An 192 bp ERV Insertion in the First Intron of TLR6 Acts as an Enhancer to Increase TLR6 and TLR1 Expression

Toll-like receptors (TLRs) play important roles in building innate immune and inducing adaptive immune responses. Associations of the TLR gene polymorphisms with diseases susceptibility, which are the basis of molecular breeding for disease resistant animals, have been reported extensively. Retrotransposon insertion polymorphisms (RIPs) were developed recently as a new type of molecular marker having great potential in population genetics and quantitative trait locus (QTL) mapping analysis. In this study, bioinformatics prediction combined with the PCR-based amplication was employed to screen for RIPs in porcine TLR genes. Their population distribution and impact on gene activity and phenotype of one RIP was been further evaluated.


Abstract Background
Toll-like receptors (TLRs) play important roles in building innate immune and inducing adaptive immune responses. Associations of the TLR gene polymorphisms with diseases susceptibility, which are the basis of molecular breeding for disease resistant animals, have been reported extensively. Retrotransposon insertion polymorphisms (RIPs) were developed recently as a new type of molecular marker having great potential in population genetics and quantitative trait locus (QTL) mapping analysis. In this study, bioinformatics prediction combined with the PCR-based ampli cation was employed to screen for RIPs in porcine TLR genes. Their population distribution and impact on gene activity and phenotype of one RIP was been further evaluated.

Results
The results showed that ve RIPs, located at the 3' ank of TLR3, 5' ank of TLR5, intron 1 of TLR6, intron 1 of TLR7, and 3' ank of TLR8 respectively, was identi ed. These RIPs were detected in different breeds with an uneven distribution among them. By using the dual luciferase activity assay a 192 bp endogenous retrovirus (ERV) in the intron 1 of TLR6 was proven to act as an enhancer increasing the activities of TLR6 promoter and multiple mini-promoters. Furthermore, the real-time quantitative polymerase chain reaction (qPCR) analysis demonstrated that the ERV insertion signi cantly enhances the mRNA expressions of TLR6, the neighboring gene TLR1, and the downstream genes MyD88 (Myeloid differentiation factor 88), Rac1 (Rac family small GTPase 1), TIRAP (TIR domain containing adaptor protein), Tollip (Toll interacting protein) of TLR signaling pathway and the in ammatory factors IL6 (Interleukin 6), IL8 (Interleukin 8), and TNFα (Tumor necrosis factor alpha) in 30-day piglet tissues. In addition, the serum IL-6 and TNFα was also signi cantly upregulated by ERV insertion.

Conclusions
Overall, ve RIPs were been identi ed in several TLRs, and the 192 bp ERV insertion in the rst intron of TLR6 can improve the expressions of TLR6, TLR1, their downstream genes, and the in ammatory factors by acting as an enhancer affecting the regulation of TLR pathways, which may be applicable in the molecular breeding of disease resistant animals.

Background
Toll-like receptors (TLRs) play profound roles in motivating innate and adaptive immune responses. Protein(s) encoded by TLRs genes are important due to their ability of recognizing different types of pathogens and associated molecular patterns [1]. Polymorphisms in TLR loci and their in uence on either susceptibility or resistance to major human infectious diseases, including tuberculosis, leishmaniasis, malaria, lariasis, some autoimmune endocrine diseases, have been reported extensively [2][3][4]. It has been reported that C1174T of TLR5 resulted in a common stop codon polymorphism and produced signi cantly lower levels of proin ammatory cytokines in comparison with individuals with the wild-type genotype which indicated that the TLR5 stop codon polymorphism is associated with protection from the development of systemic lupus erythematosus [5]. The missense mutation of TLR1 rs5743618, speci c in Europeans can change the expression of 81 genes involved in the in ammatory response [6]. Chikungunya patients with rs179010-CC genotype of TLR7 showed signi cantly high interferon alpha 1 (IFN-α) level, which might act as potential prognostic biomarkers for predicting Chikungunya susceptibility [7]. It is commonly accepted that TLRs are important candidate genes for some human diseases.
For domesticated farm animals, TLRs also have been suggested as the most promising candidate genes for immunity improvement or disease resistance by molecular breeding [8]. Genetic variants of TLRs associated with cattle mastitis, mycobacterial infection, and paratuberculosis have been identi ed [9][10][11]. Totally, ten TLRs in the pig genome were annotated and SNP screening in the pig TLRs and their expression patterns in immunity organs have been reported extensively [12]. It has been suggested that piglets with the T allele of TLR5 (C1205T) exon exhibit impaired resistance to Salmonella typhimurium infection [13]. C506W of TLR4 exon in Japanese wild boar populations caused loss of ability to induce nuclear factor-κB activation after lipid A stimulation [14][15][16]. Association between SNPs in TLR4 and TLR5 with transcription levels of cytokines indicate that these SNPs are related to the modulation of the cytokine mediated immune response. Until now, all studies were focusing on the associations of SNPs in TLRs with diseases susceptibility, reports for retrotransposon insertion polymorphism in TLRs and their genetic effects are not available.
Retrotransposons are important components of plants and mammal genomes, which account for nearly half of the mammal genomes [17,18], and can mobilize themselves to new genomic locations and generate polymorphic insertions. Retrotransposons can further be classi ed into LTR (Long Terminal Repeat elements, including endogenous retrovirus, ERV) and non-LTR (including LINE, Long Interspersed Nuclear Elements; and SINE, Short Interspersed Nuclear Elements) [19]. For a long time, transposable elements including retrotransposon has been considered as genomic parasites [20,21]. However, recently, more and more evidences support that retrotransposons contribute to genome architecture and evolution even maintenance of three-dimensional chromatin organization in mammals [18,[22][23][24].
Retrotransposon Insertion Polymorphism, RIPs, have been applied for genomic evolution and genetic diversity evaluations in plants as molecular markers [25,26]. For human, RIPs have been identi ed as causative mutations for some diseases [27]. By genome-wide association studies, an intronic Alu insertion polymorphism inserted in the 97 bp downstream of a CD58 exon is on a haplotype associated with multiple sclerosis risk due to the Alu insertion altering mRNA splicing [28,29].
For domestic animals, RIPs also have been used for evolution and population genetic analysis of sheep [30], deer [31], cat [32], chicken [33], and rabbit [34]. Several RIPs have been de ned as the causative mutations for phenotype variations. An ERV insertion in the 5' anking region of SLCO1B3 causes blue eggshell by promoting the expression of SLCO1B3 gene in the uterus (shell gland) of the oviduct in chicken [35,36] . The henny feathering allele harbors an insertion of an intact avian ERV at the 5'end of CYP19A1 [37]. The SINE insertion in follicle stimulating hormone beta (FSHβ) and protein disul de isomerase associated 4 (PDIA4) genes were associated with litter size variation in pigs [38,39].
In the current study, the contribution of RIPs to the structural variations of TLR genes, the breed distribution of these RIPs and exemplarily the genetic effect of one RIP was investigated. We identi ed ve RIPs for 10 TLR genes, and our data suggest that one RIP plays role in the regulation of TLR pathway by acting as an enhancer. These ndings will contribute to the understanding the role of RIPs in the pig genomic and genetic variation, and one RIP may be useful for disease resistance selection in the pig breeding.

Results
Five RIPs generated by retrotransposon insertions in the pig TLR gene cluster Ten TLR genes and their anking sequences from sixteen assembled pig genomes, representing lean type pigs (Cross-breed of Yorkshire/Landrace/Duroc, Duroc, Landrace, Yorkshire, Pietrain, Berkshire, and Hampshire), fat type pigs (Rongchang, Meishan, Bamei, and Jinhua), and miniature pigs (Bama, Wuzhishan, Tibetan, Goettingen, and Ellegaard Gottingen) were used for screening of structural variations by alignment with ClustalX program (version 2.0) [40]. Totally, we identi ed 53 large SVs (more than 50 bp and less than 1000bp) or large frameshift variants (more than 1000bp), and 35 of them were predicted as retrotransposon insertion polymorphisms, including 15 SINE, 11 ERV and 6 LINE insertions, which were summarized in additional le 1 (Table S1). Then all these predicted insertions were further been evaluated by wet PCR via the speci c primer pairs spanning the insertions in pooled samples. Finally, 5 RIPs, including 2 SINE insertions, 2 ERV insertions and 1 Line insertion, were obtained by using 11 domesticated pig breeds and wild boar samples (Fig. 1A). All these insertions were been con rmed again by TA cloning and sequencing.
One 288 bp and one 294 bp SINE insertions in the 3' anks of TLR3 and TLR8 were detected, respectively. Moreover a single 357 bp L1 insertion in the 5' ank of TLR5 was found, and in addition two ERV insertions were detected, one 192 bp and one 413 bp in rst intron of TLR6 and TLR7 were identi ed, respectively. We named those insertions as TLR3-SINE-RIP, TLR5-LINE-RIP, TLR6-ERV-RIP, TLR7-ERV-RIP, and TLR8-SINE-RIP, respectively ( Fig.1B and 1C).
Evidence of enhancer activity of the 192 bp ERV insertion near the core promoter of TLR6 Both of TLR6 and TLR7 genes contain an ERV insertion in the rst intron, and further analysis revealed that the 192 bp ERV insertion was near the core promoter region of TLR6, and diverse regulatory elements such as CpG islands and transcription factor binding sites ( Fig. 2A). To further evaluating the potential roles of ERV insertions in the regulation of TLR6 activity, the core promoter regions with or without the 192 ERV insertion allele was been cloned into a luciferase reporter vector (pGL3-Basic), respectively (Fig.2B), and then submitted for luciferase activity evaluation. The dual luciferase activity assay revealed that the ERV insertion signi cantly improved the promoter activity of TLR6, the luciferase activity in cells transfected with the vector with EVR insertion allele (TLR6 ERV+ -Luc(En)) were almost two times (P<0.01) than that without ERV insertion allele (TLR6 ERV--Luc(En)) in both porcine PK15 and human Hela cells (Fig. 2C), which suggested that the 192 bp ERV may act as an enhancer in the regulation of TLR6 activity. To further con rm this deduction, we evaluated the enhancer activity of the ERV insertion by cloning it into the luciferase reporter vector containing mini-promoter, but absent SV40 enhancer, which is generally used for enhancer activity evaluation. Two type mini-promoters (β-globin, and OCT4) were been evaluated, and the schematic diagram of vectors were shown in Fig.3A. Again, the luciferase activity assay revealed that the ERV insertion allele signi cantly improve all these mini-promoter activities in both PK15 (Fig.3B) and Hela cell lines (Fig.3C). These data strongly support that the 192 bp ERV insertion acts as an enhancer and may be involved in the regulation of TLRs.
ERV insertion alter the expression of pig TLR6 and TLR1 and their downstream genes in multiple tissues To further illustrate the biological roles of the 192 bp ERV insertion in the Toll like receptor signaling pathway, we investigated the mRNA expressions of TLR6 and TLR1, which are neighboring genes located on chromosome 8, and their downstream genes (MyD88, Rac1, TOLLip, TIRAP, IL-6, IL-8, and TNFα) in the same pathway by qPCR in multiple tissues (liver, spleen, lung and kidney) for different genotypes in 30-day piglets.
The qPCR results revealed that, generally, the ERV insertion improve the expression of TLR6 and TLR1 in these tissues. In detail, in the spleen, kidney and liver tissues, the expression of TLR6 and TLR1 in the pigs with ERV +/+ genotype were signi cantly higher (P<0.05) than that in those animals with ERV +/and ERV -/genotypes. In lung, there were signi cant expression differences (P<0.01) of TLR6 between the ERV +/+ animals and the ERV -/animals ( Fig.4A and 4B). MyD88, Rac1, TOLLip and TIRAP are the downstream genes of TLR6 and TLR1 in the Toll like receptor signaling pathway and play key roles in innate immune mechanisms as central molecules. The qPCR results revealed that ERV insertion enhanced the expressions of most downstream genes of TLRs. In detail, in the spleen, lung and kidney, the expression levels of MyD88 and Rac1 were signi cantly higher (P<0.05) in the animals with ERV +/+ than in the animals with ERV +/and ERV -/genotypes ( Fig.4C and 4D). Expression of TIRAP and Rac1 in liver, lung and kidney of ERV +/+ homozygous animals were signi cantly (P<0.05) higher than those of heterozygote (ERV +/-) and homozygote (ERV -/-) animals. In spleen, signi cantly (P<0.05) difference was only observed for the expression of TIRAP gene between homozygotes ERV +/+ and ERV -/animals ( Fig.4E and 4F). In ammatory factors IL-6, IL-8, TNFα are important genes in the end of Toll like receptor signaling pathway. The expression of TNFα, IL-6, IL-8 increased signi cantly (P<0.05) in the spleen, lung and kidney of homozygous of ERV +/+ compared with that in ERV -/genotype piglets (Fig.4G, 4H, 4I). These results indicated that the 192 bp ERV insertion allele near the core promoter of TLR6 increases not only the expressions of TLR6 and TLR1, but also the expressions of the downstream genes of the TLR signaling pathway.

Impact of ERV insertion on the serum immune cytokine
To investigate the impact of the ERV insertion allele near the core promoter of TRL6 on the immune response, several serum immune cytokines were measured by ELISA for different genotypes of 30-day piglets. The ELISA analysis revealed that, consistent with the higher expression of IL-6 and TNFα in the important immune tissues (spleen and kidney) of ERV +/+ piglets compared that in other genotype piglets (ERV +/and ERV -/-), the serum concentrations of IL-6 and TNFα in the animals with ERV +/+ genotype were also signi cantly higher than that in the ERV -/genotype animals (P<0.05) (Fig.5). But there is no signi cant difference of serum IL8 among different genotype.

Discussion
Retrotransposons are dominant components in most land plant genomes and mammals [18,26,41] and regarded as important drivers of species diversity and putative actors in evolution and adaptation [26,[42][43][44]. Various methods, such as inter-retrotransposon ampli ed polymorphism (IRAP) and retrotransposonmicrosatellite ampli ed polymorphism (REMAP), have been developed to exploit RIPs [45]. They have been successfully applied in the studies of plant genetic diversity and QTL mapping [46]. Furthermore, some RIPs have been proved as a causal mutation to change the plant phenotype [47][48][49]. It is believed that SINE of RIPs are "nearly ideal" genetic markers to facilitate plant breeding [50]. In domestic animals, RIPs also have been applied to genetic diversity, evolution and variety identi cation, and display great potentials in animal molecular breeding. Our previous study revealed that LINE, LTR, and SINE are highly enriched in the pig genome, and totally account for about 37.13% of the genome sequence [51]. According to the insertion age analysis, differential evolution pro les were been observed for different families and subfamilies of retrotransposons. Most retrotransposons in the pig genome are ancient and no longer jumping, and cannot generate polymorphic insertions in populations, whereas some of them were thought to be younger retroelements, such as SINEA, L1D, ERV6 subfamilies [51]. These retrotransposons still play roles in shaping genomes and gene evolution and contribute to the genomic variations and their insertions tend to generate polymorphisms, which can be used as genetic markers. In addition, it has been suggested that transposable elements affect the genome in both destructive and constructive ways [41]. Natural selection and genetic drift could shape the distribution and accumulation of TEs and insertions with destructive effects are rapidly removed from the population [52]. In the current study, we identi ed ve RIPs in TLRs genic and anking regions by alignment and combining the PCR validation. They located in different genic positions including introns, 5' and 3' ank of TLRs. Two ERV insertions, One of 192 bp and the other of 413 bp in length were identi ed in the rst intron of both TLR6 and TLR7, respectively, and based on the bioinformatics analysis, the 192 bp ERV insertion in the intron 1 of TLR6 was predicted to be in the core promoter region, which may in uence the gene expression. Since the ERV insertion fragment contained the U3-R-U5 sequences, which were thought to act as transcriptional regulators because of U3 region [53], it may act as enhancer or promoter that could be involved in the regulation of expression pattern of target genes [54][55][56]. Here, the luciferase assay analysis strongly supported that the 192bp ERV insertion serves as an enhancer, which not only can increase the TLR6 promoter activity, but also can improve the activities of diverse mini-promoters.
Population distribution analysis of these RIPs revealed that most loci are in genetic equilibrium, while some polymorphic insertion loci (TLR5-LINE-RIP) are in genetic disequilibrium, indicating that these loci may have experienced strong selection and play biological roles in gene regulation and phenotype variation. TLR6-ERV-RIP in Sujiang and Fenjing deviated from the Hardy-Weinberg equilibrium (P<0.05), indicating that this locus may have experienced strong puri cation selection considering the ERV insertion playing positive roles in the immune response. The deletion allele of TLR6-ERV-RIP were only found in China native pig breeds (Bama, Fenjing) or cross breeds of Sujiang (Duroc×Jiangquhai×Fengjing), while all analysed western pigs including Landrace, Yorkshire, Duroc, and Sicilian Black from Italy are monomorphic and only contain the ERV insertion allele (ERV + ), suggesting that the deletion allele (ERV -) seems to originate from Asian pig breeds. Base on the positive effect of this allele on the disease resistance, it can be used to improve the disease resistance performance (such as Sujiang) or to be introduced into other China native breeds with marker assisted selection.
An age of 30 days is an important stage for pig to develop the adaptive immunity [57]. We further evaluated the expression of TLR6 and TLR1, which is a nearby gene of TLR6 with only 4.2 kb far away, in different tissues of 30-days piglets between different ERV insertion genotypes. The qPCR analysis demonstrated that the mRNA expressions of TLR6 and TLR1 was been enhanced signi cantly by ERV insertion in multiple tissues of 30-day piglets. TLRs play important roles in the innate immune response by interacting with adapter molecules, such as MYD88, TIRAP, Rac1 and Tollip, which are downstream genes of the TLR pathway [58]. With the help of the qPCR analyses, we also con rmed that all/most detected downstream genes of TLRs were also upregulated by ERV insertion in multiple tissues of 30-day piglets, suggesting that the ERV insertion not only increases the expressions of TLR6 and TLR1, but also triggers the expressions of their downstream genes.
The cytokines released by in ammatory cells are essential factors in resisting pathogen infection. When activated by pathogens, TLRs recruit adapter molecules and subsequently initiate downstream signaling pathway, resulting in the activation of transcription factor nuclear factor kappa B subunit (NF-kB) and the production of downstream in ammatory cytokines [59,60]. Higher TLR1 expression suggested better prognosis in patients with pancreatic ductal adenocarcinoma (PDAC) [10]. The mRNA expressions of major TLR genes including TLR1 and TLR6 of Tibetan pigs were higher in most immune tissues than those of Yorkshires, which may attribute to stronger innate immunity for Tibetan pigs [61]. And higher expression of TLRs were also associated with stronger disease resistance [60]. Yorkshire × Landrace (YL) pigs exhibited more serious clinical symptoms when arti cially infected with porcine circovirus type 2 (PCV2) virus compared with Laiwu, which is a China native pig breed, indicating YL and Laiwu pigs display different susceptibility to PCV2 infection and Laiwu pigs seem to be more resistant for PCV2 virus. The serum levels of IL-6, IL-8, IL-12 and transforming growth factor beta 1 (TGF-β1) in the PCV2 relatively resisting pigs of Laiwu also increased signi cantly more pronounced at the early infection stages than in YL pigs [62]. Here, consistent to the increased expression of genes in the TLR pathway due to ERV insertion, we found that the expressions of the important in ammatory factors including IL-6, IL-8, and TNFα were also been increased by ERV insertion in tissues of 30-day piglets, with signi cant increment of IL-6 and TNFα in the serum.
Overall, these data strongly support that the 192 bp ERV insertions improves the expressions of TLR6, TLR1, and their downstream genes by acting as an enhancer and playing roles in the regulation of TLR pathway, which may not only alter the gene activities in the TLR pathway and in ammatory factors, but also cause phenotype variations during the immune response.

Conclusions
By using bioinformatics analysis and PCR-based veri cation, ve RIPs, located in the 3' ank of TLR3 gene, 5' ank of TLR5 gene, intron 1 of TLR6 gene, intron 1 of TLR7 gene, and 3' ank of TLR8 gene, was identi ed respectively and uneven distribution in diverse pig breeds was observed. The 192 bp ERV insertion in the intron 1 of TLR6 signi cantly increases the activity of the TLR6 promoter and multiple mini-promoters acting as an enhancer. Furthermore, the ERV insertion also enhances the expression of TLR6 and TLR1, the downstream genes (MyD88, Rac1, Rac1, Tollip) of TLR signaling pathway and the in ammatory factors (IL-6, IL-8, TNFα) in diverse tissues of 30-day piglets, as well as the serum concentrations of IL-6 and TNFα. Thus, the 192 bp ERV insertion allele is bene t for disease resistance and may be useful for molecular breeding of disease resistant animals.

RIP
The predicted core promoter regions (NC_010450.4, 30171883-30172870) of TLR6 with and without the ERV insertion was cloned from the Sujiang genomic DNA (primers were listed in Table S2), and veri ed by sequencing. Then the clones were inserted into the pGL3-Basic vectors (Ambion, Austin, American) to construct TLR6 ERV+ -Luc(En) vector and TLR6 ERV-Luc(En) Vector. In addition, β-globin and Oct4 mini-promoter were cloned from pEDV-β-globin-GFP and pTol2-Oct4-mCherry vector, respectively [63] and inserted into the pGL3-Basic vectors with or without the 192bp ERV for enhancer activity evaluation. A total of 2×10 4 PK15 and Hela cells were plated in a 24-well plates and transfected with the plasmids by using Lipofectamine 3000 reagent (Invitrogen, Carlsbad, American). After 48 hours, cells were collected for luciferase activity evaluation by using the dual luciferase reporter system (Promega, Madison, American) according to the manufacturer's protocol.

Expression analysis
The Sujing piglets were genotyped and ve piglets for each genotype (ERV +/+ , ERV +/-, and ERV -/-) were selected and slaughtered to collect tissue samples at 30 days. The mRNAs were extracted and cDNAs were prepared according to the manufacturer's protocol by using TAKARA Kit(Takara,Tokyo,Japan). Then, the mRNA expressions of TLR6, TLR1, MyD88, RAC1, TOLLip, TIRAP, TNFα, IL-6, and IL-8 mRNA were evaluated by quantitative real-time PCR (qPCR) using the 7900HT Fast Real-Time PCR System (Applied Biosystems, new York, American) in a total volume of 20 μl containing SYBR mix (10 μl), primers (4 ng), and cDNA sample (50 ng) according to the manufacturer's instructions (Takara,Tokyo, Japan). Glyceraldehyde-3phosphate dehydrogenase (GAPDH) was used as an endogenous control to normalize the target gene expression in four different tissues including liver, lung, kidney, and spleen. All gene expression was measured using the 2 − ΔΔCt method. PCR products were run on 1.5% ethidium bromide-stained agarose gels and con rmed using melting curve analyses to assess PCR product quality.
Measurement of serum TNFα, IL-6, and IL8 by enzyme linked immunosorbent assay (ELISA) TNFα, IL-6, IL8 concentration in serum of 30-day Sujian piglets were measured using the Pig TNFα, IL-6, and IL8 ELISA Kit (Solarbio Science, Beijing, China) by following the manufacturer's protocol. All measurements were performed in 3 replicates, and mean values were used for statistical analysis.

Statistical analysis
Experimental results were processed by statistical SPSS17.0 software package (SPSS, Chicago, USA) using one-way analysis of variance with Tukey's post hoc test, and the data were expressed as mean ± S.D.

Animal welfare
All treatments and protocols involving animals in this study was been strictly done in accordance with the guidelines of the Animal Experiment Ethics Committee of Yangzhou University (approval number:

Availability of data and materials
All data needed to evaluate the conclusions in this paper are present either in the main text or the Supporting information.
Ethics approval and consent to participate Not applicable.  Effect of ERV insertion on TLR6 promoter activity. A: sequence analysis of pig TLR6 and TLR1; B: a schematic diagram of the recombinant vector using pGL3-basic vector. En: SV40 enhancer; C: the luciferase activity assays. ** showed P<0.01 between groups(P<0.01).

Figure 3
Effect of ERV insertion on mini promoter activity. A: plasmid diagram of the recombinant vector using pGL3basic vector; B: the luciferase activity assays of ERV on activity of β globin and Oct4 in PK15; C: the luciferase activity assays of ERV on activity of β globin and Oct4 in Hela cell. * showed P<0.05; ** showed P<0.01.