Rapid evolution of mitochondrion-related genes in haplodiploid arthropods

Abstract

Background: Mitochondrial genes and nuclear genes cooperate closely to maintain the functions of mitochondria, especially in the oxidative phosphorylation (OXPHOS) pathway. However, mitochondrial genes among arthropod lineages have dramatic evolutionary rate differences. Haplodiploid arthropods often show fast-evolving mitochondrial genes. One hypothesis predicts that the small effective population size of haplodiploid species could enhance the effect of genetic drift leading to higher substitution rates in mitochondrial and nuclear genes. Alternatively, positive selection or compensatory changes in nuclear OXPHOS genes could lead to the fast-evolving mitochondrial genes. However, due to the limited number of arthropod genomes, the rates of evolution for nuclear genes in haplodiploid species, besides hymenopterans, are largely unknown. To test these hypotheses, we used data from 76 arthropod genomes, including 5 independently evolved haplodiploid lineages, to estimate the evolutionary rates and patterns of gene family turnover of mitochondrial and nuclear genes.

Results: We show that five haplodiploid lineages tested here have fast-evolving mitochondrial genes and fast-evolving nuclear genes related to mitochondrial functions, while nuclear genes not related to mitochondrion showed no significant evolutionary rate differences. Among hymenopterans, bees and ants show faster rates of molecular evolution in mitochondrial genes and mitochondrion-related nuclear genes than sawflies and wasps. With genome data, we also find gene family expansions and contractions in mitochondrion-related genes of bees and ants.

Conclusions: Our results reject the small population size hypothesis in haplodiploid species. A combination of positive selection and compensatory changes could lead to the observed patterns in haplodiploid species. The elevated evolutionary rates in OXPHOS complex 2 genes of bees and ants suggest a unique evolutionary history of social hymenopterans.

Keywords: Evolutionary rate; Gene duplication; Gene family evolution; Hymenoptera; Mesostigmata; Oxidative phosphorylation genes; Phthiraptera; Thysanoptera; Trombidiformes.

Fig. 1

Distribution of haplodiploidy in the phylogeny of 76 arthropod species. The tree is based on the time tree of Thomas et al. [32]. Hymenopterans are colored in green; non-Hymenoptera species are colored in blue; diploid species are in red. Different types of haplodiploidy are in parenthesis. Names of major arthropod orders with more than one species in the tree are next to the corresponding clade

Fig. 2

Root-to-tip evolutionary rates (AA/site/MY) of different gene categories among arthropod groups. Asterisks indicate significant differences among hymenopterans (green), other non-Hymenoptera haplodiploid species (blue), or diploid species (red). Note that the y-axis of mtOXPHOS genes is different from nuclear genes due to the elevated evolutionary rates of mitochondrial genes

Fig. 3

Spearman’s rank correlation of evolutionary rates (AA/site/MY) between mtOXPHOS genes and A nucOXPHOS, B nucMTRP, E nucCRP, F nucControlSingle, and G nucControl genes based on root-to-tip evolutionary rate. In addition, we also estimated the correlation of evolutionary rates between mtOXPHOS genes and C nucOXPHOS genes not from complex 2 and D genes only from complex 2. Spearman’s correlation coefficient (R) and p values (p) are used to estimate the correlation between evolutionary rates of mtOXPHOS genes and evolutionary rates of nuclear-encoded gene categories based on all four arthropod groups (black), diploid species (red), hymenopterans (bees and ants: green, sawflies and wasps: dark green), and other non-Hymenoptera haplodiploid species (blue). Linear regression represents the general trend of the correlation

Fig. 4

Root-to-tip evolutionary rates of different gene categories among hymenopterans. Asterisks indicate significant differences between bees and ants (green) and sawflies and wasps (dark green)