Deciphering the evolutionary origin of the RB1 imprint

Previously we could show that the RB1 gene is imprinted. Skewed expression in favour of the maternal allele is due to a differentially methylated CpG-island within intron 2 of the RB1 gene. This CpG-island (CpG85) serves as a promoter for an alternative RB1 transcript and is part of a truncated processed pseudogene (PPP1R26P1), which is derived from the PPP1R26 gene (previously KIAA0649) located on chromosome 9.

We could now narrow down the time interval of this retrotranspositional event by in silico analyses, which revealed that the ancestral gene PPP1R26 is present in all primates, whereas the pseudogene copy within the RB1 gene is only present in higher primates, which comprise Catarrhini (Old World Monkeys, Gibbons, Great Apes and Human) and Platyrrhini (New World Monkeys). Thus, the retrotransposition of PPP1R26 has occurred after the divergence of Strepsirrhini and higher primates, but before the split between Catarrhini and Platyrrhini. Although information for Tarsiidae as distant sister lineage to higher primates is lacking, the retrotransposition of PPP1R26 into the RB1 gene appears to coincide with the retrotranspositional explosion described by Ohshima et al., 2003. Moreover, the in silico analysis revealed that there are additional pseudogene copies on chromosome 22 in human and chimp, which must have been derived from independent retrotransposition events. Only the chimp and the marmoset have another copy on chromosome 8 and chromosome 4, respectively.

For further examination of the evolutionary origin of the RB1 imprint we compared the methylation patterns of the ancestral gene PPP1R26 and its pseudogenes in different primates (human, chimp, rhesus, orangutan and marmoset). Methylation analysis by next generation bisulfite sequencing on the ROCHE/454 GS Junior showed that the pseudogene copy within the RB1 gene is differentially methylated in all primates studied. All other copies are fully methylated except the additional copy on chromosome 4 in the marmoset, which seemed to be differentially methylated. By using an informative SNP for the methylation analysis in 8 individuals from 4 different families we could show that the methylation pattern of the copy on chromosome 4 in the marmoset is not parent-of-origin-specific, but allele-specific. We conclude that the epigenetic fate of a PPP1R26 pseudogene after integration depends on the DNA sequence and selective forces at the integration site.