Centromere remodeling in Hoolock leuconedys (Hylobatidae) by a new transposable element unique to the gibbons

Abstract

Gibbons (Hylobatidae) shared a common ancestor with the other hominoids only 15-18 million years ago. Nevertheless, gibbons show very distinctive features that include heavily rearranged chromosomes. Previous observations indicate that this phenomenon may be linked to the attenuated epigenetic repression of transposable elements (TEs) in gibbon species. Here we describe the massive expansion of a repeat in almost all the centromeres of the eastern hoolock gibbon (Hoolock leuconedys). We discovered that this repeat is a new composite TE originating from the combination of portions of three other elements (L1ME5, AluSz6, and SVA_A) and thus named it LAVA. We determined that this repeat is found in all the gibbons but does not occur in other hominoids. Detailed investigation of 46 different LAVA elements revealed that the majority of them have target site duplications (TSDs) and a poly-A tail, suggesting that they have been retrotransposing in the gibbon genome. Although we did not find a direct correlation between the emergence of LAVA elements and human-gibbon synteny breakpoints, this new composite transposable element is another mark of the great plasticity of the gibbon genome. Moreover, the centromeric expansion of LAVA insertions in the hoolock closely resembles the massive centromeric expansion of the KERV-1 retroelement reported for wallaby (marsupial) interspecific hybrids. The similarity between the two phenomena is consistent with the hypothesis that evolution of the gibbons is characterized by defects in epigenetic repression of TEs, perhaps triggered by interspecific hybridization.

Fig. 1.—

FISH with NEL BACs generates bright centromeric signals in hoolock but not in the other gibbon genera. (A) FISH with NLE BAC clones produces bright signals on almost all the centromeres of HLE chromosomes. Five chromosome pairs (indicated by the arrows) show reduced or significantly less intense FISH signals. (B) The same BACs produced the main hybridization signal only on one pair of chromosomes in gibbon species from the other three genera. One of the BACs (CH271-457L13) is represented here (see also supplementary fig. S1, Supplementary Material online). Weaker repeated signals were observed in Nomascus and Hylobates.

Fig. 2.—

FISH with PCR-derived SVA probe generates centromeric signals. PCR primers were designed to amplify a portion of the SVA element. The PCR product was fluorescently labeled and hybridized by FISH on hoolock chromosomes. The hybridization pattern recapitulates the one obtained with the BACs.

Fig. 3.—

Structure of the LAVA element. (A) Illumina reads obtained from hoolock BACs and known for containing SVA sequences were mapped against the SVA consensus sequence. Most of the reads map in correspondence of the VNTR region of the SVA element. (B) Structure of the novel composite TE based on the sequence data obtained from NLE BACs. The arrows inside the different components indicate their orientation. (C) The PCR product spanning the junction between the AluSz6 and L1ME5 regions was labeled and used as probe for FISH on hoolock metaphases. The bright centromeric signals confirm that the novel TE is expanded in the hoolock centromeres.

Fig. 4.—

Co-hybridization experiments to investigate the relationship between LAVA elements and centromeric satellites. (A) Hoolock chromosomes were co-hybridized with α-satellite and LAVA probe. All centromeres, except that of chromosome 4, show depletion of α-satellite. (B) The SVA probe co-hybridizes with a probe designed on the SATR-1 satellite. (C) The relationship between LAVA elements and SATR-1 sequences is shown at higher resolution by the fiber-FISH experiment.