Phylogenomic Studies on the Evolution of Morphological Complexity

National Human Genome Research Institute

Investigators

Linked publications, trials & patents

Abstract

Genomic sequencing of non-bilaterian animal species has provided invaluable insight into the molecular innovations that have fueled the outbreak of diversity and complexity in the early evolution of animals. The cnidarians are organisms unified in a single phylum based on their use of cnidocytes (stinging cells) for capturing prey and defense from predators, and they occupy a key phylogenetic position as the sister group to bilaterian animals. Cnidarian genomes are remarkably similar to the human genome in terms of gene content and structure, and what makes these organisms particularly attractive for study is our observation that the genomes of cnidarians encode more homologs to human disease genes than do classic invertebrate models such as Drosophila and C. elegans (1). We are leading efforts to establish selected cnidarians as new model organisms that have the potential to inform important questions in human biology and human health, laying the groundwork for translational studies focused on specific human diseases. We have sequenced and annotated the genomes of two Hydractinia species (manuscript in preparation). What makes these simple organisms particularly well-suited as a model system lies in the fact that they possess a specific type of interstitial cell (an i-cell) that is pluripotent and provides the basis for tissue regeneration, expressing genes whose bilaterian homologs are known to be involved in stem cell biology. Hydractinia is also colonial, possessing an allorecognition system that may provide insights into important questions related to host-graft rejection. Our sequencing approach involved performing both PacBio long-read sequencing and Dovetail long-range scaffolding, yielding very high coverage for both genomes; the N50 value of the assembled H. symbiolongicarpus genome makes it one of the most contiguous invertebrate genomes sequenced to date. The vast majority of a set of evolutionarily conserved single-copy orthologs can be easily identified in these assemblies, and analyses of these whole-genome sequencing data have already provided important insights into the evolution of chromatin compaction (2) and animal neurogenesis (3). Ribosome Biogenesis. We have determined the consensus sequence of rDNA repeats in Hydractinia, possessing 2,500 rDNA repeats as compared to 200-600 rDNA repeats in human. Its 18S, 5.8S, and 28S genes, as well as its external (ETS) and internal transcribed spacers (ITS), all appear in the same order as in human. However, the Hydractinia 28S gene is more than 1 kb smaller than in human, its ETS and ITS are compressed, and its intergenic spacer is 100 times shorter than in human. This structure may reflect an evolutionary adaptation necessary to meet the high demand for ribosomes (and, in turn, protein production) during regeneration. Given the structure of Hydractinia's rDNA arrays, we surmise that they may be under the control of a single promoter. Further, protein domain structural analyses indicate that Hydractinia does not possess the canonical UBF protein; its absence suggests that Hydractinia may employ a different mechanism for regulating transcription of rDNA genes than that used by higher eukaryotes, perhaps providing important insight as to the regenerative capacity of this organism. A comparison of de novo transcriptomes generated by our group across a wide taxonomic range indicates the canonical UBF protein is not present in non-bilaterians, suggesting the involvement of a novel protein or UBF precursor. An outgrowth of this study is the Animal Proteome Database (https://research.nhgri.nih.gov/aniprotdb), a resource that provides open access to the 100 proteomes that we generated from raw transcriptomic data found in the NCBI Sequence Read Archive (SRA) during the course of this work (4). Our pipeline for generating these proteomes has been made publicly available, and the web site vastly increases the utility of these data by removing the barrier to access for research groups who do not have the expertise or access to computational resources such as NIHs Biowulf supercomputer to run the requisite pipelines themselves. Allorecognition. The availability of our Hydractinia sequence data has enabled us to characterize the genomic structure of the allorecognition complex (ARC) in greater detail, revealing its inherent complexity. Bioinformatic analyses of our highly contiguous whole-genome sequence data has revealed there are 10 putative Alr genes located within the 12 Mb allorecognition complex, with fusion assays indicating that Alr4 is a putative third allodeterminant (manuscript in preparation). Fusion assays performed in collaboration with colleagues at the University of Pittsburgh identified an additional polymorphic gene (Alr4), with ongoing studies focused on identifying additional putative allodeterminants in Hydractinia. This work demonstrates that the extensive degree of gene duplication and sequence diversification within the ARC is consistent with general trends observed in other invertebrate self/non-self recognition systems. Data Sharing. We have developed the Hydractinia Genome Project Portal, located at https://research.nhgri.nih.gov/hydractinia. The scope of data available through the Portal goes well-beyond the sequence data available through GenBank, providing additional biological information intended to increase the utility of the sequencing data generated by our group. It also provides a customized, interactive JBrowse front-end for visualizing assemblies, gene predictions, assembled, transcripts, predicted functional domains, non-coding RNA sequences, and methylation data from both species. The structure of the database parallels that of our Mnemiopsis Genome Project Portal (5), providing access to value-added data that is not available elsewhere, and we are currently using the same approach in building and maintaining an allied public portal for genomic data being generated by investigators studying the highly regenerative cnidarian Hydra, located at https://research.nhgri.nih.gov/hydra. Identification of Conserved Non-Coding Elements (CNEs) in Animal Genomes. At the genomic level, spatiotemporal regulation of gene expression relies on non-coding regulatory elements that contain clusters of transcription factor binding sites and regulate transcription of their target genes at a distance of up to hundreds of kilobases. Unlike for genes, there are no computational methods to identify these regulatory regions based solely on their DNA sequence, but it is still possible to identify candidate regulatory sequences by searching for non-coding DNA regions that have been under purifying selection over large evolutionary timescales. We are performing a large-scale identification of CNEs across animal genomes and characterizing the distribution of CNEs at a range of evolutionary time scales. The emergence of more efficient k-mer based algorithms for CNE identification, along with our ability to take advantage of NIHs Biowulf supercomputing resource, has made it possible to perform thousands of whole-genome pairwise comparisons efficiently and on a reasonable timescale. We hypothesize that genes that travel in close proximity to the same CNEs over large evolutionary time may be the transcriptional targets of these CNEs. Finally, we are assessing the distribution of each CNE across the animal tree to determine at which node they likely arose, using these data to identify candidate CNEs that may be involved in the evolution of taxon-specific traits. (1) Maxwell et al. BMC Evol. Bio. 14: 212, 2014 (2) Torok et al., Epigenetics & Chromatin 9: 36, 2016 (3) Gahan et al., Dev. Biol. 428: 224-231, 2017 (4) Barreira et al., Mol. Biol. Evol., doi:10.1093/molbev/msab165, 2021 (5) Moreland et al., Database 1-9, doi:10.1093/database/baaa029, 2020