Genomes of eight populations of the copepod Tigriopus californicus show a correlation between rapid mitochondrial evolution and compensatory nuclear evolution, suggesting that mitonuclear incompatibilities might drive speciation in this system.
In eukaryotes, precise coordination of genes in the mitochondrial and nuclear genomes is crucial to ensure the success of biological functions such as oxidative phosphorylation and energy production. Nuclear DNA encodes proteins that function in the mitochondria and/or interact with products encoded by the few mitochondrial genes. Given that the products of mitochondrial and nuclear genomes need to interact for eukaryotes to function properly in their environments, mitochondrial–nuclear (referred to as mitonuclear) allelic variation has generally been thought to be under strong selection pressure. Nuclear genes that interact with mitochondrial genes (N-mt genes) must constantly coevolve with mitochondrial genes (mt-genes) to compensate for deleterious mutations in the mitochondrial DNA (mtDNA) and therefore rescue any dysfunction1. In fact, it has been hypothesized that mitonuclear incompatibilities can be a mechanism promoting postzygotic reproductive isolation. When gene flow is disrupted, rapid co-evolution of mt- and N-mt genes can result in mitonuclear interactions that are beneficial (or neutral) in their original genetic background, but if divergent populations come together, these co-evolved mitonuclear interactions can be incompatible and become detrimental in a hybrid background2 (Fig. 1). Mitonuclear incompatibilities have been shown in different organisms, including animals2, fungus3 and yeast4, but we still lack a complete picture of how they can drive incipient speciation. Writing in Nature Ecology & Evolution, Barreto et al.5 report that mitonuclear incompatibility is a potentially important mechanism driving intrinsic reproductive isolation in naturally diverged populations of the copepod Tigriopus californicus.
Copepods play crucial roles in aquatic environments. The species T. californicus shows extensive population divergence and strong mitonuclear incompatibility in different populations6,7,8. In addition, the rate of synonymous site substitutions of the mitochondrial genome is much higher than that of the nuclear DNA in this species (that is, 55-fold higher and the highest found in invertebrates)9, suggesting that mitonuclear incompatibility could potentially be a potent evolutionary force driving postzygotic isolation. T. californicus has thus been proposed as a model for studying the mechanisms underlying incipient speciation, but how such genomic patterns were generated across populations remains to be determined.
Using a combination of next-generation and Hi-C sequencing technologies, Barreto et al. sequenced a reference genome of T. californicus comprising ~190 Mb spanning mainly across 12 scaffolds. The authors further sequenced the genomes of seven populations that showed varying degrees of reproductive incompatibility when hybridized. T. californicus revealed levels of mtDNA amino acid differentiation among the highest seen in animals. To understand the evolutionary forces shaping such differentiation, they determined patterns of selection using two statistics, the ratio of the rate of non-synonymous to the rate of synonymous nucleotide substitutions (dN/dS) and direction of selection. Despite an overall pattern of purifying selection across the mitochondrial genome, the authors found four genes with an excess of amino acid divergence, probably evolving under positive selection. To test the hypothesis of coevolution of the nuclear genome in response to changes in the mtDNA, they estimated rates of coding sequence evolution in the eight populations. Mitonuclear coevolution would be consistent with elevated rates of amino acid changes in nuclear-encoded proteins that interact with mtDNA-encoded elements. Focal genes included mitochondrial versus cytosolic counterparts of ribosomal protein genes and within aminoacyl transfer RNA (tRNA) synthetases, different oxidative phosphorylation (OXPHOS) complexes and mitochondrially targeted proteins. The authors found elevated rates of amino acid substitutions in nuclear proteins that putatively interact with mtDNA-encoded products across multiple mitochondrial pathways, suggesting compensatory evolution. With the exception of mitochondrially targeted aminoacyl tRNA synthetases, the results remained unchanged after accounting for different levels of gene expression, indicating that elevated evolutionary rates are due to compensatory evolution, not to functional constraints on proteins. To further examine the diversifying selection, the authors also allowed dN/dS to vary among codons to test for patterns of amino acid changes consistent with positive selection. Four mitochondrial ribosomal proteins and five nuclear-encoded OXPHOS genes showed positive selection for at least one amino acid site. These genes probably evolved to compensate mtDNA divergence and are good candidates for functional studies in the future.
To test for adaptive evolution on different Tigriopus lineages, the authors applied the branch-sites model test to two other Tigriopus species whose genomes are publicly available. In the branches leading to these two copepod species, one mtDNA-interacting gene is found to be positively selected in each lineage; intriguingly, 28 nuclear-encoded mitochondrially targeted proteins exhibited positive selection on the branch leading to T. californicus. Altogether, these results reveal a correlation between elevated levels of sequence divergence in mitochondrial genomes and positive selection signatures in mtDNA-encoded genes among conspecific isolated populations. The implications are that each population may experience adaptive evolution of their nuclear genome as compensatory changes to their rapidly changing mitochondrial genome, resulting in mitonuclear incompatibility between the populations that can act as intrinsic reproductive barriers and promote speciation.
To our knowledge, Barreto and colleagues’ study provides the first genomic analysis of various degrees of mitonuclear incompatibilities across different populations. An important next step would be to study the genomes of hybrids. For instance, in the case of early stage of allopatric speciation in T. californicus, future work could generate inter-population hybrids to look into gene sequence and gene expression level changes of different generations. Furthermore, constructing mutants of candidate genes involved in mitonuclear incompatibilities would allow testing of whether such genes can contribute to or rescue mitonuclear incompatibility between populations. This study establishes T. californicus as a promising system to understand not only the evolutionary forces but also the genetic mechanisms that drive mitonuclear incompatibilities and how they contribute to speciation.
Hill, G. E. Mol. Biol. Evol. 32, 1917–1927 (2015).
Wolff, J. H. et al. Phil. Trans. R. Soc. B 369, 20130443 (2014).
Chou, J. Y. & Leu, J. Y. Bioessays 32, 401–411 (2010).
Yan, Z. et al. Fungal Gÿen et Biol. 44, 315–322 (2007).
Barreto, F. S. et al. Nat. Evol. Evol. https://doi.org/10.1038/s41559-018-0588-1 (2018).
Burton, R. S. Evolÿution 44, 1814–1822 (1990).
Edmands, S. Evolution 53, 1757–1768 (1999).
Peterson, D. L. J. Biogeogr 40, 1664–1675 (2013).
Willet, C. S. J. Mol.. Evol. 74, 310–318 (2012).
The author declares no competing interests.
About this article
Cite this article
Hui, J.H.L. Copepod incompatibilities. Nat Ecol Evol 2, 1203–1204 (2018). https://doi.org/10.1038/s41559-018-0615-2
Journal of Plankton Research (2020)