The relationship between the rate of molecular evolution and the rate of genome rearrangement in ani

1
The relationship between the rate of molecular
evolution and the rate of genome rearrangement in
animal mitochondrial genomes.

• Wei Xu(1), Daniel Jameson(2), Paul G Higgs(1). 
• Department of Physics, McMaster University,
  Hamilton, Ontario.
• School of Biological Sciences, University of
  Manchester, UK.

Background to this project Phylogenetic Artefacts
Gene sequences are often used in molecular
phylogenetics studies to deduce the evolutionary
relationship between species. This is difficult
with mitochondrial genomes because of the
presence of rapidly evolving problem species
with very divergent sequences (which leads to
long branch attraction), and due to the wide
variation in the frequency of bases and amino
acids among species (which causes biases in
trees). Can Gene Order Help? Gene order
rearrangements sometimes provide strong evidence
of shared ancestry of a group, e.g. the
translocation of tRNA-Leu shown in the Drosophila
genome (see 2) is a derived character shared by
many other insects and crustaceans, and this
supports the existence of the Pancrustacea clade.
The gene order of Rhipicephalus is shared by two
other ticks and no other species, which is a
strong signature of the relationship of this
group. However, there are also species, such as
Tigriopus, with extremely scrambled genomes,
where gene order tells us little. Sequence
Evolution and Genome Rearrangement are Related
Here we will show that species with high rates of
sequence evolution also tend to have high rates
of genome rearrangement. Problem species in
molecular phylogenetics also tend to be problems
in gene order studies.
3
Chelicerata
4
Methods A consensus tree topology for arthropods
was obtained from morphological evidence,
published molecular phylogenies and our own
analysis of mitochondrial sequences. The base of
the pancrustacea (P) was left as a multifurcation
as there is no reliable consensus. Maximum
Likelihood branch lengths were obtained using
this fixed topology. The protein tree (left) is
derived from a concatenation of 4 mitochondrial
proteins. The tRNA tree (right) is derived from a
concatenation of 22 mitochondrial tRNAs. For each
species, the total branch length from the root of
the arthropods (A) to the tip was measured (see
Table 1). Rates of sequence evolution vary
substantially between species. It is thought that
the ancestral gene order (at A) is the same as
Limulus. Therefore, the break point and inversion
distances from Limulus to each species were
measured.
Myriapoda
Typical animal mitochondrial genomes contain 13
protein-coding genes, 2 rRNAs and 22 tRNAs. OGRe
produces comparisons of mitochondrial gene orders
for any two species. The examples below show
comparisons between the Horseshoe crab, Limulus
polyphemus and three other arthropods. These
genomes a circular the two ends are connected
but they are shown as linear for convenience.
Each gene is shown as a block labelled by its
gene symbol. Single letter abbreviations are for
tRNA genes. Genes drawn below the central line
are transcribed from left to right. Genes drawn
above the line are transcribed from right to left
(and a sign is added to the gene symbol).
2
Crustacea
P
Hexapoda
5
Results Table 1 shows the two gene-order
distances and the two sequence-based distances
between the ancestral arthropod and each current
species. The number of deleted or duplicated
genes with respect to the ancestor is also shown.
Species are classed into four categories
according to breakpoint distance (shown by
colour). The figure on the right shows that all
four distance measures are positively correlated
with each other. The correlation coefficients are
shown by each graph. This is also demonstrated by
Table 2, which shows the minimum, mean and
maximum of the sequence-based distances in each
of the categories. Species with high break point
distances also have high tRNA and protein
distances. It is found that tRNA genes are more
frequently translocated than rRNA or protein
genes. There are many species where only tRNAs
have moved. This includes 9 species whose
breakpoint distance is in the High category and
21 species in the Moderate or Low break point
categories. In Table 2, the two bottom rows show
that, even when only tRNAs have moved, there are
higher tRNA and protein distances for species
with higher breakpoint distances. This means that
high rates of tRNA translocation are correlated
with increased rate of evolution in tRNA and
protein genes.
Acknowledgements This work is supported by
Canada Research Chairs and NSERC.
Discussion These results show that the rates of
both sequence evolution and genome rearrangement
are very non-clocklike. Species with high
evolutionary rates often have close relatives
with much lower rates. This means that rates have
increased in scattered lineages independently.
For example Holometabolous insects Bees
(Apis, Melipona) gtgt Beetles (Tribolium,
Crioceris) Hemiptera (bugs) Aleurodicus gtgt
Triatoma Maxillopod crustaceans Tigriopus,
Argulus gtgt Tetraclita, Pollicipes Spiders
Ornithoctonus, Habronattus gtgt Heptathela This
suggests a breakdown in accuracy of mitochondrial
genome replication in the fast-evolving lineages
that causes higher mutation rate and higher
susceptibility to major rearrangements, but it is
also possible to envisage selective explanations,
such as rapid adaptation to a new
environment. There are still many questions that
we do not understand. Why do tRNAs move much more
frequently than larger genes? Why are there many
examples of long jump translocations of tRNAs?
Why are there many examples of genes whose
positions are reshuffled but remain on the same
strand (no inversions)? We hope that further
analysis of the OGRe gene order database will
give some clues to these questions.
Table 1
Table 2