By Kevin E. Noonan --
The coelacanth, an aquatic animal described as a "living fossil" when discovered in 1938, was thought to have gone extinct during late Cretaceous period, ~70 million years ago. Only about 300 specimens of the African coelacanth, Latimeria chalumnae, are known to exist and a second species, Latimeria menadoensis, was discovered in 1997. These animals are morphologically primitive, resembling fossils dating ~300 million years ago, which has suggested that these are "slowly evolving" species (although this is more a descriptive than an informed characterization). There have been sporadic reports of coelacanth gene sequencing over the past decade, but last week Nature published the first report on whole genome sequencing (Ameniya et al., "The African coelacanth genome provides insighted into tetrapod evolution," Nature 496: 311-16, April 18, 2013).
The article reports that the coelacanth genome comprises 2.68 gigabases, on 48 chromosomes, with 19,033 protein-coding genes comprising 21,817 transcripts, 2,894 "short" non-coding RNAs, and 1,214 "long" non-coding RNAs. 336 of protein-encoding genes were found to have undergone gene duplication. The authors examined transcripts from 251 genes from coelacanth and compared these genes with Protopterus annectens (the African lungfish) expressed in brain, gonad, and kidney and with 15 terrestrial vertebrate lineages (dog, human, mouse, elephant, armadillo, Tammar wallaby, opossum, platypus, chicken, turkey, zebra finch, lizard, Western clawed frog, Chinese brown frog) and five modern fish species (tilapia, pufferfish, zebra fish, spotted catshark, little skate, and elephant shark). The results of these analysis, comprising 100,583 concatenated amino acid positions, was consistent with the lungfish being the earliest relative of modern tetrapods (four-limbed animals), answering a previously unresolved uncertainty as to the role of the coelacanth in vertebrate evolution.
Turning to the question of the status of the coelacanth as a "living fossil," the authors confirmed earlier conclusions (based on Hox genes and protocadherins) that this species evolves slowly, the authors assessed the data from the 251 genes used in their phylogenetic analyses. This was done as follows:
Pair-wise distances between taxa were calculated from the branch lengths of the tree using the two-cluster test proposed previously to test for equality of average substitution rates. Then, for each of the following species and species clusters (coelacanth, lungfish, chicken and mammals), we ascertained their respective mean distance to an outgroup consisting of three cartilaginous fishes (elephant shark, little skate and spotted catshark). Finally, we tested whether there was any significant difference in the distance to the outgroup of cartilaginous fish for every pair of species and species clusters, using a Z statistic.
The coelacanth genes showed 0.89 substitutions per site, compared with 1.05 substitutions/site in the lungfish, 1.09 substitutions/site in chicken, and 1.21 substitutions/site in mammals.
The authors also ascertained the "abundance" of transposable elements in the coelacanth genome, because these elements are believed to provide "templates for exaptation," i.e., to facilitate formation of novel protein exons and regulatory elements, as well as providing targets for genomic rearrangement. Transposable element content was "high" (~25%, which the authors consider an underestimate) and also showed a "wide variety of transposable-element superfamilies," with 14 such families being "currently active." The authors acknowledge that these results "contrast with the slow protein evolution observed."
"[E]xtensive conservation of synteny" was observed in a comparison of chromosomal breakpoints in coelacanth and tetrapod genomes, and "indicate that large-scale rearrangements have occurred at a generally low rate in the coelacanth lineage." Interchromosomal rearrangements indicated that "karyotypic evolution in the coelacanth lineage has occurred at a relatively slow rate, similar to that of non-mammalian tetrapods" (31 in coelacanth, 20 in salamander and 21 in Xenopus species). Comparison of the two coelacanth species, L. chalumnae (Africa) and L. menadoensis (Indonesia), showed divergence rates similar to those found between humans and chimpanzees, and the authors estimated that these species diverged "slightly more than" 6-8 million years ago, based on the slower substitution rates found in coelacanth species.
The authors then looked at estimates of how vertebrates adapted to the terrestrial environment. They identified 50 genes found in coelacanth but not terrestrial tetrapods, presumably because these genes were not needed when the animal left water for land. These genes included "components of fibroblast growth factor (FGF) signalling, TGF-β and bone morphogenic protein (BMP) signalling, and WNT signalling pathways, as well as many transcription factor genes," and specifically that 4 genes (And1, And2, Fgf24 and Asip2) not present in tetrapod genomes were indeed present in the coelacanth genome. These genes also included 13 genes involved in fin development, 8 genes in otolith and ear development, 7 genes for kidney development, 13 genes for eye development, and 23 genes for brain development. In contrast, there were only slight differences in homeobox genes. There were also changes in gene regulation, wherein 6% of "conserved non-coding elements (CNEs)" ("promoters, enhancers, repressors and insulators") had originated after divergence of the coelacanth from the ancestral lineage. Further analysis showed that tetrapod-specific CNEs were most closely (genetically) linked to genes involved in smell perception (consistent with the recognized expansion of olfactory receptor genes in the evolution of tetrapods from teleost fishes) and "morphogenesis (radial pattern formation, hind limb morphogenesis, kidney morphogenesis) and cell differentiation (endothelial cell fate commitment, epithelial cell fate commitment)" and immunoglobulin VDJ recombination.
A particular set of genes compared in the study are genes for digits, "[a] major innovation in tetrapod evolution," and specifically Hox genes for regulating limb development in ray-finned fish, coelacanths and tetrapods (mouse). Three of the six "cis-regulatory elements" showed sequence conservation limited to tetrapods, with one element being shared by tetrapods and coelacanth but not the ray-finned fish; this latter element could function in transgenic mouse assays to "drive reporter gene expression in a limb-specific pattern." Another particular set of genes compared between tetrapod and coelacanth lineages were genes for the urea cycle, because "[e]xcretion of nitrogen is a major physiological challenge for terrestrial vertebrates." Urea cycle genes involved in producing urea for nitrogenous waste disposal (such as carbamoyl phosphate synthase I) showed strong evidence of selection whereas genes (such as mitochondrial arginase) involved in arginine metabolism but not excretion did not show such selection. The authors conclude that this is evidence of adaptive evolution in the transition from water to land. Hox gene studies also indicated changes from the coelacanth and tetrapod lineages implicated in placental and other reproductive structures not found in animals living in an aquatic environment. Finally, the coelacanth genome lacks genes for immunoglobuins of the M class but did possess two IgW genes previously found only in lungfish and certain cartilaginous fish.
In addition to establishing that lungfish not the coelacanth was the common ancestor to all terrestrial vertebrate, the authors also established that the coelacanth also showed a slow rate of evolutionary change, which they speculate might be due to "a static habitat and lack of predation" and promising that "[f]urther study of these changes between tetrapods and the coelacanth may provide important insights into how a complex organism like a vertebrate can markedly change its way of life."
The authors were affiliated with the following institutions: The Broad Institute at MIT; Benaroya Research Institute and University of Washington; University of Konstanz, Germany; Universite de Montreal; University of Oregon; Institute of Molecular and Cell Biology, Singapore; Universidade Federal do Para, Brazil; Harvard University; University of Utah; Ecole Normale Superieure de Lyon; University of Kentucky; Rhodes University, South Africa; Wellcome Trust Sanger Institute; University of Trieste; University of Liege; Victoria University, Australia; University of Tuscia; University of Hamburg; Polytechnic University of Marche, Italy; University of South Florida; University of Western Cape, South Africa; Woods Hole Oceanographic Institution; Oxford University; Universitat Leipzig; Keio University, Japan; The Graduate University for Advanced Studies, Japan; European Molecular Biology Laboratory; University of Wuerzburg, Germany; University of Illinois at Chicago; National Institute of Genetics, Japan; and Uppsala University.