Hagfish genome illuminates vertebrate whole genome duplications and their evolutionary consequences
Daqi Yu, Yandong ReniD, Masahiro UesakaiD, Alan J. S. BeavaniD, Matthieu MuffatoiD, Jieyu SheniD, Yongxin LiiD, Iori SatoiD, Wenting WaniD, James ClarkiD, Joseph N. KeatingiD, Emily M. CarlisleiD, Richard P. DeardeniD, Sam GilesiD, Emma RandleiD, Robert Stephen SansomiD, Roberto FeudaiD, James F. FlemingiD, Fumiaki SugaharaiD, Carla A. CumminsiD, Mateus PatrícioiD, Wasiu A. AkanniiD, Salvatore D’AnielloiD, Cristiano BertolucciiD, Naoki IrieiD, , Juan Pascual‐AnayaiD
Abstract
Whole genome duplications (WGDs) are major events that drastically reshape genome architecture and are causally associated with organismal innovations and radiations1. The 2R Hypothesis suggests that two WGD events (1R and 2R) occurred during early vertebrate evolution2, 3. However, the veracity and timing of the 2R event relative to the divergence of gnathostomes (jawed vertebrates) and cyclostomes (jawless hagfishes and lampreys) is unresolved4-6 and whether these WGD events underlie vertebrate phenotypic diversification remains elusive7. Here we present the genome of the inshore hagfish, Eptatretus burgeri. Through comparative analysis with lamprey and gnathostome genomes, we reconstruct the early events in cyclostome genome evolution, leveraging insights into the ancestral vertebrate genome. Genome-wide synteny and phylogenetic analyses support a scenario in which 1R occurred in the vertebrate stem-lineage during the early Cambrian, and the 2R event occurred in the gnathostome stem-lineage in the late Cambrian after its divergence from cyclostomes. We find that the genome of stem-cyclostomes experienced two additional, independent genome duplications (herein CR1 and CR2). Functional genomic and morphospace analyses demonstrate that WGD events generally contribute to developmental evolution with similar changes in the regulatory genome of both vertebrate groups. However, appreciable morphological diversification occurred only after the 2R event, questioning the general expectation that WGDs lead to leaps of morphological complexity7.
02 /Specification & Classification
Click any tag to browse matching papers- Citations
- 146
- FWCI
- 33.80
- Field percentile
- top 1%
- Organism
- Assay
- Research Focus
- Approach
In-depth summary
Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 1 (8 Apr 2023) · CC BY 4.0
Chromosome-scale inshore hagfish genome places 1R before the cyclostome-gnathostome split and 2R in the gnathostome stem, and reveals two cyclostome-specific duplications (CR1, CR2).
Whether the second vertebrate whole genome duplication (WGD), 2R, came before or after the cyclostome-gnathostome split was unresolved: hagfish had no reference genome, and the sea lamprey genome had been read as supporting conflicting scenarios. A germline hagfish assembly enables macrosynteny, gene-tree and gene retention analyses that date each duplication and test whether WGDs drove morphological diversification.
Key results
ShownPhylogenetic and chromosomal-fusion evidence supports a single 1R shared by all vertebrates. 2R is confined to the gnathostome stem lineage after it split from cyclostomes.
Among gene trees anchored to 656 orthologous gene sets, 73.2% (hagfish), 79.1% (lamprey) and ~75.7% (both) fit 1R. 2R-compatible topologies were 99.5% for elephant shark vs 19.1% hagfish, 10.6% lamprey and 8.2% cyclostome. Sea lamprey retains three of four pre-1R fusions and none of eight post-1R fusions. · Gnathostomes and cyclostomes share 1R but not 2R; Phylogenetic support around 1R/2R
ShownHagfish and lampreys have six one-to-one orthologous Hox clusters and very similar chromosome-level gene retention profiles. This shows they share two extra stem-cyclostome WGDs (CR1, CR2) on top of 1R.
Median overlapping ratio (OR) of ohnologous chromosomes: 0.49 chicken, 0.54 gar (simulated fission pairs never above 0.15), 0.30 lamprey, 0.29 hagfish. 12/16 ACs map to at least three mutually ohnologous chromosomes in each cyclostome (Spearman ρ=0.92). 52 lamprey-hagfish chromosome pairs are 1:1 orthologs (median OR=0.84). E. burgeri has 40 Hox genes in 6 clusters. · Conserved Hox cluster evolution in cyclostomes; Cyclostome-specific whole genome duplications; Definition and calculation of overlapping ratio
ShownMolecular clocks place all four WGDs between the early Cambrian and the earliest Ordovician. Even so, most chordate morphological disparity arose only after 2R, which weakens the idea that WGDs generally trigger leaps in morphology.
177 genes and 33 fossil calibrations date 1R to 536-525 Ma (crown vertebrates 510-502 Ma), CR1 to 505-494 Ma, CR2 to 491-473 Ma and 2R to 498-486 Ma, 46-53 Myr before crown gnathostomes (450-445 Ma). NMDS of 577 traits in 278 living and fossil chordates: 88-97% of vertebrate convex-hull morphospace arose after 2R. · Impact of WGD events on vertebrate morphological…; Dating species divergences; Dating genome duplications in vertebrates; Phenotypic disparity analyses
ShownAfter CR1/CR2, hagfish ohnologs parallel gnathostome ones: they are enriched for developmental functions, and many narrow their expression domains. Hagfish genes overall carry more accessible chromatin regions (ACRs) per gene than amphioxus, mostly distal and most marked for developmental genes.
Embryonic ATAC-seq (stages 45 and 53) shows more ACRs per gene than in amphioxus, mostly distal and highest at developmental ohnologs (P < 2.2 × 10−16, two-sided Wilcoxon). Over 68% of hagfish and 71% of chicken ohnolog families subfunctionalized or specialized across adult organs. The brain holds the largest share of strongly specialized hagfish ohnologs. · Increase of developmental regulatory complexity; Chromatin accessibility profiling; Fate of ohnologs after WGD; Figure legends
ShownThe germline-derived E. burgeri assembly is near-complete and has gnathostome-like coding GC content and codon usage, unlike lamprey. Phylogenomic and gene-content tests strongly favor cyclostome monophyly.
k-mer genome size 3.12 Gb; 19 Hi-C clusters plus 9,295 unplaced scaffolds/contigs; 16,513 protein-coding genes (98.3% on clusters 1-18); BUSCO 96.0% assembly, 94.2% annotation; coding GC 50.0% vs 64.0% in lamprey. AU test rejects cyclostome paraphyly (log likelihood difference 7947.7, AU = 0.004). · Chromosome-scale assembly and genome annotation; Phylogenetic position of the hagfish and gene…; Genome sequencing and assembly
Methods, data and reproducibility
Testis DNA from male E. burgeri was assembled from Illumina, Chicago and Hi-C data and annotated with Ensembl. WGD history and its effects were then tested with macrosynteny against 17 reconstructed ancestral chromosomes (ACs), AC-anchored gene trees, a gene retention 'overlapping ratio', molecular clocks, embryonic ATAC-seq and a 278-taxon morphospace.
- Germline hagfish genome: testis DNA, ~240X Illumina plus Chicago library; Hi-C (~350X valid contacts) from a second male; final assembly v4.0 · Eptatretus burgeri (inshore hagfish) · n = two adult males (sequencing; Hi-C) · ENA PRJEB21290; GenBank GCA_900186335.2 (assembly v3.2)
- Adult RNA-seq of 9 tissues (brain, gills, liver, intestine, heart, skeletal muscle, kidney, testis, blood), HiSeq 1500, 127-bp paired-end · Eptatretus burgeri · n = 9 tissues from two adult males · ENA PRJEB21290
- Embryonic ATAC-seq at Dean stages 45 and 53, ~50,000 nuclei per replicate, HiSeq 4000 · Eptatretus burgeri · n = two embryos
- Public embryonic ATAC-seq of amphioxus, zebrafish, medaka, chicken and mouse for cross-chordate ACR comparison · multiple chordates (amphioxus, zebrafish, medaka, chicken, mouse) · n = two replicates each · GSE106428; DRA006971
- Proteomes clustered to reconstruct ancestral gene family gains and losses · multiple metazoans · n = 45 animal genomes
- Phylogenomic alignment of 190 single-copy protein families (84,017 trimmed sites) for the Bayesian tree; 172 orthogroups with outgroups for clock dating · 31 chordate species (tree) plus non-vertebrate outgroups · n = 190 alignments
- Macrosynteny references: chicken, spotted gar, elephant shark, sea cucumber Apostichopus japonicus, Belcher's lancelet, sea lamprey, Pacific lamprey meiotic map · multiple deuterostomes · n = 696 orthologous gene sets (2R test)
- Morphological character matrix of extant and fossil chordates used for NMDS disparity analysis · chordates (extant and fossil) · n = 278 taxa; 577 traits (578 characters in Methods)
- Methods & tools
- LACHESIS (Hi-C scaffolding), Pilon v1.22, Ensembl gene annotation pipeline, BUSCO v5.2.2 (metazoa_odb10), PhyloBayes 4.1 (CAT-GTR), MCMCtree relaxed clock, OrthoFinder 2.3.5, ALEml_undated with CONSUL approximately unbiased test, IQ-TREE, RAxML-ng v0.9.0, MrBayes v3.2.6, DIAMOND with MCL homology clustering, Bowtie2, MACS2 and IDR (ATAC-seq peaks), dispRity with NMDS on Gower distances
- Data & accessions
- https://figshare.com/s/ca9e9b23cc36cfb59a44http://ucsc.crg.eu/https://www.ensembl.org/Eptatretus_burgeri/http://www.ensembl.org/info/genome/genebuild/2018_06_eptatretus_burgeri_genebuild.pdf
- Reagents & environment
- Pilon v1.22; BUSCO 5.2.2; MAFFT 7.402; trimAl 1.2; PhyloBayes 4.1; OrthoFinder 2.3.5; IQ-TREE v1.6.3/v1.6.12; RAxML-ng v0.9.0; PRANK v150803; MrBayes 3.2.6; MEGAX 10.2.4
- Reproducibility gaps
- No code repository is given, and the code availability statement is deferred to online content behind a placeholder DOI (XX.XXXX). No accession is given for the hagfish embryo ATAC-seq; ENA PRJEB21290 is stated to hold the genome and adult RNA-seq.
Limitations (6)
- Gene duplication-rate inference cannot tell whole genome duplications apart from other large-scale duplication mechanisms, so per-lineage duplication peaks alone do not establish WGDs. · stated by the authors
- Heavy hagfish rearrangement (at least 52 fusions) makes shared fusion signals unreliable, so fusion-based 1R timing rests on sea lamprey. The AC3 pre-1R fusion could instead reflect independent post-1R fusions (30% of simulated cases). · stated by the authors
- Hagfish rearrangements and the large lamprey haploid number prevent direct chromosome-level macrosynteny inference of ohnology and orthology. 8 lamprey-hagfish chromosome relationships remain ambiguous. · stated by the authors
- The higher ACR count per gene could reflect a loss of regulatory elements in amphioxus rather than a gain in vertebrates; this alternative was not ruled out. · stated by the authors
- Hagfish chromatin accessibility comes from only two embryos at different stages (each split into halves), which limits biological replication for cross-species ACR comparisons. · noted in this summary
- Dataset sizes differ between the main text and Methods (177 genes vs 172 orthogroups for species dating; 577 vs 578 morphological characters), which complicates exact reproduction. · noted in this summary
Who should read:Read this if you study vertebrate genome evolution, the 2R hypothesis, cyclostome biology, or how whole genome duplications relate to regulatory and morphological innovation.
Reproducibility ledger
Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.
Sign in to file a reproducibility report.
Related papers
- The hagfish genome and the evolution of vertebrates
Ferdinand Marlétaz, Nataliya Y. Timoshevskaya, … Daniel S. Rokhsar·NaturePeer-reviewed·Jan 23, 2024
- Chromosome evolution at the origin of the ancestral vertebrate genome
Christine Sacerdot, Alexandra Louis, … Hugues Roest Crollius·Genome biologyPeer-reviewed·Oct 17, 2018
- Extensive lineage-specific rediploidisation masks shared whole genome duplication in the sturgeon-paddlefish ancestor
A. K. Redmond, M. K. Gundappa, … A. McLysaght·bioRxiv·May 16, 2022
- The Sea Lamprey Meiotic Map Resolves Ancient Vertebrate Genome Duplications
Jeramiah Smith·Genome ResearchPeer-reviewed·Jun 5, 2015
- OHNOLOGS v2: A comprehensive resource for the genes retained from whole genome duplication in vertebrates
Param Priya Singh, Hervé Isambert·Nucleic Acids ResearchPeer-reviewed·Oct 10, 2019
- Horseshoe crab genomes reveal the evolutionary fates of genes and microRNAs after three rounds (3R) of whole genome duplication
W. Nong, Z. Qu, … J. H. L. Hui·bioRxiv·Apr 16, 2020