Endogenous bornavirus-like nucleoprotein elements (EBLNs), the nucleotide sequence elements derived from the nucleoprotein gene of ancient bornavirus-like viruses, have been identified in many animal genomes. Here we show evidence that EBLNs encode functional proteins in their host. Some afrotherian EBLNs were observed to have been maintained for more than 83.3 million years under negative selection. Splice variants were expressed from the genomic loci of EBLNs in elephant, and some were translated into proteins. The EBLN proteins appeared to be localized to the rough endoplasmic reticulum in African elephant cells, in contrast to the nuclear localization of bornavirus N. These observations suggest that afrotherian EBLNs have acquired a novel function in their host. Interestingly, genomic sequences of the first exon and its flanking regions in these EBLN loci were homologous to those of transmembrane protein 106B (TMEM106B). The upstream region of the first exon in the EBLN loci exhibited a promoter activity, suggesting that the ability of these EBLNs to be transcribed in the host cell was gained through capturing a partial duplicate of TMEM106B. In conclusion, our results strongly support for exaptation of EBLNs to encode host proteins in afrotherians.
Endogenous retroviruses are representative of endogenous viral elements (EVEs), which are known to have occasionally served as the source of evolutionary innovations of the host. Endogenous bornavirus-like nucleoprotein element (EBLN) was the first EVE identified in mammalian genomes to have been derived from a non-retroviral RNA virus. Here we show evidence that EBLNs that were integrated into afrotherian genomes more than 83.3 million years ago have gained novel protein functions associated with rough endoplasmic reticulum in afrotherians. In the amino acid sequence of EBLN proteins, negative selection appeared to have operated more strongly on hydrophilic regions than on hydrophobic regions, suggesting that EBLN proteins may interact with other molecules in their host cells. In addition, we clarified the mechanism how EBLNs have acquired an ability to be transcribed in the host cell; they captured a partial duplicate of an intrinsic gene, transmembrane protein 106B, which retained an intrinsic promoter activity. Our findings suggest that not only retroviral EVEs but also non-retroviral EVEs may have contributed to the host evolution.