Feature: Blooming controversy

This feature appeared in the July/August 2009 issue of Australian Life Scientist. To subscribe to the magazine, go here.

Two years ago, botanists recognised 80 species of Banksia, the spectacular Australian wildflower genus named in honour of Sir Joseph Banks. But then, on February 26, 2007, Banksia suddenly exploded to more than 170 species when Dr Kevin Thiele and Dr Austin Mast used new DNA and chloroplast analysis to add the 93 members of the Western Australian endemic genus Dryandra to the Banksia genus.

Yet some traditional botanists were not impressed. It’s been long recognised that Banksia and Dryandra were closely related, but many native plant enthusiasts and nurserymen were expecteded to resist the fusion of the two genera, or at least wait and see if the molecular union becomes widely adopted. Such is the shake-up occurring in classification circles due to new laboratory-based techniques, although discoveries such as the relationship between Banksia and Dryandra reveal the potential of molecular taxonomy to aid in our understanding of the natural world.

Contradictory or complementary

Thiele, Curator at the Western Australian Herbarium, and Mast, of Florida State University in the US, based their new scheme on a comparative analysis of chloroplast and nuclear DNA sequences, and published their findings in Australian Systematic Botany in 2007.

Their DNA analysis found some lineages previously lumped within Banksia to be closer to Dryandra than to each other – for example, Banksia serrata, the founding species in the genus collected by Banks and Solander at Botany Bay, is more closely related to Dryandra than the widely distributed eastern Australian species, B. integrifolia. Rather than maintain Dryandra as a separate genus, which would have meant splitting Banksia into at least three new genera, Mast and Thiele chose the ‘minimally disruptive’ solution of sinking Dryandra within Banksia.

‘Minimally disruptive’, maybe, but many traditionalists still decided to ignore the new branch of the Banksia family tree on the grounds that few traditional botanists understand molecular systematics and the laboratory-based tools are hardly field-friendly. Superficially, it would seem to be just another collision between the comfortable familiarity of the old and the shock of the newfangled. But Thiele denies there is a dichotomy between molecular and morphological analyses – the two are complementary; and more powerful when used together than alone.

Traditional taxonomic schemes based on morphology – particularly floral characters – can be led astray by structures that appear homologous (inherited from a common ancestor) when they are actually homoplasic (independently derived by convergent evolution). Much like with beauty, homology can exist only in the eye of the beholder.

---PB---

“Molecular data has an advantage in that there’s a lot of it to work with, and there are many ways it can be interrogated to answer different questions,” says Thiele. “Another advantage is that some DNA sequences, like microsatellites, are selectively neutral and invisible to natural selection, which means taxononomic schemes are less likely to deceived by homoplasy. But if both morphological and molecular data are analysed properly, both are good data sets.”

As an example of the complementarity of molecular and morphological data, Thiele cites the congruence of Mast’s new molecular scheme with his own ground-breaking morphological analysis of Banksia a decade ago, based on highly conserved seed characters.

“To me, the interesting thing is not what type of data you analysis, but how you think about similarity and difference, which is the nub of the Banksia-Dryandra story,” he said. The old way of thinking about similarity is that all similarities are equal, so that it’s possible to do a Jussieuian averaging of similarity to come up with a set of relationships.

“The new realisation since Hennig is that some similarities are logically more equal than others. In fact, some similarities should be positively ignored no matter how in-your-face they are because they are more likely to reflect homoplasy.

“A similarity in a tiny, obscure feature, such as a stomata or a DNA base-pair, that co-occurs in two taxa because it evolved as a novelty in their common ancestor, is vastly more important in establishing a relationship than any obvious similarity that the two taxa might share simply because it’s a primitive trait that has been lost by other members of the same genus.

“This realisation, that some similarities are informative of evolutionary relatedness while others are not – or are informative in a different way – is enormously important.”

Whether a scheme is based on DNA or physical traits, or both, it’s still up to humans to draw the lines in taxonomic schemes, guaranteeing there will be arguments about what’s in and what’s out.

For Thiele and Mast, DNA homologies placed Dryandra firmly within Banksia. By sinking Dryandra within Banksia, they rendered familiar scientific names in hundreds of scientific and popular texts obsolete. Dozens of new combinations must be memorised and native plant nurseries must print new labels. It promises to cause as much of a mess as did the demotion of Pluto to a ‘dwarf planet’ a few years back.

In compensation, Thiele and Mast’s new scheme illuminates some formerly enigmatic relationships between species and lineages within the enlarged genus and indicates roughly when the major lineages diverged. DNA data adds a new dimension – time – to taxonomic schemes. Morphology and DNA can both indicate how species are related, but only DNA can indicate when they diverged from a common ancestor.

---PB---

Reconciliation

Canberra molecular systematist Dr Bernard Pfeil, of the Centre for Plant Biodiversity Research, concedes the new science is “still under debate to some degree.” According to Pfeil, “some traditional morphological systematists see molecular systematics as a bandwagon driven by the technology and question whether it provides all the answers. But others are quite enthusiastic.

“My own view is that molecular systematics has the advantage that DNA is inherited independently of environmental influences that influence things like leaf shape, and flower colour. We don’t understand the genetic basis of many of these traits, so we can’t be certain that what we are seeing in the plant is a consistent diagnostic character. Going straight to genotype provides a lot of power.”

Pfeil says that before molecular systematics came to the fore in the late 1980s, researchers used proteins to resolve evolutionary relationships. Early DNA studies focused on single genes, or chloroplast sequences.

“They were very accurate for the selected region of the genome, but genomes are not monolithic entities. People at the leading edge of molecular systematics then concluded that genomes should be viewed as mosaics, and each gene provides only a part of the picture.

“Particular populations can be impacted by hybridisation with sister species, so the morphological characters selected by the taxonomist may no longer be useful in resolving species relationships. So you can’t rely on just one gene from one individual – you need to use multiple genes and take more of a statistical approach to determine what constitutes a distinct species. And you have to keep in mind that what is expressed in the plant right now might not be expressed in all phases of development or in all environments in which the plant grows.”

Just over two years ago, Pfeil began work on the systematics of Australasia’s native citrus species. Like oranges, mandarins, lemons and limes, all belong to Aurantioideae sub-family of the family Rutaceae. Pfeil’s former boss, Dr Randy Beyer, had accumulated the largest data set of chloroplast DNA from cultivated citrus and their wild relatives. “Any genus that has been associated with cultivated citrus seems to be nested within Citrus,” says Pfeil.

“Randy’s data set covered 10 kilobases of sequence data from all but one genus in the Aurantoideae sub-family. Biogeographically, the sub-family is centred in Australasia – the Australian genera included Eremocitrus and Microcitrus and the latter also occurs in south-eastern New Guinea. Another genus, Clymenea, occurs in northern New Guinea and New Ireland, and New Caledonia has an endemic genus, Oxanthera.

---PB---

“This raises the question of when the divergences occurred, and whether they coincided with the fragmentation of [the Australasian] parts of Gondwana. The simple approach is to determine when they diverged, and check it against the geological timing. My DNA data indicates the divergences were very recent compared with the geological timetable, so it appears that trans-oceanic dispersal was involved.

“Given that all the close relatives of the Australian species are south-east Asian, the direction of dispersal seems to have been from Asia to Australasia. If you follow the biogeographic evidence and think about how they dispersed, it’s easy to fall into the trap of thinking that the Australasian species haven’t changed very much from their Asian ancestors. But we need to consider the possibility that the fruits and seeds might not always have been the same sizes and forms that we see today. Another consideration is that dispersal may have resulted from cataclysmic events like cyclones.”

Pfeil says that events of the magnitude, location and timing required to disperse large-fruited or large-seeded taxa like citrus are probably extremely rare, but just a few successful dispersals could explain the latter-day distribution of the Australasian species. Chloroplast DNA data indicate that Australasian citrus species, Eremocitrus and Microcitrus are most closely related to each other, and to Asian taxa like Fortunella and Poncirus.

Remarkably, Pfeil and Dr Mike Crisp, from ANU, have estimated from chloroplast DNA that the Australian taxa diverged from Asian relatives only 5.8 million years ago. “So we’ve probably had a single introduction to the Australian plate, which included south-east New Guinea in prehistoric times,” says Pfeil. “But it’s not yet clear whether Clymenea and Oxanthera descended from Australian ancestors, or originated in a separate dispersal event.”

The molecular systematics revolution continues apace: in the Myrtaceae, another old southern family, Melaleuca is likely soon to absorb related ‘bottlebrush’ genera including Callistemon, Regelia and Beaufortia. The Myrtaceae is already a battleground. In 2000, Professor Pauline Ladiges, of the University of Melbourne’s School of Botany, and her former PhD student, Dr Frank Udovicic, split the bloodwoods and ghost gums from Eucalyptus, and assigned them to the new genus Corymbia, treating them as a sister group Angophora, to the native ‘apples’.

Ladiges’ DNA evidence suggests Corymbia and Angophora parted company with true Eucalyptus 60-65 million years ago, but taxonomist Dr Ian Brooker, of the Australian National Herbarium, argues that both should be reinstated within Eucalyptus.

Controversies such as this are sure to continue as new technologies allow us to peer under the morphology of plants and animals, revealing previously imperceptible similarities and differences. That’s not to say the traditional botanists are out of a job. It’ll take a fusion of approaches to unravel the complex relationship between various species, a fusion that will only improve our understanding of the natural world.

This feature appeared in the July/August 2009 issue of Australian Life Scientist. To subscribe to the magazine, go here.