Retrotransposons evolution and impact on lncRNA and protein coding genes in pigs

Abstract

Background: Retrotransposons are the major determinants of genome sizes and they have shaped both genes and genomes in mammalian organisms, but their overall activity, diversity, and evolution dynamics, particularly their impact on protein coding and lncRNA genes in pigs remain largely unknown.

Results: In the present study, we performed de novo detection of retrotransposons in pigs by using multiple pipelines, four distinct families of pig-specific L1 s classified into 51 distinct subfamilies and representing four evolution models and three expansion waves of pig-specific SINEs represented by three distinct families were identified. ERVs were classified into 18 families and found two most "modern" subfamilies in the pig genome. The transposition activity of pig L1 was verified by experiment, the sense and antisense promoter activities of young L1 5'UTRs and ERV LTRs and expression profiles of young retrotransposons in multiple tissues and cell lines were also validated. Furthermore, retrotransposons had an extensive impact on lncRNA and protein coding genes at both the genomic and transcriptomic levels. Most protein coding and lncRNA (> 80%) genes contained retrotransposon insertions, and about half of protein coding genes (44.30%) and one-fourth (24.13%) of lncRNA genes contained the youngest retrotransposon insertions. Nearly half of protein coding genes (43.78%) could generate chimeric transcripts with retrotransposons. Significant distribution bias of retrotransposon composition, location, and orientation in lncRNA and protein coding genes, and their transcripts, were observed.

Conclusions: In the current study, we characterized the classification and evolution profile of retrotransposons in pigs, experimentally proved the transposition activity of the young pig L1 subfamily, characterized the sense and antisense expression profiles and promoter activities of young retrotransposons, and investigated their impact on lncRNA and protein coding genes by defining the mobilome landscapes at the genomic and transcriptomic levels. These findings help provide a better understanding of retrotransposon evolution in mammal and their impact on the genome and transcriptome.

Keywords: Distribution bias; Gene overlapping; Pig genome; Promoter activity; Retrotransposition activity; Retrotransposon evolution.

Fig. 1

Neighbor-joining polygenic tree of pig L1 based on the 5′UTR and classified L1 s into four distinct families (L1A, L1B, L1C, and L1D)

Fig. 2

Evolution of L1 s and SINEs in the pig genome. a Structural schematics of the putatively active L1 s and pig-specific SINE families (SINEA, SINEB, and SINEC). b Age distribution of pig-specific L1 families. c and d Age distribution across the subfamilies (L1D1–21) of the youngest L1 family (L1D). e Insertion polymorphism (IP) detection of the youngest L1 (L1D1) and SINE (SINEA1) subfamilies by PCR. Breed name abbreviations: Meishan (MS), Shawutou (SWT), and Jiangquhai (JQH) pigs are native Chinese pig breeds from Jiangsu Province; the Sujiang (SJ) pig is a newly established breed based on the Duroc and Jiangquhai bloodlines; Bama (BM) pigs are miniature pigs from Guangxi Province; the wild boar (WB) was from Anhui Province; and the Landrace (LD) and Yorkshire (YK) pigs were from a breeding farm in Anhui Province. Ne, negative control without DNA. Two transposon loci in each of the youngest transposon subfamilies were selected for insertion polymorphism (IP) detection and labeled as IP1 and IP2. If an individual contains SINE insertion at SINE-IP1 or SINE-IP2 site, the band size would be 629 or 676 bp, respectively, and if no SINE insertion, the band would be 335 or 382 bp. The three bands showed in the M (marker) lane are 750 bp, 500 bp and 250 bp from top to bottom. f) Age distribution of pig-specific SINE families. g and h Age distribution across the subfamilies (SINEA1–11) of the youngest SINE family (SINEA). The x-axis represents the insertion age (Million years ago, Mya), and the y-axis represents the percentage of the genome composed of retrotransposon families/subfamilies (%) in Fig. b, c, f, and g

Fig. 3

Retrotransposition activity analysis of pig L1. a Schematics of vectors used for retrotransposition assays. hL1 and mhL1 were used as positive and negative control, respectively. The pL1 vector contains 5′UTR, ORF1, IGR, ORF2, and 3′UTR of L1 cloned from the pig genome (L1D1 coordinate). The pL1-CMV is the same as pL1, but the 5′UTR of pig L1 was replaced with the CMV promoter. The phL1 is a chimeric vector derived by the CMV promoter, the two ORFs and 3′UTR were from pig, and the IGR was from human L1 (99-PUR-RPS-pBlaster1). All the vectors contain two selective cassettes (mBlast and Puro) for two-round selections. The mBlast cassette contains an inverted blasticidin resistance gene (black box) disrupted by a self-splicing intron [–51]. The introns will only splice out from a transcript generated by the L1 or CMV promoter. The spliced RNA is reverse-transcribed, followed by integration of the cDNA into the genome. The new insert contains a functional Blast gene. Blasticidin resistance will be obtained only if retrotransposition occurs. b and c Number of clones formed after puromycin and blasticidin selection. Blast^R foci were fixed to flasks and stained with Giemsa for visualization. Bars represent the mean blasticidin resistant colonies ± standard deviation, shown as error bars for each construct

Fig. 4

Evolution of ERVs in the pig genome. a ERVs were classified into 18 ERV families (ERV1–18) based on the phylogenetic tree inferred by using the Neighbor-joining method with the MEGA7 program, and the reference RT sequences from species other than pigs are included for comparison, shown with dots and described in the methods. b Structural schematics of the ERV6A and ERV6B, which featured LTR-gag-pol-env-LTR and were presumed to be active. Gag_MA: Matrix protein (MA), p15; Gag_p30: Gag P30 core shell protein; RVT_1: Reverse transcriptase (RNA-dependent DNA polymerase); RNase H-like: RNase H-like domain found in reverse transcriptase; rve: Integrase core domain; MLV-IN_C: Murine leukemia virus (MLV) integrase (IN) C-terminal domain; TLV_coat: ENV polyprotein (coat polyprotein) (c) Age distribution of pig ERV classes. d Age distribution of the youngest pig ERV subfamilies (ERV6A and ERV6B). e Insertion polymorphism detection of the youngest pig ERV subfamilies (ERV6B) by PCR. Breed name abbreviations are the same as those in Fig. 1f. The x-axis represents the insertion age (Mya), and the y-axis represents the percentage of the genome composed of retrotransposon families/subfamilies (%) in Fig. c, d

Fig. 5

Sense and antisense promoter activities of pig L1 5′UTRs and ERV6 LTRs. a Schematics of vectors used for promoter activity detection by luciferase assay. The sense and antisense 5′UTR/L1 and LTRs of ERVs from young and putatively active subfamilies of L1 were cloned into the pGL3-enhancer luciferase reporter vector to investigate the promoter activity. b Sense and antisense promoter activities of ERV6A and ERV6B LTRs measured by luciferase assay. c Sense and antisense promoter activities of young L1 5′UTRs (L1D) measured by luciferase assay. Eight sense and four antisense L1 5′UTRs from different subfamilies of L1D family were cloned as described in the methods, and two 5′UTRs (hL1–3 and hL1-M) of active L1 s from human and one 5′UTR (mL1) of active L1 from mouse were used as positive controls

Fig. 6

Sense and antisense expression profiles of pig L1D of L1 s, SINEA of SINEs, and ERV6B of ERVs. a Primer design for reverse transcription (RT) and real-time quantitative PCR (RT-qPCR) detection. The primer for sense and antisense RT are indicated by red and green arrowheads, respectively, and the primers of ORF1-F/R, ORF2-F/R, 5′UTR-F/R, pol-F/R, gag-F/R, env-F/R, LTR-F/R, SINE-F/R (black arrowheads), are used for RT-qPCR to detect the expression of 5′UTR, ORF1, and ORF2 of L1, LTR, gag, pol, and env of ERV6 and SINE, respectively. b Sense expression of ORF1 and ORF2, and antisense expression of 5′UTR of L1D in tissues and cells. c Sense and antisense expression of SINEA in tissues and cells. d Sense expression of gag, pol, and env of ERV6, and antisense expression of LTR of ERV6 in tissues and cells

Fig. 7

Retrotransposons contribution to protein coding and lncRNA genes. a The proportion of protein coding (pc) genes and lncRNA genes overlapping with retrotransposon insertions. b The proportion of TE insertions in the introns and exons of protein coding and lncRNA genes, and their flank regions. c The genomic coverage of retrotransposons in protein coding (pc) and lncRNA genic regions, and their flank regions. d The proportion of mRNAs, ESTs, and lncRNAs containing retrotransposon-derived sequences. e Sequence coverage of retrotransposons in lncRNAs and mRNAs. f The proportion of the protein coding genes generating chimeric transcripts with retrotransposons