Evolution of the Mitochondrial Cytochrome Oxidase II Gene in Collembola

Abstract

Abstract. The sequence of the mitochondrial COII gene has been widely used to estimate phylogenetic relationships at different taxomonic levels across insects. We investigated the molecular evolution of the COII gene and its usefulness for reconstructing phylogenetic relationships within and among four collembolan families. The collembolan COII gene showed the lowest A + T content of all insects so far examined, confirming that the well-known A + T bias in insect mitochondrial genes tends to increase from the basal to apical orders. Fiftyseven percent of all nucleotide positions were variable and most of the third codon positions appeared free to vary. Values of genetic distance between congeneric species and between families were remarkably high; in some cases the latter were higher than divergence values between other orders of insects. The remarkably high divergence levels observed here provide evidence that collembolan taxa are quite old; divergence levels among collembolan families equaled or exceeded divergences among pterygote insect orders. Once the saturated thirdcodon positions (which violated stationarity of base frequencies) were removed, the COII sequences contained phylogenetic information, but the extent of that information was overestimated by parsimony methods relative to likelihood methods. In the phylogenetic analysis, consistent statistical support was obtained for the monophyly of all four genera examined, but relationships among genera/families were not well supported. Within the genus Orchesella, relationships were well resolved and agreed with allozyme data. Within the genus Isotomurus, although three pairs of populations were consistently identified, these appeared to have arisen in a burst of evolution from an earlier ancestor. Isotomurus italicus always appeared as basal and I. palustris appeared to harbor a cryptic species, corroborating allozyme data.

Phylogenetic Relationships Among the Phrynosomatid Sand Lizards Inferred from Mitochondrial DNA Sequences Generated by Heterogeneous Evolutionary Processes

“…First codon positions showed the lowest bias (C = 0.042), whereas high T and low A and G frequencies at second positions resulted in an intermediate bias (C = 0.236), and high A and low G frequencies at third positions resulted in the most bias (C = 0.293) among codon positions. General patterns of bias for each codon position observed here are similar to those reported for mitochondrial genes sequences from birds (Kornegay et al, 1993;Nunn and Cracraft, 1996;, mammals (Irwin et al, 1991), and coUembolans (Frati et al, 1997). Corrected (likelihood under the GTR + I + F model) pairwise distances (Table 3) between and within currently recognized sand lizard species ranged between 0.009 and 0.508 and between 0.002 and 0.208, respectively.…”

DNA sequence analysis to study the evolution of Antarctic Collembola

“…For protein coding genes, the alignment was obtained by comparing the inferred amino acid sequences, while, for rRNA sequences, it was manually improved using conserved fragments of important secondary structure features. Phylogenetic analysis was performed with the software PAUP* (Swofford, 1998) using Maximum Likelihood (ML), Maximum Parsimony (MP) and Minimum Evolution (ME), following the strategy outlined in Swofford et al (1996), Frati et al (1997b) and Sullivan et al (1997).…”

Phylogenetic evidence for extensive horizontal gene transfer of type III secretion system genes among enterobacterial plant pathogens

“…All datasets were first analysed in PAUP* (Swofford, 2001) using the principle of maximum parsimony (MP) under heuristic search methods using tree bisection-reconnection (TBR) branch-swapping, and 100 random taxon addition replicates. For the nucleotide datasets, the shortest trees were then used as the starting topology for the evaluation of 16 nested models of sequence evolution (Frati et al, 1997;Sullivan et al, 1997;Swofford et al, 1996). The best-fit model of substitution for each dataset was determined by the likelihood score of each under a chi-squared approximation of the null distribution (Yang et al, 1995).…”