Matthew R. Goddard,1,2 Jessica Leigh,3 Andrew J. Roger3 and Andrew J. Pemberton4

“ The likelihood of these events appears greater in simpler single celled organisms, but it also seems that whist the opportunities for homing and horizontal transfer are less in higher organisms, this does not necessarily put them out of HEGs reach. It is possible that HEGs were only able to invade sea anemones because of the slow evolutionary rate of their mitochondrial DNA [27] , [28] . The extent to which HEGs have not invaded other metazoans may either be a lack of opportunity, or that most metazoans have faster substitution rates, which mean the more rapid degeneration of HEGs. ”

“ The mitochondria of animals are generally thought to be uni-parentally inherited, and not to recombine: this should deny opportunities for homing. Even if within-species HEG spread were possible for some reason, then escape from long-term degeneration would seem unlikely as one imagines horizontal transfer between metazoan germ-lines to be a prohibitively rare event.
We investigated the distribution and evolutionary genetics of this HEG within the Cnidaria in order to test the applicability of the recurrent invasion model [6] in metazoans. ”

“ For the likelihood ratio test, the log-likelihood of two data sets (e.g. A and B) is maximized under the null hypothesis where A and B share a single tree (TAB ) but each gene is allowed to have its own model and branch-length parameters (θA and θB) . The log-likelihood of this hypothesis is given by:
This work is released under the Creative Commons Attribution 2.0 license, which allows free use, distribution, and creation of derivatives, so long as the license is unchanged and clearly noted, and the original author is attributed. ”

“ This cyclical evolutionary model allows major invasions of closely related species, and predicts the patchy distribution of both functional and non-functional HEGs, and incongruence between host-genome and HEG phylogenies [6] – [8] . HEGs have been reported from a wide array of species and comparative data from phage, archaea, fungi, algae and plants support this HEG recurrent invasion model (see [1] and references within) . One clear prediction of this model is that selection will produce HEGs that are adapted for major invasion by having recognition sites that are conserved among taxa. ”

“ Overall these data suggest that the HEG within the mitochondria of the Actinaria behaves in a similar evolutionary manner to HEGs in a variety of other organisms [1] , [3] , and conforms to the recurrent invasion model [6] . The low rates of horizontal transfer among metazoans will likely ensure that HEGs degenerate long before an opportunity to ‘escape’ arises, and mean that major clade-wide invasion is prevented. ”

“ Phylogenies reconstructed exclusively from the Actinarian derived HEGs using Bayesian, likelihood and parsimony criteria all agreed (see Supporting Material 1 for model details) . A phylogeny reconstructed from the translated amino acid data was also congruent with the DNA derived phylogenies. Statistical comparisons of host and HEG data allowed us clear tests for differences in evolutionary histories. An Actinaria host tree was reconstructed using only those taxa for which we have HEG sequence (see figure 2A) . ”

“ The loss test was conducted by constructing randomised intron assignments with Mathematica (v5.2) [37] and then using PAUP* to score the number of losses. Multistate (v0.8) [20] was used to estimate the most likely transition rates for HEG gain and loss across the host phylogeny. ”