before vascular plants began to diversify — the first vascular plant macrofossils, from a genus called Cooksonia, occur in Victorian sedimentary rocks of middle-Silurian age (around 433 million years ago).

 “These proteins (in the signalling network) go back a long way,” he said.

“All plants — not just angiosperms — have them. We have tested a few plant species that are not normally exposed to fire and found they respond nicely to karrikins.

“Surprisingly, one of them is Arabidopsis, which would almost never be exposed to fire in its natural habitat.

“Conversely, while all plants seem to have the genes, not all of them necessarily respond to karrikins.

“There’s a disconnect between those plants that have been identified as fire responsive and those that respond specifically to karrikins. For example, some grasses whose seeds are very dormant do not germinate in response to karrikins, so karrikins are no magic bullet (for germinating all plant seeds).”

Dr Waters says karrikins are structurally similar to strigolactones — plant hormones that were originally identified as seed-germination stimulants produced by members of the broomrape family Orobanchaceae, which are root parasites. Strigolactones also regulate branching of axillary shoots and many other aspects of plant development.

Five new members of the karrikin family have been identified since Flematti et al isolated the first, KAR1. Of the six, KAR1 and KAR2 are the most active, so Dr Waters selected these two for his experiments.

He says both karrikins and a synthetic strigolactone, GR24, induce early germination of Arabidopsis thaliana seeds and enhance seedling responses to light. One effect is to inhibit elongation of the hypocotyl — the stem of the seedling between the cotyledons (seed leaves) and the radicle, or primary root.

But karrikins don’t stimulate seed germination in root parasites like witchweeds (Striga spp), the genus in which strigolactones were first discovered. Nor do they have any effect on axillary shoot architecture in witchweeds, which seem to respond exclusively to their own strigolactone signals.

Dr Waters concluded that while karrikins are superficially similar to strigolactones, they are functionally distinct, but both probably targeted members of the same set of protein receptors in plant cells.

Through targeted gene-knockout experiments in Arabidopsis thaliana, Dr Waters and US colleague Dr Dave Nelson created three karrikin-insensitive (kai) mutants.

Karrikin-insensitive mutants.

The first target, MORE AXILLARY BRANCHES 2 (MAX2), was Dr Nelson’s discovery. It had previously been implicated in mediating responses to strigolactones. It provided the first genetic evidence that the response pathways to karrikins and strigolactones have certain components in common.

The second karrikin-responsive protein, KAI2, discovered by Dr Waters, is evolutionarily similar to the DWARF14 (D14) protein in rice. Rice D14 mutant plants have reduced apical dominance and increased numbers of axillary branches, giving the plant a bushy appearance. The D14 protein is an enzyme that appears to be essential for the strigolactone response; it is widely recognised as the receptor for strigolactones.

KAI2 is necessary for all karrikin responses in Arabidopsis but is unresponsive to strigolactones. This supports the conclusion that, although they have certain targets in common, karrikins and strigolactones differ structurally and functionally.

The third karrikin-responsive protein in Arabidopsis, also Dr Nelson’s discovery, is SUPPRESSOR OF MAX2 1 (SMAX1). It appears to act downstream of MAX2, abolishing MAX2’s function by tagging it for breakdown by the cell’s used-protein disposal system.

Freshly harvested seeds of karrikin-insensitive Arabidopsis plants carrying MAX2 or KAI2 loss-of-function mutations show significantly reduced germination rates relative to normal Arabidopsis. Both mutants also develop elongated hypocotyls and reduced, downward-oriented cotyledons, symptomatic of impaired responses to light — in contrast, karrikin-activated seedlings grow larger cotyledons that orient towards sunlight and have more robust growth.

In his recent Plant Cell paper, Dr Waters said that the KAI2-MAX2 signalling system plays an important developmental role, even in the absence of exogenous karrikins. He stated, “Given that D14 is the likely strigolactone receptor, these observations imply that the mutant KAI2 and MAX2 phenotypes result from an inability to perceive an unknown signalling compound that is the substrate or ligand for (normal) KAI2.”

While the KAI2 ligand remains unknown, he said it will likely be a butenolide, like the karrikins.

Dr Waters said KAI2 performs a range of functions in plants. In Arabidopsis, its phenotypic effects are a by-product of other KAI2-regulated functions.

“Some plants that really do respond to smoke have become exquisitely responsive to karrikins — they work at nanomolar concentrations, so there has been strong positive selection for sensitivity to karrikins,” he said.

“The big question is: if KAI2 did not evolve as a target for karrikins, there should be another molecule that is the original ligand for KAI2. If we can find it, or develop something like it, we could use it commercially to influence germination and plant development.

“For example, we could either promote seed germination, or inhibit it.

“For a long time, there has been this idea that you could spray karrikins or karrikin-like compounds onto farmland to mass-germinate weed seeds so they could be killed with herbicide. The problem is that karrikins are not very stable when exposed to the ultraviolet light in sunshine — they dimerise within a few hours and become inactive.

“Kingsley Dixon’s group at Kings Park is looking at using karrikins for restoration ecology — for restoring mine sites, for example, or stimulating regeneration of native plant seeds in the field.

“One of the problems of spraying with smoke-water is that smoke also contains compounds that inhibit germination, so it would be much more efficient to spray with pure karrikins because they trigger germination at extremely low concentrations.”

Dr Waters has recently published a paper that examined KAI2 function in Selaginella, a primitive plant from a group allied to the club mosses, or lycopods. One of the earliest vascular plants, Selaginella appears in the fossil record during the Silurian period, 400 million years ago.

A US research team recently sequenced the genome of Selaginella moellendorffii, which has become an important model for comparative genomics research and understanding plant evolution. Dr Waters’ team found two KAI2 homologues in the Selaginella genome and had them synthesised.

The team inserted the KAI2 transgenes into Arabidopsis and showed that one of the two proteins could operate in the natural karrikin signalling system in Arabidopsis. The experiment supports the hypothesis that karrikins work their smoky magic by appropriating elements of the ancient strigolactone signalling system.

Australia’s uniquely fire-adapted flora began to flourish with the onset of locally arid conditions after the continent parted company with Antarctica, around 50 million years ago, in the final episode of the breakup of the Gondwana supercontinent. Between 25 and 10 million years ago, genera like Eucalyptus, Casuarina, Banksia, Hakea, Petrophile, Grevillea and Fabaceae pea plants underwent rapid evolutionary radiation across the continent, and eventually came to dominate the flora.

Experiments at Kings Park and Botanic Garden in Perth have identified more than 120 native plant genera in which some or all species exhibit enhanced germination in response to treatment with smoke-water — including smoke bushes (Conospermum spp, Proteaceae), which had long defied efforts to grow them from seed.

************************************************************

Karrikins and conservation

In late July, a small team of volunteers from the Australian Native Plant Society’s Dryandra Study Group spent five days touring the south-eastern wheat belt of Western Australia, taking leaf samples from fragmented populations of Banksia densa (formerly Dryandra conferta). Scientists at the WA Herbarium will run a comparative DNA analysis on the leaf samples to determine the taxonomic status of a very rare and distinctive form of B. densa with attractive silvery blue leaves and bright yellow flowers, known to enthusiasts as ‘Corrigin Blue’.

Just two small, roadside populations of B. densa Corrigin Blue were known to have survived extensive clearing of the native heathlands for grain farms in the area around Corrigin, 200 km south-east of Perth. The survey team, with your correspondent as a guest, arrived at the first location to find the area had been razed by a bushfire the previous summer. A thorough search failed to locate any surviving plants of Corrigin Blue.

But across the fire-blackened landscape, the former heathland vegetation was already regenerating vigorously. In places, seed liberated from scorched Hakea seed capsules and Petrophile seed cones had already germinated in the ashes in such numbers that they resembled a lawn.

The shrub Hakea cucullata.

Banksias — especially those species formerly included in the genus Dryandra* — are uniquely dependent on fire to reproduce. If any mature plants of Corrigin Blue existed at the site before the fire, their seeds should have released to germinate in numbers sufficient to replace or even exceed those of the parent population.

Fire is a vital force in maintaining the rich flora of WA’s heathlands and woodlands. In senescent heathland, the searing heat of its passage liberates seeds from over-mature plants, leaving a nutrient-rich ash bed that becomes the nursery for a new generation of seedlings after the first substantial rains.

Fire is more than a seed-releasing agent: the combustion process generates smoke imbued with potent molecules called karrikins that induce mass germination of the newly released seed crop. Karrikins also play a vital role in stimulating regeneration of long-dormant seeds deposited by shorter-lived pioneer species in the soil seed bank, as long-lived woody shrubs become progressively more dominant.

B. densa Corrigin Blue, one of the rarest plants in the WA flora, was in danger of extinction simply because of its anonymity — despite its distinctiveness, it was not accorded species status. Plant taxonomists — an increasingly rare breed themselves — had lumped it with a broad complex of taxa under the specific title B. densa.

Corrigin Blue is very distinct from other members of the B. densa complex in multiple ways, including the shape and hue of its leaves, its floral structures, its flower colour and its dense, mounding habit. If the DNA study affirms its unique identity, then, as the first-discovered member of the complex, it would become the holotype or foundation species for Banksia densa.

There is a proposal to hive off all the taxa formerly included in the B. densa complex, under the holding name ‘Banksia spp “Wheat Belt”’, until their own distinctiveness and relationships can be resolved.

On the final day of the survey, team member and Dryandra expert Keith Alcock discovered a third population of B. densa Corrigin Blue that more than doubles the known number of surviving plants.

The Dryandra Study Group will monitor regeneration at the burned site to see if karrikins have worked their magic. Any new Corrigin Blue seedlings that rise from the ashes will be added insurance for the species’ future.

Dryandras differ from the typical members of the genus Banksia in having free seed capsules equipped with oil vesicles along the suture between the two halves of the capsule. The oils may contribute to the incendiary brew of karrikin precursors.

If the broad end of a dryandra seed capsule is held over a candle flame for a moment, the oil vesicles ignite explosively, causing a brief flare and an audible ‘pop’ as the halves of the capsule come apart, freeing the seed.

*In 2007, botanists Dr Austin Mast, director of Florida State University’s Robert K Godfrey Herbarium, and Dr Kevin Thiele, director of the WA Herbarium, published the results of a DNA analysis that confirmed the genus Dryandra is closely related to Banksia and had in fact arisen within Banksia. In consequence, all Dryandra species are now formally — if controversially — included in the genus Banksia, although enthusiasts are free to continue referring to members of the new sub-genus as ‘dryandras’.

Top image caption: The shrub Banksia rufa.