Endogenous retrovirus-mediated genomic variations in chimpanzees

Abstract

Transposable elements (TEs) have played a significant role in the evolution of host genome by triggering genomic rearrangements. TEs have been studied in various research fields, ranging from population genomics to personalized medicines. Human-specific TEs and TEs existing in the human genome have been well studied. Unlike them, non-human primate-specific TEs remain shrouded in mystery. However, the study of TE-mediated genomic or genetic variations through comparative genomics is essential to understand mechanisms which TEs utilize to modify species-specific genome architecture and to cause species-specific diseases, Therefore, we have studied chimpanzee-specific TEs as well as human-specific TEs. At first, we identified human-specific HERV-K integrated into the human genome after the divergence of human and chimpanzee. Then, for a comparative study of HERV-Ks and non-human ERVs, we extracted chimpanzee-specific endogenous retroviruses (PtERVs) from the chimpanzee genome. We identified 256 chimpanzee-specific PtERVs and characterized them, focusing on their estimated evolutionary age, polymorphism level in chimpanzee populations, and potential impact on the difference between the human and chimpanzee genomes.

Identification of the Chimpanzee-Specific PtERVs

After the completion of human whole-genome sequencing, the whole-genome sequences of various primates including chimpanzee, gorilla, orangutan, and rhesus monkey were also been sequenced and released, which leads to comparative genomics.^1-5 The first version of human whole-genome sequence had a lot of unsequenced regions and assembly errors. Enormous efforts have been put to improve the sequence data, and most of the errors are corrected in the latest version of human reference genome sequence. Human and the non-human primate genome sequences are available with UCSC genome browser (http://genome.ucsc.edu), and species-specific endogenous retrovirus can be retrieved through the UCSC utility. However, the quality of the latest version of the chimpanzee and other primate genome sequences is not good as much as that of the human genome sequence, and the reference genome sequence data still contain unsequenced genomic regions and assembly errors. The relatively poor quality of non-human primate genome sequences is somewhat troublesome because it leads to a high proportion of false positives of TEs. Thus, after computational mining of TEs from the sequence database, the extracted elements have to be experimentally validated, which limits the study of TEs in non-human primates. Despite the difficulty, we tried to study ERVs which exist in the chimpanzee genome. Based on chimpanzee reference genome sequence (PanTro3; October 2010 released), we computationally mined a total 693 of PtERV candidates using UCSC table browser and annotated these elements using RepeatMasker. And then, these elements including their 2kb of flanking sequences were aligned with orthologous sequences of human (hg19; February 2009 released), gorilla (gorGor3; May 2011 released), and orangutan (ponAbe2; July 2007 released) reference genomes. Through these approaches, we identified 256 chimpanzee-specific PtERVs which were not present in the human, gorilla and orangutan genomes. Chimpanzee is the closest species to human and ∼95% of its whole-genome sequence is shared with the human genome sequence.⁶ TEs are partially responsible for the 5% genomic difference between humans and chimpanzees. Thus, we expected that the study of chimpanzee ERVs would shed a light in not only chimpanzee-specific ERV evolution but also the genomic difference caused by species-specific ERVs between humans and chimpanzees. We previously published an article about human-specific HERV-K.⁷ Twenty nine human-specific HERV-K insertions were identified in the study (Fig. 1). Later, we analyzed chimpanzee-specific PtERVs and their roles in triggering genomic variations.⁸

Figure 1.

The proportion of human-specific HERV-Ks and chimpanzee-specific PtERVs in the human and chimpanzee genomes, respectively. The left diagram regarding HERV-Ks in the human genome was drawn based on the Buzdin et al. (2003) and Shin et al. (2013). The right diagram regarding PtERVs in the chimpanzee genome was drawn based on the Mun et al. (2014). The number of black lollipops means the number of retrotransposition-competent ERVs in each host genome.

Characterization of Chimpanzee-Specific PtERVs

Chimpanzee-specific endogenous retrovirus (PtERV) is fossilized in the chimpanzee genome through infection into germ line and retrotransposition after divergence of the human and chimpanzee genomes.⁹ Similar to the human endogenous retrovirus (HERV), PtERV has a classical structure, composing of 3 ORFs (gag, pol, and env) encoding essential viral machineries and long terminal repeat (LTR) existing in both ends of the internal ORF region.¹⁰ We computationally extracted 256 chimpanzee-specific PtERV candidates.⁸ They exist in 3 different types in the chimpanzee genome: full-length PtERV, solitary LTR, and truncated PtERV. Among them, 14 full-length PtERVs and 23 solitary LTRs were new in that they had not been reported in any of previous studies (Fig. 1).^9,10 The chimpanzee-specific PtERVs located on chimpanzee chromosomes 7, 13, and Y; Y chromosome had 15 full-length PtERVs.⁸ It was suggested that the host genome has adapted various mechanisms to eliminate TEs or weaken their effects during evolution.^11,12 One of the mechanisms is a homologous recombination. Homologous recombination occurs between PtERVs due to a high level of sequence identity between the elements, leading to the elimination of the element. Considering that recombination rate is relatively low on the chromosome Y, compared with other chromosomes, the accumulation of full-length PtERVs on the chromosome Y is well explained.^13,14 Given that there was no preferential insertion site for PtERV insertions, it is likely that PtERVs randomly insert into the chimpanzee genome, and then are fixed or eliminated from the host genome via the host defense mechanism.¹⁵

Unlike human-specific HERV-Ks whose majority is solitary LTRs, 121 out of 256 chimpanzee-specific PtERVs are full-length (Fig. 1). It is a remarkable difference between human-specific HERV-K and chimpanzee-specific ERV.^7,8 Compared with human LTRs, there was a high level of sequence divergence between PtERV LTRs. The divergent sequence reduces the homologous recombination between PtERV LTRs. One of the mechanisms to generate solitary LTR is the homologous recombination between LTRs.¹⁶ Taken together, it can be suggested that the relatively low rate of recombination between PtERV LTRs is responsible for the higher copy number of full-length ERVs in the chimpanzee than human genome. In fact, human-specific HERV-Ks have propagated in a single lineage while chimpanzee-specific ERVs have propagated in 2 different lineages, CERV1 and CERV2 which are distinguished from one another based on their LTR sequence. However, we could not rule out other possibilities to explain the high copy number of chimpanzee-specific ERVs.

Recent Integrations of PtERVs and Their Polymorphism in the Chimpanzee Genome

PtERV retaining act ORFs and LTRs could be retrotranspositionally competent because it is still capable to produce its viral machineries which lead to retrotransposition through “copy and paste” mechanism or reinfection by making their particle.¹⁷ We closely examined chimpanzee-specific full-length PtERVs computationally and experimentally, and found 8 retrotransposition-competent PtERVs (rcPtERVs) (Fig. 1). The eight elements contained intact ORFs which are not disrupted by frameshift or nonsense mutation. Based on their LTR sequences, 3 and 5 rcPtERVs belonged to CERV1 and CERV2 families, respectively. We estimated the age of the 2 families and the result showed that CERV2 family (2.43-3.16 myrs) is younger than CERV1 family (8.80-11.45 myrs). PtERVs recently integrated into a new genomic region are likely to be polymorphic in the chimpanzee genome: presence/absence polymorphism.¹⁸ The polymorphism level of chimpanzee-specific PtERVs was examined using a DNA panel composed of 12 chimpanzee individuals. The result showed that 11 chimpanzee-specific PtERVs are polymorphic in the chimpanzee genome and the polymorphic rate of CERV2 family (22.2%) is higher than that of CERV1 family (20.9%). After full-length PtERV insertion, PtERV is likely to accumulate mutations over time or be removed by host defense mechanisms. Thus, it may as well that younger PtERV family has more rcPtERVs and higher polymorphic rate in the chimpanzee genome.

PtERVs Causing Chimpanzee-Specific Genomic Deletions

The chimpanzee genome has undergone species-specific genomic and genetic changes, since the divergence from human, its closest species. Retrotransposons are one of major sources to cause the changes in various primate genomes including the chimpanzee genome. They have increased their copy numbers in the chimpanzee genome via various mechanisms such as retrotransposition, reinfection, and recombination-mediated duplication events.^3,6-8,19 Retrotransposons are able to cause host genomic variations using various mechanisms which include insertion, gene conversion, insertion-mediated deletion, recombination and transduction.²⁰ Most of PtERVs have propagated via a classical insertion whose hallmark is target side duplication (TSD). However, some PtERVs resulted from non-classical insertions, called NCPI. The elements produced by NCPI event are lack of TSD and instead exhibit a short junctional microhomology that is shared with the host genomic region neighboring the element. NCPI accompanies host genomic deletions.⁸ We found that 5 chimpanzee-specific PtERVs were integrated into the chimpanzee genome via NCPI and deleted 51,366 bp of the chimpanzee genome.

Transcript Variants Caused by PtERVs

ERVs are able to insert into the genic region of the host genome. They can destroy host genes through the insertion into the exonic region. In addition, ERVs residing in the intronic region are also able to modify the expression of host genes by causing alternative splicing via their cis-elements, which could lead to abnormal proteins with deleterious function and genetic diseases, ultimately.^21,22 It was reported that some mobile genetic elements could provide cis-element altering the cellular environments such as chromatin structure and epigenetic modification.^22-25 Thus, ERVs are vulnerable to host defense mechanism which could delete the elements from the host genome. Due to the host defense mechanism, only small portion of the elements go to be fixed in the host genome. The host defense mechanism is probably more efficient against ERVs inserting into the genic region. In fact, most ERVs are located in intergenic and heterochromatin regions. Nonetheless, some elements are fixed in the genic region although they locate in UTR (untranslated region) or intronic region.^26-28 Some of chimpanzee-specific PtERVs exist in the intronic region of known coding genes. PtERV residing inside PNRC2 gene causes chimpanzee-specific transcripts whose transcription starts from different transcription start site (TSS). PNRC2 transcripts from non-chimpanzee primates including human have a long 5’ UTR while its chimpanzee counterpart has a relatively short UTR. It suggests that a potential alternative PNRC2 promoter was created by PtERV insertion in the chimpanzee genome.⁸ Thus, PtERVs locating either in or nearby the genic region have a potential to influence the expression of the respective genes, which could lead to phenotypical differences between chimpanzee and other primates.^22,30-33

Concluding Remarks

In this commentary, we characterize chimpanzee-specific PtERVs and describe their potential role in causing genomic changes in the chimpanzee genome. Since the divergence of human and chimpanzee, PtERVs have contributed to chimpanzee genomic fluidity and chimpanzee-specific genomic variations via insertion-mediated deletion and the homologous recombination between them. In addition, PtERV is able to lead to alternative gene transcripts by its insertion into the genic region. Further studies with chimpanzee RNAs or cell-lines will elucidate more about the impact of the chimpanzee-specific PtERVs on phenotypical divergence in chimpanzees.

Disclosure of Potential Conflicts of Interest

No potential conflicts of interest were disclosed.