Transposable Elements Facilitate the De Novo Origin of Antifreeze Protein and the Diversification of Its Gene Family in Snailfishes

Transposable elements (TEs) are increasingly recognized as important sources of genomic innovation, yet mechanistically resolved examples of how they help generate new functional genes in vertebrates remain rare. Type I antifreeze proteins (AFPI) in fishes are life-saving adaptations shaped by strong freezing selection and provide an exceptional system for studying new gene evolution under extreme environmental pressure. We recently showed that AFPI in flounder, cunner, and sculpin evolved independently through distinct partial de novo routes, converging on a nearly identical alanine-rich antifreeze protein. Here, we elucidate the origin and evolution of AFPI in the last remaining unresolved lineage, snailfishes, using a chromosome-scale genome assembly for Liparis atlanticus together with multi-tissue Iso-Seq, tissue-specific RNA-seq, and comparative genomics across AFPI-bearing and AFPI-lacking snailfishes and teleost outgroups. We show that snailfish AFPI originated within Liparis and rapidly diversified as a young gene family with multiple isoforms and lineage- and population-specific copy-number change. Genome-wide homology searches support a de novo origin of the alanine-rich coding region from noncoding sequence rather than from a pre-existing protein-coding precursor. In contrast, the surrounding regulatory architecture was assembled through sequence recruitment: a hAT-derived fragment contributes promoter- and transcription-start-site-proximal sequence, and a conserved noncoding segment together with a Ty3/Gypsy-derived long terminal repeat (LTR) contributes the 3' regulatory region. TE-rich locus structure also provides plausible mechanisms for subsequent locus expansion and translocation. Together, these results reveal a TE-facilitated, mosaic route to new gene evolution in vertebrates, demonstrating how noncoding DNA, repetitive sequence, and TE-derived regulatory fragments can be assembled into a strongly selected adaptive innovation. Author SummaryWhere do new genes with brand-new functions come from? We tackled this question using one of evolutions clearest natural experiments: antifreeze proteins, life-saving molecules favored by selection because fish without them freeze in icy seawater. In this study, we show that mobile DNA called transposable elements helped build a new antifreeze gene in stages. Different transposable elements appear to have played different roles: one helped switch on a previously silent stretch of noncoding DNA, others contributed control sequences at the beginning and end of the gene, and repeat-rich DNA around the locus likely promoted gene duplication, movement to a new chromosomal location, and rapid diversification into a gene family. This is an unusually clear vertebrate example of how a new gene can emerge not in a single leap, but through stepwise assembly from different pieces of the genome. More broadly, our work shows that transposable elements do much more than disrupt genomes. Under strong natural selection, they can help turn noncoding DNA into a life-saving adaptation and then help that innovation expand and diversify.