Leggy creatures and long branches - 蟲蟲

原文連結
http://www.nature.com/news/2010/100810/full/news.2010.397.html

Leggy creatures and long branches

Tracking centipedes and millipedes to their exact location in the
evolutionary tree.

Amy Maxmen

Biologists have amassed further ammunition in the debate over the position of
centipedes and millipedes among their spider, insect and crustacean kin.
They've positioned the many-legged group alongside insects and crustaceans, a
finding that could put warring zoologists at ease.

In a report published today in Proceedings of the Royal Society B (1), the
authors report the first use of microRNA as a genetic marker to clarify
relationships among arthropods — creatures like crabs, cockroaches and
centipedes, with jointed legs — and offer analyses that counter previous
hypotheses.

The arthropod family is immense, accounting for more than 80% of all known
animals and encompassing insects, crustaceans, spiders and Myriapoda —
centipedes and millipedes. Myriapods have been neglected and frequently
relocated in studies of arthropod relationships.

In the late nineteenth century, zoologists believed myriapods to be ancestral
arthropods because they looked like earthworms with legs. Soon this
hypothesis gave way to one that had myriapods nestling next to insects,
sharing a hypothetical air-breathing ancestor. With the advent of DNA
sequencing, crustaceans and insects were bracketed together, and the position
of centipedes and millipedes has been contentious ever since.

Enter Mandibulata

Head structures have emerged as a key criterion in organizing the unruly
arthropod family. Because myriapods, insects and crustaceans have similarly
segmented heads, mouth parts and mandibles, morphologists today place the
three groups together in a clade termed Mandibulata. Yet molecular
phylogenies often refute the existence of Mandibulata, instead positioning
myriapods beside spiders, scorpions and horseshoe crabs in a clade so
perplexing that it has been termed Paradoxopoda (or Myriochelata).

If the Paradoxopoda arrangement is correct, the intricate heads and mandibles
of myriapods, crustaceans and insects must have evolved multiple times
independently — or otherwise been lost without leaving a trace in spiders
and their kin. This is not impossible, but it is implausible, says Max
Telford of University College London, an author on the paper, which favours
Mandibulata. His team's work might have broken Paradoxopoda's last leg.

"This paper is absolutely a relief to morphologists because it is thoroughly
convincing and includes a lot of data coming in from different angles,"
comments Graham Budd at Uppsala University in Sweden, who was not involved in
the study.

Molecular data on arthropoda have delivered surprises in the past that
eventually gained widespread support. Insects and crustaceans were united
based on sequence data, for example, and the group is now widely accepted
because the creatures share other characteristics as well. Paradoxopoda,
however, has gained little support. Yet the debate has continued to rage. So
Telford and his team undertook a last-ditch effort to get to the bottom of
the Mandibulata mystery.

Spiders out

Using microRNA to trace evolutionary relationships, the team found a common
sequence in crustaceans, centipedes and millipedes that did not appear in
spiders, horseshoe crabs or ticks, raising support for Mandibulata.

"MicroRNAs tend to be very precise as phylogenetic markers because they are
rarely lost and they evolve very slowly," explains co-author Davide Pisani at
the National University of Ireland in Maynooth. Phylogenetic analyses of new
morphological characteristics and protein-coding genes supported the
Mandibulata as well, lending credence to earlier studies(2)(3) that came to
the same conclusions by analysing different sets of genes.

Telford's team went on to explore why some other studies had failed to find
Mandibulata. "Why did Dunn et al. get bits of their tree wrong?" Telford
asks, citing a phylogeny that supports Paradoxopoda(4). After running various
tests, the team concluded that Paradoxopoda came about, in part, as a result
of 'long branch attraction', a misleading effect caused by the way in which
data are analysed. The effect pulls together rapidly evolving lineages,
regardless of their evolutionary relationship.

Long branch attraction might be pulling the insects and crustaceans, each
with fast-evolving genes, close together and towards groups outside the
arthropods, stranding the slowly evolving myriapods with the chelicerates,
says Telford. When the researchers excluded quickly evolving genes or quickly
evolving species, support for Mandibulata increased. They performed the same
procedure on data from the paper of Dunn et al., with similar results.

With their enormous diversity, and a history dating back more than 520
million years, the arthropods push phylogeneticists to the limit. The clade
of the Mandibulata may finally be accepted, says Telford. But other nodes —
such as the position of insects relative to crustaceans, of sea spiders
relative to spiders, and of water bears and velvet worms relative to
arthropods — remain contentious.

Other regions of the animal tree might make for smoother sailing, but where's
the skill in that? Telford says, "If we make a tree and horses and zebras go
together, it could be for all the wrong reasons. But who cares — we'll never
know that our model doesn't work."

References
(1)Rota-Stabelli, O. et al. Proc. R. Soc. B doi:10.1098/rspb.2010.0590 (2010).
(2)Regier, J. C. et al. Nature 463, 1079-1083 (2010).
(3)Giribet, G., Edgecombe, G. D. & Wheeler, W. C. Nature 413, 157-161 (2001).
(4)Dunn, C. W. et al. Nature 452, 745-749 (2008).

--